Tanzania - Pyrethrum Project

Document of The World Bank VJ FOR OFFICIAL USE ONLY Report No.: 2749b-TA STAFF APPRAISAL REPORT TANZANIA PYRETHRUM PROJECT March 20, 1980 Regional Projects Department Eastern Africa Regional Office 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.  CURRENCY EQUIVALENTS Currency unit = Tanzania shilling (Tsh) US$ 0.12 = Tsh 1.0 US$ 1.00 = Tsh 8.30 US$ 1,000 = Tsh 8,300 WEIGHTS AND MEASURES 1 meter (m) = 3.28 feet (ft) 1 kilometer (ki4) = 0.62 miles 1 sq. meter (i ) = 10.76 sq: ft 1 hectare (ha) = 10,000 m = 2.47 acres 1 kilogram (kg) = 2.204 pounds (lb) 1 metric ton (m ton) = 1,000 kg = 2,204 lb GLOSSARY OF ABBREVIATIONS APO - Assistant Production Officer GAPEX - General Agricultural Products Export Company GOT - Government of Tanzania ICB - International Competitive Bidding IERR - Internal Economic Rate of Return KILIMO - Ministry of Agriculture MDB - Marketing Development Bureau of Kilimo MOW - Ministry of Works PMO - Prime Minister's Office RIMU - Road Improvement and Maintenance Unit RMEA - Regional Mission in Eastern Africa TAN-ZAM - Tanzania-Zambia Highway TAZARA - Tanzania-Zambia Railway TECO - Tanganyika Extract Company TIB - Tanzania Investment Bank TPB - Tanganyika Pyrethrum Board UAC - Uyole Agricultural Center GOVERNMENT FISCAL YEAR July 1 - June 30 TPB FINANCIAL YEAR July 1 - June 30  FOR OFFICIAL USE ONLY TANZANIA APPRAISAL OF THE PYRETHRUM PROJECT Table of Contents Page No. I. THE AGRICULTURAL AND RURAL SECTOR ............................ 1 A. Background 1......... 1 B. Agricultural Services . .... 2 C. Agricultural Marketing and Pricing .... 4 D. Previous Bank Group Assistance . . . 5 II. THE PYRETHRUM SUB-SECTOR 5... ..5 A. General . . 5 B. Pyrethrum Production.. ............... 6 C. Cultivation, Yields and Flower Drying . 7 D. Services to Pyrethrum Growers. . . 8 E. Producer and Export Prices . . .11 F. Future Prospects ....... . .13 III. EXECUTING AGENCIES . . .14 A. The Tanganyika Pyrethrum Board .14 B. The Uyole Agricultural Center .16 C. The Ministry of Works .17 IV. THE PROJECT AREA .17 V. THE PROJECT . . .19 A,- General Description ............................... . 19 B. Detailed Features ....20 C. Project Costs ....28 D. Financing ... .30 VI. PROJECT IMPLEMENTATION. ... 31 A. Management ..31 B. Staffing ..32 C. Procurement ..33 D. Disbursement ..34 E. Accounts and Audit ............................... 35 F. Monitoring and Reporting ..35 G. Evaluation ....... .. ...... ... 36 This report is based on the findings of an appraisal mission, consisting of J. Mullan, P. Santos, G. Onaba and K. Nathan (IDA) and R. Faun (Consultant) which visited Tanzania in April/May 1979. 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 (Cont'd) Page No. VII. PYRETHRUM MARKETS AND PRICE PROSPECTS ...................... 36 A. Uses of Pyrethrum ................. .. .................. 36 B. World Production and Consumption ........... ........... 37 C. Recent Price Trends ................................... 38 D. Market Outlook and Price Prospects .......... .......... 40 VIII. YIELDS, FARMER RETURNS, PRODUCTION AND FINANCIAL ANALYSIS .. 42 A. Flower Yields and Pyrethrin Content ....... ............ 42 B. Farmer Returns and Incomes ............................ 42 C. Production of Dried Flowers, Extract and Marc ......... 44 D. Financial Effects on TPB ............. .. ............... 45 E. Government Cash Flow Implications ........... .......... 46 F. Foreign Exchange Implications ............... . .......... 47 IX. ECONOMIC ANALYSIS AND JUSTIFICATION ........................ 47 X. AGREEMENTS REACHED AND RECOMMENDATION ...................... 50 ANNEXES 1. Supporting Charts and Tables Chart C-1 TPB Organization Chart C-2 Estimated Implementation Schedule Table T-1 Dried Flower Production, 1963/64-1978/79 Table T-2 Number of Farmers, Area Cultivated, Flower Production in Major Zones, 1978/79 Table T-3 Average Pyrethrin Content, 1962/63-1978/79 Table T-4 Summary TPB Income Statements, FY75-FY79 Table T-5 Summary TPB Balance Sheets, FY75-FY79 Table T-6 Projected TPB 'Without Project' Income Statements, FY80-FY90 Table T-7 Projected TPB 'Without Project' Balance Sheets, FY80-FY90 Table T-8 Project TPB 'Without Project' Cash Flows, FY80-FY90 Table T-9 Road Improvement Program Table T-10 List of RIMU Equipment Table T-11 RIMU Local Staff and Labor Requirements Table T-12 Phasing of Project Costs Table T-13 Costs of Smallholder Production Component Table T-14 Costs of Pyrethrum Research Component Table T-15 Costs of TPB Strengthening Table T-16 Costs of Roads Component Table T-17 Estimated Schedule of Disbursements  Table T-18 Production Costs and Returns for Pyrethrum and Maize Table T-19 Farm Incomes Table T-20 Crop Labor Profiles and Farm Labor Utilization Table T-21 Projected Incremental and Total Flower Production Table T-22 Projected Incremental and Total Extract and Marc Production Table T-23 Projected TPB 'With-Project' Income Statements Table T-24 Projected TPB 'With-Project' Balance Sheets Table T-25 Project TPB 'With-Project' Cash Flows Table T-26 Financial Rate of Return and Return to Equity Table T-27 Effects on Government Cash Flow Table T-28 Foreign Exchange Implications Table T-29 Economic Prices of Extract and Marc Table T-30 Economic Costs and Benefits 2. Selected Documents and Data Available in the Project File MAP IBRD-14436 - Pyrethrum Growing Areas MAP IBRD-14386 - Proposed Project Investments  f X f  TANZANIA PYRETHRUM PROJECT I. THE AGRICULTURAL AND RURAL SECTOR A. Background 1.01 Tanzania's population (for the mainland, estimated at about 17 million in 1979) is increasing at about 3% per annum. Per capita GNP for 1978 is estimated at US$230. Real growth of GDP averaged about 5% per annum over 1976-78. Roughly 50% of GDP is derived from agriculture and related activities, more than half of this from subsistence production. More than 90% of the population resides in rural areas, the vast majority engaged in agri- culture. About 70% of Tanzania's foreign exchange earnings from merchandize exports are accounted for by unprocessed agricultural commodities, and a further 7% by processed farm products. The major export commodities are coffee, cotton, sisal, tobacco, tea and cashewnuts. 1.02 Large-scale agriculture is confined to a small number of private estates and state farms producing sisal, coffee, tea, sugar, wheat, rice and livestock. The largely traditionally managed national livestock herd, esti- mated at 10 million head, is grazed extensively over the 40% of the country free from tsetse fly. 1.03 Recent performance of the agricultural sector has been somewhat sluggish. Over the period 1967-77, the average annual rate of growth of agricultural production was about 2.7%, less than the estimated population growth. From the late 1960's onward, food crop production also failed to keep pace with population growth and, as a result, Tanzania became increasingly dependent on imports of maize, rice and wheat. The effects of this slow growth were exacerbated in 1973 and 1974 by the effects of a severe drought which resulted in poor harvests and the need for large imports of food grains. Since 1975, overall agricultural production has recovered from the effects of the drought, increasing in real terms at an average annual rate of about 7%. At the same time there has been a shift in the pattern of production away from cash export crops toward food crop production, so that cash crop production has stagnated and, for some crops, declined. Despite food crop production increases, rising demand has meant that Tanzania has continued to depend on imports of rice and wheat. 1.04 The Government of Tanzania (GOT) has undertaken a comprehensive program to support the development of the agricultural sector, in conjunction with efforts to achieve balanced regional growth and more equitable income distribution. Within these objectives, GOT is paying particular attention to achieving self-sufficiency in food production and to supporting export crops. In the past few years, greater emphasis has been placed on the productive rather than the social aspects of rural development than was the case in earlier years. Beginning in 1970, GOT launched a wide-ranging program to resettle the country's rural population into villages in order to facilitate  - 2 - the provision of infrastructure and services and to encourage self-reliance and a community approach to rural development. There are now about 7,700 registered villages, containing over three-quarters of Tanzania's rural population. The village is intended to be the country's primary social, economic and political unit. Basic social services, including health facilities, classrooms and water supply systems, are to be located in each village. Economic services, including provision of credit, supply of inputs, organization of extension services and marketing of produce, are coordinated at the village level. Economic infrastructure, such as storage godowns, processing equipment and small-scale industries, are expected to be village- owned investments. Early experience with villagization has been mixed, and it is too early to judge how effectively villages will be able to meet their ambitious goals. B. Agricultural Services Research 1.0.) Agricultural research in Tanzania is undertaken by a variety of governmental agencies. The Ministry of Agriculture (Kilimo) is responsible for overall guidance of the national research effort, and operates a network of crop and livestock research stations. Tanzania's research system has several weaknesses. Kilimo has not been able to exercise effective control and, as a result, priorities have not always been firmly established, the regional emphasis has been unclear, and research resources have not been allocated in accordance with development priorities. There is a need for greater focus on improving knowledge about Tanzania's various agro-economic zones, and on improving the farming systems in these zones through a greater emphasis on field testing. The links between research and extension are weak, partly as a result of the failure of the Regions, which are responsib:Le for extension, to coordinate with Kilimo which is in charge of research and training. 