Zambia - Integrated Family Farming Project

Document of The World Bank FILE oPr FOR OFFICIAL USE ONLY Report No. 3746 PROJECT PERFORMANCE AUDIT REPORT ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) December 30, 1981 Operations Evaluation Department 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. WEIGHTS AND MEASURES (Metric System) 1 hectare (ha) = 2.47 acres (ac) 1 acre 0.405 hectares 1 kilogram (kg) 2.2 pounds (lb) 1 pound 0.454 kilograms ABBREVIATIONS ADS - World Bank's Agricultural Development Service CDC - Commonwealth Development Corporation MALD - Ministry of Agricultural and Land Development RMEA - Regional Mission in Eastern Africa RVP - Regional Vice President TBZ - Tobacco Board of Zambia UDI - Rhodesia's Unilateral Declaration of Independence CROP YEAR October - September FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) TABLE OF CONTENTS Page No. Preface **............................................................ Basic Data Sheet ...................................................... ii Highlights ............................................................. iv PROJECT PERFORMANCE AUDIT MEMORANDUM I. PROJECT SUMMARY .............................. 1 II. MAIN ISSUES .............................................. 5 A. Project Design ........................... 5 B. Negotiations ................................. 10 C. Project Implementation and Bank Review ............... 11 Overview ............................... .......... 11 The Early Years .................................... 11 The 1976 Review ........................................ 13 The 1976-78 Period ......................... ....16 The 1978 Review ......... ......... 17 D. Supervision ................ ........... .. ......22 ANNEX Comments by the Co-Financier .......................... 26 PROJECT COMPLETION REPORT I. INTRODUCTION ............................................. 35 II. PROJECT FORMULATION.................. o ............. 36 Identification. and Preparation ....................... 36 Appraisal and Negotiations ............. ...............37 Project Description ............. ......................... 38 III. IMPLEMENTATION ................................... 39 Effectiveness and Start-up ...... ........... 39 Scheme and Section Planning .............. ....... 40 Scheme Management and Performance ......... ..... . . 43 Physical Infrastructure .......................... 45 Project Costs ....... ............... .o.. ............ 45 Financing, Disbursements and Procurement .... ....... 47 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. TABLE OF CONTENTS (continued) Page No. IV. AGRICULTURAL IMPACT ........................... 48 Farmer Recruitment .*...................... .............. 48 Area, Yields and Quality .......... ............... 49 Production ... .... . . . ..... * ..... . . .... . ........ 49 Farmer Prices, Payments and Incentives .........o....... 50 Returns to the Farmer .. ..... ..* ....................... 51 V. ECONOMIC RATE OF RETURN .......... ......... . ........ . 53 Rate of Return ......................................... 53 Supervision Estimates of Rate of Return ................ 53 VI. ORGANIZATION AND MANAGEMENT ........ ....... 54 Staffing ........... ...................... ... 54 Senior Management's Approach to Project Concept ........ 56 Financial Management and Accounts ............. 58 Role of Board of Directors ............................ 60 VII. FUTURE OF TOBACCO INDUSTRY .......................... 61 VIII. SPECIFIC ISSUES AND BANK PERFORMANCE ......... ...... 62 Project Design ........................................ 62 Supervision Performance ........... ................... 63 Lessons for the Future ................................. 65 IX. CONCLUSIONS ..................................... 66 Tables 1 - 8 Appendices 1 - 2 MAR IBRD 3935 and Key to Map - 1 - PROJECT PERFORMANCE AUDIT REPORT ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) PREFACE This is a performance audit of the Integrated Family Farming Project in Zambia for which Loan 882-ZA in the amount of US$11.5 million was approved in January 1973. The loan was closed in March 1980 with a delay of nine months. Although the delay was small, the project was only 40% completed by that time with costs estimated 12.5% higher than at appraisal. The final disbursement was made in April 1980 and the remaining balance of US$3.8 million was cancelled. This audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and a Project Completion Report (PCR) dated June 19, 1981. The PCR was prepared by the Regional Mission in East Africa (RMEA) on the basis of a country visit in July 1980. The audit memo- randum is based on a review of the Appraisal Report (PA-151a) dated Decem- ber 13, 1972, the President's Report (P-1163) of January 3, 1973, the Loan and. Project Agreements dated February 27, 1978 and the PCR. The Project Performance Audit Report (PPAR) of the preceding project - Zambia Commercial Crops Farming Development Project, Loan 685-ZA (PPAR, OED Report 2581 dated June 29, 1979) was aLso reviewed. Correspondence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have also been consulted and Bank staff associated with the project have been interviewed. An OED mission visited Zambia in July 1980, coinciding with the RMEA completion mission. Discussions were held with officials of the Minis- tries of Agriculture and Finance, the Tobacco Board of Zambia (TBZ) and with project farmers. The information obtained and observations made during that mission were used to evaluate the conclusions of the PCR and are also reflected in the audit memorandum. The draft report was sent to the Borrower on September 23, 1981 for comments, but none were received. The co-financier (Commonwealth Develop- ment Corporation - CDC) to whom the draft report was also sent, provided detailed comments (see Annex attached to PPAM) through their regional office including the interesting comparision with a CDC-financed smallholder tobacco scheme in Malawi. The PCR is thorough and accurate with respect to analyzing the main reasons for project failure. The audit memorandum discusses in detail the role of the Bank in project preparation and supervision as well as its decision-making during project implementation. OED gratefully acknowledges the valuable assistance provided by Government and project staff, as well as the farmers affected by the project. Their assistance greatly contributed to this report.  - ii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 15.25 19.00/a Overrun (%) - 25 /a Loan Amount (US$ million) 11.50 11.70 Disbursed 11.50 7.73 Cancelled April 1980 - 3.77 Repaid ) - 1.51 Outstanding ) August 31, 1981- 6.21 Co-financing (CDC) 1.7 0.55L_ Date Physical Components Completed 12/78 n.a. Proportion Completed by Above Date (%) 100 40 Lc Economic Rate of Return (%) 20 Negative Cumulative Estimated and Actual Disbursements FY74 FY75 FY76 FY77 FY78 FY79 FY80 Estimated 4.1 6.5 8.7 10.9 11.5 - - Actual - 1.1 2.9 3.9 4.4 4.4 7.7 Actual/Estimated (%) 0 17 33 36 38 38 67 OTHER PROJECT DATA Original Actual or Item Plan Revisions Est. Actual First Mention in Files or Timetable - - 09/68 Government s Application 08/71,d Negotiations 11/72 Board Approval * * 01/16/73 Loan Agreement Date * * 02/27/73 Effectiveness Date 05/28/73 05/28/73 07/26/73 Closing Date 06/30/79 12/31/79 03/24/80 Borrower Republic of Zambia Executing Agency Tobacco Board of Zambia (TBZ) Follow-on Project Name None * Information not on file. /a PCR mission estimate (PCR para. 3.20). /b The balance of the CDC loan was cancelled (in mid-1979). /c Some aspects of physical components completed, such as curing sheds, were directly related to number of farmers recruited (38% of appraisal esti- mates). Others, such as roads, schools and clinics were independent of the rate of farmer recruitment. /d Potential Family Farming Tobacco project first discussed with the Govern- ment in 1968 when Commercial Crops (Tobacco) Farming Development Project (Loan 685-ZA) was identified. - iii - MISSION DATA Sent Month/ No. of No. of Man- Date of Item by Year Weeks Persons weeks Report Identification RMEA 04/71 0.5 2 1 07/71 Preparation RMEA 06/71 2 2 4 07/71 RMEA 08/71 0.5 3 1.5 RMEA 09/71 0.5 2 1 08/71 Appraisal HQ 03/72 4 4 16 12/72 Total 7.5 23.5 Supervision I RMEA 02/73 1 a 1 1 L! 02/73 Supervision II RMEA 07/73 1.5a 1 1.5La 08/73 Supervision III RMEA 01/74 1 /a 1 1 /. 02/74 Supervision IV RMEA/ HQ 07/74 1.5-/ 2 3 a 08/74 Supervision V RMEA 02/75 1.5/a 2 3 /a 04/75 Supervision VI RMEA 10/75 1.5.a 2 3 . 11/75 Supervision VII RMEA 03/76 1.5/a 2 3 . 06/76 Supervision VIII RMEA 06/76 1.5.a 1 0.5.L 07/76 Supervision IX RMEA 10/76 3 3 9 02/77.1 Supervision X .5. RMEA 02/77 2 3 4 03/77.1 Supervision XI RMEA 09/77 0.5 2 1 11/77 Supervision XII RMEA 02/78 1.5 2 3 04/78 Disbursement RMEA 09/78 1 1 1 09/78 Supervision XIII RMEA 09/78 1.5 2 3 10/78 Supervision XIV RMEA 12/78 0.5 1 0.5 01/79 Supervision XV RMEA 03/79 1.5 3 4.5 05/79 Supervision XVI RMEA 11/79 0.5 2 1 12/79 Supervision XVII RMEA 01/80 0.5 1 0.5 01/80 Completion RMEA 07/80 1.5 2 3 03/81 Total 24 46.5 EXCHANGE RATES Name of Currency (Abbreviation) Zambian Kwacha (K) Year: Exchange Rate: Appraisal Year Average US$1 = KO.68 Intervening Years Average US$1 = KO.74 Completion Year Average US$1 = KO.78 /a Missions also supervised Loan 685-ZA at same time. Therefore, only a proportion of supervision mission time spent on Loan 882-ZA. /b Supervision Reports dated February 1977 and March 1977. Full Review report dated April 1977. /c Follow-up to review mission. - iv - PROJECT PERFORMANCE AUDIT REPORT ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) HIGHLIGHTS Following independence, the Zambian Government decided to diversify the economy and raise rural incomes. Accordingly, the Government undertook to expand tobacco production. The Bank supported the Government's objectives with the Commercial Crop Farming Development Project (Loan 685-ZA), approved in 1970. As this project was mainly directed at expatriate and Zambian commercial farmers, the Government independently implemented a smallholder pilot scheme (1970-72). Based on this apparently successful experience, the present project was prepared. The project was appraised in 1972, and the loan (882-ZA for US$11.5 million) signed in July 1973. The project sought to increase production of tobacco and maize through recruitment, settlement and servicing of 6,300 farmers and by provid- ing the necessary supporting services (including credit, input supply, exten- sion, construction of crop extraction roads, primary schools, dispensaries and water supplies, establishing and equipping a project management unit, etc.). Project costs were estimated at US$15.2 million. Due in large measure to intensive management by experienced staff, the project started promisingly but with key management changes in the first year conditions rapidly deteriorated. The Integrated Family Farming Project, as did its predecessor, the Commercial Crops Farming Development Project, failed by virtually every criterion of success. Tobacco and maize production reached only 28% and 30% of appraisal targets; only 2,000 farmers had joined and stayed with the schemes at project closing (or 28% of targets). With benefits much lower and actual costs much higher (US$19.0 million), the rate of return is negative, as compared to 18% expected at appraisal. As measured by sustainability, the project also failed. TBZ failed to develop as an institution capable of fostering tobacco development. The project as implemented has not provided a model for future development. The major lessons emerging from the project's implementation ex- perience are: (i) the Bank should not allow a project to slide downhill with ever increasing speed without attempting to apply the brakes; where project management, or for that matter the Government, fail to make the necessary corrective measures, the Bank should take the necessary steps (such as withholding funds) until such time that the project is brought back on its intended course (PPAM para. 62 and PCR para. 8.05); and (ii) supervision missions should base their recommendations on what is best on project grounds; "country" considerations, even though legitimate during follow-up decision- making, should not influence their recommendations (PPAM paras. 65 and PCR para. 8.06). -v - Other points of interest are: - staffing was crucial for project success; although agreement was reached at negotiations that appointment of selected staff would be a condition of effectiveness and that TBZ would employ at all times qualified and experienced persons, no consultation clause was inserted in the legal documents. (PPAM paras. 22-24 and PCR paras. 2.04-2.05); - although the project was rated a problem project for five years, and thus subject to intensive scrutiny, it continued to deteriorate (PPAM para. 62); and - throughout the project the Bank accepted assurances that something would be done to turn the project around and in this case the Bank clearly accepted such assurances for far too long; this is a critical issue: when should the Bank stop accepting that assurances are given in good faith (PPAM para. 63). - 1 - PROJECT PERFORMANCE AUDIT MEMORANDUM ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) I. PROJECT SUMMARYI/ 1. Following independence in 1964, many expatriate tobacco farmers began leaving Zambia. With their departure, tobacco production fell from 24 million pounds in 1963/64 to 11 million pounds in 1968/69. The Government was concerned about this decline because tobacco appeared to offer the means to diversify the economy, including exports, while simultaneously narrowing the gap between rural and urban income levels. Accordingly, the Government undertook to expand tobacco production. The Bank supported the Government's objectives with the Commercial Crops Farming Development Project, approved in 1970 (Loan 685-ZA), which was designed as a transition program to support expatriate tenant farmers while also training and assisting Zambian farmers to replace them.2 2. The Government had wanted to devote a large part of the Commercial Crops Farming Development Project to small farms, but the Bank objected since there was insufficient experience with small farmer production and the Govern- ment ultimately agreed. However, provision was made under that project for recruiting a specialist: to assist the Tobacco Board of Zambia (TBZ) in plan- ning and implementing schemes under which small farmers would grow a limited area of tobacco (one-acre schemes). This technical component, however, was never implemented.l/ Enstead, between 1970 and 1972, Government implemented independently a pilot scheme and again requested Bank assistance for a larger project. Although only limited information regarding these pilot schemes had become available, and not all of this positive, the Government proceeded with the preparation of the Integrated Family Farming Project, with assistance from RMEA. The project was appraised in February 1972, the loan negotiated in November 1972, and signed in February 1973. The project was co-financed with the Commonwealth Development Corporation (CDC).4/ 1/ Adapted from the PCR. 2/ Project Performance Audit Report: Zambia Commercial Crops Farming Development Project, OED Report No. 2581, dated June 29, 1979. 3/ See PPAR, Zambia Commercial Crops Farming Development Project, op. cit., para. 23. 4/ CDC agreed to finance about 10% of project costs with a loan of E695,000 (PCR para. 2.09). CDC disbursed about US$0.55 million of its loan of US$1.7 million equivalent and the balance was cancelled. - 2 - 3. The project sought to increase production of tobacco and maize, the principal staple crop in Zambia, through: (i) recruitment and settlement of 5,400 new growers, plus servicing of 900 existing maize and tobacco growers;/ (ii) construction of crop extraction roads, domestic water supplies, tobacco baling centers, basic soil conservation, tree and tobacco nurseries, and a limited number of primary schools and dispensaries; (iii) establishment of a project management unit, including housing, vehicles, equipment and a tractor hire service within -TBZ; (iv) production of tobacco seedlings and the provision to farmers of critical inputs like land preparation and ploughing (by the tractor hire service), seedlings, fertilizer and insecticides and general extension advice throughout the tobacco growing and curing cycle; (v) provision of seed and fertilizer for maize; and (vi) provision of short and medium term credit to farmers to permit purchase of seasonal inputs and tools, and the construction of curing barns and grading sheds. 4. As with its predecessor, the Commercial Crops Farming Development project (Loan 685-ZA), the Integrated Family Farming project failed by virtu- ally every criterion of success. In 1978/79, tobacco and maize production2 from the project reached 28% and 30%, respectively of appraisal targets. The number of farmers participating at the close of the project was 3,050 (38% achievement), or only 2,030 if farmers who had already joined the earlier schemes prior to project start-up are excluded (details in PCR para. 4.01). The small number of farmers recruited and the significant drop-out rate among farmers indicate a high degree of farmer dissatisfaction with the program. 1/ The 6,300 growers were to be divided into three schemes, one in southern province, one in eastern province and one in western province. Schemes would comprise about seven sections and there would be three pilot sections to be located in areas where tobacco-growing had not been tried (PCR para. 2.08). 2/ Tobacco production was 1,500 tons as compared with 5,400 tons estimated at appraisal; maize production was estimated at 7,000 tons as compared with 23,000 tons estimated at appraisal. - 3 - Further, while benefits were much lower than anticipated, costs were higher. 1/ The rate of return is negative, compared to 18% expected at appraisal. The project also failed at the task of institution building and manpower training. TBZ's financial difficulties reached their height in early 1979. On the basis of a Tobacco Sector Study financed under a Technical Assistance Project,2/ the Government is now in the process of dismantling much of the institutional and field apparatus of the project. 5. There were five major reasons for the lack of farmer participation: (i) late and inadequate provision of inputs (especially land preparation services) and poor extension by TBZ depressed farmers' tobacco yields and quality, and thus profitability; (ii) it proved harder than expected at appraisal to convert the subsis- tence farmer recruits into efficient tobacco producers as tobacco is an exacting crop, and the farmers recruited had little previous experience with modern farming methods; (iii) a declining tobacco price_/ relative to other crop prices, par- ticularly maize, caused many farmers to switch away from tobacco. Others grew the minimum amount of tobacco required to qualify for project inputs which they applied largely on their maize; (iv) TBZ frequently failed to pay farmers until 6-8 months after they had delivered their crop (due to TBZ's mismanagement and poor liquid- ity), further discouraging producers; (v) all TBZ charges for growing maize as well as tobacco were deducted from the proceeds of the farmers' tobacco sales, and as the farmer was left with little cash from his crop, tobacco appeared even less profitable than it actually was. 1/ Although it has not been possible to obtain accurate data on project expenditure (PCR paras. 3.19 and 3.20), project costs were estimated at K14.8 million (US$19.0 million) as compared with K10.9 million (US$15.2 million) at appraisal, or 35% increase in Kwacha terms or 25% increae in US$ terms (PCR para. 3.20). The Bank disbursed US$7.73 million from its loan of US$11.5 million and CDC disbursed about US$0.55 million from its loan of US$1.7 million equivalent. Therefore, total donor assistance amounted to US$8.3 million so that some US$10.7 million was injected into the project from other sources, considerably more than the US$2.1 million Government contribution anticipated at appraisal (further details in PCR para. 3.22). 2/ Zambia Technical Assistance Project (Credit 873-ZA, US$5.00 million, approved December 21, 1978). 3/ A major change in TBZ marketing strategy in 1977 contributed to the price decline in that year, and perhaps thereafter (see PPAM para. 43). -4- 6. Throughout the project, the Bank was aware that TBZ management, both at headquarters in Lusaka and also at the section and scheme level, was generally inadequate; that inputs and extension were not provided effectively to project farmers; that equipment, vehicles and staff were used on other projects within TBZ; and that financial and managerial control of TBZ was sorely lacking. During project implementation, TBZ was in breach of virtually every covenant of the legal documents. These violations included the failure to: (i) submit annual reports; (ii) inform the Bank each year of the program for scheme developments so that these could be mutually agreed; (iii) establish a separately accounted-for revolving fund for farm credit; (iv) utilize funds made available solely for purposes of the project; (v) submit documentation to Government to permit application for reim- bursement; and (vi) maintain adequate financial records to reflect its operations and financial condition in accordance with consistently applied account- ing practices. 7. TBZ's financial affairs were mismanaged throughout the project. There was little or no effective financial control and most audits reported that books had not been properly kept or that the accounts did not give a true and fair view of the state of TBZ's affairs. The state of the accounts was so inadequate that the Bank was unable to obtain data on actual project expendi- ture incurred on the project until March 1979, following the employment of an economist by TBZ. The data provided by TBZ's economist showed that costs were much higher than previously thought, largely as a result of gross overstaffing at scheme level and inefficient use of human and physical capital. The overstaffing appears to have gone unnoticed by Bank supervision missions because of the failure of TBZ to provide annual plans, project costs, and evaluation reports as required in the legal documents (see PPAM paras. 60-66 on the subject of supervision). 8. Most of these severe problems became evident at an early stage and the project was declared a "problem" project in 1975 and remained so until closing. Bank actions focused on strengthening the quality and leadership of scheme managers, who were directly responsible for provision of inputs and services to farmers. Supervision missions believed that the most effective means of strengthening scheme management would be to reinforce senior manage- ment within TBZ, but this proved very difficult. The Bank regularly obtained signed and verbal agreements on issues discussed, but little action on these agreements ever ensued (see PPAM para. 63). - 5 - 9. Nevertheless, the project continued until its closing in June 1979,.1 at which time implementation was halted, but the closing date was extended, first to December 31, 1979, and ultimately to March 24, 1980, to enable TBZ, with the help of financial consultants, to update its financial accounts and review and justify already submitted reimbursement requests for project expenditures prior to the original closing date. Although TBZ had spent more funds on the project than anticipated at appraisal, the outside auditors were unable to substantiate these expenditures to fully draw down the loan, and nearly US$4 million were ultimately cancelled. II. MAIN ISSUES A. Project Design 10. The project's main emphasis was on tobacco production. It thus formed part of a larger effort to reverse the decline in Zambian tobacco production which had taken place since 1964. TBZ had been created in 1967 as a parastatal with responsibility for marketin 1 and processing tobacco, including operation of the tobacco auction center,- and for tobacco devel- opment, training and extension programs. TBZ hoped to first increase tobacco production through the Tenant Farmers Scheme, utilizing expatriate farmers as tenants on vacated tobacco farms with financial assistance from the Government and, at a later stage, augment this production by developing: 1/ CDC formally proposed to the Bank in early 1978 that project disburse- ments be suspended because of TBZ's generally poor performance, the unwillingnes of management to take effective action despite repeated past agreements, and the strong qualifications given by the auditors of the 1975 and subsequent accounts. The Bank declined, feeling that continuing to work with TBZ and the Government would be more constructive since a new Minister of Agriculture had recently been appointed who, as the Region points out, was well known to the Bank and who was expected to deal effectively with TBZ problems. CDC accepted this position (see also PPAM paras. 46 and 53-54). 