World Bank Tanzania relations, 1961-1987 (Vol. 2 of 2) : Sectoral development

Report No. I Tanzania World Bank/Tanzania Relations, 1961-1987 (In Two Volumes) Volume II: Sectoral Department January 16, 1990 Operations Evaluation Department FOR OFFICIAL USE ONLY Document of the world Bank This document has a restricted distribution and may be used by recipients o qnly in the Performance -of their official duties. Its contents may not otherwise. be disclosed withopt World Bank authorization. ACRONYMS BIS - Basic Industry Strategy CDC - Commonwealth Development Corporation CPP - Country Program Paper DEG - Deutsche Finanzierungsgeselleshaft Fuer Beteiligungen in Entwicklungslanden GmbH (Germany Finance Company for Investments in Developing Countries) DFC - Development Finance Company EACH - East Africa Common Market EIB - European Investment Bank ERP - Economic Recovery Program FMO - Nederlandse Financierings-Maatschappij Voor Ontwikkelingslanden N.Y. (Netherlands Finance Company for Developing Countries) ILO - International Labor Office KF - Kuwait Fund MIS - Management Information Systems NDC - National Development Corporation NIC - Nordic Investment Bank PCR - Project Completion Report PPAR - Project Performance Audit Report PR - President's Report SAP - Structural Adjustment Program SAR - Staff Appraisal Report SIDA - Swedish International Development Authority SSI - Small Scale Industry TAC - Tanzania Audit Corporation TANU - Tanganyika African National Union (the Party) TDFL - Tanganyika Development Finance Company, Ltd. TEXCO - National Textile Corporation TIB - Tanzania Investment Bank TISCO - Tanzania Industrial Studies and Consulring Organization U m U N TANZANIA WORLD BANK/TANZANIA RELATIONS, 1961-1987 VOLUME TWO - SECTORAL RELATIONS TABLE OF CONTENTS Page No. REVIEW OF AGRICULTURE AND RURAL DEVELOPMENT, 1961-1986 I. THE SETTING ............................................... 1 A. Background ............................................. 1 B. Characteristics of Agricultural Production ............. 1 C. Agricultural Performance ............................... 1 D. Changes in Agricultural Production ..................... 7 E. Reasons for Agricultural Decline/Stagnation ............ 9 II. GOVERNMENT/BANK POLICIES TOWARDS THE AGRICULTURAL SECTOR 11 A. Government Policies Towards Agriculture and Rural Development ............................... 11 1961-66 Post Independence .............................. 11 1967-69 Tanzanian Socialism ............................ 11 1969-1976 Increasing Government Predominance .......... 12 1976-80 The Goal of Self-Sufficiency .................. 13 1981-1986 Economic Crisis ............................. 14 B. Bank Policies .......................................... 15 1961-1972 Growth and Transformation .............. 15 1973-1979 The Project Approach ........................ 16 1979-1986 Emphasis on the Policy Environment .......... 19 C. Dynamics of Interaction ................................ 20 Bank Role in Influencing the Development Path .......... 20 Quality of Bank Analysis for Policy Dialogue ........... 23 Capacity of Bank to Influence the Development Path ..... 25 III. BANK LENDING .............................................. 27 A. Size and Phasing of the Lending Program ................ 27 B. Content of the Lending Program ......................... 29 C. Compatibility of the Lending Program with Bank Objectives .......................................... 31 D. Performance of the Lending Program ..................... 32 E. Results of the Lending Progran ......................... 38 IV. LESSONS FOR THE FUTURE .................................... 40 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. TABLE OF CONTENTS (continued) Page No. TANZANIA'S INDUSTRIALIZATION EFFORT 1961-1987 I. INTRODUCTION ............................................... 45 II. STRATEGY AND POLICY FRAMEWORK .......................... 46 Immediate Post-Independence Period (1961-66) .... .......... 46 Socialism and Self-Reliance (1967-Late 1970s) .............. 49 Retrenchment and Policy Reorientation (Late 1970s to Date) . ............................ . 55 III. PERFORMANCE AND STRUCTURAL CHANGE ......................... 56 Immediate Post-Independence Period (1961-66) ............... 56 Socialism and Self-Reliance (1967-Late 1970s) .............. 59 Retrenchment - The Late 1970s and After .................... 71 IV. SCOPE OF BANK SECTOR WORK .................................. 73 The Bank's Assessment of the Basic Industry Strategy ....... 73 The Industrial Reorientation Drive ........................ 76 Major Recurrent Issues Highlighted by Bank Missions ........ 77 Investment Planning, Coordination of Investment Decisions, and Project Selection ...................... 77 Policy Framework .. .... .. .............................. 78 Organizational Structure and Performance of Parastatals .. 79 Actions Recommended by Bank Missions ................. 80 Efficiency of the Socio-Economic System ................. 80 Policy and Incentive Framework ..................e.o...... 81 Planning and Economic Management ......................... 82 Strengthening the Productive Base of the Economy ......... 84 Export Promotion ...... .... ... . ...................... 85 V. IMPACT OF BANK SECTOR WORK ................................. 86 VI. BANK LENDING FOR INDUSTRY AND ITS EFFECTIVENESS ............ 88 The Lending Program ........................................ 88 Lending to Financial Intermediaries ....................... 89 Lending for Technical Assistance ........................... 93 Direct Lending to Industry ................................ 94m Textiles . ............. .......... ........ . ... .... 94 . Pulp and Paper .......o... ............ ............ ..... 96 Leather Products ...... .... . ........ 0 ......... ....... 98 Effectiveness .................... ................. ....... 101 TABLE OF CONTENTS (continued) Paie No. VII. EVALUATION OF BANK'S APPROACH TO TANZANIA'S INDUSTRIAL DEVELOPMENT ............................ ................. . 102 .Strategy and Policy Dimensions ............................. 102 The Political Decision-Making Process ..................... 112 External Assistance ........................................ 117 VIII. THE AFTERMATH -- PRESENT STATUS OF THE INDUSTRIAL SECTOR AND PROSPECTS ....................................... 119 IX. ASSESSMENT OF THE BANK-COUNTRY RELATIONSHIP AND DIALOGUE ... 124 X. LESSONS OF EXPERIENCE AND RECOMMENDAZIONS .................. 127 Lessons Learned .............. ....... . ................... 127 Recommendations .......................................... 131 APPENDIX: THE BASIC INDUSTRY STRATEGY ..................... 137 Theoretical Underpinning ................................. 137 Relation to Basic Needs .................................. 140 Critique ................................................. 140 BANK ACTIVITY IN THE TRANSPORT SECTOR I. BACKGROUND........................... ......... ........... 143 Introduction ........... . .. ..... ... ... .................... 143 The Transport System--Description ........................ 144 Performance, Determinants and Status of the Transport Sector ................................. 147 PerforAance ....... ..... ......... ...... 147 Determinants .......... ..... ............................. 148 Status ............ o........................o............... 150 Government's Transport Policy, Planning and Administration ........................ ......... ... 151 Transport Policy ........................... ............... 151 Planning and Administration ................... ............ 153 II. BANK'S SECTOR WORK AND VIEWS ON POLICIES ................... 156 III. THE POLICY DIALOGUE ...... .... .................... ... . 159 Impact of the Dialogue ....... 0............ .... ........ .. 165 IV.o BANK LENDING AND ITS IMPACT ... ..... . ..... ... ........ .s... 166 The Lending Program ........................................ 166 The Impact of Bank Lending ....o...................... 168 V. LESSONS _. ._........ o.... ..... . . .. . . .. . . .. . . . ..... 169 TABLE OF CONTENTS (continued) Page No. APPENDICES APPENDIX A: Chronology of Events.................................... 171 APPENDIX B: Tanzania: Bank Reports................................. 172 APPENDIX C: Notes on Transport Projects Financed by the Bank Group, 1964-1988................................... 173 THE BANK'S ROLE IN SELECTED SECTORS 1. WATER SUPPLY AND SANITATION ............................... 203 Government Policies and Sector Organization ............... 203 Bank Policies and Involvement ............................. 204 Conclusions ............................................... 206 II. POWER AND ENERGY .......................................... 206 The sector and Government Policies ........................ 206 Energy .................................................. . 206 Electric Power ............................................ 207 Bank Involvement and Lending .............................. 209 Conclusions ............................................... 210 III. TELECOMMUNICATIONS ........................................ 210 Background ................................................ 210 The Sector and Government Policies ........................ 211 Bank Involvement .......................................... 212 Conclusions ............................................... 213 IV. URBAN DEVELOPMENT ......................................... 214 Government Policies and The Sector ........................ 214 Bank Involvement .......................................... 215 Conclusions ........................................ ...... 216 TABLE OF CONTENTS (Continued) Page No. THE EDUCATION SECTOR I. THE EDUCATION SECTOR AND ITS EVOLUTION .................... 217 A. Education Policy ....................................... 217 B. Achievements and Problems .............................. 218 C. Proximate and Underlying Causes ....................... 220 II. THE BANK'S PERCEPTIONS AND ACTIVITIES ..................... 224 A. Analysis of the Education Sector ....................... 224 B. Bank Operations ........................................ 227 1. Concept and Design of Projects ...................... 228 2. Instruments to Accomplish These Goals: Civil Works and Covenants ......................... 232 3. Implementation and Supervision ...................... 233 4. Project Outcomes .................................... 234 III. LESSONS AND FUTURE DIRECTIONS.............................. 240 Table 1: Education Indicators in Tanzania and Sub-Saharan Africa, 1960-1983.................................... 244 Figure 1: Education Pyramid, Tanzania 1961, 1971, and 1980....... 245 Table 2: Enrollment Data for Public Schools by Grade and Sex, Tanzania, 1961, 1971 and 1980........................ 245 Table 3t Public Expenditure on Education in Tanzania, Sub-Saharan Africa and Other Developing Countries.... 246 APPENDICES APPENDIX 1 - World Bank Education Projects in Tanzania Summary...... 247 APPENDIX 2 - Covenants in World Bank Education Projects in Tanzania 255 APPENDIX 3 - Expected and Actual Results of World Bank Education Projects in Tanzania....................... 262 TANZANIA WORLD BANKITANZANIA RELATIONS, 1961-1987 REVIEW OF AGRICULTURE AND RURAL DEVELOPMENT 1961-1986 I. THE SETTING A. Background 1.01 The importance of agriculture in Tanzania has increased over the years: contribution to GDP rising from about 40Z in 1965 to 502 in 1985 while continuing to supply about 802 of total export earnings and providing employment for 87% of the population. Tanzania has a diversified agricul- tural base: major cash crops in value terms are coffee, tea, tobacco, sisal, cloves and cashew; maize the principal staple supplemented by cassava and bananas. Other resources include: the second largest live- stock herd in Africa; almost half the land area classified as forest land, most suitable for some type of forestry activity; extensive fishery re- sources in its three major lakes and 1,000 km of coastline. 1.02 Tanzania has a total land area of 88.4 million hectare of which 49.1 million or 55.6Z is classified as agricultural land. With a rural population of 18.4 million (1984) populption/agricultural land ratio is 0.37, which is low compared with Kenya at 1.09 and Malawi at 1.69. High- est population densities are recorded on the periphery of the country: on the slopes of Mount Kilimanjaro to the North, the surrounds of Lake Victoria to the West and increasingly in the Southern Highlands where rich- er soils predominate and rainfall is more reliable. Soils in Tanzania are generally of low fertility. Rainfall, largely monomodal, ranges between 600-2,000 mm p.a. with wide variability within and between regions from . year to year. B. Characteristics of Agricultural Production 1.03 Agricultural production is principally a smallholder activity, the majority of holdings being 2 ha or less. The principal foodcrop is maize which is grown throughout Tanzania in all but the semi-arid regions (where sorghum and millet predominate), supplemented by banana in the higher ele- vations. Smallholder cash crops similarly vary according to ecological conditions. Coffee, tea and pyrethrum are the principal crops in the moun- tainous areas, cotton and tobacco in the western and central regions and cashew and coconut on the coastal plains. Intercropping is common: maize with beans; cassava/groundnuts; bananalcoffee. Livestock is an important activity; the livestock herd is almost entirely owned by smallholders, the animals generally being kept as security. Private estates are important for tea production, private and public estates for sugar and sisal produc- tion. C. Agricultural Performance 1.04 Growth of agricultural production over the period has been variable: Table 1.1: ANNUAL REAL GDP GROWTH RATES (2) 1967-73 1974-78 1979-81 1982-84 GDP 4.4 .t.1 1.7 1.1 Agriculture 2.3 4.7 -1.0 1.8 Per Capita GDP 2.5 0.7 -1.4 -2.1 Source: U. Lele and L. R. Meyers, DRD Report No. 273, dated May 1987. Performance in the agriculture sector between 1961-66 was strong. Annual growth rates of all principal export crops (except sisal) were at least 72; increasing numbers of smallholders were being brought into cash crop pro- duction; cooperatives for large farmers (Tanganyika Farmers Association) and small farmers (coffee, cotton) were well established; and research was provided under the auspices of the East African Community. In the late 1960s export crops stagnated and quality deteriorated as production by estates declined and smallholders' share in production became more impor- tant. The latter part of the seventies and early eighties were marked by declines in the volumes and quality of six major export crops, and in- creases in the production of food crops although shortages in the preferred grains - maize, wheat, and rice contiaued. Since 1982 agricultural ;roduc- tion has been recovering slowly, with cotton and maize being the important contributors to the recovery. 1.05 Between 1965 and 1985 marketed production of major export crops, except tea, steadily declined or remained at the same level (Graph 1). Data on area and yields are considered unreliable./ production is, there- fore, measured by purchases through official marketing channels. 1/ A 1987 report states: 'Tanzania is unusual, even for a developing country, in that there is a general shortage of reliable statistical information, and an almost complete lack of statistics on crop acreages...(furthermore) the District Agricultural Development Officers usually fake data, particularly hectarages and yields' (FA0/CP report 96187 CP-URT 27 dated July 27, 1987). Marketed Production of Key Export Crops 1965 - 240 - - - - -------- - ------ ------ 220 200 - 180 - 0 0 150- 140 - c 120 - o0 - 60 40 - 20 - 0 ,-- r 1965 1967 1969 1971 1973* 1975 1977 1979 1981 1983 1985 1987 L3 Cofton '+ Coffee ,Tobacco a Tec X Sso An index of export crop purchases for --he 1072173-1986187 seasons shows the following: Table 1.2: INDEX OF QUANTITIES OF EXPORT CROPS PURCHASES a (1986187 - 100) 72-73 73-74 74-76 75-76 76-77 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 65-6 88-87 MiId Coffee (cIean) 128 10D 143 149 129 139 123 113 192 149 141 182 180 148 100 Hard Coffee (clean) 91 106 93 98 9' 108 122 124 102 102 107 94 92 108 100 Seed Cotton 108 88 97 s9 9A 71 88 93 87 62 80 56 78 81 100 Sleal * (Fibre) 421 389 846 823 265 249 221 233 200 164 128 104 88 77 100 Flue-Cured Tobacco 71 78 102 79 97 98 97 68 88 8 84 80 71 81 100 Fire-Cured Tobacco 179 129 214 164 821 284 198 293 257 286 279 160 193 29 100 Tea (Made) 94 87 99 92 108 181 125 123 116 111 125 108 118 110 100 Cashownut (raw) 761 868 712 499 592 416 346 251 847 268 195 293 197 114 100 Py-thp*m 833 27S 892 328 275 208 188 133 167 188 133 117 125 117 100 LR Annual Review of Agricultural Marketinge 1987, Marketing Development Bureau, Ministry of Agriculture, Dar as Salaam. Agricultural export earnings, in ti985) constant terms, declined from Tshs 9,074 million in 1974 to Tshs 4,442 million in 1986.2/ 1.06 Data on foodcrop production is weak: four different sources are available (Table 1.3). Typically between 70-80Z of foodcrop production in Tanzania is :onsumed in the subsistence sector. Food imports have fluc- tuated considerably, reflecting seasonality of official crop purchases (Graph III). Some data on foodcrop production suggests that maize produc- tion has kept pace with the increase in population,31 however, official figures also highlight the extent to which imports have contributed to total supplies (almost two thirds in the early 1980s).A/ Table 1.3: FOOD PRODUCTION ESTIMATES, 1965-85 /a ('000 tons) Maize 72-78 73-74 74-75 75-76 76-77 77-78 78-79 79-80 80-81 81-82 62-8 83-84 84-85 FAD 980 780 780 825 897 988 1,000 900 800 - - - - MILND 881 '..047 1.275 1.680 1,684 1.11 L,8A 1.794 1.854 1,943 1.946 1,939 - USDA - - - - - 988 1.041 1.107 1,000 1,430 1,800 1.250 1.131 Ji * - - - - - - 1,820 1.549 1.939 2,093 ]A SGreSS: FAG/CP EW OU - Early Warning Unit, Dar ea Salame 2/ Annual Rex '.ew of Agricultural Marketing g. cit. 3/ FAO/CP o. cit. 4/ Annual Review of Agricultural Marketing, 1987, p. cit. -5- GRAPP II Terms of Trade for Selected Products 1983 109 CASHEWNUTS COFFEE tco- zo 190 140 1u0 * 40 4;i '20 0 too 0 72 '4 78 78 80 81 4 8 71 74 m8 n8 m0 al 84 COTTON SISAL 120 260. Y.0 240. 220. 120. go. 290. * tco - 70 1 90 200 05 72 74 78 78 80 al la 88 n1 74 ing 78 80 61 84 ca TEA T"~ACCO' 140 170. 120 fao- -40-0 o 120 § tig * 120 100 110. 0 1"00 100 40 74 87 8 8 9o lo- 74 m 7 5 2 8 72 74 7S 78 80 5 4 85 Source: 11OA, Annual Review of Agricultural Marketing, 1987. Official Purchases of Staple Cereals 19674- 87 260 - 240 - 220 - C 200 - 180 - c 160 - 0 140 - 120 - 0 aC 100 6 :3 9. 80 60 - * O 40 - 20 -4 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 Fiscal Year 0 Wheat Flour + Rice * Sembe A Total 1.07 The livestock subsector contributes about 12% of GDP. Production is low given the extent of rangeland and size of herd (approximately 9.4 million according to the 1972 livestock census). Almost all cattle are held in the traditional sector, concentrated in tsetse free areas, most owned by the Sukuma in northwestern Tanzania, principally cotton/maize farmers and the Masai in central/north eastern parts of the country. Cattle are individually owned but communally grazed with numbers of cattle rather than quality the determining factor. Migration is an important feature of cattle husbandry. Investment in large scale ranching activities failed to result in any significant increase in livestock numbers, produc- tivity or offtake. Similar investments in dairy production, through para- statals, have proved too cumbersome and expensive to justify the costs involved, milk production has not increased as expected. D. Changes in Agricultural Production 1.08 Between 1965-85 the structure of production changed as between the estate and smallholder sector in terms of area under tea and tobacco but less so in terms of contribution to volume of marketed production: small- holders cultivated almost half tuhe area under tea in 1987 but produced only about 302 of made tea. There has been little diversification of crop activ- ity: principal export crops in value terms continuing to be coffee, cotton, tobacco, tea, sugar, and cashew (Chart I). 1.09 The decline in marketed production of crops between 1972-1986 (Table 1.2) has been most accentuated in respect of cashew, sisal, pyrethrum and flue-cured tobacco. Production of coffee and tea remained relatively stable. Between 1986-88 there was a recovery in the Index of export crop purchases but this has been almost entirely due to a recovery in cotton production. Cotton purchases have now been restored to 1965 levels due to area, not yield, increase. Coffee purchases in 1986/87 fell to their lowest levels in fifteen years; production of cashew and sisal recovered only slightly.5/ 5/ Annual Review of Agricultural Marketing, 1987 op. cit. - 8 - I Composition of Total Agricultural Export Value 1969 Ot.ers (23 SS) Coffee (21.I) Too (4.13) Cottefi (19.98) Cashe'ants (1t.45) 17 Toboco (3.24) Stess (ti.7l) (245 Caauws (?.Ia) tobaces (.8) Coffee (43.19) Sla (1.7s) 1986 ~ Others (2.73) Cas.awaet (6.2.) Tebeoo (4.6s) Sisal (3.01) Cotto. (11.3) Coffee (67.1) Sources: 1969 chart - 1974 Agricultural Sector Report; 1975 and 1986 charts - MOA, Annual Review of Agricultural Marketina - 1987. - 9 - E. Reasons for Agricultural Decline/Stagnation 1.10 Reasons for this decline in agricultural production are by now well known and have been extensively documented over a period of years.6/ They may be grouped under the headings (i) low national priority; (ii) sun- dry government policies; (iii) declining prices; (iv) low yields; (v) donor attitudes. (i) Despite the Government's avowed support for agriculture the sector had a low priority for public resources; receiving about 10Z of total budgetary expenditure during FY66 - FY76, falling to 7Z by FY80; the industrial sector receiving 112 in FY80 up from 2Z in FY72. In the distribution of development expenditure by ministries the share of Agriculture less than doubled during the 1970s while that to Industry increased more than fourfold.7/ The distribution of external resources among major ministries between FY76 and FY82 shows an even more drastic bias toward industry with its share growing from 9% to 212 and agriculture's falling from 40Z to 10%. 8/ While the diminishing share of agriculture partly reflects the rising importance of defense, it is clear that donors priorities essentially complemented national priority of the "basic indus- tries" after 1975. The low priority of agriculture was reflected in neglect of the infrastructure for research and extension. (ii) The agricultural policy environment was characterized by prolonged institutional instability due to the increasing and changing role of the public sector in control of production and marketIng. Public investment in agriculture went mainly to the crop parasta- tals in the latter half of the seventies, e.g. sugar, sisal, cashewnut and coffee. Villagization involving the relocation of the majority of the rural population; operation maduka under which private village retail shops were closed; the replacement of the cooperative marketing system with crop parastatals; and the intro- duction of communal and later planned "block, farming were the more important policy contributors to institutional instability and poor agricultural performance. (iii) The behavior of prices have been a factor in Tanzania's agricul- tural performance, though the exact impact is difficult to deter- mine. Producer prices of export crops have been adversely affected both by international prices, the overvaluation of the Tanzanian shilling implying an implicit tax on export crops, and by the pricing practices of marketing boards.9/ In real terms the prices g/ IBRD Agriculture Sector Report, 1983; Lele and Meyers, 22. cit; Frances Stewart, Economic Policies and Agricultural Performance, the Case of Tanzania, OECD, Paris, 1986; PPAR: OED Report No. 5197, dated July 12, 1984 op. cit. 7/ IBRD Current Economic Memorandum, 1984, Table 5.10 Statistical Appendix. 8/ IBRD, 22. cit., Table 5.13 Statistical Appendix. 9/ Uma Lele & Richard Meyerst MADIA study 1987, op. cit. p.24 'using the 1972 base Tanzania's exchange rate overvaluation increased sharply over time with the index reaching 33.9 in 1985'. of major export crops- cashew, coffee, pyrethrum and sisal fell by 20 - 442 between 1969 and 1974 and again by 6 - 38% between 1974 and 1987 with the exception of cashews.10/ While the index of unit values for major agricultural exports increased by about 2162 between 1970 and 1980 the terms of trade for Tanzania's principal crops, except cashew, were in decline between 1972 and 1987. Real producer prices for food crops have also declined, although to a lesser extent than for export crops, with the result that the prices of export crops have fallen significantly relative to food crops especially after 1973174. 1ll The behavior of producer prices, given that of export prices, has been due to the fact that they were the residual after deduction of costs of the relevant crop parastatal; producer incentive falling as parastatal overhead rose. Meanwhile, prices of food crops reflected the drought in- duced priority the Government gave to food security in 1974 and the consequent raising of incentives to maize, sorghum, millet and cassava. (iv) Although the data vary widely, it is evident that there has been no significant increase in yields of smallholder crops. In the case of maize for instance, unlike in Kenya where rapid uptake of hybrids has led to 602 of cultivation in improved varieties in Tanzania the corresponding figure is only 10Z, increase in produc- tion coming from increase in area cultivated. Inputs such as fer- tilizer, herbicides, fungicides are either not applied, or applied late or in insufficient amounts. Available data and mission esti- mates suggest that yields for all crops are below their potential judged either by those previously obtained or those obtained in similar ecological conditions in other African countries. One factor which affected the availability of inputs has been the scarcity of foreign exchange for agriculture brought about by two oil price shocks in 1973 and 1979, the decline in the terms of trade, the deviation of resources to defense during the war with Uganda in 1979, the impact of the 1973/74 drought on food imports, the breakup of the East African Community in 1977, and the 40% decline in the volume of export crops between 1973 and 1981. The foreign exchange requirement for food and export crops has been estimated to vary between 10% and 632, the lowest being coffee and cashew. (v) Finally, the uncritical support by the donor community, including the World Bank, until 1980 allowed Tanzania to follow a develop- ment strategy and a pattern of resource allocation inimical to agricultural development. Government and Bank policies which con- tributed to the poor performance of agriculture will be explored in detail below. 10/ Ellis, Agricultural Price Policy in Tanzania, World Development, April 1982. Annual Review of Agricultural Marketing, 1987 o. cit. 11/ IBRD, CEM 1984, .cit. I II. GOVERNMENT/BANK POLICIES TOWARDS THE AGRICULTURAL SECTOR A. Government Policies Towards Agriculture and Rural Development 2.01 For convenience it is useful to divide the history of Government policies into five periods: 1961-66 P%it Independence, 1967-69 Tanzanian Socialism, 1969-76 Increasing Government Predominance, 1976-80 The Goal of Self Sufficiency, 1981-86 Economic Crisis. 1961-66 Post Independence 2.02 Following Independence in 1961, the first significant statement on economic policy of the Government of Tanzania was the First Five-Year Development Plan adopted in 1964. In terms of agriculture and rural devel- opment, the strategy reflected the 1961 Bank reportl2 which emphasized smallholder development through a transformation approach to be employed simultaneously with an improvement approach. 2.03 The former was a long-run strategy and called for public sponsor- ship of sixty pilot village settlement schemes, to be established by 1970, fostering a modern system of cultivation and land use. New settlers, re- ceptive to innovation, would cultivate intensively holdings of economic size, following soil conservation measures and good husbandry practices. The new settlement schemes would also benefit from support services such as schools, water, sewage, and feeder roads. The long run transformation strategy did not work well. Publicly sponsored settlement schemes encoun- tered farmer resistance and implementation problems. The strategy was abandoned in April 1966 having incurred Tshs 20 million in losses. 2.04 The improvement approach was a short-run strategy. It called upon government to concentrate public resources to extension, marketing, and credit services in smallholder areas of high growth potential. Promotion of smallholder cash crop production was consistent with the government's desire to generate foreign exchange for much-needed inputs which could not be provided domestically. 2.05 In the wake of the collapse in international sisal prices, govern- ment instituted a 52 Development Levy on export crops, the first of such levies which were to continue until 1980. 1967-69 Tanzanian Socialism 2.06 In January 1967 the National Executive of the ruling party - the Tanganyika African National Union declared its commitment to socialism defined in the context of Tanzania and reconfirmed its commitment to self- reliance in national development. This Arusha Declaration was followed by more detailed pronouncements on education, rural development, public owner- ship, and the behavior of leaders, inter alia. The importance of the Arusha Declaration for Tanzania's agricultural development is difficult to over- 12/ IBRD: "The Economic Development of Tanganyika", Johns Hopkins Press, 1961. estimate. It provided the main guideline to the Government in all its fu- ture policy decisions, and was followed by concrete actions involving the extension of government controls over production, marketing, distribution, financial and other institutions, and involving the setting up of the basic rural economic and social communal unit - the ujamaa village. 2.07 The concept of ujamaa involved the application of the three bases of traditional African family living - respect, common property , and obli- gation to work, in formalizing a nationwide strategy of development. Ujpmaa villages were to avoid the mistakes of overcapitalization and premature mechanization that earlier village settlement schemes had made. They were to be self-reliant, technologically and otherwise; and the congregation of the hitherto scattered rural population into larger villages as advocated in the Bank's transformation approach was to assure economies of scale in the provision and utilization of services. Decision-making regarding agri- cultural production was to be transferred fr3m the individual to the col- lective. Villages were to promot community plots where the work and any surplus generated were to be shared, but farmers were to retain their pri- vate plots. As originally enunciated participation was to be voluntary. However, ujamaa policies vare far-reaching and complex, and their implemen- tation led to varying interpretations and to conflicts between objectives. The expectation that ujamaa could achieve a revolutionary transformation of rural Tanzania implied a juxtaposition of voluntary participation and vil- lage autonomy against a directed effort by the state using bureaucrats with an authoritarian and managerial approach, and whose performance was judged by their ability to start new ujamaa villages. Soon the allocation of public resources and services was distorted to favor ujamaa villages in order to make participation in them attractive.13/ In the years immediately following the Declaration, the rate of development of ujamaa villages was disappointingly slow in government's eyes. Despite vigorous promotion by Regional Commissioners appointed by the Party and priority given to the ujamaa villages for social infrastructure and technical support, some five years later only 2 million people were reported as having moved to ujamaa villages. 1969-1976 Increasing Government Predominance 2.08 The Second Five-Year Plan was announced in 1969 its main goal being to implement the principles of the Arusha Declaration. The Plan ex- pressed desire for the country to move away from its heavy dependence on agriculture; but agricultural production had to be increased in order to earn foreign exchange. To this end, export crops were encouraged. Follow- ing Bank recommendations, coffee and sisal, the second and third most in- portant contributors to export earnings, were officially discouraged due to declining world market prospects, diversification was proposed to tea, flue-cured tobacco, rice, wheat, cashew nuts. Fruits and vegetables, fish and livestock were enceuraged in order to alleviate nutritional problems. The emphasis on rice, wheat, and dairy products was a direct attempt at import substitution. The Plan also proposed the establishment of 29 new state-run farms. 13/ Hyden, Goramt Beyond Ujamaa in Tanzania, underdevelopment and an uncaptured peasantry; University of California Press, 1980 13 2.09 This period also marked the decentralization of government which took effect on July 1, 1972. Responsibility for the planning and admini- stration of government services devolved to the regional headquarters --whose Regional Commissioners reported to the central government through the Prime Minister's Office. Headquarters staff were transferred from Dar es Salaam to the regions and came under the control of the District and Regional Commissioners and not the respective line Ministries as before. 2.10 New policies regarding land continued to evolve. Land registration was not encouraged, the traditional tribal village authority was formally abolished, replaced with public ownership of land whereby an individual no longer could have the right of ownership or sale. Government nationalized many private estates and prevented the development of further private prop- erty ownership as well as a land market.14/ 2.11 Similarly, policies affecting the use of labor also evolved. These included active implementation of a minimum wage, restrictions on inter- regional movements of labor, encouragement of trade unions in the case of estate agriculture, discouragement of the use of hired labor by small and medium producers tending to create artificial shortages of labor. These constituted a disincentive to produce labor intensive crops, e.g. coffee, tea, tobacco and sisal, although the objective was to remove opportunities for exploitation of the peasantry by the rentier class. 2.12 A reinvigorated effort at villagization was the central tenet of the rural development policy of the Second Five-Year Plan. President Nyerere was expecting quicker progress than was being made. A 1973 Presidential Circular changed the concept of voluntary ujamaa to one of compulsory villagization for those farmers who had still not conformed; all rural people were to be living in villages of a minimum size of 250 fami- lies by 1976. By 1976 compulsory villagization had culminated in 13 million people having been moved, often forcibly, to new and much larger develop- ment villages; block farming had replaced communal farming with rules established for minimum hectarage of designated cash crops. 1976-80 The Goal of Self-Sufficiency 2.13 The Third Five-Year Plan was enacted in 1976. This Plan had the goal of achieving self-sufficiency in food by 1981. Specifically, the strategy shifted from small-scale operations to the promotion of larger- scale farming, including irrigation. On the price side, more favorable producer prices for foodcrops were accompanied by a start in phasing out many of the export taxes. In the area of crop development, the emphasis shifted back to coffee and sisal, and livestock was encouraged. 2.14 These years also saw substantial changes to agricultural support- ing services: the abolition of long established farmer cooperatives and marketing boards and the introduction of crop parastatals--the chosen in- struments of rural development--tc set producer prices, provide all sup- porting services to farmers including markoting and establish government control over foreign exchange earnings; the curtailment of consumer goods: 14/ Hadia study, 1987: 2.*it. p.67 14 in 1976, operation maduka c'osed down all the remaining private shops in the rural areas. Export taxes, heavy during the 1970s, reached their peak in 1977. 1981-1986 Economic Crisis 2.15 The Fourth Five-Year Plan (1981-86) was soon overtaken by economic circumstances. The virtual collapse of the agricultural economy reflected the sharp decline of boom tea and coffee prices (1977-78), a further dete- rioration in the availability of inputs due to balance of payments crisis aggravated by the second oil price shock, the war with Uganda,and the de- cline in external assistance. The suspension of the Extended Financing Facility with the IMF in August 1980 had the result of intensifying the dialogue with the Bank concerning the need for a structural adjustment assistance with quick disbursing funds. Although policy measures for the agricultural sector were added at a late stage to the compromise Export Rehabilitation Credit agreed upon in 1981, it was not until the following year that agriculture returned to center stage. 2.16 Government responded by forming the Task Force on National Agri- cultural Policy in 1982, as part of its Structural Adjustment Programme, 1982/83-84/85. The task force was candid in its approach and there was recognition of the impact of domestic policies as well as that of exogenous / factors on agricultural production. It identified the major problem in Tanzanian agricultural policy to be the lack of communication between the Party and government on the one hand, and the government and the individ- uals responsible for implementation, on the other. The report recommended institutional pluralism in the agricultural sector. While the Party was promoting collective farms and large parastatal entities, the task force proposed that both large private and small homestead farms could coexist and prosper together. The frequent change in agricultural marketing arrangements and institutions was cited as a barrier to growth in the sec- tor, and the report sought to legitimize private retail outlets and set the scene for the reinstatement of cooperatives. It recommended greater allo- cation of resources to agriculture, emphasized economic incentives for farmers, a reduction in the role of parastatals and a slowing of the indus- trial expansion strategy until economic recovery was achieved. However, it was significant that though the report was published, it was superseded by the Cabinet approved National Agricultural Policy of Tanzania (1983), a modified version of the report which was less market oriented and foresaw a continuing major role for the Party. 2.17 Policy measures which followed included increased allocations to agriculture for both recurrent and investment expenditure in the 1984/85 budget; higher producer prices; introduction of foreign exchange retention accounts for exporters; liberalization of the grain trade (which went beyond the NAPT); reestablishment of cooperatives, and the transfer of the remaining responsibilities of crop parastatals to newly created marketing boards. - 15 - 2.18 In 1986 a three-year Economic Recovery Program aimed at intensify- ing the reforms was launched. Its main components were: - appropriate pricing and exchange rate policies; - conservative fiscal and monetary policies; - institutional improvements to marketing and distribution. B. Bank Policies 2.19 Bank policies towards the agriculture sector are drawn from an economic sector repor published in 1961;15/ two agricultural sector reports of 1974 and 1983 respectively, prepared by the agricultural projects division; country program papers (CPPs), the responsibility of the programs division (after 1972), and published annually between 1970-75, 1977-78, thereafter biannually until the final paper in 1982. 2.20 The Bank's attitude and approach to Tanzania's agricultural sector in general has been supportive of the Government's policies even though these have been oriented toward control of most aspects of the sector. The evolution of the Bank's attitude can be usefully divided into three phases: 1961-72 Growth and Transformation; 1973-79 The Projects Approach; 1979-86 Emphasis on the Policy Environment. 1961-1972 Growth and Transformation 2.21 At the outset, Bank strategy was generally to stimulate agricul- tural production and to contribute to welfare and equity objectives, but in reality it was tentative and flexible partly because of the lack of indepth knowledge and experience in the country and partly because of uncertainty regarding the direction of political development. Its approach was consid- erably influenced by its pre-independence mission's 1961 report - OThe Economic Development of Tanganyika", which saw the possibility of agricul- tural development as depending on "improvement" of ongoing activities through better extension and credit, and on "transformation' through planned settlement schemes, research, and water management. As early as May 1963 a Bank/IDA mission noted that it had discussed village settlement schemes: "The concept of villagization as a means of engaging the African population more fully in the cash economy and of increasing the availabil- ity of economic and social services to them has the strong support of (the) President Nyerere". 16/ A 1964 memo suggests that a possible IDA project in villagization in 1964/65 was discussed in the Bank, and a letter to a Tanzanian Minister in 1965 confirmed the Bank's continuing interest in this approach.17/ The positive attitude toward villagization was not unanimously shared within the Bank, which probably explains the absence of a lending 15/ IBRD: Economic Development of Tanganyika, 1961, Johns Hopkins Press. 16/ Memo to files: July 31, 1963 17/ Letter d/d March 3, 1965 from IBRD African Dept. Director to Minister Bomani. - 16 - operation. A 1964 memo noted that the "settlement or transformation" pro- gram competed with the "improvement" program for extension staff and that "settlement" required more and better qualified staff for the numbers served. It went on to predict that no output would result from the settle- ment schemes in the next five years.1/ By December 1966 a back-to-office report said that the transformation approach through settlements had been a failure. 2.22 Even if there was no agreement on the merit of villagization, the Bank had clearly accepted the 1961 Report's strategy of pursuing growth and equity by getting peasants into the money economy by growing cash crops, by diversifying output to include new crops, and by treating cattle as a cash crop. It supported the importance of rural credit with one loan in the sixties and considered the extension service and the training of Tanzanian youth for it to be critical for agricultural development. Awareness of the acute shortage of skilled manpower was reflected in the Bank's view that as far as possible Tanzania should make existing institutions more efficient and expand their services rather than create institutions, e.g. expand the role of cooperatives in channelling credit and technical assistance to farmers. 2.23 The Bank's attitude was also reflected in its failure to reject certain recommendations of the Report. It did not reject the principle of state involvement in production and land use planning, nor did it reject the idea of land reform to provide each peasant a consolidated holding of economic size. nor did it reject the idea of acquisition by government of estates that were offered for sale. Rather than reject particular approaches the Bank was in favor of experimentation using pilot schemes to see what would really work. During this period there was little concern with the policy environment since initially the newly independe-at country had not yet put in place a regulatory framework. 2.24 When the government broadened and deepened its efforts to develop the rural sector. BAnk policies in country program papers initially re- mained rather more general: broad statements of intent towards developing the agricultural potential by (i) diversification away from sisal, coffee and cotton; (ii) intensification of existing crop/livestock activities; (iii) achievement of self-sufficiency in food production; and (iv) expanded agricultural exports to generate foreign exchange for industrialization. Toward the end of the period the project approach was beginning to dominate Bank attitude and provision was made for a number of preinvestment studies for identifying productive projects and the direction of possible Bank involvement in rural development. Meanwhile, the Bank was clearly more confident in supporting infrastructure, such as feeder and secondary roads, and manpower development through training, with the expectation of increas- ing absorptive capacity for directly productive investment in the sector. 1973-1979 The Project Approach 2.25 McNamara's speech in Nairobi in 1973 ushered in an era of Bank commitment to an action plan for rural development. The suosequent Policy 18/ Memo to files: August 3, 1964. & I Paper on rural development provided for the creation of a special depart- ment within the Bank to appraise and supervise rural development projects. These initiatives appeared to compliment well the stated objectives of Tanzania which was designated a country of concentration for the Bank's rural development efforts. Investment in the agricultural sector was to receive the highest priority. 2.26 Meanwhile it was also apparent to the Bank in 1973 that the strong record of agricultural and thus economic growth in Tanzania which had been sustained throughout much of the sixties was not continuing in the seven- ties. The 1973 CPP stated that the least satisfactory aspect of Tanzania's economic performance over the preceding five years was the sluggish growth of marketed agricultural production. Mainland exports were described as having virtually stagnated since 1966. Bank strategy, enunciated in the same CPP, to arrest this decline was to expand the size, volume and scope of the agricultural project pipeline to fill the resource gap which was increasing more rapidly than expected because of declining (agricultural) export earnings. Bank attention then focussed almost exclusively at the project level with subsequent CPPs highlighting the "substantially enlarged lending program." Projects being introduced to the Board during this period reiterated Bark support for ujamaa (Tobacco%, villagization (National Maize) and institutional change (Geita Cotton). 2.27 Within this project focus, although the Bank's overall rural development strategy regarded increased production as central to, and an integral part of, any rural development program, the Bank was sensitive to Tanzania Government thinking which wanted development resources concen- trated in the most backward areas so that all regions would participate in the development process. Debate within the Bank focussed on whether the increased flow of resources should be channeled more towards productive activities or to supporting serviceslinfrastructural programs in marginal areas, as was the wish of government. To some degree, the differing opinions expressed within the Bank were a reflection of the institutional fragmentation which followed the 1972 reorganization which created programs and projects div.isions, the rural development division in CPS and and assigned almost all supervision responsibility to the Regional Office in Nairobi. 2.28 The agricultural projects division emphasized productive invest- ment in its sector work, and cautioned against too much investment for equity/welfare objectives. The 1974 agricultural sector report which was charged by the 1973 country program paper with evaluating agricultural development since Independence to determine why performance in agriculture had been disappointing, concluded that investment in production should take precedence over support services. A dual approach was again proposed, this time in sequential fashion: with highest priority to be given to invest- ments with immediate production impact, and a lower degree of emphasis to social infrastructure and other projects with little directly productive impact. Specifically, the sector study warned of potential problems with recent Bank support of tha large, regionally-based, multisectoral projects such as the Kigoma Rural Integrated Development Project (RIDEP): in par- ticular, the tendency to support areas of low productivity, to over-empha- size social investments and to "crowd out' directly productive initiatives. - 18 - The report proposed that there be a "temporary halt" in the process of villagization so that major effort can be directed at minimizing the nega- tive impact on planted acreage and yields. It also proposed that the Government "should immedi&tely begin to implement a range of policy changes and reforms ... particularly those concerned with marketing, input supply, extension services and the proposed strengthening of the Ministry of Agriculture". The report said that improved farmers' incentives and commod- ity pricing needed special attention. That the report went on to propose a strategy for Bank intervention which left policy issues Aside and suggested two new types of projects in the dual approach mentioned above is not just surprising but an indication of the differences between technical experts and program managers and the pressures to continue - project approach, which proved an inadequate vehicle for daaling with policy issues. Further evidence of this situation is that the Management review of this sector report in October 1974 reaffirmed strong support to the Kigoma project19/ and the two follow-on projects in the pipeline for 1976 and 1977. These projects were not part of the portfolio of the agricultural projects division but of the rura. development division. 2.29 This ambivalence continued for two years. Meanwhile, warnings raised by Bank technical staff (see following chapter) at the project level were not reflected in country program papers. However, in 1978 the ambigu- ity in Bank strategy was resolved. The country program paper for that year concluded that much of the Bank's !974 agricultural sector work "has been overcome by the subsequent evolution of Tanzania's administrative and or- ganizational framework." At the same time the paper questioned the feasi- bility of the "production over equity" approach and drew attention to gov- ernment's wish to have more investment in regional infrastructure. Conse- quently a new Bank strategy emerged, not linked to any sector work, in which full priority was given to government's equity/welfare objectives by shifting the amphasis away from projects with immediate production impact toward regional development programs. Thus support for the National Agricultural Project, which had been anchored in the 1974 agricultural sector report was dropped and three new rural development projects were added to the lending program. Regional projects were no longer to stress productive activities as central to the project, support would be concen- trated on planning activities. 2.30 The focus on projects relative to policies continued. At Board presentation of new projects it was reiterated that the focus of Bank ad- vice would continue to be in the framework of project lending. This was supported by provision of technical assistance for projects, initially for preparation but latterly for implementation. At the same time, quarterly reviews of project implementation by headquarters and RMEA were instigated with monthly follow-up by the Resident MissLon in Dar es Salaam. 19/ In the annual -,eetings for two consecutive years (1974 and 1975) the Kigoma rural development project was specifically mentioned as one of the most innovative rural development projects and a model for new settlements everywhere. - 19 - 1979-1986 Emphas q on the Policy Environment 2.31 By 1979, the generally poor performance of the much increased project portfolio was causing considerable concern; the agricultural proj- ects division undertook a full review of all ongoing projects in Septem- ber/October 1980. Increasing attention was given to the impact of govern- ment policies on agricultural project performance. For the first time, technical assistance was proposed for policy analysis, however, there was no consensus within the Bank as to an agreed strategy for policy reform. Preparation missions fielded from the programs division and under different miseton leaders for a Structural Adjustment Credit(SAC), in place of a Program Loan which Tanzania had requested, outlined substantially different measures;gO/ the Bank office in Tanzania also held differing views.21/ Meanwhile, the agricultural projects division believed that reform of the agricultural sector in Tanzania had to be central to any SAC, but the j.copesal for reforms was only added at a late stage after it became evieent at the beginning of 1981 that the policy changes advocated were not to be a the core of the proposed adjustment program. Since agreement in principle had already been reached between the programs division and government the attempt to get the Government committed to these policy changes was largely ineffective. By February 1981, the agricultural projects division urged22/ that the fourth RIDEP for President Nyerere's own region of Mara appraised in January 1980 and negotiated in December 1980 should not be processed further, since this could send conflicting signals to government that "old style investment projects" were stilL %cceptable even under the new strategy which was to focus exclusivel.y on program support for policy reform. In October 1981 the programs division conceded reluctantly that the project should nnt be submitted for Board approval, after arguing that the project design took account of lessons from earlier RIDEPs, was smaller, had a production/rehabilitation and smallholder focus, and had been negotiated and preliminary implementation started.231 2.32 By 1982 and under the chairmanship of Bank President Clausen there emerged a more coordinated approach. In the final country program paper, no longer was the debate over the nature and extent of policy reform neces- sary. Instead government was urged to improve the incentives to smallhold- ers, reduce government interference in the market place, and work toward improved maintenance and utilization of existing productive capacity. At the urging of the agricultural projects division, future investment in projects was downplayed. The focus shitted from the project-led strategy of the 1970s to a more long-term program based assistance intended to achieve policy reform at the macro-economic level. There were severe cut- 20/ For the first preparation mission in early 1979, see Aide Memoire for a Framework of Program Lending in Tanzania, dated June 8 1979; for the second in March 1980, see Structural Adjustment Initiating Memorandum, dated May 9, 1980 on project files of Credit 1133-TA. 21/ Project files, Letter dated September 20, 1980 Pa. cit. 2/ Project files, Memo from Lele dated February 5, 1981. - 20 - backs in rural development expenditures and agricultural project lending in favor rif non-project lending and proposed crop rehabilitation for coffee, cotton, cashew, tobacco and sisal. Training and institution building con- tinued to be emphasized. 2.33 This emphasis on the need for domestic policy reform was also symptomatic of wider Bank thinking as developed in the 1981 Report on Accelerated Development in Sub-Saharan Africa. The recommendations of this report, which attributed problems of declining agricultural production in Africa rather more to government policies and less to exogenous factors, were underscored by the Bank's second agricultural sector study in 1983. The study, prepared by a team of over twenty Bank staff and consultants, was the first in-depth effort to analyze the overall policy and institu- tional framework as it affected project performance. In short, the report recommended that more foreign exchange earnings and other resources should be channeled into the agricultural sector, the role of the parastatals should be reduced, and the private sector should be allowed to participate in the provision of some of the support services. In addition, the Bank called for greater coordination among donors in t1-e case of Tanzania, more assistance to subsector, sector, and program support, and food aid. This emphasis to policy reform was also reflected at the project level. Nego- tiations in 1983 over the suspended Grain Storage and Milling project led to substantial liberalization of the grain trade. 2.34 The most important difference between the 1983 and the 1974 Agriculture Sector Reports reflected in the Bank's attitude, was the view in 1983 that the appropriate strattgy for Tanzanian development was one which was agriculture led. This meant hirhest priority to the sector in the allocation of budgetary and foreign exchange resources, and went far beyond the efficiency concerns of the 1974 report. Another important difference was that in 1983 the Bank made explicit its view that the role of the public sector should be reduced; a rather delayed rejection of the state intervention principle it had accepted in the sixties. Subsequent cross- country comparative work to assess the impact of the domestic policy en- vironment on agricultural performance has been interpreted as supporting the Bank's recent views.24/ C. Dynamics of Interaction Bank Role in Influencing the Development Path 2.35 In the twenty-year period between 1961 and the economic crisis of the late seventies there was a symbiotic relationship between the Bank and government, with first one and then the other taking the leadership role in the formulation of the development strategy which was then accepted by the other. Initially it was government which went along with the transforma- tion and improvement approaches as proposed by the Bank but when implemen- tation of this policy encountered difficulties, the initiative passed to government. The Arusha Declaration ushered in a new era in Bank-Tanzania relations. The Government elaborated its development strategy and struc- 23/ Project files, Memo from Dunn dated October 27, 1981. - 21 - tured the policy environment, while the Bank confined its influence to public expression of support for government policies combined with tangible support through the rapidly increasi.& lending program (see following Chapter). This close affinity which developed in the early sixties disap- peared in the late seventies when Bank views changed, and, contrary to Bank expectations, the Government was reluctant to change its policies. 2.36 In order to understand the attitude of the Bank it is necessary to look at the decision process through which its role nad strategy were de- termined. In 1970 the Bank created a new vehicle lor rationalizing its internal dialogue and for determining its strategy regarding its borrowers --an annual and later bi-annual Country Program Paper (CPP). The CPP was to (i) analyze government's objectives, (ii) review economic performance and creditworthiness, (iii) assess the role of and need for external assis- tance, and (iv) propose the level and content of the IBRD/IDA lending pro- gram. This vehicle could have been the basis for effective coordination of Bank lending and dialogue with realistic appraisal of development possibil- ities; but to be this it would have had to be critical. It would have had to derive from a process which allowed fair debate of the various, fre- quently conflicting, points of view within the Bank, and which had a cred- ible information *6ase not relying on mere statements of intent and policy announcements emanating from the country. Unfortunately, this was not the case. The Tanzania CPPs of the seventies are replete with statements sug- gesting an unwarranted stance of uncritical support for Tanzanian policies. There was a generally optimistic mood that led to the expectation that "Government would bring most of its current experiments to a successful conclusionO, (1972 CPP) and which also sent clear signals to Bank opera- tional staff.(see para. 3.22). In 1973, growth in total marketed agricul- tural production was far below plan targets, but the Bank concluded, in the country program paper for that year, "Our lending program is in harmony with the government's development priorities which are clear and in our opinion generally well conceived". 2.37 Part of the shortcoming of the CPP process derived from failure in the Bank to develop suitable measures and appropriate weights for equity among the other developmental objectives in analyzing performance. The analytical basis for Bank "strategizing" as well as for dialogue with a client for which equity was of importance was therefore deficient; and within the Bank the attitude to Tanzania seemed determined on bases other than that of cogency of analysis. The Bank resorted to measuring perfor- mance in conventional terms, but ignored the conventional conclusions flow- ing therefrom. It concluded that: "although the rate of economic growth has not been a spectacular success we are inclined to give Tanzania a good performance rating and recommend maximum Bank group effort' (1974 CPP). In the following year, although the CPP drew attention to the opposition to collective farming in the rural areas, noted the modest growth "especially" in agriculture, and the fact that the forced villagization program was having a negative impact on productivity, it again concluded 'There are no fundamental differences between the Bank and Tanzania on development objec- tives nor do we have any difference of opinion on important policy issues." The 1977 CPP paper noted the gap between government's "laudable objectives" and its performance, yet as late as 1980, the conclusion of the country program paper was that "Tanzania's development objectives are sound and in - 22 - line with the country's needs", when the relevant issues were affordability and appropriateness of the applied development strategy. 2.38 In adopting this stance of uncritical support for government poli- cies, the Bank's senior management was given strong endorsement from the Board. Even where risks were noted, the Board consistently reiterated that it was the Bank's role to be at the forefront of social innovation and change, supporting new approaches to development even at the risk of fail- ure. At presentation of one project (Cashewnut Development I) in May 1974 the Bank was urged to approve a fast disbursing program loan so that the "Tanzanian Experiment" would not be endangered by the 1973/74 economic crisis. Dissent was largely confined to concerns for domestic constit- uencies as in the case of tobacco or expropriation of assets. The excep- tion being the caution that was expressed on three occasions by one Executive Director between 1974-78 as to Tanzania's capacity to absorb the rapidly expanding pipeline and the pace of proposed agricultural project implementation. 2.39 The Bank's practice of choosing to accept and not to influence the development path led to considerable difficulty when the Bank then sought to reestablish some influence in the eighties. Given past experience when economic crises had largely been "resolved" by rapid influx of foreign exchange, the Government was understandably shocked at the long list of conditions attached by the Bank for the proposed structural adjustment lending. It was unable to understand why the Bank was then arguing that the same domestic policies which the Bank had supported throughout the previous decade were now considered inimical to agricultural growth. In reality, the dialogue based on objective evaluation was just beginning. The 1974 ASR had examined policy issues, but given the surprising projects orientation of its recommendations it failed to trigger a dialogue on pol- icy. At the beginning of the eighties there was still no viable process for reaching consensus within the Bank regarding the best strategy for devel- opment of Tanzania' agriculture. The agricultural projects division, admit- ting previous mistakes in project design and policy dialogue, was calling for policy reform in agriculture, but was unable to convince the programs division, which, at that time, did not share its view, and which was pri- marily responsible for policy dialogue. However it was recognized through- out the Bank that the very extent of support which the Bank had been sup- plying over the decade in the form of the lending program together with its influence over other donors, far from having the effect of moderating gov- ernment policies or obtaining sectoral changes through project intervention had often resulted in faster implementation of a strategy of questionable efficacy. 2.40 The preceding paragraph emphasizes that while for simplicity it is convenient to describe 'the Bank view", in fact at any one point in time Bank staff may hold many views. At the very least, one can expect some individuals or groups to hold the current "conventional wisdom" while others have already adopted revised positions which may become the new conventional wisdom. Thus the period 1980 to 1983 saw an evolution, within the Bank, of agricultural development priorities. In early 1980 it was recognized by some individuals that project problems appeared to be systemic. This led to a meeting in Arusha in late 1980 of Tanzanian and Bank staff involved with project implementation, which led to general acceptance of the "systemic" view. This in turn led to a thorough re- evaluation by the Bank of the nature and effects of Tanzanian agricultural policy in 1981 and 1982, culminating in a retreat for projects and programs staff to review the yellow cover draft of the agricultural sector report. The retreat resulted in general (but by no means complete) acceptance of the dominant role of policy (rather than resources) in explaining poor agricultural performance in Tanzania. In early 1983 the sector report was officially submitted to the Government. Meanwhile, it was inevitable that conflicting impressions and some ambiguity in the Bank's position may have been reflected in the views conveyed to the Government. Quait fBank Analysis for Policy Dialogue 2.41 The quality of Bank analysis for policy dialogue depends on (i) coverage and completeness, (ii) identification of the relevant issues, (iii) ranking of issues according to their importance, (iv) depth, correct- ness, and realism of analysis of priority issues, and (v) the guidance to policy formulation in terms of operationally feasible measures. Because policy formulation is done within the context of a political decision pro- cess to achieve certain socio/economic objectives through manipulation of economic incentives in a particular cultural context, coverage and com- pleteness require that analysis be multi-dimensional or multi-disciplinary, addressing the economic, socio/cultural and political aspects. The identi- fication of relevant issues requires an examination of the consistency among goals, the feasibility of targets given the constraints, the likely effectiveness of the strategy comprised of policies to influence private decisions and projects to provide certain service inputs to production, and the measurement of performance in order to determine where problems arise and where changes in approach are indicated. The ranking of issues accord- ing to importance is necessary in recognition of (a) the fact that policy interventions differ in resource requirements and in likely achievement of objectives, and (b) the fact that resources, including decision-making time of politicians and implementation effort of public servants, are limited. All issues cannot, and possibly should not, be attended to at once; clearly they should be dealt with according to some priority ranking for attention. Analysis of priority issues should be in sufficient depth to facilitate reliable forecast of the impact of alternative policy measures. Although policy dialogue should lead to general improvement in the understanding of issues, both on the part of the Government and the Bank, the value to the client will turn heavily on whether the Bank provides "actionable" recom- mendations, practical in being able to evoke the proper human response and backed by hard agricultural science. 2.42 Based on these criteria, the quality of Bank analysis for agricul- tural policy dialogue has been deficient. The signal deficiency for many years lay in the uni-dimenstonal bias of Bank analysis with focus almost entirely on the economic and with scant attention to the socio/cultural and the political. This was especially important for a dialogue with a client whose agri/rural development strategy was to serve profoundly political national objectives. Indeed, it was not until 1985 that 'The Politics of Agricultural Policy in Tanzania' came under scrutiny and then in the Development Research Department rather than in the Operations complex. 24 "Critical Country Issues" were identified in country program papers throug- hout the 1970s, most of which related to the agricultural sectors (i) the rapid institutional change; (ii) the question of skilled manpower shortages and the subsequent need for technical assistance. (iii) the lack of motiva- tion, inefficiency and mismanagement in much of the public sector; (iv) problems with the railways for transportation of inputs and marketed pro- duction; and (v) the negative impact of collective farming on agricultural productivity. These issues should have suggested that a multi-disciplinary approach was indispensable for proper analysis and policy advice; but this was not the case. In addition, Bank experts from time to time conceded that they had no viable technical package for Tanzanian agriculture. 2.43 The uni-disciplinary bias of Bank analysis resulted in a situa- tion in which the impact of government policies was in many instances su- perficially analyzed and poorly understood. This was the case with the villagization program, for which Bank analytical work failed to appreciate the cultural, social and political complexities given : (i) that as in many parts of Afica, affiliation is on a tribal rather than a village basis; (ii) that it proved to be more than a resettlement programme, changing the entire course of decision making from the individual to the collective/ village level; (iii) the implementation problems for procurement of inputs, services, etc., which occur when villages are expanded from their previous rather small size to a minimum of 250 families; (iv) that far from being a bottom-up approach for planning, villagization was much more a top down control of the rural population; and (v) the physical and labor constraints at the farm level when farmers are relocated many miles from their cashew trees, cane-cutters far from the cane factory, tobacco farmers with insuf- ficient family labor. It was also the case with the decentralization of public administration. The Bank did not comprehend the effects on the Ministry of Agriculture and the role of the Party; nor did it foresee the effects on the motivation of civil servants and on the delivery of exten- sion services. Dialogue on agricultural projects continued with the Ministry of Agriculture in Dar es Salaam which, in effect, no longer had control over what the agricultural technical staff were doing in the field. Subsequent projects were aimed at strengthening MOA's ability to deliver input and technical support to the regions.25/ 2.44 The quality of Bank analysis as economic analysis has not always been good, although it has improved over time. In the seventies the macro- economic work was inadequate and macro and sector work were not integrated. Meanwhile the emphasis on project identification and not on policy made for . significant gaps in the analytical work. While in the eighties there has been greater integration of and improved macro and sector economic work, there is need for prioritization of issues. The country cannot be expected to correct everything and to do so immediately. This insensitivity to the need to prioritize and to sequence is long standing, as can be seen in the list of recommendations in every report since the 1961 and in a lending program involving the Bank in almost all subsectors of the agricultural economy. A recent report pointed out that there has been no sense of 24/ Uma Lele & Richard Meyers studies carried out under DRD's entitled OManaging Agricultural Development in Africa" (MADIA). "establishing a few clear areas of lending for meeting the recipient coun- try's critical needs from a development point of view".261 Constraints have usually been identified but their implications not always analyzed; and while attention has been given to pricing there has been little work done to estimate responsiveness of production to price incentives. Apart from these omissions analysis of particular cases has been tainted by Bank concerns at the global level. For many years the Bank recommended against coffee and sisal because of poor market prospects, today coffee remains the most important smallholder crop and throughout the period has been the most important foreign exchange earner.27/ Meanwhile the Bank urged diversification into tea and tobacco to increase export earnings, and not until 1980 was it noted that unlike coffee and sisal both are very foreign exchange intensive, needing about 50% of their export earnings for imported inputs. 2.45 At the micro level Bank economic work has sometimes been of high quality and has brought about a fruitful dialogue in specific areas. An important and positive contribution has been the Bank's strong support for the smallholder, as the key element to agricultural growth. This not only fitted well with the characteristics of agricultural production in Tanzania but has also been pursued even when certain Tanzanian policy makers ques- tioned the smallholder approach and argued for large scale parastatal farms. The fact that the practical results were mixed was due to some shortcomings in terms of the Bank's financial contribution to smallholder development (see next Chapter). Capacity of Bank to Influence the Development Path 2.46 The Bank's failure to communicate its concerns over policies af- fecting agriculture in the context of the dialogue during the seventies cannot be explained in terms of the Government's reluctance to take deci- sions: throughout Tanzania's history, the Government has shown great ener- gy in the formulation and modification of its development policy. Further- more, it is evident that although the Bank was aware of general declining trends in agriculture production it did not develop the information base to quantify the extent of the problem, or to gauge the strength of needed policy adjustments. This then circumscribed its ability to suggest for discussion a range of options of needed policy actions on the part of Government. 2.47 There was uncertainty within the Bank and disagreement until 1982 between Projects and Programs Division over the integrity of the data on Tanzania. First, information on key indicators for agricultural production and the sector's contribution to GDP was weak, with little analysis of pro- duction trends, and upward adjustment of official figures accepted without question. 28/ In 1980, national income accounts were adjusted downwards 25/ Board presentation of the National Maize project, dated December 21, 1975. 26/ Aid Effectiveness in Africa: Some Implications for the Bank's Reorganization, d/d February 1987. 27/ The EEC provided some support during the period for coffee production. 26 by the Bank when annual subsistence production growth rates of 9Z were revised by half for the 1972-80 period.29/ Second, information on other performance indicators have, until recently, been lacking, e.g. data on public expenditure including actual investment trends for agriculture. 30/ At the end of the 1970s, Bank missions found it necessary to conduct extensive reviews to provide estimates of total costs of the existing and proposed public investment program to enable comparisons to be made between available government resources and permit a framework to be established to reconcile the two. Nor was data available on overall import requirements of the agricultural sector. Third, little data was available on the operations of the institutions supporting the agricultural sector. Although the Bank was financing rapid expansion of the parastatal sector there was no monitoring system built in which would enable the performance, liabilities and expenditures of the parastatals to be known and understood. A report was produced on parastatal finances at the end of the 70s but did not cover their subsidiary companies/investments. 2.48 Given the fact that until 1980 the Bank clearly saw the project as the principal veaicle for addressing identified constraints in the agri- culture, it is not surprising that the dialogue was oriented toward lending for projects and emphasized project identification and implementation rath- er than policy advice. Little effort was made to stimulate agricultural policy debate outside of government until the introduction of the Advisory group under the auspices of the Third Technical Assistance Project in 1982. There often has been poor continuity of staff; and few Bank staff have proficiency in Swahili which has been the official language since 1967. 2.49 Finally, the conduct of the dialogue when it did take place was such that frequently the Bank was not talking to the right people. Dia- logue continued largely with Bank agricultural staff, for example, talking with their counterparts in the Ministry of Agriculture even after the reor- ganization of government.31/ At the broader level, the Bank assumed that civil serv&nts were clearing proposed policy changes with the Party rather than estab..ishing any direct contact with the Party itself.321 28/ This lack of reliable data is noted despite considerable Bank support for the Marketing Development Bureau in the Ministry of Agriculture and for monitoring and evaluation in many Bank-supported projects. 29/ Although the agricultural projects division had previously protested that subsistence production had not been increasing at 92 p.a. 30/ Project files, Export Rehabilitation Credit 1133-TA, Memo dated March 5, 1981. 31/ Project Performance Audit Report: OED Report No. 5197, dated June 29, 1984 oD. cit. 32/ Project files, Export Rehabilitation Credit 1133-TA, Memo from Wapenhans to McNamara dated September 26, 1980. III. BANK LENDING A. Size and Phasing of the Lending Program 3.01 Cumulative Bank lending to Tanzania up to the end of FY86 was US$1,228.3 million of which lending for agriculture and rural development constituted US$321.0 million or 26Z of the overall commitment: Table 3.1 COMBINED IBRD AND IDA CUMULATIVE LENDING OPERATIONS BY SECTOR (As of June 30, 1986)(US$ million) Amount Z of Total Agriculture and Rural Development 320.95 26.1 Basic infrastructure 397.40 32.4 Industry 247.00 20.1 Other Infrastructure 62.00 5.0 Human Resource Development 66.90 5.4 Non Project Lending 95.00 7.7 Technical Assistance 9.00 3.2 Total 1,228.25 100.0 3.02 Bank lending to Tanzania in the agriculture sector (Graph IV) began at a modest pace. Prior to FY73, four small-scale projects primarily of a pilot nature were approved with an average loan/credit amount nf US$5.9 million. One project was dropped after appraisal for lack of a technical package. Bank/IDA Support for Agriculture By Ln/r Agrcemcnt & Fimcai Year 49- I-3 c c 20- g 10- - _ _ _ _ rt - Ti (T1 T i TTTTIfTIi il Ti iii iiTITJiiTIiilTi >6 67 68 69 70 71 72 73 74 75 7 $77 78 79 89 81 82 83 84 8S 85 FIsccil Yc9r gf Ln/Cr Agrccment Date 3.03 Thereafter the size and volume of the lending program picked up rapidly with between 2-5 projects being approved each fiscal year until 1978. Average loan/credit size also increased to US$13.5 million. No projects were rejected for technical reasons. 3.04 There followed a two-year hiatus, 1979-80; after which five new projects were approved in FY81/82, the average loan/credit amount increas- ing to US$22.6 million. Thereafter the lending program to agriculture ended: no new projects approved, loan processing for one project stopped, and poorly performing projects closed.33/ B. Content of the Lending Program 3.05 Bank lending for agriculture can be characterized by investment in support of (1) rural development in disadvantaged regions; (ii) increased production of smallholder crops; (iii) large-scale parastatal farms; (iv) capital intensive agro-industrial processing plants; (v) inputs for related industries; and (vi) program support. 3.06 Bank financing for rural development began with the Kigoma Rural Development project in 1974 followed by Tabora and Mwanza/Shinyanga RIDEPs in 1977 and 1978 respectively. All three34l/ were to support rural develop- ment in their respective regions on a multisectoral basis with primary emphasis on increasing food production and raising rural incomes. Bank support was to be channelled direct to the newly decentralized regional authorities rather than to central Ministries as before. Specifically the projects were designed to provide a range of economic and social infra- structure to newly established ujamaa villages; strengthen the regional authorities ability to deliver agricultural supporting services; provide additional technical assistance for land use planning, crop/livestock trials, monitoring and evaluation and project preparation. Investment in directly productive activities, in keeping with government wishes, general- ly comprised about 10% of base costs. Implementation of these projects, which typically had between 12-14 components was to be coordinated across a large number of agencies. 3.07 The Bank supported the introduction of smallholders into tea and tobacco production through several credits. The Smallholder Tea project in 1972 expected to settle smallholders in 8,300 ha and provide strengthening for support services and cooperatives, development of roads, water supplies and other infrastructure necessary for the industry, as well as to provide two new tea factories and expand three existing facilities. This same strategy was continued under the second phase Smallholder Tea Consolidation project approved in 1980. The Flue-Cured Tobacco project similarly ex- pected the settlement of about 15,000 smallholders in tobacco-growing areas with provision of necessary inputs; support for institutional development and infrastructure for flue-cured tobacco. It was followed in 1976 by a 33/ 10th Annual Report on Project Implementation and Supervision, dated February 27, 1985. 34/ Full details of this and subsequent projects discussed are available in Annex II. Tobacco Handling project designed to reduci post harvest losses at the farm level due to lack of barn capacity, pojor curing te&-.iques, inefficient bulking and grading. 3.08 Tea and tobacco product.son also received priority--as did cotton, coffee, dairy products, fishing and sisal--under a broad based credit proj- ect. The Agriculture Credit project was to provide funds to the National DeveloptLent Credit Agency for onlending to farmers for short-, medium- and long-term financing principally through cooperatives. 3.09 - Support was also provided by the Bank directly to cotton and to other smallholder cash crops: pyrethrum and coconut. The Geita Cotton project, approved in 1974, was expected to assist small farmers in one of the principal cotton growing areas of the country. Intensification oi production was to be obtained through mechanization, provision of necessary inputs, strengthening of supporting services including agricultural re- search, credit and extension, road construction, rural health and transport services. A Pyrethgum project in the Southern Highlands was signed in 1980 similarly to improve smallholder yields and also to stimulate production through area increase, through improved production incentives, strengthen- ing of research, extension, marketing, and development of transport ser- vices. Technical rssistance to the crop parastatal and improved district and feeder roads was also included. The Coconut Pilot project of the same year similarly provided strong emphasis toward research in an effort to rehabilitate and restore the coconut industry in coastal regions. 3.10 Bank support for livestock has been for large-scale rather than smallholder development and with heavy emphasis on processing facilities. The Beef Ranching Development project, approved in 1968, w&s essentially a pilot operation to test the establishment of cattle ranches. The second and greatly expanded phase, Second Livestock intended to complete the de- velopment of parastatal ranches started under the first project and was to establish non ranches at both the district and village level. Purchase of improved cattle, improvement of marketing, provision of two new processing facilities ani the expansion of a third existing meat plant. A Dairy Development project, which was signed in 1975, similarly comprised large scale enterprises: 17 large dairy farms incorporating modern intensive dairy practices with imported cattle and a high degree of mechanization. The project also provided related infrastructure and 20 small units on a pilot basis in 50 ujamaa villages. Similarly, Bank support for fisheries was on a large scale. Approved by the Board in 1976 with a total project cost at appraisal of US$12.4 million, the Fisheries Development project was intended to support the establishment and development of commercial fisher- ies at selected centers, improve fishing techniques and marketing data, and provide the basis for fisheries development in the smallholder sector by supporting a pilot fisheries program for ujamaa villages on Lake Tanganyika. 3.11 Of the remainder of the lending program for agriculture, eight projects directly supported large scale, agro-processing operations, with minimal emphasis given to sustaining or expanding parallel production, the aim being to increase domestic value added. - 31 - 3.12 The Cashewnut Development project, approved in 1974, expected to increase cashew processing capacity by the establishment of five new pro- cessing plants, port storage facilities and other infrastructure. A minor component supported cashew research and extension. This was followed by a second phase in 1978 which also concentrated exclusively on processing as opposed to prodr.ction: a further three new processing plank.s were to be financed with additional expansion of the ports. 3.13 A Tobacco Processing project was approved in 1976 to more than double existing capacity by both expanding existing factories and building a new facility. The Kilombero Sugar project signed in 1974 supported the development of a new 3,000 ha sugar plantation, partly irrigated, and de- velopment of about 1,800 ha of rainfed land for outgrowers, construction of a factory, roads and related infrastructure. Another large agro-industrial venture was Bank support for the National Milling Corporation (NMC) in 1981. The principal objective of the Grain Storage and Milling project was to improve NMC's capacity to procure, transport, store, and mill foodcrops through rehabilitation, construction of new facilities, and expansion of milling capacity. 3.14 Indirect support ftr agro-processing was also evident in the Bank's support for the forestry sector. A further two projects funded industrial forestry plantations to provide the necessary raw material for a Bank-supported pulp and papermill. Phase I of the Sao Hill ForesLry proj- ect was approved in 1976 to finance the establishment of about 16,000 ha of new industrial plantations while upgrading and maintaining about 11,000 ha of existing plantations, roads, and firebreaks. The second phase was ap- proved in 1982 and continued in the same vein with the further development of 10,000 ha and necessary infrastructure. 3.15 Program support for the sector began in 1981 with the Export Rehabilitation Credit. Originally intended as a Structural Adjustment Credit the project as agreed was directed at reversing the decline in vol- ume of agricultural exports and incre&sing the rate of growth of manu- factured and processed exports. This was to be accomplished through in- creased producer returns for major export crops; providing foreign exchange to finance high priority imports; improved efficiency of the parastatal sector and higher incentives for non-traditional exports. Supplementary and related measures included: improvements in the foreign exchange budge- tary and import allocation system; a review of all on-going and proposed investment projects; strengthening the planning, programming and budgeting functions of government and the preparation of an action program to improve agricultural services. The program was the largest in the Bank's portfo- lio at the time. C. Compatibility of the Lending Program with Bank Objectives 3.16 A central theme of Bank policies, particularly after 1973, has been the importance accorded to agriculture in general and the smallholder in particular. The lending program, however, has partially supported this tenet both in quantitative and qualitative terms. First, in quantitative terms, actual commitments to agriculture have fallen short of expectations embodied in the respective country program papers. - 32 - Table 8.2 Lending to Agriculturat/Rural Development as a Percentage of Overall Bank Oroup Lending Expected and Actual Commitments LendIng to 1970 1971 1978 1974 1976 1978 1978 1979 1980 1981 Agriculture Expected 88.0 46.6 68.8 42.4 86.8 89.0 81.8 27.7 18.6 12.1 Actual - - 21.6 89,8 28.9 26.6 15.7 22.4 20.1 8.1 3.17 In qualitative terms, emphasis to the smallholder sector has also been lower than might have been expected. Of the 24 agricultural projects/ programs, only about half were aimed at smallholders; the remainder fi- nanced agro-industrial prccessing projects, reflecting government priority to the "Basic Industries Strategy" or large-scale parastatal operations. 