1.06 Agricultural research in Tanzania is increasingly becoming linked with the various international agricultural research institutes under the Consultative Group for International Agricultural Research (CGIAR). IDA is intending to assist, through a proposed Agricultural Research Project, in strengthening the ability of the Tanzanian research system to carry out field testing, and to forge effective links with the agricultural extension services. The Tanzanian research system also receives support under the Tabora Rural Development Project (Cr. 703-TA) and the Mwanza/Shinyanga Rural Development Project (Cr. 803-TA). Extension 1.07 Responsibility for agricultural extension in Tanzania is divrided between Kilimo (which trains extension staff), the twenty regional admin- istrations (which are responsible for the day-to-day management of about half of the country's extension staff) and some crop parastatals (which employ the remaining extension staff). In early 1978, a total of 100 graduates, 3,000  - 3 - field officers with formal diploma- or certificate- level training, and 3,500 field assistants with little formal training, were involved in extension work. Extension efforts have been relatively ineffective. As a result of the weak research base, relatively few proven technical packages are available. Field staff are inadequately trained and supervised, and receive limited logistical support. The situation was worsened in 1978 by the decision of Government to appoint salaried Village Managers to each registered village. Of the 4,000 Village Managers already appointed, many came from the better-trained ranks of the extension services. The relationship between Village Managers, extension agents and Village Management Technicians 1/ remains unclear, and Kilimo and the Prime Mfinister's Office (which has overall responsibility for regional administration) hold differing views on how extension should be organized in the future. IDA has made some proposals to Government on the re-organization of the extension service, using the principles of the 'training and visit' system adapted to Tanzania conditions, and it is expected that the extension services in Mwanza and Shinyanga regions will be reorganized along these lines as part of the IDA-supported Rural Development Project for these regions (Cr. 803-TA). Kilimo is currently drawing up a comprehensive plan for the overall development of agricultural extension. However, progress on the improvement of extension services has been, and is likely to continue to be, slow. Credit 1.08 The major agencies involved in the provision of credit to the agricultural sector are the National Bank of Commerce (NBC) and the Tanzania Rural Development Bank (TRDB). NBC is the sole source of short-term credit to parastatals for produce marketing, and in addition provides about two-thirds of the total production credit, mainly to large estates. TRDB provides long, medium and short-term agricultural credit, administers special funds on behalf of Government, and has been supported by the Bank Group as the major credit channel for funds disbursed for direct use by farmers. A project to strengthen TRDB has recently been negotiated. Rural Transport Services 1.09 An adequate road transport system is needed to ensure the efficient movement of crops to market centers and railheads, and to support timely delivery of agricultural inputs to the farmers. Despite Government's efforts to upgrade the road network, only about 10 percent of the 33,400 km of roads provide reliable, all weather service. The remainder are, in general, little more than earth tracks impassable in the rainy seasons. Rural roads are in especially poor condition. In addition, the trucking fleet available for use by parastatals (either through ownership or hire) has been limited, thus making crop movement even more difficult. Crop marketing has suffered from high transport costs, and deterioration of crops in villages with inadequate storage facilities. Poor transport facilities cause delays in payment to villages and often dictate that villages receive sums large enough to handle crops over extended periods, leading to inadequate supervision and accounting of funds advanced for purchases. Current GOT action to address these 1/ For a description of the Village Management Technician Program, see the Staff Appraisal Report for the Fifth Education Project (Cr. 607-TA).  - 4 - problems is directed mainly at the public sector, and includes the provision of staff training and equipment under several IDA-assisted projects, 1/ and the recent creation of betterment and maintenance units, which are upgrading selected secondary roads in several regions. Assistance to the public truck- ing industry is being provided through the IDA- supported Trucking Industry Rehabilitation and Improvement Project (Cr. 743-TA). This has not proved sufficient to overcome transport bottlenecks in the critical area of crop procurement and GOT, under IDA assisted projects 2/, has provided parastatals with additional transport facilities. C. Agricultural Marketing and Pricing Marketing 1.10 In May 1976, the Primary Cooperative Societies and Unions, which formerly played an important role in agricultural marketing, were dissolved and their functions transferred to other district, regional and national orgetnizations, of which the most prominent are various parastatal bodies. At present, there are separate parastatal organizations for coffee, cotton, foodgrains, sisal, tobacco, tea, cashewnuts, sugar, livestock and dairy products and pyrethrum; many have been established quite recently. Generally, ambitious expectations have placed severe strains on their physical and managerial resources; they not only provide marketing services to villages but also provide processing services and have responsibilities for increasing production. The marketing system in Tanzania suffers from lack of comlpetition, high costs, poor service and slow payments to farmers. The Government is aware of these problems, and the Standing Committee on Parastatal Organization is reviewing the structure and managerial performance of parastatals. The Bank Group is assisting in this process through its analysis of parastatals invclved in recent and proposed Bank Group loans and credits; i.e. the Cashew- nut Authority, the Tobacco Authority, the Tea Authority and the National Milling Corporation. Pricing 1.11 The control of producer, wholesale and retail prices of major crops is an important tool of GOT policy. Producer prices of all principal export crops except sisal and coffee are controlled as are major food crops except vegetables, fruits, fish and eggs. Responsibility for formulating pricing recommendations rests with Kilimo, whose Marketing Development Bureau (MDB) 1/ These include the Hlighways Maintenance Project (Cr. 507-TA), and road components in the Tabora Regional Development Project (Cr. 703-TA), the Geita Cotton Project (Cr. 454-TA), the Mwanza/Shinyanga Rural Development Project (Cr. 803-TA) and Kigoma Rural Development Project (Cr. 508-TA). 2/ Including a number of trucks to the Tobacco Authority of Tanzania (TAT) under the Tobacco Handling Project (Cr. 802-TA) and to NMC, financed by BADEA (the Arab Bank for Economic Development) under the National Maize Project (Cr. 606-TA).  carries out annual reviews of producer prices for major crops, taking into account production costs, the need for incentives to producers, inter-regional equity and the sales revenues and operating costs of marketing agencies. The MDB, which was established in 1970 with financial assistance from UNDP and technical assistance from FAO, has progressively developed its skills in price analysis and formulation. MDB is currently being financed by IDA, on a provisional basis, under the National Maize Project (Cr. 606-TA) and under the proposed Grain Storage and Milling Project would receive further IDA funds and long-term technical assistance. During supervision of the Grain Storage and Milling Project, the Bank would continue and strengthen its dialogue with Government on pricing. D. Previous Bank Group Assistance 1.12 Previous Bank Group assistance to the agricultural sector in Tanzania has included IDA credits totalling US$145.2 million and IBRD Loans totalling US$37.0 million for thirteen agricultural and three rural development projects. Agricultural projects include one for agricultural credit, three for livestock, two for tobacco and one each for the development of production of tea, sugar, cashewnuts, maize, fish, forestry products and cotton. Rural Development Projects have been financed by IDA in the Kigoma, Tabora, and the Mwanza and Shinyanga regions. Project Performance Audit Reports for two projects (the Flue-Cured Tobacco Project and the Smallholder Tea Development Project) have recently been prepared. In both cases, the economic rate of return is satis- factory. However, implementation experience highlighted the weakness of the parastatals involved, and the consequent need for considerable investment in institution-building. These lessons have been taken into account in the design of this Project. II. THE PYRETHRUM SUB-SECTOR A. General 2.01 The pyrethrum plant (Chrysanthemum cinerariaefolium) is a small perennial growing to some 60 cm in height. It has one of the longest histories of botanical substances used to control insect pests and has been commercially available for 150 years. Its dried flowers yield an extract, the active ingredients of which, pyrethrins, have excellent insecticidal properties with no adverse environmental side effects. The crop is grown in about sixteen countries, five of which, Kenya, Tanzania, Rwanda, Ecuador and Papua New Guinea (in descending order of importance), produce over 97% of the world's output. 2.02 Growing conditions. Pyrethrum growing requires an altitude of 1,800 m or above with a well distributed rainfall of 1,000 mm or more per year and temperatures that fall to levels a little above freezing. Low temperatures stimulate flower production. The plant prefers light, loamy soils, slightly alkaline, but it has a fair degree of tolerance to acidity. It needs an open unshaded position and does not compete well with weed growth.  - 6 - 2.03 Processing. The active insecticidal ingredients referred to as 'pyrethrins' are obtained from dried crushed pyrethrum flowers by means of a solvent extraction process. The crude extract thus obtained is about 25% pyrethrins, the remainder being soluble oleo-resins also extracted from the flowers. The residue from extraction is referred to as 'pyrethrum marc,' which has some commercial value since it contains a low concentration of pyrethrins, as well as being relatively rich in protein and hence in demand as a cattle feed ingredient. Crude extract can be purified in various degrees. Some of the oleo-resins can be removed from the crude extract by centrifuging, producing 'partially-dewaxed' extract. Further purification requires refining by solution in further organic solvents. This produces a decolorized 'pale' extract. 2.04 Properties. Pyrethrum is basically a 'contact' insecticide, that is, one which is absorbed through the body wall. It has a wider spectrum of activity than many other insecticides, and is effective against all common household insects. It has extemely low mammalian toxicity, and its long and impressive safety record in actual use has led pesticide authorities to classify it as one of the very few virtually non-hazardous insecticidal materials. Pyrethrum has the ability both to paralyze flying insects ('knock- down') and to kill them after a relatively short period. Exposure of crawling insects to pyrethrum results in temporarily increased activity; the insects leave their hiding places and are thus further exposed to the chemical. Pyrethrum has a good repellent effect, so that new insects do not enter any treated area. Despite its long use, pyrethrum does not seem to have caused the development of any significant degree of immunity in treated insect populations. Unlike many synthetic insecticides, pyrethrum is rapidly de- composed after use into harmless compounds, especially when exposed to direct sunlight. B. Pyrethrum Production 2.05 In Tanzania, commercial pyrethrum cultivation began in the 1930's, on large farms and estates owned by expatriates in the highlands of the Mbeya and Iringa regions of Southern Tanzania. During the 1950's, production spread to the area around Mount Kilimanjaro, where land in high-altitude forest reserves was leased to large-scale pyrethrum growers. In the early 1960's, smallholders in Mbeya and Iringa regions were encouraged, by the distribution of pLanting material and the provision of extension and marketing services, to begin pyrethrum production. As a result, production of dried flowers expanded from 2,300 m tons in 1963 to over 6,300 m tons in 1967. This expansion coincided with a similar expansion in Kenya, and resulted in world over- production and marketing problems. The consequent reduction in pyrethrum prices, together with rising agricultural wages and the progressive expiration of forest leases, eliminated most of the large-scale producers, and reduced production by smallholders. Thus, overall production of dried flowers fell to 2,416 m tons by 1969/70. Between then and 1974-75, production fluctuated between 2,700 m tons and 4,700 m tons. From 1974/75, however, there has been a steady downward trend in production, from 4,700 m tons in 1974/75, to a low of about 1,600 m tons in 1978/79. This decline resulted from producer prices  - 7 - for pyrethrum that were unattractive in relation to other prices (particularly of maize) and from a lack of improved planting material. Detailed production figures for the period 1963/64 to 1978/79 are given in Table T-1 in Annex 1. 