2/ Prior to independence, most of the tobacco grown in Zambia was sold through auctions in Salisbury, Zimbabwe, but the Government felt this was improper. Accordingly, a large auction center was built in Lusaka in 1965. The center was designed for efficient operation at a throughput of 50 million lb., in the expectation that domestic output could be in- creased and that additional buyers could be attracted to Zambia, thus increasing the tobacco prices paid. Instead, when total tobacco produc- tion fell to much smaller levels, the center operated very inefficiently and the prices paid to producers declined. TBZ assumed, as part of its capital structure, the debt caused by financing the auction center and had also to service this debt. However, TBZ could not break even on its marketing operations unless tobacco throughput reached 44 million lb., and thus required a continuing subsidy for this operation. - 6 - (i) aw farms on which Zambian Assisted Tenant Farmersl/ were to be settled; (ii) production schemes managed and operated by TBZ; and (iii) Family Farming Schemes. 11. The soil and climatic conditions for tobacco are excellent in much of Zambia, but the crop requires high technical and managerial skills to prepare the soil, grow seedlings, transplant, harvest and cure. Each step must be followed precisely, and at the proper time, if disease is to be prevented and high yields and quality achieved. If each step i carried out correctly, the results in Zambia are indeed excellent, but minor errors reduce profits significantly. 12. In most tobacco producing countries, tobacco growing skills are learned through long experience. In Zambia, where a large number of small- holders, most of whom were only marginally literate and inexperienced at any sort of modern farming, were to be the producers, only highly qualified management and extension would have permitted acceptable levels of output. These levels of management and extension were expected at appraisal, but in the audit's view this was optimistic given the magnitude of the project to be implemented, TBZ's limited staff, and TBZ's other activities which also re- quired staff inputs. This judgment is discussed in the following paragraphs. 13. In the early years, following independence and with high copper prices prevailing, TBZ maintained a managerial staff of about 60, of which about 25 were expatriates. Nonetheless, although the Mukonchi Training Unit2/ and TBZ staff were considered to be of reasonable quality until 1972, early supervision missions for the Commercial Crops Farming Development Project reported already that there was scope for substantial improvement at TBZ headquarters, particularly in planning and management. These were pre- cisely the qualities most needed to establish the Integrated Family Farming Project. The PCR of the Commercial Crops Farming Development Project (which began in 1970/71) states, for example, that "since inception, the main theme that ran through supervision reports was the overall weakness of TBZ's manage- ment, the lack of control, the shortage of well qualified staff, and the inability of TBZ to attract or recruit expatriate staff."'3/ Thus, TBZ staff appear to have been stretched thin before the Integrated Family Farming Project was implemented. 1/ This scheme, as well as the Tenant Farmers Scheme, was assisted by the Zambia Commercial Crops Farming Development Project, PPAR, o2. cit. 2/ The training scheme where Zambians were trained over five years in the growing of tobacco. 3/ PPAR, Zambia Commercial Crops Farming Development Project op. cit. para. 5.03. Review by RMEA of supervision reports indicates, however, that this was particularly true after 1973. - 7 - 14. The future of the expatriates, upon which TBZ management was large- ly dependent,.!/ was also uncertain. With the decline in the price of copper in the early 1970s, the allocation of public funds became inadequate to meet the increasing requirements of the schemes being developed, and at the same time TBZ continued to incur deficits for the auction and processing center. The salaries and fringe benefits of civil servants, particularly expatriates, began to decline. Heightening political tensions in Southern Africa exacer- bated the situation, making it appear possible that many expatriates might depart, which would further weaken TBZ's management. 15. The project was quite large, hoping to involve 6,300 tobacco grow- ers, to be dispersed over six different locations (three main.schemes plus three pilot schemes) in previously underdeveloped rural areas. Although some consideration was given to decentralized decision-making and implementation, TBZ headquarters in Lusaka retained effective control over basic planning and management decisions and was bound to have difficulties maintaining adequate control over operations within such a dispersed project.2! 16. The project was designed for land preparation to be carried out by TBZ through its scheme-based tractor hire services. The intensive use of machinery imposed a particularly heavy burden on management, which subse- quently proved inadequate to this task, and also sharply increased project costs, both fixed and variable. Tractor hire services had been extremely difficult to organize and manage in nearly all of the Bank-assisted projects in which they had been included, because of a general lack of experience and absence of maintenance facilities. 17. One reason why the mechanized approach was adopted in the Integrated Family Farming Project (as opposed to similar projects implemented in the same period in Malawi and Tanzania in which mechanization was minimized), is that this approach had been used previously in Zambia for various smallholder development schemes which had been planned and financed by both Government and 1/ The appraisal repor: (Report No. PA-151a, dated December 13, 1972, para. 5.07) indicated that "there was an acute shortage of suitably trained Zambians for the senior supervisory posts, such as scheme managers. Virtually, all senior personnel in TBZ are currently expatriate and it seems improbable that suitable Zambians could be found to fill all these posts." Indeed, Bank staff indicate that relatively few Zambians were experienced tobacco farmers and these found private farming much more profitable than work for TBZ. 2/ It should be noted that the Government, in its project preparation report, had proposed that 9,500 farmers be developed, but this number was reduced by the appraisal mission to 6,300 farmers. - 8 - bilateral aid agencies..1/ Government had been able to afford this approach, it was associated with progress, the Ministry of Agriculture had become accustomed to it and, not least, smallholders had come to expect it. Thus, the design was entrenched whether or not it was appropriate technologically or economically. It is still somewhat surprising that so little analysis of possible alternatives was carried out, and insufficient consideration given to how the Bank ought to react in case TBZ was unable to effectively manage the machinery. TBZ utilized many tractors in its other schemes, including those of the Commercial Crops Farming Development Project, and here both management and maintenance had proven inadequate. 18. Equally important, the mechanized approach did not seem to support the view that tobacco was well suited to smallholder production. It was understood by the appraisal mission that Zambian smallholders would not be willing or able to produce tobacco if they had to prepare their land by hand. Thus, either oxen were needed (with which many smallholders had experience, and which in any case would have been managed directly by the smallholders), or tractor hire services had to be provided by the project. Selection of the latter approach required that family farmers were highly dependent on TBZ inputs and services, which ultimately discouraged smallholder initiative. --In fact, many farmers subsequently felt they were employees of TBZ, rather than independent entrepreneurs, with negative effect on their efforts. And when TBZ was unable to provide the services on which project development depended, results were correspondingly bad. 19. The appraisal report indicates that the experience gained from the small pilot scheme initiated by the Government in 1968 led to the preparation and appraisal of the project. TBZ had independently started a smallholder scheme in 1968 and had 3 years of experience when the project was appraised in 1972. The appraisal mission thus had a better technical basis on which to build than was usually available for projects in sub-Saharan Africa. The first crops marketed from the pilot smallholder scheme were of very high quality and fetched prices better than the average for commercial farmers, due to good TBZ management of these pilot schemes. Crop budgets at the time of appraisal indicate that the returns to tobacco were better than to maize both per hectare and per man-day, assuming good quality tobacco was produced at the yields and total area projected at appraisal. However, yields were assumed to rise from 650 kg/ha in year 1 to 1,000 kg/ha from year 5 onwards, while area planted simultaneously were assumed to increase from 0.5 ha to 1.0 ha per farmer. 1/ A tractor hire service had been run at a profit for several years in one region of Northern Province and although this situation was nearly unique in Zambia, the manager responsible for the scheme was in charge of TBZ's smallholder tractor operations at the time of appraisal. - 9 - 20. The two latter assumptions were optimistic. The physical achieve- ments under the pilot project prior to appraisal were as follows: ONE ACRE SCHEME FARMERS Area Production Yield No. of Growers (ha) ('000 kg) (kg/ha) 1968/69 28 11 9.4 825 1969/70 46 45 28.8 643 1970/71 312 157 100.5 641 1971/72 539 238 n.a. n.a. Source: Annex 6, Integrated Family Farming Project Appraisal Report (Report No. PA-151a, dated December 13, 1972). The appraisal report indicated that 1971/72 yields were expected to rise to more than 850 kg/ha, after two poor years. Nonetheless, the assumptions utilized reflected an improvement in output per farmer (yields and area) which had not yet been achievedl/ and such improvements over a substantially expanded number of holdings. The audit doubts that this was realistic given the high level of management and extension services which this would have required, TBZ-s limited staff, and obstacles to any expansion or improve- ment in TBZ staff due to the Zambian political and economic situation then prevailing. 21. In summary, the audit feels that the bulk of the evidence then available suggested that TBZ was incapable of planning, implementing, and managing a project of the size and complexity contemplated.!/ The appraisal placed too little emphasis on the difficulty of working within Zambia, still newly independent, where the governmental structure was relatively weak and undercoordinated, where political concerns created strong pressures to change previously conceived plans, where the future of qualified expatriate per- sonnel, considered essential to the project, was uncertain, and where lack of mechanical experience made the tractor hire service planned of improbable success. Moreover, the choice of tobacco for a small farmer crop was wrought with greater difficulty than the appraisal report suggested. The project 1/ Poor crop handling, including storage, was an additional problem to be faced. Significant losses on the pilot one-acre schemes were reported in the appraisal report: (Annex 6) and, during this same period, the Assisted Tenants under the Commercial Crops Farming Development Project had been unable to make a profit despite yields substantially above those expected at appraisal (and also above those achieved by smallholders in the pilot scheme) because of poor crop handling, which resulted in low quality tobacco. 2/ The Region does not understand this finding in view of the positive experience of the small pilot scheme, as indicated in PPAM para. 19 above. - 10 - might ha c;ucceeded had it been implemented along the lines proposed (as the -.k augg:sts para. 8.01), but the probability that it could be so implemented was very small, given the factors mentioned above. A project focusing more on training and institution building, with a much smaller production component, and undertaken as a pilot effort to determine the efficiency of smallholder tobacco production, carefully monitored to obtain exact data on production costs (e.g., with and without mechanization) therefore would have been more appropriate. The Government, with Bank approval, attempted to do too muchi!, ultimately to Zambia's loss rather than benefit. B. Negotiations 22. Following appraisal, the Bank learned that the General Manager and Production Manager of TBZ had been replaced_2 and that other expatriates were expected to leave-3/. The appraisal mission had noted previously that "should this happen, the success of the project would be in jeopardy". After these new developments, the mission proposed that the appointment of five senior staff be made a condition of effectiveness and also that these staff must be acceptable to the Bank, in order to ensure that senior positions at TBZ were filled with competent staff. Such loan conditions had commonly been utilized in many projects during the 1960's, but following a discussion of their effectiveness, the Bank had moved away from such covenants at the time of appraisal. However, given the seriousness of the staffing issue for the Integrated Family Farming Project, the Bank decided to incorporate these proposed employment conditions in the draft documents to be negotiated. The audit mission was informed, however, that discussion of the issue within the Bank was considerable. 23. The Zambian delegation pointed out that the appointment of the TBZ General Manager was the prerogative of the President of Zambia and that all 1/ Two reasons mentioned to the audit mission by some Bank staff for the project's large size (although not found in the appraisal report) were the need for a large tobacco throughput to cover the fixed administra- tive costs of the smallholder project and provide increased utilization of the auction center which had been built, thus reducing its operating deficit and, hopefully, achieving higher tobacco prices by attracting more buyers to Zambia. 2/ This was contrary to the agreements for Loan 685-ZA which required that such appointments be mutually acceptable to the Bank and the Borrower. 3/ The border with Rhodesia was closed in 1972 and many expatriates, espe- cially those with children attending schools in Rhodesia, found it difficult to remain in Zambia; the cost of living also rose as a result of this border closure, reducing the real wages paid. An internal Bank memo dated July 27, 1972 expected a further eleven expatriates to leave in the next 12 months. - 11 - other TBZ officers were to be appointed by the TBZ Board. The delegation indicated that the Government was unwilling to accept any interference by the Bank in these sovereign rights. The Bank, after additional internal consulta- tion, partially accept.ed the Government's position. Agreement was reached that appointment of selected staff would be a condition of loan effectiveness and that TBZ would employ at all times qualified and experienced persons. 24. The PCR (para. 2.05) voices doubt whether reservation of the right to approve staff by the Bank would have significantly affected project imple- mentation. The audit is not certain. Hindsight indicates clearly that the staffing issue was crucial to the project, and it had been so recognized previously. From evidence available, the Bank felt that Zambia would refuse to go ahead with the project if the Bank held firm on the staffing issue and decided that the risk should be taken to proceed nonetheless. It might well have been better to confront the Government on this issue, and to accept the ossibility of not continuing with the project, given the issue's importance. C. Project Implementation and Bank Review Overview 25. Project implementation started reasonably but rapidly deteriorated through time. Farmer recruitment, tractor hire services, the provision of other inputs, TBZ extension work, and tobacco production were always well below appraisal estimates. TBZ-s financial and production records were practically non-existent and of little use as a management tool. TBZ failed to meet nearly all of its agreements with the Bank, including both the Loan and Project Agreements major covenants, and also the successive agreements worked out during implementation. The details of these failings can be noted in their chronological sequence by reference to specific periods: (i) the early years; (ii) the 1976 review; (iii) the 1976-78 period; and (iv) the 1978 review. The Early Years 26. The first two supervision missions reported on results obtained by those farmers who had been recruited before the project's implementation. Thus, recruitment was not an issue to be considered. The yields obtained in the first year of the project, 1972/73, were reported to be good, but this was 1/ The Region disagrees. Its experience in Zambia and elsewhere, in the agricultural sector as well as in other sectors, indicates that insisting on any formal right of prior approval of staff appointments provides no additional safeguards and can often be quite counterproductive. The Region has had problems in other projects even where this formal right is - provided for, and it has had no problems in other cases where no such right is provided for. The Region would thus not wish to generalize from this particular case of lack of performance. - 12 - largely explained in the supervision mission report to have resulted from unusually good climate. No evidence was available from TBZ regarding leaf quality or overall profitability to the farmer, due to TBZ's failure to maintain adequate financial or production accounts. Both supervision missions reported that TBZ management was bad and worsening and that TBZ costs were high. 27. The situation became clearer in August 1974, when the supervision mission reported that physical progress had fallen behind appraisal schedule, that crop results were worse than had been expected (due to adverse weather), and that TBZ had been requested to furnish to the Bank detailed crop results of the 1973/74 operations (as these were not available), and also submit a "detailed program for developing project farms during 1974/75, supported by financial projections, as required by Section 3.02 (a) and (b) of the Loan Agreement." 28. TBZ had begun to implement schemes throughout the country on its own initiative, some of which were assisted with project funds and some with Government funds. The development program as afterwards implemented often diverged widely from the appraisal since location of some schemes was deter- mined unilaterally and often for non-economic reasons. Although the Project Agreement required that TBZ submit its development program to be mutually agreed with the Bank, precisely to control this, TBZ never submitted such a program, either in September 1974 or thereafter. 29. In February 1975, a supervision mission found that farmer recruit- ment was lagging even more and that experience thus far suggested few family farmers would be capable, even after five years, of producing the tobacco crop expected at appraisal. TBZ management at all levels was again reported weak and deteriorating. The mission rated the project a problem project, and recommended that a major review be carried out. The next supervision mission in October 1975, reported that conditions were improving.!1 As a result, the major review was delayed. Only after two subsequent supervision missions confirmed the deteriorating project condition (in March and June 1976), was the review actually carried out (October 1976). Given the general evidence of deteriorating management in TBZ, and the recognized importance of management for projected achievements to be realized, the audit believes that more prompt review of the project was merited. Even with the slight improvement noted by the October 1975 mission, the situation was bad. 1/ Although status was rated as improving, the project was still rated as a problem project. The same mission noted that the revolving credit fund had still not been established, that the cost benefit review had not been submitted, that most expatriates were unwilling to continue service, and that reporting from the field required improvement. The same mission also supervised the preceding Commercial Crops Farming Development Project and rated its status as deteriorating. - 13 - The 1976 Review 30. Several issues were discussed among the Bank, TBZ, and Government during the October 1976 review. First, the Bank believed that faulty imple- mentation was due principally to TBZ's staff weakness, particularly at head- quarters. The Bank thus attempted to convince TBZ and the Government that TBZ's management should be improved, and that this could best be done by hiring more qualified staff. In practice, the Government had to be convinced to hire more expatriates since there were few additional Zambians who could be attracted to TBZ. At this time, however, it was known that the Mwanakatwe Commission had recommended decreases in civil service salaries throughout Government, including a reduction in education and other fringe benefits of expatriates. These decreases were to be particularly severe in the Government parastatals, like TBZ, which had previously been allowed to pay salaries higher than those in the regular civil service. These decreases also would come on top of the erosion in benefits and working conditions which had already occurred, and which had been the reason for the resignation by many of TBZ's qualified expatriate and Zambian staff. 31. The Bank requested the Government to allow TBZ to recruit at pre- Mwanakatwe conditions. The Government was not eager to do so, but exceptions had been allowed for expatriates hired in the copper industry and the Govern- ment ultimately agreed to allow TBZ to do so as well.Y. This agreement largely satisfied the Bank. The audit believes that this agreement was insufficient to guarantee the management on which the project depended. TBZ had been unable to recruit expatriate staff at the salaries previously paid and it was hardly reasonable to expect that the Government's acceptance of pre-Mwanakatwe salaries, however welcome, would allow TBZ to replace the lost staff and arrive at adequate management levels. In addition, because resources in Zambia were scarce, the Bank should have anticipated that the Government would, even after this agreement, attempt to reduce expatriate salaries through time, as it subsequently did. Thus, the agreement reached was only temporary and in any case too late. 32. Second, the Bank was concerned with the failure of TBZ to recruit sufficient farmers for the project. The project's fixed start-up costs for items like housing, buildings, vehicles and tractors, specialized equipment, dispensaries and schools, and project management were high, and without the expected number of farmers and tobacco acreage, tobacco production would be insufficient to achieve an acceptable rate of return.2/ 1/ While the Bank was convinced that additional qualified, by necessity expatriates, staff were needed, the Government was disappointed with the performance of many of them feeling that they were insufficiently productive to justify their high cost. No good analysis was available to suggest whether this view was correct or not, but the issue has never been openly discussed with the Government. 2/ The review mission estimated that the return had declined to 6% as a result of slower recruitment and poorer yields. Even this estimate was based on optimistic assumptions as the Bank did not have adequate figures on project costs, tobacco acreage, tobacco yields, or leaf quality, all of which were overestimated as it subsequently turned out. - 14 - 33. ihe Bank, attributing slow farmer recruitment to insufficient effort Eicouraged TBZ to recruit more aggressively. TBZ agreed, but the Bank's diagnosis was incorrect. Recruitment was slow largely because of the deter- iorating profitability of tobacco compared to that of maize and other crops, particularly on a per day worked basisl/, and many farmers became dissatis- fied with tobacco and either withdrew from the project or shifted their emphasis to maize. Tobacco was less profitable partly because the price of tobacco had fallen relative to that of maize, partly because TBZ's poor performance in providing inputs and technical assistance to farmers reduced their tobacco yields and leaf quality. Thus, an effort to increase the number of settlers at that stage was likely to be counter-productive as, under the circumstances, a larger number would further dilute TBZ's resources and reduce its capacity to provide the services needed. The lengthy delay in TBZ pay- ments to farmers also made tobacco less attractive. 34. In addition, the Bank placed emphasis on total recruitment, rather than on recruitment per scheme. This, in the audit's view, wrong emphasis also contributed to increased unit costs of tobacco produced, as it offered TBZ additional justification for the proliferation of schemes, preferred for political reasons (see PPAM, para. 37); i.e., TBZ could argue that it was attempting to bring additional farmers into the project. Already by 1973/74, supervision missions indicated that the number of schemes had expanded to 12 and the project finally ended with 24, as compared with the appraisal plan for six. Similarly, while the appraisal report planned for 24 sections, the project reached 60.1/ Given this rather substantial departure from appraisal design, and given the management problems faced by the project, the Bank, in the audit's view, should have requested TBZ to sharply curtail the size of the 1/ Already in 1975, a Bank mission warned that in the light of increased maize prices there was a risk that farmers would neglect tobacco growing. After this date, it appears that maize was progressively more profitable. In 1980, a Bank mission estimated the relative profitability of alterna- tive crops in Eastern Province, the best tobacco area, as shown below: Virginia Tobacco Maize Cotton Sunflower Groundnut Net return K/ha 414 202 196 94 270 Net return K/man-day 0.53 1.5 1.4 1.1 1.2 The return per man-day is the more relevant statistic insofar as land is not scarce in Zambia, generally held on communal basis and allocated on an as-needed basis. 