3.18 Another Bank policy was diversification of agricultural activity; in this respect the lending program as designed was well tailored to this goal. During the twenty year period, Bank support covered almost all sub- sectors: export crops--tobacco, tea, cotton, cashew, pyrethrum and coco- nut; foodcrops--principally maize, the staple crop and sugar; dairy and livestock activities; fisheries and forestry. Almost the only activity excluded from Bank support was coffee, which has continued to be Tanzania's principal foreign exchange earner and traditionally an important smallhold- er crop. 3.19 The Bank has also provided considerable support in efforts to obtain geographical diversification. Projects as designed were to support the production of maize, tea and pyrethrum, in particular, in the pre- viously lightly inhabited areas such as Southern Highlands, diversifying away from areas of high concentration, notably the regions of Arusha, Kilimanjaro and Bukoba. 3.20 Food production was supported under the aegis of a number of proj- ects--the productive components of the rural development projects were almost entirely targetted at food production providing assistance for maize, beans and groundnuts; the cotton project included parallel efforts for maize, while a freestanding project in support of maize production was approved in 1976. D. Performance of the Lending Program 3.21 Seventeen of the 23 agricultural investment projects supported by the Bank have been evaluated as of December 1987. At project completion, three had reestimated economic rates of return above 1OZ, of which one was in dispute. Twelve projects had negative rates of return. With the excep- tion of one project, all had rates of return at completion substantially below those estimated at appraisal (Graph V). The sustainability of almost all these projects is either unlikely or in doubt. U 33 3.22 Agri-ultural production was consistently overestimated at project appraisal. In terms of area, there were over-ambitious planting targets in tobacco and tea--targets for the latter being far in excess of those in neighboring Kenya; in terms of yields, there were over-optimistic assess- ments of increases in sugar, cotton, maize and pyrethrum. Existing data were not analyzed sufficiently: "official" figures of the livestock herd, growth and offtake rates as contained in the preparation report were ac- cepted without question by appraisal. Historical production trends were sometimes ignored--despite steadily declining cotton production, Kigoma Rural Development project expected cotton to double in the cropping mix. 3.23 Inappropriate or non-existent technical packages were also noted in many evaluations. In the case of dairy, although 98% of milk production is derived from the traditional sector, the project supported large-scale, parastatal farms following modern intensive dairy practices with non-indig- enous cattle and a high degree of mechanization. Similarly, the livestock projects supported large-scale public enterprises although cattle tradi- tionally are owned by nomadic tribes.35/ Appraisal both underestimated in- vestment costs of the meat processing component (US$6.0 million at apprais- al compared to US$22.8 million at completion) and grossly overestimated the national herd. While in the Flue-Cured Tobacco project, labor requirements for tobacco production were completely overlooked at appraisal despite considerable research data being available.36/ In the Kilombero Sugar project, the project did not see the effect which the construction of a nearby dam would have on river flows. This then required the construction of a weir (an additional US$2.5 million) to restore pumping capability but the weir as constructed has precluded a road being built, thus cane from the original estate cannot be transported to the new factory on the opposite side of the river. In the Geita Cotton project, there was no appropriate technical package to extend to farmers. 311 This contrasts with the successful Zaire Ituri Livestock Development project, which was being prepared during the same period in neighboring Northeast Zaire, and which was aimed at small livestock owners. 36/ Research undertaken by Collinson, a Ministry of Agriculture employee at that time would have highlighted that the appraisal estimates of the area of tobacco which could be cultivated by family members were greatly overestimated, which in turn led to a serious overestimation of incremental production. Government policies at that time precluded hiring of labor. - 34 - 3.24 Intrasectoral linkage between processing and production was given little attention: additional processing facilities were supported in cashew and tobacco despite the fact that at appraisal of both projects. declining production trends were evident. Where the linkage was well co- ordinated at appraisal, as in the industrial forestry plantation project, performance was affected by subsequent delays and under-utilization of the related processing capacity. 3.25 Provision was made for management in almost all projects to re- ceive considerable support in the form of technical assistance, generally through internationally recruited staff and training. At completion, sever- al projects noted the limit to which expatriate project management can substitute for local managerial capacity. U3vernment was generally reluc- tant to hire expatriates and frequently there was overlapping technical assistance with that supplied by other donors37/ (fisheries, dairy, live- stock) or key positions were not filled (sugar) or recruitment delayed, (beef ranching, pyrethrum). On occasicn, the insistence by the Fank for international recruitment was resented by project officials (forestry). Training was often not very effective: at closing of the Kilombero Sugar project, the Bank Group's third such lending operation, eleven internation- ally recruited staff were still in place. In many projects, training should have received higher priority during supervision (forestry, cotton, Iwanza/Shinyanga, livestock, both tea projects), while additional training would have helped (tobacco). The Bank was unable to convince government of the potential usefulness of drawing on the experience of the private sec- tor; efforts in the tea sector--where productivity in the estates sector has been good--largely failed. 3.26 Other projects suffered design weaknesses in terms of institution- al shortcomings. Kigoma RIDEP as well as the follow-on projects were all too large and complex. This resulted in fundamental coordination and man- agement difficulties, further aggravated by the basic incompatibility of the projects with policy and organizational systems at the national and regional levels.381 In the Smallholder Tea project, the Bank expected that a newly created institution, the Tea Authority, would be able to implement and coordinate new tea plantings over four widely scattered areas: one in the extreme northwest of Tanzania close to the Uganda border, a second in the Usambara Mountains in northeast Tanzania, and two in the extreme south- west of Tanzania close to Malawi. In the Second Livestock project, the assumption that the introduction of parastatals would result in returns to producers being higher than the existing private system proved quite wrong. Many projects relied upon projer& units staffed by internationally re- cruited technicians which proved to be a mistake both in implementation and in terms of sustainability. 371/ There were 32 donors supporting the agric!ltural sector in Tanzania between 1970-84. 38/ Experience with these highly integrated projects contrasts with that of successful rural development projects in Burkina which include a range of sub-projects based on provision of voluntary labor, organizational simplicity and rapid realization of benefits. 35 - 3.27 Project performance was also deeply affected by the unfavorable sectoral environment. This was analyzed in depth in a combined audit of six agricultural projects which highlighted the effects of the macro- economic crisis of the late seventies and early eighties as well as the impact of rapid institutional change on project performance. As a corol- lary, agricultural pricing policies were similarly often not given enough attention at appraisal or in project implementation. Where this issue was taken up in detail at preparation and appraisal (Second Livestock) with the Bank arguing at length for increased producer prices for beef to obtain greater parity between domestic and international prices, government suc- cessfully resisted this proposal. Price policy also turned out to be a significant factor in output performance of cotton and tobacco but was only addressed by the Bank late in the day. Enforcement of pricing cove- nants, deemed critical to project success in Kilombero sugar, should have received higher priority; on the other hand, producer prices of pyrethrum and milk were raised annually over many years but had little impact on declining production trends, farmers switched to other more financially profitable crops. 3.28 On the positive side, Bank projects, particularly those appraised and implemented during the years when Tanzania was suffering acute shortage of foreign exchange, made a significant contribution to keeping crop pro- duction operating at a minimal level; this particularly in respect of the follow-on forestry and tea projects as well as for coffee and tea under the Export Rehabilitation credit. 3.29 The hurry to do follow-on projects and the Bank's failure to re- sist Tanzania's wish to make projects larger than warranted by its capacity to implement them suggest that there were within the Bank itself pressures to transfer resources to Tanzania. Although problems of the Beef Ranch Development project were already evident when the second and much larger (US$18.5 million) Livestock project was being appraised in 1972, the second project still went ahead.39/ Declining tobacco production had already been noted in the course of the ongoing Bank-supported project, yet the Bank quickly went ahead with both a Tobacco Processing and Tobacco Handling project. Many projects as presented to the Bank by Government were large and complex given human and physical resource constraints, e.g. Second Livestock, Geita Cotton and Kigoma RIDEP. Although some proposals were scaled back by the Bank, examples remain which indicate pressure to lend. In Geita Cotton, the Bank wished to focus intensively on 75 villages but Government insisted upon 250 which at completion was considered a mistake. In Smallholder Tea project, although the Bank at preparation considered 6,000 ha of new plantings to be the maximum attainable for new settlers, Government requested plantings of 12,000 ha which was then scaled back to 8,300 ha. The National Maize project as prepared was initially expected to cover the six most productive regions in Tanzania; Government wanted all 20 regions covered, the final project comprised 13 regions with a further two added in implementation. The Pyrethrum project was expanded at appraisal from one region to two. The Second Cashewnut project in 1978 financed 39/ In contrast, the Zaire Ituri Livestock Development project was similarly fully prepared in 1972 but the project was deferred for three years to permit necessary institutional strengthening. - 36 - three more cashew factories, at a total project cost of US$36.0 million, although five had already been supported during the first phase and produc- tion had been steadily declining since 1974. 3.30 Project preparation in many instances was both intensive in terms of number of missions and staffing and extensive in respect of period of preparation. The successful Sao Hill Forestry project benefited from care- ful project preparation, as did the Smallholder Tea project. The need for greater Bank qualitative input into the preparation process can be seen in both the Second Livestock and the Geita Cotton projects. The former was prepared largely by government over a two-year period with 80 days of assistance from the Regional Office in Nairobi. Despite a strong appraisal mission, the failure of this large and complex project (US$24.7 million at appraisal, US$49.1 million at completion) was in large part due to poor preparation. In the case of Geita, although discussions with government over project design and content were drawn out over a two-year period, the Bank's preparation mission spent 18 days in the field at the end of which it was suggested that a strong pre-appraisal mission be sent in view of the serious shortcomings which had been identified. Despite this, a one-man two-day pre-appraisal mission visited the country prior to appraisal. In contrast, the preparation mission for the COte d'Ivoire Cotton Area Rural Development project (see box) spent a total of 200 staff days and only a few outstanding issues remained for the appraisal mission to resolve. There have also been instances where even intensive preparation by Bank staff have not saved the project. Thus the Beef Ranch Development project was likewise prepared over a long period (the first such Bank-supported livestock project in East Africa) but project performance was disap- pointing. Kigoma Rural Development received intensive preparation: the subject of eight preparation missions from RMEA and a further five from headquarters yet this, and subsequent rural development projects, were all failures. Preparation of Kigoma RIDEP missed important aspects of the region's farming systems, neglected the agricultural potential of the high- lands and lake shore areas, and was not able to determine that villagiza- tion was being coerced in the hurry to get the project approved. 3.31 The causality between rapid project preparation and poor results is strong. The government approached the Bank in May 1973 for support for the Kilombero Sugar project and pre-appraisal took place the following month in June. The Tobacco Processing and the National Maize projects were both prepared and appraised simultaneously despite government resistance to this approach.40/ The first cashew project was prepared and appraised in eight months. All projects had low or negative economic rates of return. 3.32 There was often little involvement of Tanzanian staff in project preparation and appraisal. In the National Maize project, some regions felt they had not been fully consulted during preparation and thus were reluctant to cooperate during implementation.41/ In the Mwanza/Shinyanga RIDEP, Tanzanian technical staff argued that the forestry component should be designed to suit local conditions rather than "stereotype objectives" 40/ Project files, Dewar memo dated May 27, 1975. 41/ Project files, MOA progress report dated November 18, 1975. - 37 - but was overruled by appraisal.42/ The credit agency was frequently brought into projects without previous consultation, most notably in the National Maize project and Kigoma RIDEP, while in the Geita Cotton project the Ministry of Education had no knowledge of the education component until well after both project preparation and appraisal.43/ There was a delay of one year in the start up of Mwanza/Shinyanga project while government thought that a project unit was in charge of implementation. 3.33 Warnings by Bank technical staff at project design stage were frequently not taken into account. Appraisal was undertaken by head- quarters staff, either from the agricultural projects division or the rural development division. Projects were then reviewed by Central Projects staff and by the Regional Office in Nairobi which was responsible for proj- ect preparation and later for supervision. Central Projects Staff raised doubts regarding tobacco production forecasts at appraisal of the Tobacco Processing project44/ and counselled against supporting untested ujamaa villages in the Fisheries Development project.45/ RMEA warned that the Second Livestock project was far too ambitious to be implemented by an institution experiencing organizational and financial weaknesses. It cautioned against expanding the Maize project to unproductive regions and highlighted that the independent, entrepreneurial farmers who characterized potential beneficiaries in Geita Cotton were unlikely to cultivate cotton on a block basis. RMEA also, in the same project, warned against both the economics of, and the technical ability to administer, the proposed tractor use. Warnings were also raised that Kigoma was too large and too complex to be administered efficiently,4/ little reliable data was available for this region which was classified as one of the most underdeveloped in Tanzania.47/ RMEA warned against proceeding with the continued investment in all three meat processing plants when the substantially reduced herd numbers became available in the Second Livestock project; its suggestion that only one plant should be renovated was overruled.48/ The 1974 agricultural sector report counselled against investment in marginal areas, stressing the need to concentrate on productive activities. Despite this, the National Agricultural Development program originally expected to cover 10 projects, and which was predicated on this basis, was dropped two years later in favor of three more rural development projects, all the responsibility of the separate rural development division. The agriculture projects division argued, over a period of years, for structural reform in 42/ Memo to Director of Forestry dated July 4, 1977, mission sources. 43/ Project files, Byron, Santos and Scearce memo to files June 29, 1973. 44/ Project files, Pickering memo dated May 21, 1976. 45/ PPAR dated June 18, 1984. 46/ Project Files, Nekby memo dated April 30, 1974. 47/ Project Files, Denness, Reader memo dated June 19, 1973, Ahmed's BTO to Loh dated November 21, 1973 and Loh's memo to files dated January 13, 1974. 48/ PPAR dated December 27, 1982, para. 7.06. - 38 - the agricultural sector to be central to any additional lending to Tanzania but was overruled by programs staff who determined conditionality in the last analysis.491 3.34 Warnings were also being issued by cofinanciers. USAID expressed concern regarding the ability of 13 regions and four Ministries to coordi- nate implementation of the National Maize project. USAID also argued that the administrative and implementation capacity at the regional level would need to be substantially improved; this was addressed by the Bank increas- ing the number of internationally recruited posts in the Project Servicing Unit. Similarly CDC expressed concern in the same project regarding the government's ability to run the project effectively given the inefficien- cies created by decentralization. In the Second Livestock project, SIDA withdrew as potential cofinancier following the revised project design which supported large parastatals rather than small farmers, as had been expected earlier. SIDA was also unhappy with the fact that there was not enough training for local personnel, the inattention to long term grazing potential of project areas, and to the failure of the Bank to get agreement with government on increasing producer prices. 3.35 Bank supervision was intensive in many cases, and almost always deemed helpful by the borrower. Thus the technical assistance supplied by supervision was often particularly useful in respect of procurement (fores- try), in convincing local authorities of the potential damage of collective production (tobacco), generally in the resolution of problems (sugar). In only a few cases was it noted that the Bank could have followed up more in supervision (cotton). Greater use could have been made of agriculturalists in supervision of the agro-processing projects. Similarly the flexibility which enabled project redesign during implementation proved beneficial in more recent projects (the Second Cashew project was changed to the Cashewnut Production Improvement Pilot project with much greater focus on production while the Tea Consolidation project benefitted from greater attention to technical assistance). Mid-term technical reviews undertaken for the Tobacco Handling, the Second Livestock, National Maize, Geita Cotton and Kigoma Rural Development projects were not sufficient; all proj- ects remained problem projects. In some instances it was noted that the Bank could have acted more as a team with Tanzanian project staff. In general, however, project offic3als also felt that the supervision process provided a structured framework which facilitated decision making w.th other government agencies; continuity of staff and an appropriate skills mix was also commended by the borrower. E. Results of the Lending Program 3.36 In view of this record of project performance, it is not surpris- ing that project results in most cases failed to reach appraisal expecta- tions. 49/ PPAR, Export Rehabilitation Credit, dated June 30, 1988. 3.37 Despite Bank intervention into almost all aspects of agricultural production, production of all major cash crops between 1970-86 declined or stabilized at low levels (Graph I), with the exception of tea which main- tained a modest increase over the period, largely due to increased produc- tivity in the estates sector. Smallholders have moved out of cash crops and into subsistence agriculture. The increase in maize production is due to area, not yield, increase. Although yield data are questionable, there appears to have been no intensification of production in the smallholder sector. There has, in the past two years, been a restoration of area cul- tivated to cotton to the level cultivated in 1966. Crop diversification similarly has not taken place (Chart I), coffee has consistently remained the most important crop in value terms. Processing capacity supported by the Bank is being used at a low level or not at all: none of the eight cashew factories financed by the Bank are in operation, cashew is now being exported unprocessed; the tobacco factory is similarly not operating; the tea factories are running at half capacity; the new sugar factory has only ever reached two thirds capacity and the two new high-cost meat plants geared to exports, have low throughput, export sales are negligible. The industrial plantations for forestry have been successful but the related processing capacity is inadequate thus there is excess supply of fuelwood and saw logs. 3.38 One principal objective of technical assistance is to strengthen local capability; the results in Tanzania of institution building through the auspices of these projects have not been good. For much of the period under review, parastatal enterprises and regional authorities were the principal vehicles of rural development. Although the Bank provided direct support to the regional authorities, implementation capacity remained weak at project completion. Crop parastatals have also received substantial Bank support50/ either through a single project (dairy, fisheries, pyrethrum) or through a series of projects (cotton, cashew, sugar, maize milling, livestock, tobacco and tea) and a substantial component of freestanding technical assistance in the Technical Assistance III project, but all continually suffer from a wide range of problems including: general proliferation of staff; increased commercial bank borrowings; weak management despite the training input; overoptimism regarding implementation capacity; an inability to respond to rapidly changing economic conditions, e.g., falling international commodity prices; weak accountability; and an incentive structure which did not prove conducive to (i) reducing costs; (ii) assuring cost recovery from farmers; and (iii) resisting political interference. There is no evidence to suggest that, until 1980, the Bank questioned the use of such parastatals. This again contrasts with experience elsewhere in Africa.51/ 50/ Lele estimates this to have been almost ITS$261.0 million, Lele and Meyers, Part II, op. cit. A1/ In the Eastern Senegal Livestock project the preparation team proposed a new parastatal. The Bank proposed--and Government agreed, that an existing and efficient agency should assume responsibility instead. 3.39 Substantial efforts have been made by the Bank and Government to improve performance and accountability, including restoration of the coop- eratives and transfer of remaining responsibilities to new marketing boards. A recent government report notes that there was a slight improve- ment in performance of the eight major crop marketing boards for their financial year ending 1987. Four recorded a profit for 1987 compared to only two in 1986; their overall losses declining slightly from Tshs 1,600 million to Tshs 1,500 million. However the losses incurred by the National Milling Company increased from Tshs 700 million in 1986 to Tshs 1,840 mil- lion in 1987 while the loss incurred by the Cotton Marketing Board similar- ly increased from Tshs 845 million to Tshs 1,835 million over the same period. 3.40 The results in respect of Bank processes illustrate that a Bank policy which sought to fill the increasing resource gap after the 1973/74 crisis by providing more projects created pressure within the Bank to transfer resources as quickly as possible. This then resulted in too many large, complex projects with poor technical foundations being approved and created a momentum for follow-on projects regardless of agricultural pro- duction trends. The rapid build-up in the project pip-aline with investment in almost every subsector has been at the expense of a more measured coun- try strategy with operational integration. The attempt to use projects to influence Government policies of general applicability, rather than seeking to influence these policies at the national level through policy dialogue clearly doomed many projects to failure. IV. LESSONS FOR THE FUTURE 4.01 The continual uncertainty and frequent disagreements over the reliability of data on agricultural production, which in turn adversely affected the estimates of economic rates of return, emphasizes the impor- tance of reliable data on the agricultural sector. Good information exists at the village level.52/ Additional opportunities exist for data capture within the public sector, as well as for the coordination and cataloguing of information already in the Bank and other donor agencies, e.g. socio- economic surveys as undertaken by OED in the course of impact evaluations. 4.02 There is clear need for greater attention to be paid to intrasec- toral linkage between agricultural processing and production as well as intersectoral linkage, notably for transportation and water, since the general state of underdevelopment and the deterioration in infrastructure do not allow the assumption that the additional demand on them arising from a particular project will be automatically met. 4.03 Lack of technical packages has now long been recognized and is under study but greater thought could be given to strengthening relations between existing research stations in the region, drawing lessons from the very effective research network across francophone countries or exploring links with the estates sector. 52/ (Research work in the Bank has confirmed the validity of farmer estimates of yields.) 4.04 The poor record in technical assistance, despite substantial Bank investment, suggests that project design should be more sensitive to the scarcity of trained manpower particularly in the use of enclave projects which has implications for disbursement and implementation schedules as well as for supervision. Expatriate inputs are only effective where their need is clearly recognized by government and project personnel. Experience also underlines the importance of donor coordination for technical assis- tance activities. The continuing problems with implementation capacity point to the need for technical assistance, albeit using a different ap- proach. Development of a manpower strategy at the sectoral level, first proposed by the Bank in 1970, is overdue. 4.05 The weak record of institution building suggests several lessons. In general, the Bank has often had greater success when it has provided support to institutions already established, often with prior assistance from agencies experienced in institution building. OED experience points to successful Bank reinforcement of FELDA in Malaysia, KTDA in Kenya and various cotton companies in West Africa. Institutional stability can be an important factor in project success; it should not be lightly assumed that they will develop and that their capability will not suffer ups and downs. Experience from Tanzania suggests: (i) the need to establish the basic institutional framework for project implementation prior to project approv- al and not left to be developed during implementation; (ii) project design should not assume that existing parastatals cti easily deal with large increases in work load; (iii) management perforr,nce depends on the proper matching of tasks expected of an institution a-d its capacity; in many of the Bank-supported projects involving parasts 41s, while "management" was often identified as the principal risk in jtaff appraisal reports, few offered much in the way of remedy; (iv) the structure of incentives includ- ing overall and relative pay scales should be adequately analyzed and ac- countability mechanisms be put in place; (v) the potential for political interference assessed; (vi) institution building should be central to the project. Finally, where rapid institutional change follows project effec- tiveness there is need for a complete review of appraisal assumptions. 4.06 Experience in Tanzania shows the need to ensure that the overall project design is appropriate to project objectives. This calls for suffi- cient integration in tree crop projects, for example, which, to produce a quality product are dependent on prompt and sophisticated processing. Projects for annual crops, which also have a close association with proces- sing and international trade, similarly require close coordination to as- sure a stable relationship between the farmer and the factory. On the other hand, large and institutionally complex area development projects whose small productive components were aimed only at increased food produc- tion would have benefitted from a simpler design. 4.07 The effects of poor policy environment on project outcome suggests that project appraisal should take account of likely effects of the general economic milieu. Specifically, individual project design should be based upon a realistic rather than desirable picture of the environment and take account of likely trends in supply constraints, e.g., foreign exchange, fuel or manpower; this in turn implies sector work geared to analysis of the impact of sector policies on project design and performance. 4.08 The effects of agricultural pricing policies suggest the need to treat price policies, necessary for project success, explicitly at apprais- Al. Low producer prices have been one of a range of constraints which explain declining agricultural production, and efforts need to be made to improve the weak data base at the micro level regarding supply response. 4.09 There are several lessons for Bank processes, both in respect of the policy dialogue and the quality of the lending program. First, policy dialogue to be effective must be a mutual interchange of ideas. Government has lead responsibility for policy formulation the country it governs. What falls to the Bank, however, is good analysis of the likely and actual impact of such policies, together with implications for Bank-supported projects. Such analysis assumes even greater importance when the Bank is actively supporting "experimental" policies. This, when based on (more) reliable data on agricultural production trends, would then form the basis for meaningful dialogue with government and agreement on needed policy changes. Subsidiary measures to strengthen the process could include time- ly delivery of what has been considered to be good Bank sector work in a way which can be more readily accepted by government and internalized into its own policy making process. This assumes good understanding on the part of the Bank of who the actual decision makers are and not necessarily those drawn from the ranks of the technical leadership. In this regard it may also require that the Bank act more as a catalyst in stimulating government to prepare its own sector work rather than taking on the responsibility itself. 4.10 Second, the quality of the lending program would be significantly enhanced if there were a mechanism to internalize better the warnings raised by Bank technical staff. Senior Bank management and the Board re- lied on statements of intent from the country's highest political leader- ship, while ignoring warnings from the Bank's own technical staff. This suggests the need for a more effective system of checks and balances within the Bank to enable all views to be adequately represented to Bank manage- ment, and does not belie the need for better coordination within the Bank, the lack of which has also been noted; but to argue the case for the tech- nical strength of the Bank to be fully involved in final investment deci- sion making, which could result in greater transparency of the management decision process and its accountability. 4.11 Experience in Tanzanian agriculture also illustrates that the Bank's lending program has swung according to prevailing Bank doctrine. The pendulum effect, which is clearly evident over this twenty-year period in Tanzania, with rather abrupt swings between sectors (infrastructure to agriculture); within sectors (growth to basic needs; rural development to market forces/efficiency)131 and between instruments employed (project to non-project lending) has tended to punctuate rather than smooth the devel- opment process. Given the leadership role of the Bank among other donors, the pendulum often swings wider as others follow the Bank's lead. A more 53/ A more extreme example of this effect can be found in West Africa where the Bank has considered dismantling and privatizing successful parastatals rather than seeking the reasons for their success and how the development strategy could be replicated elsewhere. prudent approach may be to learn the lessons of experience since there will continue to be a need for future Bank investment support within the frame- work of an overall country strategy. In learning lessons of experience, the Bank could encourage greater contact among its own staff working in different countries of Africa, more particularly between anglophonelfranco- phone Africa. Successful strategies exist which could be replicated54/ (see Box). 4.12 The lending program reflects pressure within the Bank to transfer resources to Tanzania. This suggests the benefits of a more cautious and phased build-up in the lending program; one not driven by the need either to fill the savings/investment gap or to invest in every subsector. Links with Bank sector work should be more closely drawn; the lending program should be an integral part of a previously established country strategy which identifies priorities for investment and program support and seeks to ensure continuous funding over an adequate time period. 541/ Caution has to be exercised in replication of ideas. While there appears to have been little replication of these successful projects in anglophone countries, the Bank's experience with large scale livestock ranching in Burkina, Kenya, Tanzania and Malawi highlights the dangers of widespread replication of what turned out to be a seriously inappropriate technical package. ANNEX AGRICULTURE PROJECTSPROGRAMS IN TANZANIA Agreement Loan/ Amount Closina Date Cancelled ERR Date Credit Prolect Lent Appraisal Actual millions) Appraisal Actual 01/13/66 C. 680- Agricultural Credit 6.03 66/89/78 06/89/76 - * - 10/31/68 C. 182-4 Beef Ranch Development 1.61 12/39/72 0113/74 - 11B 21 10/09/78 C.0217 Flue-Cured Tobacco 8.07 09/36/76 12/80/77 - 803 14X * 63/63/72 C.6287 Seallholder Tea 10.84 12/81/76 12/81/79 - 17X 12X 65/28/73 C.882-1 Second Livestock 16.50 12/81/79 12/81/88 - 853 neg. 01/17/74 C.0454 Ceta Cotton 17.59 12/81/82 12/81/82 0.8 14X neg. 068/26/74 L.1614 Cashownut Development 7.25 12/31/85 12/31/81 - 89 neg. 01/21/74 C.650 Klgoma RIDEP 10.60 04/89/81 98/81/83 1.2 18% neg. 09/27/74) C.0518 Kilombero Sugar 9.66 12/81/79 12/81/86 - 125 4 09/27/74) L.1041 Kilombero Sugar 8.06 12/81/79 12/81/89 - 123 4X 08/15/75 C.6680 Dairy Development 10.66 64/86/81 68/81/82 1.2 18 neg. 01/29/78 C.66OM National Maize 18.66 60/66/80 8/81/88 8.5 87X nag. 07/12/76 C.652 Fisheries Development 9.66 12/31/81 69/18/88 4.8 27X neg. 07/12/76 L.1807 Sao Hill Forestry 8.85 66/80/82 68/890/88 - 183 21X * 60/18/76 C.6958 Tobacco Processing 8.66 66/68/81 12/81/81 6.5 253 neg. 95/11/77 C.0708 Tabora RIDEP 7.25 068/80/88 6/80/88 2.1 173 neg. 06/14/78 C.0802 Tobacco Handling 14.66 64/80/88 84/89/88 7.9 243 neg. 6/14/78 C.0a01 Second Cashownut Dev. 27.50 12/81/84 - * 163 - 66/14/78 C.080a Mweas/Shin. RIDEP 12.66 12/81/84 12/81/84 2.9 213 neg. 08/26/80 C.1087 Smallholder Tea Consol. 14.66 12/81/76 - 21Z - 16124/86 S.1670 Coconut Pilot 6.80 08/81/86 - None - 10/24/8 C.16W7 Pyrethrum 10.66 12/81/85 12/81/86 5.1 173 neg. 04/16/81 C.1015 Grain Storage A Mill. 48.66 06/80/88 - 881 - 64/24/81 S.1188 Export Rehab. Program 560.60 8/81/82 08/81/88 - None Failed 05/19/82 S.1229 Sao Hill Forestry II. 12.66 12/81/67 - 80 - * Disputed. TANZANIA WORLD BANK/TANZANIA RELATIONS, 1961-1987 TANZANIA'S INDUSTRIALIZATION EFFORT 1961-1987 I. INTRODUCTION 1.01 This review of the Bank/Tanzania relationship examines the coun- try's industrial development during the period 1961-87, based on available documents, statistical evidence.1 the literature and extensive interviews. The aim is to gain insights concerning the evolution of the country's in- dustrial structure, analyze in some detail the key dimensions of perfor- mance of the sector during the period under review, and reflect on the Bank's posture and approach toward Tanzania's industrialization effort. Three periods are distinguished, which display distinct characteristics and reflect major shifts in objectives, strategies and policies: the immediate post-independence period (1961-66), broadly characterized by continuation of the status quo ante; the shift to socialism and self-reliance (1967-late 1970s); and a period of unchanged political, economic and social fundamen- tals but alluding to an increased realization of past policy flaws and the need for retrenchment and policy reorientation (late 1970s to date). This distinction puts in proper perspective the stated government objectives and the strategies, policies and instruments employed for their achievement; helps to evaluate the effectiveness of the policy framework and institu- tional arrangements; provides insights on the array of factors that have impacted on the results attained; and facilitates the assessment of the respective role of the Bank, donors and governmen. entities in the coun- try's industrialization effort. 1.02 With the benefit of hindsight, an attempt is made to reconstruct and assess the Bank's perception of the appropriateness of the strategies and policies adopted by the government to achieve its sectoral objectives, the extent of perceived achievements and failures, and the critical factors that have affected industrial performance. The review examines critically the Bank's stance and approach to important issues; the appropriateness, quality and timeliness of advice and assistance; and the overall contribu- tion, including errors of omission and commission. An assessment is made of the quality of the Bank's industrial sector work, as reflected in the depth of sectoral knowledge in its various dimensions; the entire direct and indirect (through development finance institutions and technical assis- tance) lending program to industry, including its consistency with sectoral 1/ Statistical information for the sector is generally far from comprehensive and not always reliable; nonetheless, while absolute figures should be interpreted with caution, trends are clearly distinguishable. The base year for constant prices referred to in the text is 1966. priorities; rroject selection procedures; and project design, implementa- tion, performance, sustainability, and impact on policy reforms. Particu- lar emphasis is placed on the Bank's responsiveness to changing needs and circumstances, the adequacy of the analytical underpinning for conducting a policy dialogue, policy conditionality, understanding of and sensitivity to the country's political and administrative processes, and the Bank's pos- ture in the face of the government's receptivity. The review comes up with an overall assessment of the outcome of the industrialization effort, the quality of the Bank's advice in light of the configuration of Tanzania's economic system and the espoused goals of the country's political leader- ship, and the strengths and weaknesses of the total interactive relation- ship between the Bank and the responsible authorities over the years. Finally, the lessons of experience are drawn and recommendations are made, which may help reconsider certain aspects of the Bank's decision-making process and approach to sectoral issues. 1.03 Sections VIII. The Aftermath--Present Status of the Industrial Sector and Prospects and IX. Assessment of the Bank-Country Relationship and Dialogue provide a synopsis of the main findings and conclusions. II. STRATEGY AND POLICY FRAMEWORK Immediate Post-Independence Period (1961-66) 2.01 Right after independence, the government and the Bank concurred on a development strategy aiming at increasing the rate of economic growth, exploiting in the process the most obvious opportunities for industrial investment .largely simple consumer goods and building materials). It was recognized that, at Tanzania's stage of development,2 the most important single condition for further industrial expansion was the development of agriculture, which accounted for 59% of GDP and 80% of exports. Agricul- tural growth would expand the domestic market for manufactures by raising farmer incomes, supply processing materials, and provide the requisite foreign exchange for essential imports. It was felt that there was neither 2/ The immediate post-colonial period was characterized by a fairly open economy and reliance on private (largely foreign) initiative and market forces. The industrial structure was rudimentary. It concentrated on first-stage processing of commodity exports, included a few medium-scale enterprises producing consumer goods for the domestic market, and was heavily dependent on non-indigenous enterp:ise and capital both resident and foreign. need for, nor the possibility of, rapid absorption of a large volume of unemployed or underemployed labor through industrial expansion and, under these circumstances, a policy of "forced" industrialization in Tanzania would not accord with economic realities. 2.02 Furthermore, both the Bank and th? A.D. Little studies of 19613 pointed to potent constraints to industrialization, which would be rein- forced if the country's prevailing politico-economic philosophy and aspira- tions were carried out injudiciously (e.g., rapid growth, speedy indigeni- zation, higher wages, misgivings regarding the effectiveness of the free market system).4 Such obstacles included the small size of the domestic market, for most products being far below requirements of minimum economic plant size; inability to process local raw materials for export or. a grand scale in the short run; absence of indigenous entrepreneurship, managerial talent and technical skills; serious lack of trained and disciplined labor force, while developing these skills by necessity would be a gradual and drawn-o.t process; lack of indigenous financial resources, resulting in dependence on local minority groups (Asian and European) and foreign inves- tors; and inadequate infrastructure. 2.03 The First Five-Year Plan (1964-69) stressed the need for a "more balanced" economic structure to correct the "structural disequilibrium," reflected in the production of agricultural raw materials for export and the importation of manufactures. The government felt that simply to expand agricultural output would be to condemn the country "to a position of per- manent economic inferiority in the world", and would perpetuate its reli- ance on the fluctuating income from the export of primary commodities. Besides, the expansion of commercial production of agricultural commodities was viewed as being limited by "sociological" factors and by the level of foreign demand and outlets. To rectify the situation, industrialization 3/ World Bank, Economic Development of Tanganvika (Baltimore: Johns Hopkins Press, 1961), pp. 229-248, 495-499, 515-521; Arthur D. Little, Inc., Tanganyika Industrial Development (Dar es Salaam, 1961), pp. 1-41, 87-117. 4/ The government had expresseA its intention for continued support of private enterprise. "Direct state operation" was to be confined to projects of great importance to the welfare of the country and to cases where private investors could not be attracted. How- er, "the people of Tanganyika will expect rapid progress in industrial eevelopment and may press for other approaches if development through private enterprise does not show evidence of early progress." A.D. Little, 22. cit., p. 26. This statement is suggestive of the incipient impatience and ambivalent attitude vis-a-vis the role of private initiative, stemming from the lack of consensus within the political leadership and strong socialist undercurrents. - 48 - would be accelerated. Industry would grow more than twice as fast as agri- culture, producing the consumer goods needed for the domestic market, and paying wage levels that would enable consumers to buy more goods and there- by expand market demand. 2.04 The 1964-69 Plan in effect broadened the scope of import substitu- tion to include intermediates (e.g., cement, steel and aluminum rolling, tires, glass products, tanning, oil refining, truck assembly), with the market constraint to be eased by changing the rules of the East Africa Common Market (i.e. by restricting certain imports from partner states and reserving particular industries catering to the regional market to indivi- dual states). The establishment of a relatively large number of economi- cally viable consumer goods and primary processing industries was viewed as feasible, although without the benefit of adequate sectoral analysis. Despite its unpopularity, it was recognized that continued recruitment of expatriates would be necessary, while efforts would be made to instill confidence in the private sector and create an economic climate conducive to investment. Private enterprise was accorded a dominant role, and fi- nancing for the proposed ambitious program of industrial investment was to come from private, mostly foreign sources, while the process was to be fostered by guarantees for private investment and protection cum assistance to "infant industries' (negotiated tariffs on an ad hoc basis, accelerated depreciation, and financing through the state-owned Tanganyika Development Corporation). 2.05 The 1964-69 Plan unmistakingly alluded to a structural transforma- tion of the economy and reflected a shift away from the position taken in the 1961-64 Plan, where much less emphasis was placed on industrialization since industrial development was viewed as being heavily conditioned on the growth of agriculture, and where much of the capital required for develop- ment was to come from export earnings also dependent on agricultural cash crops. Furthermore, the Plan not only stressed industrial growth, but contemplated a change in industrial structure through the promotion of intermediate goods industries. The Plan did not outline a policy framework that would explicitly promote such desiderata as intra-sectoral linkages, external economies or choice of appropriate techniques in the course of the industrialization process; but the emphasis placed on intermediates and agricultural processing industries would have the positive, albeit inciden- tal, effect of advancing these goals to some degree--production (forward and backward) linkages in particular. On the other hand, considerations of comparative advantage, economic viability of productive facilities, and production cost efficiency did not receive due attention.5 5/ World Bank, Current Economic Position and Prospects of Kenya, Tanzania, and Uganda, Report No. AF-35, September 13, 1965, Annex II, para. 6; World Bank, Prospects for Economic Development in East Africa, Vol. III: Tanzania, Part Three: Annex B - Industry, Report. No. AF-58a, August 31, 1967, para. 50. World Bank, Tanzania: Basic Economic Report, Annex V - Industry: Perspective and Strategic Choices, Report No. 1616, December 1977, para. 1.5. - 49 - Socialism and Self-Reliance (1967-Late 1970s) 2.06 A major ideological shift to socialism and self-reliance took place in 1967. The political leadership perceived the policies it had adopted during 1961-66 as fostering the emergence of certain disconcerting trends in social and economic development. In the first place, there had been a growth of an indigenous urban elite and a widening of urban-rural income differentials following the substantial salary and wage increases in urban areas and the rapid increase in the absolute number of Africans with high and middle level jobs as they moved into positions previously held by expa- triates. Secondly, less emphasis had been placed on a broad-based rural development, as evidenced by the limited investment allocations in agricul- ture and the neglect of food crop production and livestock, i.e. tradition- al activities which bypassed the majority of peasants. Thirdly, there was discernment of the beginnings of class formation among peasants, and the prospect of capitalist farming leading to the formation of a large, dispos- sessed and exploited rural proletariat, stemming from the advantage taken of extension services and better husbandry practices by progressive far- mirs. Fourthly, it was noticed that an inappropriate system of education in terms of throughput, composition and content had evolved, producing graduates ill-prepared for productive work. Finally, heavy emphasis had been placed on securing financial resources, especially from external sources, to the neglect of efforts to mobilize Tanzania's underutilized resources of land and labor. Faced with these socially unwelcome trends, in February 1967, the political leadership carried out a radical reassess- ment of its fundamental development objectives and strategies, and opted for a socialist pattern of development which was enunciatc by President Nyerere in the Arusha Declaration.6 6/ World Bank, 1977, The Main Report, paras. 1-4, 1.1-1.23; J.F. Rweyemamu, Underdevelopment and Industrialization in Tanzania, pp. 78-110, 138-172. Though socialism had been part of the TANU creed since early in 1962 (J.K. Nyerere, Uiamaa - The Basis for African Socialism, 1962), it was the Arusha Declaration and the accompanying policy statements that gave tangible content to this objective (J.K. Nyerere, Uiamaa: Essays of Socialism, 1968). - 50 - 2.07 Following the resolution to pursue a socialist mode of development, state-led industrialization and import substitution strategy became the vehicles for promoting the structural transformation of the economy.7 The basic tenets of the new industrial strategy included: a broad-based and indigenously-propelled industrial development; structural change toward developing greater capability in producers' goods to promote linkages and externalities; employment creation; industrial self-reliance, in the sense of minimal reliance on world trade, external technical assistance, foreign investment and aid; more equal distribution of the development gains among regions; absence of exploitation and more equitable income distribution; and greater participation of workers in the management of enterprises.8 However, there was no consensus as to what constituted a socialist industrial strategy and about the appropriate content of industrial investment. There was concern about what products should be produced, cost reduction, inter-sectoral linkages--but no guiding doctrine or coherent pattern of policy regarding these aspects of industrial strategy had been worked out. A long-term strategy had yet to be developed.9 2.08 The distinguishing features of this strategy, which was articulated more fully and formalized in the mid-1970s as the Basic Industry Strategy (see paras. 2.11, 2.12 below), included production of heavy producer goods, limitation of consumer preferences, reduction of the share of international trade in production, and central direction of investment toward industries defined as "basic" without regard to comparative advantage and costs of production (e.g., the importance of scale economies was downplayed). In effect, considerations of economic efficiency became of secondary impor- tance. The overriding concern was to restructure the economy away from its dependence on trade and toward production for the domestic market, using 7/ The state would play a major and direct role in the country'secolomy, with a view to accelerating structural change, fostering the generation of externalities and mitigating the impact of growth on income differentials. To this end, the state would be heavily involved in the process of investment allocation, both through extensive ownership and through a price structure that would reflect nationally defined priorities and long-term objectives. 81 National self-reliance; enhancement of manpower, educational and research capacity; broad-based rural development; planned change of economic structures; ownership and control of economic activity; national mobilization to increase the effective resource capacity for development; equitable income distribution; and promotion of economic cooperation with other developing economies were considered as prerequisites for economic independence as well. R.H. Green, "Economic Independence and Economic Cooperation," in D.P. Ghai (Ed.), Economic Independence in Africa, (Dar es Salaam: East African Literature Bureau), 1973, pp. 46, 56-57. 9/ M. Roemer, G. Tidrick, D. Williams, "The Range of Strategy Choice in Tanzanian Industry," Journal of Development Economics, 3 (1976), p. 259; World Bank, 1977, Annex V, para. 1.13. 51 domestic resources. Basic needs rather than market demand would determine production priorities for consumer goods. Backward-linkage import substi- tution was clearly preferred over processing for export. Such objectives would be advanced through public investment and government ownership and control.10 2.09 Thus, in the aftermath of the Arusha Declaration, the manufacturing sector was systematically transformed from a largely private and foreign- owned into one where ownership and control were vested predominantly with parastatal companies. The larger industrial enterprises, wholesale trade, finance and insurance were nationalized.11 The state-owned National Development Corporation (NDC) became the most important instrument of in- dustrial development as a result of acquisitions of a controlling share in existing privately owned companies (operated as subsidiaries), as well as of new investments of its own. Public ownership of manufacturing enter- prises was accompanied by increasing reliance on a rexue of direct controls, which affected not only the investment decision-making process but also the operations of parastatals (e.g., through foreign exchange allocations, licensing and confinements, wage-setting, etc.).12 10/ World Bank, 1977, Annex V, paras. 1.29-1.35. il/ Full ownership and majority control enterprises that were taken over accounted for about 502 of manufacturing output, suggesting that a significant share of industrial production was still left in private hands. For a view of the positive role of the state sector, albeit with caveats, in reducing the degree of foreign control and in providing the basis for the development of indigenous economic institutions, see B. van Arcadie, "Development of the State Sector and Economic Independence," in D.P. Ghai (Ed.), Economic Independence in Africa, (Dar es Salaam: East African Literature Bureau), 1973, pp. 88-122. 12/ World Bank, 1971, Vol. III, paras. 39, 45; World Bank, Economic Memorandum on Tanzania, Report No. 1567, April 12, 1977, para. 45. Aithough the nationalization legislation stipulated that corporations shall conduct their business according to commercial principles, it gave little guidance on the economic and commercial goals of the nationalized undertakings, thus creating an uncertainty as to what these public entities were expected to achieve. A. Bradley, "The Nationalization of Companies in Tanzania," in P.A. Thomas (Ed.), Private Enterprise and the East African Company (Dar es Salaam: Tanzania Publishing House), 1969, pp. 207-288, especially 224-225. - 52 - 2.10 The Second Five-Year Plan (1969-74) aimed at a rapid growth of industry (13% per annum, or almost twice that of agriculture) based on import substitution. The bulk of the new industrial investment would be undertaken by NDC, which implied the establishment of new state-owned en- terprises.13 Rapid industrial growth, according to the planners, would foster the development of the rural economy, as the latter was dependent on the growth of urban markets composed of industrial wage earners who would buy the farm products. A fast growing pool of industrial workers would make it possible for the farmers to market an increased volume of their produce, while the growth of modern industry would make locally produced cheap manufactures available to the rural population.14 This marked a significant reversal of the early position, which held that industrial growth was dependent on the growth of agriculture. 2.11 Until the mid-1970s, the government did not have a well-articu- lated, long-term industrial strategy. The maximum growth strategy, which by default had become the official strategy, came under criticism as being difficult to reconcile with the espoused socialist objectives, as giving inadequate consideration to linkages and externalities, as basing invest- ment decisions on distorted market prices and an inegalitarian income dis- tribution, and as being biased against a major structural change because of its marginalist approach. These deficiencies were reinforced by the short term focus and the de facto decentralized decision-making process characteristic of the planning mechanism under the strategy. Furthermore, shortcomings in both the planning and the pricing systems effectively lim- ited the growth effects of the strategy. Investments were usually approved on the basis of the potential to produce and not the ability to produce efficiently.15 In 1974, after a two-year planning exercise in which several alternative long-term (1975-1995) industrial strategies were considered, the government opted for the Basic Industry Strategy (BIS) as the best suited for the pursuit of the goals of structural transformation and self-reliance. 13/ NDC rather than private foreign investment was seen as a more suitable vehicle for industrialization, in view of the purported limitations of the latter based on political, social and economic considerations, and the inexpediency of awaiting the slow emergence of a full-fledged local entrepreneurial class capable of undertaking the range of industrial investments required. E.g., see B. van Arcadie, "Private Foreign Investment: Some Limitations" in P.A. Thomas (Ed.), Private Enterprise and the East African Company, Dar es Salaam: Tanzania Publishing House), 1969, pp. 156-173. 14/ Government of Tanzania, The People's Plan for Progress (1969-74), Dar es Salaam, May 1969, p. 42. 15/ World Bank, 1977, Annex V, paras. 1.19-1.22, 1.29-1.34. 4.19, 6.4. - 53 - 2.12 In its pure version, BIS was perceived as giving top priority to industries supplying "basic needs" and promoting producer goods which enter into the production of a wide range of industrial products. Emphasis was placed on using local resources and on producing for the local market to foster forward and backward linkages, and thereby promote the structural transformation of the economy. Needs would be defined more by central planners and less by market demand. The import-substitution and anti-trade biases were welcomed as a way of reducing dependence on external markets in the long run (the self-reliance goal).16 In practice, however, the adopted version of the BIS in the 1976-81 Plan incorporated important modifi- cations. Existing commitments for expanding certain non-BIS industries, especially export-oriented processing activities, were honored;17 efficiency criteria were set to guide the choice of techniques within 16/ World Bank, 1977, Main Report, para. 5.47; World Bank, Industrial and Mining Sector Survey, Report No. 647, 1975, Vol. I, paras. 105-106. For the theoretical underpinning of the BIS see Clive Y. Thomas, Dependence and Transformation: The Economics of the Transition to Socialism, (New York/London: Monthly Review Press), 1974, and the Appendix to this report, paras. 1-11. 17/ In 1974, the government endorsed an investment program influenced by the processing strategy, which also appealed to the Bank and to donors. The 1976 mission argued that the appeal stemmed from a presumption that Tanzania had a comparative advantage in processing (e.g., hides and skins, cashew nuts, textiles, sisal), a notion which, it felt, was questionable on the basis of the findings of studies conducted in the context of the long-term industrial strategy exercise. From an efficiency standpoint, the strategy was deemed deficient because it considered only benefits and not costs, while from the structural point of view increased processing for export meant increased external dependency. The mission, however, was quick to point out that the criticism of the processing strategy did not imply lack of export possibilities, but rather highlighted the need to study the prospects on a case by case basis using cost-benefit analysis. Nonetheless, the mission supported the planners' view which perceived exports as the outgrowth from efficient import-substituting industries achieving some export penetration rather than the conscious promotion of specialized, export-oriented industries. World Bank, 1977, Annex V. paras. 1.23-1.28, 6.5, 6.8. It is noteworthy that the efficiency ratios used in these exercises were based on crude domestic resource cost estimates, rough estimates of opportunity costs of domestic factors of production, incomplete knowledge of cost structures, historical levels of productivity, conjectural assumptions regarding prospective world price trends, while no allowance was made for secular trends in labor productivity and technological change over time. See M. Roemer, G.M. Tidrik, D. Williams, "The Range of Strategic Choice in Tanzanian Industry," Journal of Development Economics, 3 (1976), pp. 260-264. In short, conclusions were drawn from an inadequate data base, static efficiency measures, and disregard of dynamic comparative advantages, long-term externalities, and foreign exchange adjustments. - 54 - subsectors, which were to be governed not just by efficiency but also by the potential for local adaptability; minimum rate of return liLits were set for BIS investments, and attention was to be paid to the sequencing of investmentt to reap scale economies and raise rates of return; certain activities were reserved for the small-scale industry;18 and consumer preferences would play a role in determining product choice.19 2.13 The Third Five-Year Plan (1976-81) constituted the first phase of the long term Industrial Perspective Plan 1975-1995. Unlike the emphasis of previous Plans on developing import-substituting industries which de- pended heavily on imported inputs, the new strategy would promote indus- tries catering to the basic needs of the majority of the people (e.g., clothing, food, shoes, construction materials, transportation) using local raw materials, which would reduce dependence on external supplies and fos- ter forward and backward linkages. In satisfying these basic needs, a group of basic (or core) industries (e.g., iron and steel, metal working and engineering, chemicals, paper and wood, textiles, leather, construction materials) would be developed whose output would be used by most other industries.20 Thus, emphasis on self-reliance, on meeting basic needs and on restructuring the industrial sector coalesced in an industrial strategy supporting the establishment of basic industries. Workers would be fully involved in the management of their respective enterprises and measures (rewards and penalties) would be taken to increase production and efficien- cy. In parallel, high priority would be given to the development of eco- nomic (power, water) and social (technical education) infrastructure.21 18/ It was argued that support for rural-based, small scale industry (SSI) would be consistent with the Ujamaa policy, extend self-reliance to the local level, promote greater worker participation, develop the skills of a much larger number of workers because of the simpler and more labor-intensive technology employed, reduce urban-rural income inequalities, use mostly local resources, and economize on capital and foreign exchange. In practice, however, there has been ambivalence toward the promotion of SSI, while implementation of SSI schemes has been stymied by the policy of indigenization, lack of planning and technical assistance, limited production possibilities, consumer resistance, pre-emption of the market by parastatals, and the unwillingness of the government to override consumer preferences for goods produced in large scale factories. World Bank, 1977, Annex V. paras. 1.36-1.42. 19/ World Bank, 1977, Main Report, para. 5.48; Annex V, paras. 4.01-4.46, 6.7. 20/ For a contrast with the Bank's notion of basic needs see Appendix, para. 6. 21/ United Republic of Tanzania: Third Five Year Plan for Economic and Social Development (1976-81), pp. i, 1i, 5, 7, 43. Retrenchment and Policy Reorientation (Late 1970s to Date) 2.14 There has been no major change in political, social and economic fundamentals during this period; however, there has been increasing reali- zation of policy flaws. The decline of the economy which started in the second half of the 1970s continued with increased intensity during the 1980s and reached crisis proportions. The inadequate policies and funda- mental weaknesses in the management of the economy ultimately took their toll. As a result of the policy framework adopted in promoting the inward- looking -industrialization and self-reliance strategy, and reinforced by external factors, serious problems continued to afflict industry throughout the 1980s. In the face of the economic malaise and the dismal industrial performance, there have been signs of retrenchment and growing appreciation of the need for structural adjustment and reorientation, as evidenced by the adoption of Economic Survival Plans in 1980 and 1981 and the Structural Adjustment Program (SAP) in 1982. 2.15 The policy and institutional measures adopted under the SAP, how- ever, were insufficient in scope and intensity to stimulate any major re- covery in the economy. Capital and recurrent allocations to agriculture did increase, foreign exchange retention accounts for importers and an "own funded" imports scheme were introduced, and a few industrial parastatals were dissolved; but the measures did not go far enough. Industrial perfor- mance continued to be impeded by the deep-seated problems of the past and, as a result, the pattern of industrial development continued to be dis- torted.22 2.16 With Bank encouragement and financial support, the government pre- pared in mid-1986 a new Economic Recovery Program (ERP), taking the first initial steps to rectify the distortions created by earlier policies in an attempt to reverse the protracted deterioration of the economy. The areas addressed include exchange rate policy. the trade regime and foreign ex- change allocations, performance of parastatals, the agricultural marketing system, pricing policies, industrial restructuring, transport sector effi- ciency, and public expenditures. The government's objectives in the indus- trial sector aim at improving capacity utilization, rehabilitation of major industries, completion of on-going projects, and at ensuring that resources are directed toward the more productive and efficient firms in the sector. The ERP envisages a recovery period of five to seven years. However, the ERP did not propose much needed interventions to bring about reforms in the management and operation of parastatals.23 The Policy Framework Paper 1987-90, prepared jointly by the government and the staffs of the Bank and the IMF in October 1987, endorsed the thrust of ERP's policy, institutional reforms and structural adjustment measures. Recently proposed policy re- forms and action programs for the restructuring of the industrial sector are detailed in paras. 4.06 and 4.07 below. 22/ World Bank, President's Report No. P-4404, Tanzania: Multisector Rehabilitation Credit, November 3, 1986, paras. 6-15. 23/ Ibid., paras. 16-53. III. PERFORMANCE AND STRUCTURAL CHANGE Immediate Post-Independence Period (1961-66) 3.01 The industrial secter experienced tangible progress during the post-independence period of 1961-66, despite the difficult environment. Starting from a low base, real value added in manufacturing grew at 11-12% per annum doubling its share to 62 of GDP, while industrial investment grew at 18Z a year largely due to several lumpy investments (e.g., oil refinery, brewery, sugar and cement plants). During the same period, real GDP grew at 52 annually. The largest projects were foreign-owned or joint ventures, including some with state participation; but a good part of the growth was attributable to the establishment and/or expansion of many small-medium scale private industries.24 The number of establishments employing 10 or more workers rose from 220 in 1961 to 438 in 1966, or by a respectable 1002, of which 21% employed 100 or more employees. Industrial employment rose from about 22,000 in 1961 to 29,390 in 1966, increasing 62 annually.25 Industrial investment was financed by new foreign capital, mostly suppliers' credit rather than risk capital, funds from development corporations, reinvested earnings and depreciation charges, and local Asian capital. Although foreign investment spurred industrial growth, there was considerable net capital outflow, amounting to T Sh 220 million (US$30 million equivalent) during 1961-66,26 a trend which had started prior to independence as a result of the uncertainties and was reinforced by the political difficulties through the mid-1960s. However, not all of this capital was industrial in origin; the bulk followed the departure of alienated non-Africans or expatriates and had its origin in commercial or egricultural activities. Z.02 There was much variation among sub-sectors in terms of efficiency and cost structure. Oversized plants, supply difficulties, inadequate infrastructure, inefficient operation, and insufficient demand had resulted 24/ World Bank, 1967, Vol. III; World Bank, The Economic Development and Tanzania, Report No. AE-7, March 17, 1970. 251 World Bank, Industrial Development in East Africa: Progress, Policies, Problems and Prospects, Vol. III, Tanzania, Report No. AE-12, April 16, 1971, p. 4 and Appendix Tables 6-11; A. Seidman, Comparative Development Strategies in East Africa, Nairobi, 1972, pp. 111-114; R. Skarstein and S.M. Wangwe, Industrial Development in Tanzania: Some Critical Issues, (Upsalla: Scandinavian Institute of African Studies), 1986, pp. 2, 12; R. Skarstein, "Growth and Crisis in the Manufacturing Sector," in J. Boesen, K.J. Havnevik, J. Koponen and R. Odgaard (Eds.), Tanzania: Crisis and Struggle for Survival, (Upsalla: Scandinavian Institute of African Studies), 1986, p. 79. 26/ J. Rweyemamu, "The Political Economy of Foreign Investments in the Underdeveloped Countries," The African Review, 1, 1 (1971), p. 115. in relatively low levels of capacity utilization (about 60Z on average), while fast rising labor costs as a result of trade union pressures without commensurate increase in labor productivity27 and the slow learning process by new establishments tended to raise production costs. Thus, while cer- tain undertakings were competitive with duty-paying imports, many were operating with high costs of production which they could cover by charging high prices taking advantage of the protected market. 3.03 The fact that private investment surged despite the post-indepen- dence uncertainties seems to suggest investor confidence in the role of private initiative in the face of Tanzania's socio-economic realities, and a lingering hope that a modus vivendi would somehow be worked out. To be sure, there were frictions as a result of the government's policy of Afri- canization of management and key personnel and the natural reluctance of private companies to surrender their prerogative of selecting their own staff. But while the work permit system which had been introduced to con- trol the inflow of expatriates created at times difficulties in getting qualified managers and technical personnel, ultimately it did not act as a major deterrent to enterprise and investment, probably because it was ap- plied flexibly or had been eroded. Incentives to private investment, which included import licensing and quantitative restrictions, special investment allowances, rebates of import duties, guarantees for compensation in case of nationalization, and provision of land and other services, apparently have had a positive effect. They were probably reinforced by other, less tangible factors. Private investment in manufacturing was officially en- couraged and it was reasonably expected that, in case of an eventual take- over, investors would most likely be compensated. Some Asian entrepreneurs appeared willing to stay on and take their chances by investing in indus- try, particularly since trade was insecure and was threatened by the co- operative movement. In the wake of the rising nationalism and the uncer- tain future of the East Africa Common Market (EACM), established Asian industrialists in East Africa decided to invest in similar activities in Tanzania to spread their risks. The eagerness of former foreign suppliers to protect their stake in their former export market, or enter into the manufacture at the source of supply of raw materials to protect the supply 27/ Between 1961 and 1966, earnings of Tanzanian workers rose by over 100% while retail prices by less than 10%. Labor productivity of Tanzanian workers reportedly was about one-third of that of European workers and one half to three-fourths of that of Asian workers. To some extent, low labor productivity could be attributed to management's shortcomings, as some industrialists lacked experience in labor relations, training methods, organization of work flow, etc. and, as a result, they were not able to use more effectively the available labor. On the other hand, the trade unions opposed on ideological grounds incentive wages (e.g., piecework) as a means both to raise labor productivity and ensure labor a fair share of the benefits of higher productivity. World Bank, 1967, Vol. III, paras. 65, 66. 58 and avert possible moves by their competitors, were significant motiva.. tions of international firms. Finally, multinationals, trading companies and other immigrant minorities saw an opportunity to get a foothold in Tanzania in case the EACM would cease to exist.28 3.04 Investor confidence was probably further strengthened by the gov- ernment's pragmatic economic policies, including assurances as to th- role of the private sector, measures to safeguard foreign investment (Foreign Investment Protection Act of 1963) and to instill labor discipline (Trade Dispute Act of 1962),29 establishment of firmer control over the trade union movement, increasing appreciation of the need to restrain wage increases and limit their growth to increases in labor productivity, and the parallel establishment in 1962 of the Tanganyika Development Corporation to foster (minority) state participation in selected industries and of the TDFL to support private industrial activities through joint ventures or joint ownership with local investors. Finally, there was a visceral sentiment that official pronouncements such as "African Socialism" and "control of the economy by the indigenous people rather than by expatriates" would not lead to a comprehensive public ownership of industry. Rather, the perception was that there would be no dramatic departure from the familiar "mixed economy" model, where both private and public enterprise would play a role in the country's development effort.30 3.05 In general, judging by the results attained, the conditions pre- vailing in the mid-1960s appeared not to have been unfavorable for indus- trial development on a modest scale, with the private sector acting as the prime mover. Certainly, this does not mean that conditions for rapid in- dustrialization were favorable. And although legitimate questions can be raised regarding the structure of incentives and the afforded degree of protection, the wisdom of certain investments, and the low operating effi- ciency of particular firms--issues reminiscent of similar results in other parts of the world where inefficient import substitution policies were also 28/ J.F. Rweyemamu, "The Historical and Institutional Setting of Tanzanian Industry" in K.S. Kim, R.B. Mabele, M.J. Schultheis (Eds.), Papers on the Political Economy of Tanzania, (London: Heinemann), 1979, pp. 69-77, especially pp. 71-74; Idem, Underdevelopment and Industrialization in Tanzania, (Nairobi: Oxford University Press, 1973), pp. 122-130; A. Coulson, Tanzania: A Political Economy, (Oxford: Clarendon Press), 1982, pp. 169-175. L9/ The Act outlawed strikes and lock-outs unless a specified conciliation procedure had first been followed. 30/ By the end of 1966, half a dozen years after independence, the share of public sector manufacturing value added to GDP amounted to about 5Z. R. Skarstein and S.M. Wangwe, o2. cit., p. 19. practiced, there were no apparent economic imperatives justifying the wholesale nationalizations that ensued in 1967, nor were there "substantial political or class demands for socialist initiatives.o31 Socialism and Self-Reliance (1967-Late 1970s) 3.06 Investment in manufacturing remained strong during 1967-79 and started tapering off from 1980 on. Industrial investment rose from TSh 174 million in 1967 to TSh 962 million in 1979, though at an uneven pace, grow- ing at an impressive average annual rate of 15.32 in real terms. The high rate of investment was induced by the goal of rapid industrialization, and was facilitated by the substantial capital inflows at concessionary terms and the constantly declining real interest rates.32 The manufacturing sector absorbed a disproportionately large share of capital expenditure: 31/ C. Pratt, The Critical Iase in Tanzania. 1945-1968, (London: Cambridge University Press), 1976, p. 176. For an opposite view of an emerging "spectre of perverse growth," that is growth which undermined rather than enhanced the potential of the economy for long-term development, and a belief that reliance on the private enterprise system as the prime mover of social development would not lead to self- generating and self-sustaining growth, see J.F. Rweyemamu, Underdevelopment and Industrialization in Tanzania, pp. ix-xiii, 38-74, 92-7, 111-137. According to this view, dependency on foreign markets, entrepreneurship, capital and technology had led to the establishment of a productive structure that was biased against the capital goods industries, utilized relatively more capital-intensive techniques of production, had limited linkage effects, fostered lop-sided development in terms of industrial location and sectoral distribution of consumer goods, and helped set up uncompetitive oligopolistic structures. On the view of surfacing class struggles, see Issa G. Shivji, Class Struggles in Tanzania, (Dar es Salaam: Tanzania Publishing House), 1975, p. 9; and Idem., "Introduction: The Transformation of the State and Working People" in ''. Shivji (Ed.), The State and the Working People in Tanzania, (Dakar Senegal: CODESRIA), 1985, pp. 1-15. 32/ Real interest rates fell from 8.5Z in 1966 to 3.42 in 1979. B.J. Ndulu,"Investment, Output Growth and Capacity Utilization in an African Economy: The Case of Manufacturing Sector in Tanzania," Eastern Africa Economic Review, Vol. 2, No. 1, 1986, pp. 14-20, Table 4, p. 18. while the sector's contribution to GDP rose from 9.5Z to 12? between 1967 and 1979, its share in total investment increased from 15? to 37Z, with the more substantial increases occurring after 1975.33 Manufacturing investment grew from TSh 211 million in 1970 to Tah 849 million in 1979, or by 12.3Z annually. But while parastatal inveptment grew at 9?, private sector in- vestment increased at 212 a year, suggesting that private entrepreneurs proved to be remarkably resilient, despite raw material and import restric- tions, price and markcting controls, and parastatal competition, exploiting market opportunities and taking advantage of a weak administrative appara- tus. This heightened level of private investment activity is reflected in the rise of new enterprise formation. The number of establishments with 10 or more employees increased from 438 in 1966 to 524 in 1976, or by 20?, but more than doubled between 1976 and 1978 reaching 1,276 enterprises. How- ever, the trend reversed itself in subsequent years and, by 1981, the num- ber of establishments had declined by 55% to 706, as closures due to the difficult economic conditions affected particularly medium-scale (10-50 employees) firms.34 3.07 Parastatal investment continued at a high level from 1976 on and peaked in 1979, despite the decline in output and returns. Most of these capital expenditures involved foreign-financed, large scale projects that had long lead times to prepare and long gestation periods. Foreign assis- tance made it possible for the parastatal sector to continue to invest in new plant and machinery, although more and more existing capacity was 33/ Achieving and sustaining a high level of investment has been a major element in Tanzania's development, and it was explicitly supported by the Bank (paras. 4.02, 4.03). The ratio of gross investment to GDP increased from 15.5Z in 1966 to 26.4% in 1971 and averaged 212 annually until 1979. But although Tanzania managed to sustain a high investment ratio, this success was not matched by a similar performance in the mobilization of domestic savings. The domestic savings rate increased from 15? in 1966 to 18% in 1972, but declined subsequently to 10% in 1979, while the rate of foreign savings rose from 0.5% to 11%. World Bank, Economic Memorandum on Tanzania, Report No. 2086, January 23, 1981, Table 1.3, paras. 1.9-1.11. This suggests that while domestic savings financed 96% of total investment in 1966, their share had dropped to 492 in 1979 (as low as 362 in 1974 and 1975, and 31% in 1978), with the balance financed with external assistance. Put differently, while Tanzania was virtually self-reliant in 1966, it had become increasingly, and substantially, more dependent on external assistance by the end of the 1970s. 34/ C.E. Barker, P.V. Mitschke-Collande, M.R. Bhagavan, D.V. Wield, African Industrialization, Gower, 1986, p. 93; D.A. Phillips, "Choice of Technology and Industrial Transformation: The Case of the United Republic of Tanzania," Industry and Development, No. 5, UVIDO, 1980, p. 92; World Bank, Tanzania: An Agenda for Industrial Recovery, Report No. 6357, June 30, 1987, Vol. I, p. 13. standing idle. Donors tended to show preference for supporting new proj- ects rather than financing rehabilitation or recurrent costs in existing enterprises. Meanwhile, the parastatals continued to generate new project ideas for which support was canvassed among, and won from, donors .35 Al- though foreign-aided investment through the parastatals did not diverge from the government's broad industrial strategy, the project distribution has been unbalanced. There has been over-investment in some sectors, lit- tle interest in the development of linkage industries, and under-spending on supporting infrastructure (transport, water, power). There has been no clear po.licy as regards technology, and each parastatal remained free to 35/ In many instances, demand projections for projects under consideration were overly optimistic and, even if they were to be taken at face value, production on the envisaged scale would be uneconomic for many years to come. International firms--machinery and product manufactarers especially--entering into minority partnership with parastatals were all too ready to undertake projects which appeared financially beneficial but which were devoid of economic merit. The usual arrangement with such firms was to put up a small equity investment which was well covered by the profit earned on the sale of machinery and equipment to the project, obtain a management contract, collect royalties, and secure a protected market. The absence of appropriate project selection and appraisal methods, and the inability of the Ministry of Planning and the parent miaistries to subject all projects submitted for approval to the rc.i.site scrutiny due to staffing constraints, resulted in less than optimal investment decisions and excess capacity. Many of the established industries were too large for the market, capital-intensive and heavily import- dependent. World Bank, Report No. AE-12, Industrial Development in East Africa: Progress. Policies. Problems and Prospects, Vol. I: The East Africa Community, April 15, 1971, pares. 39, 124, 125; World Bank, Report No. AE-26, The Economic Development Prospects of Tanzania, 1972, Vol. It Main Report, May 22, 1972, paras. 7, 26, 77; World Bank, 1975, paras. 