2.06 Over 95% of dried flower production now comes from the highland areas of Mbeya and Iringa, where pyrethrum is grown almost exclusively by smallholders. Pyrethrum production by smallholders has also recently started in Rukwa and Ruvuma regions, where physical and climate conditions are also suitable. No accurate data exist on the number of growers involved, but it is currently estimated that about 20,000 smallholder families cultivate an average of 0.4 ha of pyrethrum each, giving a total area under the crop of about 8,000 ha. A more detailed estimate of the number of growers, area cultivated and dried flower production within each of the major production zones is given in Annex 1, Table T-2. TPB operates two large farms (Mahenye and Ilinde) in the Southern Highlands. Production from these is less than 1% of total production and TPB does not intend to expand these production activities. Instead, these farms would be used under the Project for seed production and as bulking centers (paras. 5.04, 5.08). C. Cultivation, Yields and Flower Drying Cultivation 2.07 The average size of pyrethrum growing farms is about 2 ha. Culti- vation is mainly by hand; only in isolated cases are oxen or tractors used. Aside from the low availability of mechanized or animal power the steepness of the terrain usually precludes anything but hand cultivation. Trials and observations have shown that pyrethrum will come into full flower production in the second and third year and that production rapidly declines thereafter. In view of the soil pulverizing effect of the crop and the hazard of 'root-knot nematodes' it is necessary to uproot every third year, and to rotate with either a grain crop or a period of fallow. Plant propagation is by seedlings raised from seed or by splitting and replanting existing plants. Picking takes place the year round, but there are two flushes of flowers; the main flush is in September-October and the lighter one in February-April. Yields and Pyrethrin Content 2.08 No systematic data series exist detailing the yields of dried flowers. According to a survey of smallholders carried out in 1974 1/, the average annual yield of dried flowers per ha was 270 kg. Since then, staff of the Tanganyika Pyrethrum Board (TPB) estimate that yields have been falling, as a result of adverse climatic conditions and because of aging of plants in 1/ C. Bremer Jespersen, Pyrethrum in the Southern Highlands, 1974.  - 8 - the absence of replanting- Current average annual yields are put at 200 kg per ha, with yields of 100 kg in the first year and 250 kgs in both the second and third years. The pyrethrin content of dried flowers, which is measured prior to processing (para. 2.19), depends on the quality of the planting material, on climatic conditions and on the efficiency -of drying and marketing operations. Over the last sixteen years for which figures are available, pyrethrin content remained fairly stable around an average of about 1.25% until 1970/71 since when it has steadily declined to the 1.1% average overall content recorded in 1978/79. This trend reflects the decline in large scale production, the absence of improved replanting material and, for the last three seasons, adverse weather conditions. Details of average pyrethrin contents since 1962/63 are given in Table T-3 in Annex 1. 2.09 Flower drying. Pyrethrum flowers require drying before they can be collected from farmers for processing. The aim is to reduce the moisture content from around 80% to 10% with minimum loss in dry matter and pyrethrins, which are heat sensitive and subject to oxidation. Rapid drying with a minimum of heat is therefore desirable. Natural sun drying is employed as far as possible. In some areas, however, limited sunshine hours and wet weather for most of the year make natural drying difficult. Indoor drying on beams over the fire is not fully satisfactory but must be used in several areas. Drying in portable driers or barns is much more efficient, reducing the drying time from 12 days to 12 hours, or less. TPB has 50 small, portable driers on loan to villages in the Iringa Region and another 50 in Mbeya. Artifical drying is constrained by scarcity of fuel; pyrethrum growing areas have few trees and the cost of delivering alternative fuels to isolated highland stations is prohibitive. There is also scope for increasing the fuel effi- ciency of driers by design improvements. The Project would provide funds for a study by consultants of the extent of the drying problem and of possible solutions to it, and for construction of additional artificial drying capacity, should this prove feasible (para. 5.11). D. Services to Pvrethrum Growers Resea_rch 2.10 Pyrethrum research was started in the late 1950's at Igeri in Iringa Region. At that time, considerable work in plant selection was carried out by research staff with the assistance of pyrethrum field staff and growers. In 1968, research was halted, following the decline in market prospects (para. 2.05), and it was not until 1974 that research on pyrethrum was resumed. At present pyrethrum research is carried out by staff of the Uyole Agricultural Center (UAC) (para. 3.11) at Uyole itself and at three offstations. The emphasis of this research program is on the selection and multiplication of high yielding plants which have flowers with a high pyrethrin content. Agronomic research, for example on the use of fertilizers, crop rotation, weeding and spacing, is also carried out. The present selection program is understaffed and too small to produce results in a reasonable period; it would be strengthened and expanded under the Project (paras. 5.14-5.16). The status of research on agronomy is generally satisfactory.  -9- Extension 2.11 Pyrethrum growers are provided with a low level of extension ser- vice. The Mbeya regional administration employs about one hundred extension agents to cover over 500 villages, and in Iringa there are about sixty re- gional field staff to cover almost 600 villages. Until 1977, the regional extension staff were solely responsible for pyrethrum extension. Since then, the Tanganyika Pyrethrum Board (TPB) has employed some Assistant Pro- duction Officers (APO's). TPB currently has eight APO's in Iringa Region (of whom three are certificate holders while the remainder have only short-course training) and five APO's in Mbeya (of whom only one is a certificate holder). On average, each APO is therefore responsible for about 1,500 growers. TPB's long-term objective is that these APO's should work through the region's village-level extension workers, and act as specialist pyrethrum advisers. At present, the weaknesses of the regional extension service preclude this. 2.12 This low intensity of extension is in part responsible for the relatively low flower yields obtained in the Project area. However, the problems facing the extension services are manifold and deep-rooted, (para. 1.07) and it would be unrealistic to expect major improvement during the next few years. The Project therefore provides for only a modest expansion of TPB's field staff (para. 5.10) to enable TPB to manage the production and distribution under the Project of improved planting material, and accordingly only modest improvements in flower yield are projected (para. 8.01). Credit 2.13 Since pyrethrum requires virtually no material inputs other than planting material, and since planting material has been provided to farmers free of charge, there has been no need for the provision of production credit for pyrethrum growing. Some pyrethrum growers (the exact proportion is not known, but it is likely to be less than 20 percent) receive credit for maize production through the National Maize Project (supported by IDA Credit 606-TA). Marketing and Processing 2.14 Primary Marketing and Transport. TPB buys dried flowers only from licensed growers. Flowers are purchased on a cash basis. Licenses are issued to individuals cultivating 2 ha or more of pyrethrum, or to groups of farmers who collectively cultivate more than 2 ha. Individual licenses were introduced in 1978/79, to encourage farmers to care for their crop; such license holders can sell their flowers either directly to TPB or through their village. There are about 800 individuals at present with licenses and 300 active groups. A registration number is used to distinguish all flowers produced by the license holder. Most group licenses are held by villages and about 90 are empowered by TPB to act as buying agents and receive a fee on pyrethrum that they handle. To qualify as agents, villages must be registered, have a lockable store and a literate Village Manager or Secretary. Production from villages which are not TPB agents is handled by purchasing clerks operating from about 20 buying centers/stores. Purchased flowers  - 10 - are brought to the buying centers by four-wheel drive vehicles. The buying centers were originally built and owned by cooperative societies which are now defunct and in the hands of a liquidator to whom TPB pays rent. Ihese stores are generally in poor condition. New market centers/stores would be constructed in strategic locations under the Project (para. 5.12). Villages in isolated areas are served by TPB mobile purchasing units. 2.15 Dried flowers are taken in 7-ton trucks from the villages and TPB buying centers to four main stores operated by TPB in Mbeya, Njombe, Chimala and Iringa (Map IBRD 14436) and from there by truck or truck and rail to Arusha for processing. These operations are seriously constrained, however, by the poor state of the district and feeder roads and by poor management and organization of the TPB transport fleet which is, in any case, insufficient. The long haul transport situation will be greatly improved when the new Mafinga extracting plant comes on stream in 1980 (para. 2.16). The other transport problems are addressed under the Project by the provision of trucks and four wheel drive vehicles to TPB (para. 5.13), by the improvement of TPB transport management (para. 5.19) and by a program of road improvement (paras. 5.23-5.27). 