2/ The Bank had established no criteria to justify the inclusion of a scheme or section in the project, nor did it insist on concentrating them in certain areas (as it had in other countries). - 15 - project and concentrate the available resources on a reduced number of schemes and sections, even smaller than that planned at appraisal.Y 35. Third, TBZ still failed to comply with most of the legal covenants. Its financial and project accounts were in dismal condition, so that the Bank did not really know which schemes were part of the project and which not, what was being financed, or where expenditures were made. The General Manager explained to the review mission, as he had frequently explained to supervision missions before that a shortage of staff, particularly accounting staff, had delayed preparation of the promised information. As a result, CDC included an accountant on the 1976 review mission, who confirmed the situa- tion. The review mission recommended that TBZ hire the necessary staff and prepare the information as a prerequisite for the needed rationalization of the project. 36. TBZ, as before, was requested to provide the information forthwith, but the Bank did not inform TBZ about likely remedial action to be taken if the information was not fully available by some mutually agreed date. The Bank had been requesting TBZ compliance in this regard for over three years (and the same problem had been encountered in the Commercial Crops Farming Development Project, then in progress for nearly 6 years) to no avail. It is doubtful that TBZ management took the Bank's request very seriously at this point. 37. The Bank also did not directly raise the issue of the General Manager's performance with the Government during the 1976 review, despite increasing difficulties with project management, including the repeated failure of TBZ to meet agreements reached. The General Manager, who held his position from 1974 untLl TBZ's collapse in 1980, was a political appointee with no experience in running large commercial or parastatal organizations, who operated TBZ largely as a rural development agency. The project was designed to establish settlements in previously underdeveloped areas, and was to include crop extraction roads, domestic water supplies, and a limited number of primary schools and dispensaries. Moreover, project machinery could be put to use for a variety of other non-project purposes. The project thus contained elements of the integrated rural development approach, and was 1/ The Region disagrees with this conclusion as the PPAM in their view underestimates the effect that lack of data had on supervision. The region points to the full discussion of this matter contained in paras. 3.05-3.10 of the PCR and believes that the more important lesson is that drawn in para. 3.09 of the PCR. In addition, the Region adds the following clarification. In 1973/74, the expansion to 12 schemes was not considered very important, as the project was still quite successful at that stage. Later supervision missions were largely unable to monitor progress because of lack of data. Moreover, the distinction between "schemes" and "sections" went largely unnoticed as did the proliferation of staff and other overhead costs. These data should have been obtained from TBZ, if necessary by providing a deadline and then suspending disbursements. - 16 - particularly attractive to regional politicians. As a politician himself, the General Manager tended to favor those individuals who appreciated his agency-s actions, and who would support his activities. His efforts were thus aimed at utilizing the project to assist his overall endeavor. 38. Unfortunately, this endeavor was inconsistent with implementing the project according to its original design. The General Manager frequently stated that he believed the project was too sophisticated and expensive for Zambian conditions, particularly as Zambia's financial situation deteriorated. However, the General Manager did not have a coherent or viable alternative to offer. He spoke of encouraging self-help at village level, coordinated through local Party committees, but no such viable schemes were ever devel- oped. Consistent with this, his unwillingness (which became increasingly apparent as the project evolved) to hire qualified staff and to maintain proper accounts was largely designed to reduce effective opposition to this management. Normally, the lack of staff and accounts would be grounds them- selves for dismissal; in this case, it was not. 39. The Bank did not separate the issue of the General Managers com- petence from the issue of project redesign, and did not discuss both with the Government. It did not do so largely because it thought that it had reached agreement with the General Manager that he would follow the appraisal design, and because only a direct appeal to the President of Zambia could have made the General Manager's shortcomings apparent. On itself, this would have been a sensitive issue. Moreover, with no right to comment on the qualifications of his replacement, and the general weakness of Zambian paras- tatals at the time, the Bank was unsure of achieving an improvement. With the benefit of hindsight, however, taking these risks was warranted. The 1976-78 Period 40. Despite the agreements apparently reached with TBZ, the project manager resigned in late 1976 and was not replaced until early 1978. TBZ's General Manager took his time to find a replacement, despite repeated Bank protests. The Bank wanted TBZ to recruit internationally, but the General Manager indicated that expatriates were "unacceptable", too expensive, and that the Government would not agree. The Bank then discussed the issue with the Ministry of Agriculture, which agreed to look into the matter, but recruitment was still delayed. When ultimately hired, the project manager was not given a vehicle and thus was unable to supervise effectively project developments. 41. TBZ also encountered increasing difficulty maintaining its tractor hire service in operation. Poor tractor maintenance resulted in shorter tractor life (4 years on average) and the unavailability of spare parts led to many tractors being laid up for long periods. Moreover, parts were removed from some tractors to be used as replacement parts for others, and some of the cannibalized tractors never became operational again. The November 1977 supervision mission reported that 40 of 112 project tractors were inoperative. - 17 - 42. The Bank went to great effort to get TBZ to adopt a rehabilitation and use rationalization plan for its machinery and received agreement even- tually that a consultanic would be hired to visit each scheme to assess the need for spare parts to put TBZ's machinery back into operation and ratio- nalize the location of machinery among schemes. This consultant, despite repeated follow-up by the Bank, and discussion between TBZ and the Government, was never hired. TBZ then imported new tractors, of a different make; this exacerbated the maintenance problem and required additional stocks of dif- ferent spare parts. 43. During the 1977 selling year, TBZ implemented a new marketing system, purchasing all tobacco from farmers at pre-season announced guarantee prices and then reselling this to manufacturers also at pre-season agreed prices. The system established resulted in lower prices for lower grades of tobacco. Before this system was implemented, the Bank feared that it might be prejudicial to project farmers, as it actually turned out to be. Family farmers received only 39.9 n/lb., substantially lower than the overall average or than the original target price. Notwithstanding the low average floor prices received by farmers, TBZ made a healthy profit on its processing and marketing, more than had previously been made. About half of this "excess" profit was rebated to farmers at Government insistence, but the other half retained. Thus, the new system permitted TBZ higher revenues, but further reduced the price received by project farmers. The implementation of the system also occupied the time and energies of most of TBZ's most knowledgeable staff during the year of preparation for the new system, further reducing the quality of implementation supervision. 44. The Bank did not formally object to the new system, both because the decision to implement the system was taken at the highest level, and because nothing in the Loan Agreement was abrogated by it. The specific operation of the scheme, however, was clearly prejudicial to project smallholders and was further indication that the project was not likely to achieve its objective under the conditions then prevailing in Zambia, and particularly those in TBZ. 45. Throughout this period, TBZ also failed to establish the revolving credit fund which had been foreseen, or to submit annual reports to the Bank. TBZ promised repeatedly to send these reports by a given date, but on each occasion, had a new excuse for non-compliance. The Bank regularly protested; supervision missions discussed the issue with TBZ management, Aide-Memoires were left specifying agreements achieved, and Bank follow-up letters were dispatched. But the Bank did not threaten suspension. The 1978 Review 46. Bank decision-making during the 1978 review can only be properly understood in the context of the appointment of a new Minister of Agriculture who was well known to, and respected by the Bank, having previously been Minister of Finance. It was decided that the new Minister should be given the facts and the opportunity to put things right in TBZ. In the event, he was not able to do so. This is a fundamental fact which explains the motives behind decision-making by the Region. Nevertheless, the audit believes that at this point in time, given the five-year implementation history of the project and the experience with the earlier Commercial Crops Farming Project, it was overoptimistic to think that the project could be put right. With the benefit of hindsight this reflected poor judgement. - 18 - 47. The Bank considered firmer action in lane 1977, possibly to reap- praise the project or to suspend disbursements 11 Headquarters asked RMEA for a full review of the projecL2 including recomm ndations as to whether the Bank should continue to support the project. It had also earlier requested RMEA for an estimate of the incremental rate of return which would result if the project continued to be implemented with unsatisfactory staffing after January 1, 1978, as then appeared likely.. / 48. Somewhat surprisingly, the incremental rate of return estimated by RMEA for the project was based on a stunningly different set of assumptions, i.e., the considerably more optimistic one that "TBZ can retain, or replace, staff necessary for sound project implementation, and step-up rates of recruitment". Although this assumption was highlighted,.! neither it nor the methodology was subsequently discussed by RMEA or by headquarters and the incremental rate of return which resulted, 18%, was increasingly taken in Bank memoranda as an indication of the project's potential.6/ 49. As the PCR points out (para. 5.03), because of the chaotic state of TBZ accounts, the September 1977 supervision mission obtained its incremental rate of return estimate from a theoretical scheme of 400 growers rather than from an assessment of the actual costs and benefits at that time. The PCR indicates that the figures obtained were probably as good an estimate as could have been made given the total absence of actual cost data. This is a moot point, but irrelevant. Little was learned by this exercise. The model, although utilizing assumptions which partially reflected project experience to date, assumed that greater resolve, and Government budgetary support, would permit success of the original design. The results thus largely reflected unlikely assumptions. Worse, rather than suggest the futility of the original approach under the conditions then prevailing, the effort directed attention away from the manifest problems which the project had been suffering for six years. 1/ The fall 1977 problem project review recommended that cancellation of the project should be considered if no satisfactory progress on the staffing issue had been made. 2/ Telex to RMEA in December 1977 following visit of RVP to the Zambia. 3/ Terms of reference for the February 1978 supervision mission. 4/ Letter from Headquarters to RMEA, dated September 23, 1977. 5/ Supervision report dated November 18, 1977, Annex 2. 6/ The audit accepts that there are methodological problems in calculating rates of returns under assumptions of bad management. In this case, the incremental rate of return was acceptable and the February 1978 mission did not recommend discontinuing the project. However, given the terrible record of project implementation to that date these findings should have been much more closely scrutinized. - 19 - 50. These problems were clearl reflected in the report of the February 1978 supervision mission. It stated/ that: "The Commercial Crops Farming Project completion mission in Septem- ber 1977 was unable to determine from TBZ the costs associated with that project. TBZ maintained no project accounts and only limited project records. The Family Farming project is no exception. No records or accounts are available to identify recurrent or capital project costs,, or to identify items of expenditure within them. A recent project disbursement request included expenditures on family farming tobacco schemes that are not amongst the schemes that form the project. It would appear that TBZ is covering certain recurrent expense with money approved for capital expenditure. A review of project accounts, or lack thereof, was not part of the Mission's terms of reference and would have required considerable time and effort. It is clear, however, that apart from the Bank's concern that key senior staff be in post (e.g. Financial Controller, Internal Auditor, Project Accountant, and Project Administrative Officer), important provisions of the Project Agreement to Project Accounts and information are not being, and have not been observed." 51. The same report, however, also commented upon the earlier calculated rates of return: "The current mission sees no reason to depart from the previous supervision mission's estimate that, assuming improved management and increased rates of farmer recruitment, the economic rate of re- turn would be over 7%.Zl Two further observations modify this position..... To the extent that various incentives! were effec- tive, the potential rate of return would be higher than 7%. Secondly and conversely, TBZ (and Government) lack the necessary will and ability to implement the project in a sound manner". The audit finds it difficult to understand how the basic assumption behind the newly calculated rate of return - namely the assumption that improved manage- ment and consequent increased rates of farmer recruitment - could have been accepted in decission-making, even allowing for the fact that the Bank had high expectations of the new Minister. 1/ Supervision mission back-to-office report dated February 28, 1978. 2/ The overall rate of return was 7%, while the incremental rate of return was estimated at 18%. 3/ The mission had proposed a set of simple measures to make tobacco more attractive; these have subsequently not been implemented. - 20 - 52. Only shortly afterwards the PCR for the Commercial Crops Farming Development Project was completed. It concluded: "It is not clear why with the Management of TBZ being in such total disarray over a long period of time, no action was taken by the Bank to withhold disbursements of funds until such time that management was properly strengthened and accounts put in order...The Bank's apparent flexibility towards project implementation was interpreted as license for complacency and ineptitude."1/ The PCR addresses the question (PCR para. 8.06) as to why the Bank did not suspend disbursements of the project. Basically the Bank considered that it had to give the new Minister of Agriculture a chance to turn TBZ's management around, particularly after he gave a positive response in June 1978 to actions proposed by the Bank (PPAM para. 56). 53. In addition to this consideration, there was also a general feeling, based on supervision mission findings, that "the project remained technically sound but had not been given a reasonable chance to succeed".'2. The Bank, therefore, decided to place a number of choices before the new Minister of Agriculture (PPAM para. 55). At about the same time CDC had in April 1978 after obtaining a severely qualified copy of TBZ's 1975 audited accounts,.3 sent a letter to the Government expressing dismay with TBZ's financial condi- tion and indicating that it "could not imagine under these circumstances IBRD or CDC being enthusiastic about advancing further funds to finance TBZ opera- tions under their existing loan agreements". CDC then proposed to the Bank that Government be given one month to react to its letter and that, unless positive steps were taken, disbursements under both their loans be stopped. The letter to the Bank indicated that CDC "doubted if it would have any effect unless the Bank were prepared to do likewise". 54. At the time that CDC informed the Bank about the status of TBZ accounts, the Bank was in the midst of preparing a letter to the new Minister of Agriculture, setting down what it thought was wrong with the project and suggesting some alternative courses of action which the new Minister might wish to take (see PPAM para. 55 for details). The letter was expanded to cover the question of TBZ's accounts, and in an internal memorandum the Bank noted that, if the Minister took actions which would allow the project to go forward, then steps would have to be taken to isolate any further funds disbursed under the loan from TBZ's overall accounts. This approach was 1/ See PPAR, Zambia Commercial Crops Farming Development Project op. cit. paras. 5.03 and 6.04. 2/ February 1978 supervision mission report dated April 1978. 3/ The 1976 and 1977 accounts, still in preparation, were reported (infor- mally) by the external auditors to be in equally poor condition. - 21 - discussed between RMEA and CDC's field officel/ and was accepted as being more constructive than suspension of disbursements and therefore more likely to be successful. 55. Following the February 1978 supervision mission's report, and CDC's letter and proposal to suspend its loan, the Bank sent a letter to the new Minister of Agriculture. The letter indicated that the Bank understood the Government might wish to change the project's concept, from intensive and costly use of machinery and other inputs, including sophisticated management, to a less mechanized and less costly approach. If so, the Bank pronounced its willingness to go ahead with such an approach, but stated that such a decision would require virtual abandonment of the present scheme, the cancellation of the current balance of the loan, and preparation of a new project. The Bank also pointed out that in its judgement there was "nothing intrinsically wrong with the project as originally conceived, provided recruitment is increased and farmers served better." In addition, the Bank also stressed (PCR para. 6.09) that it would be essential that management problems be resolved if the original concept were to be retained. The Bank also believed mechanization was essential, and that if mechanized land preparation and intensive supervi- sion were withdrawn, 90% of project farmers would quit and also maize output would decline as these farmers were relying on the mechanized land preparation services. In the audit's view, the letter implicitly encouraged the Government to continue with the project, as the alternative option would have involved considerable further delays. 56. The letter was delivered by a Bank mission in May 1978, which discussed its contents with the Minister of Agriculture. The Minister in- formed the May 1978 mission that he had considered the options presented, that he thought the project should be continued according to its original concept and that he agreed with the steps recommended to make this possible (including additional resources for TBZ and improved management). Following this meet- ing, the RMEA/HQ. mission involved indicated that, despite the Minister's assurances, they had severe doubts that there was any chance that the Minister could indeed turn TBZ around while the General Manager of TBZ remained. Nevertheless, the mission also stated -- and this was accepted by management -- that it felt it had little choice but to accept the Minister's assurances and give him a chance to perform. In the event, the mission's judgement turned out to be right: no fundamental changes in project management were forthcoming mainly because these were still resisted by TBZ's General Manager. 57. The same Agricultural Program Mission reported (May 22, 1978) that the general economic picture in Zambia was extremely serious, partly as a result of the decline in copper prices, world inflation, and political events in Southern Africa, but also because of major institutional problems, charac- terized by the almost total lack of central Governmental control and direction 1/ Actually, CDC was represented on TBZ's Board and it continued to try and influence project implementation. CDC's representation on TBZ's Board provided valuable insights to Bank supervision missions. - 22 - of ministries and parastatal bodies. The report stated that each ministry and parastatal operated autocratically, that there was little coordination and cooperation between ministries, and that Treasury overview and control was weak. Since the conduct of Government business required a certain degree of interministerial action, the autonomy of ministerial organization led to little concrete Government achievement. The mission reported that there was no overall strategy in agriculture. The agricultural parastatals were all in serious financial difficulties and in need of review and better management. 58. In early 1979, TBZ fell into effective bankruptcy. In 1977, TBZ borrowed K6 million from commercial banks, ostensibly to finance the purchase of the 1977/78 tobacco crop. However, instead of using the revenues from the sale of the crop to repay the loan, TBZ used the funds to meet their general capital and operating expenses, including counterpart project costs. Because TBZ was not providing the Treasury with the information required to request reimbursement from the Bank, the Treasury had begun to reduce (though not cut off) the funds provided to TBZ. This control did not stop TBZ from borrowing and, because TBZ's financial records were poorly maintained and out of date, its financial position and its use of the commercial loan went unnoticed by both the Government and the Bank. By early 1979, when it was time to purchase the 1978/79 crop, TBZ had consumed the crop finance loan and commercial banks refused to lend additional funds without a Government guarantee. TBZ's finan- cial situation then began to become more fully evident. 59. The September, and especially the December 1978, supervision mis- sions had reported quite positively on project progress and on the basis of their recommendations not to close the project at that time, there was a general feeling that the closing date would be extended. The March 1979 supervision mission accepted that TBZ had made some effort to implement the agreements, reached earlier to put the project back on track. However, this mission also obtained cost data for the first time. Following an analysis of these data, the mission realized that there was no chance of the project succeeding and recommended that the closing date not be extended and TBZ's disbursement requests be rejected unless they could be verified by consultant accountants. The Bank agreed to these recommendations; the project was formally ended (June 1979) and the Bank gave TBZ an additional nine months to resubmit the rejected reimbursement requests. As a lot of expenditures could not be substantiated, the ultimate reimbursement requests were much smaller than those originally submitted. The final disbursement was made in April 1980 and the remaining balance cancelled (for further details see PCR paras. 