43, 64, 111; World Bank, 1977, Annex V, para. 6.13. choose its own techniques, usually based on narrow bureaucratic considera- tions--primarily access to finance and choice of products.36 In general, 36/ J. James, "Bureaucratic Engineering and Economic Men: Decision-Making for Technology in Tanzania's State-Owned Enterprises," in S. Lall and F. Stewart (Eds.), Theory and Reality in Development, (London: St. Martins Press), 1986, pp. 217-239; D.A. Phillips. op* cit, pp. 85-106. "If a parastatal has other sources of finance lined up (its own funds, a promise of TIB loan, prospective foreign assistance, or even a supplier's credit) the project proposal is almost never questioned--even when the proposed investment is of low strategic priority and/or has a low prospective rate of return." World Bank, 1977, Annex V, para. 5.48(i). "In practice, however, parastatals frequently deal directly with donors to initiate project proposals. The principal result of this is loss of effective central control over project selection. The Treasury is reluctant to turn down an offer of aid for a particular project--even if it is a bad project--for fear of offending the donor. Donors are reluctant to turn down a request for a particular project--even if it is a dubious project cooked up by a national equipment supplier and a Tanzaniat. parastatal--for fear of offend.ng the Government. Parastatals, on the other hand, have every incentive to try to interest donors directly in their projects because of the way the budget system works. Parastatals find it extremely difficult to secure finance if they ask for a large Government equity contribution, but they find approval almost automatic if they can show they have a commitment from a foreign donor. It is this interaction of foreign aid management and the budget system which is responsible for many of the worst industrial planning decisions in recent years." Ibid., para. 5.49. "The issue of aid coordination has become increasingly pressing, if national investment priorities are not to be subverted by donor preferences." Ibid., Main Report, para. 35. - 63 - there has been no concern for introducing appropriate technologies and for planned financing of recurrent local and foreign exchange costs once projects were completed.37 37/ World Bank, Tanzania: Country Economic Memorandum, Report, No. 5019, August 10, 1984, para. 2.38; S.M. Wangwe, "Industrialization and Resource Allocation in a Developing Country: The Case of Recent Experiences in Tanzania," World Development, June 1983, p. 490. A study of over 300 establishments in 10 major industries concluded that "despite the rhetoric, Tanzania's industrialization programme has, in general, promoted the establishment of enterprises using large-scale capital-intensive, often technically, and almost invariably economically, inefficient techniques. Its technological choice policy in industry has in most instances failed to promote the achievement of major national objectives, such as employment creation, economic self- reliance, decentralization of development, rapid growth of output, conservation of scarce development capital and efficient allocation of resources." F.C. Perkins, "Technology Choice, Industrialization and Development Experiences in Tanzania," Journal of Development Studies, January 1983, p. 231. On the choice of technology, see also D. Williams, "National Planning and the Choice of Technology: The Case of Textiles in Tanzania," in K.S. Kim, R.B. Mabele, M.J. Schultheis (Eds.), Papers on the Political Economy of Tanzania, 1979, pp. 41-46, especially pp. 43-44; Ian Parker, "Contradictions in the Transition to Socialism: The Case of the National Development Corporation," in B.U. Mwansasu and C. Pratt (Eds.), Towards Socialism in Tanzania, University of Toronto Press, 1979, pp. 53-60; J. James, "Choice of Technology in Public Enterprise: A Comparative Study of Manufac -ing Industry in Kenya and Tanzania," in Developing Coun.:ies, (Boulder/London: Westview Press), 1987, pp. 219-247; W.E. Clark, Socialist Development and Public Investment in Tanzania, Toronto, 1978, pp. 125, 173. The pattern of foreign technology transfer has evolved over the years as follows: 1961-66: through direct investment; 1967-74: through joint ventures (majority shareholding with parastatals); 1975 and on: through turn-key projects (by parastatals). M. Broden, From Transfer to Acquisition of Technology: A Study of the Industrialization Process in Tanzania, Linkoeping, Sweden, 1983, pp. 135-136. - 64 - 3.08 Capital-output ratios and ICORs have been rising sharply.38 ICORs increased from 3.6 in 1968, to 6.6 in 1973 and to 8.5 in 1979, thereafter turning negative as a result of the decline in real output, inter alia due to scarcity in foreign exchange for imports of raw materials and spares, inadequate and irregular supplies of local inputs, and infrastructural problems.39 In the private sector, the ICOR generally showed an upward trend in 1977-79, before turning negative in 1980, whereas in the public sector the ICOR turned and remained negative from 1977 on, largely because of the declining capacity utilization and the inefficient use of inputs. Nevertheless, the evidence suggests that since the mid-1960s manufacturing investment had become increasingly more capital-intensive40 and that capital productivity had declined. 3.09 Real GDP grew at 3.32 annually during 1967-79. It increased at 4% a year between 1967 and 1976, largely influenced by the growth in the ser- vices sector, but remained stagnant during 1976-1978. Real value added in manufacturing increased 6.8% annually in 1967-79, while its share in GDP rose from 9.52 to 12.0%. But while manufacturing value added grew at an annual rate of 9.2% annually during 1967-73, the rate of growth decelerated to 6% during 1973-79. In fact, value added peaked in 1978 and began de- clining precipitously thereafter through the 1980s. The share of the para- statal manufacturing enterprises in total manufacturing value added rose from 142 in 1967 (5% in 1966) to 312 in 1979, suggesting that the share of the private sector manufacturing enterprises declined from 862 to 69%, but 38/ The capital-output ratio rose from 2.6 in 1967 to 3.4 in 1979 suggesting inefficient use of capital. This is corroborated by evidence in specific sectors (e.g., textiles, cement, leather, shoes) where large investmrents failed to yield the projected increases in output. B.J. Ndulu, "Investment, Output Growth and Capacity Utilization in an African Economy: The Case of Manufacturing Sector in Tanzania," Eastern Africa Economic Review, Vol. 2, No. 1, 1986, pp. 14-30, Table 1, p. 16; S.M. Wangwe, "Industrialization and Resource Allocation in a Developing Country: The Case of Recent Experience in Tanzania," World Development, June 1983, pp. 483-492, p. 486. During the same period, the capital-labor ratio increased from 43.0 (TSh thousand per worker) to 49.3, suggesting that the average labor productivity should have increased. However, labor productivity in manufacturing remained stagnant from 1966 to 1977 (it was declining in parastatals) and declined precipitously henceforth, reflecting poor labor utilization. 39/ For details see S.M. Wangwe, "Factors Influencing Capacity Utilization in Tanzanian Manufacturing," International Labor Review, January- February 1977, pp. 65-77. A/ Lending by the financial intermediaries and the allocation of import and business licenses have also favored large scale, capital-intensive techniques. F.C. Perkins, loc. cit, p. 229. - 65 - still accounting for a major part. Also, industrial employment (establish- ments employing 10 workers or more) increased from 29,390 in 1966 to 86,500 in 1979, or 8.7% annually, while the share of the public sector in manufac- turing employment rose from 15.5Z in 1967 to 50.0% in 1979,41 reflecting the importance of the private sector in industrial employment. But despite industrial investments amounting to some US$2.5 billion in real terms dur- ing 1967-79, labor productivity remained below the 1966 level, reflecting the severe underutilization of capacity, poor overall performance and over- manning.42 Value added per worker was about one-third higher in private industries compared to parastatals. Real output per employee in manufac- turing has been declining constantly, and by 1979 was 70% of the 1966 4l/ Employment in parastatals increased independently of actual output suggesting excessive overmanning. 42/ A comparative study of parastatals and private enterprises concluded that: parastatals had lower labor and capital productivities, overemployment of labor, and lower managerial efficiency in 1970-75; and that, in contrast to private-owned enterprises which showed operating surpluses, parastatals in the same industry generally incurred operating losses. Kwan S. Kim, "Enterprise Performances in the Public and Private Sectors: Tanzanian Experience, 1970-75,0 Journal of Developing Areas, April 1981, p. 482. The study also found that, despite the lower labor productivity in parastatals, the average hourly wage rate was 271 higher compared to the private sector. Ibid., p. 478. For similar findings see also E.W. Clark, Socialist Development and Public Investment in Tanzania, 1964-73, University of Toronto Press, 1978, p. 117; F.C. Perkins, "Technology Choice, Indus- trialization and Development Experiences in Tanzania," Journal of Development Studies, January 1983, pp. 222-224; M.S. Silver, The Growth of Manufacturing Industry in Tanzania, (Boulder and London: Westview Press), 1984, pp. 270-274. - 66 - level.43 Real earnings per employee in manufacturing peaked in 1973, having increased by 11Z in 1967-73, but declined precipitously thereafter. By 1979 they were more than a quarter lower than in 1966.44 3.10 The rapid growth of investment in industry during the 1970s led to substantial additions in installed capacity which, since the early 1970s, 43/ In the early 1970s, Party guidelines were issued increasing the participation of workers in the management of Tanzania's enterprises (Mwongozo). A period of labor unrest (strikes, lock-outs) by dissatisfied workers (salaries, working conditions) ensued, which led to strong government reaction (dismissals and reinstatement of lock- outs). This has had serious effects on the operations of the enterprises. The government stepped in and crushed shopfloor initiatives, mandated compulsory arbitration in labor disputes, and introduced a system of individual labor grievance settlement. On the organization and participation of workers and the rise of industrial conflicts see Goran Hyden, Beyond Ujamaa in Tanzania - Underdevelopment and an Uncaptured Peasantry, (London: Heinemann Press), 1980, p. 163; H. Mapolu, "The Organization and Participation of Workers in Tanzania" in K.S. Kim, R.B. Mabele, M.J. Schultheis (Eds.), Papers on the Political Economy of Tanzania, 1979, pp. 272-277; W.B.L. Kapinga, "State Control of the Working Class Through Labor Legislation" in Issa G. Shivji (Ed.), The State and the Working People in Tanzania, CODESRIA, 1985, pp. 87-106; P.B. Mihyo, Industrial Conflict and Change in Tanzania, (Dar es Salaam: Tanzania Publishing House), 1983; F.L. Nyalali, Aspects of Industrial Conflicts: A Case Study of Trade Disputes in Tanzania, 1967-1973, (Dar es Salaam: East African Literature Bureau), 1975. 44/ According to other estimates, average real wages in manufacturing in 1979 were 58Z of their peak 1972 level, providing a severe disincentive to work, forcing workers to moonlight, and resulting in a substantial decline in manufacturing production. R. Skarstein, "Growth and Crisis in the Manufacturing Sector," in J. Boesen, K.J. Havnevik, J. Koponen, R. Odgaard (Eds.), Tanzania: Crisis and Struggle for Survival, 1986, Table 4.12, p. 94. - 67 . was not accompanied by commensurate growth of output.45 Capacity utiliza- tion during the period under review did not exceed on average 50% due to shortage of imported inputs, management and technical problems, machinery breakdowns,46 power and water supply failures, and project completions ahead of infrastructure.47 Rates of return on manufacturing investment declined steadily from 26Z in 1966 to 15% in 1976 and to 6% in 1980. And since protection to industry increased during this period and performance deteriorated, the decline in economic rates of return was most probably even steeper. 3.11 The slow growth of industrial output in the 1970s was exacerbated by the failure to shift foreign exchange allocations away from capital goods imports for capacity expansion and in favor of raw materials, spare parts and intermediate goods to improve capacity utilization. During 1971-79, imports of capital goods more than quadrupled and their share in total imports rose from 36% to 47%, while imports of intermediates more than doubled but their share fell from 292 to 16%. This anomaly has been attributed to the bias of external finance in favor of investments in new capacities (e.g., easier access to suppliers' credits for capital goods imports rather than intermediates; foreign aid tied to the capital goods content of specific projects); and the weak link between macro-level and 45/ It is noteworthy that import substitution played no role in the overall expansion of manufacturing. Estimates for 1965-72 show that the sources of growth of manufacturing output were: domestic demand 96%, export demand 5% and import substitution -1Z. There was import replacement of certain consumer goods, but this was offset by increased imports of other types of consumer goods and heavy reliance on imported intermediate and capital goods. The picture probably has not changed much since. World Bank Staff Paper No. 457, Industrial Strategy for Late Starters: The Experience of Kenya, Tanzania and Zambia, 1981, para. 21. Aside from the inducement provided by the protective environment, the practice of screening requests for imports according to the 'indigenous availability criterion" (i.e. only goods that cannot be supplied by a domestic producer can be imported) also contributed to the creation of excess capacity as producers, realizing that their licensed capacity determined their import allocation and hence volume of production, expanded facilities ahead of requirements. 46/ Tanzania's stock of machinery and equipment was acquired under disparate aid or suppiiers' credit arrangements which made standardization of spares extremely difficult, while their diverse origin required specialization of the limited pool of skilled operators and engineers available. Also, representing the latest in technological advance, they were ill-adapted to local needs and resources. World Bank, 1984, para. 2.40. 47/ Here again the problem lay in part in the lack of coordination between donor agencies and project sponsors. Ibid. - 68 - micro-level (project) planning, which failed to take full account of macro- economic constraints implicit in project analysis, because the process of project generation is decentralized involving parastatals, investment banks and private individuals whose investment decisions are based on short-term capital and foreign exchange requirements.48 Realistically, the link be- tween macro- and micro-planning has not been forged, primarily because of the inability of the responsible government agencies to vet project propo- sals on the basis of economic criteria rather than the expediency of access to foreign exchange, to ensure that micro-level resource requirements (e.g., foreign exchange, technical and managerial manpower, infrastructure, budgetary finance) are within the macro-level limits, to monitor and update resource requirements, and to institute an efficient and impartial resource allocation process. 3.12 The extent of import substitution and structural change achieved has not been significant. Domestic production as a percentage of total manufacturing supply rose from 442 in 1965 to 50% in 1978, while import dependency has been rising constantly.49 The share of consumer goods in manufacturing value added declined from 56% in 1965 to 55? in 1977, the share of intermediate goods declined from 40Z to 36Z, while the share of capital goods rose from 3% to 6%. This suggests that manufacturing contin- ued to be dominated by consumer goods, and that there was no noticeable improvement in the structure of manufacturing production, as indicated by the share of value added calculated at domestic prices. In fact, there may have been some deterioration, if estimates were based on world prices in view of the higher and disparate structure of protection introduced during the mid-1970s.50 3.13 Exports of manufactures during 1967-79 have been erratic, ranged from 8% to 18? of total exports, and averaged 12.6? of total exports and about 10? of manufacturing output. However, in real terms, manufacturing exports increased marginally in 1967-72 and declined constantly thereafter 48/ S.M. Wangwe, "Industrialization and Resource Allocation in a Developing Country: The Case of Recent Experiences in Tanzania," World Development, June 1983, pp. 483-492; B.J. Ndulu, op. cit., pp. 15, 20-22. 49/ World Bank, 1987, Vol. I, paras. 1.19-1.25. 50/ Ibid., Table 1.2, p. 3. I I - 69 - reaching by 1979 442 of the 1972 level.51 These trends reflect the exchange rate prevailing ideological anti-export bias, the low priority assigned to export growth, the lack of competitiveness, and the inadequacy and poor administration of incentives. Over the years, potent factors have affected export performance: most major industrial investments were directed to production for import substitution; low labor productivity and relatively high wages made a number of products uncompetitive; an overvalued exchange rate put exports at a competitive disadvantage; the structure of protection and pricing policy discriminated against exports; increased domestic consumption reduced exportable surpluses; irregular supply and inconsistent quality of agricultural materials reduced processed exports; government agencies were unwilling to enter into medium and long- term marketing agreements with foreign importers; market intelligence has been inadequate; quality control was poor or non-existent; public and private sector enterprises lacked interest in exploring new export possibilities; and, finally, administrative restrictions and cumbersome bureaucratic procedures (e.g., the import licensing system, the import 51/ The volume of exports grew at 2% annually in 1965-73 but decreased by 7% annually in 1973-77. World Bank, Staff Working Paper No. 413, Eastern and Southern Africa: Past Trends and Future Prospects, 1980, Table 6, p. 30. Between 1966 and 1979, the volume of exports declined by 70Z. On the other hand, production for exports, as a percentage of total manufacturing supply, declined from 7.8% in 1965 to 3.7? in 1978. World Bank, Tanzania: An Agenda for Industrial Recovery, 1987, Vol. I, Table 1.6, p. 9. - 70 - confinement system,52 procedures for reviewing export prices, export documentation, long delays in refunding duty drawbacks) further discouraged 52/ The Bank (and the IMF) have been criticized as having "pressured" the government in 1978 to "deconfine" imports, taking advantage of Tanzania's comfortable foreign exchange position. Allegedly, the decision precipitated a foreign exchange crisis and led to the depletion of the country's reserves, because a substantial proportion was allocated for imports of non-essential consumer goods due to administrative shortcomings and mishandling. It has been suggested further, that it would have been possible to increase import allocations for essential industrial inputs without dismantling the confinement system. Such an approach would have prevented the waste of foreign exchange on inessentials, and would also have made it easier to control the situation if circumstances changed. C. Payer, "Tanzania and the World Bank," Third World Quarterly, October 1983, pp. 797-799. The fact is that the need to deconfine imports was long overdue and basically correct. Nevertheless, the issue remains whether: (a) the decision to liberalize imports was justified at that point in time, given that it was based on the comfortable size of foreign exchange reserves in one exceptional year--the result of a "fluke" (boom in coffee prices caused by the frost in Brazil and favorable weather)--and not on the sustainability of the reserve position, which was not cushioned by a sufficiently broad export base and a rather stable inflow of foreign receipts; (b) pertinent fiscal, monetary and exchange rate measures could have been initiated to mitigate the potential risk to reserves; and (c) effective control mechanisms could have been put in place in advance of the decision to ensure that there would be no dissipation of reserves to non-essential consumer goods. Judged on these grounds, the Bank's reform proposal probably was ill-advised and premature. any potential export initiatives.53 Retrenchment - The Late 1970s and After 3.14 Real investment in manufacturing continued at about the same level during 1979-82, as major projects undertaken earlier were nearing comple- tion. Investment almost halved in 1983 and 1984, mirroring the dire eco- nomic difficulties that had ushered in, but picked up since, and in 1986 was slightly higher (Tsh 1,072 million) than in 1979 (Tsh 962 million). Capacityutilization dropped precipitously--from around 50% at the end of 1970s to about 25% in the mid-1980s--as the earlier aifficulties (paras. 3.10, 3.11) continued to impact with even greater force. From 1977 on, ICORs in parastatals turned negative as output began to decline (para. 3.08), and the trend persists to date. Real GDP during 1979-86 increased by 1.7% per annum; it virtually stagnated during the first half of the 1980s, but picked up somewhat in 1985 and 1986. The domestic savings rate declined steeply from 10% in 1979 to 2% in 1986, reflecting the mounting deficits of parastatals and suggesting an increase in the share of foreign assistance in financing industrial investments.54 Real value added in manufacturing declined throughout this period by about 5% annually, while its share in GDP fell from 12.0% to 7.3Z. Industrial employment increased 53/ World Bank, 1975, Vol. I, paras. 48, 108; World Bank, 1977, Annex V, paras. 5.66-5.71, Appendix C. For discussions on the performance of the manufacturing sector during shorter, and largely overlapping, time intervals during the 1967-79 period see: World Bank, 1981, paras. 1.31-1.36; World Bank Staff Working Paper No. 457, Industrial Strategy for Late Starters: The Experience of Kenya, Tanzania and Zambia, 1981; World Bank, 1984, paras. 2.25-2.45; World Bank, 1977, Annex V, paras. 2.1-2.24, 3.1-3.15, 5.1-5.40, 5.66-5.71; World Bank, 1987, Vol. II, Tables A, 2.3, 2.4 (pp. 101, 102); R. Skarstein and S.M. Wangwe, Industrial Development in Tanzania: Some Critical Issues, 1986, pp. 6-27 and Tables 1.2-1.4, 1.8, 1.10; A. Coulson, "The State and Industrialization in Tanzania" in H. Fransman (Ed.), Industry and Industrialization in Africa, (London Heinemann), 1982, pp. 60-76. For views taking rather kindly to the performance and achievements of manufacturing, see: ILO, Towards Self-Reliance, (Addis Ababa: JASPA), 1978; C.E. Barker, P.V. Mitschke-Collande, H.R. Bhagavan, D.V. Wield, African Industrialization: Technology and Change in Tanzania, 1986; ILO, Basic Needs in Danger: Basic Needs Oriented Development Strategy for Tanzania, 1982, pp. 209-235; R.H. Green, D.G. Rwegasira, B. van Arcadie, Economic Shocks and National Policy Making: Tanzania in the 1970s, (The Hague: Institute of Social Studies), 1980, pp. 75-81, 95-101; M. Bienfeld, "Evaluating Tanzanian Industrial Development" in M. Fransman (Ed.), Industry and Accumulation in Africa, 1982, pp. 105-141; R.H. Green, "Industrialization in Tanzania" in M. Fransman, op. cit, pp. 80-104; C.G. Kahama, T.L. Maliyamkono, S. Wells, The Challenge for Tanzania's Economy, (London: J. Currey), 1980, pp. 69-100. 14/ The gross domestic marginal saving rates were 13.9? in 1965-73, -18.7? in 1973-80, and -95.3% in 1980-85. from 86,000 in 1979 to some 106,000 in 1986, registering a net growth of 122 despite the slowdown in economic activity, and bearing witness to the fact that the sector harbors excess labor for social reasons. Real outp"t per employee in manufacturing continued to fall, and by 1985 was 56% of the 1966 level, while real earnings per employee in 1984 were only 302 of the 1966 level. The situation has probably deteriorated further in the last few years.55 3.15 The structure of manufacturing production did not change percepti- bly through the mid-1980s, as the import substitution effort has not pro- duced tangible results. Domestic production as a percentage of total manu- facturing supply rose from 50% in 1978 to 592 in 1984, but input dependency continued to grow and reached high levels by the mid-1980s, reflecting serious structural weaknesses.56 The share of consumer goods in manufacturing value added declined further, falling from 552 in 1977 to 502 in 1983, the share of intermediates rose from 362 to 402 but they were still at the 1965 level, and the share of capital goods rose further from 6.52 to 9.32. However, measured in world prices, the share of consumer goods in 1984, amounting to 852, was substantially higher than in the 1960s, while only 152 of the value added is generated by the investments sunk in producer goods industries which accounted for two-thirds of all installed capacity.57 3.16 Finally, exports of manufactures, both in nominal and real terms, have been declining in the 1980s, reflecting the continued influence of the adverse factors discussed in para. 3.13. In 1986, they stood at 602 of the 1979 level while their share to total exports had dropped from 182 to 17%, averaging 72 of total exports during 1980-86. Production for exports, as a percentage of total manufacturing supply, rose marginally to 5z in 1984 compared to 3.72 in 1978, but was still below the 7.82 in 1966. In real terms, the value of exports almost halved during this period. In 1986, the volume of exports, declining steadily, was 382 of the 1980 level; that was already 702 below the 1966 level (see para. 3.13, footnote 51). 55/ For an account of the level of economic efficiency of the industry in 1984, including the sources of inefficiency, see World Bank, Tanzania: An Agenda for Industrial Recovery, 1987, Vol. I, pp. 33-66 and the entire Vol. II. 5a/ World Bank, 1987, Vol. I, paras 1.19-1.25. The ratio of imported input costs to gross output rose from 14.62 in 1973 to 27.8% in 1984 (62% in shadow prices), while the imported input costs to total input costs from 22.92 to 52.5% (702 in shadow prices). Ibid., p. 10. 57/ ibid., pp. 3, 4. IV. SCOPE OF BANK SECTOR WORK 4.01 Besides the regular economic missions which touched on industrial sector issues, the Bank mounted special missions to review in greater depth developments in the sector. These missions commented on the government's industrialization strategy and identified major and persistent shortcomings in the structure, institutional arrangements and performance of the indus- trial system, including the investment decision-making process and operat- ing efficiency of the parastatals, and have made specific and detailed recommendations over the years for remedial action. The Bank'r Assessment of the Basic Industry Strategy 4.02 The 1974 mission supported in principle the basic industry strat- egy, but argued that the strategy must be qualified by efficiency require- ments that maximize growth of value added and foreign exchange con- tribution.58 Furthermore, the mission recommended that industrial invest- ment should: raise productivity in existing manufacturing enterprises through balancing investments, improved maintenance, additional shifts, etc.; expand consumer goods industries to meet shortages; step up total industrial investment and place more emphasis on the joint venture approach;59 implement and budget only for projects for which it is reason- ably certain that the financial resources required will be available and the sponsoring parastatals have the managerial and technical capacity to implement; and emphasize human skill development, as many parastatals and foreign management agencies are paying lip service to the training of coun- terparts.60 Finally, the mission indicated that many projects included in the Plan had no feasibility studies, estimates of resource availability were highly optimistic,61 and "given the stringency of the preaent and foreseeable foreign exchange situation, it is not at all clear whether the strategy's implicit dependence on intermediate and capital good imports can be sustained without more emphasis on accelerated expansion of manufactured exports."62 58/ World Bank, 1975, Vol. I, paras. 4, 66, 67, 110. 59/ Joint ventures between foreign companies and domestic parastatals were thought of as providing the most practical way to encourage the inflow of foreign capital and expertise and to ensure efficient and timely project implementation. Concerning deals with foreign partners, often interested in the sale of equipment, the Bank's advice was that the government should ensure that prospective partners have a strong and lasting incentive to aim at high efficiency standards during production. World Bank, 1975, Vol. I, para. 111. 60/ World Bank, 1975, Vol. I, para. 111. 61/ Ibid., paras. 112, 114, 115. 62/ World Bank, Economic Memorandum on Tanzania, 1977, para. 48. 4.03 The 1976 mission also took kindly to the thrust of the BIS, and argued the case for the strategy on the following grounds. First, BIS should not be judged solely by its effects on efficiency and growth. The political leadership is willing to sacrifice a substantial amount of growth in the medium term in order to restructure the economy and reduce external dependence (e.g., by foregoing the gains of specialization in the interest of reducing foreign dependence, by minimizing the importance of scale econ- omies and comparative advantage in investment deciaions). Efficiency mat- ters, but it is not the overriding concern. Second, most of the import substitution industries selected appeared to be relatively efficient at world prices (due to the natural protection of transport, possibility of achieving minimum scale economies within a 20-year horizon, exploitation of easy import-substitution possibilities, development of industries using the by-products of existing industries). On the other hand, the suggested export-oriented industries by alternative strategies appeared only margin- ally efficient, as Tanzania may not have the alleged strong comparative advantage in labor-intensive manufactured exports (due to low labor produc- tivity and high wages). Third, no other strategy can have more significant impact on the generation of employment opportunities, and the employment problem would have to be largely resolved by the growth of other sectors. Fourth, the potential for developing in the long run a minimal export ca- pacity by efficient import-substitution industries (assuming early atten- tion to the right price, high quality, and development of market channels) introduces a flexibility in the face of an uncertain international market. Finally, BIS emphasizes the development of the metals engineering sector which fosters technological independence and serves as a breeding ground for a wide range of industrial skills.63 4.04 The mission, however, pointed to a number of shortcomings and po- tential pitfalls in the BIS. First, the capital and foreign exchange re- quirements would be very high in the early years (capital-intensive invest- ments with long gestation periods) and costly in term:; of foregone growth. Second, the costs of a concurrent structural change in industry and agri- culture (reduction of the share of exported cash crops) would be prohibi- tively high since, to earn the foreign exchange to meet the import require- ments of the investment levels envisaged by the BIS, will require rapid expansion of agricultural exports during the next decade or so. Third, the strategy attached an unwarranted importance to the steel industry, whose development a. this stage would clearly be premature. Fourth, in the me- dium term, the BIS would increase dependence because of the capital-cum- import intensity of BIS investments and the relative technological sophis- tication of many of the proposed industries, implying a clear trade-off between less self-reliance now and more a decade or longer hence. Fifth, the development of a metals engineering industry would be very difficult, especially for a country at an early stage of industrialization. Finally, because of the linkages among industries, the BIS would make very high demands on the country's planning capacity, which is the scarcest resource, thereby threatening the very viability of the BIS. Considerable centralized control would have to be exercised to ensure that investments 63/ World Bank, 1977, Main Report, paras. 5.50-5.55; Annex V, paras. 4.18-4.34. - 75 - are sequenced properly to take advantage of linkages and to ensure that small-scale production possibilities are not pre-empted. Yet, such cen- tralized control over investment decisions would be inconsistent with the development of the inherently decentralized metal engineering industry, a3 well as with parastatal efficieny.64 4.05 On the basis of a macro-model of the Tanzanian economy, the mission attempted to simulate and compare the BIS with an alternative growth-ori- ented strategy. Quantitative guesstimates suggested that, by 1990, real industrial value added would be some 20Z lower under BIS, real GDP about 5Z lower, gross domestic savings by about one-third lower, and income distri- bution and employment growth also somewhat worse under BIS. The mission concluded that, given the importance attached to structural change by pol- icy makers, in principle, BIS seemed reasonably well-conceived, given that over a 20-year period Tanzania would tend to have (potentially) a compara- tive advantage in import-substitution industries rather than in export production (due to the higher cost of unskilled labor emanating from the relative abundance of land which raises the marginal product of labor in agriculture rather than the distortions of high minimum wage rates, the lower efficiency of labor, and the overvalued exchange rate); backward- linking import substitution, particularly if some export possibilities could be developed by efficient import-substituting industries, could im- part flexibility to the economy in the long run and prevent the emergence of a foreign exchange constraint; and the central role assigned to the metals industry appearet defensible on grounds of both growth and structur- al change. It is noteworthy that, over a 20-year period, the differences in outcome between BIS and the maximum growth strategy were not that sig- nificant as there would be substantial overlap of industries and production processes.65 On the other hand, there were some features of the strategy which called into question both its ability to meet minimum growth targets and its consistency with its own objectives. The main concerns were that attempts to restructure the economy too quickly could ultimately frustrate both growth and strr-tural change, that the proposed expansion of the steel industry would con er t little to the ultimate objectives of the BIS, and that the planning capacity of the country would be overburdened. Moreover, if too much attention were paid to long term strategy, there would be a grave danger that individual project decisions would be made without the needed hard-nosed, micro project evaluation.66 The mission concluded that "perhaps what is called for is a gradual shift towards the BIS.n67 64/ World Bank, 1977, Main Report, paras. 5.56-5.60; Annex V, paras. 4.34-4.46, 6.10. 65/ World Bank, 1977, Annex V, para. 1.45. 66/ World Bank, 1977, Main Report, paras. 5.61, 5.62, 5.64; Annex V, paras. 4.22-4.24, 4.26-4.31, 6.8, 6.10. 67/ World Bank, 1977, Main Report, para. 5.63 (emphasis added). - 76 - The Industrial Reorientation Drive 4.06 The Bank, as part of its sector work, prepared in 1987 an Agenda for Industrial Recovery following an in-depth study of the sector. The long range industrial reorientation drive proposed by the Bank aims at increasing the efficiency of resource allocation and use, with emphasis on raising production, productivity and capacity utilization. Specific ele- ments of the proposed strategy include inter-sectoral reorientation, with agriculture and infrastructure receiving a larger share of resources while industry would play a significant but subservient role; intra-industry reorientation, supporting the more productive firms and weaning inefficient ones; reorientation of resources toward recurrent and rehabilitation needs and away from new investments; allocation of resources toward smaller firms; balance between the private and parastatal sectors; reduced import- substitution and increased export orientation; and development of local technical capability. A set of policy reforms, to be implemented gradual- ly, are also recommended in exchange rate allocation, trade, and pricing regimes to provide the requisite incentives. Emphasis would also be placed on improving public investment programming (preparation of a consolidated public investment program, curtailment of new and deferment of low priority investments, rehabilitation of existing firms, improvement of decision- making process); restructuring of parastatal enterprises; development of technical skills; promotion of small scale industry; strengthening of the development banks and improving the quality of their portfolio; and promot- ing aid coordination.68 4.07 As part of Economic Recovery Program, the Bank and the government have been working jointly to develop action programs to restructure and rationalize the industrial sector, to be supported by sector lending. The proposed industrial restructuring scheme envisages three phases. Phase I will involve a diagnostic study of selected subsectors and the identifica- tion of pressing issues both at the enterprise and subnector levels. Dur- ing Phase II, rehabilitation and restructuring programs will be developed, including time-tables for implementation. Phase III will lead to the im- plementation of agreed action programs. This effort will be reinforced by measures to ensure appropriate foreign exchange rate policy, liberalize import restrictions and the foreign exchange allocation regime, limit price controlled items to specified minima, rationalize the tariff structure based on the recommendations of a study carried out by the government with Bank assistance, reform export incentives, and liberalize the internal trade (decon!inement).69 Furthermore, in 1987 the Bank conducted a compre- hensive study of the parastatal sector70 which makes recommendations that 68/ For details see World Bank, 1987, Vol. I, pp. 91-131. 69/ PR No. P-4944, Industrial Rehabilitation and Trade Adjustment Program, November 22, 1988, paras. 93, 100-103. 70/ World Bcnk, Parastatals in Tanzania: Towards a Reform Program, Report No. 7100, July 27, 1988. See also World Bank, 1987, Vol. II, pp. 36-59. - 77 - . would address, inter alia, fundamental causes of the parastatals' poor performances absence of overall strategy and clarity on goals, excessive number of parastatals given the resource and managerial capacity, weak or non-existent competitive pressures, discriminatory treatment, and absence of a regulatory system that rewards and punishes on the basis of perfor- mance. The recommendations of the study aim to initiate a process that will produce and implement a Parastatal Recovery Program that is consistent with and complementary to the Economic Recovery Program. Major Recurrent Issues Highlighted by Bank Missions Investment Planning, Coordination of Investment Decisions, and Project Selection 4.08 Bank missions have repeatedly commented that the investment plan- ning system has performed very poorly, and tbs failure has been respon- sible for the injudicious choice of parastatAi investments, the lack of structural change, and much of the disappointing performance of the indus- trial sector over the yeaes. Identified deficiencies of the planning sys- tem include: focus on the annual budget as the major instrument of control and, within that context, excessive concern with the source of finance, especially availability of external funding; inadequate attention to future cost implications of current decisions, and a strong tendency toward over- commitment with consequent underfinancing of all projects and time and cost overruns; inadequate coordination of public investment decisions and for- eign assistance; and tendency of parastatals to evade both central planning controls and selection criteria in their investment plans.71 4.09 Missions have reported that, more often than not, parastatals have initiated and decided on major industrial investments without appropriate screening by the controlling ministries, have sought out on their own ex- ternal aid commitments or suppliers' credits, and have approached the Treasury for approval and local funding only after such arrangements had been completed. This approach has undermined efforts at coordinated vet- ting and priority setting, and has accentuated the tendency toward overcom- mitment. In effect, the overriding criterion of project selection became availability of external financing, and there have been numerous instances in which aid has distorted investment priorities or the choice of tech- nique. The main motivation of parastatals in the prevailing policy 71! World Bank, 1977, Annex V, paras. 4.46, 6.15; World Bank, Economic Memorandum on Tanzania, Report No. 1567, April 12, 1977, para. 46. For an assessment of the organization, management and supervision of parastatals, see World Bank, Report No. 7100, Parastatals in Tanzania: Towards a Reform Program, Appendix C. - 78 - environment has been to create new facilities and expand production.72 And since the budget process and the production planning system does not empha- size cost reduction and efficiency, parastatals can pursue their goals unimpeded and with impunity. When commitments for priority projects ex- ceeded resource availability, the typical reaction has been not to reduce the number of projects, but rather to make smaller allocations over the initial project list. As a result, fewer projects got to be completed and with long delays.73 4.10 The choice of projects and techniques of production has been the outcome of bureaucratic processes, involving a complex interaction of the budgeting process, the practice of foreign aid commitments, and the methods and motivations of parastatal planning. The results, with a distinct his- toric bias in favor of large scale production, bear only a coincidental relation to any formal development plan or to the proclaimed goals of in- dustrial development. The uncertainty surrounding investment criteria among parastatals and the supervising ministries, the persistent acute shortage of manpower capable of preparing commercially viable projects in the parastatal sector, and the tendency for projects to be approved when they were hardly more than project ideas and full financing was not as- sured, have exacerbated the situation. Finally, the typical response to any perceived fault in the decision-making process has been either to reor- ganize a ministry or create a new parastatal as a counterweight to the entrenched interests responsible for the faulty decisions.74 Policy Framework 4.11 As early as 1971, the Bank was concerned that high levels of effec- tive protection and an array of investment incentives, including their inefficient and disparate implementation, have placed an exaggerated empha- sis on promoting import-substituting industries, have fostered a degree of capital intensity hardly warranted by the existing factor endowment, and have contributed to the establishment of many inefficient manufacturing firms with a high cost structure which was detrimental both to the coun- try's export potential and to consumer and user interests.75 In general, 72/ For instance, missions expressed serious reservations about the economics of the fertilizer and auto tire projects, both of which were already under implementation in the late 1960s, while the prospects for an integrated iron and steel, automotive and basic chemicals projects were deemed not to be promising. World Bank, 1970, paras. 113, 121-123, 131, 135, 151-154; World Bank, Report No. AE-16a, Economic Developments in East Africa, June 28, 1971, Vol. I, paras. 127-136; Vol. III, paras. 198, 203. 73/ World Bank, 1977, Main Report, paras. 34, 6.35-6.38; Annex V, paras. 5.46-5.56. 74/ World Bank, 1977, Annex V, paras. 5.53-5.56. 75/ "The compulsion to promote investment and industrial expansion has been overriding." World Bank, 1971, Vol. III, para. 65. the macro-incentive system tended to be over-protective, failed to encour- age employment and exports, was biased against both foreign and domestic investment, and did not stimulate investment in small-scale industries. The distribution system was confusing and inefficient, and the foreign exchange allocation plans discriminatory. The system of price controls, comprehensive import licensing, and confinement of imports virtually fore- stalled direct competition between foreign and domestic suppliers, and international prices lost their significance as a check on the efficiency of domestic production. At the same time, alternative institutional ar- rangements to guard against inefficiency and incentives to improve produc- tivity were extremely weak, if non-existent.76 Organizational Structure and Performance of Parastatals 4.12 Th,i creation of numerous holding parastatals, missions observed, tied up large numbers of scarce indigenous Tanzanian managers, accountants and technical specialists, spread thinly senior Tanzanian management, and created greater need for expatriates, while the overhead costs became a burden on the operating subsidiaries and reduced their surpluses. In many instances, it was not clear whether the cost of operating holding parasta- tals was justified by the accruing benefits, and whether their subsidiaries wo- .: not perform equally well or even better without them. In general, the organizational structure of the public sector has been top-heavy with general planners, advisors and supervisors, while there has been a severe shortage of experienced personnel for pro ect implementation, plant man- agers, technicians, financial analysts, etc.;7 4.13 The Bank found that the performance of parastatals has been persis- tently poor. Labor and capital productivity has been declining constantly due to inadequate project design, prolonged technical and learning prob- lems, unclear objectives and responsibilities, scarcity of management cadres and skilled technicians,78 poor maintenance and repair facilities, inadequate supplies of water and electricity, transport bottlenecks, short- age of imported raw materials and spare parts, irregular flow of local raw materials, management failures (attitude, motivation, competence), low capacity utilization, and inadequate incentives for both workers and mana- gers, as rewards for efficiency and penalties for inefficiency have been very xreak, and much reliance has been placed on political education in and 76/ World Bank, 1971, Vol. I, paras. ix, 51-99, 154-158; World Bank, 1972, Vol. I, paras. 7, 76, 80; World Bank, 1975, Vol. I, paras. 6, 7, 81-95, Annex I; World Bank, 1977, Annex V, Appendix B. 77/ World Bank, 1975, paras. 59-60; World Bank, 1977, Annex V, paras. 3.16-3.20, 6.14. 78/ Technical education has encountered many problems, including inadequate facilities, enrollments lagging behind targets, high wastage rates, and insufficient staff. As a result, throughput could not meet the rising demand. 80 outside the work place as an instrument to instill in workers and managers a stronger sense of purpose and a greater awareness of the need for dis- cipline.79 The cost-plus price control system has been a particularly ad financial losses, and the parastatal sector as a whole experienced a falling real rate of return on capital.80 4.14 The critical problem of low factor productivity in the manufac- turing sector was found to be linked to the continuing absence of a work- able system of performance indicators. There have been no clear and un- equivocal performance yardsticks and procedvres for the systematic review and comparison of the performance of parastatal enterprises. By default, financial results have been relied upon as a crude measure of performance, even though they depended more on decisions taken outside the enterprise (e.g., regarding prices, wages and input supplies), rather than on manage- ment decisions and effort applied within the firm. The bonus system for parastatal workers which had been in existence for years (but little-used) had also foundered. Thus, while material incentives had weakened, adequate compensatory non-material substitutes had generally been lacking. Although the government has been well aware of all these problems, it has not been able to devise effective solutions.81 Actions Recommended by Bank Missions Efficiency of the Socio-Economic System 4.15 Bank sector and economic missions have made specific recommenda- tions on an array of institutional and policy issues which, if acted upon, could have assured a better outcome in terms of growth, equity and achieve- ment of other government objectives, and "without compromising the funda- mental character of the socio-economic system which it has adopted.o82 79/ OThe relatively low productivity found in many public sector enterprises is not in the first place due to circumstances beyond the control of the management but to an apparent lack of motivation on the part of the managers and workers alike." World Bank, 1975, Vol. I, para. 71. 80/ World Bank, 1972, paras. 38, 89; World Bank, 1975, Vol. I, paras. 8, 13, 53, 71, 74; World Bank, Economic Memorandum on Tanzania, 1977, para. 46; World Bank, 1977, Main Report, paras. 62, 6.27; Annex V, paras. 5.01-5.17, 5.32-5.40, 6.12-6.14. 81/ World Bank, 1975, Vol. I, paras. 71, 108: World Bank, Economic Memorandum on Tanzania, 1977, paras. 46, 56. 82/ World Bank, 1977, Main Report, para. 21. Having supported in principle the centrally planned, public sector oriented approach, the institutional framework, and the basic industry strategy, Bank advice was mostly directed at making the Tanzanian system as efficient as possible. Within these basic parameters, recommendations reflected appreciation of the foreign exchange and other constraints facing the sec- tor, and the strength inherent in its still active private sector. Report after report identified the shortage of fcreign exchange as a major con- straint on capacity utilization and sector performance, and linked the decline in agricultural cash crop production to this shortage. The contin- ued importance of the private sector in manufacturing value added, persis- tently high private investment, and greater efficiency in the use of capi- tal (as reflected in ICORs) relative to parastatals wAs regarded as an important strength to be used to further sectoral growth. Bank advisors regarded the industrial economy as mixed, and their advice was directed to making both the private and public parts more efficient through improve- ments in the policy and incentive framework, reforms in the system of plan- ning and controls, and relaxation of foreign exchange and infrastructural constraints. Policy and Incentive Framework 4.16 Bank missions suggested that although the macro-incentive framework has lost much of its importance as a factor determining the size and direc- tion of industrial investment, having been largely replaced by a process of administrative decision-making, a policy framework consistent with the overall objectives could still be useful. For instance, a system of price incentives could be set up which would reinforce, and to some extent re- place, administrative controls while the number of price-controlled items can be reduced. The existing incentives structure for import substituting industries could place more emphasis on stimulating employment, efficiency and exports. Existing levels of external protection could be reduced over time, to put pressure on producers to become more competitive and efficient and to promote export capability. Import licensing and quantitative re- strictions on imports could be gradually phased out, with increasing reli- ance on a more uniform tariff structure to afford a reasonable degree of protection. Furthermore, increased selectivity in granting drawbacks of duties on imported inputs could help reduce the excessive levels of effec- tive protection and provide a stimulus to the development of producer goods industries. Finally, production licensing needed to be reconsidered as it tended to shelter inefficiency, high prices, excessive profits, and to ignore consumer interests.83 831/ World Bank, 1971, para. xxi; World Bank, 1975, Vol. I, paras. 68-71, 73; World Bank, 1977, Main Report, paras. 6.21-6.23; Annex V; paras. 4.51-4.57, 6.19, 6.20. 82 4.17 While the preceding advice on the policy and incentive framework was directed at compensating for distortions leading to misallocation of resources and, perhaps, would have been adequate for the early 1970s had it been followed, the appearance of formidable resource constraints later in the decade led to new distortions as the parastatal sector was given priority access to foreign exchange for imported capital goods and operat- ing inputs and to external finance for development projects. In addition, the allocation of foreign exchange among subsectors was made on the basis of criteria other than efficiency, suLh as the importance of subsector production to government revenues, e.g., beer and cigarettes. Given the continuing importance of the private sector in manufacturing value added, Bank missions were prompted to return to an earlier theme, urging the re- moval of uncertainties regarding the role of the private sector and the issue of an investment code to promote private (including foreign) invest- ment in the manufacturing sector.84 In the early 1980s, when the Government appreciated the need to avoid involving the public sector in areas already served by the private sector, the Bank took the opportunity to advise that access to the export earnings retention scheme would be a valuable incentive to export-oriented private manufacturers who were receiving very little from the foreign exchange allocation process.85 More recently, when the government's decision to undertake certain structural adjustments in order to rehabilitate the ecot-my signalled the opportunity to restart the dialogue on the policy framewc -k, many of the original Bank recommendations were still relevant. For in-tance, in the last decade of deep crisis, the framework of controls for administering resource allocation had not been dismantled; however, now the Government appeared disposed to accept the need for gradual change toward a less discretionary and more automatic system. Planning and Economic Management 4.18 The thrust of the Bank's recommendations on planning and economic management was that Tanzania not rely on a comprehensive system of direct controls, given the government's limited administrative capacity, the acute shortage of experienced senior and middle level Tanzanian managers, the existence of a still fairly important private sector, and the strong depen- dence on external prices, imported technology and expatriate technical expertise. Successive Bank missions suggested that a judicious and selec- tive use of indirect controls, material incentives, and decentralized deci- sion-making at the enterprise level--a change in the direction of "market socialism"--might well be compatible with Tanzania's development objectives and at the same time conducive to greater efficiency and industrial devel- opment.86 To improve the effectiveness of the agencies involved in planning 84/ World Bank, 1971, para. xxi; World Bank, 1975, Vol. I, paras. 20, 21, 50; World Bank, 1977, Main Report, paras. 31, 6.39-6.41; Annex V, paras. 5.80-5.92, 6.16, 6.17. 85/ World Bank, 1984, para. 5.31. 86/ World Bank, 1975, Vol. I, paras. 9, 15, 43, 68-70; World Bank, 1977, Annex V, para. 6.18. and policy-making, missions suggested greater stability of tenure of key personnel, selectivity with respect to what is to be controlled or managed, timely production of more consistent data, and greater concern for resolving more immediate problems to balance the tendency to focus more on strategic issues relating to the more distant future.87 4.19 Regarding parastatals, the Bank's advice was that the investment planning system could be improved by reforming the tariff and pricing sys- tem and by restructuring the incentive system to focus attention on costs and productivity. This would require that holding companies prepare me- dium-term perspective plans for early review; the Treasury take the lead in securing foreign finance for projects rather than accepting whatever hap- pens to be offered; establish better project selection procedures to ensure that proposed projects fit in with Plan objectives, to avoid inconsisten- cies, poor phasing and sub-optimal investments, and to facilitate proper sequencing; base project gelection on cost-benefit analysis to establish their economic merit, while scrutinizing suppliers' credits and contractor financing arraigements entered into by parastatals; and introduce better project costing and budgeting procedures to establish better cost control, speedier implementation, and more realistic projections of finance require- ments. Missions also recommended that multi-year planning include consid- eration of recurrent expenditures and the outlook on foreign exchange availability, in order to reverse the pattern of overcommitment of re- sources. Finally, Bank missions recommended that greater attention be paid to aid coordination in light of the major role played by donors in the direction of industrial investments. 4.20 Consistent with its view that decentralized decision-making at the enterprise level was preferable, the Bank suggested that improving the operating efficiency of parastatals (e.g., by reducing excessive labor force, improving capacity utilization, raising labor productivity), should largely be the corcern of individual parastatals and their holding com- panies.88 This would be done within the context of a coherent framework of policies aimed at influencing, monitoring and evaluating the performance of parastatals. Monitoring could include periodic audits of parastatals to root out overmanning and ensure greater financial accountability of man- agers. It was further recommended, that company maragers be given greater discretionary power to deal with disciplinary problems, that greater parti- cipation of workers could be allowed on an experimental basis,89 and that 87/ World Bank, 1970, Vol. I, para. 126; World Bank, 1971, Vol. I, paras. xxi, 98-104; World Bank, 1977, Main Report, paras. 33-35; Annex V, paras. 5.57-5.65, 6.25; World Bank, Economic Memorandum on Tanzania, 1977, para. 49. 88/ Interestingly, although missions to date have questioned the role of holding companies (para. 47), they have not suggested any reforms, including the possibility of their dissolution and allowing parastatals in the same sector to compete with each other. 89/ In light of the potential for abuses (see para. 28, footnote 43), it is not clear why missions would venture such a recommendation. parastatals could experiment with payment by result schemes in order to increase worker productivity. Abolitiovn of subsidies from the Treasury for loss-making parastatals was also sugges..ed, although potentially more dras- tic measures for dealing with losses were never specified.90 91 Strengthening the Productive Base of the Economy 4.21 Mindful of the fact that an import substitution industrial develop- ment strategy is crucially dependent on the level and growth of domestic demand and, in turn, on the level and growth of income, Bank industrial sector missions have been concerned about the productive base of the econ- omy, primarily agriculture and its sluggish development. Over 80Z of Tan- zanian households depend on agriculture for their livelihood. Agriculture also accounts for about 802 of the foreign exchange from exports, and therefore is crucial to securing imported inputs for industry. In addi- tion, agriculture could provide the inputs for a dynamic agro-industry. Infrastructure should also be regarded as part of the productive base of the economy, and its parallel development is equally important. Bank mis- sions have recommended strengthening agricultural production through im- provements in management and resource use, and have urged a more balanced approach toward the inter-sectoral distribution of investment. Noting that inadequate maintenance was leading to premature destruction of infrastruc- ture which could impede industrial expansion, missions advised that less emphasis should be placed on new investment in the public sector and great- er on budgetary allocations for recurrent expenditures, particularly main- tenance.9 90/ This suggestion was in line with the Finance Minister's 1975/76 Budget speech which stated that "Those parastatals which fail to generate surpluses will have to be closed down...." But it wts not followed through in earnest. 91/ World Bank, 1972, Vol. I, paras. 6, 38; World Bank, 1975, Vol. I, paras. 77-80; World Bank, 1977, Main Report, paras. 25, 32, 6.29-6.31; Annex V, paras. 5.18-5.31, 5.41-5.45, 6.21, 6.22, 6.24. 92/ World Bank, 1977, Main Report, paras. 22-26. - 85 - Export Promotion 4.22 Bank missions early on had stressed the need for export promotion.93 suggesting that the export potential of industry could be improved through phased policies and programs designed to spur competition and improve effi- ciency in manufacturing. This would require the reorientation of protec- tion and incentives policies to maintain pressure on producers to achieve production economies and lower costs, supplemented with positive incentives (e.g., exemption from sales and excise taxes, tax relief on export profits progressing with value added, simplification of cumbersome export marketing procedures, access to export credits and insurance, streamlining procedures for quick reimbursement of import duties, development of export promotion schemes) and frequent adjustments in the foreign exchange rate.94 4.23 An export strategy based on agricultural exports was outlined in 1976 by the Bank mission, following an assessment of the country's poten- tial. "For at least the next decade, a successful export strategy must rest on boosting agricultural exports. This is consistent with a develop- ment strategy centered on rural development, which offers the best long- term prospect for transforming Tanzania's underexploited resource of land into output, incomes and exports. While there is scope for expanding some resource-based manufactured exports, Tanzania does not, at present, have any pronounced comparative advantage in labor-intensive, manufactured ex- ports. For the next 10 to 15 years, the major contribution of the indus- trial sector to reducing external imbalances should lie with import substi- tution. But to ensure that such import substitution takes an efficient path will require certain reforms in the structure of tariffs and taxes and the opening up of the industrial sector to more external and internal com- petition."5 The mission indicated further that there was scope for gen- erating more exports from existing manufacturing industries. However, this would require, inter alia, increased investments in export-oriented indus- tries, improvements in factor productivity, favorable changes in the insti- tutional and price policy framework for manufacturing exports, and admini- strative measures to remove bureaucratic hurdles--all of which had been neglected.96 93/ 'If Tanzania does not succeed in encouraging and diversifying exports, the country's development will probably be hampered by a scarcity of foreign exchange." World Bank, 1972, Vol. I, para. 8; see also paras. 120, 121. 94/ World Bank, 1971, Vol. I, paras. 159-167; World Bank, 1975, Vol. I, paras. 50, 52; World Bank, 1977, Annex V, paras. 4.56, 6.23. 95/ World Bank, 1977, Main Report, para. 29. See also para. 15, footnote 17 above. 96/ World Bank, Economic Memorandum on Tanzania, 1977, para. 49; World Bank, 1977, Annex V, paras. 5.72-5.79. - 86 - V. IMPACT OF BANK SECTOR WORK 5.01 The Bank mounted special missions at frequent intervals (1959, 1966, 1970, 1974, 1976 and 1985) to review in-depth and assess developments in the indastrial sector. These missions identified major weaknesses in the structure, institutional set-up and performance of the industrial sys- tem, and made specific and detailed recommendations over the years for remedial action (paras. 4.08-4.23). The Bank's industrial sector work generally has been of high quality in terms of coverage of issues, depth of analysis, and overall diagnosi- and prescription. Sector reviews have been helpful in carrying out a dialogue with the government on policy issues- -albeit with very limited success, in identifying project possibilities and, on occasion, in dissuading the government from undertaking ill-advised projects, in analyzing the performance of major industrial subsectors and, in the 1980s, in the formulation of an agenda for action to rehabilitate the sector. 5.02 Industrial sector reports produced in the 1970s--as opposed to those issued in the 1980s--basically supported the government's strategies and, with notable exceptions, the types of projects included in the Plans, without full appreciation of the gravity of binding constraints. Presenta- tion of the findings and recommendations of sector missions through the late 1970s was almost invariably followed by qualifying statements of opti- mism, empathy, and appreciation of laudable efforts by the government and its agencies to rectify the situation at hand when, in fact, such initia- tives seldom were expected to be forthcoming, thereby diluting the force of the argument and the urgency for action, and paying lip service to the advice offered. In particular, growth of industrial investment per se and deceptive employment generation (due to excessive overmanning) received praise, while the quality (i.e. the efficiency) of the new investments undertaken received only belated attention (mid-1970s)--and only when the expected results had failed to materialize and nagging problems had sur- faced. Evidently, the Bank did not fully comprehend the implications of the speed, content, and all-out effort of the government at rapid indus- trialization in terms of requirements in foreign exchange, infrastructure, skills (technical and managerial) and inter-sectoral links (agriculture and infrastructure in particular), and was not prepared to question openly the underlying conflict between ideological intentions and what was realistic- ally feasible. 5.03 In all fairness, sector missions did report objectively and candid- ly on persistent issues, and did draw attention to the inefficient way policies were implemented as well as to their potential consequences; how- ever, particularly before the mid-1970s, findings and recommendations were scattered in the text of long reports and, often, were not presented stark- ly up front in the summary and conclusions (or in a separate document) and in the form of actionable programs. The findings and recommendations of sector missions (the coverage of industry in regular economic reports was rather limited and usually of an updating nature) were discussed at the hierarchically appropriate levels of the policy-making bodies, although few Of - I firm commitments were obtained (or if any commitments were made these were not adhered to). Thus, Bank-government exchanges have had limited lasting impact, in part because the true locus of decision-making power rest with the more doctrinaire and less visible Party apparatus. Moreover, the re- currence of the issues identified by sector missions, normally years apart, suggests that there has been no systematic follow-up by the ensuing and more regular economic missions, or through other channels of communication, thus emasculating their potential impact and, in effect, disrupting the continuity of the sector dialogue. The hiatus was further widened by the frequent change in the personalities involved on both sides. In short, the open-ended and largely inconclusive and non-commital dialogue, the inappo- site presentation of findings, and the lack of follow-up on important is- sues tended to undermine the effectiveness of the sector work and to rou- tinize the dialogue. 5.04 The potential impact of the sector work was further diluted be- cause, at the decision-making centers where Bank policy is formulated, the findings and recommendations of sector missions were not adequately assimi- lated, their implications were not fully appreciated and, as a result, they did not receive the attention they deserved. Rather, a highly accommodat- ing bureaucratic, if not self-serving, posture was taken which was at vari- ance with the dictates of the economic realities of the country--at least until the early 1980s. The CPPs produced during the 1970s, in their rendi- tion of economic and sector work, kept praising effusively the country's leadership for their commitment to development and their social achieve- ments, and disregarded serious policy shortcomings and fundamental weak- nesses in the management of the economy. Planning, implementation, and, generally, absorptive capacity were rather lightly dismissed as not being absolute concepts but functions of a country's willingness to recognize bottlenecks and to deal with them. In this respect, CPPs lacked forth- rightness and perspective, glossed serious matters over, and tended to reflect entrenched Management positions, thereby undermining their effec- tiveness as a policy instrument. Little attention was paid to the fact that many times what was planned had little relation to what was ultimately implemented and, as a result, no attempt was made to communicate to the political leadership at the highest level privately the true dimensions of the situation--at least not until the late 1970s. 5.05 Receptivity to Bank mission recommendations was low, as such advice was perceived as undermining the status quo and as against the grain of a socialist course of development. Perhaps, the magnitude of growth and efficiency to be sacrificed to achieve the goal of self-reliance by subor- dinating economic calculus to bare socio-political objectives had not been fully fathomed and, as a result, it was not emphasized strongly enough so as to impress on the responsible leaders its colossal socio-economic costs and thereby force a reassessment of the situation--at least not until the early 1980s when things had gotten out of har.d. And lack of understanding of the impact of government economic policies can be a powerful reason for the continuation of inefficient policies and inertia. Perhaps, Bank sector reports should make reference to the lessons 'f experience--especially - 88 - failures--from the industrialization effort of other countries with similar socio-political regimes which have employed analogous policy instruments and institutional arrangements to make the point.97 5.06 In this regard, it should be noted that the composition and person- al idiosyncracies of staff on missions (and of resident missions) played a role in assessing situations and in devising actionable institutional and policy reform recommendations. However, the endorsement and follow-up of their findings was a conscious Management decision, often colored by per- sonal, socio-political and bureaucratic expediencies and interpretations. Thus, while economic and sector reports, particularly in the 1970s and the 1980s, made proposals and suggested specific courses of action, these were not fully endorsed or pursued vigorcasly by the Bank's Management. The optimism and reassuring attitude of tie missions during the 1970s--in con- trast with the 1980s--apparently reflected more the biases of the Bank's Management (e.g., as witnessed in the CPPs) and less the thrust of their findings. VI. BANK LENDING FOR INDUSTRY AND ITS EFFECTIVENESS The Lending Program 6.01 The Bank's investment in industry began in the mid-70s and faded out by the early 1980s (see Table 1 below). Total direct and indirect (through DFCs and technical assistance) lending to industry during FY74-FY83 amounted to US$250 million, of which US$80 million (32Z) in IDA funds, accounting for about 82 of Tanzania's industrial investment during 1967-87. It supported five operations involving one state-owned (TIB) and one foreign government agency controlled (TDFL) financial intermediaries, four resource-based industrial projects, and two technical assistance credits to finance pre-investment and special studies, and an assortment of training activities. Three of the indastrial projects (pulp and paper, shoe factory, Morogoro textiles) were large greenfield facilities, while the Mwanze textile project involved the expansion of an existing plant. The total investment cost of the industrial projects amounted to US$540 million, of which US$161 million was the Bank's contribution, US$250 971 For instance, concerning the serious problems encountered with holding companies and the entire structure of controls over parastatals in other developing countries with socialist leanings see: World Bank Report No. 6080, Bangladesh: Review of the Experience with Import Program Credits I-X, February 24, 1986, paras. 3.02, 3.46-3.49, 4.02, 4.29, 4.30; World Bank Report No. 1975a, Syrian Arab Republic: Development Prospects and Policies, February 22, 1980, Vol. I, paras. 2.13-2.16, Vol. III, Annex 3; World Bank (draft) Industrialization in Egypt: Performance and Policies, October 1984, paras. 5.22-5.30; World Bank Report No. 6305, Egypt: Public Industrial Enterprise Efficiency ItSd., July 18. 1986, Vol. I, paras. 2.01-2.107. - 89 - million bilateral and suppliers' credits, and the remainder local financing. Total DFC and technical industry-related assistance lending reached US$72 million and US$17 million, respectively. Table 1 INDUSTRIAL SECTOR LENDING BY BANK, 1961-87 (Approval*: USS Million) Total 61- 84- Project Direct FY*: 78 74 75 76 77 78 79 8 81 82 88 87 Total Investment Mwanza TextlIle 15 15 8 Morogoro Industrial Complex 28 28 40 Morogoro Texti les ( 25 46 111 ( 26. Muindt Pulp and Poper (80 18* 78 851 ( 86* Sub-total 181 640 Tech. Assistance I 6* 6 Tech. Assistance II 11* 11 Sub-tota1 17 Indirect (DFCs) TDFL 11 11 TiB 6. 15 16 25 61 Sub-total 72 Total Lending -6 15 21 286 6 71 26 11 - 18 - 250 *IDA Credits. Lending to Financial Intermediaries98 6.02 TIB's establishment in 1970 aimed at channelling all public indus- trial investment through a single, state-controlled banking institution, thereby ensuring adequate vetting of projects, independent judgment on the 98/ For details see OED, PPAR No. 3881, Tanzania - TIB I (Credit 460), March 30, 1982, and OED, PPAR No. 7744, TIB (Loans 1172/1498/1750) and TDFL (Loan 1745), May 4, 1989. - 90 - technical, financial, and economic merits of the projects submitted for financing, and appropriate design of financing packages.99 In supporting TIB, the Bank sought to strengthen the institution, particularly in such dimensions as project analysis, policies, procedures, MIS, to transfer resources for industrial investment within the framework of the basic in- dustry strategy, and to build up TIB's capacity to extend technical assis- tance to project sponsors. TIB has made progress in building up its organ- izational structure and appraisal/supervision capability. However, the fact that it operates in a difficult environment has affected dramatically its performance and sectoral contribution. During the critical period of the 1970s, TIB was involved only in some 202 of the new industrial invest- ments, having been adroitly by-passed by investing parastatals which ar- ranged their own financing, largely with external sources (see paras. 3.07, 4.07). In many instances, projects were presented to TIB at a very late stage of preparation, or even after the government had committed funds and equipment had been ordered, when rejection was not possible and changes in scope were difficult to make. Quite a few projects were also financed at the government's behest and risk. Furthermore, since its board has been dominated by high ranking government officials, TIB was not entirely immune from institutional pressures in its own decision-making. Obviously, such ' practices defeated the most important pvrpose of TIB's institution, namely the exercise of an independent quality control function over project in- vestment decisions. 6.03 In general, during the investment spree of the 1970s TIB did not pay enough attention to design and issues affecting the longer-term viabil- ity of projects (e.g., technology, market size, scale economies, management capability, capital structure, import dependence, export potential) as well as to emerging constraints (e.g., foreign exchange shortage, irregular flow of domestic raw materials, wanting infrastructure). On occasion, projects sponsored by parastatals were approved prematurely and without sufficient investigation regarding availability oi financial resources and implementa- tion capability. Cost and demand projections tended to be over-optimistic, as were implementation schedules. Appropriate feedback mechanisms were not developed to ensure that the lessons from earlier operations were taken into account in appraisals of new projects. And it was not until the early 1980s that TIB began to place emphasis on project rehabilitation, moving away from financing new projects and using part of its foreign exchange resources to finance working capital requirements. As a result, TIB's actual impact on project selection and design and, by extension, on the pattern of industrial investment in the country, has been substantially attenuated. iF9/ The establishment of TIB, as well as of sectoral holding companies in the early 1970s, also meant to curb NDC's dominance whose effectiveness had been questioned. - 91 - 6-04 Bank appraisals were unjustifiably upbeat and reassuring, asserting that TIB was "in a position to allocate resources to financially sound and economically viable projects."100 And although they voiced concern about the condition of TIB's portfolio since the mid-1970s and acknowledged the existence of general policy and operational problems in the industrial sector which affected TIB-financed projects, they expressed optimism that these were being addressed by the government and that they were not expected to endanger TIB's viability or to prevent it from operating effectively as a financial intermediary.101 Bank appraisal and supervision missions did not question early on TIB's pattern of appraisals and depth of economic analysis, in particular market analysis, sourcing of inputs, and the a3sumptions underlying calculations of economic rates of return, did not address the issue of working capital availability, let alone provide for it, and did not appreciate fully the impact of poor project conception, competent management, and economic policies and conditions on project performance. Apparently, missions did not ascertain early on the full extent of the problems faced by TIB's clients and, consequently, the Bank had no clear perception of their status and TIb's financial condition. The Bank has not been circumspect enough in its subproject review and, as a result, financially and economically unviable projects were approved. Finally, the Bank has not been able to influence TIB's lending pattern, e.g., by suggesting better balance between large and m'dium size subprojects, capital-intensive and labor-intensive, import-intensive and local material-intensive. In the circumstances, the performance of the subprojects financed by Bank funds has been disappointing, paralleling that of the sector as a whole. 6.05 Though marginally profitable on paper, TIB's financial position is not sound, since almost three-quarters of its portfolio is affected by arrears (involving two-thirds of the total number of projects) because of poor subproject performance. TIB has not been in a position to influence the operating efficiency of parastatals, and finds it extremely difficult to collect due to the weak financial position of most clients, resulting from the general deterioration of economic conditions, the excessive and inefficient controls over economic activities, deficient project design, inexperienced management, poor maintenance, technical, local resource and infrastructural problems, and mounting indebtedness due to successive and dramatic devaluations. Quite a few projects in its portfolio are under liquidation and, in view of the severe structural, management and liquidity problems faced by many parastatals, TIB's financial condition can be ascer- tained only after a full assessment of their true financial situation and prospects. TIB's sustainability as an institution remains uncertain as it faces the prospect of a financial collapse in view of the amount of pro- spective write-offs, and is in dire need to rehabilitate its portfolio. 100/ SAR, TIB II, No. 849, October 6, 1975, para. 6.01; SAR, TIB III, No. 1730, November 21, 1977, para. 6.01; PR, TIB IV, No. P-2606, July 12, 1979, para. 51. 101/ PR, No. P-2173, November 21, 1977, para. 47. But this would require, inter alia, weeding out unviable parastatals, resolving the issue of the denominated in foreign currencies debt, finan- cial restructuring, employing expatriate management to man key positions on a long term basis, ensuring access to working capital, resolving pressing industrial, trade, and financial policy issues, and assuring operational independence--a tall order requiring .ourageous political decisions. 6.06 TDFL was established in 1962 under the joint sponsorship of CDC and DEG, with FMO joining in 1965 and EIB more recently. TIB is a minority (24%) shareholder. TDFL supports primarily medium scale, privately owned manufacturing enterprises (702 of portfolio). The Bauk's involvement aimed at increasing TDFL's effectiveness in resource allocation by providing policy, institutional and operational advice, and at transferring resources in support of private enterprises catering to basic needs. As in TIB's case, the difficult economic environment and the alignment of TDFL's opera- tions to the government's industrial strategy and policy framework have had an adverse impact on the performance of the subprojects it financed and on TDFL itself. TDFL's institutional weaknesses, coupled with the acute sys- temic distortions, have undermined its financial position and effectiveness in resource allocation. Although TDFL has made progress in developing managerial and staff capacity, appraisals, supervision and collections have been inadequate. Over three-quarters of its portfolio (one-third among its larger projects) is affected by arrears, while its financial position re- mains critical as it has been experiencing operating losses since 1982. TDFL faces the same set of issues as TIB, i.e. rehabilitation of its port- folio, improving the operating efficiency of its clients, and restoration of its own financial viability. 6.07 While correctly emphasizing institution-building, DFC opetations did not address policy and strategy questions, taking as given the policy environment and supporting import-substituting industrialization as per- ceived and applied by the government. Yet, macroeconomic conditions and policies, as well as the degree of autonomy of the financial interme- diaries, have had an overwhelming impact on the success or failure of sub- projects. There has been no macroeconomic or sector conditionality in the Bank's DFC operations, and lending to financial intermediaries was not used vigorounly as a vehicle to press._for ref9rms on issue' affecting directly enterprise performance (e.g., the pattern of and administrative mechanisms for foreign exchange and credit allocation, cost-plus price setting prac- tices, licensing, infrastructure bottlenecks) and, by extension, the DFCs themselves. Such matters were thought of as best be taken up in the con- text of the macroeconomic dialogue. But the dialogue has been tentative and ineffective during the conception, preparation and disbursement of the loans. Nor have the DFCs been able to pursue on their own effectively these issues with the respective ministries, even though some of their top managers and board members were participating in influential committees (planning, budget allocation) that reviewed and approved parastatal invest- ment programs and allocated resources to individual projects. TIB-client seminars, conducted annually and attended by representatives of the minis- tries concerned, dwelt on these issues but did not elicit action either. Finally, the financial intermediaries have not been able to develop at the I - 93 - requisite level the so crucial in irvestment banking intimate bank-client relationship, that would have enabled them to influence early on project formulation and the operating efficiency of the parastatals. possibly be- cause of a hands-off attitude, not unusual in relations among public enti- ties, and/or resistance by parastatal managers (e.g., because of the per- ception that the lending institution is encroaching on the management's autonomy). Thus, the DFCs' potential influence as financiers was not exer- cised fully. All in all, it emerges that Bank follow-on operations were launched hastily and in disregard of the deep-seated problems afflicting the sector, in the hope that somehow things would turn around. As it turned out, the Bank's objectives largely were not met during its long association with TIB and TDFL, and its contribution to improving investment allocation has been marginal. Lending for Technical Assistance 6.08 The two technical assistance credits were channelled through TIB, and financed consultancy services for the preparation of pre-investment and feasibility studies, productivity and capacity utilization improvement studies, and training in project preparation/implementation, management and related techniques in an effort to strengthen the capacity of local insti- tutions (e.g., parastatals, government agencies, development banks), in- cluding the establishment of a technical assistance unit in TIB. After some initial problems, stemming from the lack of experience by both the Bank and TIB in the management of a complex program, were sorted out, the first credit financed a total of 21 feasibility studies (some involving agro-industry, transport and tourism activities, most of which were includ- ed in TIB's project pipeline), 7 special studies (most of which were imple- mented), and training programs for some 400 public sector managers (ac- counting for 9Z of total disbursements).102 The second credit (still not fully disbursed) has financed to date some 40 feasibility and special stud- ies (mostly in the energy, industry and transport sectors) and training for over 800 managers covering subjects in management, export marketing, and project implementation. 6.09 The rationale for extending technical accistance in these forms was valid. The country's scarce human resources have been spread thin across a wide spectrum of development activities, and the shortage of people trained to formulate and supervise the implementation of productive projects and take up management positions remains acute. Although Tanzania has benefit- ted from a high level of externally financed technical assistance, their focus has been largely outside the industrial sector and more on broad- based institutional support rather than the preparation of new industrial investments, or studies designed to improve efficiency, analyze operational (e.g., technical, management) problems of parastatals, or devise work in- centive schemes. The credits have helped alleviate critical manpower bot- tlenecks and upgrade technical, management and accounting skills, although it cannot be ascertained whether the entire throughput of trainees 102/ For details see PCR, No. 6732, April 8, 1987, Tanzania - First Technical Assistance Project (Cr. 601). - 94 continues to make full use of the skills acquired. TIB has benefited from the training provided to its own staff, and TIB's Feasibility Studies Unit has developed the capacity to prepare terms of reference for consultants to be engaged by public sector agencies, to evaluate feasibility studies pre- pared by consultants, and to assist in providing training to the staff of parastatals. The pre-feasibility studies helped identify some potentially viable projects, and were also useful in identifying risky or non-viable ones. On the other hand, the impact of the special studies on capacity utilization and efficiency improvement has largely been vitiated by the persistent economic difficulties and the results are far from visible. Direct Lending to Industry Textiles 6.10 The two integrated textile projects in Mwanza and Morogoro aimed at promoting the government's goal of self-sufficiency in woven cotton and blended fabrics, based on the country's ample cotton supply. The mills have been in operation since 1979 and 1986, respectively, but their perfor- mance has been disappointing. Both mills have been unable to build up production at the projected levels since cammission, operating at 1OZ of rated capacity in 1988. The Hwanza plant, in particular, has been crippled by persistent shortages of power103 and water, and both mills have been affected by the unavailability of imported spare parts, consumables, dyes and chemicals due to foreign exchange shortage; weak domestic market and stiff competition from legal and "unrecorded" imports; lack of qualified personnel due to high turnover among trained staff, the replacement of highly qualified expatriates in key operational positions by nationals with limited professicnal experience; overstaffing due to low work assignments, lack of operating skills, and social reasons; weak marketing set-up; inadequate working capital due to the credit squeeze; and rising indebtedness due to devaluations. The Mwanza mill has accumulated losses exceeding TSh 800 million while the Morogoro mill TSh 5 billion, and both cannot service their financial obligations. Recalculated financial and economic rates .f return are negative, and the viability of the projects remains uncertain. The objectives of both projects have not been met, and it is doubtful that they will be met in the foreseeable future.104 1031 It is noteworthy, that supply of electric power sufficient to ensure the uninterrupted operation of the Mwanza plant was assured under a covenant in the Loan Agreement, a commitment which the government has not been able to honor. 1041 For details see PCR, Tanzania - Mwanza Textile Project (Loan 1128), OED Report No. 5187, June 29, 1984, and PPAR, Morogoro Textiles Project (Loan/Credit 1607/833), forthcoming. - 95 - 6.11 At the time of the appraisal of the Mwanza project (November 1974), the Bank had impressed on the government the need for undertaking a market and marketing study to establish the long-term potential for cotton fabrics before embarking on another project.105 Because of resource constraints, shortage of trained technical staff, low productivity, and high production costs of existing mills which presented serious obstacles to launching a major export drive, it was suggested that the government would be well advised to develop the textile industry at a more realistic pace. In the ensuing years, and in addition to the Bank-sponsored Morogoro textile mill, three other cotton textile projects were undertaken by TEXCO (the holding company) with no apparent regard to the excess capacity being built and without consulting with the Bank. At present, annual demand for textiles is estimated at around 140 million meters,106 while the existing production capacity is about 246 million meters and actual production 75 million meters. This suggests a capacity utilization rate of 30%. It is noteworthy, that 75Z of the production capacity is less than 10 years old and cannot be viewed as outdated. With total annual production in the past few years limited to only 75 million meters, the unsatisfied demand of some 65 million meters was largely met through legal and illegal imports which, aside from the low level of production, have been further encouraged by the high domestic prices due to hIigh costs of production and heavy sales taxes. 