2.16 Processing. Flower processing is currently undertaken by the Tanganyika Extract Company (TECO) (para. 3.10) in Arusha. The factory, which was built in 1961, has an installed capacity of around 6,000 tons of dried flowers. Another processing plant with a capacity of about 4,600 tons is under construction by TPB at Mafinga (Iringa Region) and is expected to be operational by mid-1980. When the Mafinga plant starts to operate, the Arusha plant will no longer be required, since practically no production now comes from the Arusha area, and since the Mafinga plant, which is designed for ultimate expansion to 8,500 metric tons capacity, will be sufficient to handle production for the foreseeable future. During negotiations, GOT indicated that, once the Mafinga plant is operation, the Arusha plant would be closed, its Staff transferred to Mafinga, and any flower production from the Arusha region sold as pyrethrum powder, which would be produced by an existing small powder plant and for which there is a local market. 2.17 Refining. Tanzania currently exports crude extract containing 25% pyrethrins. This product has a high wax content and requires refining for use in aerosols. The possibility of constructing a refinery in Tanzania is under consideration by GOT, but a detailed feasibility study has yet to be carried out. While in the long run it is probably desirable for Tanzania to establish its own refining capacity to diversify its market outlets, no marketing problems for crude extract are expected over the next five years, during which pyret:hrum will be in short supply in world markets (para. 7.15). Construction of a refinery should be delayed until flower production has increased sub- stantially, until TPB's financial position (para. 3.09) can support such an irnvestment and until the operations of a new refinery in Rwanda can be evaluated. This refinery is due to come on stream at the end of 1980 and incorporates a technology which is considerably cheaper than available alter- natives.  - 11 - 2.18 Export Marketing. In 1974, responsibility for export marketing of pyrethrum extract and marc passed from Mitchell Cotts to TECO and the General Agricultural Products Export Company (GAPEX), a parastatal. In 1977, TPB assumed responsibility for these activities. Pyrethrum products are sold on a contract basis, at prices announced in advance by TPB after consultation with Kenya and Rwanda, the world's other major producers. The volume of exports over the last five calendar years for which data are available is tabulated below: Tanzanian Extract and Marc Export Volume, 1974-78 Pyrethrum Extract Pyrethrum Marc ------------------ (m tons) ----------------- 1974 86.3 1,286 1975 137.2 2,280 1976 118.4 1,810 1977 95.8 1,658 1978 74.1 2,812 Extract exports have been falling since 1975, as flower production has dec- lined. Exports of marc have been erratic, as sales are less regular, and stocks are often carried over from one year to the next. At the present time, 80% of Tanzania's exports of pyrethrum extract go to the US, and 16% to Canada. Almost all of the pyrethrum marc exported goes to Japan. E. Producer and Export Prices Flower Prices 2.19 The price paid to growers for dried flowers depends on the pyrethrin content of the flowers, as measured when the flowers arrive at the processing factory. At the time of purchase, farmers are paid in cash the price for the lowest grade; a second payment is made as appropriate after the pyrethrin content has been determined. In 1974/75, TPB experimented with payment of a uniform price, irrespective of pyrethrin content. This practice was discon- tinued after a year, since it appeared to have an adverse effect on grower incentives, and a quality-related price structure was reinstated in 1975/76. 2.20 The current grading system provides for five grades, ranging from flowers with less than 1.10% pyrethrins by increments to flowers with pyre- thrin contents in excess of 1.40%. Producer prices are set by GOT (para 1.11)  - 12 - taking into account TPB's sales realizations and costs of operations, and farmer incentives relative to other crops. Recent producer prices are de- tailed below: Grade and Price Structure, 1977/78-1980/81 1/ 1977/78 & 1978/79 Price (Tsh/kg) 1979/80 & 1980/81 Price (Tsh/kg) Pyrethrin Content 1977/78 1978/79 Pyrethrin Content 1979/80 1980/81 over 1.51% 6.50 6.50 over 1.40% 7.50 11.00 1.31-1.50% 5.50 6.00 1.30-1.39% 7.00 9.50 1.21-1.30% 4.21 5.00 1.20-1.29% 6.50 8.50 less than 1.20% 4.00 4.60 1.10-1.19 6.00 7.50 Less than 1.10% 5.50 6.50 1/ These dates refer to marketing years, running from July 1 to June 30. Prices have been increased substantially since 1977/78. In particular, prices of the lowest grade, which accounts for over 80% of flower purchases, have risen by over 60%. At current prices, pyrethrum provides attractive returns to smallholders (para. 8.04). Assurances were obtained during negotia- tions that, in order to maintain producer incentives, the Borrower would continue to review and adjust producer prices annually taking into account pricess, TPB costs and producer prices of other crops. These matters would be periodically discussed by GOT and IDA in the course of IDA's involvemenlt with MDB (para. 1.11). JXPOrt Prices 2.21 Export prices for extract and marc over the period 1974-1979 are given below: Extract and Marc Unit Export Prices, 1974-1979 Extract Marc --- (Tsh/m ton) --- 1974 135,400 766 1975 144,400 725 1976 168,800 744 1977 185,300 879 1978 218,000 836 1979 345,900 1,500  - 13 - The export price of pyrethrum extract has risen by over 150% since 1974, and the price of marc has almost doubled over the same period. The increase in prices has been particularly marked in 1979. These increases reflect the growing shortage of pyrethrum extract and marc on world markets, as production in Kenya and Tanzania, which accounts for about 90% of total world production, has declined (para 7.04). F. Future Prospects Prices 2.22 An analysis of future price prospects (see Chapter 7 of this report for details) concludes that the export price of pyrethrum extract and marc is likely to remain in real terms, at current levels through 1983/84, reflecting continued supply shortages in world markets. For the following five years, prices are expected to fall by 5% per year in real terms as the world supply and demand situation progressively returns to balance. Correspondingly, prices paid to producers for dried flowers are expected to remain at present levels in real terms through 1983/84, and thereafter, in order to preserve TPB's financial position, to decline in real terms. Analysis indicates that to give TPB adequate margins, producer prices need to fall by 2% per year in real terms and would then stabilize in 1988/89 at levels equivalent to about 90% of present producer prices. Analysis of farmer returns (para. 8.03) indicates that pyrethrum production would remain attractive even at these reduced levels of producer prices. Production 2.23 Production of dried flowers is expected to increase from the low levels of 1978/79 as a result of three factors. Firstly, producer price increases for the 1979/80 marketing year are beginning to result in an in- crease in area planted to pyrethrum. Secondly, flower yields have undoubtedly been depressed during the past three years by adverse weather conditions. Thirdly, a pyrethrum development program is being undertaken in Iringa Region as part of a broader program of development in Iringa, supported by the European Community. Under this program, sufficient planting material for up to 4,000 ha of pyrethrum is expected to be made available to farmers over the next three years. The expected future pattern of production, (without Project investments) is expected to be: 1978/79 1979/80 1980/81 1981/82 1982/83 1983/84 onwards Number of growers 20,000 22,000 28,700 29,100 31,100 33,100 Pyrethrum area (ha) 8,000 8,700 11,200 11,800 12,600 13,400 Dried flower yield (kg/ha) 200 235 235 235 235 235 Dried flower prod. (m tons) 1,600 2,050 2,555 2,758 2,961 3,078 Pyrethrin content (%) 1.10 1.10 1.10 1.10 1.10 1.10 By 1983, a total of about 33,000 farmers are expected to be growing pyrethrum, producing almost 3,100 m tons of dried flowers annually.  - 14 - III. EXECUTING AGENCIES A. The Tanganyika Pyrethrum Board Background 3.01 The Tanganyika Pyrethrum Board (TPB) was established by the Pyrethrum Ordinance (Amendment) Act of 1965. Until 1974, TPB functioned solely as the body responsible for licensing growers. In 1974, TPB acquired the shares of the Tanganyika Extract Company (TECO) (para. 3.10) from the former owners, Mitchell Cotts, and thus assumed responsibility for flower processing. In 1976, TPB took over from cooperative societies (which were dissolved by the Government in that year) responsibility for flower purchase from farmers. In 1977, TPB took over responsibility for pyrethrum product marketing from GAPEX and TECO. 3.02 The broad objectives of TPB are the promotion and supervision of the pyrethrum industry, and the development of increased production. Its present activities include the purchase of pyrethrum flowers, the marketing of pyrethrum products, the promotion of research, the operation of two large- scale farms, the development of production and, once the Mafinga extract plant (para. 2.16) is commissioned, the processing of pyrethrum flowers. Organization and Management 3.03 TPB is controlled by a Board of fifteen directors, appointed by the MinLster of Agriculture, and chaired by an Executive Chairman who is appointed by the President of the Republic of Tanzania. The Board includes representa- tives of the Ministries of Agriculture (Kilimo) and Finance and Planning, the Regional Development Directors of Iringa, Mbeya and Arusha Regions, the General Manager of TECO and representatives of growers. Day-to-day respon- sibility for TPB operations rests with the Executive Officer, who is also a Presidential appointee, and to whom all department heads report. 3.04 TPB's headquarters is presently in Dar-es-Salaam. This siting is inappropriate, since it is distant both from the production areas and from the processing facilities. Provision would be made under the Project (para. 5.21) for the head office of TPB to be relocated to Iringa. TPB currently employs 230 persons, and is organized into four headquarters depart- ments and three regional offices; a Manufacturing Department will be added once the Mafinga plant is commissioned (Chart C-1, Annex 1). The Admiinistration Department is in charge of personnel management, publicity and the provision of general services such as procurement. The Finance Department is responsible for maintaining accounts, for controlling cash flows and for moniitoring TPB's overall financial position. The Marketing Department is responsible for sales (both domestic and export) of pyrethrum extract and other products. The Planning and Production Department is responsible for overall planning of TPB operations, for operating TPB's two large farms, and for providing marketing and extension services to producers. TPB's Regional Offices are in Arusha, Mbeya and Iringa Regions; the Iringa Region would have  - 15 - two District Offices, in Mufindi and Makete Districts, once regional and district office accommodation has been constructed under the Project (para. 5.21). 