6.10-6.12 and PPAM para. 9). D. Supervision 60. The PCR argues that the Bank was unable to supervise the project effectively because of a lack of data on which to make judgement (PCR para. 8.05). Contrary to the PCR, the audit believes this argument to be incorrect. Supervision at its most fundamental level involves ascertaining whether or not a project is evolving in a reasonable and promising fashion. This was done and is well documented. The supervision reports, while missing some of the - 23 - finer points of TBZ's decline, and although containing an incremental rate of return estimate which was quite unrealistic, made abundantly and consistently clear that TBZ had failed to meet practically every legal covenant, that project records were practically non-existent (and the accountants had stated that they could not guarantee against fraud or misuse of funds), that recruit- ment and production were well below appraisal estimates, and that project management was disastrously inept, at best. With this at hand, the Bank had the information needed to take decisions. 61. It is true that supervision missions and RMEA vacillated between recommendations that the Bank take firmer action such as the threat of suspen- sion, and on the other hand, that the Bank continue the project urging the Government to improve management in the hopes of resurrecting the project. However, these obvious contradictions were, in the audits view, not due to a lack of information so much as to changes in judgement. There was ample evidence in the supervision reports themselves and in Bank internal memoranda to indicate that supervision mission recommendations might well be in error. There is little evidence in the files that these recommendations were seri- ously questioned. 62. The PCR (1978) for the Commercial Crop Farming Development Project concludes that: "It is not clear why with the management of TBZ being in such total disarrary over a long period, no action was taken by the Bank to withhold disbursements of funds until such time that management was properly strengthened and accounts put in order ......if there is one lesson to be learned from this project it is never to allow a project to slide downhill with ever increasing speed without attempting to apply the brakes. The Bank's apparent flexibility towards project implementation, was interpreted by TBZ as a license for complacency and ineptitude. Where project management, or for that matter the Government, fail to make the necessary corrective measures, the Bank should take the necessary steps (such as withholding of funds) until such time that the project is brought back on its intended course."!/ The attached PCR confirms those conclusions (PCR, para. 8.05). The audit finds itself in total agreement with these conclusions. Although the project was a problem project for five years, and thus subject to much more intense management scrutiny, it continued to slide downhill. In this case it raises questions about how successful problem project reviews are in problem solving at the specific project level. The audit does not wish to convey the message that the Bank should stay away from problems; rather, as partners in develop- ment, the Bank should actively participate in problem solving. However, in this case, the Bank's indulgence was certainly excessive. 1/ PPAR, op. cit., paras. 5.03 and 6.04. - 24 - 63. Throughout the project implementation period the Bank accepted assurances by the Government that something would be done to turn the project around. This is a critical issue: when should the Bank stop accepting that assurances are given in good faith? The decision as to when the Bank actually suspends disbursements is never an easy one. Moreover, at the time there were few precedents to follow. Clearly in the case of this project, the Bank continued to accept assurances for far too long and with the benefit of hindsight this was an error. 64. The PCR places the principal responsibility for the Bank's failure to act on the supervision missions, implying that it was their responsibility to recommend firmer action. The PCR states (PCR para. 8.06): "There appears to be a sound explanation of why the Bank was reluc- tant to take firmer action in the period 1976 to early 1978 (although the PCR does not concur with the reasoning behind the explanation). Supervision missions understood that because of strained relations that existed between the Bank and Government of Zambia during the period that they should not exacerbate tensions by recommending action that involved suspension of disbursements. Accordingly, they were always willing to give TBZ the benefit of doubt and to allow further time for past promises to be met. The PCR mission was unable to determine whether the understanding of supervision missions was based on specific instructions from within the Bank or whether they were perceptions of what missions believed was Bank policy on Zambia. The lesson that emerges is that super- vision missions should recommend what is best on project grounds. Where firm action is warranted, it should be put on record. If their assessment is overruled on country grounds that is another matter." 65. The audit agrees that supervision missions should have recommended firmer action. The audit also agrees that supervision missions should recom- mend what is best on project grounds; other considerations, even though legitimate in follow-up decision-making, should not influence supervision recommendations. The PCR also raises a serious issue about supervision missions, understanding that they should not recommend suspension of disburse- ments. A review of the files does not reveal any instructions from within the Bank to that end.11 The audit therefore concludes that the alternative explanation is valid, namely that missions perceived what was Bank policy on Zambia. For the period 1976 to early 1978, such perceptions included a con- cern not to exacerbate tensions between the Zambia and the Bank and a concern to support a project in which the primary goal was to help smallholders. To 1/ A review of the files also indicates that from 1978 onwards, active consideration was given to suspension of disbursements and supervision missions were asked to specifically include recommendations as to whether the Bank should continue to support the project (PPAM para. 47). - 25 - the extent that such perceptions existed - there is no real evidence to prove their existence, but neither is there evidence to prove their non-existence, and the audit believes that on balance, and partly on the basis of the PCR's own analysis, such perceptions might well have existed at the time - they are dangerous. First, a clear understanding of Bank policy is always better than a perceived perception; second, a perception to not make certain recommenda- tions might influence the judgement of supervision missions; and third, these perceptions related to concerns outside the scope of the project - the audit labels these "country" considerations as opposed to strict project considera- tions - and such "country" considerations should not influence supervision mission recommendations. 66. Another perception also needs review, namely that firm action would have caused marked deterioration in Zambian-Bank relations (PCR, para. 8.07). Although a hypothetical question, the audit agrees with the PCR that the Bank might do more harm in the long run by continuing a project which is performing poorly, and this project performed poorly to say the least.  -26 - ANNEX Page 1 of 7 COMMONWEALTH DEVELOPMENT CORPORATION OFFICE OF THE REGIONAL CONTROLLER COMMERCIAL BANK BUILDING Telephone 732033 P0 Box 30157, LILONGWE 3, MALAWI Telex 4559 MI ROM 506/2005 1st December 1981 The Director, Operations EvalUation Department, World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. Dear Dr. Kapur, ZAMBIA INTEGRATED FAMILY FARMING PROJECT Thank you for your letter of 2nd November which I am afraid became even further delayed as we have during November moved our offices to the above address in Lilongwe. Karl Edwards has however now been able to produce some comment on the Project Performance Audit Report. He has until recently been Regional Agricultural Officer in my Region and is far better acquainted with the project than any other member of CDC here. I am therefore passing copies of his comments to you as they stand and I hope you may find them useful. Yours sincerely, 3. R. Tuckett Regional Controller. Enclosures Note: A draft report was submitted to CDC on November 2, 1981. The present report incorporates revisions JRT/bw following review of the report by Bank staff. CDC's letter was based on an earlier version of this re- port and this should be taken into account when reading their assessment of the project. COMMONWALTH DEVELOPMENT CORPORATION - 27 - ANNEX Page 2 of 7 PROJECT PERFORMANCE AUDIT REPORT ON ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) SUMMARY OF MAIN ISSUES A. Project Design Government of the Republic of Zambia, GRZ, became concerned at the rapid decline in tobacco production following independence in 1964 and the steady departure of mainly expatriate large-scale commercial farmers. Tobacco was the only major agricultural export crop from Zambia and following UDI in Rhodesia (now Zimbabwe), GRZ considered that copper revenues should be used to diversify the country's economic base and maintain or expand tobacco production. Soils and climate in much of the Southern, Eastern and Central provinces of Zambia are very well suited to the production of high quality flue-cured tobacco. World Bank initially undertook to support the Commercial Crops Farming Development Project under which expatriate tenant farmers would be assisted to produce tobacco and maize, while Zambian farmers were to be trained and assisted to replace them, from 1970. GRZ were anxious toinclude small farmers in this scheme but the Bank maintained there was insufficient experience with small farmers. Between 1970 and 1972, GRZ implemented a separate pilot scheme for small farmer production and this was used as the basis for the Integrated Family Farming Project, which was appraised by the Bank in 1972, and commenced in 1973 with CDC as a co-financier. However, it is clear in retrospect that at the time of project appraisal in 1972, information available concerning the pilot scheme and the operations of TBZ already indicated:- (1) TBZ did not have the technical (tobacco and extension specialists) nor financial expertise to-handle the project. Because TBZ carried full finance charges for the auction floors designed for 50m lb (way beyond the various national production targets), its cash flow could not cope with the pressures of an annual production season. (2) The GRZ small farmer pilot scheme failed to achieve production, -quarterly or income targets but this was not taken into account in project preparation. (3) Mechanisation was an essential component of the scheme, although experience with Commercial Crops Project in Zambia and small farmer schemes elsewhere, had already indicated that mechanisation introduced high costs, a demand for a higher degree of specialist supervision and a lower level of small farmer labour input, resulting in a much reduced farmer commitment and an "employee" mentality. (4) The project was ambitious in concept planning to involve 6,300 new tobacco growers dispersed over six different locations. B. Negotiations Immediately after appraisal, the General Manager and Production Manager of TBZ were replaced and following the closure of the Rhodesian border in 1972, - 28 - COMMONWEALTH DEVELOPMENT CORPORATION ANNEX Pige 3 of 7 other expatriate employees began to leave. Despite the already weak position of TBZ management, CRZ refused to provide special incentives for appropriate staff and insisted on applying its own criteria in appointing senior staff including the new General Manager. The Bank reluctantly agreed to accept this position and agreement was concluded with GRZ without any enforceable conditions concerning TBZ staff competence. It is clear from this moment onwards that Bank staff considered this loan in the context of overall Bank policy in the Region and in Zambia and not primarily on the project's own targets and achievements. C. Project Implementation and Bank Review As can be seen from para B, the Integrated Family Farming Project began by accepting conditions contrary to appraisal proposals and ignoring the known track record of TBZ and as a result it was no surprise that conditions deteriorated. Farmer recruitment, tractor hire services, input provision, extension work and tobacco production were always well below appraisal. TBZ's financial and production records were consistently totally inadequate. Repeated supervision missions (17 altogether from 1973-1980) documented TBZ's non-compliance with virtually every loan pre-condition but the Bank never seriously confronted TBZ and GRZ with the possibility of suspending the loan. (It is interesting to note that this was only finally done in 1979 following strong representations by CDC to GRZ in 1978 after CDC participation in a Bank supervision mission and the receipt by CDC of the 1975 audited accounts of TBZ, which were very heavily qualified.) PROJECT DEVELOPMENT Early Bank supervision missions identified poor TBZ management and reporting and rising production costs. By 1974 appraisal targets were clearly not being met and TBZ's problems were increasing, with management and reporting deteriorating and costs rising. At the same time, TBZ embarked on an expanded programme of scheme implementation on its own initiative throughout the country without reference to the Bank or the project proposal. Inevitably project funds were diverted to some of these additional schemes, which at no stage were submitted to the Bank for approval. In an attempt to strengthen TBZ management, the Bank opened negotiations with GRZ to permit the recruitment of qualified experienced expatriate staff. Limited agreement was obtained but high costs and poor performance were continually cited as grounds for not implementing even the limited agreement. TBZ management performance at scheme and HO level continued to decline. As farmer recrUitment fell behind schedule, Bank supervision mission recommended a more aggressive scheme expansion policy. This indicated a failure to appreciate that since 1970, with rapidly escalating costs, most of which were not controlled by the farmer, e.g. barns and other infrastructure, seed beds, fertilizer, etc., the profitability of tobacco production for peasant producers had seriously declined, relative to alternative crops such as maize, cotton, groundnuts and sunflower. The failure to appreciate the economics of the individual small farmer tobacco producer can be seen as a very serious one. With deteriorating TBZ management performance, the position of the General Manager was a serious source of concern. The incumbent from 1974-1980 [ . . . . was a political appointee with no previous large-scale management- - 29 - COMMONWEALTH DEVELOPMENT CORPORATION ANNEX Page 4 of 7 experience, who operated TBZ as a rural development agency rather than follow- ing project guidelines. Unfortunately, on every occasion when policy was discussed by Bank and CDC officials with TBZ General Manager, both financiers were led to believe that the project would in future adhere more closely to agreed implementation, but this was never the case. Finally, as conditions continued to deteriorate and particularly after CDC had declared its intention to GRZ to suspend its loan, the Bank also suspended its loan at 31.12.79 and separately provided funds to GRZ to under- take a complete Tobacco Sector Study during 1980. The report prepared by Minster Agriculture Ltd. recommended the complete dismantling of TBZ production department, and the transfer of potentially viable schemes to Dept. of Agriculture. Unfortunately this recommendation was not fully implanted, and on 1st September 1981 all TBZ Integrated Family Farming Schemes were transferred to Dept. of Agriculture but without making available either funds or staff to ensure success. K. D. Edwards Regional Agricultural Officer. (Signed in his absence) 25.11.81 KDE/bw - 30 - COMMONWEALTH DEVELOPMENT CORPORATION ANNEX Page 5 of 7 COMPARISON OF KASUNGU FLUE-CURED TOBACCO AUTHORITY (KFCTA) MALAWI WITH INTEGRATED FAMILY FARMING PROJECT SUMMARY OF MAIN ISSUES A. Project Design Malawi Government, MG, considered that tobacco production had been neglected during Federation but that with suitable soils and climate, produc- tion would be rapidly increased. Following UDI in Rhodesia (now Zimbabwe) in 1965, MG considered tobacco could provide substantial export earnings for the economy. Taxation and remittance incentives were offered to local and foreign companies wishing to develop tobacco estates, and recruitment of experienced tobacco growers (mainly Rhodesian) on expatriate terms was made relatively easy. A training, credit and service scheme (for Malawi growers) was developed by CDC at Kasungu in Central Region from 1964 and this led to the formation of the Kasungu Flue-Cured Tobacco Authority, a statutory body charged with developing tobacco in the area, in 1968. Initially Malawi growers were alloca- ted 10 acres (4 ha) or 20 acres (8 ha) of tobacco each year, but in 1970 a pilot scheme was introduced with small farmers growing only 2 acres (0.8 ha) of tobacco each year. The original 10/20 acre growers scheme was based on a scaled-down version of standard tobacco estate practice with a high degree of mechanisation (from a central tractor hire service), a high level of inputs, and employed labour carrying out field, barn and grading work. The small-scale (0.8 ha) farmers were intended to carry out virtually all the field, barn and grading work themselves or with members of their family but were still dependent on centralised tractor hire services for land preparation and transport. The detailed project proposals for KFCTA growers and smallholder schemes were prepared by CDC in close co-operation with MG and no major differences of approach emerged at this stage. According to the CDC proposals, all develop- ment costs could be met from expected revenue. MG accepted an expatriate management team of CDC's choice for KFCTA despite the fact this was a statutory authority, and day-to-day management was left very largely to the management team. B. Negotiations The KFCTA, situated in the politically important Central Region, was given considerable publicity during the negotiations to establish the Authority. The site chosen consisted of old tobacco estates south of Kasungu, some of which had been developed by CDC, but prior to UDI none had been very successful. MG clearly saw the scheme as a means of achieving substantial development in the area. C. Project Implementation From the beginning, as indicated in A above, the appointed members of the Authority (the so-called Board) confirmed the strong interest of MG in the implementation of the project and although day-to-day management was largely left to the CDC expatriate management team, policy was a matter for - 31 - COMMONWEALTH DEVELOPMENT CORPORATION ANNEX Page 6 of 7 the Board. The CDC project proposal had indicated that tobacco revenues would meet all development costs but there was misunderstanding at all levels as to how this was to be achieved. Direct production costs were to be charged to smallholders or growers using actual costs incurred. However, the Board insisted that charges be notified in advance and strongly resisted building in inflation factors. During the 1970's when costs, especially fuel costs, rose drastically, KFCTA charges always lagged behind real costs. The admini- stration and development costs were subject to continual discussions at Board level. As these costs are effectively fixed costs completely outside the control of growers or smallholders, they were scrutinised more closely as tobacco prices fell in the late 1970's and growers and smallholders' margins declined. With very low tobacco prices in 1980, many KFCTA farmers made net losses after meeting direct, administration and development costs. At best, KFCTA could avoid losses if all farmers made profits and covered costs, and this would mean management controlling leases and evicting poor performing farmers. The Board insisted on being itself responsible for evictions, and refused to approve evictions of many poorly performing farmers. PROJECT DEVELOPMENT Surprisingly, CDC did not insist on routine agricultural supervision of KFCTA. CDC's own internal reporting requirements were considered adequate for project monitoring with specific reappraisal missions every 2-4 years. Since 1977 this policy has been modified and agricultural projects within CDC are now visited once or twice a year by an appropriate specialist. However, from 1970, physical settlement, production and tobacco quality standards were largely met, although a divergence between 10- and 20-acre growers on the one hand, and 2-acre smallholders on the other hand became apparent. The average yield and quality of smallholders' tobacco was generally above expectations while yield and quality of growers' tobacco was generally below expectations. In addition, labour supervison of growers was not as effective as expected and costs were consequently much higher. Unfortunately, although gross returns kept pace with costs until 1975, costs increased while prices stalilised in 1976-1978 and costs continued to increase in 1979 and 1980 when tobacco prices collapsed. The very low prices in 1979 and 1980 highlighted the basic problems of a flue-cured tobacco scheme for smallholders and small-scale growers:- (1) Management requirement is very high. Good quality technical (tobacco and extension) and financial management is costly. Small farmers and growers can only afford to pay for this management when crop prices are high. Without this management yield and quality targets will not be achieved. (2) Development costs are high. Small farmers and growers can only afford to pay towards these costs when tobacco prices are high. (3) On a settlement type scheme, as KFCTA certainly is, a rigid production package, aimed at high yields, is costly and the smallholder or grower has no control over these costs. Where a smallholder has family labour available, the use of labour instead of mechanisation is obviouslyin the settler's interest. (4) With very large crop price fluctuations as with tobacco, and steadily rising costs, an authority such as KFCTA needs to be able to charge smallholders and growers with the intention of making a surplus in good years to compensate for bad years OR needs a mechanism for Government subvention in bad years. - 32 - COMMONWEALTH DEVELOPMENT CORPORATION ANNEX Page 7 of 7 (5) Commercial criteria must be applied to tenants and poor performers must be evicted if returns for the Authority are to be maintained. Malawi Government is now considering how best to deal with the problems, while continuing to allow the CDC management team to manage on a day-to-day basis. K. D. Edwards Regional Agricultural Officer. (Signed in his absence.) 25.11.81 KDE/bw  - 33 - ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) PROJECT COMPLETION REPORT June'19, 1981 Eastern Africa Projects Department Regional Mission in East Africa  - 35 - I. INTRODUCTION 1.01 The economy o:. Zambia was and remains dominated by the copper industry with average incomes in the modern sector several times those of rural areas. Following Independence, Government decided to diversify the economy and raise rural incomes. With this goal in mind, Government in 1966 announced its intention to develop Zambian tobacco production. Several factors motivated Government to embark on such a tobacco expansion program:- (a) Zambia wished to reduce its dependence on one commodity, copper; (b) Zambia has large areas of light sandy soils suitable for tobacco production; its climate is most favorable and the country had considerable experience in tobacco growing; (c) tobacco is a crop well suited to production by enter- prises o:' all sizes; its production would thus favor participation of large numbers of Zambians; and (d) the UN sanctions against Rhodesia following UDI gave Zambia an excellent opportunity to strengthen its posi- tion in the market. 1.02 The Bank supported Government's objectives with a project designed to increase production of tobacco by commercial growers. The project, the Commercial Crops Farming Development Project, was approved in June, 1970. It also included provision of a specialist in smallholder tobacco develop- ment to assist the Tobacco Board of Zambia (TBZ) in planning and implement- ing a pilot smallholder scheme. The results of the pilot schemes were most encouraging and experience gained formed the basis for the present project. The present project--the Integrated Family Farming Project--was the fourth agricultural project in Zambia supported by the Bank. The Bank loan (Loan 882-ZA) of US$ 11.5 million covered 75% of the estimated project costs of US$ 15.25.million. It was signed on July 26, 1973. The loan was closed in March 1980, nine months later than anticipated, at which time US$ 3.8 million were cancelled. 