6.12 Aside from the fact that the Bank's demand projections were opti- mistic, Bank appraisals (and the government) neglected to consider the alternative of rehabilitating the four older mills under TEXCO (Friendship Textile Mills, Tanganyika Dyeing and Weaving (SUNGURATEX), and Kilimanjaro Textiles (TILTEX)--2 plants in different locations) before proposing the installation of new plants (replacement of machines and equipment, balanc- ing, and rationalization of processing facilities). The more so, since by the mid-1970s it was quite evident that the economic situation had been deteriorating and that the difficulties of the textile sector were com- pounding.107 Such an approach would have been more economical, saved foreign exchange, made the existing mills more competitive, and avoided the 105/ Rapid expansion of the cotton-based textile industry was a key element of the basic industry strategy. World Bank, 1975, Vol. III, Annex IV, para. 3. 106! Consumption of textile fabrics has grown at a substantially lower rate than projected in Bank appraisals, even allowing for the impact of the deterioration of economic conditions on demand. 107/ World Bank, 1975, Vol. III, Annex IV, pp. 1-21. Reflecting Bank concerns, a dated covenant was included in the legal agreement of the Morogoro project obligating TEXCO to carry out a study of the productivity and capacity utilization of all its operating mills, and see to the implementation of its findings. The recommendations of the study were partially implemented, but the impact cannot be assessed due to the adverse effects of factors beyond management control which have severely influenced individual mill performance. - 96 - creation of excess capacity. Also, given that scale economies in textiles beyond the minimum economical size, which is relatively low, are marginal and the severe constraints in human skills and infrastructure, no consider- ation was given during appraisal in building smaller plants with provision for phased expansion. This would have avoided straining scarce resources, including foreign exchange, and would have made the management of these plants more wieldy. Furthermore, since the country was facing foreign exchange problems, as is witnessed inter alia by the change in the scope of the Morogoro project to enable production of all-cotton fabrics, one won- ders why the Bank, in sponsoring the Mwanza and Morogoro projects, did not insist on their partial export orientation to take advantage of the abun- dance of good quality cotton, modern technology, expatriate management, and lower wages, while ensuring higher productivity by supplemental training of a selected group of workers drawn from existing mills under TEXCO's con- trol, introduction of worker incentives, and an export subsidy to compen- sate for the unrealistic foreign exchange rate and price distortions. 6.13 With regard to institutional development, production management systems (e.g., emphasizing process and cost controls, conservation/recy- cling of energy, quality control, etc.) have been weak in most firms under TEXCO's control, while TEXCO has provided limited guidance on improving production planning and management. The sub-optimal and low value added product-mix in most enterprises attests to significant weaknesses in mar- keting. In the absence of a feedback and intelligence system, most enter- prises continue to produce run-of-the-mill items for local consumption, including the modern mills which are in a position to produce superior quality fabrics for domestic and, potentially, for export markets. More- over, there has been no serious effort to upgrade quality, while the mills make their own procurement and marketing arrangements (though not always without external interference). Finally, concerted efforts at training, human resource development, and succession planning leave much to be de- sired. These observations cast doubts on the role of the holding company concept in advising, coordinating and directing the entities under its control. For its part, the Bank also has not been able to influence per- ceptibly the pace of growth, reorientation and institutional development of the textile parastatals, albeit in part due to lack of receptivity. Pulp and Paper 6.14 The Mufindi integrated pulp and paper project, with a capacity of 60,000 tons per year of various grades of paper and board, was justified on grounds that it was given high priority by the government, as involving the processing of a domestic rerource and catering to basic needs; that it would provide a dependable supply of paper to the country at stable and reasonable prices; and that it would save foreign exchange. The mill began production in 1985, and is currently operating at 492 of its designed ca- pacity. The reasons for its poor performance to date include optimistic demand projections, incorrect estimates of the composition of product de- mand, weak domestic market, product variation in terms of moisture, weight WI and grade affecting quality, oversights in mill design, transport bottle- necks, 'nadequate maintenance due to shortage of spares, and insufficient expatri-te sup-ort in key positions. The mill has accumulated substantial losses, is seriously ..iort of working capital and unable to meet its cur- rent obligations, and its financial position is precarious. A new injec- tion of some US$40 million is urgently needed for investments to improve the operating efficiency of the mill and for the financial restructuring of the company.108 6.15 It is notable, that annual consumption of paper products at ap- praisal was only 23,000 tons and that for the most part of the 1980s, even allowing for domestic market growth, the project would have had to export a good part of its production at a price below production costs because the small size of the plant, even under ideal operating conditions, would not make it competitive at world prices. As scale economies are significant in pulp and paper production.109 a minimum economic size plant in Tanzania would have to be much larger than could be justified by domestic demand fo. many years to come. In grappling with this issue, the final configuration and scale of the project reflected significant modifications in concept, including reducing the size and/or the number of products produced, to reach an "optimum" between project cost and complexity on the one hand, and economic benefits on the other. Yet, as the project was to have the capa- bility of supplying the whole range of the paper needs of Tanzania. "of necessity, this has meant that diverse and sometimes complex processes have been included. Moreover, the cost of chemicals, most of which will need to be imported, and environmental considerations, have required that the mill be equipped with full chemical recovery and effluent treatment facilities. This has led to a mill concept which is expensive in relation to the econ- omy of the country, and to the scale of production."110 Furthermore, it was acknowledged that, given the very large size of the project and its 108/ There is need to improve the product mix and quality (installation of process computers, increase of bleaching and sheet cutting capacity), to remove bottlenecks (increase of roll wrapping capacity, improvement of roll handling system), to procure materials handling equipment, to replace worn out logging equipment, to erect warehousing facilities, to provide f--r townsite development, etc. 109i See M.A. Amsalem, Technology Choice in Developing Countrits: The Textile and Pulp and Paper Industries, MIT Press, 1983, pp 111-6; M. Roemer, "Resource-based Industrialization in the Developing Countries: A Survey," Journal of Development Economics, 6 (1979), pp. 177-9, and R. Skarstein and S.H. Wangwe, Industrial Development in Tanzania: Some Critical Issues, p. 150, and references cited there. 110/ SAR, Tanzania - Mufindi Pulp and Paper Project, No. 1029, December 14, 1978, para. 5.02 (emphasis added). For more details see also Ibid., para. 5.03. modest economic rate of return, the risks in proceeding were significant.111 Internally, the Bank's management justified the project by arguing that: "development of this project was Justified; only its appropriate timing was in question. The project is consistent with the Government's Basic Indus- try .trategy and will lead to significant foreign exchange savings through import substitution...in view of the limited absorptive capacity of other priority sectors and the lack of other industrial project alternatives, it was appropriate from both the point of view of the Government and the Bank to commence a large-scale industrial project such as this at this time. No pipeline of incremental smaller industrial projects with higher priority existed." (Decision Memorandum, 11/811977, emphasis added). This suggests that the rationale for supporting such an inordinately high capital and skill intensive project which defied basic economic and market considera- tions was shallow, and that the Bank's technical staff was compelled to come up with a project design reflecting major technical and economic com- promises which seriously affected the viability of the project. Indeed, the recalculated economic rate turned out to be unsatisfactory. Clearly, in the Tanzanian circumstances and strictly on project grounds, the project should have never been initiated. Institutional pressures to lend, rein- forced by a government request and pressure from a donor community eager to assist, are the only plausible explanations for the Bank's involvement in this inefficient operation. Leather Products 6.16 The Morogoro industrial complex comprised the infrastructure for a small industrial estate, a large footwear factory, a small plant to manu- facture other leather products which were supported by the Bank, and a tannery and a canvas plant financed through other external sources. The basic objective of the project was to use domestic hides, skins and cotton, which were exported unprocessed, to produce finished products of high qual- ity for export markets, and thereby increase employment and foreign ex- change earnings. The main component of the complex was the shoe factory, that would produce 4 million pairs a year of which 80% for exports mostly to Europe. World market demand was estimated to be so large and growing that the key factor for effective market penetration (aside from quality and price) was the development of a successful marketing strategy. The appraisal foresaw no major problems in establishing external commercial relationships, and felt that the modest increments to world supply from the 111/ PR, Loan/Credit 1650/875, Tanzania - Mufundi Pulp and Paper Project, Report No. P-2436, December 20, 1978, paras. 43-49, 76. The project was co-financed with CDC, KfW, KF, NIB, OPEC Special Fund, and SIDA. In addition, IDA financed a Technical Assistance and Energy Conversion Project (PR No. P-3546, May 2, 1983) to provide for expatriate management services for six years, woodharvesting and transport equipment, and imports of pulp, chemicals and other consumables, and the conversion of the oil/coal power boiler into fuelwood using. - 99 - project would be easily marketable. By Board presentation, the project sponsor had selected a "leading world distributor" with "proven" marketing capabilities in international shoe inarketing.112 6.17 The shoe factory, commissioned in 1980, has never operated at more than 4Z of its capacity (in the last several years it has been virtually idle), and has never developed export capability. A number of factors account for its poor performance. In the first place, the crucial omission was that a joint venture ensuring both experience in export marketing &nd a stake in the enterprise--a condition sine gua non for the success of an undertaking in luch a highly competitive industry--was not formed early on during project preparation. Inability to involve a capable and willing partner from the very beginning in the project should have signalled to the Bank that further efforts in designing and preparing the project in its original export-oriented configuration must be discontinued. Despite the fact that the necessity of a joint venture operation with equity participa- tion by the party responsible for marketing had been recognized early on (Decision Memorandum, December 9, 1975), in the process, the importance of the issue eluded the Bank, and project preparation proceeded even though the project could not elicit interest. In the end, the simplistic view was taken, despite the fact that the project sponsor was inexperienced in shoe making and export marketing and was relying entirely on the Bank's guid- ance, that contracting a company experienced in management and marketing, even without a stake in the venture, would be sufficient to ensure success in the export drive. 6.18 As it turned out, the contracted firm did not have the experience they claimed they had, the Bank having failed to investigate sufficiently the firm's assertions regarding their experierce with implementation of shoe making plants and international marketing contacts. The consulting firm's inexperience in operating plants of the Morogoro's size, coupled with inadequate training, acute shortage of experienced operators and middle-level managers, and poor product design and quality, production costs were well above world prices and the product was not marketable. Furthermore, the contract that was signed recorded only intent and did not stipulate penalties or incentives for export performance. No wonder the managing firm had no interest in increasing production and promoting ex- ports, and the contract failed to achieve the objectives it sought. Final- ly, the possibilities of successful penetration of international markets had been over-optimistically assessed. During project preparation only a very general worldwide market analysis was made, but no specific export outlets had been identified. 6.19 The project concept and design were excessively ambitious--4 mil- lion compared to 500,000-1,500,000 pairs in developed countries, suggesting the existence of marginal scale economies. In view of the difficulties in developing export capacity in the surrounding circumstances, it is striking 112/ PR, No. P-2014, Morogoro Industrial Complex, March 15, 1977, paras. 69, 76, 77; Loan Agreement, April 6, 1977, Section 3.04(a) and (b). that no thought was given to designing a smaller project to serve initially the domestic market and build up ezport capability gradually. Also, the professional competence of the holding company as well as of the parastatal operating the plant was thin, so much so that no efforts were made over the years to attempt to capture at least a share of the domestic market, or come up with a solution to utilize the equipment (e.g.. through leasing). Finally, the appraisal brushed off the issue of the quality of leather rn . grounds that exports were being made; failed to investigate major (and known) problems with the branding, flaying, drying, preparation, transport, storage and marketing of hides and skins which, given the quantities rs- quired for the operation of the shoe factory at full capacity, would have created severe shortages of high quality leather; and paid attention neith- er to the adequacy of leather supply, despite the iact that the country's herd numbers and growth rates had been overestimated, nor to price con- trols, which have led to extensive unrecorded exports of hides and skins to neighboring countries. 6.20 In sum, the shoe factory was an over-dimensianed project, built in an environment lacking physical, technical and managerial infrastructure. Given its magnitude and the risk involved, the feasibility of the project did not receive the attention it deserved. And in the absence of a partner experienced in shoe manufacturing and export marketing, it should have never been undertaken. Ultimately, the operation of a new export-oriented project was entrusted to a parastatal, and in a country, where there was no experience in the type of the product to be exported--a product destined for a highly competitive and sophisticated market where superior quality, and keeping up with ever-changing fashion, style and design are critical elements of success. As would be expected, !inancial and economic rates of return for the complex are negative. Under the present management struc- ture, know-how, market conditions, cost structure, availability of raw materials, and caliber of operating staff, the likelihood of the now virtu- ally idle (1% capacity utilization) shoe factory operating as it was origi- nally platined and on a financially and economically viable basis is remote. There can be no question that the Bank exercised poor judgment in going ahead, and in such scale, with this project.113 1131/ The performance of the other two components to date does not allow for optimism concerning their future financial viability either. The proposed plant for leather goods was a small-scale operation which, it was felt at appraisal, did not warrant carrying out a detailed market survey. It ne er took off due to poor management, and is currently vegetating, producing small quantitiis of tarpaulin and other canvas products for the army. On the other hand, the organizational structure and management of the industrial estate leaves much to be desired, while utilization of the standard small-enterprise buildings in the estate has been modest. For more details on other dimensions of the project experience see PPAR, Morogoro Industrial Complex Project (Loan/Credit 1386/1385), forthcoming. Effectiveness 6.21 The Bank's lending strategy to industry was based on three con- siderations: Tanzania was one of the least developed countries in the world; the country's leadership had shown a high degree of commitment to growth and development; and Tanzania faced an enormous resource gap. The emphasis of the Bank's direct industrial lending in Tanzania has been on support of specific investments in resource-based, import-substituting industries within the precepts of the basic industry strategy and the goal of self-reliance. The projects supported would rely on important domestic raw materials largely exported unprocessed, generate value added, create job opportunities, and save foreign exchange by substituting for imports. The link between industrial sector work and lencling operations has been somewhat tenuous, particularly concerning policy issues and their implica- tions for the operating efficiency and financial viability of the projects. In designing industrial projects, the Bank did not conqider more economical and more suitable for the borrower's conditions alternatives, and did not always respect fundameatal technico-economic principles concerning scale of plant. The genuine need to support the projects financed by the Bank (i.e. whether the project was justified because of its benefits and not just because it supported the government's industrial strategy) was not always established convincingly. This suggests that, in many instances, the transparency of Bank appraisals can be questioned, in the sense that lend- ing in the industrial sector was often dictated more by institutional pres- sures under the guise of much needed resource transfers and less by the project's economic merit. Co-financing by eager donors apparently rein- forced such pressures. Finally, in most projects the Bank's supervision effort has been perfunctory and of questionable effectiveness. 6.22 ':.- ank-supported industrial projects have not met performance expectati_:.s due to a confluence of factors, including the adverse policy environmpnt, poor design, difficulties in implementation, weak markets, low capacity utilization, insufficient managerial and labor skills, and their impact on industrial development has been marginal--at best. The Bank did not introduce macro- or micro-economic conditionality on policy issues affecting parastatalr, pricing in particular. Rather, the Bank extracted vague promises that the government would ensure that parastatals earn suf- ficient revenues to cover all their costs, service their debts, and earn a reasonable return on their invested capital (e.g., Memo 5/15/78, Morogoro Textiles). The severe and persistent bottlenecks in infrastructure (power and transport in particular) which have afflicted its own projects seem to suggest that the Bank has not been able to achieve a proper balance and inter-sectoral coordination in its lending operations. The Bank's impact on investment through indirect lending to industry has not been tangible either. Lines of credit through financial intermediaries in support of the basic industry strategy also ran into difficulties, largely associated with problems encountered by sub-borrowers and in part due to inadequate ap- praisals. Although top management in both institutions was knowledgeable and experienced, their efforts to steer their respective inistitutions on the right course during these difficult times have been frustrated in a large measure by compelling exoganous factors. Indeed, the fact that TIB and TDFL operated in a difficult environment has affected dramatically their performance and sectoral contribution. These problems, coupled with internal weaknesses of the intermediaries, have limited substantially the success in achieving the original institution-building goals, while the financial position of the institutions is %:ritical and their sustainability remains uncertain. In short, dll Bank-financed projects share the fate of other parastatals, and are in dire need of rehabilitation and/or financial tastructuring along the lines suggested in para. 6.05, in fine. On a less tangible plane, since the Tanzanians were looking up to the Bank for guid- ance and advice concerning project identification and design, the poor outcome of particular proje.s and the ensuing disappointment have cast doubts on the soundness of the Bank's counsel, and have undermined the trust in the judgment of the Bank's technical staff. VII. EVALUATION OF BANK'S APPROACH TO TANZANIA'S INDUSTRIAL DEVELOPMENT Strategy and Policy Dimensions 7.01 The Bank's position on Tanzania's strategy for industrial develop- ment has evolved over the years, reflecting the change in the Bank's lead- ership in the late 1960s, and again in the early 1980s, and the attendant philosophy toward economic development. The Bank proposed a strategy just prior to Tanzania's independence which was officially adopted by the new government in 1961. But alreLdy by the mid-1960s, and before the Arusha Declaration in 1967, the Bank took a reactive stance and, until the early 1980s, has been supportive of the strategies propounded by the political leadership. 7.02 The thrust of the Bank's position and advice at the time of Tan- zania's independence was that emphasis should be placed on economic growth, to be sustained by a circumspect and circumscribed import-substitution industrial strategy perforce heavily conditioned on agricultural growth and exports, private initiative, moderate and of limited duration protection based on sound economic analysis, expatriate management and skills, exter- nal funding, more positive attitude toward non-Africans, and improved in- frastructure. This approach was dictated by the confluence of potent con- straints to industrialization, not least being the small size of the domes- tic market and limited skill availability, and the limited substitution possibilities between domestic and foreign-produced inputs, producer goods in particular. In retrospect, this appears to have been a realistic and sensible approach to have followed. laproved methods in agriculture, say using fertilizers and machinery, could be as powerful generators of induced I investmen in the chemical and engineering industries as any other manufac- turing activity. As incomes and markets grow over time, ever. smaller coun- tries can invest in producer goods industries that achieve scale economies. Skills can be developed and there is a distinct, albeit slow, process of "learning by doing," which can bring reasonably protected "infant indus- tries" closer to world competition. But the proceis cannot be unduly stretched and accelerated with impunity. 7.03 The 1954-69 Plan, unlike the 1961-64, stressed the need for struc- tural transfnrmation of the economy to correct organic disequilibria and broadened thd scope of import substitution to include intermediates, usher- ing in an era of irrational import substitution under a misguidedly protec- tive system. The Bank did not appreciate at the time the significance of tnis turnabout, which emphasized accelerated industrial growth and played down the role of agriculture- -an approach which was contrary to its own advice only a few years earlier.114 Despite forceful constraints to indus- trialization which had been highlighted not long before and the serious implications of the Plan's new direction and focus, the Bank did not voice concern over tAe growing signs of impatience shown by the political leader- ship. 7.04 The Bank accepted Tanzania's road to socialism enunciated in 1967 as a fait accompli, and did not challenge the leadership's industriali- zation strategy--its concept, its underlying premises, its potential for success in the surrounding circumstances. And later, in the mid-1970s, when a somewhat more articulate longer-term basic industry strategy was develoned by the government, the Bank hastened to endorse the strategy and continued to confine itself largely to following up on its implementation, pointing up deficiencies, and suggesting corrective action. In fact, all Bank industrial lending has been justified in support of the basic industry strategy. The mission's report, issued in 1967, limited itself to high- lighting :he dangers of any significant withdrawal of non-African capital 1141 The importance of a progressive agriculture for initiating and sustaining the industrialization process, particularly at the early stage of a country's development, cannot be over-emphasized. The following remarks are pertinent on the issue: "What happened, or prevented, the process of industrialization outside a group of favored countries?...If one is looking for some general cause which is common to most countries...it is the ba.kwardness and stagnation of agriculture...the growth of the agricultural surplus is an essential condition for providing the growth of purchasing power necessary for sustaining industrial expansion. The increase in the demand for manufactured products cannot be wholly self-generated...It is no accident that...the 'agricultural revolution' historically preceeded the 'industrial revolution '" N. Kaldor, Strategic Factors in Economic Development, (Ithac.i, New York: Cornell University, 1967, pp. 55-57, emphasis added). and entrepreneurship, the prevailing uncertainties in the investment cli- mate and their potential impact on private investment (availability of risk capital in particular), the shortage of high-level personnel due to rapid indigenization and the need for retaining/recruiting competent management and technical personnel for the nationalized undertakings and planning units, the expediency of clarifying the government's position on the future role of the private sector, and the need for greater participation of pri- vate enterprise and investment (e.g., by not insisting on majority public participation). Despite the unambiguous ideological sway, the sweeping declarations in favor of public ownership of the major means of production, and the ambivalence on the future role of private initiative, the Bank apparently entertained the hope that the leadership will appreciate the potential contribution of the private sector to the country's industrial development within a "mixed" economy, and exhorted the government to con- sider such endorsement.115 7.05 Since 1967, and at least until 1980, the Bank has been viewing Tanzania as coming close to being a model developing country, in the sense that the leadership was committed to economic development, had maintained a climate of political stability, and hac, given the country a clear sense of direction with the long-term perspective of creating an egalitarian soci- ety. This belief was shaped by Tanzania's stated social and economic poli- cies which were to be guided by three fundamental objectives: (i) the achievement of a participatory, decentralized socialist economic order; (ii) the eradication of absolute poverty and progress toward greater income equality; and (iii) rapid, long-term economic growth, sustained by high rates of growth of industry, with full participation of all regions and population groups in the development process. The Bank considered these development policies and priorities as well conceived, fully thought through, and worth supporting (CPP 1973, para. 36). Considerations of "need, of commitment to what the Bank stands for, and of desire for close cooperation with the Bank to achieve these goals," it was felt, justified iull Bank support (CPP 1975, para. 4). In restructuring the political, economic and social life of the country, the Tanzanian leadership was cred- ited with having introduced impressive aLd far-reaching institutional re- forms, demonstrated its dedication to economic and social justice, and willingness to undertake ambitious programs in pursuit of its goals. Mis- givings of successive missions concerning the effectiveness of the planning mechanism, the workability of industrial policies, and the viability of an array of industrial projects initiated in the late 1960s-early 1970s did not evoke sober Bank reaction. CPPs (1973, 1975) could state that there has been no tendency to waste resources in ill-conceived or over-capital- ized projects. 7.06 By 1976, the Bank continued to be impressed by Tanzania's continu- ity and stability in political structure, goals, and leadership. The ob- jectives of social equality, self-reliance at local and national levels, and economic and social transformation were still unexceptionably extolled. Contrary to earlier pronouncements, the Bank slighted a decade of poor 1151 World Bank, 1967, Vol. III, Annex B, paras. 73-77, 80, 81, 92-96. economic performance arguing that "economic growth has been an important but not an overriding goal; the leadership has been willing to forego short-term income gains for longer-term structural change or more equitable distribution" (CPP, para. 7). Not until 1978 would the Bank state that "while the leadership has been willing to forego short-term income gains in pursuit of long-range structural change and/or more equitable distribution of income, it has recognized that attainment of these objectives is depen- dent on performance with respect to more narrowly defined economic objec- tives" (CPP, para. 7). And while admitting that "Tanzania had apparently not lived up to expectations" (Memo, 12/23/77), the Bank concluded that "the increasing pragmatism in Tanzania, observed across a wide variety of issues (ministerial choice, the role of the private sector, the increased emphasis on managerial responsibility, etc.) could make an important con- tribution to ensuring improved economic performance within the "African Socialist" framework established by its leadership" (CPP 1978, para. 51). Still in 1980 (CPP, para. 78), the Bank determined that "Tanzania's basic development objectives are sound but it is necessary that a more appropri- ate set of policy means be developed." 7.07 It was not until 1981 that the Bank acknowledged that the situation was getting out of hand. The CPP issued during that year stated that "the structural weaknesses of the economy are closely interwined with the devel- opment philosophy of the Government" (CPP, para. 14), and that "Tanzania is currently facing the worst economic crisis of its history. While the situ- ation has been exacerbated by the impact of external shocks, these are weaknesses of a more long-term nature in the structure and performance of the domestic economy. In addressing these problems, the Govermm nt will have to make a frank reassessment of its development strategy, especially in the areas of public investment and expenditure, incentives, trade and exchange policy, and government involvement and regulation of economic activity. Further, the sectoral emphasis of government policies and pro- grams will need to more clearly reflect the importance of agricultural production, especially in the smallholder sector" (CPP, para. 61). This belated recognition (if not admission) of the existence of a chronically ailing economy casts doubts on the transparency of the Bank's decision- making proces3, the logical integration, consistency, and intelligibility of the communication system along the successive tiers of the managerial hierarchy, and the effectiveness of the presumably forthright CPP as a management tool (see para. 5.06 above). 7.08 It is arguable whether the espoused Basic Industry Strategy, in whatever variation, could have been efficiently implemented at the pro- jected pace, considering the demanding requirements of BIS in terms of planning and administrative capacity, institutional arrangements, organiza- tion, management, technical skills, foreign exchange, disciplined and moti- vated labor force, etc.; the importance of scale economies and optimal plant scale; the conflict that encouragement of basic heavy industry (e.g., steel, chemicals, heavy machinery) entails on the goal of appropriate tech- nology; the questionable notion of availability of efficient, easy import- substitution industries; the anti-trade bias and disregard for the gains - 106 - from specialization; the lack of an indigenous managerialltechnical capa- bility to run efficiently parastatals; and the difficulty to devip) and implement appropriate arrangements to safeguard managerial autonomy in a highly centralized environment. Developments since the mid-1970s, which have vitiated the implementation of its first largely diluted phase, have brought out forcefully the impact of these constraining factors.116 7.09 It is noteworthy that no effort was made to dissuade the leadership from embarkii.g on ambitious industrialization programs when fundamental preconditions for such undertaking identified by the Bank earlier were lacking; resources in abundance--land and labor--remained underemployed and their utilization dictated a different development strategy; and emphasis on industry required human capital and cultural qualities of non-existing caliber, a planning apparatus which placed heavy demands on organizational and institutional structures, and administrative talent well beyond the local potential. And all this at a time that a policy of indigenization was pursued in full force. Not .,nly was undue and premature emphasis placed on industry, but the wrong industries were promoted and at an un- justified speed, reflecting growing impatience and unfounded aspiration to telescope the development process from centuries to decades. Yet, the Bank appeared reluctant to operate in the forefront and counsel on the impera- tive of a more pragmatic industrial strategy; instead, it endorsed the government's blueprints, suggesting accommodating adjustments, some lacking practicality (see para. 7.11) given the established politico-economic framework (e.g., discouragement of state involvement in economic activi- ties, encouragement of private initiative), while bearing witness to the rapid disintegration of the sector. 7.10 The Bank remained an apathetic observer and an impotent bystander of unfolding developments as Tanzania became the testing ground for an allegedly "unique" socialist experiment--and one that was poorly thought out at that. In the absence of a proactive attitude, the Bank has not been able to influence the investment program in the industrial sector, in terms of inter-sectoral linkages, subsectoral distribution, choice of technology, capital-intensity, optimality of scale, local sourcing and import depen- dence, export orientation, balance between public and private sectors, even when it was advising on economic grounds against the undertaking of partic- ular large projects. Thus, project selection and design was unwittingly relegated to the parastatals, the whims of the bureaucracy, and the suppli- ers of equipment. The three-tier project evaluation process that was in- troduced proved ineffective, aL scrutiny was biased by bureaucratic expedi- encies and interventions. Many industrial projects virtually evaded pro- fessional vetting either because investing parastatals adroitly by-passed the scrutiny of financial intermediaries by arranging their own financing, or presented their projects at a very late stage of preparation or when equipment had already been ordered and only for complementary financing. On the other hand, financial intermediaries, though supported by the Bank and other multilateral and bilateral agencies, did not pay sufficient at- tention to project selection and design, and failed to consider factors 116/ For an elaboration see Appendix, paras. 7-11. affecting tha longer-term viability of the ventures they supported in the face of changing economic conditions and mounting constraints. Socio-po- litical considerations and institutional pressures also have colored their decisions in financing industrial undertakings. And since the local capac- ity to screen projects has always been inadequate, inadvertently even proj- ects supported by donors (usually of fairly large scale, capital-intensive and import-intensive) were implemented without due attention to their eco- nomic merit. 7.11 The Bank did noL grasp fully the inherent ideological contradiction betw4en the espoused socialist pattern of development and the appeasing official pronouncements concerning the role of private sector as perceived in Tanzania. The nationalization decisions announced on February 6, 1967, were a reaction against the growth of private business which the same po- litical leadership had helped to foster earlier. Evidently, it reflected a growing concern on the part of the leadership about the increasing domi- nance of the private sector and of the AsiLn and European interests so largely identified with it. And to ensure consistency between precept and practice, a code of ethics was promulgated calling upon "leaders" (e.g., political leadership, civil servants, parastatal employees, etc.) to avoid involvement in "capitalist practices." Because all citizens were induced to remain or become TANU members, this effectively meant that no existing (or would be) Tanzanian entrepreneur could own or be a director of a pri- vate company other than a small farm or a small shop. The Acquisition of Buildings Act of 1971, which accelerated the emigration of Asians, further undermined the official position on the potential role of the private sec- tor. Therefore, Bank efforts to impress on the government the virtues of private initiative, an institution which was perceived by the authorities as inimical to achieving their socio-economic objectives, though sensible, were not likely to elicit a favorable response and open support.117 The more so, since the chances for a speedy emergence of an indigenous entrepreneurship were remote. The same holds for the Bank's advice that the government reduce its pervasive involvement in and regulation of the economy. Even if injudicious, active state participation and control were quintessential elements of the leadership's chosen path of development. 7.12 In the same vein, the Bank did not appreciate fully that the speed of indigenization, given the goal of rapid economic and industrial growth, was overly ambitious, that organization and human and physical resource endowment were severely constraining the country's absorptive capacity, and that external technical assistance and training were poor substitutes for indigenous managerial talent and in situ experience. The Bank never ques- tioned openly the thrust and speed of industrial development and the insti- tutional set-up that was put in place to implement it, in effect identify- ing industrialization with development, and reverting on its earlier posi- tion and advice that de-emphasized "forced", rapid, import-substituting 117/ The views of the government were clearly reflected in the following quotation from a Bank sector report: "Tanzania's political leadership firmly believes that the system of a private enterprise economy is incompatible with the country's desire for genuine political independence and social equality. It bases this belief on the conviction that a private enterprise system--because of Tanzania's poverty and shortage of entrepreneurial skills--would inevitably lead to dependence on private foreign investors and multi-national corporations which would entail an increase in foreign political influence, the consolidatioA or creation of elitist social structures and a pattern of development that is contrary to the interest of the majority of the people in Tanzania." World Bank, 1975, Main Report, para. 11. No wonder that the "delineation of public and private sector roles has lost much of its clarity since 1967" (Ibid., para. 18), that uncertainty as to the future role of private investment persisted, that the official pronouncements inviting private initiative to participate in the industrialization effort remained unconvincing, and that the Bank's exhortations (Ibid., paras. 18, 20, 21 and elsewhere) remain unanswered to date, while a scheduled small scale industry operation in the Bank's lending program never materialized. I - jVV, - industrialization. In essence, the Bank has inadvertently shared the po- litical leadership's excessive optimism, impatience, and tendency to play down the counteracting effects of potent constraining factors.118 7.13 The possibility of pursuing a gradualist approach to industriali- zation (and economic development) apparently was never considered either by the political leadership or the Bank.119 From the mid-1960s to the late 1970s, the government made an all-out effort to forge ahead haphazardly with investments in producer goods industries at a hardly warranted pace and without the benefit of an economic calculus, under the rubric of basic in- dustry strategy. Yet, the daunting preconditions for successful implemen- tation of the BIS with its emphasis on promotion of heavy producer goods industries eluded the political leadership, the government's advisers, and 118/ It is seldom appreciated that economic and social development follows, or at best parallels, the pace of cultural change, which in turn depends on the resilience of traditional values. The need to place the Problematik of the development process in its proper historical context and institutional and cultural realities is illustrated in the following quotation: "The fact that a society is the captive of its own history and that history has its own phases determined by the material conditious prevailing at any given time ought not to be difficult to accept for many of those who think and write about Africa. Yet, the virtually total absence of any recognition of this point is a testimony to how far our own time has become insensitive to the significance of history in development. Whether on the right or on the left, the tendency for writers on Africa has been to take readily for granted the predominance of the international environment and the power of science and money. In a Promethean urge, there has been little inclination to pause and examine what the objective conditions of Africa really are. The attempt by many African leaders, supported by the compulsive international donor community, to cut historical corners, however, has proved abortive. This is the root of the present crisis in Africa. To be sure, the dismal state of the world economy contributes to it but is not the real cause." Goran Hyden, No Shortcuts to Progress: African Development Management in Perspective, University of California Press, 1983, p. 213 (emphasis added). 119/ Yet, the Bank had identified the strong linkage between human capital development and economic progress as early as the late 1940s when it stated that "there are many reasons why development is necessarily a gradual process. Of fundamental importance is the low level of education and health prevailing in most underdeveloped countries. Without intelligent, skilled and vigorous manpower, the economic progress of any country is likely to be slow, however amply it may be endowed with natural resources and however substantial the assistance of foreign capital.' World Bank, Annual Report 1948-49, p. 8 (emphasis added). - 110 - even the Bank. The Bank is not on record for ever having voiced concern about the high level and composition of industrial investment in the public sector undertaken over the years. In point of fact, both the 1974 and 1976 Bank missions supported the BIS and recommended that industrial investment be stepped up through the promotion of joint ventures, unwittingly over- emphasizing the role of investment and de-emphasizing that of institutions and human skills in the development process.120 Even some of the Bank's own indastrial projects have been far from exemplary in terms of conception and design (e.g., shoe factory, pulp and paper, cashew nut processing) for that matter. The adverse impact during t .e short-lived implementation of the BIS, even in its diluted form, became evident in the late 1970s-early 1980s- -although realistically the situation reflected the cumulative and compounded effects of the ill-conceived investments undertaken under the import-substitution policy applied in earnest since the mid-1960s. 7.14 The Bank made little, if any, effort to impress on the leadership the need for a more circumspect approach to industrialization, point out the pitfalls of coloring economic decisions with strong ideological elements, and draw the line beyond which it would be disinclined to support inopera- tive and counter-productive government policies. Rather, for more than a decade, the Bank kept cautioning on the distorting effects emanating from the severe shortcomings of the control, allocative and coordination mecha- nisms established to operate and monitor a centralized industrial system, on the deleterious effects of the various constraints (e.g., acute shortage of manpower in the parastatal sector to identify, prepare and operate commer- cially viable industrial projects), or on the potential role of private initiative, without taking a firm stance and calling the government's atten- tion to the dire consequences of policies and actions which were already surfacing. 7.15 Traditionally, the Bank's approach in dealing with socialist re- gimes has been to accept the economic, social and political realities and, since it could not aspire to change the status quo, to try and operate with- in the parameters of the system. In the case of Tanzania, the Bank rather impulsively identified itself with the objectives and the resultant social and economic policies of the political leadership, and adapted its country 120/ National Manpower Surveys conducted by the Ministry of Labor and Manpower Development in 1977 and 1984 indicated that some 50Z and 85%, respectively, of the expatriates had no local counterparts to understudy them, largely because of the non-availability of candidates with proper qualifications. There has been a persistent acute shortage of managerial and technical staff with a broad knowledge of project preparation, evaluation and implementation, and of the technical, organizational and financial aspects of industrial mandgement. While there is a small core of competett staff, there are no adequate professional cadres capable of providing an in-depth managerial structure in most parastatals. strategy, policy and lending objectives to the ideological posture and pol- icy framework of the government. The Bank's overriding concern has been to help the government carry out its objectives. But, in the process, no at- tempt was made to contest the soundness of the underlying premises and work- ability of the system in light of the experience gained, or hint the need for a dramatic departure from enshrined philosophical, political and econom- ic principles. Instead, the Bank's efforts were confined to impressing on the government the beneficial role of particular vestiges of the free market system (e.g., macro-incentives, export trade, private initiative) and to suggest the potential of a gradual "drift" toward market socialism. But even such modest suggestions proved to be against the grain. Thus, over the years, sound economic policies and common sense succumbed to ideological imperatives. 7.16 Yet, even if expediency dictated that the Bank should work within existing political and economic structures, the question ought to be raised whether the Bank should have continued to support policies which were not working, could not reasonably have been expected to work in the circum- stances, and which the Bank was not in a position (or was reluctant) to influence. Apparently, the view within the Bank was that, at least, if the projects supported would succeed, that would be good for the country--an attitude conveniently reinforced by institutional pressures to lend.121 The question whether a project can succeed in the wrong policy environment was not fully explored, and this raises a more fundamental general issue: if conditions are uninviting and are likely to impair project implementation and sustainability, should the Bank be lending rationalizing that somehow it has to accept the socio-political status quo? Is such an approach consonant with prudent banking policies and responsible development assistance? 121/ A strong correlation has been suggested between price distortions and growth rates in 31 borrowing countries, including Tanzania: the greater the degree of price distortions in their economies, the lower their GDP growth rates were. "Pricing for Efficiency," Chapter 6 in World Bank, World Development Report 1983, 1984. In the context of its Annual Review (1987), OED found, not surprisingly, a similarly negative correlation between price distortions and the performance of (Bank- assisted) projects implemented in the 1970s in the same group of countries. These findings confirm the conventional wisdom that country economic policies and management do influence project performance and can undercut the positive zontribution of individual projects to economic growth. The Political Decision-Making Process 7.17 On a different plane, the Bank had not fully fathomed the implica- tions of the sub rosa involvemeut of the Party apparatus in the decision- making process, both at the policy level and at the level of project selection and implementation.122 At the project level, perhaps the r.asoning was that a price had to be paid to maintain political stability and keep open channels of communication, and that acquiescence would pay dividends in the longer haul. But at the policy level, more than a decade after the Arusha Declaration the Bank still assumed that, in its dealings with high ranking government officials, their decisions had been sanctioned by the Party (Memo by RVP to President dated 9/26/80), thereby obviating the need to establish direct contact with the Party leadership itself. Nevertheless, this intangible dimension was bound to (and did) affect the quality and effectiveness of an open and potentially proactive dialogue on important issues. 7.18 Similarly, the Bank has not been able to appreciate the limitations and full implications of the political leadership's policy-making style and rhetoric for the effectiveness of the decision-making process. The "we must run while others walk" approachl23 advanced through ideology, political power and the Party apparatus to $rovide the much needed impetus for the 122/ In a speech made in 1968, President J.K. Nyerere pointed out that "It (the party) has to speak for the people....It is the government which is the instrument through which the party tries to implement the wishes of the people and serve their interests. And the party has therefore to determine the basic principles on which government should act; it has to determine the policies its government will follow." "The Party Must Speak for the People," in J.K. Nyerere, Man and Development (Dar es Salaam: Oxford University Press), 1974, pp. 14-18, p. 16. Experience with the role assumed by the Party over the years illuL-rates the point further. It has been argued convincingly that the Party and the state in Tanzania are so closely related organizationally that it is very difficult to make a distinction between them. Over the years, there has been a trend of concentrating important powers in the Party rather than other state organs. In fact, the Party has assumed a greater role in terms of decision-making after the promulgation of the Arusha Declaration, and the supremacy of the Party has weakened the role of the parliament. A.K.L.J. Mlikuka and P.J.A.M. Kabudi, "The State and the Partyu in Issa G. Shivji (Ed.), The State and the Working People in Tanzania (Dakar, Senegals CODESRIA), 1985, pp. 57-86, especially pp. 83-84. See also A. Coulson, Tanzania: A Political Economy, 1982, p. 323. 123/ Goran Hyden, "We Must Run While Others Walk: Policy-Making for Socialist Development in the Tanzania-Type of Policies" in K.S. Kim, R.B. Mabele, M.J. Schultheis (Ede.), Papers on the Political Economy of Tanzania, 197P, pp. 5-13. I desired behavioral transformation of the masses, i.e. to effect a change in the peoples' attitude toward life and work, has had inimical effects at the policy-making level since it tends to foster uncritical and injudicious decision-taking postures. This stems from the fact that this tactical weap- on creates a strong inducement among policy-makers to try and do everything and all at once, without understanding fully what is realistically achiev- able given the available resources and the prevailing constraints, and with- out due consideration of the consequences of their decisions in terms of economic and social costs. By ushering in a "motivation-outruns-under- standing" behavioral pattern,124 whereby motivation to solve a problem arises in advance of adequate understanding of the issues involved and not the other way around where understanding paces motivation, this style breeds overconfidence in the solvability of all problems and results in impulsive solutions to problems. The ultimate objective is viewed as so important that the real costs to attain it become a secondary matter--e.g., witness the readiness to accept lower akd never quantified rates of economic growth in the interest of building socialism, or the emphasis on human dignity without due regard to the potential impact on enterprise efficiency. 7.19 The inherent inadequacies of this tactic are reinforced by the attitude of policy-makers to ignore past experience and rule it out offhand as irrelevant--in fact an experience from which a break is sought, albeit for understandable reasons. Thus, because of the tendency to look abstract- ly at the more distant future and the emphasis on developing new initiatives and "getting things off the ground", reflection, rationality, optimization, economic calculus and development of effective institutions give way to ideology and political calculus, which become the crucial variables to deci- sion-making and problem solving. More often than not, therefore, decisions 124/ For a discussion of the experience with this policy style in Latin America, see A.O. Hirschman, Essays in Trespassing: Economics to Politics and Beyond, C&mbridge University Press, 1981, pp. 152-3. have been based on broad fprinciples" rather than cost-benefit analysis.125 and frequently taken within the framework of the key decision-making organs of TANU in camera rather than the National Assembly. Moreover, fragmentation of the government structure at the implementation level, unar- ticulated mandates to the executive, and absence of effective implementation mechanisms, reinforced by poor organization and management, has entailed far-reaching compromises, with the result that the outcome of policy deci- sions has been quite modest compared to the original promises and expecta- tions. No doubt. this emotionally charged syndrome has been effective in arousing empathy and in mustering foreign assistance. But, ironically, this policy style was taken by bilateral and multilateral agencies, including the Bank, at face value and as implying strong political commitment and determi- nation to bring about economic and social change, and no attention was paid to the attendant untoward effects on the decision-making process and the ability to develop the capacity to achieve the intended results. 7.20 Throughout the 1970s, the Bank's attitude reflects a determination to support an influential African leadership strongly endorsed by the donor community. This steadfast commitment derived largely from the Bank's world view, the perceived commonality of purpose, and the mounting institutional pressures to lend. Availability of concessionary, including IDA, funds perhaps made it easier and convenient for the Bank to be supportive of a 125/ It has been argued persuasively that the vagueness and diffuseness with which development goals are articulated in Tanzania is far more than a technical problem, i.e. the willingness of the decision-makers to specify the trade-offs between multiple and conflicting goals of development. Rather, it is symptomatic of a fundamental political 4issensus centered around basic ideological issues, such as the type of society that Tanzania wishes to create. Because of these furamental political divergences of opinion, the Party has been deliberately issuing directives on the most general plane and, thus, it has never developed the capacity to concretize goals to both guide the bureaucracy and hold it effectively to account. J. Loxley and J.S. Saul, "Multinationals, Workers and the Parastatals in Tanzania", Review of African Political Economy, No. 2, 1985, pp. 62-71. Another observer remarks that "the ideology was applied mechanistically, the most fundamental elements often ignored or bypassed because they could not be 'translated' into detailed schemes by decision-makers and resource- controllers." Idrian N. Resnik, The Long Transition: Building Socialism in Tanzania, (New York/London: Monthly Review Press), 1981, p. 274. The diversity and often conflicting objectives set for parastatals and the political difficulties of formalizing trade-offs has had another important implication: it has rendered extremely difficult the task of devising an effective accountability and control system and, thereby, ensure the effective implementation of government policies by parastatals. - 115 - regime which espoused appealing social objectives but questionable develop- ment/industrialization policies and programs, in the sense that it obviated the need to take a firm stance on important matters, to press for solutions to difficult and controversial issues, and to resort to conditionality. In consequence, the time horizon and the effectiveness of the means to achieve the stated ends of the political leadership--even if laudable--were not seriously challenged by the Bank's top managers. 7.21 The Bank debated issues relating to iadustrialization strategy and policies with government officials and their advisors, but the influence and impact until the mid-1980s has been marginal--at best. Evidently, the Bank did not press firmly enough on fundamental issues, confining itself to ad- vising. probably because it had discounted heavily the government's degree of receptivity. Also, at least until the late 1970s, the Bank's Management appears to have entertained the view that the problems of industry, despite their intractability, were basically transitional in nature and not systemic and deep-rooted.126 This implied that they were manageable and that they could be potentially resolved through piecemeal policy reforms. However, such optimism is difficult to rationalize, considering that Bank missions had been invariably critical of the effectiveness of the planning mechanism, organizational structures and policy framework, of the choiLe and design of industrial projects supported by the government as well as of their operat- ing inefficiencies, and of the absence of political will to take conscious remedial action. This attitude perhaps explains why it took the Bank so long (mid-1980s) to face up squarely to Tanzania's micro- and macro-economic problems, to communicate in a more assertive manner its cot.cern about the seriousness of the situation and, more importantly, to devise a sectoral policy package and concrete actionable proposals. Tanzania's critical eco- nomic situation and the change in the Bank's leadership ultimately provided tae requisite catalyst that led to a change in the Bank's posture and ap- proach in the early 1980s. 7.22 Over the years, the political leadership of the country has been excessively preoccupied with strategic issues pertaining to the long run outlook of the society and the required structural changes to achieve this transformation. The inordinate allocation of effort to map out long-range strategies for the more remote future meant that, inadvertently, pressing issues of more immediate concern did not receive due attention, as reflected in the widespread institutional indecisiveness and inert;a. Such an atti- tv-ie was bound to frustrate the Bank's efforts to develop a meaningf-al agen- da for action and follow it through. This suggests that the notion enter- tained within the Bank of worki.g closely with the borrower, and within the 126/ For instance, in the late 1970s, the Bank's resident representative in Tanzania was quoted as having stated that "the manufacturing parastatals were generally performing well, once their early 'teething problems' were resolved." M. Bienfeld, "Evaluating Tanzanian Industrial Development" in H. Fransman, ip* cit* p. 130. - 116 - existing economic, social and political system, to the end of inducing piecemeal changes, has not worked in the case of Tanzania. The results achieved heretofore hardly suggest that ths approach has helped promote the long-term interests of the borrower, and may well imply that this conception of Bank-country relationship may have to be reconsidered. 7.23 Given that the heavy state intervention had increased ineffici- encies, it is striking that the Bank, in the face of the inertia of the planning and implementing author!ties, did not feel the need to take it upon itself to devise early on a practical and cost-effective mechanism to moni- tor the inter- and intra-sectoral allocation of investment and recurrent resources based on in-depth reviews and assessment of investment programs and resource availability, relative factor scarcities and costs, strategic objectives and priorities (e.g., skewing resources toward operating existing plants instead of building new ones, favoring greater allocations for agri- culture and infrastructure), to signal potential pitfalls, and to seek more vigorously appropriate government responses. The Bank did "ppreciate and communicate its concern about policy-induced distortions, such as the un- workability of the extensive substitution of direct controls in allocation decisions for the market mechanism and the traditional indirect controls (e.g., taxes, subsidies, exchange rate). But it failed to point out in no uncertain terms that, particularly in Tanzania's circumstances, public man- agement was a poor substitute for private management (and ownership) at the enterprise level, that management contracts, aside from being controversial and expensive, were mere palliatives, and that industrial expansion based on this model was bound to create serious difficulties. More generally, the Bank was not explicit in its recommendationa that government intervention in any form entails Otransaction costs" in vcquiring, processing and transmit- ting the relevant information, in designing public policies as well as in enforcing compliance, and that, in consequence, there may well be (and there have been) many instances of "bureaucratic failure" militating against the attainment of the desired efficient outcomes. 7.24 In fairness to the Bank, policy-makers in Tanzania were hardly unanimous in their approach to industrial development, with the various factions frequently espousing views encompassing a wide spectrum. MoreoveL, with the passage of time, vested interests had become more entrenched. For practical reasons, the Bank's approach was to try to work more closely with technocratic elements who appeared to be more receptive to the Bank's views. However, given the strong ideological biases, the fractious political struc- ture of the country, and the dominant role of the inner circles of the Party, such effort did not yield tangible results. By the same token, how- ever, the Bank could have reconsidered its position much earlier than it did and, thereby, possibly mitigate the inimical effects of this protracted indecisiveness. - 117 - 7.25 Programs of policy reforms and the attendant conditionality pre- scribed by bilateral or multilateral agencies are often viewed, perhapi uncritically, as 'by nature" interventionist. When they are perceived nar- rowly as encroaching on prevailing economic and political ideologies, they pose severe barriers to reforms. Both the Bank and the borrowers do not take kindly to intervent'ons with explicit political and social conditions. Nevertheless, in many instances, as in the case of Tanzania, proposed eco- nomic reforms are not likely to be initiated and sustained unless they also address related political, social and cultural aspects--in effect question- ing the ideologiLcal underpinning of the who. approach to economic develop- ment and industrialization and, thereby, inviting growing resistance by entrenched bureaucracies and vested interests. But if the political leader- ship is perceptive, receptive and responsible, that is if the process of policy reform can be internalized, then the "odium" disappears and the en- actment of reforms can no longer be seen as the product of external inter- vention. To be sure, the line is hard to draw and this dilemma confronts the Bank frequently, putting its Management in a difficult posttion and forcing it to consider taking decisive actions, including suspension of lending. Yet, the issue cannot be brushed off in view of its significant repercussions for all parties concerned; rather, it needs to be addressed squarei.y at sone juncture, particularly if efforts to persuade the political leadership about the pressing need for reforms and ideological compromise have failed--i.e. if the process of policy formulation cannot be internal- ized. External Assistance 7.26 Tanzania's unprecedented access to concessionary flows of external capital has allowed it inter alia to maintain a high rate of largely ill- conceived and uneconomic industrial investment. The cushioning effect of this access to funds, at least until the late 1970s, reduced the govern- ment's receptivity to external advice for badly needed reforms in the coun- try's economic plannin- and management system, policy framework, and insti- tutional arrangements that were required to streamline industrial invest- ment, improve industrial performance, and generate foreign exchange earn- ings. Inertia helped perpetuate a morbid situation, compound the existing problems, postpone the hard political decisions, and render the eventual decisions even tougher. This cushioning effect, by easing the pressure and need for immediate action, has limited the Bank-country dialogue to convey- ing diagnostic results unilaterally, to suggesting caution, or to making recommendations without eliciting a reciprocal commitment to action. 7.27 The Bank apparently shared--at least concerning the immediate future--the leadership's pessimism about export trade prospects which, aside from ideological predilections, were seen as severely limited by world de- mand, declining prices and administrative restrictions. Limited prospective export earnings were perceived as a binding constraint both on domestic production, whose growth would be adversely affected by the limited avail- ability of the requisite capital and intermediate goods, and on the capacity to transform even an increased level of savings into investment, perforce necessitating recourse to external assistance--notions which were formalized - .-. I - in the "two-gap" approach to aid and development. This approach, coupled with the recurrent balance of payments crises and the view that foreign aid would still be required to supplement the patently low level of domestic savings (largely due to the poor performance of the parastatals), led to a situation where the Bank became a vocal advocate of official aid to sustain the government's high rates of industrial investment. 7.28 The attitude of the po itical leadership toward the issue of ex- ports and external assistance is intriguing. The volume of Tanzania's com- modity exports have been steadily declini.-g since the mid-1960s, as a result of the slow growth -f agriculture, the ambivalence toward encouraging pri- mary commodity exports which in some quarters were regarded as symptomatic of the colonial trade pattern, and the tepid support for manufactured ex- ports also based on doctrine. But, if exports reflect a nation's dependence on the world economy, foreign loans at concessionary terms and unrequited transfer of resources do more obviously so. Yet, it was to such external assistance that Tanzania has had to turn const-ntly in order to sustain a high level of industrial investment and to alleviate persistent balance of payments problems. This ideological contradiction to the highly proclaimed self-reliance did not evoke second thoughts either among local policy prac- titioners or among bilateral and multilateral agencies. 7.29 The Bank's endorsement of Tanzania's industrial development objec- tives and strategy, although with some reservations concerning the effec- tiveness of the supportive policy framework, has facilitated the mobiliza- tion of funds on concessionary terms from donors127 in substantial amounts, and has supported investments which is unlikely that they would have been undertaken without such external financing. Furthermore, the Bank's continuous exhortation of the donor community to provide assistance, coupled with their well-meaning eagerness to help, has sustained a constant inflow of official aid that helped maintain irrational domestic policies and, argu- ably, has hampered the country's ability to borrow on commercial terms abroad. Finally, lack of aid coordination, in part because of the govern- ment's policy to play off one donor against the other, and the fact that the Bank failed to see that both the donors and the government, for different reasons, were embarking on ambitious projects which each one wanted to pro- mote, led to commitments that could not be honored because of the limited absorptive capacity of the country (e.g., infrastructure, local financing, skills) and to supporting projects of questionable economic merit. Thus, 1271.In 1980, foreign aid was 18% of GDP, 107% of total tax revenues, and 1532 of export earnings. D. Lal, The Poverty of "Development Economics," (Cambridge: Harvard University Press), 1983, p. 55, footnote 1; P.T. Bauer, Reality and Rhetoric: Studies in the Economics of Development, (Cambridge: Harvard University Press), 1984, p. 47. - &&V. w easy access to aid inadvertently contributed to the installation of exces- sive and at times inefficient industrizl capacity, which produced temporar- ily euphoric investment and GDP figures but ultimately created severe bot- tlenecks, resulted in poor performance, and stifled industrial growth. The Bank's approach toward and unconditional support of Tanzania's industriali- zation effort did little, if anything, to forestall the sector's present morass. 7.30 In light of the above, the economic benefits of the external assis- tance in relieving fundamental constraints impeding industrial growth and structural transformation (e.g., increase and improvement of human skills, rise in the level of investment and savings, change in the composition of output and employment, adoption of more productive technology, development of new and efficient institutions, etc.) appear to have been minimal. Ex- ternal assistance has sustained a respectable growth in industrial invest- ment, but the pattern of intra-sectoral distribution, project selection and plant scale have been far from optimal and, as a result, the productivity of the new investment expenditures has been very low (see para. 4.08). Furth- ermore, this nve-tment rate has not been accompanied by sufficient improve- ment in technical skills and organization to make effective use of the addi- tional capacity that became available, to accelerate and sustain growth, or make fuller and better use of domestic resources. Whatever increments in production and income have been achieved, they have not been channeled into increased savings, especially in the public sector. Investment has not been allocated in activities that would promote export growth (or efficient im- port substitutlin) and thereby alleviate (let alone avoid) balance of pay- ments problems and reduce dependence on aid. Foreign private investment represents a miniscule fraction of external capital inflows. And there has been no tangible improvement in the efficiency of the supporting institu- tional framework and in the effectiveness of the government apparatus. Perhaps the major contribution of external assistance has been the mainte- nance of political stability during the formative years of the nation. But, in retrospect, and on purely economic grounds, the effectiveness of the external assistance, in particular its sizable financial component, has been limited as the country has failed to develop the requisite capacity to use massive external transfers efficiently and may even have fostered a self- defeating attitude by having adopted injudicious policies in its haste to industrialize. This suggests that unmeasured external financial assistance may inadvertently have a desensitizing effect on the recipient's behavior. VIII. THE AFTERMATH--PRESENT STATUS OF THE INDUSTRIAL SECTOR AND PROSPECTS 8.01 Tanzania's industrialization strategy in the 1960s and 1970s was colored by its colonial past and the contemporary thinking about economic development. The political leadership was anxious to achieve economic as well as political independence by moving away from its dependence on im- ported manufactures in exchange for primary exports. A socialist strltegy of industrial deveiopment and policy framework was adopted in earnest during the late 1960s to advance the objectives of self-reliance, economic growth and structural transformation. Industry was seen as a powerful engine of growth that could modernize the economy through more capital-intensive, higher-productivity processes, promoted within a protected environment. The political leadership firmly believed that a rigorously pursued industriali- zation strategy with emphasis on resource based, producer goods industries that cater to basic needs, dubbed as the basic industry strategy, coupled with an extensive state involvement in the economy, could accelerate growth and achieve a more equitable socio-economic development. The Bank endorsed this approach to industrial dvelopment from its inception, and even became its staunch supporter throughout the 1970s. 8.02 The implementation of the adopted basic industry strategy implied inter alia: creation of effective organizational structures and planning processes to ensure appropriate inter-sectoral (e.g., industry, agriculture, infrastructure) and intra-sectoral (e.g., rational project selection) allo- cation of resources; the substitution of an administrative apparatus for the market mechanism to effect allocation decisions and implement policy mea- sures; development of public management capability to ensure efficient proj- ect implementation and operation; significant levels of investment in rela- tively large, capital-intensive and import-intensive industries; a constant and dependable stream of foreign exchange to finance imported capital and recurrent import requirements; a relatively large domestic market to reap scale economies; and an adequate supply of technically trained manpower, including managers, to operate industrial undertakings--a tall order. Con- sidering Tanzania's level of development and the demanding requirements of the basic industry strategy, it is arguable whether the strategy could have been effectively implemented at the projected pace and time frame. 8.03 Major setbacks, indeed, have set in and the achievement of these goals in the immediate future is far from assured. After two decades of mis-perceived inward-looking industrialization, and despite the injection of an inordinate amount of external financial and technical assistance, the industrial structure that has evolved to date is not significantly different from that which existed in the early 1960s, real per capita income and wages in the mid-1980s are lower than in the mid-1960s, the balance of payments situation remains critical, and heavy dependence on foreign inputs, finan- cial resources, technology and expertise persists--in defiance of the stated goals. A confluence of factors, encompassing an excessive and poorly man- aged planning system, inadequate policies and ineffective institutional arrangements conditioning behavior, extremely low productivity of the work force and of the investments undertaken (some US$3 billion in real terms) and, to a limited degree, external shocks, has led to the development of an inefficient industrial structure which is incapable of generating the hoped for sustainable growth and transformation of the economy. 8.04 The string of intractable problems afflicting industrial perform- ance, and whose severity assumed egregious proportions during the 1980s, includes oversized or sub-optimal plant scale; completed but inoperative plant capacity due to lack of infrastructure; shortage of technical and managerial skills and continued reliance on expatriates; acute shortage of - 121 - raw materials and spare parts due to the shortage of foreign exchange; over- staffed establishments and low factor productivity due to rigid labor laws, poorly maintained equipment, and absence of incentives to reward perfor- mance; inadequate economic infrastructure (shortage of power and inadequate transport system in particular) reflecting the poor inter-sectoral alloca- tion of investment and recurrent resources; and under-capitalized and heav- ily indebted, if not virtually bankrupt, parastatals due to poor management and performance, reinforced by successive drastic devaluations.128 The poor condition of the parastatals and their mounting indebtedness has in turn undermined the viability of the financial intermediaries which have an ex- tremely weak portfolio and are striving to remain afloat. 8.05 In essence, the government has failed to develop the requisite organizational and technical capacity to devise, articulate and implement effectively a realistic industrialization strategy and investment program consistent with the country's factor endowment and resource potential. An alien industrialization strategy was transplanted uncritically, prematurely, and without regard to binding constraints on the basis of shcer ideology. Socialist policies were pushed beyond the capacity of the highly constrained system to manage, resulting in poor economic performance.129 Structural change has never been adequately articulated. Thus, in effect, import sub- stitution became a device for substituting imports by imports.130 According to one, sympathetic, observer, "Tanzania was approaching a classic 'import substitution crisis' even without the changes which followed 1973..131 The 128/ Between 1966 and 1988, the shilling has depreciated dramatically: 1966-74 T Sh 7 = US$1; 1983 T Sh 12 = US$1; 1988 T Sh 120 - US$1. 129/ For an early warning see G.K. Helleiner, "Socialism and Economic Development in Tanzanias, Journal of Development Studies, January 1972, p. 202. 130/ "The industrialization policy has tended to follow the classic import substitution pattern with the result that import dependence has not been reduced but simply shifted from dependence in the consumer goods to dependence in the intermediate and capital goods sectors." John Loxley, "Monetary Institutions and Class Struggle in Tanzania," in B.U. Mwansasu and C. Pratt (Eds.), Towards Socialism in Tanzania, University of Toronto Press, 1979, p. 81. 131/ J.M. Bienfeld, "Evaluating Tanzanian Industrial Development" in M. Fransman, op. cit, p. 135. evidence clearly suggests that domestic policies and poor management are largely responsible for the current economic crisis, despite the infusion of inordinate amounts of external financial assistance.132 8.06 Inadequate allocation of resources and deficient incentive struc- tures for the development of the agricultural sector,133 excessive administrative controls over economic activities, and the continued growth in the size of the public sector without due regard to the limited administrative capacity available distorted the pattern of development process and stifled progress. Ineffective organizational structures and planning processes incapable of ensuring proper coordination134 and inter- and intra-sectoral allocation of resources; ambitiolis investment programs, largely supported by the largess of the Bank and sympathetic donors; poor project screening procedures and haphazard project selection; underestimation of the human, institutional and organizational capacity of the country; poor organizational structure and performance of parastatals; and a distorted policy framework resulted in the development of an industrial sectir which is over-extended in relation to the size of the market and the country's technological and skill (managerial, technical and labor) capabilities. The policies adopted ended up favoring industries linked to imported capital, skills and materials, rather than the use of domestic resources, thereby creating mounting pressures on the balance of payments at a time when, because of the anti-export bias, foreign exchange earnings have been dwindling. Industrial investments made in the 1970s lacked selectivity, were grossly unproductive, and did not succeed in changing perceptibly the industrial structure, as no effective mechanism was developed to ensure that they interlocked and made the best use of available resources. The flaws and underlying problems of the strategy became more forcefully evident in the late 1970s and persist since, suggesting that external dependence will continue well into the 1990s. The specific problems currently affecting industry at the enterprise level and some of the policy issues involved are presented in a summary form in Table 2 below.135 132/ "At the level of project by project evaluation--ex ante and ex post- -there can be no question that Tanzania has made serious mistakes" which the political leadership has openly admitted. However, these are viewed as a price to be paid for "learning by doing." R.H. Green, "Industrialization in Tanzaniaf in M. Fransman (Ed.), Industry and Accumulation in Africa, 1982, p. 96. 1331 The complementary nature of the relationship between agricultural and industrial development is acknowledged even by structuralists. A.F. Ewing, Industry in Africa, (London: Oxford University Press), 1968, pp. 4-5. 134/ After two decades, a decision was taken early in 1988 to establish a National Planning Commission to tighten the planning mechanism and to coordinate all economic activities. 135/ See also, World Bank, Tanzaniat An Agenda for Industry Recovery, 1987, Vol. I, pAras. 3.60-3.93, 4.32-4.43. - 123 - Tahis.2 PROBLBS AFFECTINO INDUSRY: OTTLEIMS. CAUSES AND POLICY ISSUiS Eproa sagna Main Polai Tesues 1. SCARCITY OF RAW Reduced import capability to service an Since: (I) Import capability will remain MATERIALS AND overextended industrial sector, the result of insufficient to service neds of overestended SPARE PARTS anti-agricultural policy blases, excessive industrial sector, and (Ii) many firma would investments In import intensive industries, be inefficient at attainable capacity, the and exogenous factors. allocation of resources should be selective. 2. INADEQUATE Due to Inaufficient past Invstents, Overall infrs"ructural Improvesenbs are required. INOtASTRuCTURS inadequate maintenance and overall Hwaver, specific debottlenecking for particular (POtWM, WAISt, deterioration, combined 6ibh locational enterprises (e.g., provision of road and power TRANSPOR choices of firma which often responded to generators) justified only if benefits would ron-economic considerations. In addition, in exceed (potentially high) investment coats. transport, there are large inefficiencies in the parastatal sector. 3. EQUIPMENT Wrong technological choice, lack of preventive Reformed policy framework required to reaward INADEACY maintenance, poor quality control, all related effective maintenance and upgrading (rather to weak technological capability and inadequate than continuing purchase of new equipment), for incentive structure. Role of aid in provision adoption of appropriate technology, and mesures of inadequate equipment. to enhance technological capability. Redirection of aid flows. 4. INADEQUATE Overextended industrial sector for the countryea Improvement in macroeconomic, trade and industrial TE .OLDOICAL, technological capability; wrong choices of policies necessary but not sufficient. Specific SKILLED LABOR industrial activities, and lack of capability measures required for technological learning and AND MANACERIAL growth over time. Macroeconomic and trade education process, which is a slow process. CAPASILITY policies and role of expatriates often were not conducive to indigenous technological capability growth. 5. WRONC PLANT SCALE, Exchange and Interest rate and price control Viability of operating large projects ought INAPPROPRIATE policies which rewarded large-sized import and to be masessed consldering capital costs as IMPORT AND CAPITAL capital-intensive investments; non-economic sunk. For new projects. adequate policy INTESITY factors In parastatal investments. Neglect in framework, investment appraisal capability assessment of recurrent needs and market size. and aid coordination is essential. Role of Role of aid in the design and implementation of 65E development. large projects. 6. INSUFFICIEfT Depressed state of the econoty, fow product Reformed policy framework important to DEMAID quality, and high production costs resulting in increase GDP growth and aggregate demand high prices and excessive capacity. These, in and to lower costs (and prices). turn, are due to past policy inadequacies, over- expansion of capacity, and external factors. 7. WASTEFUL Due to pricing policies that do not encourage Improved overall policy framework and better UTILIZATION eavings in resource use, and to inadequate technological, skilled labor, and OF IN'IIS technological choice, equipment conditions and managerial capability. technological capability. ouge: World Bank, Report No. 6367, Tanzania, An Agenda for Industrial Recovery. June 30, 1987, Vol. I, p. 63. 8.07 The proposed industrial reorientation scheme (paras. 