3.05 TPB's current staffing is not commensurate with its responsibili- ties, which have increased markedly since 1974. Weaknesses are particularly evident within the Finance Department, which lacks sufficient qualified personnel and effective accounting systems and controls, and within the Planning and Production Department, which has insufficient qualified technical staff. TPB's staffing, particularly within these two departments, would therefore be strengthened under the Project (para. 5.18). Finances 3.06 Largely as a result of staff weaknesses within the Finance Depart- ment, TPB's accounts are not up-to-date. The last audited accounts relate to the year ending June 30, 1977 (FY77), and accounts for FY78 and FY79 are undergoing audit at present. Appraisal of TPB's current financial status is thus difficult. Summary income statements and balance sheets for FY75 - FY79 are presented in Tables T-4 and T-5 respectively, in Annex 1. Figures for FY75, FY76 and FY77 are audited, figures for FY78 are from draft accounts and figures for FY79 are based on TPB and IDA staff estimates. 3.07 TPB made after-tax profits of Tsh 104,000 in FY75 and Tsh 497,000 in FY76. In these years, TPB's income was derived only from management fees (based on export realization) and a levy (based on TECO throughout) which were paid to TPB by TECO. In 1977, TPB purchased flowers from growers and sold them to TECO for processing. TPB made a small loss (of Tsh 53,000) in that year. In FY78, when TPB began marketing pyrethrum products on its own account, a loss of Tsh 1.2 million (equal to 5% of revenue) was made. These losses resulted from TPB's taking over responsibility for marketing (and thus raising its costs) while producer prices (set by Kilimo) had not been adjusted to reflect this. In FY79, despite rising export prices, TPB is estimated to have a net deficit of about Tsh 3 million, as a result of declining flower production (para. 2.05). 3.08 TPB's net worth (representing the sum of an initial capital grant from GOT, capital grants for smallholder development and for construction of the Mafinga plant, and accumulated net income) rose from Tsh 2.3 million in FY75 to an estimated Tsh 32 million in FY79. TPB had no long-term debt in FY75, but at the end of FY79 had accumulated, Tsh 21.6 million in long-term loans from the Tanzania Investment Bank (TIB) to finance construction of the Mafinga plant. However, the long-term debt to net worth ratio remains satis- factory at 0.6:1. TPB had an overdraft estimated at Tsh 18 million at end FY79, and a current assets/current liabilities ratio of 0.9:1. Overall, TPB's liquidity is thus somewhat strained. 3.09 Projections, in the 'without-Project' situation, of TPB's income statements, balance sheets and cash flows (in current price terms) for the years FY80 through FY90 are given in Annex 1, Tables T-6, T-7 and T-8 respec- tively. These tables have been drawn up on the basis of producer prices,  - 16 - export prices and production figures given in paras. 2.21 and 2.22. Over this period, except for FY80 and FY81 (when losses of Tsh 2.7 million and Tsh 1.1 million respectively are expected), TPF is expected to earn consistently positive net incomes, which total Tsh 19.7 million (in current prices) over the period. The balance sheet is expected to improve. Total n. worth would rise to Tsh 58.3 million, all long-term loans would repaid, and the current assets/current liabilities ratio would remain satisfactory throughout the period. The cash flow statements indicate a cumulative cash inflow over the period of Tsh 49 million which wgould result in TPB's accumulating surplus cash reserves of Tsh 30.9 million, wnich would be available for further investment or distribution as dividends. TPB's future financial prospects are therefore satisfactory. The 'Tanganyika Extract Company 3.10 The Tanganyika Extract Company (TECO) was established in 1962 as a private company controlled by Mitchell Cotts (a British company). TECO's function was to purchase and process dried flowers and to market pyrethrum products. In 1974, TECO was nationalized, and its shares held by TPB on beha:Lf of GOT. TECO-s main asset is an extract plant in Arusha. However, the need for this plant will be eliminated once TPB's Mafinga plant comes on stream, and TECO would thus have no part in Project execution. B. The Uyole Agricultural Center 3.11 The Uyole Agricultural Center (UAC) was initially established at UIyole in Mbeya Region by GOT with assistance from Denmark, Finland, Iceland, Norway and Sweden as an Agricultural Research and Training Institute, intended to serve the regions of Mbeya, Iringa, Rukwa and Ruvuma. Most of the research departments were established by 1973, and by the end of 1975, the training program was fully operational. In 1976, UAC assumed its present legal status as a public corporation, w7holly owmed by GOT. TJAC is controlled by a Board of Directors, appointed by the Minister of Agriculture. Responsibility for the day-to-day operations of UAC rests with the Managing Director, who is a presidential appointee. The objectives of UAC are to carry out a production- oriented program of applied research, to deal with the problems of surrounding regions, and to provide diploma-level, certificate-level and short-course training for agricultural staff. 3.12 The research activities of UAC are carried out by nine research departments, covering crop husbandry, plant breeding, agro-economics agricul- tural engineering, horticulture, plant protection, soil science, pasture, agronomy, and livestock. UAC carries out research work at six offstations (of which three are used for pyrethrum research) in addition to Uyole. The training activities at UAC cover eleven subjects - agricultural extension, agricultural science, agro-economics, agricultural engineering, agronomy, animal production, food processing, home economics, horticulture, land use planning and veterinary science.  - 17 - C. Ministry of Works 3.13 The Ministry of Works (MOW) is primarily responsible for the plan- ning of highway investments and for the reconstruction and maintenance of trunk roads. Maintenance of the secondary and tertiary road network is the responsibility of the regional administrations, which also carry out the maintenance of trunk roads, using equipment and funds from MOW. MOW also gives administrative and technical support to several units set up under externally-financed projects for the improvement and maintenance of feeder roads in rural areas. These include units set up under the IDA- financed Third Highway Project (Cr. 265-TA) and Tea Project (Cr. 287-TA), and feeder road betterment units set up with Norwegian aid. MOW's long-term policy is to disengage itself from involvement in feeder road improvement and maintenance, but the serious problems of organization, staffing, training and lack of equipment that affect regional roads departments make such disengagement infeasible in the short term. 3.14 MOW is organized into six operating divisions: Buildings Division, Mechanical Division, Roads and Aerodromes Division, Manpower Department and Administration-Division, Planning Division and Government Stores. Respon- sibility for road construction and maintenance rests with the Roads and Aerodromes Division, whose Director is a road engineer. The Division suffers from a shortage of qualified staff. Hence, technical assistance staff would be financed under the Project to implement the Project's proposed feeder road improvement and maintenance component (paras. 5.23-5.27). IV. THE PROJECT AREA Altitude, Climate and Soils 4.01 The Project would be implemented in the pyrethrum-growing highland areas of the Iringa and Mbeya Regions of Southern Tanzania (see Map IBRD 14436), which comprise six major zones: Umalila, Mporoto, Uwanje, Ukinga, Ubena and Ludewa. The Pyrethrum growing areas range in altitude from 1,700 m to 2,700 m. Rainfall, which occurs in a single rainy season (December to April at the lower altitudes, November to May at higher altitudes), ranges from about 1,000 mm per year in parts of Umalila and Ubena, to over 1,800 mm in parts of Ukinga. Soils in the pyrethrum growing areas encompass red clays, yellow clays, dark-brown loams of volcanic origin, sandy loams and red sandy soils which are slightly acidic. These conditions are good for pyrethrum cultivation (para. 2.02). Farming Patterns 4.02 Average cultivated area per family on pyrethrum - growing farms ranges from about 1.4 ha in Mporoto zone to over 2.6 ha in Ludewa zone. Other widely-grown crops are maize, wheat and Irish potatoes. Maize is particularly important in the Umalila and Ubena areas. Cropping patterns on a typical pyrethrum-growing farm in each of the main pyrethrum-growing areas are esti- rated to be as follows:  - 18 - Typical Cropping Patterns CroF Umalila Mporoto Uwanje Ukinga Ubena Ludewa ------------------------- Hectares ----------------------- Pyrethrum 0.47 0.34 0.32 0.40 0.53 0.53 Maize 1.21 0.50 1.20 0.53 1.28 0.95 Wheat 0.03 0.31 0.10 0.58 0.11 0.09 Peas 0.43 0.12 0.02 0.07 0.19 0.21 Others 1/ 0.06 0.14 0.11 0.60 0.47 0.87 2.20 1.41 1.75 2.18 2.58 2.65 Transport 4.03 The main transport arteries in Mbeya and Iringa Regions are the Tanzania-Zambia (TAN-ZAM) highway and the TAZARA railway also linking Tanzania to Zambia (Map IBRD 14436). The TAN-ZAM highway has been upgraded to two-lane asphalt standards with assistance from USAID and from the Bank Group under the Second Highway Project (Loan 586-TA and Credit 142-TA). Pyrethrum traffic uses this highway in preference to the TAZARA railway, which is experiencing severe operational problems, and this preference is likely to continue in the future, since the new extract plant at Mafinga is not located on the railway. 4.04 In the pyrethrum-growing areas, a network of district and feeder roads provides access to the villages where pyrethrum is grown (Map IBRD 14386). This network connects with the TAN-ZAM highway and with the paved Nbeya-Lake Nyasa road via local main and regional roads, which are in a relatively satisfactory condition and passable in all weathers. District and f'eeder roads, however, generally consist of bare tracks which can be negotiated only with difficulty in the dry season and which are impassable during the rains. There is extensive rutting and potholing, culverts are missing, and drainage ditches are inadequate and badly maintained. As a result, transport costs are high and vehicle life shorts 4.05 Pyrethrum flowers are produced, and thus require to be marketed, year-round, and yields are in fact highest during the rains. The delays in marketing and processing which occur as a result of the poor condition of the feeder road network lead to a marked reduction in the pyrethrin content of dried flowers because of oxidation. Under the Project, therefore, selected district and feeder roads serving pyrethrum growing areas would be upgraded and maintained to all-weather standards to facilitate pyrethrum marketing (paras. 5.23-5.27). 1/ Mainly potatoes, but also millet, sorghum, beans, sunflower, groundnuts and vegetables.  - 19 - 4.06 Responsibility for the maintenance of all roads in Mbeya and Iringa regions rests with the Works Division of the respective regional administra- tions. However, these are constrained by lack of finance, equipment, work- shops and workshop tools, trained staff and spare parts, and as a result maintenance is only sporadic, and often carried out only on an emergency basis following damage caused by heavy rains and flooding. In both Mbeya and Iringa regions efforts are underway, supported by external financing agen- cies (Norway in Mbeya and the EEC in Iringa), to strengthen the capabilities of the Works Divisions. While such efforts will need time to come to fruition (and for this reason the road improvement and maintenance component proposed under the Project would be administered by MOW), it is anticipated that the regional road maintenance capacities would be sufficient, after the end of the Project's road component, to continue the maintenance of the roads which would have been improved under the Project. Socio-economic Conditions 4.07 Although Mbeya and Iringa regions are, overall, amongst the weal- thier regions of Tanzania (Mbeya ranking fifth and Iringa tenth of the twenty regions, on the basis of per-capita GDP), the pyrethrum-growing regions are relatively poor, mainly because of their isolation. Little systematic data exist on farm incomes in the Project area, but it is estimated, on the basis of cropping patterns, yields and producer prices, that the average annual per capita income (including subsistence) of pyrethrum growers is about Tsh 650 (US$80), and that annual per capita cash income is Tsh 425 (US$50), all figures in 1979 prices. The absolute rural poverty level in Tanzania is about $105 in 1979 prices, so that the majority of pyrethrum- growing families (probably about 85%) live in absolute poverty. V. THE PROJECT A. General Description 5.01 The Project was originally prepared by GOT with assistance from the Bank's Regional Mission in Eastern Africa (RMEA) as part of a larger project, the National Agricultural Development Program Phase I (NADP I). NADP I was appraised by IDA in June 1977. However, it was not possible to complete appraisal of the pyrethrum component at that time because the roads program had not been adequately prepared and there was insufficient data available on the financial status of TPB and on the costs of pyrethrum marketing and processing. During 1978 and early 1979, consultants undertook detailed reviews of these aspects, and project preparation was completed in February 1979 by GOT with further RMEA assistance. 5.02 The Project would, over a five year period, have the primary objec- tive of raising the cash incomes of smallholder pyrethrum growers in the highlands of Mbeya and Iringa Regions by assisting these farmers to expand  -20 - their output of dried pyrethrum flowers through the provision of planting material and improved research, extension, marketing and road transport services. Specificaliy9, the Project would provide for: (a) the expansion of smallholder production through: the production and distribution of improved planting mate.al and, on ar. interim basis, unimproved planting material; the improvement of pest control; the expansion of TPB's field staff; the improvement of flower drying techniques; the construction of additional TPB market centers/stores; and the expansion of TPB's transport fleet; (b) the strengthening of the existing research program of the Uyole Agricultural Center (UAC) through: the construction and equipping of additional laboratory facilities; the funding of additional research staff and staff training; the establishment of three new off-stations; and the financ- ing of research operations; (c) the strengthening of TPBs operations through: the provision of both long-term and short-term technical assistance and stafL training; the construction of new office accommodation for TPB; and the provision of vehicles for TPB headquarters staff; (d) the improvement and maintenance of feeder roads through: the establishment of a fully staffed and equipped Road Improvement and Maintenance Unit (RIMMU); and the financing of the costs of operations of the RIMU. 5.03 Responsibility for implementation of the smallholder production component would rest with TPB as strengthened under the Project. UAC would implement the expanded research program, and the RIMU, administered by and receiving technical support from MOW, would carry out the roads program. The location of proposed Project investments is shown on Map IBRD 14386. B. Detailed Features Smallholder Production Program 5.04 Improved Planting Material. Six existing bulking centers, (three in Mbeya and three in Iringa Regions) would be expanded and four new centers established in Iringa; suitable sites have been identified, and land is available. The centers would be used to multiply selected improved material, obtained from UAC or TPB's two large farms, for distribution to farmers. The distribution and size of the centers reflects the fact that two-thirds of the planting material produced under the Project is expected to be distributed in Iringa Regior. Within Iringa, it is estimated that 75% of the material would go to Njombe and the recently created Makete Districts with the balance of 25%  - 21 - going to Iringa and Mufindi Districts. The planting material to be bulked would be a clone which, although selected twenty years ago, is superior to planting material now in general use. By the end of the Project period, the bulking centers would have access to improved clonal material from UAC. 5.05 Planting material produced by bulking centers would be further multiplied in village nurseries before being distributed to farmers. In Year I of the Project (PYI), 45 ha of village nurseries would be established. In PY2, this would rise to 135 ha and in PY3 and succeeding years, to 200 ha. These nurseries would be located close to villages where planting material is required, in order to reduce transport costs and plant mortality. The size of each nursery would vary according to demand; the average size is expected to be 2 ha. They would not be permanent, but would be relocated periodically in accordance with farmer demand. Each nursery would be managed by a headman, recruited locally on a temporary basis by TPB, and supervised by TPB field staff. TPB has had experience with the operation of village nurseries since 40 ha were established and operated, generally satisfactorily, during the 1978/79 season. The Project would finance the costs of operations of the bulking centers and village nurseries, which comprise the costs of fertilizers, nematocides, pesticides, transport, tools and labor. 5.06 The output of village nurseries would be sufficient to plant 160 ha of farmers' fields in PYI, 610 ha in PY2, 1,080 ha in PY3 and 1,650 ha in succeeding years. This planting material would be treated prior to dis- tribution to ensure that it would be nematode-free, and would be distributed to farmers free of charge, in accordance with TPB's current practice. There is no established credit system for pyrethrum farmers (para. 2.13) and given the low input requirements for pyrethrum production and the scattered and isolated locations of pyrethrum producing villages, it would be prohibitively expensive to establish such a system. Without credit, however, many farmers would not be able to afford the initial cost of establishing pyrethrum plots if improved planting material were charged for, and would therefore continue to use their present inferior material. Production and distribution of planting material through bulking centers and village nurseries would be a permanent long-term program to continually upgrade planting material through the dissemination of improved material selected at UAC. Assurances were obtained during negotiations that TPB would continue to provide this service to farmers after the completion of IDA disbursements. In effect, the costs would be met by a reduction in producer prices, since these are set taking in account TPB revenues and costs. 5.07 Unimproved Planting Material. Substantial quantities of improved planting material would become available only in PY3 and subsequent years, since little improved material is currently available to be bulked. To meet the present demand for planting material in areas where even unimproved material is in short supply, the Project would fund, in PYI, PY2 and PY3 only, a program to produce and distribute unimproved planting material grown from seed. This program would complement the similar EEC program in Iringa Region (para. 2.22).  - 22 - 5.08 In PYI, 40 ha of seedling nurseries would be developed, using seed produced at TPB's Ilinde and Mahenye farms in 1979/80. In PYI and PY2, the Project would finance the production of seed at these farms on 125 ha and 100 ha respectively. The resultant seed would be sufficient to plant 50 ha of seedling nursery in PY2 and 40 ha in PY3. These nurseries would be 1-2 ha in size, an- would be located close to a source of water for irrigation and close to the point of demand. These nurseries would produce sufficient seedlings for 500 ha of farmers' fields in PYI, 600 ha in PY2 and 500 ha in PY3. Planting material would be made available to farmers free of charge. Project investments would cover the costs of fertilizer, nematocides, tools and labor. As a result of this program, it is expected that the number of pyrethrum growers would be increased by 4,000. 5.09 Pest Control. One hundred knapsack sprayers and a supply of in- secticide (probably Sumithion) would be provided to combat "thrips," an insect infestation which occurs sporadically in lower, warmer areas. Only about 5% of farmers' fields are affected in any one year and for this reason, routine spraying is uneconomic. The problem can become serious, however, if outbreaks are not controlled. The sprayers would be held by TPB staff at bulking centers and at selected villages in vulnerable areas. They would be lent to farmers free of charge when needed; farmers would pay for the costs of insecticide. 5.10 Incremental TPB field staff. At present, only four of TPB's Assistant Production Officers are fully trained to certificate level (para. 2.11). To assist in supervision of planting material production and distri- bution, TPB would be allocated by the Ministry of Agriculture a further 16 certificate holders and two diploma holders, to be employed by TPB as field staff. Twelve of these certificate holders and the diplomates would be available, having completed training, in July 1980, and the remaining four cert:ificate-level staff in July 1981. This level of field staffing would be sufficient for implementation of this Project, but would still be less intensive than desirable. GOT has indicated its intention to strengthen regional extension services (para. 1.07), but no decision has yet been made on how this would be done. Assurances were obtairled during negotiations that Kilimo, TPB and the Iringa and Nbeya regional aut_rities, riot later than June 30, 1982, would review and discuss with !iDA the adequacy of extension services for the Project area including the number of extension staff and arrangements for supervision of village level staff, and if necessary, pro- posals for the improvement of such services. Amongst possible ways to improve extension services would be the further expansion of TPB's field staff coupled with the transfer of the costs of operation of these field staff to the Kilimo budget, as has been done recently for other parastatals (most notably, the Tobacco Authority of Tanzania). 5.11 Flower-drying Techniques. Present techniques used for flower-drying in the Project area are sun-drying, barn-drying and drying in small portable driers (para. 2.09). A considerable proportion of pyrethrins is lost with all of these methods. Provision would therefore be made for TPB to obtain from a consulting firm the services of a team of three consultants (a drying  - 23 - specialist, an economist and a draughtsman) to review the needs of the various parts of the Project area, which differ climatically, to assess the economic viability of alternative methods of drying given the limited availability of fuels in the Project area, and to draw up detailed proposals for the imple- mentation of improved drying procedures where necessary. The coxisultai.zs would be employed for a total of about 10 manmonths. Agreement was reached at negotiations that their recruitment would be in accordance with IDA guidelines and that their terms of reference and conditions of service would be acceptable to IDA. The likely outcome of the consultants' studies would be recommendations for the establishment of pilot drying facilities, and a sum of Tsh. 2.0 million (US$250,000) would be made available to implement such recommendations. During negotiations, it was agreed that these funds would be disbursed by IDA only after agreement has been reached between IDA and TPB on the consultants' recommendations and their detailed implementation. 