1.03 This PCR was prepared by staff of RMEA. It is based upon: a field visit to Zambia; extensive discussions with staff of TBZ and concerned Ministries; and a review of project files and the following reports: identification, preparation, appraisal, supervision and the PCR for the Commercial Crops Farming Development Project. Two additional sources of information proved pa:rticularly helpful: first, data compiled by TBZ's economist for the PCR mission and, second, a major study of Zambia's tobacco sector undertaken by a firm of international consultants. The study was undertaken in 1980 primarily to determine (a) the future of Zambia's tobacco industry,and (b) the most appropriate means of production. The study was financed by the Bank out of the proceeds of a Technical Assistance Project (Loan 873-ZA), and the report is quoted extensively in this PCR. - 36 - II. PROJECT FORMULATION Identification and Preparation 2.01 In the late sixties, Government decided to expand the production of tobacco by supporting commerical (expatriate) farmers and by examining alternative means of production better suited to the economic and political climate of the country at the time. As a first step, Government started a tenant farming scheme, under which mainly expatriate farmers, many of whom came from Rhodesia, would be allocated farms for five years. Once this period was over, it was proposed that Zambians would take over as tenants. Zambian farmers were also encouraged to start growing tobacco through an Assisted Tenant Farmer Scheme, whereby selected farmers were to be put through a comprehensive training course before being assisted to grow some 20 acres of tobacco and 40 acres of maize. As a third means of increasing production of tobacco and to assist smallholder farmers, Government also proposed to establish 8 schemes of 100 smallholders each in various provinces of the country. A project was prepared by Government and submitted to the Bank in late 1968 requesting support for the above proposals. 2.02 The project was appraised in December 1968. Because of the lack of experience with smallholder tobacco production, the appraisal mission reduced the proposed settlement of smallholder tobacco farmers by 3 0%. Regarding the remainder of the project, one source of difficulty remained. It became clear at appraisal that 80% of the number of tenants and management personnel required for the project would have to come from Rhodesia. Government felt unable to give assurances that Rhodesian tobacco growers would be allowed entry into Zambia. Accordingly, the project was reappraised in June 1969, The reappraisal mission significantly reduced the tenant farming scheme to be implemented by expatriates, increased the number of assisted tenants, and dropped the smallholder scheme in view of the lack of experience. In its place, the project provided for the recruitment of a specialist in small farmer tobacco groving. He was to assist TBZ in planning and implementing schemes in which small farmers would grow a limited area of tobacco. The project--the Commercial Crops Farming Development Project--was approved in May 1970 and declared effective in October 1970. 2.03 The relatively cool reception in the late sixties by the Bank to the smallholder scheme did not deter Government. In 1968, Government started a small pilot scheme for smallholders. By 1970, it had expanded from 28 growers to over 300 and showed every sign of success. Based on this, Government proposed to submit a second tobacco project to the Bank. A Government preparation team was established and, with assistance from the Bank's Regional Mission in Eastern Africa, submitted its report in August 1971. The proposed project aimed to benefit 9,500 farmers over 4 years. It was proposed that the Bank would expand the 7 smallholder tobacco schemes that had been started already and support 2 new schemes. Each scheme, in turn, would be divided into 5 sections of 240 farmers each. At full development, therefore, there would be 9 schemes and 45 sections. - 37 - Appraisal and Negotiations 2.04 A Bank appraisal mission visited Zambia in February 1972 by which time there were some 500 smallholder tobacco growers, who had demonstrated that they could grow and cure Virginia tobacco successfully of a quality better than the country average. Despite this, the mission cut back the rate of fanmer development proposed by Government on the basis that small- holder tobacco growing in Zambia was still in its infancy. The rate of development was reduced from the Government's proposal of 9,500 growers in four years to 6,300 in five years. The number of schemes was reduced from 9 to 3, and the number of sections from 45 to 24. The reductions were introduced to make for more efficient management units and thereby reduce overhead costs. The appraisal mission required 27 management units against the 54 proposed. In retrospect, there can be little doubt that this was the correct decision. 2.05 Following the appraisal mission, there were three meor isues raised within the Bank and which were expected to affect project implementa- tion. The first concerned senior TBZ and project management. Until 1972, senior TBZ management posts both at headquarters and in the field were filled largely by experienced expatriates. Following appraisal, it was learnt that the General Manager and Production Manager had been replaced and that other expatriates were also expected to leave. The appraisal mission noted that should this happen the success of the project would be jeopardized. The number of suitably qualified and experienced staff available in Zambia to fill these posts were severely limited. To ensure that senior positions were filled, the mission proposed that appointment of five senior staff be made a condition of effectiveness and these staff must be acceptable to the Bank. At negotiations the Zambian delegation pointed out that (a) appointment of the General Manager was the responsi- bility of the President of Zambia and (b) that all other TBZ officers were appointed by the Board. It would not be possible, therefore, to accept interference by the Bank in these established rights. After much consultation within the Bank, the negotiating team yielded on this point. Agreement was reached that appointment of selected staff be a condition of loan effectiveness and that TBZ would employ at all times qualified and experienced persons. It was also agreed that TBZ would inform the Bank of any proposed appointments in time to allow the Bank to comment before the expiration of staff's respective probationary periods. In short, appointments were subject to Bank comment but staff no longer had to be acceptable to the Bank. The departure of senior staff at that time was, in fact, the start of an exodus of senior qualified and experienced personnel that was to continue through 1976, by which time most had left. Many of these positions remained vacant or were filled with staff of lesser qualifications and experience than was envisaged by the appraisal mission. However, it is doubtful that even if the Bank had the right to approve staff that it would have made a significant difference to the outcome of the project. Nevertheless, staffing was a crucial issue and it is unfortunate that the Bank ceded on this point. 2.06 The second issue was the doubtful viability of TBZ because of large outstanding loans to construct its marketing facilities. The apprais- al mission recommended that loans taken out by TBZ to develop its market- ing facilities should be converted to equity. Government was originally - 38 - opposed to this proposal but, surprisingly, at negotiations informed the Bank that it had revised its policy which was now to put all parastatals in good financial order. It was agreed, therefore, that by June 1973 Governrent would submit proposals to the Bank for the recapitalization of TBZ. 2.07 Finally, the Bank was concerned about the relatively poor market prospects for tobacco. To respond to this, the appraisal mission carried out a detailed sensitivity analysis which showed that if the worst scenario should develop (if Rhodesia re-entered the market), tobacco prices could fall by 30% but that even then the rate of return would still be 10%. The appraisal report did not, however, compare the effect such a decline relative to the price of maize might have on the comparative 'advantage of these two crops. This is discussed in paras 4.12 and 4.13. Project Description 2.08 Apart from the staffing issue discussed above, there was basic agreement and no major changes were introduced at negotiations. The main project features as agreed were:- (a) settlement of about 5,400 new growers over the project period, and some 900 existing maize and tobacco growers serviced. Farmers were expected to cultivate 0.5 ha tobacco and 1 ha maize in their first year, rising over 5 years to 1 ha tobacco and 1 ha maize; (b) division of the 6,300 growers into 3 schemes, one in Southern Province, one in Eastern Province and one in Western Province. In turn, schemes would comprise about 7 sections each. There would also be 3 pilot sections to be located in areas where tobacco grow- ing had not been tried. There would thus be a total of 3 schemes and 24 sections; (c) provision of credit through the establishment of a revolving credit fund. Short-term credit would be used to purchase seasonal inputs. Medium-term would be used by farmers to construct curing-barns and grad- ing sheds; (d) establishment of tree and tobacco nurseries by TBZ, and provision of domestic water supplies, tobacco baling centers, a limited number of primary schools, and dispensaries; (e) establishment, within TBZ, of a tractor service to haul fuelwood, assist farmers destump their fields and undertake land preparation; and (f) provision of TBZ extension and management services to farmers. To this end, the project would include housing for staff, vehicles and equipment. - 39 - 2.09 Project costs were estimated at US$15.2 million. It was agreed that the Bank would fiaance about 75% of project costs with a loan of US$11.5 million. The Commonwealth Development Corporation (CDC) agreed to finance about 10% of project costs with a loan of £695,000 (Sterling), leaving Government to cover the remaining 15% of project costs (US$2.1 million). 2.10 The project represented one of the major thrusts of Zambia's development programs aimed to benefit smallholders and bring them into the monied economy. The project's major benefit would be increased pro- duction of tobacco and maize. At the end of the project period in 1977/78, 6,300 farmers were expected to have benefited; it was anticipated that by full development in 1981/82 nearly 9,000 1/ farm families would have improved livelihood as a result of the project. The project would also provide full time employment to some 30,000 adults. 2.11 The project was approved by the Board on January 16, 1973. III. IMPLEMENTATION 3.01 The project failed for a host of reasons. Principal amongst these were poor scheme and section management; poor headquarters manage- ment and financial control; and declining profitability of tobacco vis-a-vis maize, To rectify the situation, supervision missions tended to concentrate on boosting the rate of farmer recruitment and improving management at all levels. Missions believed that the most effective way of improving scheme and section management would be to strengthen senior management within TBZ. Bank efforts to put the project back on track by addressing issues related to senior management are discussed under "Organisation and Manage- ment". This chapter, on Project Implementation, deals primarily with per- formance at scheme level. Effectiveness and Startup 3.02 There were three conditions of effectiveness, apart from the usual legal steps, which included:- - execution, delivery and ratification of the Project Agreement and the Project Financing Agreement; - submission to the Bank of detailed proposals to carry out project evaluation; and - employment of qualified staff to fill the positions of Project Manager, Provincial Managers in Eastern and Southern Provinces, and Scheme Manager in Western Province. The project became effective in July 1973--two months later than expected. The additional two months were required because of delays in furnishing the legal opinion and the Project Financing Agreement. Proposals for project evaluation were submitted on time and adequately qualified staff were appointed to fill key positions; at effectiveness, TBZ emplo,red some 1/ 21 schemes with 400 growers each. 3 pilot schemes with 120 growers each. - 40 - 20 expatriate staff. Thus the conditions of effectiveness caused no undue delays. In hindsight, it would have been helpful had the appointment of one or more accountants also been a condition of effectiveness, since shortage of qualified accounting staff seriously hampered project implementa- tion. 3.03 The Loan Agreement also included a provision that Government would submit by June 30, 1973, a detailed proposal to the Bank for the recapitaliza- tion of TBZ to establish TBZ on a sound financial basis. This covenant should, therefore, have been met by the time the project was effective. It was not and, as with most other covenants in this project, was never fully complied with. As a result TBZ was not put in good financial order; details are discussed in Section VI. The Bank's failure to enforce this covenant was the beginning of a series of accommodations and acceptance of promises of action (that did not materialize) that would eventually contribute to the collapse of the project and the probable wind-up of most of TBZ. 3.04 Experience with the Family Farming Schemes at the time of project effectiveness continued to be encouraging, with a further 510 farmers join- ing the scheme in 1972/73. Yields and the quality of leaf produced also remained good. But the early promise was not to be realized. In March 1975--only some 18 months after project effectiveness--a mission reported that the project was no longer progressing well. The project had recruited only 950 new farmers in the two seasons since it became effective. Far from picking up, the rate of farmer recruitment was declining. Many farmers had abandoned their plots due to unsatisfactory earnings. Moreover, expe- rience was beginning to suggest that few family farmers would be capable, even after five years, of growing more than about 0.4 ha of tobacco, as against the appraisal target of 1 ha. In short, appraisal targets no longer appeared realistic. The mission reported also that management at most levels was weak and deteriorating. Because future prospects looked un- favorable, the mission rated the project a "problem project" and recom- mended that a major review be carried out. The next mission reported that the project was improving and the review was postponed. However, in 1976, the Ministry of Finance, in response to worsening economic condi- tions in Zambia, cut TBZ's budget by 50% and devalued the Kwacha. Respond- ing to this and the continuing deterioration in management reported by supervision missions, RMEA mounted a major review mission in October 1976. Its findings are discussed throughout this PCR. The project remained a "problem project" until it was closed at the beginning of 1980, Scheme and Section Planning 3.05 In 1972/73, there were some 1,020 farmers growing tobacco in 8 schemes. By 1973/74 the number of farmers had expanded to 1,590 and the number of schemes to 12. At the time, no great attention was paid to the number and location of schemes since the project was successful. On the other hand, the appraisal mission had provided for the rational and orderly development and location of schemes and sections. The Project and Loan Agreements included provisions that TBZ would, prior to January 1 each year, prepare and submit to Government and the Bank for their agreement a program for development of project schemes and sections. The program - 41 - was to spell out the proposed locations of schemes and sections and their estimated financial requirements. TBZ would then implement agreed programs accordingly. This covenant was never complied with, although breach of the covenant was raised in several letters to Government and TBZ. Failure to insist on compliance led to the haphazard proliferation of schemes and sections so that at times neither TBZ nor the Bank knew which schemes were included in the project and which were not. More importantly, failure to plan the orderly development of schemes and to spell out in detail their location and development in terms of staffing, vehicle needs, financial requirements and the like was to cause three major problems. First, schemes were frequently opened up in response to political directives so that several schemes were located in areas not well suited to tobacco growing. Second, failure to spell out annual physical and financial requirements meant that the Bank was not able to monitor, for example, how many staff were employed on each scheme to enable checks of staff:farmer ratios and unit costs. Third, failure to draw up written plans for comparison with apprais- al targets, meant that a confusion between schemes and sections appears to have gone unnoticed. The confusion was not semantic and it was to have costly implications on unit overhead costs. 3.06 The appraisal report anticipated that the project would be divided into 3 schemes and 24 sections. In practice, sections became reported as schemes and the project ended up with 24 schemes which were compared to the appraisal target of 24 sections. The project also ended up with some 60 sections. The appraisal report envisaged that in 1978/79 schemes and sections would contain an average of 2,400 and 300 farmers respectively 1/. In practice, they contained 135 and 55 respectively. Although schemes were much smaller than anticipated at appraisal, most had almost as many TBZ staff as intended for the larger appraisal schemes. A similar situa- tion developed with the sections (which were not reported on by supervi- sion missions). Accordingly, there was gross overstaffing at scheme and section levels. 3.07 Scheme Staffing: The extent of overstaffing in TBZ generally-- the bulk of whom were employed on family farming schemes--is illustrated by the following: in 1980, with its 1,765 permanent staff and 1,810 casual laborers, TBZ employed more staff than there were tobacco growers, in the whole of Zambia. In 1977/78, the anticipated and actual staff ratios were:- Anticipated Ratio of ratio at Appraisal Scheme level a/ Appraisal Actual to Actual Area tobacco grown per TBZ staff member employed (ha) a/ 11.8:1 1.5:1 7.8 Number ofr farmers per TBZ staff member employed a/ 17.6:1 3.1:1 5.6 a/ Scheme level only, excluding casual labor. Staff employed at T3Z Provincial and Headquarters levels are also excluded. 1/ Appraisal mission estimated that there would be 7,310 farmers by 1978/79 divided into 3 schemes and 24 sections. This would grow to 9,000 farmers at full development. - 42 - The appraisal report anticipated 1 staff member for every 18 growers. Instead, there was 1 staff member for 3 growers, 6 times as many as anticipated per grower. There were 8 times as many staff as anticipa- ted per hectare of tobacco grown. The ratios would be even less favorable if staff at the Provincial level and of TBZ's Production Department were included. 3.08 The extent of overstaffing and overhead costs that resulted was only brought to light when actual cost and staff data were provided in early 1979 by TBZ's economist. Until that time there was little or no actual data on which to make judgements. Had TBZ provided the annual plans for scheme and section development, spelling out/locations and financial requirements, the PCR mission believes that the proliferation of schemes and sections without due regard to cost would have been picked up earlier. Instead, supervision missions, starting primarily with the mid-term review in 1978, focussed on methods for increasing farmer recruitment--which was correct--but only half the story. Increased farmer recruitment could never bring the project back on track because of the large and scattered number of schemes. It needed to be accompanied with a rationalization of schemes and reduction in staff. 3.09 In addition to the failure of TBZ to provide annual plans and data, TBZ was also permanently in breach of Sections 2.06 and 4.01 of the Project Agreement. These provided that:- (a) TBZ would maintain records to record progress of the project (including project costs); (b) TBZ would furnish to the Bank all such informa- tion as shall be reasonably requested; (c) TBZ would prepare annual evaluation and progress reports for submission to the Bank; and (d) TBZ would maintain adequate accounting records. In short, none of the covenants to provide quantitative data on performance was being complied with. The lesson that emerges is that it is impossible to monitor, supervise effectively, or indeed manage a project without adequate physical and financial plans, progress reports and data on actual performance in terms of expenditures, staffing and the like. The PCR mis- sion believes the Bank should have taken a firmer stand to ensure that this data was provided (paras. 8.05-8.07). 3.10 Finally, with regard to the distinction between schemes and sections it might be argued that each scheme and section proposed at appraisal con- tained too many farmers to manage as one unit (they contained more farmers than had been proposed in Government's application). The PCR mission, how- ever, believes that schemes and sections of the size envisaged at appraisal were not unreasonable. This is illustrated by a comparatively successful family farming tobacco project in Malawi, financed by CDC. It had one scheme of 1,350 farmers divided into 4 sections, each with about 330 farmers and which is expected to expand to 600--very much in line with the targets set by the appraisal mission for the Zambian project. With competent management in TBZ - 43 - at headquarters and scheme levels, it should therefore have been possible to implement the project as designed. 3.11 Physic Planning: The appraisal report provided for 3 land use planners (one in each scheme). The objective was, subject to suitable landuse classification, to set the family farms out on a grid with farmers adjacent to one another. This would have facilitated TBZ extension staff visits to ensure the constant supervision that appears to be an essential feature of successful smallholder tobacco schemes. In practice, not only the schemes and sections, but the farms themselves were frequently scattered over large areas so that, for example, tractors had long distances to travel between one farm and the next, increasing operating expenses, and the likeli- hood of breakdown. The lack of physical plans, therefore, was an additional factor increasing costs, again indicating a lack of discipline by scheme and section management. In retrospect, it would have been helpful had the appraisal report and legal documents stipulated that submission of land- use plans (grids) be a condition of disbursement against expenditures in each section, other than expenditures on management-related activities. Scheme Management and Performance 3.12 The project's early success was due in large measure to intensive management by experienced staff. Unfortunately starting around 1973, sub- stantial numbers of TBZ's qualified staff (many of whom were expatriates) left TBZ for reasons explained later. TBZ was unable or unwilling to offer attractive salaries to facilitate recruitment (Section VI). Also, with the proliferation of schemes and sections, the number of managers to be recruited far exceeded appraisal targets and this number of trained staff was simply not available. The shortage of qualified scheme and section managers had serious consequences for the project since the success of farmers, many of whom had little or no experience in growing tobacco, depended largely on the quality and leadership of the particular scheme and section manager. 3.13 Scheme management was responsible for raising tobacco seedlings and the provision of inputs to farmers: land preparation and ploughing (by the tractor hire service), seedlings, fertilizer and insecticides. Scheme management was also responsible for the provision of seed and fertilizer for maize and general extension advice throughout the tobacco growing and curing cycle. Nearly all supervision reports noted that farmer inputs were provided too little and too late. Farmers' fields were frequently not ploughed until late December or January, causing substantial decreases in yields and quality of tobacco (land preparation should have been completed by mid-October and, preferably, ploughing undertaken in March and April). Because of late provision of inputs, farmers tended to pay more attention to their maize crop (para. 4.13). Fesponsibility for the late provision of inputs was partly that of scheme and section managers and partly that of TBZ head- quarters management. For example, there was little that local managers could do when headquarters provided funds too late to enable them to pur- chase their inputs such as fuel and spare parts on time (one of the reasons that the farmers' fields were prepared late was that 35-50% of tractors were inoperative for lack of day-to-day maintenance and spares). The farmers' problems did not end with growing; they generally had to wait several months - 44 - after they had delivered their crop before receiving payment, which was the responsibility of TBZ headquarters (para 4.09). The inability of TBZ to provide adequate inputs, services and payments in a timely manner was a strong disincentive to tobacco growing. 