4.06, 4.07) appears to be in the right direction and realistic, although success would in part be conditioned on devising a strategy emphasizing a two-pronged agro-industrial development, including the requisite infrastructural support. But more importantly, since the required adjustments are more than marginal or of a tidying up nature, the effectiveness and sustainability of the envisaged adjustment process will depend inter alia on a total reorientation in economic thinking and in political and social attitudes, the unwavering commitment and support of the Party and the political leadership, the internalization of the adjustment process, the potential resistance by entrenched vested interests, the consistency and complementarity of proposed policy changes, and the pace of implementation of the suggested institutional and policy reforms--a tall order. In this respect, the government's expressed intention to address the structural problems "within the socio-political framework chosen by the people of Tanzania" (Policy Framework raper, para. 6) is a matter of concern. The Bank has been working for ever two decades within Tanzania's socio- political parameters without measurable success. And as long as, according to the Management's own admission (CPP 1981, para. 39), ideological considerations play a major role in decision-making and sensitivities and suspicions persist regarding the role of the Bank, the extent to which fundamental development strategy and policy issues can be meaningfully pursued with the government remains limited. Furthermore, opportunities to raise strategy k,sues with the Party, which plays a dominant role in the decision-making process, are rare and historically have been shunned. In light of this experience, it would seem that, unless there is a major shift in the ideological position of the political leadership in telling socio- political parameters, it is doubtful whether the drive towards industrial reorientation will produce tangible results. Resolve to introduce political cum economic reforms in tandem is a condition sine gua non for the success of any program to reorganize/rehabilitate the industrial sector. This is a major hurdle which nevertheless needs to be fully appreciated by all parties concerned. IX. ASSESSMENT OF THE BANK-COUNTRY RELATIONSHIP AND DIALOGUE 9.01 The Bank's posture over the past quarter of century vis-a-vis the country's approach to development, particularly during the early formative and critical years, has been reactive. The Bank elected to operate within Tanzania's socio-political parameters and support impulsively its indus- trialization effort, reversing its previous position; and it did not change its attitude even though the socialist pattern of development was founder- ing. The Bank has not been able to curb the leadership's unrealistic as- pirations and growing impatience. Nor did it challenge the government's industrialization strategy or present the leadership with potentially more viable and sustainable alternative approaches to development--e.g., a grad- ualist approach to industrialization, concurrent pursuit of efficient im- port substitution and export promotion strategies, or adoption of a two- pronged agro-industrial development strategy based on exploitation of the country's abundant resources. Furthermore, the Bank has not been able to have an impact on the size, composition and funding of the industrial in- vestment programs, or on the inter-sectoral allocation of resources. All in all, the Bank's involvement through economic and sector work, industrial lending, and intercession in external assistance has had no tangible impact on Tanzania's industrial sector, and no success in foreclosing trends and developments that led to its present morass. Inadvertently, the Bank's attitude reinforced the delusion of shortcuts in the development process and the simplistic notion that somehow the stage of development of a country does not matter, and that all it takes to accelerate industrial growth and economic development and promote the desired structural trans- formation is a five-year plan, central direction, and availability of ex- ternal (preferably from *neutral" sources) aid and know-how. 9.02 The Bank has been tentative in its sector dialogue with the gov- ernment and has shunned from taking up strategic issues with the political leadership, although it was aware that ideology and the Party apparatus played a key role in the decision-making process. The Bank kept a low profile, and elected to remain an inert bystander while major debilitating events were unfolding. It did caution on the distorting effects of the controlling and allocative mechanisms employed, but fell short of taking a firm stance and emphasizing the consequential implications of the govern- ment's policies and actions on industrial performance. Ironically, through the late 1970s, the Bank continued to extoll Tanzania's commitment to cre- ating an egalitarian state and its achievements in social transformation, while glossing over its poor economic and especially industrial performance and voicing unfounded optimism. Bank missions had identified major and persistcnt shortcomings in the structure, institutional arrangements and t.erformance of the industrial system, including the investment decision- making process and operating efficiency of the parastatals, and had made specific and detailed recommendations over the years for remedial action. Nevertheless, the Bank's Management refrained from reassessing its strategy and involvement (until the early 1980s) and opted for an accommodati.,g posture which was clearly at variance with the dictates of the economic realities prevalent in the country--largely brushing aside the telling findings of economic and sector missions. Apparently, a confluence of factors including a narrow perspective, expediency, directives from top Management, a rigid hierarchical structure, a tenuous balance between the power of the decision makers and their moral obligation to take under ad- visement the views of the technical staff who carry out their decisions,136 corAplacency and inertia, a culture of loyalty to the organization and ruspect for authority, and a short horizon of lower and middle level man- agers restricted the capacity for rational action, and prevented mid-course adjustments in the Bank's position and overall strategy. In the process, the Bank's operational effectiveness was unwittingly compromised--a grie- vous instance of bureaucratic failure. 136/ For the importance of authority to the large organization and the conditions for its recognition, the value of the responsibility of authority in the organization and the conditions for its achievement, and the need and possibilities for achieving a reasonable trade-off between authority and responsibility, see K.J. Arrow, The Limits of the Organization, (New York: W.W. Norton & Co.). 1974, pp. 63-79. - 126 - 9.03 The Bank's overriding concern in industrial lending has been to help the government carry out its objectives, with which it identified, and to this end it adapted its strategy, policy and lending objectives accor- dingly. As a result, resource transfer rather than the true economic merit of the particular project became the guiding principle, as is evidenced by the Bank's support of unsustainable industrial projects whose selection was dictated from above and was reinforced by external pressures--and for which the Bank certainly cannot pride itself. The Bank continued to support new projects even after the country's economic situation had become very worri- some and the government was not responding effectively, slighting the po- tential impact of a hostile environment on project implementation and per- formance. All Bank supported industrial projects are in grave difficulty, partly due to the local conditions but partly because of inappropriate conception and design, and their impact on industrial development has been far from tangible. The Bank's impact through support of financial interme- diaries on the pattern of industrial investment has not been positive eith- er, while its contribution to their institutional development has been very modest. Finally, the Bank has not been able to achieve internally a proper balance and the requisite inter-sectoral coordination (power and transport in particular) in its industrial lending operations; nor has it been able to develop effective feedback mechanisms to utilize fully past project experience on an inter-sectoral basis. 9.04 The Bank-country relationship to date has not been fully produc- tive and the net benefit to the country remains an open question. This is due to the fact that the Bank's advice and prescriptions were not in tune and did not square with the ideological milieu and the course chosen by the political leadership for the country's industrialization. As long as ideo- logical considerations loomed large in the country's decision-making pro- cess, the extent to which fundamental strategy and policy issues could be influenced perforce remained limited. The Bank-country dialogue did not serve as an instrument to induce political choices. The Bank's accommo- dating attitude, failure to challenge openly the concept, premises, worka- bility, and potential consequences of the industrialization strategy and policy framework propounded by the political leadership, and neglect to present alternative and more realistic approaches to industrialization, reinforced this inherent limitation and frustrated progress. Ac a result, the Bank-country sector dialogue remained ineffective, having been down- graded to unilateral communication of findings and advice, largely going against the grain, and without eliciting reciprocal commitment to action. Ostensibly, the Bant has rendered service to the country through its moral support, technical and financial assistance, and as an active intercessor with the donor community. But the benefit of the Bank's financial assis- tance and its role as a vocal supporter can be contested, if seen through a different prism. The Bank's blank endorsement of the country's mode of industrial development, coupled with the donors' eagerness to assist, has sustained an inordinate flow of external aid that, inadvertently, not only has supported unviable industrial projects, but has helped the political leadership maintain injudicious domestic policies and postpone the hard political choices and decisions, with dire economic and social conse- quences. In the absence of an interactive Bank-country relationship, it was only a matter of time that the resultant impasse would have led to a dramatic reversal in the Bank's posture from one of enthusiastic support to one of virtual stand-off. Only in recent years have the Bank and the Government entered into a constructive dialogue, working jointly to develop action programs to rationalize the industrial sector. X. LESSONS OF EXPERIENCE AND RECOMMENDATIONS Lessons Learned 10.01 The review of the experience of the Bank-country relationship in Tanzania offers instructive lessons and suggestions that may provide a better defined framework for shaping the Bank's posture and approach to country macro- and micro-economic issues and lending operations. 10.02 Concerning issues pertaining to industrial strategy, the experi- ence suggests the following. (i) Industrial development should be pursued in parallel with and be firmly anchored in agricultural development and adequately sup- ported by the requisite human and physical infrastructure. The path chosen should reflect the prevailing socio-economic reali- ties, resource endowment, and binding constraints, while the op- portunity cost of ideological biases injected into the strategy should be quantified and acknowledged. Impatience with and unwar- ranted acceleration of a country's industrialization effort are likely to entail heavy economic and social costs. Government intervention should be circumspect and consistent, in appreciation of the fact that the evolutionary pace of development of human capital, institutional capacity and cultural parameters determine the pace of the industrialization process and set constrictive conditions for the sustained success of the industrialization effort. Infusion of external financial and technical assistance can be complementary--at best. (ii) There are insurmountable difficulties in pursuing an industrial strategy stressing the rapid development of an indigenous producer goods industry at an early stage of industrialization, since the deepening of the industrial structure in the direction of inter- mediate inputs and capital goods faces crucial threshold relations vis-a-vis important requirements (e.g., size of plant and market, technology and skills, attitudes and values, organizational struc- tures, institutional arrangements). The process cannot be unduly stretched and accelerated with impunity. (iii) Even if a "rational" inward-looking strategy is pursued, every effort wouli have to be made to exploit fully the existing and develop new export possibilities--an approach far from necessarily alluding to a commitment to an expurt-led industrialization strat- egy. 10.03 The experience with policy reforms affirms the following conclu- sions. (i) The crucial element in successful policy reforms is a strong po- litical will and commitment on the part of the policy-makers, the full support of highly motivated technocrats in core economic ministries, and the development of a broad consensus within the country. (ii) There is a far greater likelihood that policy reforms will be successful and sustained if the process of policy formulation is internaliz2d, in the sense that reforms are initiated, formulated and implemented through institutional processes and mechanisms originating within the country and integrated in its industrial development program, than if the reform emerges from the exercise of external leverage. (iii) Inertia and tentativeness on the part of the government, and con- tinuation of the status guo by adopting palliatives or fitful measures, only helps to perpetuate a morbid situation, to postpone the hard political choices that inevitably will have to be made, and to make the eventual decisions even tougher. (iv) Institution building is as important as the transfer of resources and extension of technical assistance; but it is a particularly difficult and drawn out process, and commitment, patience, and perseverance are required to achieve long-lasting results. More- over, sustainability of effort is crucial in solidifying the prog- ress already achieved. (v) It is extremely difficult to reorganize, let alone dismantle, unesseitial entities and services in the public domain because of the .sistance of entrenched and concerned bureaucracies. On the other hand, economic problems can hardly be solved through legis- lative means. The government often is under the illusion that it can actually solve a problem by setting up a new institution (e.g., investment bank, agency, holding company, corporation) to which it delegates the task. This new entity is presumed to be free of political interference, able to attract and retain the kind of talent it wants to secure, and have the capacity to cir- cumvent cumbersome and slow administrative procedures. More often than not, however, this practice has blocked real progress. Identifying and addressing squarely the root cause(s) of the issue at hand may have greater chance of rectifying the situation. 10.04 The experience with the effectiveneas of external financial assis- tance leads to the following observations. (i) External assistance can be helpful in promoting reform when it is conceived and presented as a reinforcing agent and as a means of reducing the cost of a reform to which the policymakers in the recipient country are already committed. Aid is likely to be self-defeating if it is proffered as a quid pro quo for the reform commitment itself. (ii) Only those forms of aid and investment that have the effect of changing economic, social and organizational structures can pro- vide essential benefits to the recipient. External assistance therefore should be circumspect and discrete to avoid thwarting government efforts to address pressing economic and sector issues by providing a convenient cushion. (iii) State intervention and policy-induced distortions (e.g., ill-de- signed industrial, fiscal, monetary and foreign exchange policies) tend to undercut the positive contribution of individual projects to economic growth. Therefore, monitoring of the entire invest- ment program of the recipient country with respect to aggregate size and composition, resource availability, design and implemen- tation of the policy framework is essential if there is to be some assurance that aid funds are put to productive use. 10.05 The experience with actions that addressed specific issues on parastatal performance provides useful insights on a number of issues. (i) There is no inherent reason why parastatals cannot be as dynamic and efficient as privately owned enterprises. But this can happen only if they are run by qualified managers who are allowed the requisite degree of autonomy in decision-making and operate in an inviting environment. In practice, however, and with few notable exceptions, state-controlled industrial enterprises and financial intermediaries suffer from intrusive political interventions, the appointment of unqualified top executives and board members, poor staff compensation and inadequate reward of high performers, and the difficulties imposed by a variety of social tasks often thrust upon them (e.g., overmanning) that distort their behavioral pat- tern, reflecting narrow if not self-serving perceptions of their role by the political leadership. Institutional inertia and lack of receptivity, reinforced by a weak policy framework and the Bank's inaility or reluctance to press more forcefully these issues, tend to further stultify progress in developing sound parastatals and financial intermediaries. (ii) It is important that unequivocal rules be established regarding the relationship between parastatals and the government authori- ties concerned, and such understandings should be adhered to. Parastatals should have clear objectives, and be assured of a degree of autonomy sufficient to enable them to make management decisions free from political pressures. At the same time, a well-designed system of ex pobt accountability, as opposed to ex ante controls, should be instituted based on appropriate and monitorable indicators for each industry to assess their perfor- mance. Introduction of standard cost accounting and establishment of profit centers at key stages of the production process are important management tools for cost and efficiency control and for taking remedial action. If such elemental conditions cannot be assured, the implication is that state ownership is not a workable option and privatization may be the only viable alternative. (iii) Given that the composition, quality, continuity and modus operandi of the Board of Directors of a parsstatal greatly affect the en- terprise's effectiveness and performance, due attention should be paid during selection to the qualifications, experience, non-alle- giance to special interests, and diversity (e.g., in terms of affiliation, skills, background) of the appointees to enable the Board to provide guidance free of self-serving influence and en- sure impartial decision-taking based on strictly commercial prin- ciples. In this regard, the usefulness of holding companies needs to be reconsidered. 10.06 Attempts to streamline the imports control regime and to devise arrangements to promote exports yield some interesting insights. (i) Restrictive import policies, and the attendant reliance on admin- istrative regulations and controls to ensure their effectiveness, tend to adversely affect production and resource allocation, are open to abuse, take a heavy toll on the country's scarce admini- strative resources, and stifle initiative. As a result, such policies should be used sparingly and only as an interim expedien- cy until a rational tariff structure and a well-designed incentive system can be put in place. (ii) Speedy import liberalization may not be expedient, and is likely to be strongly resisted, in the absence of a sufficiently broad export base and, generally, of good prospects for a stable inflow of foreign receipts. (iii) The existence of a transparent rationale for instituting an export incentives scheme is very important. Incentives should basically be compensatory in nature, counterbalancing discriminatory dom- estic levies and/or equalizing advantages of competitors due to subsidies, and should not aim at making up for production ineffic- iencies. On the other hand, the effectiveness and serviceability of well-conceived incentives can be easily stultified by poor implementation practices. (iv) The strong link between an efficient production framework and export potential, in view of the constraining effects of supply inelastictties in the short and medium term, has to be duly appre- ciated in efforts to stimulate export growth. (v) Reform of export procedures and documentation can be distressingly slow x.i their implementation, as it affects deep-rocted attitudes and vested interests, entails far-reaching changes in the internal procedures of agencies involved and the flow of documents, neces- sitates the establishment of new controlling mechanisms, and re- quires the development of inter-agency trust. Recommendations 10.07 The Bank's relationship over the years vith Tanzania suggests the need for the Bank's Management to rethink and reconsider certain aspects of its decision-making process and approach toward sectoral issues, fashion particular monitoring tools to follow up more closely on country macro- and micro-economic developments, and consider the possibility of greater direct involvement of its top managers in the development of strategies, in ad- dressing policy issues, and in discussions of important economic and sector work with borrowers. Furthermore, the experience with project design and implementation leads to a number of observations that warrant the Bank's attention and suggest ways for potentially more effective Bank interven- tions to foster the promotion of sustainable industrial projects and finan- cial intermediaries. Specifically, the following cources of action are recommended. (i) As long as ideological considerations play a major role in deci- sion-making, the extent to which fundamental strategy and policy issues can be meaningfully pursued with the government remains limited. Therefore, unless there is a major shift in the ideolog- ical position of the political leadership in key socio-political parameters, it is doubtful whether the present drive towards in- dustrial reorientation will produce tangible results. Political cum economic reforms in tandem are required for the success of any program to reorganize/rehabilitate the industrial sector and help move it off dead center. (ii) Industrial structure and performance are strongly affected by the macroeconomic policy framework, as well as by policies wiLhin the sector and within other sectors--agriculture and infrastructure in particular. It is important, therefore, to devise a mechanism for coordinating industrial sector work and operations, including indirect lending through financial intermediaries, with that in other parts of the Bank and on a continuing basis. Preparation of a Country Industrial Sector Strategy Paper and/or a Sector Memo- randum based on policy-orie..ted sector work and addressing strat- egy, policy and operational issues may be considered. Such docu- ments, which should reflect the Bank's position on fundamental issues and be updated as appropriate, would also provide inputs to the Country Policy Framework Paper. Furthermore, the present country focus of the Bank's organizational structure lends itself to other devices to coordinate effort, such as the formation of country work groups, pre-mission issues meetings, etc. (iii) The Bank should assess, say at three to five year intervals, the accomplishments in terms of the objectives set out in the Country Policy Framework Paper and its lending operations in the sector, set new goals, make mid-course strategy and policy adjustments, and determine what its posture vis-a-vis the political leadership should be in the near future. Such a stock-taking exercise could provide an opportunity to redefine objectives, reassess the effec- tiveness of alternative lending instruments, focus attention, redirect effort and resources, and elicit action on key policy issues to be decided at the highest levels of the Bank's manageri- al hierarchy, and strengthen responsibility and accountability. (iv) The Bank-country dialogue normally deals with central economic and policy issues. Given the centralization of the decision-making process and the thinness of the local elite, these matters ordi- narily can be decided only at the very top of the political struc- ture. This necessitates some equivalence of rank, i.e. the con- duct of the dialogue by high ranking Bank officials (RVPs, Direc- tors), to avoid the irksome difference of grade of the partici- pants in the discussion when the Bank is represented by relatively low-ranking managers or mission leaders, to show the genuine con- cern and weight that the Bank attaches to the dialogue, to make more authoritative and convincing the arguments put forward, and to enhance the chances of reaching purposeful agreements and ob- taining official commitments. Furthermore, it is advisable that high-ranking Bank managers visit more frequently the countries of their responsibility and conduct personally the dialogue at the highest levels of government, i.e. act as principals and not through surrogates, taking up the crucial issues directly with the inner circles of the political structure where the locus of power rests. More frequent contacts with the political leadership and persona: involvement (e.g., through visits, by organizing local seminars) would also make it possible for the Bank's top managers to obtain first hand and unfiltered knowledge on important policy and operational issues, understand better the nuances of the de.- velopment process and ;Lts determining elements in the country surroundings, suggest mid-course adjustments in the Bank's posi- tion and strategy in the face of changing circumstances, strength- en the Bank-country relationship and, conceivably, prevent the escalation of issues and abrupt policy shifts and stand-offs. The new organizational structure of the Bank, ensuring greater country focus at higher hierarchical levels, is conducive to such proac- tive attitude and initiatives. (v) The Bank's experience with Tanzania, where the President's unwa- vering support and instinctive commitment to the country's social- ist experiment set for a long and critical period the tone on policy matters and lending operations, suggests that the Bank's integrity and effectiveness of intervention can be compromised if the Bank's top Managers impulsively and pre-emptively make policy and lending commitments, and the technical staff is then called upon to make good on such commitments--thus allowing no room for airing important policy and technico-economic issues. Discretion- ary authority, an effective system of "checks and balances," and an internalization mechanism are called for to prevent erosion of the staff's professionalism, stifling of management and staff initiative and forthrightness, thwarting of dissenting views, and undermining of the credibility and effectiveness of the in-house decision-making process. Within the Bank's internal organiza- tional structure, a better balance between the power of the deci- sion makers and their moral obligation to respect the views of the technical staff who are called upon to implement their decisions- -between authority and responsibility--would therefore be of advantage to the institution, as it would minimize the likelihood of unnecessary error (e.g., failure to make use of information available somewhere in the Bank but not to the decision makers, access of decision makers to unfiltered knowledge and advice), or permit corrective action early on in the decision-making process. Furthermore, given the hierarchical and bureaucratic organization of the Bank, a two-way exchange and emphasis on the corresponding moral responsibility of decision makers would enhance the effi- ciency of the institution by heightening the staff's trust, dig- nity, loyalty, work attitude, pride, openness, and similar values. (vi) Policy-oriented economic and sector work is of critical importance in appreciating the modus operandi of key economic and institu- tional parameters, in identifying deficiencies in performance, institutional structures and policies, and in enhancing the Bank's capability to design workable policy reforms and purposeful proj- ects. Moreover, a deep understanding and appreciation of the cultural background, political realities, institutional arrange- ments, attitudes and, more generally, of the way decisions are taken and implemented in a country can lead to less dogmatic ap- proaches to institutional and policy reforms and more practicable solutions. However, in presenting their findings and recommenda- tions, sector (and economic) missions should question openly un- derlying conflicts between perceived ideological imperatives and what is realistically feasible; reinforce their conclusions by citing the lessons of experience of other countries with similar socio-political regimes which have employed analogous policy in- struments and institutional arrangements; refrain from expressina undue optimism, empathy and appreciation of efforts when not jus- tified, thereby weakening the argument and urgency for action; and present their findings unequivocally and concisely up front in the report, or in a separate document, and in an actionable form, to facilitate assimilation and decision-taking by top managers, oper- ational staff, and country officials. Finally, it is important that the issues identified by sector missions be pursued vigorous- ly by ensuing and more regular economic missions, or through other channels of communication, in order to maintain the continuity of the sector dialogue and enhance the chances of eliciting positive response and promoting a more orderly development of the sector. (vii) Poor project performance can be ascribed largely to the Bank's hasty and weak preparatory work and its inability to assess ade- quately the implications of Tanzania's socio-political and econom- ic conditions for project selection, design and implementation. The Bank's shortcomings in this area generally relate to inade- quate market analysis and optimistic demand projections; failure to respect the critical importance of scale economies (Mufindi pulp and paper mill) and to eschew project over-design (shoe fac- tory); and lack of a determined effort to assess the implementa- tion capacity of the borrower and to detect weak borrower commit- ment, as reflected in the borrower's inability or disinclination during implementation to carry out undertakings, e.g., provision of infrastructure. Poor outcomes in turn have reflected on the Bank and tarnished its image, and modification in the Bank's ap- proach and course of action therefore are called for. Since cove- nants committing the government to provide infrastructure in a timely fashion do not necessarily guarantee compliance, there is need to make sure early on that such facilities are being imple- mented and that progress is being monitored. Furthermore, project preparation and design need to be strengthened by more discerning diagnosis of what commonly goes wrong in the industrial sector and by devising imaginative and country-specific solutions. Haste to appraise and prematurely commit Bank funds in the absence of fun- damental preconditions assuring project success (e.g., partners with experience in export markets and a stake in the venture as in the case of the now idle shoe factory) tends to create difficult problems in implementation and operation. In addition, pressure to lend, when it does not lead to supporting unviable projects, does not allow sufficient time and effort to improve on project quality. Finally, closer link with sector work and greater empha- sis on detail during the preparatory phase is likely to ensure more innovative project designs, deeper analysis of the broader sectoral policies and issues impacting on performance, more accu- rate assessment of project risks, and better quality of apprais- als. (viii) Project aid often implies biases and perverse incentives: it encourages the recipient of tied aid to prepare (often with the help of suppliers of equipment) large capital and import intensive projects, to bypass thoughtful vetting, and to play off one donor against the other in the hope to get a better deal. On the other hand, the usefulness of external assistance can also be compro- mised when project aid forces the recipient country to substitute - 135 - the investment preferences of the donor for its own. Thus, hap- hazard infusion of external financing provides no guarantee for judicious industrial investment, while it might inadvertently legitimize ideologically colored but economically unsound indus- trialization policies. Furthermore, access to unmeasured external assistance may well hamper the industrialization effort because of its desensitizing effect on the recipient's behavior, in the sense that, by providing a cushion, it inadvertently weakens the deter- mination of the political leadership to undertake crucial reforms. Since aid does have an opportunity cost, bilateral and multilat- eral aid agencies need to act in concert to ensure congruity of purpose and consistency in policy reforms, impress the need for economic calculus as opposed to political calculus (access to resources at notional cost and easy terms) in investment deci- sions, coordinate the total effort and synchronize actions, pre- vent the transmission of conflicting signals and recommendations, and avoid duplication and dissipation of effort. (ix) The long-standing notion entertained within the Bank of working closely with the borrower and within the existing economic, social and political structures, to the end of inducing piecemeal changes, has not worked in the case of Tanzania. The results achieved heretofore hardly suggest that this approach has helped promote the long-term interests of the borrower--or of the Bank for that matter. This finding raises a number of sensitive issues affecting the Bank's posture and approach vis-a-vis its borrowers. * Should the Bank continue to support policies which are not working, cannot reasonably be expected to work in the circumstances, and which it is not in a position to influ- ence? Or, in the face of radical policy shifts or uncom- promising borrower attitude, should the Bank retrench, question the wisdom of its further involvement, and even suspend lending? * If ideology poses severe barriers to reforms, as the expe- rience with Tanzania strongly suggests, should the Bank adapt its strategy and lending unquestioningly to the ideological posture of the government, or should it press for a compromise, challenging the ideological underpinning of the political leadership's approach to economic and industrial development? * Since it is unlikely that a project will succeed in an unconducive policy environment, should the Bank be lending to sectors (or at all) when it disagrees with the strategy espoused by the government, is unable to exercise any influence, and the prevailing conditions are very likely - 136 - to impair project implementation and sustainability? Would lending under such circumstances be consonant with prudent banking policies and responsible development assistance? No doubt, these are difficult questions with no easy answers. Yet, they have to be addressed at a certain juncture of the Bank- country relationship since the Bank's image and prestige are at stake, and reflection on them would force more explicit behavioral choices. In the process of deciding, one thing is certains while the political economy of the reform process must be understood, the Bank can only help policymakers think through, assess alterna- tives, and formulate appropriate action programs based on sound economic and sector analysis. Beyond such advice and assistance, the Bank cannot intervene in the political process. However, to tread that fine line and strike a balance between antithetic views and interests would require thoughtful analysis, imaginative ap- proaches, and delicate negotiations involving Bank managers at the highest levels. - 137 - APPENDIX Page 1 of 6 THE BASIC INDUSTRY STRATEGY Theoretical Underpinning 1. Exercises and discussions over the appropriate industrial strat- egy--always within the socialist framework--were undertaken by the planners and their foreign advisors, with the common unifying feature being the systematic underplay of the fundamental constraints and weaknesses of the economy emanating from its stage of development and their potential impact on the effective implementation of such strategies. On the ideological plane, the Tanzanian leadership viewed "African capitalism" as impractical, the real choice being between 'external capitalism" and "internal socialism."1 Development was seen not only as a process of accumulation, i.e. of increasing the productive capacity of the economy, but also as a transformation of the institutional structure of the society. Tanzania's development, being pivoted around "man", required the introduction of ac- tivities that were basic needs oriented and favored innovatile processes. The economy was viewed as a "dependent" one and its structure by definition as "deformed." It was unable 'to generate self-sustaining development and an economic system that would display a reasonable symmetry between produc- tion and consumption structures. Attempts at modernization had been "proved" abortive as the industrialization pattern (import substitution strategy) implied merely the adoption of more sophisticated alien patterns of consumption (both private and public) without the corresponding process of capital accumulation and technical progress. In effect, the techniques embodied in the imported equipment were not related to the level of capital accumulated but to the demand profile of the modernized sector of the so- ciety, and the existing s stem of project selection by parastatals tended to perpetuate dependency. 2. According to this influential strand of thought, the import sub- stituting industrialization strategy as applied heretofore reinforced per- manent dependence and was incompatible with Tanzania's economic development strategy of self-reliance, since the introduction of new products and pro- cesses from the "center" will always require more sophisticated techniques and higher levels of accumulation than could be sustained. Under the basic industry strategy, import substitution would be incidental to and in con- formity with the strategy--not because the level of imports had reached a certain threshold. In the same vein, some of the activities implied by the basic industry strategy may be export-oriented, but only as the logical extension of the domestic market. Since development is a crucial element 1/ R.H. Green, "Industrialization in Tanzania" in M. Fransman, (Ed.', Industry and Accumulation in Africa (London: Heinemann), 1982, p. 86. See also D.P. Ghai, "Introduction" in Idem., (Eu.), Economic Independence in Africa, (Dar es Salaam: East African Literature Bureau), 1973, pp. xi-xxii; R.H. Green, "Economic Independence and Economic Cooperation," in D.P. Ghai, op. cit, pp. 45-87. 1 J. Rweyemamu, "The Formulation of an Industrial Strategy for Tanzania," Africa Development, Vol. VI, No. 1, (January-April 1981), p. 5. - 138 - APPENDIX Page 2 of 6 in effecting the transformation of the institutional structure of the so- ciety, the industrial strategy to be adopted should include certain key elements to foster the process.3 3. In the first place, there should be selectivity in industrial activities to be undertaken--establishment of basic goods industries to promote backward and forward linkages; promotion of engineering industries because they embody technical progress and can sustain autonomous indus- trialization on a sequential basis; and support for small scale industries which, though not necessarily labor-intensive, are an important carrier of technology transfer. Second, within the selected activities, techniques should be chosen not on the basis of their capital or labor intensity, bu'. on the basis of the level of sophistication that can be institutionally supported and the potential for acting as a catalyst for further institu- tion building and institutional diversification and integration--creation of new skills, capabilities, and organizational structures; contribution to technological autonomy and integration; compatibility with resource manage- ment at the national level; cooperation between research institutes and producing enterprises. Third, the strategy should further specify the requisite institutional, organizational and policy framework for the indus- trial sector, which involves inter alia public ownership and control, ef- fective participation of workers in the decision-making process of the enterprise, and the establishment of a policy on technology development. The latter should aim at creating an indigenous technological capacity for producing basic goods, regulating imported technology, promoting intercon- nections between indigenous technological activities and productive pro- cesses, increasing the local technological absorptive capacity, and foster- ing demand for local technology. Finally, the strategy should determine the sequencing of the various activities over the entire planning horizon to balance the basic needs of the consumers and the requisite industrial activities to be promoted.4 4. The basic industry strategy derives theoretical support from the "structuralist" approach to economic development. According to the struc- turalists, whose views have influenced the thinking of the political lead- ership in Tanzania, the manufacturing sector provides the "dynamism" for increasing productivity and growth, as the income elasticity of demand for manufacturing is considerably higher than that for agricultural products, generates static and dynamic economies of scale, and has an infinite scope for technological progress through a continual increase in specialization and the subdivision of productive processes, thereby imparting a technolog- ical dynamism to other sectors of the economy. The growth of manufacturing output is generally associated with a relatively more rapid rate of growth fj Rweyemamu, Ibid., pp. 6, 9, 13. For more details on the rationale for pursuing a socialist industrialization strategy to eliminate underdevelopment and to position the economy on a course of self- sustaining growth, as well as ior perceptions on the nature of an array of issues on planning and production relatirns, see Idem., Underdevelopment and Industrialization in Tanzania, pp. 175-198. J/ J. Rweyemamu, "The Formulation of an Industrial Strategy for Tanzania," loc. cit., pp. 6-17. 139 Page 3 of 6 in the capital and intermediate goods sectors than in the consumer goods sector, as the growth elasticity of manufacturing (i.e. the responsiveness of changes in manufacturing value added for a given industrial branch to changes in per capita GDP) is greater in capital and intermediate goods industries. Moreover, the lower the level of a country's per capita in- come, the gre5ter the growth elasticity of manufacturing. For both theo- retical and empirical reasons, a successful process of industrialization requires the building of an economic "structure" that is conducive to the raising of productivity throughout the economy. This requires, inter alia, that manufacturing value added constitute a significant proportion (around 25Z) of GDP, and that capital and intermediate goods production contribute substantially to manufacturing value added. 'his structure, together with the attendant development of human skills, aill provide the dynamism for increasing productivity and capital accumulation. 5. To create the necessary economic structures that are conducive to rising productivity, the state, far from planning according to the dictates of world market prices and comparative advantage, often will have to inter- vene precisely in order to insulate the economy from international market forces which in some instances may inhibit or even prevent the process of structural change. A specific example of such interventions might be the protection of "learning processes" which form an inherent part of the de- velopment of a country's potential to produce technological changes. Pro- tectionism, and its alleged ninefficiencies," far from yielding non-optimal patterns of resource allocation, are an inherer part of the process of developing conducive structures that will yiele optimal returns over the longer run. Therefore, from a short-run static iewpoint, encouragement of intermediate and capital goods industries wou" seem to conflict with eco- nomic rationalism. But from a long-range viewpoint, these are precisely the industries where income elasticity of demand is high, technological progress is rapid, and labor productivity rises fast. It is inadequate as well as misleading to judge the "efficiency" of a country's industry en- tirely by the difference between domestic costs and prices and internation- al prices. Any useful concept of economic efficiency requires proper con- sideration of much broader issues (e.g., high growth rates of per capita industrial production and consumption, high levels of employment, relative- ly stable prices).5 5/ Ajit Singh, "Industrialization in Africa: A Structuralist View," in M. Fransman (Ed.), Industry and Accumulation in Africa, (London: Heinemann), 1982, pp. 24-31, and M. Fransman, "Introduction," loc. cit., pp. 3-5. Page 4 of 6 Relation to Basic Needs 6. While the BIS focusses attention on the structure of production, the strategy is also concerned with the structure of consumption, sug- gesting an affinity with the basic needs approach to development espoused by the Bank.6 According to one analyst, there is a complex relationship between Obasic needs" and industrialization. Not only there is no neces- sary contradiction between the two, but fast expansion of industry is an essential condition for meeting the basic needs of the poor on a sus- tainable basis. The main reason for this is that, even allowing for con- siderable income redistribution, the achievement of basic needs in a typi- cal developing country requires a very large increase in national income. And although industrialization does not ensure that the basic needs of the people will be met, neglecting it or according it low priority means that they cannot be fulfilled on a long-term basis. Accelerated industrializa- tion and a substantially redistributive fiscal policy (i.e. a more equal distribution of gains from economic growth by fiscal means) must be the two pillars of any well conceived basic nends program. In this context, it would be wrong for a basic needs type strategy to withhold aid for modern industrial projects in intermediate and capital goods industries, if appro- priate criteria are met. As part of an integrated development strategy, such projects would lead to faster economic growth and a greater ability to meet basic needs on a susta: ed basis in the future.7 Critique 7. The implementation of the adopted basic industry strategy implied inter alia: creation of effective organizational structures and planning processes to ensure appropriate inter-sectoral (e.g., industry, agricul- ture, infrastructure) and intra-sectoral (e.g., rational project selection) 61 But the affinity is more apparent than real. Conceptually, the Bank's basic needs strategy is concerned with alleviating poverty by assisting the poor to become more productive. The Bank's notion of basic needs emphasizes better access to essential public services--basic education, primary health care, clean water and sanitation, coupled with better nutrition and shelter. The accent is on providing such services and increasing their affordability by lowering costs (e.g., greater use of traditional building materials and the adaptation of new techniques to traditional building methods), improving the policy framework (e.g., pricing policies, access to credit), promoting community involvement, or developing appropriate administrative structures. For the Tanzanians, in addition to catering to these needs, the notion implied a strong link with industrial development: provision of transport meant development of a transport industry to produce locally transport equipment; shelter was related to the development of a construction materials industry; clothing connoted the promotion of the textile industry; and rural development alluded to the creation of an agricultural equipment and implements industry. / Ajit Singh, "The 'Basic Needs' Approach to Development vs. the New International Economic Order; The Significance of Third World Industrialization," World Development, June 1979, pp. 585-606. -141 - Page 5 of 6 allocation of resources; the substitution of an adminAidtrative apparatus for the market mechanism to effect allocation decisions and implement pol- icy measures; development of public management capability to ensure effi- cient project implementation and operation: significant levels of invest- ment in relatively large, capital-intensive and import-intensive indus- tries; a constant and dependable stream of foreign exchange to finance imported capital and recurrent import requirements; a relatively large domestic market to reap scale economies; and an adequate supply of tech- nically trained manpower, including managers, to operate industrial under- takings--a tall order. 8. Considering Tanzania's level of development and the demanding requirements of a basic industry strategy in terms of planning and adminis- trative capacity, responsive institutional arrangements, organization, management, technical skills, foreign exchange, disciplined and motivated labor force, etc.; inability to take advantage of scale and external econ- omies and linkages from the grouping of neighboring economies, following the disintegration of the East Africa Common Market; the conflict that encouragement of basic heavy industry entails on the goal of appropriate technology; the questionable notion of availability of efficient, easy import-substitution industries; the anti-trade bias and disregard for the gains from specialization; and the lack of an indigenous managerialltechni- cal capability, it is arguable whether the BIS, in whatever variation, could have been efficiently implemented at the projected pace and time frame. The fact remains that there are insurmountable difficulties in pursuing an industrial strategy emphasizing the development of -n indige- nous capital goods industry at an early stage of industrialization (see also para. 39 in the main text). 9. The deepening of the industrial structure in the direction of intermediate inputs and capital goods faces a crucial threshold relation between market size and the economic size of plant, the degree of "techno- logical alienness"8 of the new economic activities in relation to the on- going ones mitigate the backward and forward linkage dynamic, the requisite levels of scientific knowledge and technical skills are not available, and a large amount of capital and foreign exchange are required due to scale and capital intensity of the producer goods industries. Even structuralists acknowledge that the extremely low level of development, the small size of the market, the attitudes and values of the society as a whole, the lack of an appropriate social organization and requisite insti- tutions to harness science and technology for industrial development, and 8/ For an elaboration of the differential technological impact of the "alienness" or "strangeness" on the strength of the linkage process and as a potential handicap, see A.O. Hirschman, "The Political Economy of Import Substituting Industrialization in Latin America,' Quarterly Journal of Economics, February 1968, pp. 23-24, and Idem., Essays in Trespassing, pp. 71-5. Page 6 of 6 the inability of the state to carry out effectively tasks of economic de- velopment are significant constraints on industrialization.9 Indeed, the performance of the sector heretofore has brought out forcefully the impact of these constraining factors. 10. In view of these severe constraints, it is debatable whether it is possible to start with capital goods as the primary focus in an industrial- ization effort, or whether better results would be achieved in the long run by concentrating on other subsectors for some time, in order to create a base while the capital goods sector is being built up from an initially marginal (e.g., spare parts) and peripheral (e.g., agricultural implements) position. Also, it is unlikely that there would be a sufficiently large spin-off from a domestic capital goods industry to offset the initial heavy surplus losses of having very expensive capital goods. Furthermore, it is questionable that a capital goods industry consisting basically of machine building without the supporting intermediate goods subsectors, which in this case includes an integrated iron and steel industry, would be a pur- poseful industrial strategy. Finally, given limitations on available na- tural resources and the small size of the market, production of a complete range of machine tools would necessitate the negation of scale economies and specialization.10 11. Finally, ridding the country of technological dependency would face insurmountable obstacles which were grossly underrated. Given Tan- zania's state of development and the rather sequential pattern of the in- dustrialization process, fundamental adaptations of technology to its re- source endowment is effectively foreclosed for a considerable time, as it would require indigenous capability to design and manufacture specialized equipment, produce inputs according to required specifications, overcome technical problems in the use of new products, etc., well beyond its capac- ity to provide. Perforce, industrialization initially would have to depend on the importation of machinery fmbodying existing technology and tested processes which, admittedly, are not conducive to sustained technological experimentation and training--a route that nonetheless Tanzania has no choice but to follow. 2/ E.g., Ajit Singh, "Industrialization in Africa: A Structuralist View," in H. Fransman (Ed.), Industry and Accumulation in Africa, 1982, pp. 32-34. 10/ R.H. Green, "Industrialization in Tanzania," in M. Fransman, op. cit., pp. 96-7. - 143 - TANZANIA WORLD BANK/TANZANIA RELATIONS, 1961-1987 BANK ACTIVITY IN THE TRANSPORT SECTOR I. BACKGROUND Introduction 1.01 The Tanzanian territory covers 945,000 km2 and contains a popula- tion of about 23 million (mid 1986), increasing at an estimated annual rate of 3.4%. About 86% of the people are widely dispersed in rural areas; and depend on agriculture, which accounts for 59Z of GDP.1 By contrast, the industrial sector employed 5% of the Labor Force and contributed about 10% of GDP in 1986, about the same as the transport sector (in 1982 transport and communications contributed 10.5%).2 Exports are still the traditional (coffee, cotton, cashew nuts, sisal, tea ard tobacco), and are produced far inland at considerable distances from the main port and city Dar es Salaam, where one-half the urban population live. This port also handles cargo for Zambia, Burundi Rwanda, Malawi--land locked countries separated by Tanzania from the Indian Ocean. 1.02 The development of Tanzanian transport becomes clearer when seen from a geographical and historical perspective. For more than half a cen- tury, the territory of present-day Tanzania was part of British East Africa. When Kenya, Uganda and Tanzania were granted independence in the early 1960s, they shared common services for taxation, customs, posts, telecommunications and transport. Centrally administered mechanisms set up by the colonial power were expected to continue to work smoothly under three different national governments. Indeed, it was intended that Kenya, Uganda and Tanzania would achieve increasingly higher levels of political and economic integration under the auspices of the East African Community; and the possibility that the three countries might not cooperate as planned was not contemplated. On the contrary, it was assumed that national gov- ernments would willinly give up portions of their sovereign authority to supra-national bodies, such as the East African Railways, the East AfricAn Airways, and the East African Harbours Corporations. It was also assumed that the regional planning approach to public investment and administration would continue to be possible under the three national governments. These were unrealistic assumptions. 1.03 The thrust of regional transport planning in East Africa was pri- marily manifested in a preference for railways over trucks. Intricate arrangements (including preferential tariffs and postponement of construc- tion of main road arteries, such as the Mombasa-Nairobi road), safeguarded for a time the pre-eminence of the railway, but road transport came into its own after Kenya, Uganda and Tanzania became independent. Two factors contributed to this. First, there was extensive road-building throughout East Africa. Second, railways ceased to be means of efficient transport 1/ IBRD: World Development Report 1988. (1985 and 1986 data). 21/ IBRDs Tanzania, Country Economic Memorandum, August 1984; Statistical Appendix. - 144 - and became instruments for polftical prestige or patronage, and in the process lost many of their advantages (reliability, safety, low cost over long distances) over, and much of their traffic, to trucks. Even so, rail- way survival has been prolonged through subsidies; and interminable studies on "intermodal coordinetion" recommended assigning by fiat traffic to the railroad--when, in fact, market forces dictated a massive shift towards road transport. 1.04 Up to the 1970s, Bank assistance to transport in East Africa sup- ported railways, regional transport planning, and intermodal coordination. Plentiful resources were allocated to the transport services of the East African Community, the presumed tool for the region's political and econom- ic integration. Much of the assistance extended to railways was justified on the grounds that Tanzania and Kenya were important corridors for land- locked countries further west (Uganda, Rwanda, Burundi, Eastern Zaire, Zambia, Malawi) and that the railroad would be the best means to serve this traffic. Most of these views, especially those related to regional inte- gration and to the role of railways, have not withstood the test of time. The East African Community collapsed and the East African Airways, East African Railways, and East African Harbours Corporations were broken up among the former Partner States, Tanzania, Kenya and Uganda. Railroads have not proven to be the best means of serving the traffic needs of either the landlocked or the coastal countries. In Tanzania, the railway cannot compete with road transport, regional transport planning has not worked, intermodal coordination is a utopia, and transit traffic of landlocked countries has not been properly served. In sum, earlier concepts informing the approach to transport development were based on assumptions that have proven to be unworkable. Future plans may benefit from a narrower and less ambitious scope. The Transport System--Description 1.05 Long-distance domestic movement of goods is characteristic of Tanzanian transport. The network also serves the external trade of Zambia, Malawi, Burundi, Rwanda and eastern Zaire. The transport system consists of: a road network of about 82,000 km; two rail systems which do not con- nect because their tracks are of different gauges (the Tanzania Railways Corporation (TRC), operating 2,640 km; and the Tanzania/Zambia Railway Authority (TAZARA), jointly owned by Tanzania and Zambia, with 970 km in Tanzanian territory); a port system (operated by the Tanzania Harbors Authority), centered in Dar-es-Salaam, and some minor ports and lake ports (operated by TRC); a national airline (Air Tanzania, serving domestic and international routes);3 several airEr (two international, Dar-es-Salaam 3/ The Air Tanzania Corporation (ATC), established in February 1977 after the dissolution if the East African Community, is under the jurisdiction of the Ministry of Communications and Works. It is the national airline and is expected to provide both domestic and international services. It owns one Boeing 707 and, in addition, it operates two Boeing 737s, 7 Fokker Friendships, and 4 Twin Otters. ATC's schedule includes about 19 domestic and 12-14 regionallinternational destinations. ATC has not operated profitably since it was established and has had to rely on extensive government subsidies since 1979. - 145 - and Kilimanjaro, and about 50 others, mostly unpaved); and a pipeline carrying crude oil and petroleum products frLn Dar-es-Salaam to Ndola in Zambia. This review deals with the principal modes: road, rail, port and coastal shipping. 1.06 Roads and Road Transport. Up to the mid-1960s, roads mainly pro- vided transport links where rail and lake transport were not available. In the mid-1980s there are 82,000 km of roads, subdivided into Trunk Roads (10,000 km); Local Main Roads (28,000 km); and Regional, District, and Agricultural Feeder Roads (about 42,000 km). Only about 3,000 km out of the 82,000 km are paved. District Roads are mostly earth tracks. Trunk road sections which have been paved are the Dar-es-Salaam to Arusha Road; the TANZAM highway from Dar-es-Salaam to the Zambia border; and the Morogoro to Dodoma Road. The TANZAM highway is the main road artery for Zambian traf- fic and constitutes an alternate corridor for Malawian traffic. Daily traffic on trunk roads is relatively light, usually not exceeding 200 ve- hicles per day. The most heavily trafficked arteries are the TANZAM high- way and the Dar-Arusha Road which, despite their poor condition, have sec- tions with up to 800 vehicles per day. 1.07 All paved roads are in poor condition, and most of the secondary and feeder roads become impassable during the rainy season. Road surface degradation, with concomitant increases in vehicle operating costs, has resulted from truck overloading because axle-load regulations have not been effectively enforced. More important, inadequate road maintenance has led to a steady deterioration of the entire network. In the early period after Independence, Tanzania espoused the policy of the colonial government to build "low cost roads'.4 However, not many feeder roads or aligned earth tracks were built.5 The quality of new road-building between 1962 and 1967 was lower than in the pre-Independence period because of two beliefs: that average axle load was less than two tons, and that there would be regular, timely and adequate maintenance. Experience has shown that neither belief was justified. 41/ 'Low cost roads" have been defined as follows at the Tenth International Road Congress in Istanbul (1955): "A low cost road is one which, having regard to considerations of climate and traffic, has been located and built to geometrical standards commensurate with future requirements, but has been constructed with bases and surface to meet the present traffic requirements. It is, however, one which should be so designed, constructed and maintained that it allows for stage construction when traffic requires it and improvement if economic conditions permit." 5/ Construction of feeder roads has not been a genuine priority of the Government. It was formally proclaimed in the 1960s when agriculture was being developed extensively, rather than intensively, because, given the rapid growth of agricultural production and rural population, transport was turning into a bottleneck. Proclamations continued in the 1970s, during the Government's uiamaa villagization; by then, however, agricultural growth had stopped and the increase in the vehicle fleet had also slowed from the 8.4 per cent annual rate at 1965-1970. The construction of village access roads on a "self-help' basis was another proclaimed policy in the mid 1970s but was not implemented. 1.08 Tanzania's vehicle fleet more than doubled during the first decade after Independence and by 1970 road transport accounted for most of the domestic traffic. Statistics for the next fifteen years are almost non- existent, and since 1981 the Central Transport Licensing Authority has ceased to compile new registrations. The vehicle fleet more than doubled between Independence in 1962 and 1972, and has since remained at around 100,000 units. Between 1962 and 1972, an increase of more than 130 per cent in government vehicles exceeded the rate of growth in the private sector. Between 1972 and 1980 the number of passenger cars fell by over 10,000 while the number of light commercial and other vehicles more than made up the loss. Up to 1972, the number of trucks also increased faster than the general trend, their average size more than doubled, and the per- centage with trailers increased significantly. The 1982 truck fleet num- bered about 13,000 units, including trailers with an average capacity of 8 tons per vehicle. According to the Central Transport Licensing Authority, the truck fleet in August 1983, comprised 11,282 trucks, 8,780 of which were privately owned (nearly 80 per cent of these by single truck owners). A few years later, because of spare parts shortages, the number of vehicles in operation was down to about 8,000 trucks, 1,500 trailers and 1,750 buses. The number of trucks, which is presently uncertain, may total any- where from 11,000 to 13,500 vehicles. About 80Z are privately owned, with the remainder belonging to parastatal organizations, including six Regional Transport Companies (RETCOS).6 1.09 Railways7 The Tanzania Railways Corporation, a parastatal or- ganization under the Ministry of Communications and Works (MCW), connects three port areas: Dar-es-Salaam and Tanga on the Indian Ocean; Mwanza and Musoma in Lake Victoria; and Kigoma on Lake Tanganyika. The network serves the regions which produce two thirds of all exported agricultural output and which contain almost all milling and processing facilities. The main commodities carried by TRC are petroleum, foodstuffs, cash crops for ex- port, fertilizer, and cement. For the last ten years there has been no pronounced change in traffic composition except that transit traffic for Zaire, Burundi and Rwanda has become increasingly containerized. In the early 1980s, annual freight traffic fluctuated between 1 million and 800 thousand tons. The Tanzania-Zambia Railways Authority was created after the 1965 Unilateral Declaration of Independence by Southern Rhodesia. In late 1967, the Peoples' Republic of China agreed to provide Tanzania and Zambia with assistance to construct a railway between the two countries. Construction began in October 1971 and TAZARA became operational in August 1976. A/ The figures in the paragraph above are rough estimates. 7/ Traffic carried by TRC and TAZARA over the 1981 to 1986 period is given below: (Thousands of metric tons) 1981 1982 1983 1984 1985 1986 TRC 1,028 926 791 916 952 989 TAZARA 732 754 782 913 1,043 921 - 147 - 1.10 Ports and Coastal Shipping.8 Dar-es-Salaam is the principal port. It is served by both rail systems, handles almost 90% of the coun- try's cargo traffic, and serves as a regional port for Burundi, Malawi, Rwanda, eastern Zaire and Zambia. Up to 1956, Dar-es-Salaam operated as a lighterage port and the lighters were discharged on quays adjacent to the city. These quays have been preserved, partly rehabilitated, and extended to serve lighter traffic, coastal vessels and ferries. 1.11 The ports of Tanga and Mtwara are served by coastal vessels from Dar-es-Salaam. The port of Zanzibar handles small coastal vessels and provides lighterage for ocean going vessels. Coastal shipping is handled almost in its entirety by one parastatal organization, the Tanzania Coastal Shipping Lines (TACOSHILI), while some minor activity between Zanzibar and the larger ports is carried out by private dhows. TACOSHILI serves Tanga, Mtwara and Dar-es-Salaam, with calls to Bagamoyo, Kilwa, Lindi and Mafia. Occasional shipments go to Mombasa and to Maputo, but this is rare. Most of the coastal traffic is with Tanga (cement and fertilizer). The traffic to Mtwara is mostly passenger because poor roads keep the area isolated during the rainy season. In 1985-86, total traffic carried by TACOSHILI was about 115 thousand tons. Most of the traffic is carried by Dar-es- Salaam, Tanga and Mtwara, all three of which are ocean ports playing a role in Tanzania's import/export traffic. The numerous lake ports, operated by TRC, serve transit traffic and domestic agricultural purposes. Performance, Dt. .--rminants and Status of the Transport Sector Performance 1.12 During 1970-80, the transport and communications sector accounted for about one-third of fixed capital formation in the economy, and trans- port equipment about one-third of the economy's investment in equipment. With fixed capital formation being between 18 and 22 percent of GDP the investment in the transport sector was a significant share of total product.9 The output performance of this and earlier investment in the sector depends on the the impact on the sector's capacity to render ser- vices to other productive sectors, households and the public sector, and on the demand by these users for its services. Because of the slowdown in growth after 1972 and the stagnation after 1978 the derived demand could have been a factor retarding performance of the transport sector by re- stricting usage of capacity, if such were available. Unfortunately, there is no independent measure of capacity for which Tanzania has generated any 8/ Rough estimates for ocean-going dry cargo port traffic are given below for Dar-es-Salaam and Tanga. The totals do not include petroleum imports (about 2 million tons in the late 1980s) and coastal shipping. (Thousands of metric tons) 1981 1982 1983 1984 1985 1986 2,000 2,300 1,600 2,000 2,000 1,600 9/ IBRD: Tanzania CEM 1984 Op. cit. Statistical Appendix. - 148 - data. However, expert observers have assessed this capacity and judged that transport has been a serious hindrance to the growth of other sectorst "After the scarcity of foreign exchange, transportation is perhaps the most serious bottleneck to output growthO; and 0Inadequate transportation is thus becoming a major cause of the accelerating deterioration of the economy".10 The Government's Structural Adjustment Program (1983-85) ap- proved by Parliament in June 1982 correctly identified the removal of the transport bottleneck as an important component of the recovery program. Determinants 1.13 The principal supply side factors which have determined the per- formance of the transport sector have been: (i) the deterioration of roads due to poor maintenance; (ii) the size, condition and intensity of use of the vehicle fleet; and (iii) the haulage capacity of the railroads. (i) Maintenance has never been a high priority for the Government, as may be inferred from the inadequacy of funds allocated.11 Personnel practices also hampered the establishment of a sound maintenance program. The rapid implementation of Africanization increased senior and middle level civil service posts filled by inexperienced local citizens (from 25% of the civil service at the end of 1961 to 66% by mid-1966). Difficulties of transition were aggravated by the departure of experienced expatriates. In- service training programs were either not provided, or were inadequate to meet the need for engineers and technicians. By mid-1967, there were 300 unfilled vacancies in the Ministry of Works, and skill shortages further undermined the effectiveness of already reduced expenditure on both maintenance and construction. The stress on self-reliance in the early 1970s led to an exodus of Asian professionals. Although the Second Five Year Plan, launched on July 1, 1969, accepted the need for foreign personnel in principle, expatriate departures exceeded arrivals over the next fifteen years. Road maintenance has been adversely affected by frequent changes in road administration over the past twenty-five years. From 1961 to 1970, District Councils were responsible for the 17,000 km of tertiary roads, while primary and secondary roads were maintained and administered by the Roads and Aerodromes Department (RAD) of the Central Government's Ministry of Communications and Works (COMWORKS). In 1970, with the abandonment of the District 10/ IBRD: CEM - Tanzania, 1984. 11/ After a firm of US consultants, commissioned in 1972 to review road maintenance, recommended increased efforts, Government, on the contrary, reduced the five year US$140 million maintenance program prepared by the consultants to a US$50 million total. Th3 point is that the Government's own emphasis on "low-cost" roads had increased maintenance needs, as did the general non-enforcement of legal axle load limits. - 149 - Councils, COMWORKS took over responsibility for all roads, includ- ing the tertiary network. Field operations were carried out through the Ministry's 17 regional offices, but unclear lines of authority resulted in lack of equipment and materials, and priori- ties set at headquarters were not always followed in the field. The regional offices were expected to fulfill all of COMWORKS' civil construction responsibilities (including airports, communi- cations and buildings), and this was partly the reason for reduced construction and insufficient maintenance. In 1973, after the formation of Regional Authorities and in line with the Government's general decentralization policy, the Regions were given full responsibility for carrying out maintenance on all roads. COMWORKS retained administrative authority over primary roads but the actual work was to be done by the Regional Authorities with funds from COMWORKS. In 1975, when COMWORKS was split into two ministries, the new Ministry of Works (MOW) was given administrative jurisdiction over primary roads and the un- clear lines of authority, which had prevailed in COMWORKS prior to the 1973 decentralization, now became even less clear. MOW had to work through the complex Regional bureaucracy for the implementa- tion of its programs and scarce maintenance equipment was fre- quently diverted to construction and other uses. Other problems, already in evidence when COMWORKS was responsible for all roads, were exacerbated after the split. For example, Regional Authorities were responsible for all sectoral programs within their regions. Consequently, demand on their capacity was more intense than on COMWORKS between 1970 and 1973 (a period during which COMWORKS was unable to maintain the road system properly) even though it was responsible for only three sectors, transport, communications and works. In late 1979, MOW again resumed respon- sibility for trunk road maintenance but it took over from the Regions inadequate work teams and insufficient equipment. Second- ary and tertiary roads remained the responsibility of Regional Authorities, and until 1984 continued to carry out some of the works on the primary road system although the MOW was administra- tively responsible for them and allocated funds to the Regions for this purpose. (ii) The main issues concerning the transport fleet are that there has been no growth in the trucking capacity since 1972, and that the inadequate fleet has been underutilized. For all categories of transporters, vehicle age and availability of spare parts have been significant determinants of truck productivity and shortage of replacement vehicles over the last ten years has led to an aging fleet. Fleet replacement has averaged an annual 4Z, much lower than the 15% judged appropriate for countries with road conditions as poor as those of Tanzania. The only reliable road transport statistics pertain to operations by RETCOS whose parent - 150 - agency, the National Transport Corporation (NTC), has a comprehen- sive reporting system based on monthly data by vehicle which al- lows annual aggregations. Large inter-regional operators, includ- ing RETCOS, achieve the highest productivity, with trucks regis- tering between 40,000 and 70,000 km per year, while other parasta- tals and small operators, achieve as little as 10,000 km per truck annually. However, RETCO traffic is but a small percentage of the total: private trucks haul 852 of the total traffic, other para- statals about 102, and RETCOS about 5%. Traffic hauled by road on non-Tanzanian trucks cannot be estimated with even approximate precision. 12 (iii) For Tanz&tia Railways Corporation locomotive availability has become a major problem in recent years. Although TRC has been supported by a number of bilateral sources which fund equipment, technical assistance, and training, efficiency remains low and the financial situation is weak. In the case of TAZARA, the railway has not been able to handle traffic demand because of inadequate locomotive capacity, due to the unsuitability of the available locomotives for the mountainous central section of the route, and to poor maintenance. Long turn-around times for wagons compound the difficulties, and TAZARA has been using Zambian wagons. While the introduction of additional locomotives could reduce wagon turnaround time, TAZARA would continue to experience operational problems without concomitant upgrading of its staff. Status 1.14 Despite voluminous resources invested in transport, much of it provided to Tanzania by both multilateral and bilateral aid agencies, the sector is in a parlous state. Neglect of maintenance has reached critical proportions, the road and rail networks require extensive reconstruction, and the quality of service is poor. Lack of data on vehicles and road condition do not permit intertemporal comparisons, but even casual inspec- tion of paved sections of the road network suffices to show that instead of improvement, deterioration has taken place. 1.15 The main reasons for the present status of the transport sector are: poorly trained and limited local manpower resources; the falling share of fixed capital formation in transportation infrastructure in total capital formation, especially by the public sector; poor rtilization of those resources; severe morale problems in the civil service; exodus of 12/ The Ministry of Communications and Works does not collect the border tolls for the transit vehicles and consequently has neither vehicle count nor vehicle load data. The Treasury does not release toll receipts, on the grounds it is collected in foreign currency. I - 151 expatriate staff and aversion to technical assistance; and economic poli- cies not conducive to development. While on the average, the sectors share of fixed capital formation was about one-third, the trend has been down- ward. During 1970-80 the sectors share fell continuously from over 40% to 25%. The share of Government development expenditure going to the trans- port and communications sector fell from 28% in FY71-74 to 4% in FY76 re- covering to only 10% in 1980. The transport sector, with its fairly high import requirement also suffered, along with the economy, a severe shortage of foreign exchange. Other reasons are related to factors which influenced Tanzania's development for the last two decades. First, the policy envi- ronment was determined within a public sector led strategy of development. Second, the interpretation of independence espoused self-reliance in man- agement of the economy, and led to a circumscribed role for expatriate skills and a negative preference for their use. Third, the expanded role of the Party in all national economic management decisions and in implemen- tation functions cut across traditional lines of control and command in the public sector. Fourth, vested interests, by taking advantage of systemic weaknesses, contributed to the spread of lethargy in the public sector, and corruption in most aspects of life. Finally, the environment, physical as well as domestic and international, visited drought, war and recessions upon the country's meager resources. Government's Transport Policy, Planning and Administration Transport Policy 1.16 The evolution of transport policy mirrored economic change over three periods--1960-67, 1967-73 and after 1973. Between 1960-1967, Gov- ernment expenditures for both investment and maintenance in transport which had already begun to fall before Independence, experienced a further de- cline, with maintenance the principal casualty as demands of other sectors increased. 1.17 The 1967-1973 period began with the Arusha Declaration and ended with the oil crisis. It was a neriod when the transport and communications sector was accorded very high priority and accounted for 43 perc.)nt of all fixed capital formation, reflecting major infrastructure projects--road, pipeline and railway.13 However, this was a period when transport policy was essentially subservient to foreign policy considerations. Tanzania supported Zambia against the Unilateral Declaration of Independence by Southern Rhodesia, which meant there was a large increase in the movement of goods to and from Zambia via Tanzania (particularly oil and copper) after the May 1968 embargo on trade with or through Southern Rhodesia. In response to the increased traffic, large investments were made to expand the port of Dar es Salaam (five new berths), improve the road to Zambia, 13/ IBRD: Tanzania Basic Economic Report 1977; Main Report page 56, para. 2.69. - 152 - and build a new pipeline and a new railway. Formal emphasis on rural de- velopment notwithstanding, there were few attempts to improve access to the rural areas and no significant construction of feeder roads. Sometimes, transport policy contradicted national economic policy by opposing the formally declared objective of decentralization.14 Also, the continuing heavy reliance on foreign aid and the failure to coordinate Conor assis- tance, created administrative, procurement and equipment maintenance prob- lems. Toward the end of the period, absence of clear and consistent poli- cies led to degradation of earlier infrastructure through lack of mainte- nance, to immobilization of vehicles through lack of spare parts, and to a deterioration in the qLality of services offered by the ports, railways, and air transport sub-sectors through erosion of staff commitment to their tasks. 1.18 Since 1973 transport policy has been mainly a by-product of con- cern over the protracted economic crisis beginning with the first oil price increase and major drought of 1973/74 with only slight relief due to the coffee boom in 1976177. The influence of economic concerns was to shift investment priorities toward the directly productive sectors (especially basic industry and food crops, away from infrastructure, in response to the slowdown in productive sectors and the near famine caused by the drought. This was reflected in declines in the share of the transport/communications sector in fixed capital formation by 25Z between 1975 and 1980, and in its share of Government's development expenditure by 852 between FY74 and FY76. In the allocation of scarce foreign exchange the sector was not favored; imports for maintenance and fleet replacement were postponed and availability of fuel was curtailed to the point where the sector could not conveniently service even the diminished derived demand for its services. Besides the influence of the economic situation, foreign concerns continued to be a factor. This time it was the breakup of the East African Community in 1977 which necessitated the creation of Tanzanian entities to oversee policy and operations of the railroad, ports, and domestic air services and airports, thereby extending Tanzanian government ownership and control in the sector. This was consistent with the overall national goals and strat- egy enunciated at Arusha, and which influenced transport policy in other ways, for instance Government participation in the trucking industry facil- itated by a World Bank loan to set up five Regional Transport Companies. Consistent with the Government's policy of decentralization of administra- tion after 1973 responsibility for planning, construction and maintenance of secondary, district and regional roads was shifted to the regional authorities leaving only truck roads to the MOW. Toward the end of the seventies, Tanzanian authorities began to recognize that after foreign exchange scarcity, transportation was the most serious bottleneck to output growth. Recurrent budgetary provisions for the sector, formally inadequate to meet normal maintenance needs, were substantially increased. The Struc- tural Adjustment Program prepared by Government in 1980 identified the 14/ For example, in the case of centralization of road maintenance between 1969 and 1973, and in the establishment in 1969 of the National Transport Corporation (NTC) to administer state transport operations. - 153 - removal of the transport bottleneck as an important component of the recov- ery program, and recognized that this would "require" increased recurrent imports to rehabilitate the existing road network and vehicle fleet; in- creased availability of fuel, better management of the various subsectors and improved transport planning. With this restoration of priority to the sector several donors came forward with assistance to recondition locomo- tives and renovate track beds, to rehabilitate the truck fleet, and to strengthen the newly created planning unit in the Ministry of Transport. Planning and Administration 1.19 Since Independence, a variety of planning arrangements have pro- liferated on paper, but implementation of these structures has been poor. The result has been that, while there has been plentiful financial and technical assistance from a variety of donors, resource transfers did not ensure sound operation of the sector. Transfers were uncoordinated, and conflicts among the positions on reform taken by different donors remained unresolved. Principal impediments to effective planning included: admini- strative weaknesses, lack of specificity of goals (i.e., the pursuit of "political" goals at the expense of "economically justified" projects); inadequacy of information base due to a general disregard for record keep- ing and statistics; and shortages of skilled manpower. There has been no intermodal planning;15 little planning of road transport,16 with no assessment of its social and economic costs vis-a-vis rail; and no coordination between domestic road policies and the policies of the neighboring countries. In the mid-1970s, the delayed Third Five Year Plan was little more than an aid shopping list. In 1978, the Bank noted that transport planning was weak in all relevant ministries. 15/ Intermodal planning was, and is, minimal and the Second Five Year Development Plan, published in 1969 by the Ministry of Finance and Planning, regarded EARC and EAHC investment and recurrent expenditure plans as exogenous. COMWORKS did not have any responsibility over the negotiations or construction of the Tan-Zam highway or of TAZARA. Within COMWORKS, several divisions were involved with various aspects of highway administration and the Staff Appraisal Report of the Fourth Highway Project describes in detail the unsatisfactory lines of responsibility and coordination. The publication in 1971 of the TANU Guidelines (Mwongozo) was more far reaching than the 1967 Arusha Declaration and increased worker and party control over management, blurring lines of control further and leading to a sharp fall in productivity. Then, in 1973, it became government policy to increase the proportion of public road transport at the expense of private operators. Between 1973 and their dissolution in 1976, many cooperatives had their own transport for such commodities as tea, coffee, maize, tobacco, and beer, while other parastatals also ran their own trucking fleets, many of which were idle for a large part of the year. 16/ The Bank's first economic mission to Tanzania in 1960 found there was little planning of the road network and minimal coordination between modes. - 154 - 1.20 Transport administration has been characterized by frequent changes, associated with public sector growth, decentralization, and disso- lution of EAC. In 1960, the Roads and Aerodromes Department (RAD) of the Ministry of Works (MOW) was responsible for the primary and most of the secondary road network, with local and district councils looking after the rest. RAD parent ministry was sometimes the Ministry of Works (MOW) and sometimes the Ministry of Communications and Works (MCW or COMWORKS). The Central Government assumed responsibility for new road construction and in 1969 took over responsibility for the maintenance of the 6,000 miles of district roads. The National Transport Corporation (NTC) was established in 1969 to administer the growing number of vehicles owned by state trans- port companies (most of which were at this time involved in trucking to Zambia). On July 1, 1969, the East African Railways and Harbours Corpora- tion was split into a separate East African Railways Corporation (EARC) and East African Harbours Corporation (EAHC). 1.21 A major decentralization reform in 1972 created the position of Development Director, appointed by the President with cabinet rank, to oversee all activity in the Regions and Districts. The regions were given the responsibility for all road maintenance (against Bank advice to cen- tralize maintenance), for secondary and tertiary roads, and for physical works on the primary network. The Ministry of Communications and Works (COMWORKS) retained administrative authority over primary roads but not the staff to perform its functions when its nucleus of trained personnel were dispersed throughout the regions. All regional staff theoretically reported to the Development Director. Decentralization in effect reduced the control over staff who no longer had to follow directions from parent ministries or have their work inspected and, as a consequence, the quality of work declined sharply. Output fell because of the diversion of equip- ment (maintenance equipment in particular) to other uses. 1.22 The progressive dissolution of the East African Community produced further changes. The Tanzania portion of the East African Railways ,orporation (EARC) was decentralized informally in 1974, and formally in 1977, when the KenyalTanzania border was c'osed. Earlier, in 1975, COMWORKS was split into two separate ministries: a Ministry of Works (MOW) and a Ministry of Transport and Communications (MCT). MOW had administra- tive responsibility for primary roads and airports. MCT had a theoretical responsibility for all other modes but, in practice, supra-national trans- port organizations remained quite autonomous, as did the Tanzania Railway Corporation (TRC) after its formal creation in October 1977, while TAZARA is administered separately, jointly with Zambia. In the course of prepar- ing the Fifth Highway Project, Government agreed to establish a central maintenance directorate within MO. In late 1979, MOW resumed responsibility for trunk road maintenance, but had little staff or equip- ment to work with. In July 1984 MOW was merged with MCT to form, once again, a combined Ministry of Communications and Works. The outcome of these successive changes was that planning and administration both of new works and of maintenance remained deficient, as did public sector trucking operations which had grown to one-third of all truck transport. 