5.12 TPB Buying Centers/Stores. TPB's existing buying centers/stores are often in poor repair, leading to some deterioration and spoilage of dried flowers. The present number and capacity of these centers/stores is not sufficient for safe storage of dried flowers and various inputs during the rains when truck collections can be delayed. Provision would therefore be made for the construction of stores in 12 locations where existing facilities are 4udged inadequate and in two new locations. All the stores would be 90 m wish a capacity of 25 tons of flowers except for Ndulamo which would be 175 m with a capacity of 50 tons of flowers. The stores would be located in Uwemba and Ndulamo, in Njombe District; Ujuni, Matamba and Ipeleke in Makete District; Kilolo and Lupila in Iringa District; Kipanga in Mufindi District; and Santilya, Ilembo, Simanbwe, Ileje, Igoma and Lufilya in Mbeya District (Map IBRD 14386). Construction of these stores would be phased over the first three years of the Project. TPB would continue to operate the four main collecting stores it now operates in Mbeya, Chimala, Njombe and Iringa, whose capacities are adequate. 5.13 Transport. Twelve 7-ton trucks would be provided to TPB for trans- port of dried flowers from TPB's buying centers to its main stores. The trucks would have specially constructed high sides since pyrethrum flowers are bulky in relation to weight. Such trucks are too specialized to be attractive to either private truckers or to regional trucking companies. Five four-wheel drive station-wagons would also be provided to reach pyrethrum producing villages inaccessible to trucks, and to transport buying clerks and other administrative officials. All these vehicles would also be used for input supply, in particular for supply of planting material from the bulking centers (para. 5.05). In addition, 22 motorcycles would be provided for the use of existing and incremental TPB field staff. It was agreed during negotiations that these motorcycles would be purchased by the field staff using loans from TPB, and that TPB would pay these staff monthly operat- ing allowances. Pyrethrum Research Program 5.14 The main focus of the present pyrethrum research program (para. 2.10) is on the selection and clonal propagation of improved planting mate- rial. Promising plants from farmers' fields are collected and put through  - 24 - a seven-year, four-stage process of single-plant observations, line obser- vations, initial yield trials and adaptability trials. The first two stages are conducted at Uyole, W4ith the thi-d and fourth stages also involving three offstations, at Igoma in Mbeya Region and at Kinyika and Igeri in Iringa Region. The first recomviendations of improved selections are rLU expected until 1982, reflecting the fact that research of this nature did not begin until 1974. 5.15 The present research program is understaffed (only one part-time researcher is currently assigned to it). The period between initial collection of planting material and recommendations of improved material is relatively long (seven years); the present number of offstations does not adequately reflect the ecological range of pyrethrum-growing areas, and the program is short of equipment and operating funds. 5.16 The Project would fund the employment of an internationally- recruited Research Officer, who would be stationed at Uyole and who would work full-time on pyrethrum research. A tissue-culture unit (which would include an operations room, a laboratory, an evaluation room and greenhouse) would be constructed at Uyole, and would be staffed by two laboratory technicians financed under the Project. To assist UAC in the introduction of tissue- culture techniques (which would be new to Tanzania) funding would be provided under the Project for UAC to obtain, on a contract basis, long term research support from an institution (probably a university) with expertise in the use of tlhis technique, and to employ, for about four man-months, a specialist to design suitable facilities and draw up a list of equipment required. The introduction of tissue culture techniques would, by facilitating the rapid multiplication of clonal material, shorten by at least two years the time taken to produce research recommendations. A further three offstations would be established (at Santilya in Mbeya, Kitowa in Iringa and at one other location to be determined); these, plus the existing offstations, would be giveni an adequate range of growing conditions. A special unit (staffed by two laboratory technicians) for the speedy analysis of flower samples would be established in the laboratory of the Mafinga extract plant, to provide rapid and accurate feedback to research staff on pyrethrin content of research material. The Project would provide funds for the interchange of plant material with other pyrethrum-producing countries, for visits to other pyrethrum producers by research staff and for training of Tanzanian research staff. Six houses, a four-wheel drive vehicle and six motor-cycles would be provided for incremental staff, and during the Project period, the operating costs of the expanded research program would be met under the Project. Assurances were obtained at negotiations that the research support services would be obtained under terms of reference and conditions of contract satis- factory to IDA, and that GOT would provide, after the period of IDA disburse- ments, sufficient funds for the continued operation of the expanded research program. It was also agreed during negotiations that GOT would give UAC assistance as necessary to acquire sites for the construction of housing and facilities.  - 25 - Strengthening of TPB 5.17 TPB's responsibilities have grown over the past few years and would be further expanded under the Project. To assist TPB to manage these respon- sibilities, a series cf institution building measures would be financed under the Project. 5.18 Incremental TPB Staff. A Financial Controller would be recruited internationally, who would have overall responsibility within TPB for account- ing and financial administration. The present Chief Accountant would serve as the Financial Controller's deputy, would receive training under the Project (para. 5.20) and would be expected to become Financial Controller after a minimum period of five years. Two agronomists would be recruited inter- nationally. A Planning and Production Agronomist would work at TPB HQ under the Production and Planning Manager, with overall responsibility for the planning and management of the Project's smallholder production component, and a Field Agronomist would work in close collaboration with the Iringa and Hlbeya Regional Production Managers, and would assist them in the day-to- day implementation of the smallholder production component, in particular the planting material programs. The costs of employment of these inter- nationally-recruited staff are estimated at US$70,000 per annum, including salary, international travel and benefits. A Transport Officer, a Cost Accountant, an Internal Auditor, a Planning Officer, a Statistician/Evaluation Officer and a Procurement Officer would be locally recruited and financed under the Project. 5.19 Short-term technical assistance. Six man-months of short-term consultancy would be financed under the Project at an estimated cost per man-month (including salaries, overhead costs, fees, international travel and subsistence) of Tsh 75,000 (US$9,000), for a study of TPB's transport management systems and the formulation of improvements. This study would be carried out by staff of the National Institute of Productivity in Dar-es-Salaam, which has expertise in this area. Completion of this latter study and agreement between TPB and IDA on implementation of its recommendations would be a condition of disbursement for trucks, four-wheel-drive vehicles and motorcycles purchased by TPB. It was agreed at negotiations that these consultants would be employed under terms of reference and conditions of employment satisfactory to IDA. 5.20 Staff training. Funds would be provided under the Project for the training of Tanzanian TPB staff. In PY1, all TPB field staff would receive a two-week training course at Uyole. In each succeeding year, refresher courses would be given to half of TPB's field staff. UAC is satisfactorily staffed and equipped to provide such training. Funds are also provided under the Project for TPB staff to participate in both overseas and local training courses and seminars, covering subjects such as administration, finance, marketing, factory management and personnel management. In addition, the internationally-recruited staff (particularly the Financial Controller)  - 26 - wou:Ld carry out on-the-job training of Tanzanian counterpart and subordinate staff. It was agreed at negotiations that, no later than December 31., 1980, TPB would draw up, for discussion with IDA, a detailed training program to utilize the funds provided under the Project. 5.21 Offices and Houses. Under the Project, TPB's head office would be relocated to Iringa (closer to TPB's field operations). TPB's Iringa Regional Office would then move to a new office in Njombe. The Project would also finance the construction of a new district office in Makete, and a new Regional Office block in Mbeya, where existing accommodation is totally in- adequate. In view of the chronic shortage of staff housing in these locations, housing would be constructed and furnished for TPB stafS in Iringa, Njombe, Makete and Iringa. The new Head Office would ue 1150 m in size, the Iringa and Mbeya ReFional Offices would each be 190 m , and the Makete dkstrict office 100 m . A total of ten senior sWaff houses (each of 108 m ), fourteee middle level staff houses (each of 82 m ) and one junior staff house of 60 m would be constructed. It was agreed during negotiations that GOT would assist TPB as necessary to acquire the sites for these houses and offices. 5.22 Vehicles. The purchase of five four-wheel drive vehicles would be financed under the Project, for the use of TPB's Executive Officer, Planning and Production Manager, two Agronomists and Financial Controller, and as necessary by other TPB staff involved in Project implementation. Road Improvement and Maintenance 5.23 The Project would finance the improvement and maintenance of about 200 km of selected feeder roads in the Project area. The roads to be improved (listed in Annex 1 Table T-9 and shown in Map IBRD-14386) were selected by GOT following a survey of over 400 kms of road in the Project area by consul- tants in 1978. Criteria for inclusion in the program included the present road condition, the volume of pyrethrum traffic carried and the economic rate of return on each road. The roads included are roughly equally divided between Iringa and Mbeya Regions. 