3.14 It is instructive to compare briefly how different schemes per- formed within the constraints imposed by poor headquarters management (Tables 1-3). While average yield levels and the rate of farmer recruitment were undoubtedly influenced by soil and climate, the overriding factor was management. For example, Kaoma had consistently good yields though it is in an area of the country that is isolated and where ecological factors are not particularly favorable. Equally, there are a number of schemes, such as Singani and Sipatunyana, with poor performances though they were in areas where ecological conditions are ideal. Performance in several schemes changed markedly with changes in management. A good example is Kabompo where yields were good from 1974-76, when there was capable management, yet dropped significantly the following year when management changed. Clearly also management affected the rate of farmer recruitment and reten- tion. For example, from 1973/74 to 1979/80, more farmers left Luwerezi scheme than joined it, while at Kaoma (well managed), the number of farmers increased from 90 to 411 in the same period. 3.15 Table 3 provides selected indicators of performance by schemes. It has not been possible to provide fully consistent data; some refers to 1978/79 and some to 1979/80. Fortunately, the inconsistencies are unlikely to affect conclusions. The table shows that settlement schemes generally performed better in terms of yield than extension schemes. 1/ For example, Zemba, Lumezi, Kaoma, Kabile, Chipangali, Mpongwe, Mkushi West, Nyawa and Chimsali, all primarily settlement schemes, achieved yields of over 1,000 kg/ha. This suggests that settlement schemes were either easier to manage or had better managers. The ten largest sclemes achieved higher yields on average than the 14 smallest schemes. This is partly explained in that several of the smaller schemes started later, though in most of these schemes, yields have fallen over time. The correlation between scheme size and yields would again suggest that management was the major factor. 3.16 Schemes with high yields often had lower staffing levels in contrast to the assumption that appears to have prevailed in TBZ: that management at scheme and section levels could be improved with the recruitment of more staff. Table 3 shows considerable variability in:- - the ratio of scheme salaries to area of tobacco grown (from K168/ha in Sindenisale to K3,860/ha in Kashima); and - the ratio of farmers to staff (from 5 farmers per TBZ staff member on Zemba, to 1 per staff member at Chikauta; these figures exclude casual labor). The appraisal report anticipated there would be 18 farmers per staff member, so that even the best schemes were considerably overstaffed. 1/ It was anticipated that 5,400 new growers would be settled and some 900 existing growers serviced (by 1977/78). - 45 - 3.17 In sum, despite constraints imposed by headquarters, some schemes performed better than others, although not well enough to grow tobacco economically. During the life of the project more and more missions com- mented on the deterio:ration in scheme and section management, The major cause cited was lack of direction due to weak management at TBZ headquarters and is discussed under Organisation and Management. Physical Infrastructure 3.18 The project provided for the construction of barns, roads, water supplies, schools,dis-pensaries and staff houses. About 8,000 barns were built to service 1,500 ha of tobacco, slightly more than the 5 barns per ha envi- saged in the appraisal report. Most, however, were poorly designed and built. Project supervision missions made numerous requests for information on con- struction of the other infrastructural components. This data was repeatedly promised by TBZ but was not provided to supervision missions and was only sent to RMEA following the PCR mission. It has, therefore, not been possible to verify the data. Based on the data provided, 1,270 km of roads were con- structed under the project, against an appraisal target of about 3,000, 40% achievement. 7 schools, 9 clinics and about 300 houses were also con- structed. The appraisal report does not contain information on numbers of buildings, only targets expressed in expenditure terms. Actual expendi- tures on these items was generally about 75% of the amount provided at apprais- al. Using roads as an illustration, which achieved only 40% of the target, it would indicate considerable cost overruns on construction. Two lessons emerge from the above. First is the importance of including physical tar- gets in an appraisal report. Second, the PCR mission believes that it is' unacceptable that missions in 1978 and 1979 should not have been supplied information on physical progress. Without such data, missions cannot carry out their responsibilities effectively. Project Costs 3.19 Project costs were estimated at appraisal at KlO.9 million (US$15.2 million) over a five year development period. Because of the general state of TBZ's accounts (Section VI), it has not been possible to obtain accurate date, on project expenditures. 3.20 The PCR mission for the Commercial Crops Farming Development Project, implemented by TBZ, noted that "all efforts by the completion mission to arrive at actualproject costs proved fruitless". If the present PCR mis- sion had to rely on the accounts depaitment of TBZ, the situation now would be no different. However, following a Bank recommendation, TBZ recruited a project economist in 1978. He was then requested by the Bank to determine actual expenditures on the project so that the Bank could estimate a rate of return based on actual cost data. To do so, the economist had to go back to scheme accounts, ledger by ledger, to extract the necessary data. He reports that the accuracy of the accounts varied considerably scheme - 46 - by scheme and that many expenditures were charged to the wrong account head. He was not able to determine precisely which staff and costs of TBZ's Production Department in Lusaka and the Provincial Headquarters, could be ascribed directly to development of Family Farming Schemes. Based on an analysis of tobacco production and time spent by the Headquarters and Provinces on project-related activities, the economist estimated that 70% of the cost of these activites should be added to scheme costs to derive pro- ject costs. The estimate of project costs, is, therefore, only indicative of levels of expenditure and does not represent actual expenditures. Project costs so estimated are in Table 5 and are summarized below:- Estimated Project Costs Actual as Appraisal Actual % of appraisal ------ (K '000) ----- A. Credit 1,921 - a/ - a/ B. Project Investments Vehicles and Tractors 1,046 1,290 123 Specialized Equipment 540 510 94 Housing and Buildings 1,077 1,020 95 Dispensaries and Schools 169 100 59 Water, Roads, Soil and Forestry 1,340 850 63 Operation of Tractors 179 - - Other - 540 Sub-total 4,351 4,310 99 C. Establishment Costs Wages and Salaries 3,000 6,230 207 Administration 367 270 74 Vehicle Operation 1,042 2,800 268 Road Improvement 70 200 285 Improvement to Buildings 135 - - Other - 980 227 Sub-total 4,614 10,480 227 Grand Total 10,886 14,790 135 ---- (us$ '000) ---- Grand Total b/ 14,000 19,000 135 a/ TBZ were unable to provide data on the amount of credit issued to -farmers. Credit costs are subsumed under individual headings, for example, "vehicle operation".would include operating costs of tractors required to plough a farmer's field. Similarly, fertilizer costs are included under "other" establishment costs. Both these were credit items. b/ Based on exchange rate prevailing at project completion. - 47 - Estimated project costs are Kl.8 million (US$19.0 million). Thus actual expenditures on the project exceeded appraisal estimates by about 35%. Establishment (operating) costs comprised over 70% of expenditures as against 42% estimated at appraisal. Predictably, the two major increases were staff costs, from K3.0 million to K6.2 million, and vehicle operating costs, which increased by Kl.8 million. Together these two overhead costs represented over 60% of project costs. 3.21 Operating Results: While total project costs exceeded appraisal forecasts, the number of farmers recruited was lower. As a result, the actual costs of development per farmer were three times that forecast. The appraisal report projected that capital costs over the project period per,farmer would be K690; in practice they were Kl,300. Establishment costs over the project period were estimated at appraisal at K730 per farmer; they were over K3,000. A more detailed analysis for the first and last years of the project high- lights how much the overhead costs increased over time:- Actual as % Appraisal Actual of appraisal ----- Kwacha ------- Overhead Cost Per Grower First Year 410 48o 117 Final Year 180 660 367 ----- Kwacha/Kg ------ Overhead Cost Per Kg. Tobacco First Year 0.9 1.2 133 Final Year 0.25 1.4 560 As shown, the appraisal mission anticipated that overhead costs per grower and per kg of tobacco would fall as the number of farmers per scheme increased. Instead, they increased until the overhead cost per kilo of tobacco was over 5 times that projected at appraisal-. The extent of costliness of TBZ's operations is perhaps best illustrated by the following: the cost of produc- tion to TBZ for tobacco produced by family farmers (including the payment made to farmers for tobacco) was estimated at K3.2/Kg in 1977/78 and K2.7/Kg in 1978/79. This compares with an auction price for tobacco produced by family farmers of Kl.h/Kg in 1977/78 and K1.3 in 1978/79. Thus, it cost TBZ considerably more to produce tobacco than the crop was worth. Financing, Disbursements and Procurement 3.22 The Bank disbursed US$7.73 million from its loan of US$11.5 million. US$3.77 million was cancelled. The Bank thus financed 4l% of actual project costs in contrast to the 75% expected at appraisal. CDC disbursed about US$0.55 million from its loan of US$1.7 million equivalent; the balance was cancelled. Total donor assistance was, therefore, US$8.3 million, so that some - 48 - US$10.7 million was injected into the project from other sources. This is considerably more than was anticipated for Government's contribution, esti- mated at appraisal at US$2.1 million, and is surprising in light of Govern- ment's budgetary cuts. The additional funds will have come from TBZ's own funds and commercial bank loans. Because of the manner in which TBZ's accounts were kept, it has not been possible to determine how much was pro- vided by each of these sources. This is discussed in detail under the section on TBZ's Financial Management and Accounts (paras. 6.13-6.17). 3.23 From the estimates of project costs given above, it should have been possible to draw down the loan amount. However, in breach of the Project Financing Agreement, which required that TBZ would, within one month follow- ing the month in which the expenditure was incurred, submit to Government documents to enable it to apply for reimbursement, TBZ failed to prepare disbursement applications because of the lack of accounting staff. At the Bank's request, TBZ's auditors were requested to help from time to time but could not always prepare the necessary documentation since they were unable to reconcile the accounts. Also, TBZ could not make any claims against the revolving credit fund from 1975 because it was never set up (in breach of Section 2.02 of the Project Agreement), despite numerous requests by the Bank. Since it was impossible to reconstruct the revolving fund, the mid- term review agreed a balance with TBZ's auditors so that TBZ could start with a clean slate in 1977. Despite this, the revolving fund was not established demonstrating a lack of interest on behalf of TBZ management and lack of follow-up by the Ministry of Finance. 3.24 A schedule of disbursements is in Table 6. Procurement procedures for project goods and services were generally satisfactory. IV. AGRICULTURAL IMPACT Farmer Recruitment 4.01 After a promising start the rate of farmer recruitment declined so that from 1976/77 onwards there were less than 50% of the number of farmers anticipated at appraisal. At the close of the project, there were some 3,050 farmers participating against an appraisal target of 8,100 (38% achieve- ment); details are in Table 4. This, however, overstates project achieve- ments since at project start-up there were already some 1,000 farmers who had joined the scheme (Table 1). The incremental number of farmers during the life of the project was, therefore, only 2,030 or 28% of appraisal targets after adjustments for pre-project farmers. The total number of farmers recruited was more than shown because of the high drop-out rate amonst far- mers: an analysis of data shows an annual drop-out rate in the range 10-20%. Together with the low recruitment rate, it indicates a strong degree of far- mer dissatisfaction with the program. 4.02 The poor rate of farmer recruitment was highlighted by most super- vision missions--it was one of the main reasons cited for the mid-term review. The review mission, and others that followed, recommended that the rate of - 49 - farmer recruitment be stepped up and, at various times, targets for farmer recruitment were set and agreed with TBZ. They were never achieved. Indeed, several scheme managers reported that they did not recruit actively since they knew that TBZ was unable to provide farmers timely inputs and services. Most of the problems that led to slow recruitment were identified by mis- sions but TBZ was unabLe or unwilling to resolve them: poor returns to tobacco versus other crops (paras. 4.12-4.13); late and unsatisfactory payment methods to farmers (paras. 4.0;3-4.1o); untimely and inadequate provision of farm inputs which in turn depressed yields and profitability; and proliferation of schemes and sections which led to unacceptably high overhead costs. Area, Yields and Quality 4.03 Based on data provided by TBZ (Table 4), the area planted to tobacco per farmer was also below that anticipated at appraisal--about 0.45 ha as against an average of o,8 ha anticipated at the end of the project period (at full development farmers were expected to grow 1 ha of tobacco). These figures probably understate the area planted by each farmer: Minster checked several of the TBZ farms and found that frequently the size of farmers' fields was recorded incorrectly. Fields were found to be larger than recorded-- in some cases double. The average area of tobacco planted is estimated at about 0.65 ha. There Ls little data on areas planted to hybrid maize but it appears to have been some 0.8-0.9 ha per farmer. 4.o4 Average tobacco yields were generally reported in the range 700-900 kg/ha, which compares favorably with appraisal projections (Table 4). The yield estimates provided by TBZ, however, are almost certainly over-estimated since the size of farmers fields were generally larger than recorded by TBZ. The key to increased farmer returns to tobacco lay in improving yields. Average yields of 1,200 kg/ha should have been possible but this would have required that inputs be provided on time and husbandry practices closely super- vised. Missions focussed on means to provide farmers with timely inputs including drawing up a schedule of incentives to encourage farmers to destump fields early to lengthen the time available to TBZ to carry out land preparation. As with most other mission recommendations, they were not implemented. The quality of tobacco deteriorated throughout the life of the project, reflecting progressively poorer supervision of farmers by TBZ. 4.05 TBZ made no attempt to measure maize yields. It is estimated, however, that the better farmers achieved yields of 20-25 bags/ha (90 kilo bags) at the beginning of the project and 30-35 bags/ha at the end. This compares to appraisal projections of 35-45 bags/ha. Production 4.06 Tobacco: The impact of the project in terms of incremental tobacco production was disappointing:- - 50 - Tobacco Production ('000 Kg) Actual as % Appraisal Actual of Appraisal 1972/73 a/ - 500 - 1973/74 770 641 83 1974/75 1,4o0 1,061 76 1975/76 2,230 1,070 48 1976/77 3,180 1,245 39 1977/78 4,260 1,041 24 1978/79 5,370 1,503 28 a/ Pre-project. As shown, in 1978/79 tobacco production from family farms reached 1.5 million kilos, or only 28% of the appraisal estimates. Incremental production, exclud- ing production prior to project start-up, was only 1 million kilos, or 20% of the appraisal estimate. Future recommendations for the tobacco industry (Section VII) suggest that many of the smaller schemes under the project will be closed. Thus the level of production that is likely to be maintained as a result of the family farming schemes is less than that shown. 4.07 Maize: It has not been possible to determine with any degree of accuracy the impact of the project on maize production. Assuming farmers grew 0.8 ha hybrid maize and that average yields were 30 bags/ha, total maize production would be about 80,000 bags or 7 million kilos. This compares with an appraisal forecast of 263,000 bags (30% achievement). The appraisal report considered all maize production to be incremental. Clearly, whether farmers were settlement or extension farmers, they would have grown local maize pre- viously. Assuming local maize yields at 10 bags/ha, incremental maize pro- duction as a result of the project would be 20 bags/ha, giving total incre- mental production of only 52,000 bags or 4.7 million kilos. This presupposes that the switch from local to hybrid maize can be fully attributed to the project. However, TBZ made little effort to provide extension to encourage maize growing and it is likely that a large part of the "incremental production" would have taken place without the project. Farmer Prices, Payments and Incentives 4.08 Tobacco: The farmer price of tobacco increased only slightly over the project period, from KO.88/Kg to Kl.04/Kg. Based on an annual inflation rate of 13%, the leaf price received by the farmers decreased, in constant terms, by about 35% over the project period. This decrease was brought about by three factors. First, the decline in quality of tobacco grown by family farmers in Zambia. Second, a decline in world market prices of tobacco of about 15% in constant prices--a factor over which TBZ had no control. And third, the intro- duction by TBZ of a new system formarketingof tobacco. In 1976, TBZ announced that they would abandon the auction system. TBZ management believed that the decline in Zambia's production of tobacco meant that the auctions were attracting too few buyers for healthy competition. To overcome this problem, TBZ proposed that it would classify the tobacco itself and would then sell by negotiated con- tract. One of the objectives of the new system was to facilitate early delivery - 51 - to Lusaka, immediate classification and quick payments to farmers in the year following the introduction of the newmarketing system, a supervision mission noted that TBZ presold the entire crop at 20% below prices obtained at auctions in neighboring countries. The mission estimated that had the auction system been maintained farmers would have gained 22 ng/kg. In short, the new marketing system, although introduced for ostensively good reasons, failed to meet is objectives. 4.09 In addition to low farmer prices, farmers were always paid late, sometimes up to 8 months after the farmer had delivered his crop. This matter was raised by most missions since it was felt to be.one of the main factors affecting farmer recruitment. The mid-term review report notes that under the existing payment system, and given the traditional orientation to maize, it was likely that if maize had been provided without the pre-con- dition of growing tobacco, more than 80% of farming families settled on to- bacco schemes would have ceased to produce tobacco. The new marketing system was introduced to speed payment but it did not, primarily because of mismanage- ment of TBZ's financial affairs so that TBZ did not have the liquidity to pay farmers. 4.10 In addition to late payment, a further disincentive to tobacco growing was that all TBZ charges for growing tobacco and maize were deducted from the proceeds of the farmers' tobacco sales. Thus the farmer was left with little cash from sale of tobacco and it appeared to the farmer, because of the accounting system, that tobacco was not worthwhile. It is difficult, if not impossible, to devise a system whereby maize costs would not have been deducted from tobacco since there is no controlled marketing channel for maize. The February 1978 mission examined this issue in depth and recommended that the best way to overcome the problem was to stop provid- ing project farmers inputs for maize on credit. The recommendation was not implemented by TBZ so it is not possible to conclude if it would have helped. It is possible, of course, that had TBZ stopped giving inputs for maize, many farmers would have ceased to grow tobacco. 4.11 Maize Prices: In contrast to tobacco, the price farmers received for their maize increased by about 150% over the project period, represent- ing an increase in constant prices of about 40%. The real increase in price reflects Government policy to boost maize production since it happened at a time when maize prices were falling (in constant terms) on the world market. Returns to the Farmer 4.12 Based on an analysis of actual farm data, the gross returns to the farmer have been estimated for tobacco and maize as follows (details are in Table 7). - 52 - Farmer Returns to Tobacco and Maize (1978/79) Tobacco a/ Tobacco Hybrid Maize a/ Local Maize 1'ha 1 ha 1 ha 1 ha Yield (Kg) 850 850 3,000 1,080 Price (K/Kg) 1.0 1.0 0.10 .1 Gross Return ( K) 850 850 300 108 Farmer Costs of Production excluding farm labor (K) - 226 1 110 8 Gross Margin (K) 624 375 190 100 Man-days 780-600 780-600 90 85 Return/man-day (K) 0.8-1.0 0.5-0.6 2.1 1.2 a/ March, 1979 supervision report. b/ Tobacco Sector Study. c/ .uoes not reflect economic costs since farmer was not charged by TBZ for all of his inputs. The budgets show that the farmers' return was about KO.5-K1.0 per man-day for tobacco, K2.1 for hybrid maize and Kl.2 for local maize, At the wage rate prevailing in 1978/79 (Kl.l per man-day) the farmer and his family would have had to work for less than the daily wage to grow tobacco. In Zambia, the constraint to production is labor, not land; it is therefore to be expected that farmers would grow the crop with the highest return per man-day, subject to risk, after meeting subsistence requirements. On this basis, maize would appear to be a better crop than tobacco. This is in contrast to Malawi, where the major constraint to production is land, not labor. Under Malawi conditions, therefore, farmers would tend to grow the crop with the highest return per ha, explaining in part why tobacco fared better in Malawi than in Zambia. 4.13 Since the appraisal report does not include a separate farm budget for tobacco and maize, it is not possible to compare the relative return to the farmer of tobacco and maize at the time of project appraisal. Given that the price of tobacco increased by 25% over the project period, while that of maize increased by 150%, there has certainly been a major deterioration in the return to tobacco relative to maize. The decline in profitability vis-a-vis maize compounded problems of low yields and untimely provision of farm inputs. It should be pointed out that maize and tobacco are in part complementary making up a "farming system"-. However, although the peak labor requirements fall at different times, land preparation and planting fall at the same time. They do therefore compete at certain times of the year and the farmer tended to give priority to his maize, over which he had greater control, The poor return per man-day partly explains the reluctance of farmers to adopt and expand the area of tobacco. Many farmers - 53 - who grew tobacco planted little more than the minimum required for inclusion in the project (0.4 ha) so as to qualify for the supply of inputs (fertlizer, seed and tractor power) for his maize. Others switched to alternative crops which gave a better return. The recentTobacco Sector Study comnares the returns to various crops in 1980 and its figures explain, in part, why farmers drop- ped out of the program. Based on 1980 prices, returns per man-day for the different crops are:- Yield Return per Crop (Kg/ha) Manday (K) Virginia Tobacco 1,000 0.91 Burley Tobacco 480 0.94 Sunflower 800 1.00 Groundnuts 960 1.38 Maize 3,600 3.29 Cotton 750 5.20 V. ECONOMIC RATE OF RETURN Rate of Return 5.01 The economic rate of return of the project is estimated to be negative compared to 18-20% forecast at appraisal. Benefits in terms of tobacco and maize production (in 1978/79) were 28% and 30% respectively of appraisal targets. Production in 1979/80 was lower than 1978/79 and, under existing TBZ management, the PCR mission is unable to project that production would increase significantly in the future. While benefits are lower than forecast at appraisal, costs are much higher due to overstaffing, heavy vehicle operating expenditures, and inefficient use of human and physical capital. Supervision Estimates of Rate of Return 5.02 The mid-term review mission estimated the rate of return for the project at 6%. In 1978, the Bank was considering taking firm action, possibly to reappraise the project or to suspend disbursements. It was decided to estimate the incremental return to future investments. The incremental return was estimated at 18% and, based on this and other factors, it was decided to continue financing the project. In retrospect, it is clear that these estimates were in error and led to the wrong decision. 