1.23 Rigorous economic analysis has not been a prominent element in planning and policy formulation. This is why, despite its importance for domestic transport, the respective roles of road and rail transport have not received the attention they deserve, and why policy implementation, beset with contradictions, had to be ineffective.17 For example, starting in the 1960s, Government restrictions on road vehicle licensing were strengthened in support of an unacknowledged policy to keep road transport subservient to the railway. This was not consistent with the considerable investments for trunk highways aud, in any event, regulations to curb the growth of road transport were flouted, partly because in Tanzania (as in Kenya), East African Railways used a differential tariff based upon what the traffic would bear. This tariff system, (plus the fact that heavy trucks did not pay the full economic cost for using the new highways), led to increasing traffic losses for the railway, eventually reaching levels far above those which the comparative advantage of trucks might have justi- fied. In this connection, it must be stressed that pricing distortions, and their undesirable effects, had been identified many years earlier,18 but all theoretical arguments that pricing should be cost-based were overridden by the ideological principle that rates should be low enough "to serve the public.n19 1.24 The conclusion is that, starting in the 1960s and continuing into the 1980s, transport policy formulation suffered from instability in the institutional arrangements for administration and from deficiencies in planning. Incompatibilities persisted between ideological desiderata (Africanization, equity, decentralization, worker participation) and opera- tional requirements (statistics, donor assistance coordination, workable staffing arrangements, planning, and pricing). There was no rationalization process to assign relative priorities to these goals, and to identify and accommodate their implications for each other. The alloca- tion of resources was not guided by maximizing principles, and the sectoral output and contribution to other sectors output and to GDP undoubtedly suffered. It is not clear that this loss was offset by the achievement of political objectives. The inadequate specification of political goals make it difficult to assess this achievement. 17/ For example, the economics of the Tanzania pipeline have not been thoroughly examined, although they must have affected the justification both of the Tanzania-Zambia highway and of the TAZARA. 18/ The East Africa Transport Study of the late 1960s recommended replacing the railways' discriminatory tariff with a cost-based one. TRC did not accept that recommendation until 1979, at which point it lacked the administrative capacity and statistical base to implement it. 191 Revenue from road user taxation was erratic: in some years, it greatly exceeded the often inadequate government expenditure on roads and, in other years, it fell below it. Passenger cars paid more than their fair share and, in the 1970s and 1980s, road user taxation was further complicated by a government policy giving preferential treatment to diesel-powered heavy trucks. II. BANK'S SECTOR WORK AND VIEWS ON POLICIES 2.01 Bank views on transport poliey are reflected in economic reports and sector memoranda. A report prepared in the late 1950s on Tanganytka concluded that the main transport network was approaching completion al- though some road sections had to be brought up to all-weather standards.20 At the time, the Bank was much concerned over duplication: the lines of communication by road, rail and water complemented each other and it was important to maintain this pattern as new links were added. Roads sho.uld not be built where railways already existed. With regard to the railways, the tariff structure was judged to be fundamentally sound but the railway should, nevertheless, be allowed more freedom to depart from this tariff and to offer special rates to increase its competitiveness with road trans- port. Furthermore, the impact of the increasingly important road transport on the railways reinforced the case for effective regulation. 2.02 Subsequent reports, prepared in the 1960s, showed that the Bank was concerned not only about road-rail competition but also about competi- tion within the road transport industry, between *public" and "private" truckers. One suggested remedy was a revised licensing policy and better enforcement of the regulations. Neither the perils of regulation nor com- petition were overlooked, and emphasis was placed on encouraging African enterprise and preventing public carriers from stifling the growth of local (and mostly private) transport services. Regulation was to be reserved for the main long distance routes where public carriers operated; complex and restrictive regulation of local transport in the rural areas was regarded as both impractical and undesirable.21 Bank missions to Kenya and Tanganyika in the late 1950s and early 1960s regarded the railway as "the 20/ The chapter on transport noted the widespread belief in Tanganyika that lack of transport facilities is holCing back the economic development of the territory. The Mission's investigations suggested this view to be commonly exaggerated. The main routes by land, water and air reach every corner of the territory, and their capacity were underemployed. Deficiencies were largely local in character, affecting inland access routes to the main transport arteries. 21/ The reference to the basic acceptability of the value-for-service tariff by the Tanganyika mission was implicitly an acceptance of cross- subsidization between traffics, with high-rated commodities under the tariff subsidizing the low rated. The report on Tanganyika refers briefly to losses on branch lines, but not in the context of the railway tariff, and it makes no reference to the possibility that the main lines in Kenya were subsidizing both main and branch lines in Tanganyika. I backbone of freight transport, especially for long-haul traffic." Roads were to be built and improved to "reflect plans and prospects for the fur- ther development of agricultural output" and to 'facilitate farm to market movement of produce and distribution of goods.'22 The highways program was to focus on roads important for agricultural development and this was to involve a shift from main roads to primary and secondary all-weather roads, including those needed to move the tea harvest to the factories. 2.03 Thus, more than twenty-five years ago, the Bank identified many of the concerns featured in subsequent policy work. These issues include: pricing (railway rates and subsidies); investments (main roads vs. rural roads); and management (road maintenance and road transport licensing). Thought was given to the optimal allocation of traffic between rail and road and to the enforcement of truck weight and speed regulations. At a time when maintenance had not yet become a major problem, the Bank pointed to its significance and identified the related problem of axle load con- trol, foreshadowing much subsequent attention. The Bank did not anticipate how intractable these problems would turn out to be, nor did it anticipate the extent to which investment for main roads versus investment for rural roads would be tied to the maintenance issue. With hindsight, the connec- tion is obvious: if maintenance is weak, or totally absent, road recon- struction would take up more of the investment budget in future years. Hindsight has also shown that weak maintenance reduces the developmental impact of rural roads which, if not properly maintained, quickly become impassable. 2.04 Similarly, the Bank did identify the problems created for the railways by the expansion of road transport but the full implications were not properly grasped. Pricing was an unquestionably important variable but the fundamental problem, in Tanzania and elsewhere, was the railways' abil- ity to compeLa and to stay in bt `.ness. However, the Bank's policy work in the early 1960s did not envisage the large transport investments needed to keep up with population growth and socio-political change, and did not anticipate the personnel difficulties which would arise after the departure of European and Asian staff. To have done either, Bank staff would have had to be exceptionally clairvoyant because, at the time, population growth and domestic politics did not seem to entail great difficulties. 2.05 While the evolution of Bank views on transport policy and, in particular, of the underlying intramural debate, cannot be traced in de- tail, the CPPs provide some insight. The year 1970 seemed to have been an important turning point in Bank views. Although the Tan-Zam Highway had absorbed most of Bank lending to the sector up to that point in time, the intention was expressed that in the future, lending would be increasingly concentrated on secondary and feeder roads which would directly serve the needs of the productive sectors, particularly agriculture. By the 22/ The Kenya mission's allocation of a lower priority to the trunk rvads, (including connections to Uganda and Tanzania), is worth emphasizing, particularly in the context of the East African Community. Low priority was %ssigned to the nain road to Tanganyika (the Nairobi- Arusha road), although the conclusion should not be drawn that the delayed improvement of this artery was in response to the views of the mission. following year Bank concern turned to the inadequacy of maintenance capa- bility of RAD of COMWORKS caused by shortage of staff, and to the possible role of subsequent Bank lending in providing technical assistance and training. Yet, the priority of rural roads remained strong, and in 1973 when note was taken of a study of trunk roads, including the Dar/Mtwara to which the Government was expected to give high priority for security rea- sons, there was no mention of Bank involvement. In 1974 the Bank took note of the nationalization of important road transport elements and of the need for urgent clarific&tion of the role of private entrepreneurs. Its view of the appropriate priorities for road transport was set out plainly in the CPP of that year--(i) expansion and upgrading of feeder and secondary roads, (ii) improvement of maintenance (iii) development of an efficient road transport industry. This last item no doubt reflected concern that not only were private operators being forced out as cooperatives were pushed to provide transport services and were later replaced by parasta- tals, but that the state owned National Transport Company was poorly man- aged and the utilization and maintenance of trucks sub-standard. The 1975 CPF explained why the Bank's priority for feeder and secondary roads was not reflected in the dialogue and lending program for transport. It noted that financing of feeder roads had been included in nearly every loan or credit for the agricultural sector, and that this had been found to be the most effective way of expanding rural roads. Meanwhile the transport sec- tor operations focussed on a comprehensive highway maintenance program. y 1976, Bank views were reflecting a keener awareness of the important trans- port sector institutional and policy issues that had emerged and/or were emerging. This awareness was due in part to the fact that the GOT had asked that the Bank participate in an overall sector review especially of EAC related transport. Earlier the GOT had approached the Bank for assis- tance in its attempt to address problems of the trucking industry. The issues viewed by the Bank as being most important were: (i) organizational weaknesses relating to road maintenance; (ii) ways to increase road user charges; (iii) relative investment priorities between various modes. 2.06 After 1977, Bank views on the transport sector reflected the Transport Sector Memorandum of that year, which was precise in stating what the primary objectives and strategy should be. The main changes occasioned by this work included emphasis on (i) the need for better planning and coordination; (ii) preservation of the railway infrastructure, through rehabilitation investments and training of staff, and provision of better service; (iii) the liberalization of the importation of motor vehicle parts, monopolized by the State Motor Corporation created in 1974. This memo presented a more strident picture of Bank views. On the trucking industry it stated that, "Despite Government's declared policy of increas- ing public sector participation in road transport, existing private trans- port operators should be allowed to continue their operations which are sound," and that there was "need to redress deterioration in fleet associ- ated with delayed renewal caused by foreign exchange scarcity and Govern- ment's policy of discouraging private services." On road maintenance, the memorandum called for a clearer delineation of responsibilities between central and local authorities, and more adequate funding of maintenance operations, pointing out that while MOW maintenance load had doubled since 1970 maintenance allocations were unchanged in money terms in spite of - 159 - higher costs. The CPPs subsequent to this memo continued to echo the im- portance of strengthening transport planning capability, justifying this focus, initially, by the fact that the collapse of the EAC would imply a shift in responsibility from the regional to the national level, and, la- ter, by the need for better intermodal balance in investments. 2.07 The 1985 Transport Sector Memorandum did not add much to the 1977 in terms of policy measures considered necessary. However, by 1985 Tanzania had been several years into its most severe crisis of scarcity of foreign exchange, and the allocation of foreign exchange became an impor- tant policy tool. The memo called for an increase in the allocation to the transport sector, to permit it to buy spares to keep its rolling stock in repair. The continuing concern over railroads was reflected in the recom- mendations (i) that Tanzania bring railway rates (which had not been cover- ing costs) closer to commercial rates and (ii) improve utilization rate of wagons and (iii) interface with roads. These more reflected the Banks increasing concern over the deterioration in the financial condition of parastatals than its concern with low utilization of railroad capacity and the need to make it more competitive with road transport. 2.08 This picture of an almost orderly evolution of Bank views on Tanzania's transport sector obscures the sometimes sharp differences of opinion within the Bank over transport sector issues and policy. General- ly, there was one side which argued for vigorous application of traditional investment criteria, while the other side insisted that major socio-politi- cal objectives ought to take precedence over the strictly economic. Thus while the 1977 CPP noted that the Government was addressing the problems of the trucking industry and that a mission had just done a pre-appraisal of a proposed trucking project; the 197' Memorandum was critical of the Govern- ment's policy of discouraging private transport operators by not allowing liberal access to spares. Even the Bank's long standing support for feeder roads was disputed by some who felt the emphasis should not have been to facilitate integrated rural development but should have been oriented toward export agriculture. In the absence of appropriate institutional managerial arrangements to resolve these and other differences, fundamental conflicts of views among Bank staff visiting Tanzania led to expressions of widely different opinions to the client. This introduced an element of confusion into an already difficult policy dialogue. III. THE POLICY DIALOGUE 3.01 The policy dialogue in the transport sector has been conducted within the context of lending operations. Since 1964 the Bank's lending program would allocate a certain amount of resources to the sector, ana Bank staff were encouraged to listen to Governments requests and respond with "appropriate" projects. Staff tried to do so, incorporating economic justifications and conditionality as and when -ray could. After the Transport Sector Memorandum of 1977, Bank-supported transport projects were abundantly equipped with covenants, conditions, and action plans in the hope that Tanzania would take what the Bank considered suitable action. However, project identification and preparation, and policy planning re- flected efforts of Bank staff and Government technocrats, but implementa- tion was subject to political intervention and expedience. The result has been significant shortcomings for Bank supported projects and the policy dialogue alike. The Bank's official unwillingness to make the distinction between the Government and the Party and to treat with both has circum- scribed the realism and effectiveness of the dialogue by omitting the im- portant political dimension. This becomes especially clear from the fact that the main problem issues in the dialogue have had a significant ideo- logical aspect. 3.02 The principal problem issues have arisen where Bank views have been the opposite of the policy directions dictated by political considera- tions. These were not issues related to objectives since the Bank and Tanzania were in agreement on the need to have good construction, regular maintenance, strong institutions, and the creation of a transport planning capability. The differences were over the strategy for pursuing these objectives and the main issues were: (i) the staffing of the highway agen- cy; (ii) the centralization of maintenance; (iii) liberalization of road transport (iv) investment priorities, specifically some large projects e.g. Dar airport; and (v) tariffs and intermodal competition. The Bank was for long concerned over the weakness of MOW from the departure of expatriates after Independence and more so after the decentralization of administration to the regional directorates and the departure of Asian technicians in the early seventies, which had impeded the capacity to plan and implement the development of the transport sector and to maintain it. The Bank felt the situation could be improved through technical assistance and training and by more rational use of the resources available for maintenance. This meant "depending" on foreign experts and assigning foreigners to senior policy and technical positions which ran counter to the Tanzanian policy of "self-reliance" and "Africanization". The re-centralization implied a more rational use of maintenance resources, and was clearly in opposition to the decentralization and integrated regional development approaches preferred by Tanzania, (with the regional approach also supported by the Bank). The liberalization of road transport, implying few controls on private sector operations and access to inputs, clearly flew in the face of a strategy based on controls over private sector activity and public sector dominance. Similarly the Bank's effort to convince Tanzania of the desirability of cost based tariffs could not gain acceptance in a society wedded to uniform national pricing of goods and services through a system of cross subsidies. With su,h fundamental ideological differences between the Bank and Tanzania it is not surprising that the dialogue and the closeness of the relation- ship depended on the varying degree of obstinacy on both sides. 3.03 Up to the early seventies the Bank refrained from expressing an unequivocal opinion either on the soundness of the Tanzanian approach, or on the compatibility of its principles and those of the Tanzanian leadership, although in its official reaction to the Tanzanian Strategy23 in 1972 it was parsimonious with praise and explicit with warnings. 23/ IBRD: The Economic Development and Prospects of Tanzania, 1972. - ." ' - I However, these admonitions were impaired because, although the country was in a state of institutional instability for a quarter of a century, the Bank's official position, as reflected in published documents, seemed to imply unrealistically24 that the instability was exogenous to transport, that it did not really affect the integrity of institutions, and that the commitment by Tanzanian civil servants to development remained strong. Evidence which began to accumulate since the early 1970s clearly indicated that the effectiveness of Government agencies was being eroded. Equally evident was the fact that the commitment of Tanzanian civil servants left much to be desired because they were and still are poorly paid.25 In its policy dialogue, the Bank would not openly concede that ideological objectives and Party priorities were causing major economic distortions and that, consequently, design and implementation of sound policies would take a long time. 3.04 During the next four or five years after 1972 the Bank became more accommodating to the views of the Tanzanian leadership, and relaxed the strict efficiency criteria for project identification and preparation.26 Relations between the Bank and Tanzania became closer, as evidenced by the approaches made to the Bank for assistance in solving the trucking industry problems and the transport problems associated with the breakup of the EAC. 24/ Presumed assumptions are hard to document because they are not reflected in specific statements but implied in tone and style. The following excerpts from the President's Report on the Sixth Highway (Rehabilitation) Project may illustrate the Bank's attitude towards Tanzania as late as April 1986: "The new priorities, enunciated in the Arusha Declaration and related policy statements, were directed toward establishing a socialist society, with emphasis on broad-based rural development, self-reliance in development efforts, and mass education. To accomplish these ends, the State, with guidance from the Party, was expected to play the leading role, especially in the reform and creation of appropriate institutions. This led in the late 1960s to the nationalization of large-scale industry, commerce and finance, the creation of numerous parastatal bodies, the formation of Ujamaa (cooperative) villages, the decentralization of Government (1972), and the mass campaign of villagization (1974-1976). "Despite some disruption arising from these major institutional changes during the period, Tanzania managed to show improvements both in social welfare and in macroeconomic performance... 25/ In mid-1988, the monthly take-home pay of senior staff does not exceed the equivalent of US$50. 26/ The Third, Fourth and Fifth Highway Projects are cases in point. 3.05 After the 1977 Transport Sector Memorandum, the Bank became more aware of the problems in the sector and had firmer ideas regarding their genesis and solution. Bank dialogue aimed at the institutional weakness of the main ministries especially MOW, which became evident from effort to restart a highway maintenance program that had run into implementation difficulties. In 1979 this seems to have been interpreted as personal criticism of Tanzanian technocrats and resulted in the breakdown of commun- ications between MOW's senior staff and Bank staff.27 The pro-forma "policy dialogue" continued but, against the background of a change in official Bank attitude to Tanzania's development strategy. The 1980 CPP stated that "there is finally a need to recognize better than in the past the limita- tions of centralized administrative control over the economy. The existing institutional framework of governmental and parastatal organizations must be consolidated and made more efficient. The over-extended system of con- trols has to be cut back and rationalized to economize on the use of administrative manpower--the scarce resource, and to provide the incentives necessary to stimulate desirable increases in production." It was prepared to shift its investment in the transport sector to agencies which were formerly EAC corporations and which were perceived as having a stronger nucleus of trained staff than the Ministry. As the economic crisis contin- ued into the 1980s and the deterioration of physical infrastructure contin- ued apace, the emphasis of the dialogue turned to the need for rehabilita- tion which was regarded as essential to the recovery process, and which proved less contentious than the maintenance and implementation issues of 1979. Bank efforts to convince the MOW to accept assistance to improve its planning capacity and data base were positively received. Fairly recent Bank documents (for example, the 1985 Transport Sector Memorandum), being 27/ In April 1980, at the request of the Division Chief, the four staff members of the Bank's Highway Division dealing with Tanzania prepared a 14-page internal memorandum on "Continuing Problems in Project Implementation". The Fourth and Fifth Highway Projects, and the Trucking Industry Rehabilitation and Improvement Project were discussed in detail but the thrust of the argument was on Bank-Tanzania relations. Para 11(b) reads in part: '... Communications between MOW's senior staff and the Association have now broken down. During four different visits to Tanzania made by IDA staff and consultants between June 1979 and February 1980 none of the staff or consultants was able to obtain an appointment with appropriate senior MOW officials for the purpose of discussing and resolving the project's problems....during the most recent mission (the Principal Secretary) in response to IDA staff efforts to help identify and resolve the project's problems, bluntly stated that he did not want his Ministry run by Bank staff from Washington." - 163 - crisp and carefully crafted, make no explicit refe.rence to difficulties encountered by Bank staff attempting to conduct a policy dialogue on sub- stance. They decline to acknowledge that no congruence has been achieved between Tanzania and the Bank on transport policy, and they exorcise major problems with anodyne statements: Government lacks a long-term strategy for the transport sector.28 3.06 With hindsight, it is possible to point at the basic reason why policy dialogue on transport has been difficult. From the early 1960s, the Bank's advocacy of free market solutions has been clashing with the Party's socialist rhetoric, with the discriminatory measures favoring public sector and parastatal provision of transport, and with Tanzania's attitude towards private and expatriate contractors and consultants. The Bank professes that its market-criented philosophy can be applied in socialist settings without overt attempts to reconcile the two positions. Over the years, attempts have been made to compromise, but the impasse has not been re- solved. Notwithstanding assertions that solutions will soon be found, unresolved differences of opinion on transport policy abounded in the Bank, especially during the 1970s. The contrasting attitudes and shifting posi- tions of the Bank and its staff must have mystified local officials. Cer- tainly they did nothing toward encouraging a fundamental revision of the Party leadership's ideological orientation. On the contrary, the mixed 28/ The three-volume 1985 Transport Sector Memorandum sacrifices description for delicacy. Para 27 of Volume 1 reads as follows: "The Government has not yet defined a long-term transport sector strategy aimed at the pending problems in the sector and to further the sector's vital role in economic development. The Government's short- term objectives in the sector are set out in the country's general fivc-year development plans. The planning document for the period FYs 1982-86 is still being prepared." In translation, this says that after more than twenty years of centrally planned efforts, extensively assisted by the Bank and the Donor Community, Tanzania had still has not settled either on investment or on policy priorities. It also says that the five-year plan for 1982-1986 was not ready in 1985. signals made the Bank appear vacillating and ineffectual and weakened the stance of negotiating project officers and the quality of the dialogue. This does not now, nor did it ever, augur well for the development of clear policy options. Differences of opinion continue between the Bank and the Government abound not only on general policy29 but, also, on investment analysis,30 on transport administration, on the use of consultants, on staff training,31 on highway maintenance,32 and on transport planning.33 These are areas where the Bank and Tanzania have theoretically been 29/ Solution of policy problems related to the railway tariff and to intermodal competition are crucial for the sound operation of the sector. However, despite debates carried out over the past two decades, statistics are so meager that it is impossible to do more than note that the problem still exists. The Bank has argued for a liberalization of road transport from the early 1960s and, recently, one encouraging sign is that while entry into the industry remains officially restricted, it is not effectively controlled. Pricing of interregional trucking services is largely negotiated although some maximum official rates do exist. 301 The Bank has disagreed strongly with the Government over some large investments. In the 1985 Transport Sector Memorandum, the Bank notes its disagreement with the Government's proposal to lengthen the runway at Dar es Salaam airport, which the Bank reckons to be adequate for all present and likely future needs. The Bank also disagreed with the Government's decision to essentially rebuild the Dar es Salaam international airport. 31/ The question of staffing the highway agency was a source of contention between the Government and the Bank over many years. The SAR of the (first) Highway Project (Report No. 70-396a) of January 21, 1964, stated (para. 15) that "the pressure to Africanize government departments which is increasing gives rise to some concern over the rate at which Africans are being promoted over non-African employees with longer experience" and (para. 16) that the roads agency would therefore "be dependent on expatriate administrative and technical personnel for many years." The Government was guided by general policies of self-reliance and opposed the use of expatriates and consultants which the Bank urged to run the roads system. The Government resisted these urgings on the ground that they would subvert its basic policy and the Bank succeeded in persuading the Government to its view only in 1980. 32/ The Bank cut across a clearly enunciated Government policy when it urged centralization of road maintenance. The Bank believed that the technical needs of maintenance could only be met in this manner but, subsequently, it had to compromise. Maintenance operations remain a major problem. 33/ Despite the Government's commitment to socialism, it has not, in the Bank's view, formulated adequate plans for the transport sector, mainly due to shortcomings in the personnel assigned to planning work. - ;65 - collaborating for many years. But a recent study,34 which reviewed Bank activities over several decades, has revealed, once again, considerable divergence between the policies advocated by the Bank and those followed by Tanzania. While it should not be expected that the Bank is decisive in determining policy within any country a constructive interchange of ideas should narrow the gap over time. This has not happened to any great degree in this case. Impact of the Dialogue 3.07 The nearly twenty-five years of Bank/Tanzania dialogue on trans- port policy has borne some fruit. In respect of pricing, road transport has been de-licensed (though the public sector preserves a major presence); some price anomalies have been eliminated but significant distortions re- main. Price of interregional trucking services is largely negotiated al- though maximum official rates exist. There are fewer major transport in. vestments of doubtful economic worth. While entry into the road transport industry remains officially restricted, it is not effectively controlled. These successes, albeit limited, have been due to the fact that the Bank's voice was one of several, both domestic and foreign, representing the classical concern for efficiency, competition, technical optimality, and economic feasibility. The role of domestic pressure was particularly im- portant. The combination of vocal private operators who enjoyed political backing and of weak regulatory agencies did more for road de-licensing than any number of policy statements by either the socialist government or the capitalist Bank. Economic circumstances have also encouraged changes in the direction advocated by the Bank. Foreign exchange and budgetary strin- gencies have become the critical factors in investment decisions. Resource shortages caused by larger forces both inside and outside the country, drought, war and world recession, have all tended to focus the minds of Tanzanian leaders on the need to pursue efficiency in the use of resources in all sectors, including transport. 3.08 However, some management concerns raised by the Bank at the time of Independence are still on the agendas planning, maintenance, staff training and overloaded vehicles. This situation argues for a review of Bank strategy in its policy dialogue on transport issues, to ascertain whether the Bank asks the right questions and formulated and argued its points convincingly. If results alone were the test, the answer would be negative. But, and this bears repeating, the Bank does its work in the midst of many powerful internal and external influences. The Bank's con- cern with economy and efficiency accorded with its statutory role, and the emphasis on better management is in line with this concern. However, it is not clear that the Bank's effort to promote specific approaches to invest- ment analysis were appropriate. The debate over railway tariffs suggests that not only did these approaches raise theoretical problems, undermining the persuasiveness of the advice offered, but they created demands for data and administrative resources which could not reasonably be met. Here, the lesson is clear: if prescriptions lack apparent reasonableness and if they do not accord with what a government sees as necessary, they are not likely to be accepted. 34/ Policy Review of Institutional Development in Africa, Case Study on Tanzania Highway Projects (World Bank, 1984), Report No. 5085, pp.137-212. - 166 - IV. BANK LENDING AND ITS IMPACT The Lending Program 4.01 For the last thirty years, the Bank has extended, initially to the East African Community and subsequently to Uganda, Kenya and Tanzania sep- arately, financial and technical assistance to strengthen transport ser- vices. Since 1964 the Bank Group has financed the following projects in Tanzania: Loan/Credit Year Amount ERR Approved Project Name (US$ million eq.) Estimated Reestimated 1964/68 Highways I 17.0 11% li 1969 Highways II 22.5 20% 16% 1969/72 Harbors II/III 61.5 NA NA 1970 Railways III 42.4 NA NA 1971 Highways III 6.5 141 4? 1974 Highways IV 10.2 22% Negative 1977 Trucking 15.0 37% 10% 1979 Highways V 20.5 70% 10% 1984 Ports Rehabilitation 27.0 33? Ongoing project 1986 Highways VI 50.0 38? Ongoing project 4.02 In the 1960s, Bank loans to the port and railways corporations of the East African Community were to support the economic integration of Kenya, Uganda and Tanzania and to develop and modernize transport infra- structure in the three countries. Bank assistance to the Tanzanian trans- port sector was initially extended through the First Highway Project in 1964 and has since encompassed all modes except air. Completed and ongoing projects are summarized in the following paragraphs and described in detail in the Appendix C (Tanzania: Notes on Transport Projects Financed by the Bank, 1964-1988). The Revised First Highway Project: Credit 48-TA; Credit 115-TA of 1964 would finance construction of eight road sections. Because of imple- mentation delays and increased costs, the Bank reappraised the proj- ect, and provided a US$3 million supplementary credit in 1968. The project suffered from major delays and cost increases, largely because of scarcity of technical staff. The Second Highway Project: Loan 586-TA; Credit 142-TA of 1969 would assist the construction of two sections of the Tanzania to Zambia highway. The project was economically successful only because of fortuitous circumstances. An important issue on distribution of bene- fits was raised upon completion: at appraisal it was expected that most benefits (85?) would accrue to Tanzania; on completion it has become evident that 45? of the benefits had accrued to Zambia. East African Community -- Second and Third Harbour Projects: Loans 638 and 865-EA of 1969 and 1972 were for development of facilities in Mombasa, Dar es Salaam and Tanga. They registered significant over- runs in both time and cost and the audit suggested that future proj- ects ought to emphasize productivity, manpower development, and main- tenance. East African Community -- Third Railway Project: Loan 674-EA of 1971 was to rationalize tariffs and bring down costs. Implementation of a cost-based tariff was not achieved because the economic, political and operational consequences of implementing the new tariff were not fore- seen. Implementation of the cost reduction programs failed to materi- alize because of the Borrower's financial troubles, caused by politi- cal differences among the Member Governments in the East African Com- munity. By the planned completion date only 5 per cent of the project had been carried out, and it never was completed. There were substan- tial cost increases though the extent of the overrun could not be documented in the absence of adequate cost records. The Third Highway Project: Credit 265-TA; Loan 586-TA of 1971 (later supplemented by a transfer of a US$1.9 million surplus from the Second Highway Project) would construct a section of trunk road in the south- ern part of the country; improve and maintain feeder roads in the north; and strengthen maintenance. There were significant cost over- runs; improvement of feeder roads was short lived; training of Tanzanian engineers met with limited success; and the staffing program was not carried out. The goal of stimulating agricultural production was not met. Investments in the trunk road had a 4Z rate of return, and those in feeder roads had negative rates of return. The Fourth Highway (Maintenance) Project: Credit 507-TA of 1974 was the first of two phases intended to establish a trunk road maintenance organization but failed to achieve its physical and institutional objectives. The Trucking Industry Rehabilitation and Improvement Project: Credit 743-TA of 1977 was an uncommonly successful operation. The project improved public trucking operations and helped liberalize procurement of spare parts for the industry as a whole. The Fifth Highway Project: Credit 876-TA of 1979 was the second phase to establish a road maintenance organization but its scope, like that of the previous phase, proved to be too large. The need to reduce the project area became obvious during implementation but Government was reluctant to scale it down. Implementation capacity was affected by the Executing Agency's unwillingness to employ expatriate staff in line positions and these were filled by inexperienced and sometimes unqualified staff. The Port Rehabilitation Project: Credit 1636-TA of 1984 (ongoing in mid-1988) is to rehabilitate and modernize the port of Dar-es-Salaam. The Sixth Highway (Rehabilitation) Project of 1986 (ongoing in mid- 1988) is to rehabilitate the most important sections of the highway network and to improve road maintenance by encouraging work by con- tract rather than by force account. The Impact of Bank Lending 4.03 With one exception (the 1977 Trucking Project), project results did not match expectations. In virtually all cases, inception reflected sound objectives,35 but project evaluation, selection and preparation were defective and implementation was problematic. Appraisal was often dominated by arithmetical exercises aimed at high rates of return, which is perhaps the reason why freight traffic forecasts, particularly for the railways, were usually optimistic. In appraisals of rural roads, the pro- ducer surplus technique resorted to convenient, but not pragmatic, assump- tions regarding increased production attributable to the availability of a new or better road. During implementation, new construction works were delayed; maintenance targets fell short; institutional reforms did not take place; transport planning was not strengthened. 4.04 The relative success of the 1977 Trucking Project must be attrib- uted largely to the high degree of Governments interest in the project, (having requested Bank advice in 1974) and the low potential for conflict over ideological matters inherent in this type of project. Here the Bank was supporting public sector trucking through a progrem of financial re- structuring of the RETCOs, rehabilitation of their fleet, provision of workshop equipment and replenishment of their spare parts inventory, and technical assistance to strengthen management and training; and for devel- oping a project implementation, coordination, monitoring and policy advice capability. Among these components those involving purchase of equipment and physical fleet rehabilitation went well and with minimum delay, while those involving institution building and technical assistance went less quickly. This was no doubt due to the general sentiment against use of expatriate skills and to Bank underestimation of the difficulties in bringing about institutional change and of delays in attracting diverse investors to participate in government-controlled commercial companies, a clear case of overoptimism and misjudgment of human response in a con- strained environment. 4.05 This review revealed the importance of four topics. Starting with the 1964 Highways I and continuing with all projects thereafter, staffing weaknesses must be singled out as a major contributing factor to unsatis- factory results. Political interference played a major role, particularly 35/ Specifically, the Third Railway Project was aimed at the introduction of a cost-based tariff; management strengthening and staff training components were provided for in virtually all projects; maintenance was a common theme in the last three highway projects as well as in the Port Rehabilitation Project; and all highway projects devoted much attention to axle-load limit enforcement. in the 1971 Railways III and in the 1979 Highways V. Complex and ambitious project scope hindered implementation of Highways III, IV and V, as well as of Railways III. The transit function served by Tanzanian facilities was a prominent feature of all railway projects, of Highways II, of Harbors III, and of the 1984 Port Rehabilitation Project. V. LESSONS 5.01 Given the evident basic disagreement with the Tanzanian authori- ties on (a) the role of the State, (b) the priority to be assigned to main- tenance, and (c) the conditions of employment needed to make local staff fully productive, could the Bank have achieved more? Could it have ex- panded the volume of its lending? Or reduced it? Or channeled it differ- ently? Also, if these ftndamental disagreements continue, will the Bank be justified in extending more resources to the transport sector? 5.02 This review has shown that Bank impact might have been more pro- nounced if closer attention had been paid to the historical, economic and political context. Also, that the Bank could have presented its case bet- ter by deferring to the complexities of issues and to worldwide cyclical changes. These are feasible adjustments for Bank staff presently working on Tanzania. Assuming that, on its side, the Party is willing to allow ideological adjustments, the Bank will have a large role to play in Tanza- nian transport, and the experience of the past twenty-five years should help define both the policy dialogue agenda and the project work that re- mains to be done. Bank-Tanzania relations, time for preparation and imple- mentation, scope of new initiatives, and personnel matters are the four areas where redefinitions and reconsideration seem to be warranted. 5.03 Bank-Tanzania relations need to be expanded to include, on a reg- ular basis, the Party leadership. This will be critical for the success of any policy initiative, and indispensable for the continuing fulfillment of the coordinating role that the Bank has recently assumed on behalf of neighboring land-locked countries (e.g., the 1988 Northern Corridor Transport Project for Malawi). 5.04 Larger allowances have to be made for time to prepare and imple- ment both policy initiatives and projects. Experience has shown that hasty preparation and optimistic time-tables invariably lead to disappointing results. 5.05 The scope of new initiatives, both in project- and in policy-re- lated work, would best be kept simple: for roads, railways and ports, the greatest benefits would be in improving maintenance operations; for plan- ning, major studies ought to be postponed until a minimum statistical base has been established. 5.06 Because of low salaries, personnel issues will continue to be problematic and, consequently, deserve priority attention aimed at provid- ing effective incentives for better performance. 5.07 Some observations on the 1977 Trucking Project ought to serve as the conclusion of this review. The project created five public transport companies and helped liberalize spare parts procurement for public as well as private truckers. The Executing Agency collects and publishes the only reliable cost data in Tanzania, and, surprisingly enough, the activity of the public transport companies established under the project actually en- courages competition, because, otherwise, transport users would be at the mercy of quasi-monopolistic private truckers. The project's succesfful outcome is due in large measure to the professional and personal qualities of the head of the Executing Agency and of the technical assistance staff, which proves that high-caliber management talent is available locally, and that properly selected technical assistance can achieve remarkable results under very difficult circumstances. L IX APPENDIX A Chronology of Events 1898 - Construction of railway Inland from Mombasa begins. 1899 - Tanganyika Northern Line construction begins. 1901 - Railway from Mombasa reaches Lake Victoria at Klsumu. 1984 - Tanganyika Central Line construction begins. 1917 - Kenya-Uganda Custems Union established. 1919 - East African Currency Board established. 1927 - Tanganyika joins Kenya-Ugands Customs Union. 1928 - Spur constructed from Tanganyika Central Line to Lake Victoria at Mwanza. 1931 - Railway from Mombasa-Kisumu extended to Kampala. 1938 - Common Posts and Telegraph Administration established for East Africa. 1946 - East African Airways established. 1648 - Tanganyika Railways and Ports Administration amalgamated with that of Kenya-Uganda. 1949 - Tanganyika Customs Administration amalgamated with that of Kenya-Uganda. 1961 - Tangaryika independence. 1982 - Uganda independence. 1988 - Kenya Independence. Link line between Central and Northern Lines completed in Tanganyika. 1964 - Tanzania adopted as name of the United Republic of Tanganyika and Zanzibar. 1965 - Rhodesia Unilateral Declaration of Independence. 1988 - Kenya and Tanzania Central Banks established. 1967 - Arusha Declaration. Nationalization of banks and other foreign-owned enterprises in Tanzania. Treaty for East Africa%. Cooperation. 1988 - Mombasa-Nairobi road paved. Dar Es Salaam-Zambia pipeline opened for refined petroleum products. 1971 - Construction begins of Uhuru Railway from Der to Zambia. Nairobi-Tanzania border road paved. 1978 - Ndols oil refinery in Zambia comes on stream and pipeline from Dar es Salaam used only for crude oil. 1974 - Tanzania imposes ban on heavy road vehicles operating from Kenya. 1976 - Uhuru Railway operational. 1977 - Tanzania closes border with Kenya. Collapse of East African Airways. End of East African Community. Separate Kenya and Tanzania Railways Corporations established. 1979 - Tanzania at war with Uganda. 1980 - Tanzania Airways International services suspended. APPENDIX B TANZANIA- Bank Reports May 1958: The Economy - Tanganvika. December 1962: The Economy - Tanganyika. November 1963: Current Economic Position and Prospects. March 1970: Economic Development and Prospects. May 1972: Economic Development and Prospects. December 1972t Recent Economic Developments. April 1977: Economic Memorandum. December 1977: Basic Economic Report. January 1988: Parastatals in Tanzania --Towards a Reform Program. The Northern Rhodesia- Tanganyika Rail Link, May 27, 1964, Report No. TO-415a Prospects for Economic Development in East Africa, August 31, 1967, Report No. AF-58a, Vol. III, Tanzania, Part 5, Annex D, Transport. Erol Haker, Tanzania --Project Identification Mission, August 15, 1967. Transport Sector Memorandum, January 28, 1977. Transport Sector Memorandum, Vols. I, II, III, January 4, 1985 (also December 27, 1982), Report No. 4042-TA. The Economic Development of Tanganyika, Report of an IBRD Mission, The Johns Hopkins Press for IBRD, 1961, Chapter 13, Transport, Communications and Tourist Trade. Economic Developments in East Africa, July 30, 1971, Report No. AE-16a, Vol. III, Tanzania, Part I - A Program of Preinvestment Studies, Chapter 5, Transportation. Agricultural and Rural Development Sector Study, Vols. I, II, III, December 10, 1974, Report No. 541a-TA, Vol. II, Annex 8, Roads and Road Transport. Economic Memorandum on Tanzania, January 23, 1981, Report No. 3086-TA, pp. 19-24. Agricultural Sector Report, August 19, 1983, Report No. 4052-TA, Chapter V, Agricultural Support Services, F. Services of the Transport Sector. Institutional Development in Africa: A Review of World Bank Project Experience, May 17, 1984, Report No. 5085, Volume II, Chapter III, Selected Case Studies in Tanzania, C. Highway Projects. Country Economic Memorandum, August 10, 1984, Report No. 5019-TA, Chapter 2, Declining Sectoral Productivity, D. Transport and Communications, Chapter 5, Sectoral Policies, C. Transport. EAST AFRICAN COMMUNITY- Bank Reports April 1953: The Economies - British East Africa. September 1965: Current Economic Position and Prospects. August 1967t Prospects for Economic Development. March 1968: Economic Memorandum. June 1971: Economic Developments. Prospects for Economic Development in East Africa, August 31, 1967, Report No. AF-58b, Part 1, Common Markqt, Common Services and Common Problems (pp. 4-7, 25), Report No. AF-58a, Vol. I, East Africa, Part 2, Annex A, Common Transport Services. Economic Developments in East Africa, July 30, 1971, Report No. AE-16a, Vol. V, A Programme of Regional Preinvestment Studies, Chapter 4, Transportation. The Work of the Bank on Transport in Kenya and Tanzania, 1960-1980 (Draft Special Study for OED, October 1986). Other Reports and Documents Antony G. Hopkins, The World Bank in Africas Historical Reflections on the African Present (World Development, Vol. 14, No. 12, pp. 1473-1487, 1986). Louis Berger International for USAID, Tanzania Transport Sector Study (March 1987). The United Republic of Tanzania, Programme for Transport Sector Recovery (December 1987). - 173 - APPPENDIX C TANZANIA NOTES ON TRANSPORT PROJECTS FINANCED BY THE BANK GROUP 1964-1988 - 174 - PROJECT NAME Revised First Highway Project (Cr. 48-TA/ Cr. 115-TA) THE OED REPORT HAS NO BASIC DATA SHEET OED REPORT NO. 791 of June 26, 1975 PROJECT BACKGROUND AND OBJECTIVES The US$14 million credit (48-TA) of February 1964 was to finanze the foreign exchange cost of a First Highway Project which included detailed engineering and construction of eight road sections. Shortly thereafter, most expatriate staff in the Roads and Aerodromes Division (PAD) of the Ministry of Communications and Works left the country. RAD's professional manpower was depleted and project implementation seriously delayed. In addition, the work quantities of the project, based on preliminary engineering, had been underestimated, prices had escalated, and, by 1961, project costs had increased considerably. Because of delays in implementation and increased cost, the Ministry of Finance asked IDA to defer the credit's scheduled closing date and to supplement tie credit to meet the increased cost. IDA reappraised and revised the project, and provided a US$3 million supplementary credit (115-TA) in March 1968. The revised project was to: retain detailed engineering and construction of 533 m. of six of the eight road sections; extend detailed engineering of the section of the Tanzam Highway between Morogoro and Iringa from 104 mi to 142 m. but omit construction; include detailed engineering of feeder roads (110 m. of four secondary roads and 98 m. of tertiary roads) on the Geita Peninsula as well as a five-year program of staffing and training for RAD; and omit construction of the Musoma-North Mara road section. PROJECT IMPLEMENTATION Road construction was completed satisfactorily by 1970, and the detailed engineering work was carried out as envisaged. Engineering of the Morogoro-Iringa section of the Tanzam Highway provided the detailed design for its construction, subsequently implemented under the Second Highway Project. Engineering of the Geita Roads, after further revision, became the basis for their upgrading under the Third Highway Project. The staffing and training program was carried out to a limited extent because of slowness in recruiting expatriates and shortage of African staff to be trained. The closing date for the credit was postponed several times, and the final disbursement was made on July 15, 1973, two years behind schedule. About US$16.6 million was disbursed from the two credits - 175 - together, and about US$440,000 of Credit 115-TA was cancelled at closing. The actual cost of the revised project was US$22.2 million, or 3% less than estimated. The available information suggests that the overall economic return approximates the 11Z estimated in the 1968 appraisal. Three of the roads had lower rates of return at audit that at appraisal and one a substantiall higher return. Because of truck overloading, the effect of additional vehicle operating cost savings would be offset by the consequent damage to the roads. Traffic overestimation was the main factor in the shortfall from projected economic returns. The primary reasons were: (a) IDA's appraisal assumption that decreased vehicle operating costs would be enough to induce substantial generated traffic, and (b) absence of an assessment of complementary investment in the regions wrrounding the roads. Unstable political conditions, as well as low growth 3f the national economy and the vehicle fleet, contributed to traffic shortfalls. The staffing and training program was not formulated carefully. Success was too dependent on one man, the then Permanent Secretary of the Ministry, who was transferred shortly after Credit 115-TA was signed. To get the roads built it was essential to hand over more responsibility for detailed engineering and construction supervision to consultants, and this was eventually done in accordance with IDA recommendations. The sudden departure of Tanzanian engineers of Asian origin in 1971-72 contributed to the aggravation of the staffing problems. Three of the project roads, accounting for about 402 of total cost, show an ERR of less than 10Z. This, combined with the limited manner in which the project has contributed to the development of African capabilities, suggests that the original project, about which reservations had been expressed at the inception phase that it was too big, but which was approved as a special form of support to a newly independent country, should have been cut back in the 1967 revision even more than it was. IDA's objective in the staffing and training program was to ease RAD's transition to Africanization. This objective was not achieved. A manpower analysis at the time of appraisal/ reappraisal might have brought out the absolute shortage of Tanzanian staff, and the problems which would arise out of the sudden departure of the Asians in 1971-72. Furthermore, after the departure of the Permanent Secretary neither expatriates nor Africans were recruited in quite the way planned. A more gradual approach (perhaps a reduction in the number of roads), might have allowed local personnel to develop and use their capabilities and might have reduced the need for relatively costly outside consultants. Scope for greater use of stage construction would have been relatively limited in this case since most of the improvements were from earth to gravel or gravel to paved. Reduction of the number of roads in the project could have been a sound approach because economic returns on several of - 176 - them are now estimated to be quite low (three roads, accounting for 42% of total construction investment under the project, with returns fairly clearly below 10Z). For this approach to have been successfully used, more reliable data would have been necessary. For example, if this approach had been adopted, the Dar-es-Salaam-Kibiti road would have been the first to be deleted because it had the lowest economiz rate of return at appraisal and was the one about which the 1968 appraisal report showed the most doubt. But this road, benefitting from the proximity of the national capital, now shows the highest return of the six project roads. Hence, while it is clear that the original 1964 project was too large, and that the 1968 revision suffered the same faults in lesser degree, it is difficult to see how the latter could have been better reduced at reappraisal in 1967 given the poor quality of data available at the time. LU - 177 - PROJECT NAME Second Highway Project (Ln.586-TA/Cr.142-TA) BOARD DATE February 1969 EFFECTIVENESS April 1969 CLOSING December 1972 (Cr.); December 1978 (Ln.) ESTIMATED PROJECT COST US$38.0 million ACTUAL PROJECT COST US$35.20 million LOAN/CREDIT AMOUNT US$22.5 million equivalent PLANNED COMPLETION DATE September 1971 ACTUAL COMPLETION DATE May 1972 ERR AT APPRAISAL 20% ERR AFTER COMPLETION 161 OED REPORT NO. 4030 of June 30, 1982 PROJECT OBJECTIVES AND IMPLEMENTATION The project assisted the construction of two difficult sections, aggregating 499 km, of the Tanzania to Zambia highway. The goal of providing Zambia with a reliable outlet to the sea was effectively achieved. However, appraisal expectations that the project would be justified by benefits accruing to Tanzania alone were not fully met: the rate of return to Tanzania was only 7%. When benefits accruing to Zambia are included the rate of return of the project increases to an acceptable 16%. This is because, contrary to appraisal expectations, Zambia traffic did not abandon the highway when the TAZARA railway was opened in 1975. An important issue on distribution of benefits between benefitting countries is raised by the project. At appraisal it was expected that most benefits (85%) would accrue to Tanzania; on completion it has become evident that an important share of the benefits (45%) has accrued to Zambia. The PCR suggests that the issue should be reviewed and that Tanzania might consider recouping part of the benefits from Zambia. The audit supports the suggestion, although it is clear that the reason for the unexpected distribution of benefits is the failure of Tanzanian traffic to materialize as expected. Project implementation highlights the handling of uncertainty about traffic volumes. Against the wishes of the Government and the advice of consultants, the Bank initially recommended using a thin asphalt concrete surfacing, risking that faster traffic growth would require earlier than expected strengthening of the road and also risking that the road might - 178 - fail. In the opinion of the Bank, uncertainties about future traffic volumes made it worth running those risks and postponing the additional cost of a stronger surfacing. As a result of protracted discussions between the Government and the Bank, the road was built with a considerably thicker surfacing than initially suggested by the Bank and only slightly thinner than proposed by the consultant. The pavement as tuilt lasted just about as long as the Bank had predicted the pavement with the thinner surfacing would last. However, this was one of the first projects in which risk analysis was applied in the Bank and the PCR (paras. 5.30 to 5.34) contains an interesting discussion of its application, limitations and means to improve the use of the technique. The risk analysis concentrated mostly on risks associated with the economic aspects and not those related to the engineering and administrative components. No account was taken of the quality of construction, or of the quality of the soil tests used, both of which have a significant effect on pavement durability. Similarly, the very thin asphaltic concrete surfacing suggested by the Bank was prone to fail early because of poor adhesion to the base, another aspect to which the risk analysis made no reference. Further, no reference was made to the amount, quality and timeliness of maintenance operations or of the effectiveness of vehicle weight controls. A recent study by the University of Dar Es Salaam and a visual inspection of the road by OED staff confirm that the road needs reinforcing and strengthening soon. Delaying this work will lead to a considerably more deteriorated highway on account of its use by overloaded trucks. Covenants to the effect that axle weight limitations would be strictly enforced were not complied with, and the aging of the road is mostly due to the overloading of the few trucks that use it. CONCLUSIONS Project evaluation by OED was delayed long after the physical works were completed and long after evaluations are typically undertaken for Bank Group projects. The delay allowed carrying out the evaluation with nearly a decade of actual experience in the use of the road. Yet, for lack of simple traffic counts it was not possible to measure the impact of the project, and instead the impact had to be estimated on the basis of projections. It is clear, however, that the project had a fundamentally different impact from what was foreseen at appraisal. Tanzanian traffic was less than forecast, reflecting the project's reduced development impact as well as general Tanzanian conditions; Zambian traffic continuel at high levels for longer than expected, reflecting a misjudged diversion of traffic to the TAZARA railway and thus a misplaced traffic allocation at appraisal, a detrimental development in an economy short of resources and foreign exchange; and, axle loads were greater than projected, prematurely aging the road and reflecting unrealistic design parameters, as well as poor law-enforcement and lack of adherence to Government regulations mostly by the para-statal trucking organizations. The fact that the economic rate of return of the project is comparable to appraisal expectations (when Zambian traffic is included) reflects the concurrence of factors which significantly influenced by chance. m1 - 179 - PROJECT NAME East Africa Community Second and Third Harbour Project (Loans 638 and 865-EA) BOARD DATE July 1969 (II); January 1972 (III) EFFECTIVENESS December 1969 (II); April 1973 (III) CLOSING Neither fully disbursed in June 1982 ESTIMATED PROJECT COST US$105.74 million ACTUAL PROJECT COST US$136.73 million LOANICREDIT AMOUNT Ue$1.5 million equivalent PLANNED COMPLETION DATE December 1972 (II); December 1976 (III) ACTUAL COMPLETION DATE March 1979 ERR AT APPRAISAL Not given in the SAR ERR AFTER COMPLETION Not calculated. OED REPORT NO. 4029 of June 30, 1982 PROJECT OBJECTIVES The two projects were to help implement the East African Harbors Corporation (EAHC) 1969-72 and 1972-74 development programs. In addition to various smalle: modernization and improvement works, the programs included, at Mombasa. the construction of a specialized wharf to handle bulk cement, two deep-water berths and a cold storage building and, at Dar Es Salaam, five deep-water berths, a single buoy tanker mooring, and the completion of three berths started under a previous Bank-financed project. In addition, at Tanga the programs included improvement of lighterage facilities. PROJECT IMPLEMENTATION Completion of the civil works significantly increased the capacity of these ports and assisted the East African Community, until its demise in 1977, in expediting its foreign trade. Without the projects the economies of Kenya and Tanzania, but also of Uganda, Zambia, Rwanda and Burundi, would have faced serious constraints, and congestion at both Mombasa and Dar Es Salaam ports would have reached levels well beyond the proportions which prevailed in the early 1970s. Long delays occurred with the completion of the back-of-port area for berths 7 and 8 in Dar Es Salaam. This was caused by problems with the Government-controlled contracting company. On berths 9, 10 and 11 in Dar - 180 - Es Salaam, delays were caused by extra works. The two projects had joint cost overruns of about 302 counting only the items which were completed by December 31, 1981. Fortunately, nearly all construction materials had been purchased and delivered before the sharp worldwide inflation that followed oil prices increases in 1973. Many project items were either deleted, not implemented or work was still in progress on December 31, 1981. In Dar es Salaam there were delays because of additions to the works. Delays were also caused by the low priority given by the government controlled contractor. At Tanga there were contractor inadequacies but they were due to the Kenyan contractor had in operating once the Kenya-Tanzania border was closed in 1977. Administrative inadequacies were also responsible for delays in the commencement of work on cold storage facilities and on the Tanga lighter wharf. Economic returns are likely to be significantly below the optimistic assessment contained in the PCR because of the higher costs determined by the audit (compared to those shown in the PCR) and doubts concerning the methodology applied in the PCR. While the PCR equates economic impact to first-year-returns, the audit maintains that this view is too narrow in that it fails to take into account developments after the first year of operation of the new facilities. The audit also questions the hypothetical alternative used for economic analysis purposes, and the PCR's approach of considering each item of investment as a marginal investment. It considers that the projects should be comparei to the least cost feasible alternative, and that for evaluating t .e project as a whole the only reasonable approach is to consider all pioject investments jointly. Given the poor impact the projects had on productivity and on port operations and maintenance (i.e., the longer term effect of the investments), and taking into account traffic developments subsequent to the collapse of the East African Community in 1977, the audit questions the optimistic economic assessment presented in the PCR. The principal lessons are: (i) productivity improvement is an important means of achieving increased port capacity but is unlikely to be attained if the project places primary emphasis on the addition of physical facilities ard does not include a specific productivity program; and (ii) the use of first year returns is not a substitute for other, more general tools of project evaluation. In addition, the following points ought to be kept in mind: - neither project included manpower development (training or technical assistance) nor productivity betterment components nor made their improvement an integral part of the projects; this was one of the reasons why productivity did not increase and the long- term economic impact of the projects was meager; - the lack of direct project involvement with the improvement of administrative and operational aspects of cargo handling ended up restricting project impact to just the addition of physical capacity; - the economic analysis methods used in the PCR are questionable and do not reflect the true economic performance of the two projects; - the experience of the countries of East Africa with regional economic integration and with common services, and of the Bank's contribution to this effort deserves detailed review and analysis. CONCLUSIONS Despite cost overruns, implementation delays and changes in project scope, the two projects are considered successful in that the additions to physical capacity provided essential facilities to the foreign trade of Kenya and Tanzania and neighboring landlocked countries. Although the projects dealt effectively with the capacity problems of the ports, they may not have been the cheapest and mout economical solutions. The audit suggests that future projects ought to place special emphasis on productivity, manpower development, and on the maintenance of facilities - and equipment. Because of the recent decreases in traffic brought about by the relatively depressed performance of the ports' hinterlands, there is now, comppred to earlier years, adequate physical capacity at both ports. It would seem to be the appropriate time to assist the port authorities in improving productivity and catching up with badly needed deferred maintenance and operational improvements. PROJECT NAME East Africa Community: Third Railway Project (Ln.674-EA) BOARD DATE March 1970 EFFECTIVENESS October 1970 CLOSING December 1979 FSTIMATED PROJECT COST US$90.84 million ACTUAL PROJECT COST Unknown LOAN/CREDIT AMOUNT US$42.4 million equivalent PLANNED COMPLETION DATE December 1972 ACTUAL COMPLETION DATE Project was not completed ERR AT APPRAISAL None calculated at appraisal ERR AFTER COMPLETION None calculated OED REPORT NO. 4533 of June 3, 1983 PROJECT OBJECTIVES The project was to assist East African Railways (EAR) in implementing a policy package designed to streamline the Railway's tariff structure and bring down its costs to allow it to recoup profitable business it had lost to ro;ad transport and effectively compete with trucking in the future, while &t the same time remaining a commercially viable operation. The project involved actions on two fronts: realignment of tariffs, and implementatico of an investment and cost reduction progrim. PROJECT IMPLEMENTATION On completion the project had failed to achieve full implementation on either objective. Full implementation of a cost-based tariff structure was not achieved, because the economic, political and operational consequences of implementing the new tariff were not foreseen. Full implementation of the investment and the cost reduction programs failed to materialize because of the Borrower's financial troubles, caused by political differences among the Member Governments in the East African Community. While the audit agrees with the PCR that the project resulted in some important railway investments in Kenya, Tanzania and Uganda, it does not shar3 the confident assessment the PCR has made of EAR's performance during the years before 1974. The audit questions whether by restricting the analysis to the years before 1974 the PCR has not failed to consider the full impact and consequences of the project. The bulk of the investments under the project occurred after 1973, and almost all events that marred implementation of the project took place after that time. It happens, however, that 1973 is the last year for which full financial returns for EAR are available. For this reason, no quantified assessment can be made of the project; the audit is unable to confirm the optimistic judgments made in the PCR about the high economic returns of some project investments; and the audit is reluctant to agree with the PCR that the project's unspecified effect on maintenance demonstrates it high economic return. It is true that freight traffic increased almost in accordance with appraisal forecasts, that passenger-trains utilization rose, but operating costs rose sharply during project implementation, the operating ratio increased instead of falling, and the current ratio deteriorated. It is conceivable that overall railway costs rose so much that shortly before its de facto dissolution, the Railway was carrying all ttaffic at higher costs than road transport. However, the lack of proper accounting prevents determining the full extent of EAR's losses. The audit believes that the project's shortcomings can be attributed to poor preparation and design and to inadequate management, both due (in part) to the Bank's view of EAR's network as a single system --when it was not one but two, and perhaps three seoarate railways, and (in part) to the interplay between the project and the operation of the East African Community. Among the reasons for the project's limited success, the audit points at the following: - the project's multinational nature (for which no allowance was made in project design, and which resulted in serious political interference as well as difficulties during implementation) - the poor definition of many project components and an imbalance in the project's Investment Program (which resulted in serious implementation delays and lack of political btipport) - the poor definition of project consequences, particularly those resulting from changes in the railway's pricing policy (which resulted in lack of political support for the changes and eventually contributed to the financial collapse of the Railway) - the lack of quantification of expected project impact, i.e. research on whether the project could do what it claimed it could achieve (which not only contravened well established Bank practice but resulted in the implementation of a project with uncettain goals) - the limited usefulness of joint and several guarantees by sovereign governments in multinational projects particularly when not accompanied by a debt sharing agreement worked out beforehand among the governments. Failure to recognize this fact resulted in service payment defaults on Bank loans which, in turn, reflected adversely on the Bank, the Borrower and the Guarantors. - the difficulty in implementing a project with a Borrowing Agency which is owned and controlled by more than one country. The East African Community, presented as a model of economic cooperation, consisted of three Partner States with differing political and economic philosophies. For EAR, the result was that vital changes, which required unanimous approval, could not be made. Adding to the diffculties was that the three regions of EAR di not keep separate accounts and financial data was available only for the corporation in total. - prozurement problems arose from two sources: from absence of an agreed, and detailed, list of project items at appraisal, and from the unsatisfactory procurement procedures within the railway. CONCLUSIONS From the beginning, the project was beset with political difficulties. The administrative and management structure for the East African common services, including the railways, -,hich had been established under the 1967 Treaty for East African Cooperation, was already under strain from the efforts of the individual Partner States to support what they conceived to be their national interests. The formula-:ion of the investment program was delayed by disputes over the distribution of investments between the three countries. Although there was no justification whatever in the provisions of the Treaty, there was pressure for the same amount of investment to be andertaken in each, despite the fact that the volume of traffic was widely unequal. Because of disagreements among the Partner States, the Bank loan took an entire year to become effective. The investment program was subject to further revisions for pclitical reasons and to implementation difficulties because of its imprecision. The program had been prepared in haste and it left differences between the Partner States unresolved. At that time the operation of the Community machinery to resolve disputes had been brought to a standstill by the coup in Uganda. Later amendments to the investment program, in 1972, making it more specific and detailed, enabled implementation to proceed, but the cost of the program had substantially increased by then, despite the elimination of some items. Meanwhile, the financial position of the railways worsened. There was further loss of the most profitable traffic, particularly of petroleum products, because of the railways' inability to handle them. Management had reached a new level of inefficiency and wagons to carry petroleum products were not available. The Bank's decision to support and to assist in the finance of the Mombasa-Nairobi pipeline made it clear that this profitable traffic would not be recovered. In addition, the regional offices of the railways began to hold back the revenues they should have transferred to headquarters and eventually stopped such transfers altogether. Various "rescue operations" were attempted but the railways effectively split into three national systems, and this was followed in 1977 by the final collapse of the Community itself, when three national railways were formally established. Disbursement of the Bank's loan had been suspended on several occasions, and it was finally disbursed to meet the cost of mediation on the distribution of the Community's assets between the former Partner States. Poor project preparation and design is responsible for "the poor definition of items in the Investment Program" (PPAM, para. 2.01), as well as for the failure to take into account the fact that a switch of petroleum traffic back to the railways in .esponse to a lowering of the tariff could not be handled in the disorganized and depleted state of the railway system (PPAM, para. 3.05). It was also responsible for the over design of the Changamwe marshalling yard and for the need to redesign the Investment Program, making it more precise and specific, at a time when costs had substantially increased. However, part of the increased cost of the redesigned program was accounted for by additions to the project, and the superficiality of the program design resulted from pressure on time caused by politically created delays. The project designers could be held responsible for their failure to analyze the effect of project investments on operating costs, but even here improvements in efficiency were delayed and diminished by political interference, as in the procurement of diesel locomotives. The main deficiency in the project design, in the view of the PPAM, was its failure to understand the provisions of the Treaty and to take account of the territorial impact of the project and the political response. The Appraisal Report treated the railway system as a whole, working with systemwide averages, without reference to such considerations as the wide variations in traffic within the system and between the different Partner States. The PPAH points out that the Treaty specifically allowed for cross-subsidization between parts of the system (Treaty Article 72.3) and argues that cross-subsidization would be impossible with a cost-based tariff. It can now be guessed that there was no project design that could have satisfied each Partner State enough for political interference and obstruction to be avoided. The fact is that the Partner Satets were not content to operate within the letter and spirit of the Treaty. But that was not evident at the time, to the Bank or to any other observer. The supportive attitude of the Bank towards the continuation of the railways was justified by its importance as a component of the whole set of common arrangements under the Treaty. Though (always with hindsight) the Bank's optimistic assumption that the Treaty could be made to work is seen to have been unjustified, this was not an obvious conclusion at the time. To allow the railways to collapse without any attempt to save them would have been a serious blow to the Community's surv4val. It may therefore be concluded that cost increases played but a small part in the failure of the project, and the same applies to deficiencies in project design. The genuine villain was political interference which could not have been taken fully into account in the project design. If it had been, the project would never have been begun. PROJECT NAME Third Highway Project (Cr.265-TA/Ln.586-TA) BOARD DATE July 1971 EFFECTIVENESS October 1971 CLOSING December 1978 ESTIMATED PROJECT COST US$9.5 million ACTUAL PROJECT COST US$25.18 million LOAN/CREDIT AMOUNT US$6.5 million equivalent PLANNED COMPLETION DATE June 1974 ACTUAL COMPLETION DATE November 1978 (partial completion) ERR AT APPRAISAL 14Z ERR AFTER COMPLETION less than 42 OED REPORT NO. 4031 of June 30, 1982 PROJECT BACKGROUND AND OBJECTIVES On July 27, 1971, the Board approved a Third Highway Project, authorizing Credit 265-TA for US$6.5 million, which was later supplemented by a transfer of a US$1.9 million surplus from Loan 586-TA originally approved for the Second Highway Project. These funds were to help finance: the construction of 200 km of the Mtwara-Mingoyo-Masasi road in southern Tanzania; the improvement and maintenance of 475 km of feeder roads in Geita district (310 km) and Mara region (165 km); and preinvestment studies, consultant costs for supervision, technical assistance and training. Total project cost was estL-ated at appraisal at US$9.5 million of which IDA was to finance the foreign exchange cost of US$6.5 million. The original Credit Agreement was signed on August 6, 1971 with a Closing Date of December 31, 1974. The Credit became effective on October 12, 1971. The project's objectives were: to construct an economically viable section of trunk road to serve the southern regions of the country; to improve and mainta3n feeder rods in two economically important areas in the north; to develrp institutional strength and capacity for future betterment and maintenance; and to provide a basis for future road investments by carrying o-it feasibility and engineering studies. PROJECT IMPLEMENTATION The construction of the trunk road in the south constitutes a technical and physical achievement of the first objective. However, the economic objective was not achieved. Regarding the objective for feeder roads, although most of the roads were improved, due to the lack of subsequent maintenance, much of the value of the work has been lost and hence the economic objectives of this component were not achieved either. Further, the institutional development objective of the feeder roads component was not achieved because training was unsuccessful, and betterment and maintenance units were discontinued. Feasibility studies were carried out successfully and the resulting reports have since been used in making investment decisions by other aid agencies. Lessons arising from this project reconfirm lessons already learned from other Bank Group projects in a number of countries. - to ensure accurate specification of the amount and type of work to be done and reliable cost estimates, detailed engineering of civil works should be complete before appraisal; (this has been standard practice for Bank Group projects, and this project has reconfirmed the soundness of this practice, especially in countries like Tanzania with weak institutions and histories of difficult working conditions for consultants and contractors) - irrespective of a consultant firm's good reputation and past performanre, the crucial criterion for its future performance in a new project is the experience, ability and calibre of the specific staff assigned to that project. - occasionally the Bank Group will ba subjected to pressures from Borrowers to retreat from its policies, guidelines and procedures. The Bank may consider two options for dealing with such eventualities: (i) the Bank Group should stand firm and not give in to the pressures to reduce its standards; or (ii) where it yields to pressures, it should expect increased risks and reduced objectives and should accordingly be explicit in its presentation of expected results. - reappraisal of a project, undertaken because of changed circumstances. deserves as much attention and scrutiny as the original appraisal - recommendations and conclusions of a reappraisal deserve the same respect as those of the original appraisal even if they imply drastic revamping of the original project concept; - consultant contracts for supervision of civil works contracts should be specific about obligations and conditions for continuing beyond the original contract termination date if the civil works contract is not completed by that time; 188 - in Tanzania, start-up periods have been long; due allowance should be made in the design, analysis and scheduling of future projects - technical assistance in executive line positions can be extremely effective for carrying out a given work program provided the individual expert is experienced, competent, motivated and compatible with the local people. - where projects include the dual objectives of carrying out a program of work as well as institution building and training, separate technical assistance is needed for helping with ea-h objective. The project has shown that for carrying out a work program, the direct executive role is effective. However, for institution-building and training objectives, people responsible for carrying out a work program cannot effectively carry out a training and institution-building function as well; - where institution-building is a major objective, it is crucial that the institution to be developed is one which will retain continuing responsibility and involvement in the functions for which it is trained; (that is, if a feeder road unit is formed, it should be part of an organizalion with continuing responsibilities and programs in feeder roads) - Government's institutional weaknesses and deficiencies contribute to ineffectiveness of expatriate experts. Provision of expert technical assistance at the project level does not guarantee effective and efficient implementation of an overall institutional strengthening scheme. - upgrading of feeder roads can be undertaken by construction units but unless regional road maintenance ability and capacity are also improved, the upgraded roada will deteriorate rapidly; CONCLUSIONS The project was not a success, although two of its components (paving the road between Mtwara and Masasi and carrying out preinvestment studies), were realized. However, the first was accomplished with significant cost overruns (215%), %nd the second wi-h significant qualifications. In contrast, the improvement of feeder roads in the Geita peninsula and the Mara region were short lived, training of Tanzanian engineers met limited success, and the staffing program was not carried out. The goal of stimulating agricultural production was not reached as cashew nut production in the Mtwara-Masasi area has dropped (despite this and two parallel Bank Grcup projects designed to more directly assist it). Cotton production in the Geita Peninsula has alea dropped (despite a parallel IDA project to increase it), and milk and diary production in the Mara region has failed to materialize (despite a parallel Danish Government project aimed at assisting it). Investments in the Mtwara-Masasi road had a very low rate of return, (42) and those in feeder roads had negative rates of return, making the overall rate of return of the project less than 42. In summary, the audit highlights the following points: - the absence of clear objectives led to the implementation of a misconceived project - the absence of an implementation plan for the different components of the project negatively affected its execution and later prevented the materialization of a maintenance capability for feeder roads - an overly optimistic assessment of the staffing of the executing agency precluded addressing the problem properly during project preparation and ended up adversely affecting the execution of this as well as subsequent IDA projects; - the combination of executive, training and institution building functions in the the same group of consultants prevented effective execution of these functions - the non-observance of many IDA guidelines, procedures, principles and policies was deleterious to the efficient design and execution of the project and contributed to the low level of project achievement; - IDA's consistency in yielding to Government on most points of disagreement contributed to the perception that in the case of