5.24 The roads would be improved so as to allow 7-ton truck traffic to pass in all weathers. The roads would have a 5.5 m maximum formation width. As far as possible, gradients would be limited to 14% with no restriction on their length; the radii of horizontal curves in mountainous areas are expected to be at least 30 m. Existing bridges would be repaired and new bridges would consist of simple masonry or gabion abutments with sawn and treated timber decks. Drainage would be improved with adequate masonry or cement concrete culverts and side ditches with lead-offs. Generally, roads would have an eartlh surface except in water-logged areas and along sections with excessive gradients where they would be gravelled. Roads completed under the Project would be maintained as part of Project activities until the improvement program was finished. The minimum acceptable maintenance standards would be set to insure that the roads are passable in all weather. This would entail regular repairs to avoid large pot holing and corrugations, grading to retain cambers and shape across road width, keeping side ditches, culverts and lead--offs clear, repair of culverts and bridges and protection of slopes.  - 27 - 5.25 The road improvement program would be carried out, over a three- year period, by force account by a Road Improvement and Maintenance Unit (RIMU). It was agreed during negotiations that the RIMU would be estab- lished not later than March 31st, 1981 under the control of the Director of Roads and Aerodromes in MOW. Machinery-intensive techniques would be used. Labor intensive methods would not be suitable in this case because of the shortage of labor in the mountain areas where roads would be constructed and the high cost of importing labor. The relatively high wages for unskilled labor (US$2 per day) and the dearth of managerial skills would also make the use of labor intensive methods uneconomical in this case. Until the establish- ment of the RIMU, MOW would be responsible for procurement and staff recruit- ment. To allow adequate time for procurement of equipment and for construction of housing, office and workshop accommodation, (which would be handled by MOW, which has adequate expertise in these respects), RIMU management staff would not be recruited until the third quarter of PY1, and road improvement would not start until PY2. The Project would finance the procurement of equipment and spare parts for the RDMU (a full list of equipment is given in Table T-10 of Annex 1). A Roads Engineer, a Construction Supervisor and a Mechanical Supervisor, would be internationally-recruited to manage the operations of the RIMU. The costs of employing these internationally-recruited staff are estimated at US$70,000 per annum, including salary, international travel and benefits. About 135 local staff and labor (a list is at Table T-11 of Annex 1), would also be employed by RIMU. 5.26 The RIMU would be based in Mbeya, which is the major town most convenient to its operations. The Project would finance the construction of three houses for senior RIMU personnel, and a small office and workshop in Mbeya. It was agreed at negotiations that GOT would give MOW assistance as necessary to acquire sites for these facilities. The Project would finance the costs of operation of the RIMU, including the costs of salaries, wages and materials such as cement, steel, lumber, tires, fuel, oil and lubricants. 5.27 Several alternative means of carrying out the proposed road improve- ment and maintenance component were explored during Project preparation and appraisal. There are few private road contractors in Tanzania; they are based in Dar-es-Salaam and do not have the capacity to extend their operations to carry out the proposed program. Because of the scattered character of the road works proposed, and their location in mountainous areas of difficult access, international contractors are unlikely to be interested. Neither TPB nor Kilimo has the management capacity or expertise to be able to carry out the program. The regional works divisions in Mbeya and Iringa, although receiving support from external agencies (para. 4.06) are likely during the next few years to remain unable to carry out such a program in addition to their broader regional responsibilities. However, by the end of PY4, when the road improvement program would be completed, it is anticipated that the Regional roads divisions would, as a result of the external assistance they had received, be in a position to continue routine maintenance of the roads improved under the Project. During negotiations, agreement was reached that, not later than March 31, 1984, GOT would review the capacity of the Regional works divisions in Mbeya and Iringa to take over from RDMU main- tenance of the roads improved under the Project, and would furnish to IDA  - 28 - proposals regarding the responsibility for continued maintenance of the roads improved under the Project, and that RDMU equipment would, after completion of the Project's road component, be used for the maintenance of roads improved under the Project or for the development and maintenance of other feeder roads in Mbeya and Iringa Regions. Agreement was also reached during negotiations that, throughout the Project period, adequate funds would be budgeted by GOT for the maintenance of all main and local main roads used for the transportation of pyrethrum in the Project area, and that adequate funds would be budgeted for the maintenance after the construction period of roacts improved under the Project. Implementation Schedule 5.28 A Chart showing the anticipated rates of implementation of the various Project components is at Chart C-2, Annex 1. C. Project Costs 5.29 The total costs of the Project would be Tsh 109.4 million (US$13.2 million), of which US$7.8 million (60 percent) represents the foreign exchange costs. These total costs include about Tsh 4.0 million (US$0.5 million) of taxes and duties, comprised mainly of import duties on trucks and motorcycles purchased by TPB. The phasing of Project costs is summarized in Annex 1, Table T-12, and costs for each of the components are given in more detail in Annex 1, Table T-13 through T-16. Project costs are summarized below:  - 29 - Local Foreign Total Local Foreign Total Exch % Cost % ------ Tsh '000 ------- ------ US$ '000 ------- Smailholder Production Improved Planting Material 3,435 2,285 5,720 414 275 689 40 7 Unimproved Planting Material 2,466 340 2,806 297 41 338 12 4 Pest Control 113 337 450 14 41 55 75 1 Incremental TPB Field Staff 1,108 --- 1,108 133 --- 133 --- 1 Drying Study and Improvements 880 1,870 2,750 106 225 331 68 3 Flower Stores 1,856 1,076 2,932 224 130 354 37 4 Vehicles 674 3,281 3.,955 81 395 476 83 5 Sub-total 10,532 9,189 19,721 1,269 1,107 2,376 47 25 Research Program Civil works 1,591 1,076 2,667 192 130 322 40 3 Equipment 24 467 491 3 56 59 95 1 Vehicles 26 219 245 3 26 29 90 - Staff 1,030 2,465 3,495 124 297 421 71 4 Operations 1,570 740 2,310 189 89 278 32 3 Training & consultants 622 3,228 3,850 75 389 464 84 5 Sub-total 4,893 8,195 13,088 590 987 1,577 63 16 Strengthening of TPB Incremental Staff 2,555 7,845 10,400 308 945 1,253 75 13 Technical Assistance 350 - 350 42 - 42 - - Training 380 500 880 46 60 106 57 1 Offices and housing 8,108 5,210 13,318 977 628 1,605 39 17 Vehicles 31 589 620 4 71 75 95 1 Sub-total 11,424 14,144 25,568 1,376 1,704 3,080 55 32 Roads Component Establishment of RIMU 1,166 6,456 7,622 140 778 918 88 10 Operations of RIMU 3,934 9,354 13.288 474 1,127 1,601 70 18 Sub-total 5,100 15,810 20,910 614 1,905 2,519 76 27 Total Base Cost 31,947 47,338 79,287 3,849 5.703 9,552 60 100 Physical Contingencies 3,973 5,804 9,777 479 699 1,178 59 12 Price Contingencies 8,283 12,013 20,296 998 1,447 2,445 59 26 Total Project Cost 44,204 65,155 109,360 5,326 7.849 13,175 60 138 Total Project Cost 40205 65,155 105,360 846 7,849 12695 62 133 (net of taxes and duties)  - 30 - Project costs have been estimated on the basis of prices prevailing in January 1980. Physical contingencies have been included at a rate of 20 percent on the costs of civil works and equipment for the research component and at a rate of 10 percent on all other costs except the lump sum for improved dryers. Price contingencies on imported vehicles and equipment and on inter- nationally-recruited staff and consultants have been included at annual rates of 10.5% in 1980, 9% in 1981, 8% in 1982 and 7% thereafter. For local equipment and staff costs, price contingencies have been included at annual rates of 11% in 1980, 10% in 1981 and 9% thereafter. For civil works, higher price contingencies have been included at annual rates of 12% in 1980, 11% in 1981 and 1982 and 10% thereafter, reflecting the pressure on construction industry capacity. Price contingencies total 26% of Project base line costs. D. Financing 5.30 The financing of Project costs, net of taxes and duties, would be as follows: US$ Million % IDA 10.0 79 Government of Tanzania 2.7 21 12.7 100 The proposed IDA credit of US$10.0 million would be on standard terms to the Government of Tanzania and would cover all the foreign exchange costs of the Project (US$7.8 million) and 44 percent of local costs, or about 79 percent of total Project costs. GOT would finance the remaining US$2.7 million,.net of taxes and duties. 5.31 Credit funds and GOT contributions for the smallholder production program (Tsh 27.2 million, including contingencies) would be channelled by GOT to TPB in the form of equity. Credit funds and GOT contributions for the pyrethrum research program (Tsh 18.4 million, including contingencies) would be channelled to UAC on a grant basis, through the Kilimo development budget. Credit funds and GOT contributions for the road component (Tsh 27.8 million, including contingencies) would be channelled to the RIMU on a grant basis through the development budget of the MOW. The remainder of credit funds and GOT contributions for the strengthening of TPB's operational capa- city (Tsh 36.0 million including contingencies) would be on-lent by GOT to TPB for a period of 10 years, including 5 years of grace, at an annual interest rate of 10%, which is currently the interest rate charged by GOT to parastatal borrowers. In the case of TPB, the division of funds between equity and loan is in accordance with GOT's policy that 'non-commercial' parastatal invest- ments be financed by grants or equity, while 'commercial' investments (which  - 31 - are expected to generate a financial return to the parastatal), should be financed by loans. The terms and conditions of this financing plan were agreed at negotiations. Completion, in a form satisfactory to IDA, of an on-lending agreement between GOT and TPB would be a coaditioni of credit effectiveness. 5.32 To ensure that the housing and offices for TPB (which are required for effective Project implementation) and the study of TPB's transport systems (completion of which is a condition of disbursement of funds for vehicle pur- chase) are completed as expeditiously as possible, IDA has urged TPB to enter into contracts, prior to Board Presentation, for the design of the houses and offices and the preparation of tender documents, and for the conduct by con- sultants of the study. It is recommended that part of the costs of these contracts incurred after February 15, 1980, be financed retro-actively out of the proposed credit. The maximum sums likely to be involved would be: Total Retro-active Contracts Financing