5.03 Because of the chaotic state of TBZ accounts, supervision missions until March 1979 had no access to actual cost data. TBZ's economist had not been employed at the time of the mid-term review, so the rough data provided to the PCR mission was not yet available. Accordingly, the mid- term review mission estimated the rate of return from a model based on a theoretical scheme of 400 growers rather than on an assessment of the costs and benefits of the 14 schemes that existed at the time. The model assumed a rate of farmer recruitment that was better than had been achieved to date, although not overly unrealistic based on data available through 1975/76. The assumed recruitment rates should, however have been sub- stantially reduced when the rate of return was rerun in 1977 and 1978. - 54 - In brief, estimates of project benefits were somewhat overstated, but cost estimates bore no relation to reality. For example, costs of ploughing, discing and ridging of farmers' fields by TBZ were assessed at the rate TBZ charged farmers, not at the real cost to TBZ. In 1978, for example, the hectare charge was K22.5 for ploughing, K11.2 for ridging and Kll.2 for discing. These costs included heavy subsidies which were borne by TBZ and, should have been included in an assessment of economic costs. Based on data provided in 1979 by the TBZ economist, it appears that the actual costs of tractor operation (excluding depreciation) were about K100/ha,--over 5 times the cost included in the supervision model. Actual staff costs were grossly underestimated, as were overhead and administrative costs at Provincial and Headquarters levels. For example, the review report estimated Provincial and Headquarters overhead expenditures at KO.25/kg of tobacco sold. Based on data available to the PCR mission, actual costs (without shadow pricing) should have been about KO.75/kg-- 3 times the rate used in the supervision model. 5.04 The supervision model was probably as good an estimate as could have been made at the time given the total absence of actual cost data. One strong lesson emerges: the need for up to date monitoring and evalua- tion data and for the maintenance of records to record progress of the project, including project costs. The PCR mission believes that in a situation where there was no actual cost data, rather than run a rate of return that led to the wrong decision, a deadline should have been set for data to be nrovided. If not produced, this PCR mission would con- clude that disbursements should have been stopped since: the Bank had insufficient data to monitor the progress of the project effectively; there was inadequate financial control in TBZ (paras. 6.13-6.16); and the Bank could no longer justify continuing its support if the cost-benefit relationship could not be evaluated. VI. ORGANIZATION AND MANAGEMENT Staffing 6.01 A continuing theme that ran through supervision reports was that successful implementation of the project (and the Commercial Crops Farming Development Project) depended largely on the quality and dedication of management. Supervision missions believed that the best way to improve the quality of services to farmers was to ensure that TBZ had sufficient numbers of competent staff at the top. With the large number of schemes and sections it would not have been possible for missions to focus on management needs of each scheme and the assumption was therefore--and the PCR mission believes correctly so--that with good senior management the necessary steps could have been taken to implement the project as agreed. From the start of the project, there was permanently a shortage of staff at senior levels in marked contrast to staffing at scheme and section levels (the overstaffing at scheme level appears not to have been picked up by supervision missions because-of the confusion between schemes and sections and the absence of actual data). The February 1974 mission, only 8 months after project - 55 - effectiveness, commented that unless the staffing position improved, it was difficult to see how the project which had been developed successfully until that time could be satisfactorily maintained, let alone expanded. Rather than improve, the senior staffing position became steadily worse. 6.02 By 1975, when the project was first rated a "problem project" it was already clear that TBZ could not attract and keep competent managers at existing salary scales. A supervision report noted that, while staff vacancies were a problem, the general lack of direction from existing senior management was the main cause of TBZ's weaknesses and thereby the deteriora- tion in the project. Missions from late 1974 to mid-1975 felt constrained to tackle the salary issue pending the outcome of a Salaries Commission established by Government. 6.03 The Mwanakatwe Commission was established in mid-1974 in response to Zambia's deteriorating financial position and, in particular, the poor financial state of most Government parastatals. By late 1975, it was clear that the Commission would recommend decreases in management salaries and that many of TBZ's qualified staff would leave. On learning this, the Bank proposed to TBZ and Government that one way round this problem would be to recruit staff through the Bank's Agricultural Development Service (ADS). Staff costs would be charged to the loan and staff could thereby be offered attractive salaries. This proposal was rejected by Government. The Commission issued its report in mid-1976 and confirmed that salaries and benefits of senior staff would be reduced to bring staff working for parastatals onto the same scale as Government servants. Education and other fringe benefits of expatriates would be cut. The result was predict- able. A mission found that 9 Zambian and 6 expatriate senior staff were expected to leave in the near future. The mid-term review mission mounted in 1976, partly in response to the deteriorating staff position, noted that by then most of TBZ's qualified expatriate and Zambian staff had either resigned or served notice that they intended to resign. The Bank brought to Government's attention that the recommendations of the Commission would seriously affect the implementation of the project. After 8 months, Govern- ment indicated its agreement (in December 1976) to allow TBZ to recruit at pre-Mwanakatwe conditions. By this time, however, most staff had left or made their decision to leave. 6.o In light of Government's waiver, the Bank again proposed that key staff be recruited by ADS., As an alternative the Bank proposed to use loan funds to top up salaries. Both proposals were rejected as being too costly. Thus, although there appeared to be some accommodation to the Mwana- katwe proposals, in practice it continued to thwart recruitment of key staff. The following is illustrative of Bank concern at the time and action taken. In late 1977, the Bank learnt that the Financial Controller and the senior Extension/Training Officer, both expatriates, were planning to leave. The Financial Controller's contract included a provision for termination after six months if TBZ failed to arrange for an increase in salary pay- able in foreign exchange. The Bank felt that both officers were crucial to TBZ's operations. Accordingly, the Bank sent a telex to Government expressing its concern that several key staff were about to resign. Government responded that "we have tried to find out which officers are - 56 - leaving (TBZ) but have been unable to locate them". The mission report comments that the response was symptomatic of the on-going lack of concern by Government and TBZ to face the issue of TBZ staffing. 6.05 In the fall of 1977, at the Bank's internal review of problem projects a decision was taken that if satisfactory staffing was not likely after January, 1978, the Bank should estimate the economic value of any incremental investment under these conditions and, if the prognosis was poor, the Bank should consider suspension. As discussed, the rate of return was estimated and deemed acceptable, although it is now clear that returns in 1978 were already headed for a loss. The mission noted that 10 key senior staff needed to be recruited as soon as possible. However, the Bank accepted recruitment of a Project Manager, an Economist, and a Financial Controller as encouraging signs that TBZ was more willing than formerly to improve project implementation and decided against firm action. Throughout 1978 the Bank continued to discuss with TBZ steps to assist recruitment. Memo- randa of Understanding were agreed and some minor progress was made, although the accounts department continued to be starved of competent staff. The improvements were insufficient, however, to turn the project around and to justify extending the project period. 6.06 Despite the Mwanakatwe Commission and the lack of resolve of Government to assist TBZ, the PCR mission concludes that management of TBZ must bear the bulk of the responsibility for the vaccum in project staffing. A more determined effort by TBZ in putting its case to Govern- ment would.undoubtedly have helped. The lack of concern by senior management on staffing appears to have been caused by a fundamental disagreement on approach to project design and implementation, as much as it was to weak management. The lack of concern to fill key posts on the accounts side was however, inexcusable and had serious consequences for TBZ (paras. 6.15-6.17). Senior Management's Approach to Project Concept 6.07 Although the project got off to a good start in terms of farmer recruitment, the elements of future proliferation of schemes and lack of discipline were, with hindsight, evident early. Three of the schemes initiated before the project was effective were started-up unplanned by enthusiastic agricultural officers. Annual plans and scheme layouts (or grids) were not prepared and there appears to have been little discipline or system in the location of settlement farmers. In 1974 a new General Manager was appointed, a largely political appointment with no experience of running large commercial or parastatal organizations. This was a Presidential Appointment and was not subject to Bank approval. At much the same time as his appointment, Zambia's economic conditions deteriorated sharply with falling copper prices and there was a concerted effort to Zambianize. Government wanted, therefore, to move away from "costly" projects requiring sophisticated management and announced a drive to pro- mote agricultural development through decentralized and self-reliant local village level committees. Reflecting this change in policy, TBZ announced plans in 1976 to expand family farming schemes but proposed that future schemes would not receive direct management services from TBZ; they would be run by Village Committees. The Committees would be responsible for - 57 - the provision of most inputs while TBZ would confine itself mainly to extension advice. The proposal constituted a major change in concept and was one of the reasons for the mid-term review. 6.08 The mid-term review concluded that successful implementation of the project required sound management and discipline during the settle- ment period covering at least the first five years. The report noted that once fully established it should be possible to maintain performance by providing farmers with competent extension only. In short, the mis- sion concluded that, to succeed, the project must be implemented as appraised. For the remainder of the project period, the Bank attempted to get the project back on track. As the marked lack of success on staff recruitment and the proliferation of new schemes show, the Bank's efforts did not prove fruitful. Although management agreed to steps to bring the project more in line with appraisal design, clearly management never accepted that the project should be implemented on a high-input, high-output basis. Missions felt that management could not be relied upon to carry through on its agreements. 6.09 At a meeting with representatives of the Ministry of Lands and Agriculture in February 1978, the mission was informed that (a) Govern- ment was totally opposed to paying the levels of salary necessary to retain or attract staff to the key positions being insisted upon by the Bank, (b) TBZ's future development should comprise the provision of extension to farmers rather than management of sophisticated schemes. Following this and the publication in January, 1978 of the PCR for the Commercial Crops Farming Development Project--which made some harsh comments about project management and the lack of firm action by the Bank (para 8.05)--a letter was sent under signature of the Bank's Projects Director to the Minister of Agriculture. It sought clarification of the Government's views on the future of the project. The letter pointed out that if Government wanted to change the design of the project to less intensive management and to withdraw mechanical and technical ser- vices to cut costs, then the Bank would be obliged to cancel the loan, but would be willing to consider a new project. If, on the other hand, the Government wished to continue the project as originally designed, then it was essentialthat management problems be resolved and the accounts improved. The Minister indicated that the project should go ahead as originally designed and that Government would assist TBZ to recruit key staff, arrange purchase of essential spares and put the accounts in order. Nevertheless, there remained in the Bank some doubt as to the ability of TBZ to put its house in order under existing management. It was agreed that the next mission should determine whether the loan should be suspended. 6.10 The September 1978 mission reported that there was some progress on staffing and accounts and that the project should not be closed at that time. The mission was, however, disturbed by comments from the General Manager that he was still not wholly convinced about the project concept. He commented also that, despite the Minister's assurances, he had not been working within appraisal guidelines and the legal docu- ments, since he still felt that the project concept was wrong. Memoranda of Understanding were drawn up during this and a follow-up mission in December on steps that TBZ agreed to take to put the project back on - 58 track. The memoranda included a commitment to follow the legal docu- ments, not to expand the number of schemes, to operate separate project accounts, to identify project equipment, reassess the needs of spare parts, employ specified staff, and monitor and evaluate the project. In short, some five years after effectiveness the Bank was still in a position of requesting that the project be implemented as agreed, and this despite violation of virtually every agreement in the legal documents. For the first time also, the Ministry of Lands and Agriculture appeared to be taking an active role in monitoring the work of TBZ. 6.11 The March 1979 mission accepted that TBZ had made some effort to implement the agreements. However, actual cost data provided by TBZ's economist became available to the Bank for the first time. From an analysis of actual costs in relation to benefits, it was clear that no adjustments to the project under existing poor TBZ management and excessive overhead costs would be likely to make the project viable. The mission pointed out that it was almost certainly costing Govern- ment more to run the smallscale sector of the tobacco industry than the returns to the economy. This view was supported by the findings of the audit of TBZ's 1977 accounts which had just been completed (para 6.17 and Appendix 1). In short, there had been insufficient improve- ment from the situation described in the PCR for the previous project, published over a year earlier, to justify on project grounds an extension of the closing date beyond December 1980. The mission endorsed the need for a full study of Zambia's tobacco industry and recommended that the project be stopped but that an extension be granted of six months to allow TBZ to draw down the loan against expenditures already incurred. The mission's recommendation was accepted. .12 Because of the state of the accounts, the Bank Informed Guvernment that all disbursement claims would have to be certified by qualified accountants. It was agreed that TBZ's auditors,would assign staff to assist TBZ. The outcome was disappointing. Contrary to agreements reached in March 1979, TBZ allocated insufficient accounting staff to prepare claims and these staff were frequently diverted to other tasks. TBZ's auditors requested concerned Ministries to ensure that TBZ provide adequate staff. Follow-up from concerned Ministries was also disappointing and ultimately some--US$3.A million were cancelled. Thus even at the final stages of the project there was a lack of follow-up and intercst that is difficult to explain. Financial Management and Accounts 6.13 The Tobacco Sector Report notes that "since 1974 there-has been no effective financial control of TBZ because the information needed to manage the Board has not been available within any reasonable period. Therefore, all the decisions that have been taken since that time have been approved by the Board in ignorance of the cost and revenue implications of such decisions. Thus the management of TBZ can only be described as irresponsible... Bearing in mind the opportunities and funds available to it, the performance - 59 - of TBZ as an institution has been appalling". The report states also that blame for this state of affairs should also be shared by the Ministries of Agriculture and Finance and the Bank. The PCR mission concurs with the general thesis and, in particular, believes that the Bank should have taken firmer action earlier (paras. 8.05-8.08). The mission would, however, add the Board of Directors of TBZ to the list of those partly responsible for TBZ's state of affairs (para 6.18). 6.14 The lack of financial control and planning is illustrated by the manner in which TBZ utilized funds received from Government to cover costs of special schemes, such as the project. Funds were allocated every three months, althouEh there were sometimes considerable delays. These "Identifiable" funds were then deposited in TBZ's general account where they were merged with other funds. TBZ's liquidity position and financial control was such that once in the general account, funds were used to meet the most pressing needs, usually wages and bills outstanding. There was little or no attempt to plan future expenditures and no attempt to keep project funds separate from the rest of TBZ's activities. Because the Bank was only able to monitor Government inflows, which were repeatedly cut, the Bank was under the impression that TBZ had spent considerably less on the family farming project than it had. The picture only clarified when TBZ's economist completed his work in March 1979. The following demonstrates where the additional funds to finance the project came from. In 1977, TBZ borrowed some K6 million from commercial banks to finance the purchase of the 1977/78 tobacco crop. However, instead of using the revenues from the sale of the crop to repay the loan, TBZ used the funds to meet their general capital and operating expenses, including financing the project. By early 1979, when it was time to purchase the 1978/79 crop, TBZ had "consumed" the crop finance loan. Commercial banks refused to lend addi- tional funds without a Government guarantee. TBZ was effectively bankrupt. The use of crop finance to cover TBZ's operating costs was serious mis- management of TBZ's financial affairs. 6.15 Accounts: The shortage of experienced and qualified manpower took its toll in the Accounts Department. As early as February 1974, super- vision reports noted that TBZ's accounts were neither prompt enough nor complete enough to enable adequate controls over expenditures. The remedy lay in recruitment of additional accounting staff. Unfortunately, senior TBZ management showed little interest in the accounting aspects of TBZ's operations so that, despite repeated promises to supervision missions, the Department was never adequately staffed. Predictably the quality of TBZ's accounts declined throughout .the life of the project. 6.16 The 1974 audit report contained serious qualifications with the auditors stating that they were unable to verify figures relating to certain fixed assets, stocks and write off of fertilizer stocks. At the Bank's suggestion, TBZ!s auditors were employed by TBZ to provide staff on a short term basis and to recommend a new simplified accounting system. While preparing the new accounting system, the auditors produced a series of status reports that repeatedly made reference to the shortage - 60 - of accounting staff and the lack of qualified personnel. For example, in 1975 they recommended the immediate appointment of a senior accountant to be responsible for the day to day operations of the Head Office Accounts Department. Such an appointment was considered unnecessary by the Ministry of Finance and was, therefore, not made. 6.17 The audited accounts for 1975 stated that the system of internal control was not adequate to minimize the risk of error and fraud and ensure the accuracy and completeness of proper accounting records, due mainly to shortage of qualified staff. The aueitors noted that they were unable to report that the books had been properly kept and that the accounts gave a true and fair view of the state of the Board's affairs as at December 31, 1975, a most damaging qualification. The Bank con- sidered that, with the introduction of the new accounting system, adequate steps were being taken to improve the situation, with the notable exception of staffing and, therefore, decided to take no firm action. The 1976 accounts again covered a period before the new accounting system was introduced. The publication of the 1977 accounts, however, in March 1979 was the final straw. Desoite the new accounting system, they were more qualified than ever before, stating, inter alia, that the auditors remained unable to report that the accounts gave a true and fair view of the state of the Board accounts; details are in Appendix 1. The extent of qualifications was such that there was still no effective financial control within TBZ. The March 1979 mission cited this as one of the main factors when it recommended that the project not be extended. The 1977 audit report showed that TBZ had made a loss of K4.1 million in the year, bringing its accumulated losses, after Govern- ment subsidies and interventions, to K18.2 million. Role of Board of Directors 6.18 Overall policy guidance for TBZ should have been provided by a Board of Directors, which consisted of 7 members nominated by the Minister of Agriculture and Natural Resources. In practice, the Board was largely ineffective and had little influence on the working policies and strategies of TBZ. To all intents and purposes, the General Manager rantheorganisation himself. Because of the poor state of TBZ's accounts and the lack of actual project cost data, the Board approved actions without the necessary information to enable it to assess fully the implications of their decisions. Moreover, when the General Manager, a Presidential appointee, announced decisions to, say, expand the number of schemes the Board was informed that the decision was approved by or in line with Party policy. Board members felt themselves unable to question such policy and there was thus little or no Board discussion on such matters. In short, it appears that the political environment was such that the Board could not provide an effective check on the proliferation of schemes. The use of parastatals to fulfil political objectives with little or no regard for financial or economic costs is discussed as an issue in paragraph 8.03. - 61 - VII FUTURE OF TOBACCO INDUSTRY 7.01 In 1979 Government learnt that TBZ had utilized its commercial bank loan to cover operating expenses (para 6.14) and it would, therefore, have to provide some K5 million, either directly or in the form of a bank guarantee, to enable TBZ to purchase the tobacco crop in 1980. Government also learnt from TBZ's economist that TBZ had spent far more than previously thought on family farming schemes. Coming at a time when Government was having difficulty financing its recurrent budget, tobacco production in Zambia was falling (Table 8), and Zimbabwe appeared likely again to become a major producer of tobacco, Government became increasingly concerned at the cost of keeping TBZ going since the returns to the economy were so small. In addition, two Bank-financed projects had manifestly failed to find means to produce tobacco economically under existing Zambian conditions. It was, therefore, agreed to commis- sion a study of the Tobacco Industry to make recommendations for the future. The major recommendations of the study, as it concerns the Family Farming project, are as follows:- (i) retention of 6 of the 24 virginia tobacco schemes (Zemba, Kaoma, Mkushi West, Nyawa, Kabile and Serenje). In 1978/79 these 6 schemes contained 55% of the far- mers and produced 70 of the tobacco on virginia schemes; (ii) conversion of 3 schemes (Sinde, Lumezi and Vubwe) to burley tobacco; these 3 schemes contained 15% of the farmers and produced 12% of the tobacco; (iii) closure of all remaining schemes; (iv) closure of all Provincial offices; (v) rationalization of staffing on the 9 schemes re- tained to reduce staff numbers substantially; (vi) winding up of TBZ and replacement by a new tobacco organization to include responsibility for pro- duction and marketing; (vii) progressive replacement of tractors by oxen on schemes not subject to tsetse-fly; (viii) establishment of a tobacco regulatory authority; and (ix) adherence by management to production methods designed to obtain high quality tobacco. 