Tanzania the need to transfer resources had taken precedence over the achievement of reasonable economic returns; this adversely affected project control and supervision. PROJECT NAME Fourth Highway (Maintenance) Project (Cr. 507-TA) BOARD DATE September 1974 EFFECTIVENESS November 1974 CLOSING December 1982 ESTIMATED PROJECT COST US$12.45 million ACTUAL PROJECT COST US$10.52 million LOAN/CREDIT AMOUNT US$10.2 million equivalent PLANNED COMPLETION DATE June 1978 ACTUAL COMPLETION DATE December 1982 ERR AT APPRAISAL 22% ERR A7TER COMPLETION Negative OED REPORT NO. 6483 of November 7, 1986 PROJECT BACKGROUND AND OBJECTIVES ACCORDING TO THE PCR The project was the first phase of a two-phase effort to establish a trunk road maintenance organization. The first phase covered about 3,300 km of primary roads in the southern half of the country and the second phase which was financed under the Fifth Highway Project (Credit 876-TA) covered about 5,100 km of primary roads in the northern half of the country. The project was the outcome of a "Tanzania Highway Maintenance and Organization Study" which was carried out in 1972 b) consultants financed by the United States Agency 'for International Development. IDA assisted the Government in reviewing the reports at various stages of the study. The Government completed project preparation with the assistance of an advisor financed by the Canadian International Development Agency and a Government request for financing the project was made to IDA in September 1973. Appraisal took place in December. PROJECT IMPLEMENTATION Although the Credit Agreement became effective in November 1974, technical assistance financed under the project was not in place until late 1977. Initially, COMWORKS was not convinced of the need for technical assistance and considered its cost as too high, despite the fact that two-thirds of the engineering and technicians posts were vacant. Further delay in securing technical assistance services occurred from unsuccessful attempts to obtain the services through bilateral aid. COMWORKS also did not succeed in negotiations with a consulting firn to provide individual experts as opposed to a team of experts proposed by the firm. A second firm was then contacted but negotiations were not concluded until late 1976. By November 1979, the contract with consultants providing technical assistance and training services had expired and all technical assiotance and training staff had left the country. COMWORKS had not been keen to retain consultants, and poor housing and working conditions resulted in the majority of the staff remaining only for short periods in the country. Counterparts were appointed late and most of them were found to be not suitable. Hence, performance of technical assistance and training staff was not satisfactory. Procurement of equipment and spare parts was slow as a result of non- fulfillment by COMWORKS of Bank procurement procedures particularly over the requirement for prior review by IDA of bidding procedures and documentation for International Competitive Bidding. When procurement was completed in 1978, war broke out between Tanzania and Uganda and the Government diverted the project's heavy equipment to military operations. The project showed no progress and COMWORKS proposed a recasting of the project which was eventually approved by IDA in 1981. The final agreement called for the procurement of spare parts, few pieces of equipment and technical assistance. COMWORKS undertook to execute a highway maintenance program consisting of resealing 214 km, regravelling 186 km, pavement repairs for 640 km, and routine maintenance of a network which had expanded from 3,300 km at the beginning of project in 1974 to 4,900 km in 1979 with the incorporation of more roads into the network to be maintained by COMWORKS. The project was closed on December 31, 1982 with US$0.86 million of the original Credit of US$10.2 million cancelled. There was little improvement in routine maintenance and standards of resealing and regravelling. The project has a negative economic rate of return. A number of lessons can be extracted. - successful highway maintenance depends mainly on the quality of local management and availability of requisite number of trained field personnel. - the executing agency should be committed to employing necessary technical assistance staff, and should have the ability to administer and monitor the contracts for such services. - technical assistance should play an advisory role only if there are suitable counterparts. - there is need to develop the local contracting industry which could then participate in road maintenance and rehabilitation works as an alternative to force account. CONCLUSIONS The project was a logical development after a decade of involvement in the highway sector to upgrade the highway network. The departure of expatriate staff after independence in 1962 and the exodus of indigenous staff of Asian origin in 1972 brought into focus the decline in road maintenance capability that had been going on for some time. Both COMWORKS and IDA realized the need to stem the deterioration of the highway network but while COMWORKS regarded the need for new construction and paving of the highway network as of equal priority, IDA considered road maintenance as the higher priority and initiated the idea of a highway maintenance project to establish institutional arrangements for trunk road maintenance, to train COMWORKS staff and field personnel, and to implement a road rehabilitation and maintenance program. Although the project covered only about 3,300 km, it was ambitious. The project area encompassed almost the whole of the southern half of the country where communications have always been difficult and suitable housing for expatriates scarce. For a pilot project, especially one with obvious problems, the project area was too large. The project should have been tried out in one or two of the ten regions included in the project located closer to Dar-es-Salaam. The project aimed to implement a program of road maintenance and rehabilitation and to build an institution to plan, organize, execute, and monitor road maintenance activities. However, it did not define the road maintenance and rehabilitation program; it did not provide for training of engineers and technicians other than artisans; and it did not consider the alternative of carrying out road maintenance activities by contract. COMWORKS had an inadequate implementation capacity and found it difficult to fill vacant positions, particularly engineers. Senior staff were engaged in the day-to-day running of the ministry with not much time left to administer and attend to the requirements of the technical assistance staff financed by the project. For technical assistance to be effective in Tanzania, several conditions should be met: - the Government, particularly the executing agency, should be committed to employing technical assistance staff. - the role of technical assistance should be clearly defined (e.g. if it is to be advisory or executive), and an advisory role should be agreed upon if suitably qualified local counterparts are not available. - the Government should have a satisfactory capability to administer the contracts for technical assistance services. - before technical assist&;ce staff and their families arrive in the country, the Government should be satisfied that suitable housing, office facilities, and local transportation have been earm,-rked. - 7.) - - each technical assistance staff member should produce at the beginning of each half yearly period a detailed work program, satisfactory to the Government, in terms of actual tasks to be performed to achieve technical assistance objectives. - the Government should monitor performance based on the individual technical staff member's half yearly work program to see if it is justified to continue the technical assistance. Project experience suggests that COMWORKS should limit its force account operations and that most of the road maintenance woLk should be executed by contract. - 194 - PROJECT NAME Trucking Industry Rehabilitation and Improvement Project (Cr. 743-TA) BOARD DATE October 1977 EFFECTIVENESS April 1978 CLOSING June 1985 ESTIMATED PROJECT COST US$18.2 million ACTUAL PROJECT COST US$21.4 million LOAN/CREDIT AMOUNT US$15.0 million PLANNED COMPLETION DATE December 1982 ACTUAL COMPLETION DATE December 1984 ERR AT APPRAISAL 372 ERR AFTER COMPLETION 10% OED REPORT NO. Under preparation PROJECT BACKGROUND AND OBJECTIVES The Trucking Industry Rehabilitation and Improvement Project was prepared in March 1976 and appraised in September 1976. The IDA Credit of US$15 million was to finance the foreign exchange costs of trucks, technical assistance and workshop facilities for selected public companies, provide foreign exchange to Government in support of the liberalization of spare parts imports and provide assistance to a transport industry training institute. The project was to support Government's efforts to reverse the deteriorating trend in truck transport, a mode on which the predominantly agricultural economy depended heavily. This was to be achieved mainly through assistance to the public sector segment of the industryt (i) to strengthen the operating capability of selected companies in regions where agricultural activity was severely constrained by inadequate truck transport services through procurement and rehabilitation of trucks and through strengthening of management and management information systems; and (ii) to put these trucking companies on a sound financial footing through both operating improvements and covenanted requirements for adequate rates to ensure that they can be self-financing. Project components included: (i) provision of trucks and spare parts for the rehabilitation of trucks for five selected RETCOs; (ii) technical assistance for the management operations and training of staff of the selected RETCOs; (iii) provision of workshop equipment for the selected RETCOs; (iv) strengthening of the recently established National Institute of Transport through the provision of instructors, teaching aids and equipment; (v) replenishment of the country's inventory af spare parts; and (vi) provision of technical assis*ance to the project implementation agency, NTC, for project coordination, monitoring and policy advice. PROJECT IMPLEMENTATION The project experienced considerable delays because of delays in capitalizing the RETCOs. The project was the first one undertaken by the Region's Transport Division that attempted to attract a diverse set of shareholders as investors in commercially oriented companies. It was also the only commercially oriented project appraised by the then Highways Division. As a result the legal and financial ramifications of the formation of companies were not fully appreciated. Indeed, it was only during the latter stages of project implementation that financial analysts were involved in project supervision. Much closer involvement of Bank financial staff is now required in all projects, particularly commercially oriented ones. Technical assistance has been invaluable in developing systems. However, there has been little transfer of knowledge. This points to the fact that without both a strong specific skills training component in projects and an appropriate incentive system in the assisted organization, technical assistance aimed at strengthening indigenous management can prove ineffective. CONCLUSIONS The project represented the Bank Group's first direct assistance to the Tanzania trucking industry. As such it was a logical extension of its close involvement in the development of Tanzania's road network. Assistance to the industry was provided in the context of strong Government commitment to the development of the public sector. Given this commitment, the Association was compelled to direct the major part of its assistance to the public sector segment of the industry notwithstanding the existence of a dominant private sector. The project contributed to the establishment of potentially viable and self-sustaining public sector companies. But this potential will only be realized if the expatriate staff now responsible for much of the day-to-day management of operations is replaced by indigenous staff and the companies ability to price their services appropriately is increased. There has been little discernible progress made in attracting suitable financial and workshop management personnel for the RETCOs. This problem can ultimately only be satisfactorily resolved through greater national effort in training of staff, particularly engineers and accountants and through the provision of competitive remuneration for such staff. In respect of the RETC0s' pricing of their services, Government has, in fulfillment of one of the conditions for disbursement of a second tranche of the Multisector Rehabilitation Credit. announced a new policy of allowing the rates for the intra-regional transport of goods to be determined by market forces rather than by rate fixing boards. This should give the RETCOs greater pricing flexibility and ensure that they maintain their primary commitment to intra-regional trucking. This will, however, only happen if their shareholders who are also their primary users will permit the exercise of such flexibility. The RETCOs' operating performance has been satisfactory given the constraints under which they operate. However, this performance has been attained and is, at least for the near future, sustainable only with a significant input of technical assistance. Nonetheless, given the much poorer operating performance of other parastatal transport operations, the assistance to the RETCOs has been a satisfactory use of the Association'; resources. - 197 - **************** ************ *j********************************************* PROJECT NAME Fifth Highway Project (Cr. 876-TA) BOARD DATE January 1979 EFFECTIVENESS December 1980 CLOSING December 1985 ESTIMATED PROJECT COST US$25.71 million ACTUAL PROJECT COST US$22.54 million LOAN/CREDIT AMOUNT US$20.54 million equivalent PLANNED COMPLZTION DATE June 1984 ACTUAL COMPLETIOW DATE December 1985 ERR AT APPRAISAL 70Z ERR AFTER COMPLETION 102 OED REPORT NO. 6938 of September 15, 1987 PROJECT BACKGROUND AND OBJECTIVES ACCORDINC TO THE PCR This was the second phase of the effort to establish a trunk road maintenance organization in the country. The first phase was implemented under the Fourth Highway (Maintenance) Project and the Fifth Project included the establishment of routine maintenance capability for about 5,100 km of primary roads in the northerr. half of the country, the rehabilitation and regravelling of about 2,040 km of trunk roads, procurement of road maintenance and rehabilitation equipment, construction of road maintenance camps, workshops, and housing and technical assistance and training. PROJECT IMPLEMENTATION The Credit Agreement was signed on March 2, 1979, but was not declared effective until December 3, 1980. The delay occurred as a result of disagreement between IDA and Government over the selection of three key advisors to the Ministry of Communications and Works. The Government attempted to recruit them on an individual basis, but IDA urged selection of a consulting firm to provide two key advisors Twelve advisors, including one key advisor, were appointed on an individual basis. Two key advisors were provided by a consulting firm. They all arrived separately over a period of two years. Thus, from the beginning it was difficult to adopt a coordinated approach. At the advice of successive IDA supervision missions, attempts were made by the key advisor to the Senior Maintenance Engineer to coordinate and direct the work of all tezh- nical assistance advisors, but results were unsuccessful. Effectiveness of technical assistance was also undermined by the Government's attitude in providing housing, office facilities, and transport. Procurement of equipment, vehicles, and spares was affected by initial delays due to differences of views between the Government and IDA over the sufficiency of notiLfication for ICB. When this was resolved and delivery was due, the 1983-1984 suspension of disbursements by IDA for all projects in the country caused further delays and most of the items of equipment did not arrive on site until late 1984. The project fell short of completing its physical objectives and some of the works carried out have fallen into disrepair due mainly to the poor quality of the original works. Overall, the ERR dropped from an appraI.sal value of 70% to about 102. However, the trunk road maintenance organiza- tion has been established in each of the 10 regions covered by the project and budget allocations for road maintenance were stepped up to satisfactory levels. Maintenance of roads has been instituted but the quality of the work still does not keep the road in acceptable condition. Training at the Highway authority's facilities improved but local personnel were not yet able to take over as instructors at the end of the program. A number of lessons can be extracted: - it would have been preferable to reduce the scope of the project when it was realized late in 1984 that the Government could not maintain the entire trunk road network. Better planning for less ambitious but critical maintenance operations could have produced better results. - there are limitations to the effectiveness of technical assistance to strengthen road maintenance operations in the field. Many of the critical road maintenance tasks are simple technologically and their successful execution depends on close supervision, quality of leadership, and a better macro-economic environment; local resources are the best ingredients for the development of a sustainable road maintenance capability. - the project further illustrated the Government's ambivalent attitude to expatriate technical assistance staff. The Government recogni ed the need for technical assistance, but was concerned with the high costs and the burden of administering a large number of expatriates. - the project demonstrates the success of foimal training programs at education and training institutions as opposed to counterpart training in the work place. CONCLUSIONS The project was a logical development after a decade of involvement in the highway sector to upgrade the highway network. The exodus of indigenous staff of Asian origin in 1972 brought into focus the decline in road maintenance capability that had been going on for some time. Although the project covered only about 5100 kms, it was ambitious. The project area encompassed almost the whole of the northern half of the country. Distances are vast and communications especially with tae regions around Lake Victoria have always been difficult. During the rainy season, road access was cut off. In retrospect, the project area was too large. The project should have been tried out in one or two regions since the first phase was P failure for partly the same reason. The dqsirability of limiting the project area soon became obvious but Government had already deployed equipment and vehicles and allocated funds and other resources to all the ten regions and was reluc--ant to concentrate work on two regions. The implementation capacity of COMWORKS was limited by its reluctance to employ expatriate staff in line positions. Key vacant positions in the ministry were filled by inexperienced and sometimes unqualified staff. Technical assistance staff were employed as advisors to local counterparts but their presence added to COMWORKS administrative burden and some of the advisory positions were never filled. Work on building houses did not commence until after the advisors had arrived and transport was not ava'.i-ble to enable the advisors to do field work. Sometimes, the local cou-.--ziparts did not avail themselves of the services of their advisors. Thus, the effectiveness of the technical assistance staff was not evident until towards the end of the project period when loc,- and technical assistance staff collaborated better. The training component was successfully implemented both in respect of training of engineers at Indian universities and of road field personnel at the Moragoro Training School. With regard to the latter, there was inordinate delay in the selection of consultants. The Government agreed with IDA on the need for integration of the proposed new courses into the normal curricula of the school and admission of trainees from all government agencies besides COMWORKS and contractors. A major disappointment was that the quality the roads did not improve as expected although the funds were spent. What went wrong was that a few critical elements needed to actually perform the maintenance work were missing. When the labor and equipment were available, the bitumen was not there. When the labor and material were available, the equipment was not operational for lack of spare or of fuel or -f tyres, etc. In a shortage situation, it is hence essential that an order of priority be established to ensure that the more valuable activities can be carried out first. As a result, some of the roads have crumbled to the point where reconstruction at a high cost is required. Hence, it would have been preferable to reduce the scope of the project when, in late 1984, it was realized the Government could not maintain that entire trunk road network. This would have enabled the project to concentrate on key activities to preserve the paved road network, ano ensure that main trunk roads are kept open during the rainy season. Better planning for less ambitious, but critical maintenance operations could have produced much better results. It appears it is not economical to develop a full force account capability to execute road maintenance works. COMWORKS should limit its force account operations to routine .aintenance and adopt a policy of giving out the periodic and heavy maintenance works to contractors. A start has been made in this direction by Government in agreeing to carry out by contract, the rehabilitation cf gravel roads financed under the ongoing Sixth Highway (Rehabilitation) Project which includes a component to develop the local contracting industry. Ongoing Project as of June 1988 PROJECT NAME Port Rehabilitation Project (Cr.1536-TA) SAR DATE November 8, 1984 ESTT1MATED PROJECT COST US$91.0 million CREDIT AMOUNT US$27.0 million equivalent PLANNED COMPLETION DATE December 31, 1989 ERR AT APPRAISAL 33Z PROJECT BACKGROUND AND OBJECTIVES The project is designed to provide for the rehabilitation and modernization of the port of Dar-es-Salaam which, besides being Tanzania's major port, also serves as an important regional port for Burundi, Malawi, Zaire, Rwanda and Zambia. The project would provide special container handling facilities and equipment, replace some deteriorated general cargo equipment, modernize grain handling facilities, rehabilitate lighterage facilities and the petroleum jetty, repave sections of berths, and provide for a review of additional rehabilitation requirements, technical assistance, training, and consultancy services. The container and grain facilities would lower transport costs for Tanzania and land-locked neighbors, while the jetty. lighterage and paving investments are necessary to restore facilities to acceptable condition. Technical assistance and training would improve the Hardour's Authority ability to train its personnel, and improve its operational and maintenance capahility. One risk faced by the project is that traffic could fail to increase as forecast; however, container traffic is likely to continue to grow rapidly, and grain and petroleum products are unlikely to decrease significantly. Institutional benefits may materialize only slowly but training and technical assistance should riduce this risk. Ongoing project as of June 1988 PROJECT NAME Sixth Highway (Rehabilitation) Project SAR DATE April 2, 1986 ESTIMATED PROJECT COST US$107.7 million CREDIT AMOUNT US$50.0 million equivalent PLANNED COMPLETION DATE December 31, 1991 ERR AT APPRAISAL 38% PROJECT BACKGROUND AND OBJECTIVES The US$50.0 million equivalent project was approved in 1986. The Executing Agencies are the Ministry of Communications and Works, the Ministry of Education, the National Construction Council, and the National Transport Corporation. Cofinancing from DANIDA (US$2.0 million equivalent), NORAD (US$6.tl million equivalent) and a loan from the ADF (US$19.4 million equivalent). The project's fundamental objective is to reduce constraints by transport on the economy by rehabilitating the most important sections of the highway network. An important project objective is to improve the system of road maintenance by encouraging the MCW to carry out major maintenance and rehabilitation works by contract rather than by force account. It would also increase the capabilities of the local contracting industry, provide assistance to expand the capacity and efficiency of the trucking industry, and increase the availability of trained engineers in Tanzania. The main risks are institutional. The Ministry's own maintenance capability, as well as that of domestic contractors who are expected to participate in the project, may develop more slowly than anticipated. Implementation of the gravel roads rehabilitation program may je slower than expected. The project consists of: (a) A road rehabilitation program consisting of the rehabilitation of about 295 km of the TANZAM Highway, and 700-1000 km of gravel roads; (b) An equipment rehabilitation program; (c) A maintenance spares supply program to the trucking industry; (d) Assistance to the local contracting industry; (e) TA and consulting services for construction supervision, agricultural feeder roads study and feasibility studies and detailed engineering, and (f) Training. TANZANIA WORLD BANK/TAN2ANIA RELATIONS, 1961-1987 THE BANK'S ROLE IN SELECTED SECTORS I. WATER SUPPLY AND SANITATION Government Policies and Sector Organization 1.01 The Water Utilization Act of 1974 states that all water rights are vested in the government, sets the conditions and rules for the use of water and established a Central Advisory Board for the apportionment of the national water resources. Long term government policies aim at having all urban water supplies eventually operate on a self financing basis. The intention is that urban consumers with private connections should pay the full economic cost of water while water is free to public standpipe users. All water revenues are transmitted to the treasury and, in turn, funds for investments and operation and maintenance are provided from the central budget. Water supplies in rural areas are free of charge with all expendi- tures born by the government. 1.02 Due to the persistent economic difficulties of Tanzania, budgetary allocations for the water sector have been limited and the government has relied heavily on external, principally bilateral, financial assistance. While assistance to urban water services has been limited, the Nor-c coun- tries (particularly Sweden) have been providing continuous and massive grant funding to the rural water supply sector since Tanzania's indepen- dence in 1961. Over the period 1965 to 1984 some 70% of all water supply funding came from external sources, mostly grant funds. Current pledges are running at about US$ 16 million per year. 1.03 Technical responsibility for the sector is presently vested in the Ministry of Water Development (MND) created in 1987 as successor to the Ministry of Witer, Energy and Minerals. Under Tanzania's decentralization policy this responsibility is exercised through Regional Water Engineers who, however, report to the Regional Development Director on administrative and financial matters. The National Urban Water Authority (NUWA) was cre- ated in 1981 to carry responsibility for all urban water supplies under the MUD. In practice NUWA, at present, is only responsible for the Dar es Salaam water supply system but will take over all other urban systems when its capacity permits. 1.04 Sectoral planning for urban water systems is done on a town-by- town basis, as and when external financial assistance is available. The scarce funding situation effected not only expansion but, more seriously, maintenance of existing facilities. As a result virtually all urban water systems are operating at maximum capacity and, as a rule, are in dilapi- dated condition. By contrast, a long term rural water supply program was drawn up in 1972 to provide permanent and reliable water supply in all rural settlements by 1991. In 1974 the government decidei to oring the target date forward to 1980. The target was not achieved. 1.05 Planning and implementation responsibility for sewerage and sani- tation in urban areas rests with the Ministry of Lands, Houaing and Urban Development (ARDHI) and the individual towns are responsible for operation and maintenance of the facilities. Rural sanitation is the responsibility of the Miiistry of Health. 1.06 All sector organizations are characterized by weak management and acute shortage of skilled manpower. Agencies like NUWA and the municipal authorities (re-created in 1978) possess a degree of autonomy on paper but, in the absence of control over their finances and revenues, this is com- pletely ineffective. 1.07 Service coverages in water supply are claimed to be 85% in urban and 26% in rural areas. Actual service levels and, in particular quality, are far lower. Only six towns, Arusha, Dar es Salaaam, Moshi, Mwanza, Tabora and Tanga have sewerage services, serving between 10 and 20% of their respective population. About 90? of the urban population relies on pit latrines or septic tanks, many of them insanitary. There are no co- herent plans for the development of the sanitation sub-sector in either the urban or the rural areas. Bank Policies and Involvement 1.08 The Bank has never developed specific plans or policies for in- volvement in the water supply and sanitation sector in Tanzania. Lending policies, as discussed in the Country Program Papers (CPP) in the 1970s and early 80s discouraged all assistance for urban services partly to demon- strate full support for Tanzania's rural development policies and partly on the assumption that this will reduce rural to urban migration. Tanzania's persistent urban growth,ranging -rom 7 to 11? per year over the last 20 years (depending on the town), in spite of the poor urban services, clearly demonstrates that this assumption had no validity. First mention of a pos- sible urban water supply project appeared in the 1973 CPP. The same paper stated that "...there seems to be no pressing need for our involvement (in rural water supply] since bilateral assistance should prove to be ade- quate". In fact, the Bank ne-er formally participated in rural water supply developments in Tanzania and it would be a matter of guessing whether it should or could have contributed to the very large, mostly Nordic, bilater- al assistance which achieved only limited success. 1.09 Notwithstanding its negative attitude to financing urban services in Tanzania, the Bank reacted favorably to the Government's request in 1973 to assist in financing an urban water supply project to alleviate acute water shortages in seven towns. Following preparation and appraisal several cofinancing agencies expressed interest in the project. In the end, the Bank project was limited to only one town, Morogoro. With the drastic re- duction of the scope of the project the Bank clearly lost most of its lev- erage to influence sector policies and to shape sector institutional devel- opments. 1.10 The Urban Water Supply Project (Ln.1354-TA) was approved in 1976 and consisted of the construction of the Mindu Dam and associated treat- ment, transmission and storage facilities as well as limited improvements ti the Morogoro town distribution system. The small scale of the distribu- tion improvements was to considerably restrict the benefits accruing from the project. The bank loan of US$ 15 million had to be supplemented in 1982 by an IDA Credit of US$ 4 million (Cr. 1271-TA) and a Canadian International Development Agency (CIDA) loan of US$ 3.2 million to cover substantial cost overruns. These were mainly due to the delays resulting from the dismissal of the contractor on the Mindu Dam and the increased prices of the replacement contractor. The project was eventually completed in 1985 but as no completion report has been prepared, evaluation of the project's performance is not yet possible. 1.11 The Bank's efforts of improving the broader aspects of the sector under the project centered on a reorganization study prepared by consul- tants. The principal recommendation of the study, to create a national water and sewerage corporation, were not accepted by the Government and, apparently, the Bank did not press the issue. Instead, the Government cre- ated an Urban Water Supply Fund for the management and financing of the sector. In 1981 the Fund was converted into the National Urban Water Authority with initial responsibility for the Dar es Salaaam water supply system and the longer term prospect of taking over responsibility for all urban water supplies. Neither organization ever really got off the ground due to ineffective management, lack of trained staff and virtually no fi- nancial support. Although the Bank provided some intermittent and limited support to NUWA through technical assistance financed under the Second Site and Service (Cr.732-TA), the Dar es Salaam Sewerage and Sanitation (Ln. 1312-TA) and the Power Rehabilitation Project (Ln.1687-TA) NUWA remains an ineffective and underfunded organization. It has no real autonomy and the prevailing tariff pGlicies do not provide sufficient funds to cover opera- tion and maintenance. NUWA's insolvercy spread its effects to other sectors of the economy. In particular, as a major electric power consumer which does not pay its bills, NUWA has become a major factor in the financial problems of TANESCO, the power company. 1.12 The second Bank financed project in the sector, the Dar es Salaam Sewerage and Sanitation Project was approved in 1982. It was designed to assist in the implementation of the first stage of the Dar es Salaam Sewerage Master Plan completed in 1979. The project was modest in scale and short on sectoral objectives. It consisted of the rehabilitation of the totally inoperative Dar es Salaam sewer system and improvements of the on- plot sanitation and pit emptying systems which serve some 90% of the city's population. The project is still being implemented. Conclusions 1.13 Even such a brief review of the water supply and sanitation sector in Tanzania and the Bank's involvement in it paints a sobering picture. The major part of the unevenly populated country is dry with limited surface and ground water resources. Its humid coastal area with the heaviest concentration of people is conducive to the spread of water borne diseases. Lack of skilled managers and technicians, misguided economic and social policies, underinvestment and years of neglect of existing assets created a parlous state of affairs in the sector. The fast population growth is outstripping the meager investments; and very likely service coverage is dropping. The depressed economy is clearly unable to generate the funds for the needed investments. If urban services, particularly water supply and waste disposal are further neglected an emphatic adverse impact on the economic and social environment will be inevitable. The Bank clearly had no coherent policy of support for these sectors in the past and few are the signs that this might change. Overall Bank support for Tanzania has accel- erated recently and it is time that the role of the water supply and sani- tation sector and urban services in general be apprnpriately recognized in the country's overall development. II. POWER AND ENERGY The Sector and Government Policies Energy 2.01 Although det&iled analysis of the country's energy resources did not take place until the early 1980s, the economic advantages of exploiting the country's relatively abundant hydroelectric potential versus thermal generation based on imported oil had been clearly recognized in long range planning for the sector. Studies in the early 1970s identified about 1315MW hydro-generating potential, small and remote coal reserves and noted the extensive use of wood fuel. Successive investigations, culminating in the 1984 study: "Tanzania: Issues and Options in the Energy Sectorf (UNDP/World Bank) provided a comprehensive assessment of Tanzania's energy resources. The principal exploitable sources identified to date are as follows: (i) hydro potential- 4,500 MW; (ii) natural gas-800 billion ft3; (iii) coal-300 million tons. Oil explorations are continuing, so far, unsuccessfully. These studies also highlighted the fact that over 80Z of Tanzania's total energy consumption is in the form of firewood and charcoal. And, while the country has substantial forest areas, the uneven population distribution threatens total deforestation in some areas. An intensive program of studies on this issue is in progress including reforestation and pilot projects on more efficient charcoal production and cooking stoves. Studies are also under way to assess the most economic use of the natural gas and coal reserves. WWH Electric Power 2.02 Public supply of electricity in Tanzania was first provided in Dar es Salaam, Dodoma, Kigoma and Tabora by the German East African railway, in the period 1907-1914. These facilities were taken over, during the First World War by the British and operated under the Tanganyika Public Works Department. In 1932, the Tanganyika Electric Supply Company Limited (TANESCO) was formed, with capital provided by the East African Power and Lighting Company Limited of Kenya (EAP&L), primarily to develop the Grand Pangani Rapids hydroelectric facilities to supply electric power to the sisal estates in the Tanga Province. At the same time, the Dar es Salaam and District Electric Supply Company Ltd. (DARESCO) was formed to take over all government owned electricity utilities in the country. In 1957 the two companies were amalgamated under the name TANESCO, by which time they op- erated 15 separate undertakings, each with its own generating facility. 2.03 Soon after independence the Tanzanian government purchased the shares of TANESCO from EAP&L and TANESCO became the only licensed electric- ity distributing company in Tanganyika. To this day, the government remains the sole shareholder of TANESCO. The company's Board of Directors consisted of a chairman, deputy chairman and six other members, predominantly govern- ment officials and a business representative. The company's structure and commercial operating orientation, with changes in Ministerial responsibili- ties and internal organization, essentially remained the same to this day. TANESCO operates under a license issued in 1957 which is due to expire in 2012. The license defines the rights and obligations of the company, makes provisions for the exploitation of hydroelectric potential of rivers throughout the country, provides for the right to export electric power to neighboring countries and sets a ceiling for the tariffs that TANESCO may charge for the supply of electricity. In its early years TANESCO was well managed and staffed and operated satisfactorily in every respect. 2.04 For a long period TANESCO's expansion plans were in response to the demand for power implied by the Governments industrial development strategy. The Government also laid strong emphasis on rural electrifica- tion. However, it was recognized that the relatively small demand would not ensure economic or financial viability for such a program and, it was agreed that any such program would be implemented outside TANESCO's activi- ties. 2.05 Sector development strategies were, basically, straight forward, concentrating on a realistic expansion of the system, based on reliable demand forecast, accurate location of demand centers and, in the longer term,the creation of a national grid system to permit central control and the most economic use of generating sources. 2.06 Market analyses and demand studies in the 1960/70 period indicated that TANESCO's sales, on the average, were distributed as follows: in- dustry-65%; domestic-20Z and commercial-15Z. Annual demand growth rates averaged 8Z between 1962 and 1980 with periodic industrial growth rates reaching 25Z per annum. Investment programs followed these trends and in- stalled generating capacities grew from 70.5 MW in 1967 to 401 MW in 1986 of which some 327 MW was hydro power. Expansion of the transmission network followed the demand with the additional consideration of replacing indivi- dual diesel generation with hydroelectricity. However, in spite of the steady growth, in 1980, still only 7% of the country's population had ac- cess to electricity. At that tine, Tanzania was already in the midst of severe economic depression. Altaough development was being retarded across all sectors, power sector growth forecasts continued to be optimistic and expansion of generating capacity and transmission and distribution networks continued. 2.07 Throughout the period up to the oarly 1980s the institutional performance of TANESCO continued to be rated good. Starting from a sound base, with periodic reorganizations, effective staff training and prudent financial management the organization matched its physical growth. Its expatriate staff was reduced from 50 in 1971 to 6 in 1983. The financial position was sound, TANESCO was able to increase its tariffs any time (without government approval) by 1OZ to match inflation and fuel costs. Billing and collection were efficient with arrears ranging from 40 to 60 days billings. Revenues .increased, between 1971 and 1980, by an average of 21% and TANESCO was able to contribute to investments from cash generation in the range of 20-40%. System losses averaged a creditable 15?. Foreign exchange financing was forthcoming, initially from the World Bank, and subsequently from a number of bilateral and multilateral sources. Project implementation was generally timely and efficient although substantial cost overruns occurred on the two Kidatu hydro projects due to unforeseen con- struction difficulties and world wide inflation due to the oil crisis. 2.08 The deepening economic crisis after 1978 took its toll on the sector. Production cutbacks, due to shortage of foreign exchange to buy imported inputs, undermined the growth in demand for electricity. Mean- while the lack of foreign exchange to buy spare parts resulted in the post- ponement of maintenance. In 1985 the bubble burst. A comprehensive assess- ment of the status and performance of the sector revealed a shattering scenario. Installed generating capacity was found to be significantly ex- ceeding demand. Virtually all thermal generating facilities, transmission and distribution systems were on the verge of collapse. Major outages and breakdowns reached 500 per year and were rising. In Dar es Salaam the breakdowns frequently paralysed the city's water supply. System losses jumped to 21?. Financial management of TANESCO collapsed with the dismissal of the entire accounts department in 1983. Accounts receivable increased to 213 days of sales (601 due to government) and audited accounts were delayed by over a year. Staff moral was low due to lack of incentives and mainte- nance was virtually non-existent. At the government level, the creation of new agencies and the involvement of non-sector ministries confused the areas of responsibility and diverted resources from the rational develop- ment of the sector. 2.09 In 1986 a major and comprehensive program was launched to rehabil- itate the sector. Development plans were revised, discussions were started with neighboring countries with a view to arranging the export of surplus energy, and institutional and financial improvement programs were mapped out. Detailed studies got under way to determine the most economical utili- zation of the natural gas resources (fertilizer v. energy) and pilot proj- ects were initiated to assess the most efficient use of fuelwood. Investi- gations are under way for the domestic production of low cost electric cookers to take advantage of cheap hydro power. Extensive assistance is being provided by the Nordic countries in investigating the feasibility of min-hydro installations. On the institutional side, reorganization and improvements are under way within TANESCO. Staff has been cut by 17Z. The government has reduced its arrears and more stringent disconnection prac- tices been put in force. A tariff study is being prepared with particular attention to the uniform tariff structure under which more efficient areas cross subsidize expensive local generating supplies. The rehabilitation program is all encompassing but its results will not be seen until some time in the futuve. Bank Involvement and Lending 2.10 Bank involvement in Tanzania's power sector started in the mid 1960s and turned into a sustained support through four ioans (Ln. 518/67; Ln.715 I&II/1970; Ln.1306/76) and two credits (C.1405/83; Ci.1687/86). From 1980 Bank assistance expanded to the energy sector as a whole through four more credits: Cr.S27/80-Songo-Songo Gas Exploration; Cr.1199181-Petroleum Exploration; Cr.1371183-Coal Engineering and Cr.1604185- Petroleum Technical Assistance. The total lending amounted to US$ 220.5 million equivalent. 2.11 After 1980 Bank assistance took on a comprehenive sectoral as- pect, and the country's assessment of the overall energy resources started. Up to that point Bank assistance was a straight forward financial support to an, apparently, satisfactory borrower, TANESCO. High demand projections seemed to justify strong expansion of the power sector and Country Program Papers (CPP) continued to treat the sector specifically, merely recording the intention to continue to lend. Bank involvement, eventually, attracted substantial cofinancing and, by all accounts, the power sector did not lack the resources to expand even in the environment of a progressively deterio- rating economy. Bank influence was significant in the areas of improved planning, procurement, institutional improvement and, in particular, train- ing of national staff. 2.12 The history of Bank/Tanzania relationship in the power/energy sector appears to have been generally harmonious. Criticism that surfaced relates to the early Bank attitude of proceeding project-by-project with little attention to overall planning. Strong and, by all accounts, valid criticism was raised regarding unreasonable Bank covenants, in particular, relating to the revaluation of TANESCO's assets. Bank requirements produced asset value increases of 55% in 1976, 26% in 1978 and 33Z in 1980. These - A.- resulted in excessively high tariffs and more than required cash genera- tion. The matter was eventually rectified with the modification of the appropriate covenants. High turnover of Bank staff has been cited as a cause of lack of continuity in dealing with problems. While not documented, this may well have contributed to the astonishing revelations in 1985 which depicted a previously Osatisfactory" sector being on the verge of total collapse. Neither the documentation of the 1986 Power Rehabilitation Project (Cr.1687-TA) nor contemporary country briefing papers offer any clue as to a) how this situation developed and b) why it was not identified and tackled by the Bank much earlier. The proverbial Tanzanian economic morass is not a sufficient answer and, furthermore, the country was sup- posed to have turned the corner by that time. Conclusions 2.13 The review of the Bank's role in the Tanzania power sector clearly indicate that TANESCO relied heavily on the Bank for advice and support. It also seems clear that in the critical period of late 1970 and early 1980 this support was inadequate in that the deteriorating trend of the sector was either not identified or was ignored. The Power Rehabilitation Project includes a comprehensive package of remedial actions but its effect remains to be seen and only the future will show whether the damage is lasting or reparable. It is difficult to escape the conclusion that the Bank's over- riding concern with policy adjustment programs (however vital these might have been) resulted in the neglect of essential infrastructure needs. The striking similarity of the "progress" of the telecommunication sector to that of the power sector, the desperate state of the water supply/sanita- tion sector and the chronic shortage of housing, all of them most severely affecting the poorest section of society, seem to testify to this neglect. A more balanced Bank policy of assistance is sorely needed if the standard of living of the people of Tanzania is to improve. III. TELECOMMUNICATIONS Background 3.01 The East African Post and Telecommunication Administration (EAPT) was organized in 1948 to manage and operate the services in Kenya, Uganda and Tanzania. EAPT also owned 60% of the shares of the East African External Telecommunication Co. Ltd. which provided international service beyond the borders of the three countries. The remaining shares of this company were owned by the British company, Cable and Wireless Ltd. Follow- ing independence, with the signing of the East African Cooperation Treaty in 1967, the East African Common Services Authority took over responsibil- ity for a number of services in the three countries, including telecommuni- cations. EAPT, renamed East African Posts and Telecommunications Corpora- tion (EAPTC), operated and managed the service. The Authority's executive arm was the Ministerial Communications Committee, consisting of the Communications Ministers of the three countries. The head offices of the common services were located in the three capitals, EAPTC was in Kampala, Uganda. 3.02 Following the break up of the East African Community, Tanzania established, in 1977, the Tanzania Posts and Telecommunications Corporation (TPTC), an autonomous government-owned corporation under the Ministry of Communications and Transport, to manage and operate its national services. The Sector and Government Policies 3.03 The earliest available documentation, regarding the posts and telecommunication services in Tanzania, cover the mid 1960s and the infor- mation mostly relates to the East African Community as a whole. Beyond some basic data on service coverage there is little information regarding Tanzania itself. Altho-u2h EAPT carried responsibility for postal services, these operations were separate from the telecommunication services and, on the whole, paid their own way. On the telecommunication side, EAPT ran a well managed and financially viable operation. 3.04 In 1967, the three East African countries, with a total population of about 25 million, had a low but, compared to other African countries, reasonable telephone density as follows: Kenya 0.53 telephones per 100 inhabitants Uganda 0.23 Tanzania 0.19 " a a a Ethiopia 0.08 " a a a Nigeria 0.15 a " " a South Africa 6.02 a a a a The majority of the traffic was between major urban centers and economic growth points. International telex was available in Dar es Salaam, Kampala and Nairobi. International telephone services operated by high frequency radio telephones. Radio call services operated privately under licence were introduced in 1960 to remote areas. EAPT's expansion program was based on operational improvement requirements and projected growth of demand with waiting list of subscribers as the main indicator. Most available data refers to aggregate expansion plans with only limited breakdown on a coun- try by country basis. However, investments appear to have been, by and large, evenly distributed. 3.05 In 1977, TPTC took over the Tanzanian part of the telecommunica- tion system and generally followed the same sotnd policies of its predeces- sor. Priority for investments fluctuated between the business and govern- ment needs in urban areas and the desire to bring the remote rural areas into the communications network. On the break-up of the EAC, Tanzania started with the halidicap that it could not easily obtain its records from Nairobi and that few Tanzanian nationals were employed (and trained) in EAPT's operations. In subsequent years the shortage of skilled manpower, the progressive deterioration of Tanzania's economy and the persistent shortage of foreign exchange combined to severely limit the efficiency and - 212 - growth of the telecommunication services. The investment programs up to 1987, while reasonably continuous and well planned, had to be tailored to available resources, and could not keep up with the high growth rate of population, the accelerating demand (64,000 on waiting list in 1986) or even with the deterioration of the existing facilities. Break downs and service outages, due to lack of spare parts and effective maintenance, reached such proportions by the early eighties that they effactively hin- dered overall economic development. A telling statistic is the number of telephones per 100 population which, in Tanzania, shoved the following "progress": 1967 1970 1977 1981 1987 0.19 0.27 0.40 0.21 0.20 3.06 In terms of organization and management and financial position TPTC had an excellent start. It had a well qualified top management (al- though grave shortage of technical staff), sound organization and, general- ly, adequate autonomy. Although its assets could not be determined accu- rately (the Nairobi records were not available), its tariff system provided an estimated rate of return of 49% on historic cost of telephone plant and a 55% operating ratio in 1979. Its billing and collection performance was reasonable. Over subsequent years, this performance has steadily deteri- orated due mainly to the lack of long term availability of foreign ex- change. This not only prevented systematic planning and execution of neces- sary expansion programs, it also severely limited replacement and mainte- nance work both in the field and in operational management. Lack of compu- ter facilities slowed billing and collection and the absence of efficient management information system hindered internal progress. In 1987 a major rehabilitation project was launched with Bank assistance with a comprehen- sive program of improvements. Its results are awaited. Bank Involvement 3.07 Bank involvement in the Tanzania telecommunication sector started in the mid 1960s with the preparation of the first East African Community telecommunication project (Ln.483-EA, US$ 13 million) and continued through two more loans to EAPT (Ln.675-EA, US$ 10.4 million and Ln.914-EA, US$ 32.5 million) and two credits to Tanzania /TPTC (Cr.1173-TA,US$ 27 million and Cr. 1810-TA, US$ 23 million); a remarkably well sustained support. All operations were designed to support a time slice of EAPT's and subsequently TPTC's long term development program, except the 1987 operation (Cr.1810-TA) which is aimed at the comprehensive rehabilitation of the telecommunication services in Tanzania. All early operations were straight forward provision of financial assistance to a well functioning agency. Thare was little, if any, policy content (not really required), but available evidence points to substantial technical advice and assistance being provided by the Bank. - 213 - 3.08 Completion Reports prepared for the first three projects are a revealing in many ways. All three projects suffered from delays in imple- mentation of 2, 1.5 and 6.5 years respectively. The inordinate delay on Ln.914-EA was largely due to political turmoil within the EAC. The benefits and adverse effects of the Bank's procurement requirements feature promi- nently in the history of all projects. There is ample evidence of the gains from international competitive bidding in terms of lower prices and quality contracts. On the other hand substantial delays resulted from unfamiliarity with the process, disputes and need to rebid contracts where the outcome proved unacceptable to the Bank. National policies of africanization of expatriate positions were mired in conflicts between the needs for quali- fied experts and the acute shortage of trained local manpower. 3.09 It is clear that the Bank's role in the sector was immensely beneficial in may ways. Technical advice resulted in appropriate technol- ogy, desirable standardization, better procurement procedures. Sound finan- cial and tariff policies were implemented and the Bank's sustained support was almost the only source of foreign exchange for the sector. Only in the last project (Cr.1810-TA) was there significant cofinancing from bilateral and multilateral agenc-es and for this, the Bank was the catalyst. At the same time the Bank was also the subject of criticism. Unreasonable strin- gency on standardization, lack of sensitivity to local conditions, insuffi- cient supervision were some that were voiced. Staffing of Bank missions appear to have been a significant factor in country/Bank relationship. There remains a much stronger imprint of a Bank mission leader than any general Bank policy or practice. Conclusions 3.10 The telecommunication sector presents a somewhat unique situation as far as Bank assistance to a developing country is concerned and Tanzania is no exception. Amid the general clamor for low cost solutions, this sec- tor, rightly so, presses for state of the art technology. In that area, the Bank's advice (if anything, insufficient) was all important and highly appreciated. The progress from the early 1960s can be traced as going vir- tually from the proverbial "cleft stick' to space age communication. At the same time, such a situation Lonfers a high degree of responsibility on the Bank which must be reflected in the staffing of such projects. Earlier, and more extensive, training efforts would also have paid high dividend. 3.11 Compared to other sectors in Tanzania, telecommunication got off a flying start. A sound organization, good tariff and financial policies and a reasonably well prepared long term program facing high demand and will- ingness to pay promised a bright future for TPTC in 1978. There were few political problems and long term Bank support appeared to have been com- mitted. What went wrong? There can be few sectors where the overriding importance of the deteriorating economy of the country and, in particular, the chronic shortage of foreign exchange is so clearly demonstrated. Demand for the services continued to grow during the worst years of economic de- pression, the (local) financial situation of the agency remained good, yet the service continued to deteriorate. A clear conclusion can be drawn for the Bank from this, which is that in selecting priorities for support, more attention should pe paid to such potentially successful and economically vital sectors such as telccommunications, especially where the sustainabil- ity of earlier projects can be enhanced by further Bank support. IV. URBAN DEVELOPMENT Government Policies and The Sector 4.01 At independence, Tanzania was one of the least urbanized countries with only 4.6% of the population living in urban areas. While the level of urbanization is still comparatively low, 141 in 1985,1 the rate of urban growth over the last 25 years at 8.52 was one of the highest in Africa. One of the main reasonn is the proliferation of smaAl urban communities which grew out of the Governments general policy of decentralization of manufacturing and service centers to support rural development and increase the standard of living of the rural population. This rate of growth ap- pears likely to continue for the next two decader. Dar es Salaam , with 1.4 million people has one-half the total urban population, and is more than six times bigger than the next largest town, Mwanza. 4.02 The Government has not developed comprehensive urban policies to cope with the growth. In 1971 the nationalization of private properties not owner-occupied virtually eliminated the housing rental market.2 The abolishing of local authorities in 1972 removed even the limited urban management as the incipient technical staff of the urban councils was re- placed by party bureaucrats. The taxation system was designed to ensure a better income distribution and all public finances were centralized. The newly created District Development Councils (DDC) had no authority to raise revenues and the variety of local property taxes were consolidated, in 1974, into a single levy, the Land Rent and Service Charge. With this move, the link bitween revenue co)lection and expenditures in urban areas has been broken. 4.03 Due to the extreme shortage of skilled manpower and financial resources and inefficient housing policies, the rapid urban growth resulted in acute shortage of (particularly low cost) housing and community facili- ties. In Dar es Salaam alone, 652 of the population lived in unplanned squatter areas in the early 1970s. The National Housing Corporation (NHC) which produced rental housing, had low output, high standards and costs which were unaffordable to the majority of the low income population. In 1973, President Nyerere launched a National Sites and Service program for 1/ IBRD: World Development Report 1988, Table 32; Oxford University Press. I/ Acquisition of Buildings Act, April 1971. ten regional centecs with the aim of providing 12,500 serviced plots an- nually, with basic infrastructure and community facilities. To facilitate self-help housing construction the Tanzania Housing Bank (THB) was estab- lished in 1973 to provide loans for low income households. To improve the management of the program ARDHI was reorganized, it established a Site and Services Section and transferred its implementation responsibilities to the regions. 4.04 The chronic manpower shortage, inefficient and fragmented manage- ment and pilicies and the economic crisis of the late 1970s all contributed to the limited success of the urban programs. The reestablishment of the local authorities in 1978 and the partial transfer of government admini- stration to Dodoma, the proposed new capital, did little to improve the situation. The Government's interest rate policies virtually wrecked THB wbich was forced to extend housing loans at 42 at the time of 302 inflation an coamercial bank rates of 13%. In addition THB faced extraordinarily low deand for its loans (only 3Z of households) as well as low recovery. In 1986 about 80Z of payments due remained uncollected. Cost recovery in the housing sector has been unsatisfactory, the lowest rates occurring in Dar es Salaam at 132. In summary the sector has neither organizational, policy or financial base on which to operate. Bank Involvement 4.05 In 1972 the Government requested the Bank to provide assistance to the implementation of the Site and Service program. The First National Site and Service Project (Cr. 495-TA) was approved in July 1974. The US$ 8.5 million IDA credit was to support the provision of some 10,600 serviced plots in three towns (Dar es Salaam, Mbeya and Mwanza), community facili- ties, consulting and technical assistance services, studies and tr!aining and financing for THB's soft loan program. 4.06 With the first project still under (slow) implementation the Second National Site and Service Project (Cr. 732-TA) was approved in July 1977. The US$ 12 million IDA credit was to continue the support for the site and service program in five towns (Dar es Salaam, Iringa, Morogoro, Tabora and Tanga) with increasing emphasis on squatter area upgrading. The project was to upgrade 15,800 squatter houses, provide 19,000 surveyed plots, community facilities, further funding for THB's construction loans, technical assistance and preparation for a follow up project. In addition a pilot effort was included to assist small-scale induatries to generate employment in urban areas. 4.07 Both projects have suffered substantial delays and the second " project was considerably reduced in scope during implementation. Few if any of the institutional goals were achieved and the audit of the second proj- ect was Onon-performing on financial grounds". In view of the poor perfor- mance of the project the Government cancelled the preparatory work for the follow-up project. 216 Conclusions 4.08 Probably the only real success achieved by the Bank's two urban operations was that they helped to establish the validity of squatter up- grading and site and service programs as viable methods of providing hous- ing on a large scale to low income population in urban areas. By hindsight it appears clear that the Bank moved into the sector with too much optimism and too little preparation. Launching the second project without the les- sons learned from the first one was a major error. Tanzania's economic and political situation and administrative unpreparedness ought to have dic- tated a more cautious approach and reduced complexity in the projects. The Bank was much too -eady to be flexible on all fronts, in particular. to accept financial sector policies which implied generous and unaffordable subsidies under the banner of equitable socialist policies. Since the clos- ing of Cr. 732-TA in June 1984 the Bank has been inactive in the sector in Tanzania. While the two projects were less than successful, Bank involve- ment generated many useful initiatives, particularly in the national plan- ning, project implementation and manpower der lopment areas. The direct technical assistance given by Bank staff had lasting effect. Tanzania's urban service delivery problems seemed almost insurmountable and there was little doubt that the Bank was the only source of effective assistance as virtually all bilateral aid was directed at rural development. An urban crisis was looming, and continues to loom, and the possible consequences are not fully appreciated. Sustained Bank assistance may be justified and should be accorded appropriate priority for further study, given the cli- mate of renewed support for Tanzania. U 217 TANZANIA WORLD BANK/TANZANIA RELATIONS, 1961-1987 THE EDUCATION SECTOR I. THE EDUCATION SECTOR AND ITS EVOLUTION 1.01 The education sector is an important part of the story of Tanzania Bank relations for three reasons. First, an important explanation for the difficulties experienced in all other sectors is the shortage of adequately trained manpower, especially at middle and higher levels. What the Govern- ment and the Bank did or did not do in this sector has implications for all other aspects of the development effort. Second, while efforts by all parties--the Government, the Bank and other donors--to solve this problem over the last 25 years have been substantial, the results are disappoint- ing. The education system is now turning out graduates at a dramatically expanded rate* but quality of the output is low and the same complaints about shortages ot adequately trained manpower are still heard throughout the economy and still used to justify the large numbers and expense of foreign experts and technicians. Moreover, the costs are high given what is being achieved. Have these efforts been inadequate or misdirected; or is too much being expected too soon? Third, the education sector provides useful, concrete examples of the nature of the relationship between Tan- zania and the Pank over this extended period of time. 1.02 Part I reviews the history of the education sector since Indepen- dence. Part II reviews Bankl/ involvement in this sector, focusing on the seven education projects it has financed plus related policy dialogue and sector work that has supported these efforts. Part III discusses lessons and future directions. Training efforts associated with non-education projects are not covered in this review. A. Education Policy 1.03 At the time of independence in 1961, Tanzania had few African citizens of labor force age with more than a year of education, extremely low primary and secondary enrollment rates2/ and only a handful of univer- sity students (all studying abroad in the absence of local facilities). Government administration funct.oned adequately by utilizing large numbers of expatriates in middle and higher level positions. In 1961, three- fourths of established middle and senior level civil service posts were filled by non-citizens. The number of vacancies at these levels began to increare rapidly, however, as Government expanded its functions and expa- triates began to depart. I/ Throughout, the word Bank is used interchangeably with IBRD and IDA 2/ Primary and secondary enrollment rates in 1960 were 24 and 2 percent, respectively, compared to 36 and 3 percent for Sub-Saharan Africa and 80 and 18 percent for all low income economies. - 218 - 1.04 In 1962 the first of a series of carefully prepared manpower pro- jections was completed. In the light of these projections, the First Five Year Plan (1964-69) gave highest priority to the expansion of secondary and higher education, and within those areas to teacher training and science- based subjects, the goal being to fill middle and higher level manpower gaps as quickly as possible. The focus in primary education was less on expansion than on quality improvement (subventions designed to guide local authorities in their development plans, replacement of the post-primary teacher training program with a post-secondary program and improvements in curricula and teaching materials). 1.05 Then, in 1967, the Arusha Declaration and the policy paper "Educa- tion for Self-Reliance" (ERS) were published. The Arusha Declaration is significant in the present context because it greatly expanded the need for middle and higher level manpower by expanding the functions of Government, nationalizing all major private enterprises and at least indirectly encour- aging a more rapid exodus of expatriates and European and Asian citizens. The ERS document established a philosophy for the education system that emphasized self-reliance (rapid replacement of expatriates), the right of everyone to a basic education, collective as opposed to individual achieve- ment, citizenship, and community service. As translated into policy mea- sures, this emphasis has meant (1) a crash program to establish universal primary education by 1989 (amended by the "Husoma Declaration" of 1974 to be achieved by 1977, and which was in fact achieved for the entering age group shortly thereafter); (2) restriction on expansion of secondary and tertiary education to the minimum necessary to serve manpower requirements; (3) establishment of several significant adult and community education programs; and (4) significant curricula changes that introduced practical income earning activities at all levels, courses in civic and socialist values at all levels and vocational subjects into secondary education. As general directions thrse policy measures have remained in force from 1967 to the present. There have been, however, many twists and turns related to how these policies are implemented and who pays for them which have at times been quite disruptive. These include decentralization, introduction of self-help measures, institutional reorganizations, mobilization of per- sonnel for crash programs, and changes in the nature of the adult and com- munity programs. B. Achievements and Problems 1.06 For the most part the policy directives of the ERS document have been fully implemented. Expatriates in established middle and senior level civil service posts were 75 percent of the total in 1961, 27 percent in 1967 and 6 percent (of an establishment three times larger) in 1974. How- ever, these figures do not reflect the increase in technical assistance ~1.. H personnel who to some extent substituted for expatriates in established posts. The primary enrolment rate, which was 24 percent, (compared to 36 percent for SSA), increased to 93 percent in 1980, (compared to 76 percent for SSA)I/ Secondary and university enrolments have been severely re- stricted; while the numbers have increased rapidly from their very low bases, enrolment rates remain extremely low--indeed, no other country in the world except Rwanda has a lower secondary enrolment rate (22 in 1982, compared to 201 for SSA and 17% for all low income economies)--and the progression rate is only 72 (down from 122 in early 60s, in part because of expansion of primary enrolment) (see Table 1 & 2 and Figure 1). Adult literacy has also improved dramatically, from recorded figures (which should only be used as an indication of orders of magnitude) of 10 and 32 percent in 1960 and 1967 to 93 percent in 1986. And all the planned cur- ricula changes have been implemented. 1.07 But along with these achievements have come a variety of severe problems. First, quality of education, while a cause for concern since the sixties, appears to have deteriorated, particularly at primary level. It is especially poor in reading in Kiswahili (the medium of instruction in primary school), understanding as well as reading of English (the medium of instruction in secondary school) and mathematics and science subjects, just where strengths are most needed.4/ 1.08 Second, unit costs at the secondary and tertiary levels have been high and rising. In 1970 per student recurrent costs alone for these two levels of education were 2.5 times and 30 times GNP per capita, compared to 3/ Since 1980, however, the rate has declined. At least part of this decline is likely to be due to a decline in the numbers of older students who were encouraged to enroll and have now passed through the system. A/ A summary of recent evidence indicates extremely high failure rates (in 1986 nearly 50 percent of those taking Standard IV examinations failed with marks below 25 percent) especially in reading and mathematics (a 1985 study indicates that 2/3 of Standard VII students are unable to read and understand any written text, whether in English or Kiswahili and at least 95 percent have virtually no understanding of English, despite five years of at least some classroom instruction. So far as secondary schools are concerned, a 1981 study found that 15 year-olds scored lower than 10 year- olds in other parts of the world on standardized reading and mathematics tests, a 1985 study concluded that knowledge of English (the medium of instruction) is so poor that "throughout their secondary school career little or no subject information is getting across to about 50 percent of pupils in the sample, and a review of "A" level science examinations indicates that average success rate in mathematics, chemistry and biology declined from an already low 54 percent in 1982 to 27 percent in 1986. See UNESCO, The Financing of Education in Tanzania, Overview, 1988, paragraphs 38 and 68-71. 1.1 times and 11 times GNP per capita in SSA as a whole; in 1983, they increased to 3.4 times and 35 timee GNP per capita whereas they have de- clined substantially in other SSA countries (see Table 3 and discussion of financial problems below). 1.09 Third, severe shortages of manpower requiring secondary and ter- tiary levels of education, especially in occupations including teaching that require mathematics or sciences, persist. As late as 1987, only 4 percent of the labor force had any education beyond primary. Even among middle and upper level manpower (craftsmen machine operators, clerical workers, professional and technical workers, teachers, managers and admini- strators) only a third had any education beyond primary. These results follow from the Government's policy of rapidly expanding primary and only slowly expanding post primary education. While beneficial for equity as defined in Tanzania, this distribution of educational benefits must be having serious adverse effects on productivity and economic growth. C. Proximate and Underlying Causes 1.10 The proximate causes of these problems are well understood and need only be briefly summarized here. - It was deliberate Government policy to restrict expansion of secondary and tertiary education. Expansion was supposed to occur to the (minimum) extent necessary to meet manpower needs. But these needs were underestimated (see discussion of manpower pro- jections below); and no one (at least according to the evidence available) seemed to have seriously challenged these projections, probably because they provided a justification for the slow expan- sion of secondary and tertiary education that the Party decided it wanted in any case. - Instructional materials (textbooks, exercise books and teacher guides) are all extremely limited, in part because not enough are produced but also because distribution is poor; that is, even when produced they often fail to get into student hands, a result of bureaucratic procedures, poor management and shortages of funds. A rough estimate indicates that in 1986 the stock of textbooks was sufficient to provide 40 percent of primary students with a Kiswahili textbook and 15 percent with a mathematics textbook; needless to say, fewer than these numbers actually had textbooks. A 1981 study indicated that in secondary schools and Teacher Training Colleges the supply of textbooks was adequate for only one in four students. The situation had to be worse in earlier years. - Teacher competence and motivation are poor, a result of poor training, lack of instructional materials and inadequate remunera- tion. At primary level, the worst period was 1979-81 when some 35,000 mostly primary school leavers were given crash courses to force the pace of achieving universal primary enrolment; most are still in the system with only that initial training. So far as the secondary level is concerned, even as late as 1986 only 25 percent of secondary teachers had degrees from Teacher Training Colleges or from the University; the rest were graduates of two- year post-secondary programs with significantly lower qualifica- tions. And Teacher Training Colleges are themselves poorly en- dowed, lacking in equipment, library facilities, qualified staff, instructional materials and means of transport for teacher prac- tice. Not all graduates of Teacher Training Colleges go into teaching or remain in teaching very long. Attrition rates are high especially for science and technical teachers who find they are in high de- mand in the private and parastatal sectors. This has led to a perpetual shortage of science and technical teachers in secondary schools, underutilization of science and practical facilities and insufficiently qualified applicants to enter training programs in these fields. Curricula and time schedules are overloaded. Primary students devote over a third of their time to Kiswahili and English lan- guage instruction (which is too low), a quarter of their time to arithmetic (which would be adequate if good materials were avail- able) and the remaining 40 percent to everything else, including cultural subjects, physical education, political education and school product.on projects (ERS activities). The situation is more serious at the secondary level where, at considerable ex- pense, pre-vocational training courses in agriculture, commerce, technical and domestic science have been added to the curricula. The point of "diversifying" the curricula in this way, at least according to Bank documents, was to better prepare secondary grad- uates for direct entry into the labor force. But the attempt to do that as well as prepare students for further training, given shortages of resources, qualified teachers and time in school, has meant that neither goal has been accomplished well. Buildings, furniture, equipment and maintenance, especially at the primary level, have been inadequate. Local communities are sup- posed to provide the walls and flooring for simple primary schools while the Government provides the roof (generally corrogated sheeting) and other imported items. More often than not the roof- ing materials fail to arrive and the walls (generally made of pressed earth) begin to deteriorate. The shortage of primary classrooms is such that it is not uncommon to find 80 to 100 stu- dents and their teachers all in one room. At the secondary and higher levels, shortages of classrooms and boarding facilities have also been a problem, but at least in part because of the way in which the facilities are used. A 1966 UNESCO report indicated that, if used properly, secondary classrooms and boarding facili- ties were adequate to cover expansion for several years into the future. A second survey funded by the Bank a decade later reached the same conclusions. Deterioration of plant and equipment, how- ever, have been serious because of continual neglect of main- tenance. 1.11 Underlying causes include (a) financial problems, (b) inadequate planning and policy making procedures, (c) the general operating environ- ment in Tanzania which impinges upon the education system in critical ways, and perhaps also (d) the health status of primary students. 1.12 (a) Financial issues. As Table 3 indicates, public expenditures on education have been substantial and compare well with other countries in SSA through the sixties and seventies; since 1983, however, there appears to have been a decline in part because of general budgetary stringency and in part because of increased devolution of responsibilities to local com- munities, a devolution that has not been entirely successful. 1.13 But there have been problems with the allocation of these expendi- tures between various sectors and uses. Expenditures on adult and commun- ity education have at times been significant, cutting into what has been available for more tradition=I areas. Recurring expenditures per primary student have been lower than the average for SSA and, more serious, have been declining throughout the seventies. On the other hand, recurring ex- penditures per student at the secondary and tertiary level have been much higher and increasing. Much of these expenditures go for non-academic purposes. Boarding and other services to students at secondary and ter- tiary levels, for example, absorb 40-50 percent of these budgets, substan- tially higher than in most other SSA countries. In addition at university level, the proportion of non-teaching to teaching staff is high and stu- dent-teache- ratios are low compared to most other countries. While inter- nal efficiency is not worse than in most other SSA countries, expenditures on students who drop out before graduation or fail to pass exams for con- tinuation to the next level are substantial: in recent years, a third of primary students have dropped out before completion and 75 percent of high- er secondary students in mathematics and science have failed to pass final examinations. 1.14 Funding of these expenditures for the most part comes from public revenues which in aggregate have been declining over most of the 70s and early 80s. Foreign loans and grants to the education sector over the period 1981/2 to 1987/8 have averaged 70 percent of the development budget. Of this amount, SIDA has provided 30 percent, the Nordic countries as a whole 60 percent and the World Bank (IDA) 25 percent.5/ Over the period 5/ Nordic Countries, 1988, Annex II. I 1981-83 concessional assistance alone has provided 18 percent of the capi- tal and 12 percent of the recurrent budget. figures that are modestly lower than those in other Anglophone countries and substantially lower than those in Francophone countries.6/ All these figures are underestimates because they do not take adequate account of technical assistance. Since the early 70's cost recovery efforts have expanded, but they have not been very effective.7/ 1.15 Private expenditures on education are limited primarily to financ- ing private secondary schools. Enrolment in such schools accounts for 40 percent of total secondary enrolment and has been increasing. With a few exceptions, quality is considered to be poorer than in public schoels. 1.16 (b) Planning and Policy Making Procedures. Tanzania's leaders have had a clear-cut vision of where they wanted to go and what measures they wanted to use to get there. But feed-back from facts and empirical studies has played little role in the formulation and implementation of these policies. Indeed, in large measure the role has been limited to man- power studies and projections which, in retrospect, appear to have seriously underestimated the need for secondary and tertiary graduates. The most likely reasons for this underestimation was failure to take adequate account of the manpower implications of the dramatic institutional changes that took place in the decade after the Arusha Declaration plus over- optimism (which got built into manpower coefficients) about the ability of staff without adequate formal training to learn on the job. 1.17 But lack of adequate studies and information on which to base them is only part of the problem. Of more fundamental importance is the top- down decision making system with substantial distance between the very top (the Party) and the next links in the chain, in particular the Ministries of Education, Manpower and Local Government, certain offices in the Presi- dent's Office, and Regional and District Education Officers. The fact that responsibility for the education sector is divided between all these agen- cies, with no one fully responsible for overall planning and coordination within the sector greatly exacerbates the problem. 1.18 (c) The overall operating environment is one in which few alterna- tive sources of supply have been available or permitted. Parastatals construct school buildings, write and produce textbooks, and transport and distribute them. To "avoid waste and duplication" there is generally only one such firm in each functional or geographic area. Private schools, which account for 40 percent, and an increasing percent, of secondary en- rollment, have been completely regulated, in terms of fees, salaries, cur- ricula and enrollment. Without alternative sources of supply, there is no 6/ Education in SSA, Table A-25. 7/ UNESCO, 1988, para 19. way around delays and poor quality. The result is apathy at all levels: problems remain unsolved and deterioration proceeds apace. Somewhat more flexibility has been introduced during the last two years; but this situa- tion characterizes much of the period under review. 1.19 (d) Poor health and nutrition. An additional underlying problem may be lack of capacity, especially on the part of primary students, to efficiently absorb knowledge because of poor health and nutrition. At any point in time substantially more than half of all primary school children can be found to be undernourished, or suffering from diarrhea, measles, or malaria. As a consequence absenteism is high and ability to concentrate while in school is low. II. The Bank's Perceptions and Activities A. Analysis of the Education Sector 2.01 How much of this picture of the issues and problems of the educa- tion sector was known to Bank staff when the various education projects were being formulated? What has been the Bank's contribution to an under- standing of these issues and problems? Have Bank staff played a proactive role in the formulation of policies for the sector, or have they tended to follow, in reactive fashion, the lead taken by other agencies and by Gov- ernment? Answers to these questions can be distilled from the six general economic reviews of Tanzania undertaken by Bank staff, three sector reports undertaken at the behest of the Bank (two by UNESCO and one by the American Council on Education), and various other Bank documents (Country Policy Papers, Staff Appraisal Reports, Project Completion Reports, etc.). 2.02 The earliest review of Tanzania by the Bank was undertaken in 1958 and published in 1961.8/ It concluded that the economy was growing well and had good prospects for future growth "until easy development possibilities (like agricultural expansion using new land and traditional techniques] were exhausted." In its view, the primary constraint on develpment at that time was financial, not physical, administrative, managerial or trained manpower. The civil service was described as efficient and the administrative apparatus adequate to absorb and effectively utilize addi- tional resources. No mention was made of shortages of trained manpower, apparently because of the presence at that time of European and Asian na- tionals and expatriates. While it was pointed out that efficient operation of the administration required continued service by these individuals for many years to come, that was as far as the point was carried. There was no discussion of what might happen to absorptive capacity if a number of them left, no assessment of what it would have taken to induce them to stay and no contingency plan, or discussion of the need fCor one, in the event this assumption proved incorrect. 8/ IBRD, The Economic Development of Tanganyika, John Hopkins University Press, 1961. 2.03 As Europeans and Asians began to leave and the Government expanded its functions, Bank perception of the capacity of the civil service wat significantly altered. This shift is dramatically illustrated in the docu- mentation relating the the first education project which funded an expan- sion of the secondary education system. In 1963, the Staff Appraisal Re- port (SAR) noted that this project, which had already begun implementation before the Bank decided to assist with funding, was well-designed and ad- ministered. At one stage, doubts were raised because of the absence of a facilities utilization survey and an explanation about how this project fitted into an overall plan for the sector. But since it was obvious, just by looking at the education pyramid (see Figure 1). that an expansion of secondary schools was needed, and since all other elements for a well-im- plemented project were thought to be in place, the decision was made to proceed anyway. It was only one year later that a supervision mission reported that the executing agency for the project was filled with new, inexperienced staff, that they were having problems implementing the proj- ect and that they and the Ministry of Education were not capable of under- taking a facilities utilization survey or developing an overall plan for the sector. Their recommendation was to ask UNESCO to assist the Govern- ment with these undertakings. Some months later another back-to-office report raised doubts about the wisdom of proceeding with a second project until a utilization survey was available and it was clear Lhat personnel were adequate to effectively administer another project.9/ This report also passed on a request from the Ministry of Finance that IDA provide 85 percent of project costs because it was having problems obtaining local currency for these projects--a request that should have raised doubts about the ability of the Ministry to honor its pledge to meet recurring costs. 2.04 Thus, within a two year period most of the optimistic judgements about absorptive capacity made in the first country review were thrown into doubt. All subsequent reviews presented a picture that was more or less in line with the picture presented in the previous section of this report. In sector after sector and report after report, the capacity to identify, plan and implement projects and to manage and operate completed projects was described as being extremely poor. From the late sixties onward, the most critical manpower shortages were said to be at middle and higher levels, in science and technical subjects and fields relevant for agriculture. Unit costs at secondary and tertiary levels were noted as being high, but they were used only as a basis for estimating project costs; they were not seen to be a problem needing correction until the 80s. 9/ It is interesting to note that in April 1965, Treasury urged the Bank not to hold up processing of a second project because of concerns about lack of qualified personnel. Since a recruitment team in Europe had successfully signed up over 200 persons to work in Government, Treasury expressed confidence that the Ministry of Education would be fully staffed by mid-1966. 2.05 Many of the proximate causes oPf these pro1l,'es were also recog- nized from the mid 60s onvard. Certainly, the poos quality of teaching (though perhaps not the low morale and motivation of teachers), the short- age of science and technical teachers, and the limited supplies of instruc- tional materials were all well recognized. The inadequacy of buildings and equipment was also recognized. Overloaded curricula and schedules and lan- guage problems as causes of poor quality were importantly featured for the first time in the 1981 UNESCO sector review. 2.06 The way to quickly expand absorption capacity, it was believed, was through technical assistancc--in effect hiring other Europeans and Asians to take over from those who left. Thus, by this logic, financial resources continued to be considered the overarching constraint. Over the years it became clear that technical assistance was not a perfect substi- tute: the new generation of European and Asian technicians would no longer be in decision-making roles, recruitment of qualified persons proved diffi- cult, many found it difficult to adjust to Tanzanian conditions especially in the countryside, and costs were high. But until well into the eighties, the solution to most absorptive capacity problems was to increase the num- bers of technical assistants. 2.07 Of the underlying causes, financial problems were probably the most well recognized. The 1977 Basic Economic Report undertook a careful study of the financial implications of the crash program to achieve univer- sal primary education and concluded that the program was doable but only at the expense of quality and other programs; and public expenditure reviews of the 80s became increasingingly strident about recurrent cost problems. On the other hand, the weaknesses of self-help programs and the high unit costs of secondary and tertiary education was accepted well into the eight- ies more or less as facts of life, not something to be studied to search for alternatives or to be discussed and negotiated with the Government. 2.08 In the 60s and early 70s, manpower planning in Tanzania was con- sidered to ?e a model and its results tended to be accepted by Bank staff without second-guessing. In the late 70s doubts began to be raised and the 1977 Basic Economic Report undertook a review of methodology and findings. However, there is no evidence that any alternative projections were devel- oped by the Bank which might have served as counterweights to the Govern- ment's underestimates of the needs for secondary and tertiary graduates. Nor is there evidence that the Bank ever suggested to the Government that the results were implausible given the miniscule number of secondary gradu- ates in the work force, the extremely imbalanced enrollment pyramid, con- tinual complaints about skilled manpower shortages and the high cost and U large inputs of technical assistance. The staff also appear to have ac- - cepted without investigation other reasons given by the Government for restraining growth of post-primary education. 2.09 The following quotation from an Aide Memoire to the Government dated July 1977 is indicative. After praising the Government for success- fully "resisting growing pressures in regard to the social demand for in- creasing secondary school places," it notes that expansion of secondary does not appear to be a high- priority area for major additional investment. The most recent "Manpower Report to the President" indicates the current difficulties in absorbing secondary school leav- era in wage employment. In addition, a number of other agencies are providing assistance to both the expansion of secondary education capacity and the convertion of academic into technical secondary schools. Although when the secondary school [physical facility] survey is fin- ished the data may indicate the need for investment in the secondary education sub-sector to improve quality of instructions, now it is difficult to justify new proj- ects in support of secondary schools. 2.10 Given the findings of the 1966 UNESCO facilities survey, it was likely that this new survey would also indicate that expansion of physical facilities was unnecessary. That conclusion, by the logic of the preceding quote, provides support for not expanding secondary schools. B-it that is only because "new projects in support of secondary schools" meant construc- tion projects. Expansion of the system within existing facilities, as the UNESCO survey recommended, might still have been justified. 2.11 The limitations of the general operating environment in Tanzania were sometimes noted with regard to other sectors but not perceived as a problem for the education sector until quite recently. The likelihood that health problems are a serious impediment to learning in primary school has not, to the author's knowledge, been prominently mentioned to date, cer- tainly not in any operationally significant way. B. Bank Operations 2.12 The Bank has helped finance seven education projects in Tanzania, the total cost of which is US$ 112.40 millions. 10/ The Bank has provided IDA credits of US$ 67.10 millions and arranged cofinancing in the amount of US$ 23 millions. The appendix provides details on individual projects. This section focuses on several main themes suggested by a project-by-proj- ect review. 10/ Based on actual expenditures for the first five projects and appraisal estimates for the last two. 1. Concept and Design of Projectsll/ 2.33 Eclectic components. Judging from the content of these seven proj- ects and the documentation explaining and justifying them, the Bank ap- proach, at least through 1980, appears eclectic, lacking in a well-worked out strategy consistently applied over the period. All projects except the first included multiple components with little continuity from one project to the next. In the end nearly every type of education and training was supported to some degree in one project or another, but not necessarily consistently and adequately. 2.14 The first project, which went to the Board in 1963, simply and straight-forwardly assisted an ongoing Government program to expand the secondary school system, just at the heart of what most observers at the time believed to be the priority need in the education sector. The second project continued the expansion of secondary education and introduced "diversification" of secondary school curricula in 25 schools by providing additional facilities and equipment for the teaching of practical subjects. The project also provided assistance to expand and equip two primary and one secondary teacher training colleges. 2.15 The third project shifted assistance to the Ministry of Agricul- ture, providing funds for the construction of four new Rural Training Centers (RTCs), the conversion of five District Training Centers into RTCs and the construction of two and extensions to two Ministry of Agriculture Training Institutes (MATIs). While RTCs were to provide courses in agricul- ture, carpentry, metalwork, heath and nutrition as well as citizenship to farmers and their wives, MATIs were to train sub-professional personnel (certificate and diploma holders) for employment as field and assistant field officers. The project also provided 14 man-years of technical assis- tance and 7 man-years of fellowships for counterpart staff. 2.16 The fourth project initially included the design, construction, furnishing and equipping of three general secondary schools (dropped during implementation because of cost overruns), two pre-service and one in-ser- vice primary teacher training colleges, two vocational training centers, eight community education centers, one rural health center and a medical school for the University of Dar es Salaam. The community education centers were supposed to be used for both primary and adult education, but during implementation they were converted into Folk Development Colleges which concentrated on adult education and shifted emphasis from agricultural training to "community education". In addition, the project funded the total operational expenses of the Project Implementation Unit (PIU) and provided 2 man-years of technical assistance for a Management Training study, 35 man-years of university professional staff, 35 man-years of fel- lowships for Dar Faculty of Medicine and 12 man-years of technical assis- tance for the PIU. 11/ See Appendix 1 for a summary of relevant materials on each Bank education project in Tanzania. 2.17 The fifth project had three different components. The first was a program ta train about 1,500 cooperative inspectors and community develop- ment workers to become Village Management Technicians. It included the furnishing, and provision of vehicles for five RTCs, provision of about 2,000 pedal-cycles, materials for bookkeeping and local purchases, con- struction of about 1,500 modest houses, and provision of about 500 motor- cycles for supervisory staff at district headquarters. The second component added specialized classrooms and equipment to 15 secondary schools for the teaching of science and practical subjects. The third component included the provision of technical assistance personnel for the preparation of a survey of training facilities for accountants and auditors, a survey of educational facilities in all secondary schools and a primary education sub-sector review. Three different agencies were charged with pr;jject im- plementation: the Project Executive Group of the Prime Minister Office, the Project Implementation Section (previously called Project Implementation Unit) in the Ministry of Education, and the National Bureau of Accountants and Auditors (NBAA). 2.18 With the sixth project, the Bank shifted from previous emphasis on the rural sector to the provision of assistance for the development of critical middle-level manpower and skill training for the modern sector, and shifted from emphasis on secondary school diversification to the provi- sion of opportunities to women in secondary schools. The project included the provision of 160 hostel places for women in each of four secondary technical schools, improvements to two Vocational Training Centers the construction of one Instructional Training Center, expansions to Dar School of Accountancy, renovations and equipping to a Seminary building for the creation of a Management Training Program for education personnel, the provision of funds for a study of the Institute of Education, and funds for MOE planning section to support project monitoring and evaluation. 2.19 With the seventh project, the Bank, for the first time, provided assistance to upper secondary education, and assisted directly in the pro- vision of educational materials and qualified teachers instead uf calling on the Government to assure a supply of these items. The project included: i) primary education: construction, furnishing and equipping of about 360 primary school classrooms and the provision of primary teaching materials to about 16 rural districts; ii) secondary education: construction, fur- nishing and equipping of one demonstration school for the University Department of Education; the construction equipping and furnishing of about 26 new form V-VI science/math streias in 10 existing and one new secondary school; reequipping, refurnishing and rehabilitating of teaching facilities in 34 existing science/math streams in nine existing schools; the construc- tion, furnishing and equipping of extensions to one lower secondary day school; and provision of 100 man-years of graduate teachers of mathematics, physics and chemistry; iii) primary teacher training: provision of equip- ment for science laboratories in four primary teacher training colleges and Dar Technical College, construction, furnishing and equipping of a new primary teacher training college and renovating and refurnishing an exist- ing college; iv) quality support services: provision of storage facilities, vehicles and fellowships to Tanzania Elimu Supplies (TES), provision of buildings, equipment, technical assistance and fellowships to the National Examination Council (NEC), provision of fellowships and equipment for the Coordinating Unit for Research and Evaluation (CURE) and provision of con- sultants, inctemental operating costs and vehicles for evaluating ongoing and completed projects and prepare future projects. 2.20 A variety of factors explain this eclectic picture: requests from the Government, sector studies undertaken by different agencies (UNESCO, American Council on Education, Bank staff) at different times, changes in ideas about what was needed on the part of Bank staff, abundance of donor funds for much of the period which resulted in scrambling for bits and pieces to support in different programs and lack of aid coordination. Underlying these factors may have been a lack of strong conviction on the part of Bank staff about what was needed or an unwillingness to argue forcefully for these convictions; while not demonstrable, these more fun- damental explanations were at least consistent with the facts. 2.21 Focus on Secondary and Diversified Secondary Education. If there was any focus it was on secondary and diversified secondary e tion. Thirty eight percent of expenditures in these seven projects wer, 't on these components, less on expansion per se than on adding spec I fa- cilities (e.g., laboratories, workshops and boarding facili._s) and courses in practical subjects (agriculture, home economics, commerce and wood and metal working). 2.22 The focus on diversification appears to have started with the second project, signed in 1969 after four years of negotiations. At the outset, the Government wanted a continuation of support for the secondary school construction program, started in the first project. But the 1966 UNESCO facilities survey led the Bank to suggest a smaller expansion pro- gram and use of the funds saved to develop special facilities for agricul- ture and other practical subjects. Within a short period, this notion was transformed into an educational philosopy that argued for the introduction of practical course so that secondary students would be better prepared to enter the labor force upon graduation. While accepting the idea that the curricula shoald be diversified--though probably for different reasons--the Government cont!-ued pressing for expansion of the system and resisted serious consideration of the UNESCO recommendations. The result appears to have been a compromise in which some expansion and some diversification were agreed to. The Bank provided funds for construction and equipment and other donors provided technical adsistance to hire specialists to help with the curricula changes and to provide overseas fellowships to train Tan- zanians in the practical subjects. 2.23 The third and fourth projects included no significant secondary components (the fourth because they were dropped during implementation because of cost overruns). The fifth project seturned to the diversifica- tion theme. It basically included the provision of hardware for the teach- ing of science and technical subjects at fifteen schools.12/ The component was not successful mainly because insufficient attention was given to fac- tors such as software and teacher requirements, utilization rates in rela- tion to school size and student options. Even though the appraisal mission findings .showed an adequate number of teachers for the teaching of practi- cal subjects, actual shortages prevented the appropriate utilization of workshops and equipment. Furthermore, students prefered other options such as chemistry and physics since examinations favored academic subjects. By 1979 when the sixth project was appraised, considerable scepticism about the wisdom of this approach had developed and, despite the Government's desire to continue, the Bank refused to finance any more expansion in that direction. 2.24 The concept of diversified secondary education--to better prepare secondary graduates for the labor market--appears to have been imported by the Bank and other donors from Europe where it made sense in the sixties and seventies because of high secondary enrolment rates and low progression rates to colleges and universities. But it is strange in the Tanzanian context where secondary students were, and still are, an elite who either continue their education after graduation or enter the labor force at pro- fessional or pre-professional levels where these practical subjects are of little value. In Tanzania, it is the large majority of primary students who do not proceed with further education and could use special courses to prepare them for labor force entry. Indeed, after the 1967 ERS document, it became Government policy to provide such courses. These facts, plus the high cost of adding special classrooms, equipment and teachers to secondary schools, should have been evident at the time. But it took many years of accumulated evidence plus a major studyLS/ (which, among other things vast majority of their countrymen earn a living. But for this purpose, specialized classrooms and expensive equipment would hardly be necessary. Later the Government came to accept the notion as an adjunct to vocational training. All along, however, the Government wanted to proceed more slowly because of concern that specialized teachers and teaching uiaterials would not be available and quality of training would suffer--which is, of course, just what happened after it succumbed to Bank pressure to move faster. 121/ Bilateral agencies were also asisting diversified secondary education, the main donor being NORAD which was constructing extensions to 48 schools for the teaching of agriculture, commerce and home economics. 13/ Psacharopoulos and Loxley, Diversified Secondary Education and Development, Job-s Hopkins University Press, 1985. 2.25 Other components and activities are discussed below under accom- plishments. 2. Instruments to Accomplish These Goals: Civil Works and Covenants 2.26 As with Bank education projects elsewhere in the world during the 60s and 70s, a very large percentage was spent on buildings and equipment. In Tanzania 87 percent of the Bank's funds was spent on hardware, 64 per- cent on civil works alone. The implicit assumption behind this focus was that the Government and other donors would provide other inputs, and to some extent that was a correct assumption--other donors, for example, were far more interested in providing technical assistance. The result was a neglect of "software" components not only so far as spending was concerned but in the design of projects and in the Ppecification of covenants. 2.27 Covenants covered such topics as international competitive bidding for larger contracts, adequate provision for recurring costs and mainte- nance, establishment and staffing of project implementation units, techni- cal assistance to prepare new course materials and fellowships to train teachers in specialized subjects, and a few studies, e.g., of facilities utilization. Covenants did not cover a number of important proximate or underlying causes of the problems with the education sector, as one might expect. For example, despite the fact that the system was continually losing teachers, often upon graduation, to other occupations, the Bank accepted assurances that adequate teachers would be available and never called for a study of salaries and other conditions of employment, let alone any change in conditions that might have improved the situation.141 2.28 Typically, covenants were stated in general terms, often just as goals to be accomplished, without specifying terms of reference for techni- cal assistance and studies and without specifying the means by which the goal was to be accomplished. There is no indication in the documentation that the feasibility of achieving these goals was explored. There is no indication that these conditions were required to be met prior to signing the loan. 2.29 For example, a number of projects committed the borrower to pro- viding the necessary qualified teachers to staff the new facilities. De- spite growing evidence in early projects that the Government was unable to do so, future projects went forard with the same general kind of covenant. The same is true for mainterance. An effective maintenance program requires, at a minimum, a schedule of when routine operations will occur, the assignment of responsibilities and the earmarking of funds for doing so. There is no evidence that the Bank ever required the Government to undertake these actions before or even after signing, and no evidence that after the experience of early przjects it tried any different approaches. 14/ For a summary list of covenants and outcomes see Appendix 2. 3. Implementation and Supervision 2.30 As Appendix 1 indicates, implementation was difficult except for the first project (which was fully developed and under way by the time the Bank got involved). The principal explanations have to do with project design, shortcomings of the Project Implementation Units (PIUs) and a vari- ety of exogenous factors. 2.31 Project design was often not well advanced at the time of credit approval, making it difficult for the implementation unit to coordinate activities in a timely fashion. The Bank relied on the appointment of tech- nical assistance personnel for the preparation of components during imple- mentation. The Government did not fully use this assistance and often made appointments very late during implementation which created probloms and delays, for example in procuring equipment. 2.32 The PIU that has implemented most projects was established in the Ministry of National Education (MOE) in January 1974, five years after credit effectiveness of the second project and six months after credit effectiveness of the fourth project. Lack of a full time project coordina- tor, a procurement officer and an accountant delayed the implementation of the second project and created serious procurement problems. Similarly, implementation of the fourth project was delayed by the late arrival of the technical assistance team provided by DANIDA. The team was to complement the activities previously undertaken by the PIU (issue tenders, evaluate bids and award contracts), by providing architectural services for all project institutions. Even though the PIU (later called Project Planning Section (PPS) was eventually well-staffed and directed, it was made re- sponsible for the administration of projects sponsored not only by the Bank, but also by DANIDA, the United Nations Commission for Refugees and SIDA. The simultaneous implementation of these projects weakened the PIUs capacity to implement IDA-assisted projects. Other complications included poor preparation of components, geographical dispersion (especially in the fourth project), the establishment of a special PUI in the Ministry of Agriculture in the third project, and the involvement of three project units for the implementation of the fifth project. 2.33 The Third Project was administered by a different PIU established in the Ministry of Agriculture, Food and Cooperatives (MOA). This Unit suffered from lack of support from MOA, shortages of qualified staff and from poor communication with other Ministries involved in the implementat- ion of the RTCs. The Unit was provided with a full-time project coordinator eight months after effectiveness and with a full-time architect by the time of the original closing date. Furthermore, the Government did not hire the technical assistance personnel required for the establishment of training and equipment needs for the project 2.34 A variety of problems outside the control of the PIU also plagued these projects: policy changes during implementation were often proved to be disruptive, devaluation that made contractors unwilling or unable to fulfill obligations without substantial price adjustments that were slow in - 154 -U coming; poor transport and communication services, inexperienced domestic contractors and shortages of financial and physical resources. The latter, a product of macroeconomic problems and policy and regulatory measures used to cope with these problems, were especially serious from the late seven- ties onward. It is largely for this reason that the sixth project was completed in December, 1988 instead of four years earlier as originally planned, and that the completion date of the seventh project has to be extended by a year. 2.35 Given these problems, greater than the average intensity of super- vision for education projects should have been provided. Instead, sometimes the opposite occurred. Supervision missions were sometimes too infrequent. At a crucial point in the life of the second Droject, for example, there was a thirteen month interval between supervision missions. Frequently, they tried to do too much (often related to other projects) in too little time. Frequently also they were inadequately staffed. For example, only 2 of the 19 missions for the second project included a general educator and only one for the third project included an agricultural educator. The result was that procurement problems and physical rather than education aspects received nearly all the attention, a consequence that sent an un- fortunate signal to the PIU about which components were most important. 4. Project Outcomesl5l 2.36 Primary Education. Apart from teacher training (see below) the Bank assisted primary education only in the fourth and seventh projects. The fourth project helped construct and equip four out of a planned eight Community Education Centers (CECs) each of which included a primary school along with facilities for adult education. These schools were successfully completed and, as of 1981, were in full operation, although over-enrolled, which created serious shortages of equipment and furniture. 2.37 The seventh project is rehabilitating 360 classrooms and dis- tributing teaching materials and textbooks to 16 rural districts. To date, only 60% of civil works under the seventh project has been completed, and contracts for construction of eleven schools need revision. According to supervision reports, the project unit has provided equipment and books unsuitable for rural schools. The equipment is too sophisticated (schools need paper and pencils) and many textbooks are in English rather than Swahili, the medium of instruction in primary schools. Despite objections by principals and teachers, the project unit has continued the procurement of imported kits and books. 15/ For a summary of projects results compared to expectations see Appendix 3. 2.38 Secondary Education. Assistance for the expansion of secondary education was fairly successful. Two new boarding schools and two new day schools were constructed, and fifty three existing schools were extended. The first project provided 6,845 additional student places (of which 95% were for boarding students), the second project added 3,242 places (from an expected 4,672), and the fifth project expanded the capacity of 13 schools from 6,064 student places to 6,422. 2.39 Diversification, assisted under the second and fifth projects, resulted in additions of facilities and equipment for the teaching of prac- tical subjects in thirty seven schools. In addition, seventy two staff houses at seventeen schools were improved. Utilization of these facili- ties has been unsatisfactory because of shortages of qualified teachers for these subjects. While the Bank refused to help finance additional diversi- fication after the fifth project, practical subjects and agriculture con- tinue to be important parts of the curricula and examination materials in all public secondary, and to some extent as well in primary, schools. 2.40 With the seventh project the Bank is addressing the present high level manpower bottleneck by increasing the production and quality of math and science students from upper secondary schools. Twenty six new Form V and VI science/math streams in ten existing schools are being constructed, equipped and furnished; thirty four existing science/math streams in nine schools are being reequipped and rehabilitated; and 100 man-years of expa- triate volunteer graduate teachers of math, chemistry and physics are being provided. In addition, one lower secondary day school is being constructed, furnished and equipped. 2.41 Teacher Training. Fourteen percent of project expenditures have gone to teacher training, 13 percent to primary and one percent to secon- dary. The Bank constructed, equipped and furnished two new Colleges of National Education for primary teachers under the fourth project, extended and equipped one primary teacher training college under the second project, and one in-service primary teacher training college under the fourth proj- ect. The seventh project is rehabilitating and extending Nkrumah Primary Teacher Training College, and is equipping science labs at four teacher training colleges and at Dar Technical College. While the construction part of these components was successful, actual teacher training was often inef- fective due to lack of instructional staff and equipment. 2.42 A severe shortage of qualified secondary school teachers, particu- larly in science and mathematics, has been a problem since independence. In the early years it was exacerbated by the outflow of expatriate teachers and in later years by the increased demand for practical subject teachers. For the most part, the Bank relied on assurances from the Government that teachers would be supplied to the schools its funds built. fsssistance was provided only under the second project for extensions to Dar College of National Education and Dar Technical College. While the component was completed and an adequate number of teachers was trained, the Government was unable to retain them in the education system. The fact that this shortage was not anticipated had serious repercussions on the ability of the fifth project to achieve its goals. 2.43 Agricultural and Rural Education. This focus appears especially in the third project which concentrated on the improvement of facilities for agricultural education and training. In particular, it provided funds to expand the existing two Ministry of Agriculture Training Institutes and to build two more. These institutes were designed to meet the increased demand for trained manpower at the sub-professional (diploma and certifi- cate) levels to serve as field and assistant field officers. At the time of audit 82 percent of planned enrollment and 61 percent of planned output had been achieved. 2.44 Other efforts in this field were generally unsuccessful, in large part because of Government policy changes. Under the third project, the Bank constructed four new Rural Training Centers (RTCs) and converted five District or Farmer Training Centers into RTCs. These centers were meant ta provide farmers and their wives with short training courses in village centers. During implementation, these were converted into Folk Development Colleges that provided less technical and more general subjects. The fourth project assisted in the construction and equiping of four (out of the eight planned) Community Education Centers (CECs). These centers, which were established on an experimental basis in this project, included a primary school with additional facilities for instructing resettled adults. While the primary education component was successful, it was only partially additive since it replaced more scattered schools; and the skills provided to adults proved to be inadequate for the villages conditions. In addition, the costs involved in these centers prohibited their generalization. 2.45 A principal objective of the fifth project was to expedite rural development through the training and provision of logistical support and housing to Village Management Technicians. These efforts were lost in large part because the government changed its priorities regarding the "villagization" program, but also because available technical packages were not always appropriate. While the technicians were trained, they were rejected by villagers, and the centers were transformed into cooperative training colleges and vocational centers. 2.46 In retrospect, the Bank's involvement in most of these rural edu- cation schemes was ill-conceived. The principal shortages were in occupa- tions requiring secondary and tertiary education; the technical packages were not always present; and the 3overnment and the Bank did not always agree on the type of training to be provided, the Government often wanting to use the facilities to provide general and politically oriented educa- tion. In any case, little training in agriculture resulted from the build- ing of these facilities; and nothing remains of these projects today. I 2.47 Vocational/Technical Education. The second project provided assis- tance for extensions to Dar Technical College and provided 182 teaching and commlinal student places and 278 boarding places. The fourth project constructed, furnished and equipped two Vocational Training Centers which were in operation for basic training (post-primary training). The centers assumed an important place within the national vocational training scheme because there was an urgent demand for skilled workers which made it imper- ative that the facilities were used immediately, even if the equipment was not fully adequate. Trainees and instructors showed such dedication that initial shortcomings were easily accepted ar )vercame. 2.48 The sixth credit provided additional equipment and accommodations for staff and students to the two Vocational Training Centers built under the second project and constructed, equipped and furnished one new Instruc- tor Training Center at Morogoro as part of the National Vocational Training Program. The total enrollment is expected to amount to 160 trainees com- prised of four pre-service groups of 30 each in eight-month courses, and two in-service groups of 20 each. The project also included boarding facil- ities for students and staff in order to avoid handicaps already experi- enced by other Centers. 2.49 The implementation of these components has been successful in terms of people trained and trainees have o'tained jobs shortly after training. However, insufficient or inappropriatf equipment made the teach- ing of mechanical engineering not possible e,ring implementation of the fourth project. 2.50 Post-secondary education was assisted in the sixth project which constructed, equipped and furnished a new Instructor Training Center (ITC) at Morogoro as part of the National Vocational Program. The institution has a vital role within the system since it will satisfy the demand for quali- fied staff for the promotion and supervision of in-plant training programs, the training of industrial trainers, the trade traiaing programme and the staffing requirements of Tanzania African Parents Association (TAPA) and Post-primary Technical Centers (PPTCs), as well as mission/church schools and private trade schools. 2.51 University Education. The Bank has provided assistance Lo higher education only through the fourth project for extensions to the University of Dar es Salaam Faculty of Medicine and the Bagamoyo Rural Health Center. All the facilities provided were geared to medical teaching and consisted of laboratories, workshops, seminar rooms, boarding facilities for students and staff housing for a small proportion of the staff. 2.52 The project goal to increase the annual intake of medical students from 30 to 50 students was achieved. Furthermore, the quality of graduates remained high, the faculty increased in numbers and was predominantly Tan- zanian. Bank-sponsored construction of faculty and student living quarters and of classrooms contributed to the increased number of students and fac- ulty who could be accommodated. 2.53 The Rural Health Center at Bagamoyo proved to be a success. Train- ing at this site became an integral part of the medical student curriculum, and was enthusiastically received by the students. The center provided students from the University of Dar with necessary training in community health problems. 2.54 Accounting and Management Training. The sixth project included extensions to Dar-es-Salaam School of Accountancy. The component responded to needs in priority areas indicated by the April 1975 Management Training Study financed with Bank assistance under the fourth credit. This project also provided assistance to the Ministry of Education for the creation of a Management Training Program for Education Personnel (MANTEP). The program was a response to the Ministry's need for managerial improvement and strengthening and to the additional training program required in order to respond effectively to the rapid expansion of the education system. 2.55 Special Studies. The Bank provided assistance for four special studies. Two studies were geared towards institutional development and included the evaluation of management and accountancy training. They have been considered two of the Bank's most successful studies in institutional development in the region. ' "e other two studies entailed a survey of secondary school facilities at.. a primary education sector review. (i) Management Training Study. Since Independence, there was a rapid growth of management training programs, but the creation of insti- tutions and programs were not adequately coordinated, leading to duplication and over-lapping. The Bank included 2 man-years of technical assistance in the fourth project to survey all on-going programs, evaluate their effectiveness and prepare a master plan for management training. The results and recommendations of the report were welcome by the Government and led to the establishment of The National Management Development Advisory Council (NMDAC), within the Ministry of Man- power Development and the Standing Committee on Awards (SCOA) within NMDAC. The purpose of NMDAC was to review the planning, development, operations, education and specialized needs of man- agement training; the terms of reference of SCOA included respon- sibility to standardize awards and entry qualification for each course and to ensure the quality of syllabi and teaching. (ii) Accounting and Auditing. The fifth project included eight man- years of technical assistance to assist the National Board of Accountants and Auditors (NBAA) in the areas of curriculum devel- opment and standardization, instructional materials, and the assessment of existing facilities and staff. The survey of facilities for accountancy training, carried out by the Ministry of Finance, partially facilitated the inclusion of the rehabilitation of the Dar-es-Salaam School of Accountancy - 239 - (DSA) in the Sixth Project and the construction, by the Govern- ment, of a School of Accountancy at Arusha to provide National Book-keeping Certificate (NABOCE) level graduates for the parasta- tals and commercial sector. The specialists provided under the project examined the existing facilities at DSA and assisted in preparing technical papers, study materials, in conducting courses, seminars and workshops, and in reviewing examinations structures. In addition, other accomplishments can be attributed to this study: i) the establishment in 1974 of a national accountancy training and examination scheme; ii) the initiation of new courses to train primary school leavers to become account clerks; iii) the approval of the Tanzania Statements of Standard Accounting Practice in 1983; iv) NBAA has continuously held work- shops and seminars on professional matters; v) the Board has en- couraged the Ministry of National Education to introduce commer- cial classes in general secondary schools. (iii) The Survey of Educational Facilities. The fifth project included the provision of four man-years of technical assistance to cover the services of an architect and a general educator for a compre- hensive survey of Tanzania's general secondary schools. The survey required six man-years of technical assistance and covered only seven private schools, nine primary teacher training colleges, and only 56 of the 83 existing Government secondary schools. Neverthe- less, the resulting two volume document 16/ contains important information. Among the findings, the following are worth mention- ing: (a) Government schools were underutilized and about 21,000 addi- tional students could be accommodated with some physical adjustments, additional teaching staff and books, and the use of the rotation system; (b) all Government schools had major maintenance problems and most have electrical, water and sanitary deficiencies; (c) of the 56 Government schools surveyed, 53 were found to have inadequate hardware and software for the teaching of science; (d) building costs of Government schools were far too high. (iv) Primary Education Subsector Review. The review was undertaken by Tanzanian officials with some assistance from a Bank Group consul- tant. The results of the review were to be based on a survey of 16/ Survey of Educational Facilities in Secondary Schools and Teacher Training Colleges in Tanzania, MNE, Dar-es-Slaam, 1981. 300 primary school throughout the country. The actual survey con- sisted on only 170 schools or about 1.7Z of all primary schools in the country. While the survey suffered from some initial problems 17/ and the data drawn from such a small sample may not be represen- tative of primary schools throughout the country, the report con- firmed what was generally known about the primary system. On the one hand, that net enrollment ratio had improved due to the UPE campaign, that the country had satisfactory attendance rates, and the quantitative success of supplying primary education teachers during the UPE campaign. On the other hand, the report cites over- crowded classrooms, incomplete buildings, unsanitary conditions, shortage of school furniture and instructional materials, low educational level of a high percentage of the teachers, the need for more staff housing, and the inadequate coverage of the cur- riculum in all grades. III. Lessons and Future Directions 3.01 The Bank has made substantial and impressive contributions to the expansion of Tanzania's education system. But the assistance provided was far less effective than it could have been. Projects were piecemeal and scattered; they were seldom based on prior investigations of the approach being supported; project preparation was often incomplete at the time of approval, reliance being placed on technical assistance (that was not al- ways wanted and used) to correct this shortcoming; projects frequently failed to specify policy relevant covenants which were then enforced; su- pervision often neglected policy relevant components; later projects often failed to take into account results of earlier projects; and dialogue on meaningful education policy issues was limited, particularly when those issues were politically sensitive. 3.02 Project completion reports, audits and sector reviews have noted most of these shortcomings and have made recommendations to correct them. Among other things, they have recommended that: - appraisals exclude components that are not adequately prepared; - more care should be taken in selecting outside agencies or consultants used in sector work and project identification; - projects should be simpler in design; 17/ Weaknesses in the questionnaire including the lack of means to identify within school variations which might influence educational outcomes and the reliance in one person, the headteacher, to complete the questionnaire. - supervision missions should be better staffed, of longer duration and not be sidetracked into identifying follow-on projects, and devote more time to 'software" components and policy issues; - covenants should be specific and monitorable and be included only when the Bank is prepared to enforce them; - monitoring and evaluation should be built into all projects, especially into those components identified as being experi- mental; - more careful investigation should be undertaken to determine whether recurrent costs can and will be met and whether physical inputs--teachers, textbooks, etc. will be available before projects are approved; - put more effort into coordination with other donors, particu- larly in circumstances where technical assistance for a proj- ect is provided by such agencies. 3.03 Many of these recommendations have been made over and over again, with little effect at least until recently. At one level, this lack of progress can be explained by failure on the part of the Bank to ensure that careful thought and study preceed decisions to lend. The Bank failed to question Government manpower studies that underestimated needs for secon- dary and tertiary graduates; appears not to have questioned the innovations the Government wished to introduce into rural education, nearly all of which failed; allowed equipment lists specified by the American Council on Education to be set to inappropriate North American standards; and more generally and seriously, failed to develop--or ensure that others devel- oped--an overall, agreed-to strategy to guide lending operations. Admit- tedly, development of such a strategy was made more difficult in Tanzania by the fact that the Bank at times relied on other agencies--UNESCO, FAO and the American Council on Education, each with a different approach--for sector reviews and project identification. This approach has considerable merit. It fails, however, when the Bank uses it as a substitute for assum- ing overall intellectual leadership, at a minimum for second-guessing and asking hard questions about the judgements and proposals of these disparate agencies. 3.04 Part of the explanation lays in the difficult political and eco- nomic problems faced by Tanzania throughout the period under review, prob- lems that lay outside the education sector but which made it difficult to obtain pertinent decisions and to implement projects within the sector. It remains the case, however, that the design of projects and decisions about going ahead with projects failed to take adequate account of these con- straints. A contributing factor to this latter failure may have been the practice of assigning responsibility for supervision to the Nairobi region- al office with the result that headquarters staff responsible for appraisal had inadequate knowledge of field conditions and implementation problems, a practice that continued throughout the seventies and early eighties. 3.05 Underlying these explanations may have been the general over- eagerness on everyone's part to lend to Tanzania, particularly for educa- tion, during the sixties and seventies. It was conventional wisdom during much of this period that Tanzania had proper goals and policy directions, that what was needed was to work out details and provide adequate resources to implement. Indeed, Tanzania was held up as a model for other countries in the region. As a consequence, there has been a reluctance to implement recommendations of the type listed above, since they undoubtedly would have slowed down lending operations. 3.06 Since 1986. several events have occurred that suggest that the future may differ from the pas", in significant ways. First, one of the major causes of implementation delays particularly in the eighties is being removed by the Structural Adjustment Program initiated in 1986 with help from the IMF and the World Bank. This program has substantially increased the flow of financial and physical resources to projects so that the pace of implementation, which slowed nearly to a halt in the mid-eighties, can pick up.18/ Second, new management teams with an interest in rethinking the education program have taken over in both the Government and the Bank.19/ Third, several new studies of the education system which provide a solid basis for that rethinking process have recently been completed20/ and are under discussion with the Government and the donor community. And fourth, the Government has been forced by pressures from parents with children of secondary school age to reconsider its policy of severely restricting secondary school places. 181 During the 60's and 70's a new Bank-funded education project was started in Tanzania every 2 1/3 years. The seventh and last project was started in 1980 and another one is unlikely to be signed before the end of 1989. The sixth project has just been completed, four years later than pl.nned, and the seventh project has been extended by an additional year. An eighth project was prepared for Board presentation in 1983 but was withdrawn because of doubts about ability to implement given macroeconomic and budgetary constraints at the time. 19/ A new Permanent Secretary for education assumed office in the spring and the Bank reorganization took effect July 1, 1987. 20/ Most notably, a new comprehensive Sector Review undertaken for the Bank by UNESCO plus a detailed study of the textbook situatior became available in 1988. The sector review vias a major source for much of the information contained in Part I of this chapter. 3.07 These events have led to a number ef promising developments. Since 1987, the pace of implementation of the sixth and seventh projects has substantially increased.21/ Supervision has become more frequent and better staffed to cover software and policy issues.22/ More effort is being put into aid coordination.231/ Most important, the next Bank lending operation, scheduled for FY90 and currently being prepared, will focus on (i) strengthening the institutional, analytical and managerial capacity of the education system, (ii) improving educational quality by developing textbooks, new science curricula and greater access by teachers to classroom materials and equipment and (iii) providing the foreign exchange necessary to finalize construction of a large number of privately financed and built secondary schools. All these initiatives focus on areas that have been identified as critical by past reports and reviews. It remains to be seen whether they will be sufficient to overcome the systemic weaknesses that have plagued past reform programs. 211/ As of June, 1987, the two projects appeared likely to end with $23 million unspent and numerous activities unfinished. By November, 1988, evidence indicated that funds would be fully disbursed and all components successfully completed by the end of 1989. 22/ In FY88, at least 15 staffweeks were devoted to stnervision of each of the two outstanding projects, compared to an averige of 10 staffweeks per project in past years. 23/ Preparations are under way for a conference of donor, Bank and Government officials to develop a coordinated approach to the education sector and discussions are under way to develop collaborative programs with SIDA and other donors. TABLE 1 Education Indicators in Tanzania and Sub-Saharan Africa 1989-1988 Te"al Females 1968 1970 1980 1988 1986 1960 1988 Total Enroliment - Tanzania ('00) 478 903 3,451 3,641 3,268 163 1866 Gross Primary Enrollment Ratio Tanzania 24 34 93 87 81 16 84 Sub-Saharan Africa B6 48 76 76 24 83 Gross Secondary Enrollment Ratio Tanzania 2 3 3 3 1 2 Sub-Saharan Africa 8 7 16 26 1 18 Gross Tertiary Enrollment Ratio Tanzania . 0.2 0.3 0.4 . 0.1 Sub-Saharan Africa 0.2 0.6 1.2 1.4 0.1 0.4 Progressive Rate from Primary to Secondary Tanzania 16 . 8 14 7 Sub-Saharan Africa 39 . 43 88 42 Average Years of Education of Working Age Population Tanzania 0.8/ 0.7 1.6 . . Sub-Saharan Africa 1.2 a/ 1.8 2.8 3.2 . . a/ Figures for 1986. Saouc: for 1960-83, World Bank Education In Sub-Saharan Africa Tables A-8 - A-9, A-12, C-8. For 1988. Ministry of Education, Baic EdUgation Statistics. June 1987. .:rei Educanonai Pryamsd. Tanzansa. 1961. 1971. and 1980 MAL£ F~EML Unverserv irst veari VI Secondarv Form v IV Ill VII Primary Standard - v Iv IV 400 350 300 250 200 150 100 50 0 50 100 150 200 250 300 350 1980 1971 1961 a ~u~ Table 2 En;'ofmant Data for Public Schools by Grade and Sw, Tasmasa, 1961. 1971, and 1980 Mak. 1961 lm97 1980 ToWa Grad 1961 1971 1980 Nuber Per~ Nuwbr Pe~t Number Per 1980 Uivr~y (lagyear) 70 624 693 6 8 75 11 216 24 909 ,ky edtooform VI 251 1,240 1,230 25 9 196 14 407 25 1,645 V 211 1,354 1.340 25 11 254 16 413 25 1,761 IV 1,121 5,274 6.550 482 30 1.170 25 2,770 25 9,320 il 1,540 5,373 5,908 548 26 1,947 -27 2,737 32 8,645 11 2,514 5,426 5.827 1,019 29 2,197 29 2,785 32 8,612 S2,967 5,554 5,887 4.229 29 2,016 27 2,960 33 8,847 Primary school standad vi 11,322 46,563 123,749 - - 23,939 34 86,954 41 210,703 Vi 13.061 54,011 210,534 - - 31,256 37 167,445 43 377,973 V 14,737 63,300 246,579 - - 39,308 38 209,329 46 45,908 IV 65.152 82,396 269,474 30.239 32 55,850 40 237,363 47 506.837 65,53 89,616 413,125 34,787 35 59,689 40 383,375 48 706,500 fl 67,647 98.954 268,827 41,345 38 67,636 41 257,609 49 526,436 72,773 111,018 246,827 48.613 40 79,873 42 240,038 49 486,865 Source: Psacharopoulos and Loxley, Piversifled Secondary Education and Development Baltimore, J. Hopkins 1. Press 1985, p. 142-3. TABLE 3: Public Expenditure on Education In Tanzania, Sub-Saharan Africa and Other Developing Countries 19.6 125 18 1288 i1 Total Education in 1983 USS (millions) Tanzania 138.6 216.4 281.9 258.2 Total Education as N of Govt. Expenditure Tanzania 16.9 17.3 14.8 16.8 Sub-Saharan Africa 16.7 16.6 18.2 11.9 Other Low-income Countries 16.4 / 5.9 / 9. Other Middle-income Countries 18.9 a/ 14.2 h/ 14.6 Total Education as X of ONP Tanzania 4.6 5.4 6.9 6.6 1.7 Sub-Saharan Africa 4.1 4.9 5.5 4:3 4.2 Other Low-income Countries 8.4 A/ 2.0 b/ 2.0 Other Middle-income Countries 3.9 l/ 8.6 W 8.* Public Recurrent Expend. per Primary Pupil Constant '93 US$ Tanzania 6 46 88 86 Sub-Saharan Africa 87 B6 Be 62 Percent 4NP per capita Tanzania 24 16 12 12 Sub-Saharan Africa a/ 16 1o 26 is Public Recurrent Expend. per Secondary Student Constant '08 US$ Tanzania 678 6 89 628 Sub-Saharan Africa 808 278 825 287 Percent of CNP per capita Tanzanta 249 246 15 348 Sub-Saharan Africa s/ 111 93 62 62 Public Recurrent Expend. per Tertiary Student Constant '83 USI ' Tanzania 8891 6967 12857 8865 Sub-Saharan Africa 4880 6481 569 5866 Utritiple of CNP per capita Tanzania is 29 47 86 Sub-Saharan Africa s/ 11 12 7 8 a/ 1972. h/ 1981. S/ Median, sil other figures are weighted averages. d/ Figures are probably incorrect. Sourac: For 1970-83, World Bank, Education In Sub-Saharan Africa, Tables A-14, A-T, A-18, A-19. Figures for 1986, World DevelomMent Report, 1988, Table 283 figures for all lower-Income and lower middle-income countries from world Bank World Davelopment Report, 1984, Table 28. Immiluimim I Appendix 1 World Bank Education Projects in Tanzania Summary EDUCATION I. Credit Number: 45 TA Total Project Cost (US$ million): 6.00 Agreement: December 19, 1963. Credit Amount (US$ million): 4.62 Closing Date: December 31, 1967. Credit fully disbursed. Completed on schedule. PCR: - PPAR: - 1. Objectives: expand and improve secondary education. 2. Executing Agency: existing services. 3. Implementation experience: successfully completed. supervision: minimal. Four man-weeks, four one-week missions. 4. Outcomes: enrollment target met. 2.5% of secondary school age population enrolled in secondary school system in 1968. EDUCATION II. Credit Number: 149 TA Total Project Cost (US$ million): 7.20 Agreement: Nay 29, 1969. Underrun (Z): 0 Original Closing Date: 12/31/73. Credit Amount (US$ million): 5.00 Actual Closing Date: 12/31/76. Amount disbursed (US$ millions): 5.00 Identification: UNESCOIIDA. Preparation IDA. PCR: May 7, 1977. PPARs April 11, 1978. 1. Objectives: a) increase enrollments in Govt-assisted secondary schools; b) diversify secondary school curriculum; c) expand primary teacher training; and d) develop agriculture, commerce and technical subjects in teaching training Colleges and secondary schools. 2. Executing Agency: Ministry of National Eucation. 3. Main Issues: a) Changing national priorities and effects on educational planning; b) Weak educational planning, lack of data and staff shortages; c) Neither Govt nor IDA had full understanding of needs and priorities; and d) Disagreements between IDA and Govt on needs. 4. Implementation Experience: a) Project unit inadequately staffed, dependent on expatriates, little carry-out of project staff from Education I to Education II; b) Lack of Govt counterpart funds during 1971172; c) Importation delays of construction materials; d) Water supply and sewage works deficient; e) Lack of maintenance of facilities; f) Procurement of equipment problematic; and g) supervision: 61 man-weeks, 3.6 man-weeks per mission. 5. Outcomes and Impact: a) Diversification of curriculum to include "practical" subjects in secondary schools and teacher-training colleges was achieved; and b) Target increase in secondary schools enrollment was not achieved due to underprovision of boarding student places. Only about 70% of target increase was met. 6. Lessons: a) Need for sector study prior to project preparation b) Need for better communication between IDA and Govt in order to achieve effective preparation; c) IDA should have financed larger share of costs to relieve Govt of some of the burden of local capital counterpart funds; and d) Provision for post-project monitoring should have been considered to assure adequate maintenance. EDUCATION III. Credit Number: 232 TA Total Appraised Project Cost (US$ millions)t 4.69 Agreement: February 5, 1971. Total Actual Project Cost (US$ million): 4.65 Original Closing Date: 12/31/74. Underrun (%)t 0.5 Actual Closing Date: 4/26/78. Credit Amount (US$ million): 3.30 Time Overrun (%): 83 Amount disbursed (US$ millions): 3.30 Project Identification: Cancelled: 0 Project Preparationt FAO/IDA. PCRt may 26, 1978. PPAR: August 1, 1979. 1. Objectives: improvement fo facilities for agricultural eddcation and training to increase the supply of trained manpower to meet agricultural sector targets during the Second Development Plan (1968-1974). 2. Executing Agency: Ministry of Agriculture. 3. Main Issuess a) Project preparation was not far advanced at the time of project approval. Detailed work on training was left to be done during problem implementation. While this was not unusual in education projects it was risky in the rapidly changing circumstances of Tanzania; and b) Govt decided to establish broader training programs in the agricultural institutes. 4. Implementation Experience: a) The actual implementation period took three additional years. Most of the delays were of the type that could have been avoided by a more advanced project preparation previous to approval; b) Project implementation was handicapped by an unexperienced and understaffed PIU. The Unit did not have a full-time director until January 1972, an architect until August 1975, and it did not have an accountant during the first three years; a) The Govt was reluctant to appoint technical assistance personnel in general, and in particular specialists who were to determine equipment and teaching materials needs and prepare equipment lists; d) Administrative changes rithin the Govt especially related to the administration of Rural Training Centers exacerbated communication problems between PIU and the Govt; e) Project implementation suffered a major setback early in 1973 from the inadequate response from contractors for a tender for the construction of 13 facilities and from excessive high quotations; and f) The project also suffered from major instructional materials procurement problemst periodic shortages of shipping space, slowness of custom clearance and suppliers' raising prices on delayed items. g) Supervision efforts were insufficient in many respects. First, the average intensity of field supervision (1.2 man-weeks every 5 months) was below the average for education projects and only one mission (out of 14) included an agricultural educator, and two others (in mid-1975 and early 1977) included a sociologist and an adult education specialist. Second, there was inadequate site and construction quality supervision. Third, reports were silent with respect to GVT educational policies. Thus IDA was out of touch with changes such as the ones related to RTCs. Fourth, no machinery was established for the continuous evaluation of RTCs, an important component included in the Credit Agreement. 5. Outcomes and Impact: a) RTCs were transformed into FDCa which provided longer courses with little emphasis on agriculture and were directed to a wider population; b) The number of farmer training courses were fewer than expected and the use-factors were low and recurrent costs high; c) While the total number of diploma trainees exceeded appraisal estimates, the number of certificate graduates fell below expectations; and d) While the training institutions were completed, the standard of construction varied from adequate to poor. Inadequate site layout led to user complaints. 6. Lessons: a) Need for more advanced planning and project preparation prior to project approval; b) Need for adequate supervision by the range of specialists (architects and educators) required to cope with the problems of project implementation; c) Closer attention by IDA and Borrower to establish agreed evaluation machinery; and d) IDA was flexible in adapting the project to the evolving policy environment. EDUCATION IV. Credit Number: 371 TA Total Appraised Project Cost(T. Shs.millions): 104.1 Agreement: April 13, 1973. Total Actual Project Cost (T.Shs. millions): 117.30 Original Closing Date: 6/30178. Overrun (%): 13 Actual Closing Dates 12/31/80. Total Appraised Project Cost (US$ millions): 10.30 Project Identification Overseas Liaison Committee. Total Actual Project Cost (US$ millions): 10.20 Project Preparations IDA Underrun (M): 2 PPAR: - Credit Amount (US$ million): 10.20 PCR: September 21, 19831 UNESCO Cancelled: 0.1% 1. Objectives: a) the expansion of lower secondary education; b) the increase in the production of primary schoo. teachers; c) qualitative Improvement of primary school teachers particularly in the field of health and science; d) increased output of skilled craftsmen; e) an expanded supply of doctors and their supplementary training in rural health activities; f) the establishment of Community Education Centers on an experimental basis; and g) a study to rationalize the management training system. 2. Executing Agency: Ministry of National Education. 3. Main Issuess a) It was the first project to require a substantial PIU. While PIU's effectiveness was slowed by delay in the provision of TA, it became effective and was assigned to implement projects funded by other donors; b) Lack of coordination between program development preparation of equipment lists for the CNEs and VICs; c) Difficulty ensuring sufficient recurrent funding to purchase raw materials for workshops in the CECs; and d) the success of Bagamoyo Rural Health Center in integrating practical rural concerns into formal health training. 4. Implementation Experience a) The project suffered delays that were never recovered; b) Financial constraints and costs overruns due to long delays and higher than anticipated inflation led to the three secondary schools being dropped; and c) Project preparation was not well advanced in some areas at appraisal; d) Components were too varied and geographically dispersed. e) Insufficient and inappropiate equipment, lack of maintenance, and delays in the start-up of operations in some CNEs and VTCs; f) TA to health component substantially reduced; and g) Supervision: 16 missions. Supervision missions were too short and not adequately staffed to supervise the numerous, separated and varied project components. Supervision was sometimes combined with identification, preparation and appraisal activities as well as coverage of earlier and later projects. 5. Outcomes and Impact: a) The provision of three new secondary schools was cancelled due to project cost escalation; b) The two new CNEs were operational. However, due to staff shortages and the lack of appropriate equipment, the Colleges had not provided the prescribed program of studies; c) The upgrading of primary school teachers in the fields of mathematics and science at the Bagamoyo CNE had not been implemented when the completion mission visited Tanzania; d) Internal efficiency and external productivity of VTCs were satisfactory. Insufficient, and some cases inappropriate equipment hampered the Implementation of the instructional programs; e) Only four of the eight CECs were operational. The adult education programs were not conducted as planned. There were serious maintenance problems, and the location of all CECs in one region was questionned. The costs involved in this item prohibited its generalization; f) The health education component was relatively successful. The non- completion of the two operating theatres at the Muhumbili hospital reduced the effectiveness of the instructional program; and g) The Management Training study was successful. The management training system was improved as a result of the study recommendations. 6. Lessons Learneds a) Project preparation was in some instances not well advanced at appraisal. Future appraisal should exclude items not well prepared if there is a risk that project preparation will not be accelerated early in project implementation; b) The project required a significant strengthening of the PIU which was beyond IDA or Government since TA was provided by bilateral sources. The Bank could have anticipated such a problem and taken a more aggressive approach during project preparation; and c) PIU suffered from low administrative position. PIU should have been given a departmental status due to its heavy responsibilities in project planning and implementation. EDUCATION V. Credit Number: 607 TA Total Appraised Project Cost (US$ millions)t 15.00 Agreement: January 29, 1976. Total Actual Project Cost (US$ million): 8.80 Original Closing Date: 6/30/82 Underrun (%): 41 Actual Closing Date: 6/30/82 Credit Amount (US$ million): 11.00 Project Identificationt UNESCO Amount disbursed (US$ million): 7.06 Project Preparation UNESCO Cancelled (US$ million): 3.94 PCR: May 27, 1985. 1. Objectives: a) provide logistical support and housing to Village Management Technicians; b) assist in the diversification of secondary school curricula; c) rationalize and upgrade the training of accounting and auditing personnel; and d) improve data base for planning primary and secondary school development. 2. Executing Agenciess a) Ministry of National Education (through PIU); b) Executive Group of Prime Ministry Office; c) National Board of Accounting and Auditing. 3. Ma & Issuess a) Change in the Govt. policy regarding "Villagization Program"; b) Govt.'s limited capability to provide practical subject teachers and recurrent funds; and c) Implementation was hampered by the existence of three project units and the fact that each component involved country-wide activities. 4. Implementation Experiences a) Credit Agreement was amended to give more support to VMTs component; b) Implementation was hampered by: i) lack of counterpart funds; ii) lack of building materials, transport facilities and construction equipment; iii) inadequate supervision by regional and district authorities. c) The project was supervised by 15 missions which were timely and the Bank demonstrated a continuous understanding of the implementation problems and took actions to alleviate them. I 5. Outcomes and Impact: a) The VMTP component: outcomes were affected by the Govt change in policy regarding the "Villagization Program". The training element produced more VMTs than anticipated. The quality of the training was questionnable; b) Secondary School component: not successful because of the shortage of practical subject teachers, poorly designed workshops, substandard equipment, and low enrollments; and a) The objectives of the studies on primary and secondary education and accountancy were achieved. 6. Lessons: a) need to exert great caution in the formulationlappraisal stage of projects which involve sweeping social/political reforms despite apparent benefits; b) necessity to assure through appropriate covenants, the minimization of risks; and c) necessity to closely assess the Govt. capability in terms of providing specialized teachers as well as resources for training materials when providing workshops and specialized facilities as was the case of the Secondary School component. EDUCATION VI. Credit Number: 861 TA Total Project Cost (US$ millions): 16.00 Agreement: January 22, 1979. Credit Amount (US$ million): 12.00 Original Closing Date: 12/31/84. Amount disbursed (US$ millions): 10.02 Actual Closing Date: 12/31/88 (as of 8/31/88) Project Preparation IDA. PCR: to be undertaken. 1. Objective: assist in the development of critical middle level management and skill training for the modern sector. 2. Executing Agencys Ministry of National Education (through the project unit). 3. Implementation Experience: a) Project well managed by capable coordinator, Technical staff, augmented by DANIDA, was adequate although the procurement section was weak; b) Implementation delays caused by general shortages of materials, fuel and transportation; and c) Civil works component experienced delays due to contractual disputes with contractors. Contractors were modified to reflect substantial increases in the cost of labor and materials. 4. Outcomes: a) All civil works completed except for Dar School of Accountacy which is scheduled for completion in December 1988. - Z3L4 - EDUCATION VII. Credit Number: 1056 TA Total Project Cost (US$ millions): 55.5 Agreement: 09/28/80. Credit Amount (US$ million): 25.00 Closing Dates 12/31/88. Cofinancing (DANIDA)t 20.00 Revised Closing Date: 12/31/89 Credit Amount Disbursed (US$ millions): 7.24 (as of 6/31/88) 1. Objectives: a) Improve the quality of primary education, and to provide poorer districts with greater access to primary schooling; b) Improve the quality and increase the output of upper secondary education; and c) Enhance the capacity of key Governmental and parastatal agencies to manage the sector. 2. Executing Agency: Ministry of National Education. 3. Implementation Experiencet a) Implementation of construction slow due to: i) Budget procedures: local govts. difficulties allocating necessary funds to permit construction to proceed from one fiscal year to the next. Payments delayed by Treasury and local offices; ii) Bidding procedures: contracts awarded to firms not capable to carry-out the work; and iii) Contract pricess because of devaluation too low to motivate contractors to finish their work. Aft-r long delays adjustment were granted in mid-88 and wor. resumed. b) Procurement of furniture: i) Contracts awarded to companies not able to do the work; and ii) Payments delayed, and contracts outdated. a) Procurement of equipments project unit has procured equipment and books not suitable for rural schools. Acoeneis 2 Covenantse in World Bank Education Projects in Tanzanie roiatse in Tnasmai Educataon I (963-67) Implementation Credit Agreemens 4.01 - Cause the Projeat to be carried out with due Fulfilled. diligence and efficiency. ane the schools to be operated in accoreance with sauno educational practices. - Cause the schools and equiomens to be adevately ? maintained. - Make available promptly neeeed m for carrying Fulfilled. out the Project. - Maintain adequate recoda. Fulfilled. 4.02 * Furnish to IDA all necssry information with regard Fulfilled. to the operation of the school system and progress for educational develosment. - Furnish to IDA, upon ereparation. programs for Fulfilled. the construction of new schools and the easension of existing schools. 4.0 * The Gorroawr ad ZDA should cooperate fully to Fulfilled. aeare the accomtishens of the purposs of the Project. 4.08 - The Agreement shall be free of taxes. Fulfilled. Closing Date: 0e. 81. 1967. December 81. 1967. - � i - твniм�в: Мог1� WMt Q+iвsвЧвw РгоlвеУ - eduas0•т I12 (1071•107л). Isвlммиs•м �гвli4 �Огеммв �.аs;м д ел. Ргоj.о. � . 3.0� - P.ariяim д 1иМв. 1вai1iбiи. ип;аи м1 еtАвг Iлo�lieiвпf вwвNв вав �iвм fe гивrгsвв моinв tвг амв ►ro1м. P1U b моА. 9.0� • Ew1q вг�•i{NWpi м� мМiаигiр аоnв�иl{вsiв for ° In MiM в/ вгвjw. виргооiм впв вввгв;иl tАв вгввагввiм о/ вгaAiNafopt Угiдв. •a�iwм � t•!в. ва,ввввввsв fAвs Сав.вгt• aRo11 aos �• bi•д, Wад�iпв ow;вis вм 1uмiWrot смве �иi�n oi thв if ив вввеiwlв� b иг�егs tM вогh. виi�iеь.. а» •г.ввnим ef в;иiм вowи.sv. _ 3.0� • Садв Na11 Ув ввN вsеlwiввlу fег О�в Угеjвав. А3 iоиl iw iывituliaи (iОСв) вви wвд гог .eм.rroa►o■i wno... � оии со..в..в. __............ a.oL - ивiwe.iw •..вг✓в •iW ввввiвfвввllr оr.в ваввwsiл0 R.o.d. веа вО.вwиllr oiafsiaвв Чавов вгмiа. вм .вмвiаа w fiпмai•I owaisiaв д ваввrвввsв .а w1а�.в вhм sал в�аи дМ INjвiOв огl�Mi �дI0{в�т. . •.Ф • авв а i.Os ♦.03 • 1hв вММвliвоl iпо�iМ.0iввв д �Ав Ргоlви 6о И 0�lвsfiwв д�мовi +мгiцо iввlввмNМаn. ОрА0М М N r вАвввв УМ в�овf�Аlвi e►jaЙws R1Ов вwtввгвМ iaM /Ы� Овлвlввовд W1lвpvs д Ми bмввм. вw1 qs вW1 tAa вitA вoiiliв/ (вwв рввгвl йио�iвs). м� Мг. Tnlaiпp lвввУвriв вМ NвfдiвlгМвгв iд NвМвМ виаУвгв. СММ/д Vвiвв� орМгвllи. j - ввьlвfМв, tв.wiвои вм мтамв а и.в.o.rlr вnilвlмв мв д.во0.Чr os.s.ia.r. oiwMlмi вМ гврiвд• м/ o�liвh вргвдгiеМ linмвiвl вгеsNrмв 1ег вhiв рг�w- ♦.04 � FsfN11в1ь мl.оiМв{в в Моiввв ивi0 вitАlл f1и N1w 1�ьlti 1в11•1{вв 1{гввМг врвiвр/ in .haoq д iyгiw�wи .амiиiи� s1 в ваеiив ееогвiм�вг i0l�. вгw+виs in wpp a07s ов веваавмв s вrвjвМ вгд�i0ввi. в Nвlilid ваswмОМ1 аМ iп 1q4. wiw вWt и •вдвв. 4.аб • ОгвМ 1в11агаЬiр М Ч вгегi�. On1y 4.3 д!М вмМ вввqоq агв аМ. •$ilqoвf O/ Маiвliв4 00 вaii/ iд thв вгвргвliФ 1O.s Мд'lЧгi Nгд 1д� У�р1вв/. д Мв вwгво д Hnq а«1 мrio амiwlв виr вwри иiмfвЧм Iег Ми Иеjва�'в .rwввfвмl iiWiwвiвдв. - цгwо�Ам в1и Nвiwiвlоьlrв вsrовwгв д е1а Rвиег+вb Nвв k�11i11вo. мд ТлiвiМ Oiвisia� д QАв N1e. д АвгfwlWгв. '� L .� а аб - Ргвввдв. •iбд�n в�. воnив вiбвг Аргввiиnв. tдв Иоа 6u1Ii11•д. бn�niM вгарувв iог васа АТС. - Ргвввnf. •ieJ�in одв вввг вitвг Аргвимв. еигмаиlв Иоб 1oNiI1N. оаг вЧ еаигвр W Ув вгоvьвN ов Арг. Тгsiлiав tМО�бвtи. - Evв1wM sonO�avoysly Чи вtlвббiвмрв oi рТГ,. Pby fu11i I1в�. - G/гfiCYla IGГ АрГ. Тгвiаiд0 iдв0irwбвs маи1� pirв Оbiвабlввв едlдвw вУ Qo•гвм�мв еУг�др Сгавiав,Кв io Oгsiaina in fвгв qмрвввnв вао tвН iваlвввwt�{�an. '� ввtlqaiqвiea. Мввsввв ы+вбiWбв вnв11 •rov�iв • саигавв �n адiовl •гarrstiOn вб tАв ¢вгб�liеiбо iвввl мi ia vвMгiasry ов в1и �iвlaes 1ввв1. • • ЕffваСiвв Os6вs Gaп�i�ionв • . 8.03 - Eвlqrвws e/ Pгojвsf СоаNiав/ег вдв вгемiаваf. FЫ 1вNillaa. gi!sasiв• ааМ воввваnМ. а¢.1до aas о.и�.. ai. ivн. ао.таа wa� iv�s. . - � с - �апкмьва wгьд BsnY ёдиавеьал ProjwOa :дvм•woa (Сапе.1 � еа„еаеьоп ta (ioвwto7вl. -- 2леlа.аьовоьм Gdi{ Аргвввмо иы •i Ргоерав о1 thв CrNie з.са - тод• оо и вмарг.д •л т• ь.а:• ot �ло.пг.о:мн еми.isьw wиtllи • � biЧiap. 3.W - Сооlв finвna•а b bв иввд вааlивiввlr in евгггiлр ои0 !hв • Fu11i11d1. OгЫM• акгоiеиlег Соrммав в.01 � Proj•а► iмoioreioм ао i• вгaridN аiеИ аи•iifiвd еваамг• Inads•wo• пwЬег ot ,atw3sмwм. иив.г. (•и и1а.). - Bti8ldiapв. 1амiWм мд мiввмв ао ►• вeiMeiwM Мв{дtмма а�ое seпiM овs. мi м•:М вМ •вopYliab Nвiwi.v.otw аи1 _ ♦1•рмiвl вг•авNг•в 1ог о1а wгвавв о! auah М10е•RM1a•. � б1М►liвh 1л 1qa в•вавiаl pwii h•едвд �s • вгоiва� Р1ц qWlidк! iл iов вр«дiыеог еьк� арвоiwевмо о/ вn вваlыеаае. вл tp11 аиw ia дмrвр вsМrдамв. мд wдвгвiМ eвsbwiasi. вааомsiлр 1УТ4. Fu11.0iaв мд а•вг•V►i•1 •ОвТ� С0р/'д3лвоаР •Ч�:ле•д iл 107д. . . ' бвlq ы ева•в Ма •ввlа�r aыlidiad •М •у•гiмаМ Сал•вlрдfв рввiдуд. ргрвве{r. eawiMnM •гаАi{ыМ as dripa аМ �уwгвiи Pbpr w�•wiвiaд а1 Ti•Аiв� •огkв ' веhрвi aaь�w0ia►. ваавомsв eI iлlвгiог Mqliyr ввМгiвlв. aner/�iatмo ' аОМ:д•г•оlад ptOM b рlOV3aiQA о/ авtв •воа мд евввра. в.0'! - Одваlw pгiaalWni. еаввwlеl мд овдаNа•1 aubj•аев Qрв•геввае aNpiaN ТА nввд•9 1от in ib Ммhiлр eгeiwi•р iaeoiepoiaм мд aвaвwfayr tввМиг огаiw3лр. АА•врвр аwйвв •аУр1а. Еа�1ал ij i0 awlitlod aвasiвli•ев fo pвiN о1 МрА•гв а•г• егаiавд. hapraг. ia •aвYlfahipp рд•алаs fпiliela• 1ог eгsiлiee s•aaq•n tгpЫr oг.iw•д оиааиа d.wraN ia ргNеiввl в�ьУjааев вМ ia давslsвiup apгiwlWгal. Iаг еЧ+вг tya•в tI •м1•уввwв. еа�виа:еi. .ад esa�issl adaиaian in .кмдегц •дюоlа� W1leыbia а•г• в1м в�а.:дN ii) pewiЧ а0 f•1lasahiи 1ег бniлinp Тмааiмв о••гwо eaвNiap ор аав•nмо. - и вв•piвlistв 1n �гвавiдl auYj•аtв. •.OS � sмаавthм Ча wtI iп eh• амвваоемо• iл насi �r 4 .вrвiввi,и ра рд•.wи аwьвг w ииiн i.e. ia вгмiавl .иь).sь.. сlмtм д•а� оа..ь.г si. aoTS. с1м1м д.и R.ь.дргТ sx. avss. гвnsм,r �ог�• g�nk Едие•е,аа Projwи Соrвмnав сса•е. ) _�ие•е�ол iv (1os3-ia80) I•е1и••е•e.an Сгвдiе Аргввв•no � Еаоб•eial о4 `Ав Рrolвае � '- 3.01 - Ево•М вn1 а:•esip еhв Рго1оа9 tlлi0 eoиieeie® pni Futlillo/. л вгоiвае eaor/iмеаг. е Оао•egr вгоiвае ааа►/inвеаг ве ваеоиееме, • огеаягввме ofliпr. fоиг егд�ibпев. • aiвil мОinsвв, е а1вгУ of гогkв. вn sd•wsaв n�•ввг � Of �W{1t1f/ /г•te®МА• 611Гр Й1�1/1др .11а••BiOro, •1!/ - аГLог Mpi••гг iУг•ОIИдр iL•�7. 3.03 - Ргавi/в в•.ЧУ1в вгваiвев for РIЧ Fu11i11•д. 3.0� • Еавlа�r вгд+iеваWnl еепамlЧлев еМ аи•лsiб�r wr+qiwo Fu1t111e/. ►ае аiев виовг�iвiаи еав ean�t) емМ. i мММ�вМ. s.� • Аргi о!М лдiгв/ Iв110во11{М мf М1Оу iеооСlвl)вФо1 :� ав ...-,..г. ог т/ в« �а. о* и./{.{•.. г) м/ ::) а1.о.+ w Ра•в. о.г. влs{а.11, ii) s6 вво-sввгs о/ 1в11а•а4iы for огоевоsеiвв в•вУвго егN (TAj 11.� qw.lrигоt Rв1i 10.0 ввw.ррл). fJю Pes. о/ Na/ieiм. � i7i) Q ow�an oF ТА 1ег {lаарввме Tniainp $Ч�у• 1оав1. г•) Рro1м un:e� •гамiеаав <а .муал). ir) рвl.lrв iw г.вw�iавввв .ав /iNaaiae. о! aiвil б�iыqiuoJ Q.5 в•sqrup. варiвiеiи /ив{w� tiгM 10 веsеlу rвlврг/ аlвгfг еР оогУвl l.8 вм-sачв. orвjвsf iвр{. Ni� Wвlier о1 вtаН г}агУiев/. • 3.00 • Gыв Уи •ам�гоеiт в4 CFtв b Ч аеггiЧ еие Qa1o аnв ва�ilв rnie ви oeiliaN siдlr /ив Ьу воУilв аопвегоаsiм rnieв о1 Сhв гимl емtил. Ое 1ва4 в/ ii11M •егtвл. а.О? • Св<ие раЬ iмfiev<im ее дгго амв i!в аоввееiодв А11 1asslMhiaaa in aыrwвian па+ввь 1ог еhв • in атоеNааи вieb вawN вдвiдiвблеiгв. ♦inмeiвl W11врв вF Neeieлti �eip� вhieh иw иаьвiавl. вn1 вlwвеiОмl оо1 ieia. un/мвааfвв/. {.вд� М вМеiввив ое УГ�Св вв/в ie iвовввi�l• �е ееnИМе вiвi�tв ааилвв. м/ tве аовгвеiррl УивМгв in еАв Мваlliдо hewieвl nл иов ia we. ♦.03 • М•ioeвia м/ гвыiг thв Ргоjвае рыil/f•р. вдиiыввь 1имiМ.л. tад е/ в/вwвов в•iwевввwи. вивsiвllу ое fM СВв и1 Cllв. � в.04 • Смlиое. ne lвем аАи tво sал rIa•► Аог+мв. Na� aeniN аиа. м вsвlweian о/ д» мrsаг вгiаsго еаеаАвг eлiдino олвгвв• � •.Об • Qrьвi� в1м 1ы Ми мвгssiад о1 bhв oWhlвYili Морiрвl ТМ UwiM1 ла woe 1ut1y qeqeianвi. и в Овегbiао hевоiЧl. Cterвiu� iвМ: Junв i0, 197�. ОввввУвг ,1, 10l0. � С " iПfM1• �ОГ{О Qвf1Y Е'AYQie1011 Т10)вСбi Совммав Ссове.) �ирslм V (1V78-1а85р imoiиwaolм Сгв/ь0 АУгввввдо Eaвcunian of еhв Proiвes _ � . 3.02 - Gгry оие VйI► Соllваnме throыaM thв РгоjИге Вавеирlьв. АваiвСвд iy NoNia Counмiq. buб вi i ариiвiiа0в Уие Уnв Ь�д 1Ме !у еhв вМ ` ot 19�0 - мд Sваоnввry bаlтlв аеввоомв сАгоиоА PIU. 2n yia о1 sq9f вlтгNвll vorY ввв SвeiaAoбnryr. - Сеnеiмгоивiу воwlsог vи» а.аии Чв Ргоiв�о 6ваивlvо t Ргав�ьlу вом wlei 1 14FO. - Wпiвh вввry вis вenw ал гвоога м wв1. of VИТР. 4 Рго�в►1r дам ипоil 1980. 3.03 � бввlоq вiw:ви futl-Ofr satt 1ог PIU. ttв4Уiгвд seвt1s91i вговiдвдs40. $.Oi � 6as�OJ вРи1s{isti W Qa1гi ОУе в:Удiи. FV{ti{Iвд. ij t ивеiвliвОв (8 ам�иn) fог NBAR вWу. i i) 1 виsiеl iве (2 вм� an) fм вво. вdwslri wгвву. iif) 1 ииiвliМ (II вмsигв) fог aris. вдУавеiоw sqNS. 9.ОВ � Аввi�n рвмвгiвМ ивlкn ef сааь4вгргев b еивМiвевв tb! 1У11{tto�f .е оМ Ргоiвев бвawis. вМ PJU. 5.0� � РгеУiдв вдвrивев АУУвiв� ев вqвг0в Т 9.0� - Mмisaвie вдваУвЧ гвеегrв. мов IY1fi11Ч Yr Proiиe 6вeveivв unie. a.pg - WУво еУ oiaMiд вл1 грвiг вroiвае enildilapв МвiвОми¢в е9 иеоМвry/ в¢мроьв 1ае вдмУ:ев вМ��аМе �М �УlА1Wгв. сtиiм �а: ,iим ю. iaBS. ciиiYO о.аг suгw ю. i9a1s. �� ' � �sлрт• �ог�д 8апУ Ёдиав0�оп г'гоlваМ � Саrммев (Canf.) _auue�an vI (1979-198�). :■в�ммае�аn � егыiв аQги.ме 9.0? � Eawiey nваwвгв вадf tег аввгвъiм oI Inвегиааог �. Tniwiiц СмМг мд tor 9вг 8ahool of Аеаамавмs. 9.W - 8иЬвiе а0в�t !w. оlввв 1ог 2лвегиаLог Тгвiпiпр GnMг. Dвг Sehaol о1 Ааеоиnавпа�r вnв ltвввsвввnа ' Tniaina Р►аgгр ву Ова. 81. Ъ979. 3.0� - BuMie игвв of гд•гмав iог аавоi•fi•е гваиiгN •11 а�ммав Iu11i11в0 и 8/18/б7 •гаа• ter Nnro�ee Tniaiea fог Е4иао�оп Рвгммв� �ог 8.0В. Еrвlааеiм to Ч иnдвгWм ву . Inв9iWeв м1 Faea Afгiaвn Ммваввме lnaeieueв СооНiмеiда i/nie Iег RЧввгм anf 6вlweian �иигwеог eniwfдa вгеаг.. и, о.а. s1, ко7о, (cLL1�, вад с�1оу epaiвi iвf Y1r МвгаА 81. i980. s.0� • Lю1пlв iw Р1{1 а►рг�веwМеiвв at еhг Nfe. е1 1.вЬаг. S.Oi • Nвiahiв вр�вргlвМ гвавNа. • Мви{0вг мi wвiwM Нк Projвa►. вМ виМiе а1 Мвгвnsв �ег м1. Ьу Gaa• 81. 1070. ♦.О8 • 14IatвI11 YYilliдpв ГУг1liеУгв гМ ваУiвввво. ♦.Oi • Gwв Na РЛ! {о мв/р evвliliвd sWP. влд • sniwiiq 1ог 1еав1 веМ1. - а.рв • CeYw Projвsf iмeiMieiaю ео авмм ввааrвиу.. Ctes{р Овfв� ОвивУвг i1. i0da. ,рр� jp. 19у. Appendix 3 Expected and Actual Results of World Bank Education Projects in Tanzania ProJect/Results Expected Actual 1. Education I Create 6,845 additional places Expected results seem to have been for see. school students by accomplished . 2.5% of see. school 1968 (2.5% of see. school-age age pop. was enrolled in sec% pop. to be enrolled by 1968) school system in 1968. through expansions to 27 schools, constr.-of 2 new schools, completion of 7- schools, adding boarding to 9 schools, conversion of 10 prim. schools and teachers colleges to see, schools. Il. Education 11 1) Secondary Schools: Create 5,072 new student Places built: 4,228 places, (of which 4, 672 not underprovision of boarding new enrollment: boardings facilities not new enrollments 3,290; day studentas 1,382) 3,242 through constr. of 1 now boarding: 2,832 school and extensions to 24 day students: 410 schools. Most remodeling work wais *14-Justed to keep project costs within appr. estimates; inadequate water supply, some sanitary facilities were closed. ii) Secondary School Teacher Trainingt Create 80 add. places in Dar Places built: 90 See. Training College through Covenant fulfilled, but teachers extensions to the College. did not stay in school system. Covenants obtain TA for Only 16% of teachers could teach teacher training program. practical subjects Instead of required 30Z. III) Technical Educations Create 450 add. places at Dar Technical College (280 Places built: 460 boarding). Boarding: 278. iv) Prim. Teacher Training: Places built: 468. Creation of 365 add. places. Zd:. ducation I:. i) 9 Rural T:aining Centers (ROCs): Objectives changed during Each center was to offer oe implementation. While construction or two-week agricultural work was cwmpleted, RTCs were courses, 40 coursesiyear with transformed into FDCs. FDCs 2,000 to 3,000 participants provided little agricultural (farmers & wives) at each training to a wider target center. population. Low utilization rate (34%). ii) Min. of Agriculture Training Institutes (MATIs)t Expand 2 existing MATIs to Constr. completed. increase enrollment by 42 and 139 respectively, replace 100 boarding places, establish 2 new MATIs each for 185 students. Project would meet 54Z of 611 of the planned output and 821 demand for animal husbandry by of planned enrollments were 1979 and 181 for crop achieved. husbandry. Enrollments Certificates 815 488 Diplomas 125 214 Others 50 outout: Certificates 385 204 Diplomas 125 106 iv) Technical Assistance, fellowshipst 14 man-years of 10.5 man-years of TA not used only TA and 7 man-years of 4.5 man-years of fellowships were fellowships. used. ?rojectlResuits 7xnectea .Actua. LV. 7ducation IV. i) Prim. Teacner Training: -increase production oi prim. 960 new places were constructed but school teachers to meet 400 staff shortages hampered training p.a. deficit by creating 480 program quality. new places through constr. of two new Colleges. Facilities used for other purposes -quality improvement in the (courses assisted under Sixth fields of science and Credit). mathematics through extensions Program expected to train 720 to Bagamoyo in-service CNE. teachers annually. ii) Vocational Trainings Alleviate shortfall in supply 600 add. places built. of skilled workers by creating No info. on output, however, PCR 300 add. places on each center shows number of trainees dropped and an annual output of 150 after first year of operation. craftsmen. iii) Coimunity Education Centers (CECs)i 4 centers completed an 4 under 8 new CECs to be constructed construction by the time of project and provided with furniture completion. Furniture and and equipment. equipment incomplete. No. info on number of trainees, however, PCR mentions inadequate training provided. Lv) 3 new secondary schools. Dropped during implemenration. v) Study of Management Study successfully undertaken. Education and Training. vi) Health Education: Increase annual output of Annual output increased fr'.a 30 to doctors from 30 to about 60, 50. and training to assistants Delays in TA, and funds drastically through extensions to Dar. Fac cut. (see below) of Medicine and Bagamoya Rural Nevertheless, program was Health Center and TA. successful. v) Technical Assistance & fellowshipss - 35 man-years of TA for Fac. TAs 11.4 man-years; Fel: 19.9 man- of Medicine. years. -35 man-years of fellowships for prospective members Fac. of Medicine. - 2 man-years of TA for Local Management Study. ? -Project Unit: architect (8 man-years) civil engineer (2.5 man- years); clerk of works (1.5 man-years). U cecr.iResu.ts _xpecred Actua. . cation V. i) Village Management Technicians Program: Train 1,500 cooperative 'More than 1,500 technicians were inspectors and community trained, 400 houses were built, 900 development workers. pedal-cycles were bought, and out Equipping, furnishing, and of the 500 motorcycles provided, 40 provision of 2,000 pedal- were commandered by the police. cycles, contr. of about 1,500 houses and provision of 500 motorcycles. ii) Secondary Educations Expansion of capacity of 15 Expansion of capacity of 13 schools schools from 6,900 students to from 6,064 to 6,422 (7,860 9,000. expected). i11) Studies: - Survey of existing Undertaken. facilities for accountants and auditors - Survey of educational Undertaken. facilities in all see. schools. - Primary education sector Undertaken. review. iv) Technical assistances - 4 specialists (8 man-years) 5 specialist, 9.8 man-years. for NBAA study. - 1 specialist (2 man-years) 2 specialist, 6 ran-years for sec. schools survey. - 1 specialist (2 man-years) 3 specialist, duration n.e.. for prim. education study. ?roiectiResuits Expected Actua. VI. `iducazion VI. i) Vocationai and Technical Educations - additional equipment to 2 Completion Report to be undertaken. VTCs and 150 student boarding All work completed except for DSA. places and 15 staif at each center. - new Instructor Training Center for a total enrollment of 160 trainees. ii) Secondary Educations 160 hostel places to each of 4 technical secondary schools to raise women participation from 3Z to a planned 252. iii) Management and - Accountancy Training: Expansions to Dar School of Accountancy to increase DSA enrollment to 1,015. iv) Management Training for Educational Personnels Creation of a training institution in Bagawayo to train about 500 students per anm. About 2,000 MOE staff ambers to be trained during the first five years. v) Project Related Studiess Institute of Education Pro- Iavestment study. vI) Support to project monitoring and evaluation.