7.02 In sum,the study recommended closing all but the largest and best run schemes which between them accounted for over 80% of production. They proposed also that those schemes which were to continue be run by a small - 62 - but highly qualified management team aiming at constant supervision of farmers. These recommendations are in line with the 1972 appraisal mis- sion's proposals which envisaged a small number of schemes with a high degree of management and inputs--along the lines of the CDC-financed tobacco scheme in Malawi. The consultants would have preferred to see the joint participation of the private tobacco sector, and Government currently is discussing the possibility of management contracts with private firms. Most of the other recommendations have been accepted, and are being implemented. The key one is the abolition of TBZ's Production Department, which has begun and will be completed at the end of the present tobacco season. The study shows that under such conditions tobacco can be grown profitably and that it will compete favorably with other crops. The key, not surprisingly, is increased yields and reduced overheads. VIII. SPECIFIC ISSUES AND BANK PERFORMANCE Project Design 8.01 The PCR mission has no reason to believe that had the project been implemented along the lines proposed, it would not have been relatively successful. The change in profitability of tobacco vis-a-vis other crops would have reduced achievements somewhat but with improved yields, lower TBZ marketing charges and prompt payment, farmers could likely have con- tinued to grow tobacco successfully. This is certainly the conclusion reached in the Tobacco Sector Study.Unfortunately, the project was not run along appraisal lines for sufficient time to test whether the appraisal concept was sound. Nevertheless, the promise shown until about 1974(when the project was implemented rather closer to appraisal methods) and the more succes- sful CDC-financed smallholder project in Malawi would indicate that the project could have worked. While the PCR mission concludes that the appraisal design was basically sound, a number of issues are worth explor- ing in greater depth. First, whether to succeed the project required levels of management that were not available in Zambia. Even prior to nego- tiations the Bank learnt that several key staff were planning to leave TBZ and appointment of these key staff was made a condition of project effectiveness. The Bank also included a legal covenant requiring that it be informed of all appointments for the positions of Production Manager, Scheme Managers and Provincial Managers to enable it to comment. As it turned out TBZ did not inform the Bank when staff were appointed to these posts and the Bank did not enforce the covenant. The PCR mission finds that the covenant was too weak to ensure that staff posts were filled with competent individuals and that the Bank should have retained its right of approval, as envisaged prior to negotiations. The mission also con- cludes that, until mid-1975 when the findings of Mwanakatwe Commission were apparent, there were sufficient staff available in Zambia to imple- ment the project as designed. From mid-1976, for the project to have succeeded would have required either that the Mwanakatwe recommendations were waived or that the project be reappraised. - 63 - 8.02 The appraisal concept was based on a high level of inputs (management, tractor power, extension supervision, and farm inputs) to generate high outputs (yields) which gave a satisfactory economic rate of return, Nevertheless, the project was relatively costly in terms of development expenditures per farm family, about US$2,600 in 1972 prices. This would be equivalent to some US$7,000 in 1980 prices, assuming an annual rate of inflation of 13%. Could the project have been designed to benefit more farmers at lower cost, while achieving a satisfactory economic rate of return? This is what senior management in TBZ believed. As stated, neither the high-input, high-output or the low-input, low-output uethods were effectively tested. The project suf- fered from the worst of both, very high-input costs and relatively low yields (once allowance is made for TBZ overestimates of yieldi. The Tobacco Sector Study concluded that tobacco cannot be grown economically on a low-input basis. The PCR mission does not feel itself equipped to comment on whether the low-input approach favored by project management was feasible. Certainly, it would not have been possible to s-itch from one approach to another without a reappraisal. 8.03 The final issue regarding design, relates to the implementation of projects through parastatals. The appraisal mission anticipated the need for the recapitalization of TBZ to set it up on a sound financial footing and Government agreed to provide plans accordingly. This covenant was never complied with and, at the end of the project, TBZ was effectively bankrupt. The mission believes that even if TBZ's capital structure had been put right in 1973, TBZ's financial performance would have been little better. A General Manager was appointed on political grounds, an expatriate in the Party, who had no experience in managing large commercial organisations; in fact, he ran it akin to a Government Department and opened new schemes in remote areas to satisfy political objectives. Regrettably TBZ's performance is all too typical of the performance of other parastatals in Zambia and Eastern Africa. Based on this experience, the question the PCR mission poses is under what conditions can a parastatal be expected to perform satisfactorily. The mission proposes that a review be carried out of parastatals sup- ported by the Bank in Eastern Africa to assess (a) what criteria (politi- cal, managerial, financial) are necessary to ensure that they can operate as financially responsible organisations and (b) if some of these criteria cannot be met initially, whether the Bank should invest through them. Supervision Performance 8.04 The project was supervised regularly with 17 formal supervision missions visiting the project over a 7-year period. The first 8 missions supervised the project jointly with the Commercial Crops Farming Develop- ment Project (Loan 685-ZA). There was a major mid-term review mission in late 1976. With the exception of the proliferation of schemes and - 64 - sections, which was allowed to proceed largely unchecked, the PCR mis- sion finds that supervision missions focussed on the major issues. Despite doing so, the project was a failure. The following summarizes why the Bank's efforts proved fruitless:- (i) Relative Returns to Tobacco and Maize: Throughout the project, the return to tobacco declined vis-a-vis other crops. (ii) Project Concept: Following the deterioration in Zambia's economic conditions, Project Management, and to a lesser extent Government, believed that the project concept and design were too sophisticated and expensive for Zambian conditions. The Bank, on the other hand, indicated fol- lowing the mid-term review that, for the project to suc- ceed, it should be implemented, as agreed, on a high-in- put, high-output basis. In this light, the Bank made recom- mendations to bring the project back on track. Project Management appeared to accept mission recommendations and missions would report that remedial action had been agreed. But TBZ rarely made serious efforts to imple- ment agreed actions. Examples are TBZ's failure to: set up a revolving credit fund, carry out a study of spare parts required for tractors, and recruit staff as agreed. It is not clear why management agreed to these and other stens if it had no intention of carrying them out. (iii) Lack of Support and Follow-up: Until the final stages of the project, concerned Ministries showed little interest in the project and, for example, did not assist TBZ to overcome its staffing problems. TBZ's deteriorating financial status does not appear to have been monitored by the Ministry of Finance although brought to its atten- tion by the Bank. Moreover, TBZ's Board of Directors proved ineffective in providing policy guidance and in monitoring the affairs of TBZ. (iv) Mismanagement of TBZ: Despite the disagreement over project concept, TBZ's affairs have been mismanaged over the past six years, as illustrated by the absence of financial control and the state of TBZ's accounts today. TBZ management did not accept that TBZ needed to be run as a financially sound institution and saw TBZ as akin to a Government Department. (v) Lack of Data: Until the project economist produced cost data in early 1979, neither the Bank nor Government had information on actual expenditures, physical construction, and staffing numbers and costs. Nor did TBZ submit evalua- tion reports or annual plans for the development of schemes - 65 - and sections. Thus for 5 years the Bank supervised a project without the benefit of actual data on performance. It is instructive that as soon as the Bank received this information, a decision was taken to close the project. Lessons for the Future 8.05 The PCR m"ssion believes that the conclusions from the above are clear: that the Bank generally identified the problems but lacked quantitative data to assess just how costly and inefficient TBZ's opera- tions had become. Without such data, supervisions could not be fully effective and, could not make, for example, a sound judgement of continued economic viability of a project. Instead the Bank tried to accommodate TBZ and went along with a model that subsequently led to the wrong deci- sion. The Bank should have taken action to ensure that cost and other data including, annual plans, was provided. In this context, the Bank should not have allowed: the breach of covenants, the lack of management follow-up on agreed actions, and the chaotic state of TBZ's accounts to have continued without taking firm action. At the end of the project, virtually every major covenant. in the project was breached; details are in Appendix 2. The PCR mission's conclusions are very much in line with those reached by the PCR for the Commercial Crops Farming Development Project which stated "It is not clear why, with the Management of TBZ being in such total disarray over a long period of time, no action was taken by the Bank to withhold disbursemerts of funds until such time that management was pro- perly strengthened and accounts put in order .... If there is one lesson to be learnt from this project, it is never to allow a project to slide downhill with ever increasing speed without attempting to apply the brakes. The Bank's apparent flexibility towards project implementation was interpreted by TBZ as a licence for complacency and ineptitude. Where project management, or for that matter the Government, fail to make the necessary corrective measures, the Bank should take the neces- sary steps (such as withholding disbursements of funds) until such time that the project is brought back on its intended course". 8.06 It might be asked why the Bank did not suspend disbursements of the Family Farming Project following the comments in the January 1978 PCR about applying the brakes. As noted, shortly after the PCR was issued, the Bank wrote to the Minister asking whether Government wished to continue with the project as appraised or to redesign the project, in which case the existing loan would be cancelled and a new project prepared. Following a positive response from the Minister in June, 1978, the Bank felt it had to allow sufficient time to monitor what, if any, remedial action would be taken by TBZ management. To this end, a mission visited Zambia in September, 1978. It drew up a Memorandum of Understanding covering action to be taken by TBZ, recommended that the project should continue and proposed that an interim mission visit Zambia in December, 1978 to check on implementation of the agreements - 66 - reached. The December mission drew up a further :emorandum of Under- standing and again recommended that the project continue. In short, discussions about firm action were prompted by the earlier PCR and throughout 1978 the Bank kept the project under close review, in particular whether or not it should suspend disbursements. The final straw came in March 1979 when the TBZ economist provided data on actual expenditures and it was clear that existing TBZ management would not be able to take the drastic steps necessary to make the project viable. The Bank decided not to extend the closing date. There appears to be a sound explanation of why the Bank was reluctant to take firm action in the period 1976 to early 1978 (although the PCR does not concur with the reasoning behind the explanation). Supervision missions under- stood that because of strained relations that existed between *the Bank and Government of Zambia during that period, they should not exacerbate tensions by recommending action that involved, say, suspension of dis- bursements. Accordingly, they were always willing to give TBZ the benefit of doubt and to allow further time for past promises to be met. The PCR mission was unable to determine whether the understanding of super- vision missions was based on specific instructions from within the Bank or whether they were perceptions of what missions believed was Bank policy on Zambia. The lesson that emerges is that supervision missions should recommend what is best on project grounds. Where firm action is warranted, it should be put on record. If their assessment is over- ruled on country grounds that is another matter. 8.07 The PCR mission was unable to find evidence that firm action would have caused a marked deterioration in Zambian-Bank relations. Indeed, some Government officials informed the mission that the Bank should have been tougher. It would then have highlighted to senior Government officials the need to take drastic action. Even if it had caused a deterioration, does the Bank do more harm by continuing with a project that is performing poorly than by forcing Government to look squarely at a deteriorating situation? IX. CONCLUSIONS 9.01 The objectives set out in the appraisal of this Project were not achieved; the project generated neither the production nor the rate of return envisaged. A project's success must also be measured by its sustainability--whether accomplishments, despite being smaller than anticipated, can be sustained and expanded in the future. The project as implemented has not provided a model for future development and it is likely that 15 of the 24 schemes started under the project will be closed. On the institutional front, TBZ failed to develop as an institu- tion capable of fostering tobacco production and providing marketing services. - 67 - 9.02 The project concept was based on a linkage between high- inputs and high-outputs. In particular, it relied upon intensive management by qualified and experienced staff. The project was never implemented as designed so it is not possible to conclude definitively that the project would have succeeded. The evidence, however, suggests that the project design was sound. In that belief, the Bank tried to encourage TBZ to implement the project as agreed. Neither TBZ nor Government were willing or able to take the necessary steps to put the project back on track. The absence of firm remedial action resulted in the untimely provision of farm inputs and late payment of farmers. This, added to a deterioration in the profitability of tobacco when compared to other crops. meant that farmers were not particularly interested in growing tobacco except to obtain inputs for maize. The lesson from this project is similar to that for the Commercial Crops Farming Development Project. The Bank should not allow a project to slide downhill indefinitely without applying brakes. The absence of annual plans, data on actual project costs, andevaluation reports meant that the Bank could not effectively supervise the project. This led to the proliferation of schemes and sections with little regard to cost. Had this data been available to supervision, the PCR mission believes the situation would not have been allowed to deteriorate to the extent it did. The PCR mission finds that the project made too little use of deadlines. Too often, for example, Aide Memoires specified that TBZ undertook to a6here to legal covenants. Rather they should have provided a deadline for covenants, such as the provision of cost data, to be met. If they were not, the Bank should have suspended disbursements. The Banl's unwillingness to take firm action appears to have stemmed from a concern not to upset Zambian-Bank relations which were reportedly poor in the period 1976 to 1978. The PCR mission believes that in any event recommendations should, as far as possible, have been made on project grounds. The failure to stem the tide has resulted in the current winding-up of TBZ--an institution which the Bank has supported through this and the Commercial Crops Farming Development Project.  - 69- Table 1 ZAMBIA INTEGRATED FAMILY FARMING PROJECT COMPLETION REPORT Buildup of Farmers and Schemes Scheme 1/ Original Scheme 2/ 71/72 72/73 73/74 74/75 75/76 76/77 77/78 78/79 79/80 Name - PREPROJECT - ------------------------- PROJECT ------------------- Zemba Zemba 243 402 587 720 832 909 949 891 874 Sindemisale Sendemisale 58 109 146 163 204 262 299 (165 (114 Vubwe(15 14 Lumezi Lumezi 16 47 79 102 126 131 158 171 1,25 Luwerezi Luwerezi 22 120 186 129 98 119 149 162 176 Singani ) (126 ( 75 Sipatunyana )Gaela 71 144 252 314 353 331 338 42 8 Chikanta )Gea71- '' (24 (15 Siachitema ) 88 ( 97 Kaoma Kaoma 22 90 176 227 259 343 378 411 Serenje Serenje 75 144 184 198 214 252 257 249 176 Kabile )abile 29 35 45 51 78 88 105 (117 (131 Kaindu ( aie2 5 45 5 8 8 0 57 (76 Chipangali Chipangali 8 16 24 39 37 52 28 Mpongwe Mpongwe 8 16 21 34 31 30 50 Mkushi West Mkushi West 8 33 59 100 134 ill 112 Munkongwe Munkonge 3 26 42 53 80 100 79 Sinda Sinda 19 30 68 87 92 96 Kabompo Kabompo 15 12 21 40 44 44 Nyawa Nyawa 16 48 89 107 111 Mulungushi Chibwe 8 12 19 20 Chinsali Chinsali 10 26 Kashima Kashima 7 11 Mansa Mansa 40 58 TOTAL FARMERS RECRUITED 514 1,023 1,596 1,978 2,336 2,722 3,108 3,235 3,057 TOTAL RECRUITMENT EXCLUDING PREPROJECT 3/ - - 573 955 1,313 1,699 2,085 2,212 2,034 APPRAISAL ESTIMATES OF RECRUITMENT 910 1,650 2,790 4,050 5,190 6,290 7,310 8,110 APPRAISAL RECRUITMENT EXCLUDING PREPROJECT - - 74o 1,880 3,140 4,280 5,380 6,400 7,200 ACTUAL RECRUITMENT AS % OF APPRAISAL (BOTH EXCLUDING PREPROJECT FARMERS) - - 77 51 42 4o 39 35 28 Number of Schemes 7 8 12 14 15 16 16 2h 24 Source: TBZ economist, ' Tobacco Sector Re:ort and Appraisal Report. 1/ Schemes presented in same order as TBZ statistics. This is generally the order in which schemes were started up. 2/ Three schemes (Sindemisale, Gamela and Kabile) were subdivided in 1978/79 as shown. 3/ Shows number of farmers recruited during the project period, i.e. excludes those recruited prior to project startup. ZAMBIA INTEGRATED FAMILY FARMING PROJECT COMPLETION REPORT Tobacco Yields (kg/ha) 1/ Scheme Province 71/72 72/73 73/74 74/75 75/76 76/77 77/78 78/79 --------------------- Yields 1/ (kg/ha) --------------------------- Zemba Eastern 819 992 775 953 762 912 582 1,027 Sindemisale) Eastern 686 1,108 995 1,222 856 791 634 837 Vubwe ) (atr1,0'(1,o8o Lumezi Eastern 443 830 661 1,014 619 618 756 558 Luwerezi Eastern 1,324 593 516 773 762 656 496 402 Singani ) ( 606 Sipantunyana Southern 888 744 638 550 595 496 879 Chikanta ( 713 Siachitema ) (1,135 Kaoma Western 1,116 1,137 1,070 1,244 943 1,611 Serenje Central 987 1,298 601 1,104 960 1,038 691 852 Kabile Central 681 638 747 750 705 760 (1,234 Kaindu ) Cnrl70'~~ ( 860 Chipangali Eastern 1,443 923 777 731 1,049 Mpongwe Copperbelt 674 674 644 576 710 1,055 Mkushi West Central 1,626 1,569 1,397 957 1,271 0 Munkonge Northern 674 831 683 937 556 523 Sinda Eastern 996 1,056 578 513 741 Kabompo North Western 873 1,362 1,441 745 520 Nyawa Southern 1,337 1,412 1,260 1,240 Mulungushi Central 751 151 621 Chinsali Northern 667 Kashima North Western 1,669 Mansa Luapula 362 Average 837 963 755 942 797 880 668 986 Source: TBZ economist and Tobacco Sector Report. I A 1/ Actual yields are lower than indicated, be.cause of TBZ underestimates of farm size (see para, 4.04 of' PCR). ZAMBIA INTEGRATED FAMILY FARMING PROJECT COMPLETION REPORT Indicators of Efficiency by Schemes Ratio of Scheme Total scheme tobacco Ratio of Farmer Salaries & cost per kg planted farmers to Number Yields Price wages per ha tobacco to salaried salaried Farmers Settlement Extension Kg/ha K/Kg tobacco (K/ha) (K/kg) staff (ha) staff 1979/80 1979/80 1979/80 1978/79 1978/79 1979/80 1978/79 1979/80 1979/80 (1) (2) (3) (4) (5) (6) (7) (8) (9) Zemba 874 780 - 1,027 1.17 350 0.8 2.7 5.3 Sindemisale ) 114 - 114 837 0.98 168 0.5 2.0 4.1 Vubwe ) 114 114 - 1,080 0.82 - - - Lumezi 125 116 9 558 0.77 1.3 Luwerezi 176 22 154 402 0.90 1.3 Singani ) 75 - 75 606 0.78 7.7 Sipantunyana) 38 - 38 879 0.73 1,277 3.8 0.9 1.9 Chikanta ) 15 - 15 713 1.36 1,387 7.2 1.0 Siachitema ) 97 - 97 1,135 1.28 992 2.5 0.9 1.6 Kaoma 411 311 100 1,611 1.05 1.2 Serenje 176 132 44 852 1.01 1.6 Kabile ) 131 109 22 1,234 1.02 808 2.1 1.4 2.2 Kaindu ) 76 - 179 860 0.97 1,245 2.9 0.7 1.7 Chipangali 28 28 - 1,049 0.80 1,393 1.4 0.7 1.5 Mpongwe 50 50 - 1,055 0.69 1,180 4.6 0.95 2.1 Vkushi West 112 100 12 1,271 0.95 790 0.7 1.4 3.4 Munkonge 79 51 28 523 0.66 2.8 Sinda 96 79 17 741 0.98 520 2.3 1.6 3.0 Kabompo 44 44 - 520 0.77 2,389 2.5 0.7 2.2 Nyawa 111 63 48 1,240 1.06 1.6 Mulungushi 20 - 20 621 1.32 898 3.1 1.0 2.2 Chinsali 26 16 10 667 1.13 975 - 0.9 2.6 Kashima 11 11 - 1,669 1.05 3,860 - 0.2 1.1 Mansa 58 41 17 362 1.17 2,924 - 0.5 1.9 Total/Average 3,057 2,067 990 986 1.04 670 1.5 3.1 Note: (1) Selection of Years. Breakdown of number of farmers into settlement and extension is only available for 1979/80. Similarly, data on number of staff per scheme and wage bill required to calculate columns 6, 8, 9 also only available for 1979/80. Other data only available to 1978/79. (2) Source: TBZ economist and Tobacco Sector Report. ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) COMPLETION REPORT Key Indicators 1973/74 1974/75 1975/76 1976/77 1977/78 1978/79 1979/80 Number of Schemes 12 1/ 14 15 15 16 24 24 Number of Families - actual 1,596 1/ 1,978 2,336 2,722 3,108 3,271 3,057 Review projections 2/ - - - 2,600 2,900 3,200 - Appraisal estimate 1,650 2,800 4,050 5,200 6,300 7,310 8,110 % of appraisal estimate 96 71 58 52 49 45 38 Total Area (ha) - actual (tobacco) 850 1,127 1,343 1,414 1,560 1,532 1,385 Review projections 2/ - - - 1,550 1,750 1,900 - Appraisal estimate 1,050 1,850 2,800 3,900 5,000 6,060 7,020 % of appraisal estimate 81 61 48 36 31 25 20 Area per Farmer (ha) - actual .53 .57 .57 .52 .50 .47 .45 Review projections 2/ - - - .60 .60 .60 - Appraisal estimate .64 .66 .70 .79 .79 .83 .87 % of appraisal estimate 83 86 81 66 63 57 52 Tobacco Yield (kg/ha) - actual 3/ 755 942 797 880 668 981 n/a Review projections 2/ - - - 820 850 875 - Appraisal estimate 735 770 810 815 855 855 910 % of appraisal estimate 102 122 98 107 78 114 Tobacco Production ('000 Kg) - actual 641 i,o61 1,070 1,245 1,041 1,503 n/a Review prajections 2/ - - 1,270 1,490 1,660 - Apnraisal estimato 770 1,400 2,230 3,180 4,260 5,370 6,400 , of appraisal e1tv'b 83 76 8 39 2h 28 .obacco Price (W/Kg) 88 76 101 101 115 - Source: TBZ Economist 1/ In 1972/73 there were 8 schemes with a total of 1,023 farmers. Thus, for example, in 1975/76 there were 1,313 incremental families joining the tobacco/maize schemes as a result of the project. 2/ Mid-term review mission: October 1976. 3/ Actual yields are lower than indicated, because of TBZ underestimates of farm size (see para 4.04 of PCR). - 73 - Table 5 ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) COMPLETION REPORT Estimated Project Costs (K '000) Lusaka and 70% of 1/ Provinces HQ Cost Schemes Total Appraisal A. Credit 2/ Medium Term - - 737 Incremental Seasonal - - 1,067 Baling Sheds - - 117 Sub-total 1,92-1 B. Project Investments 3/ Vehicles and tractors 640 450 840 1,290 1,046 Specialized equipment 80 60 450 510 540 Housing and Buildings 220 150 870 1,020 1,077 Dispensaries and Schools 10 10 90 100 169 Water, roads, soil conservation and forestry 210 150 700 850 1,340 Operation of tractors - - - - 179 Other 490 340 200 540 - Sub-total 1,650 1,160 3,150 4,310 4,351 C. Project Management/Operating Costs 3/ Wages and Salaries 3,840 2,690 3,540 6,230 3,000 Administration 240 170 100 270 3t7 Vehicle Operating Costs 1,280 900 1,900 2,800 1,042 Road Improvement - 200 200 70 Improvement to buildings - - - - 135 Other 500 350 630 980 - Sub-total 5,860 4,110 6,370 10,480 4,614 Total Project Costs 8,380 5,270 9,520 14,790 10,886 1/ Assumes that 70% of activities in TBZ's Production Department and in Provincial headquarters related to work on Family Farming Schems (estimate by TBZ Economist). 2/ Project economist was not able to estimate credit provided under the project. Credit components sub- sumed under other items. 3/ Allocation between items is not accurate because many ledger entries were made to wrong account head. Notes: 1. Source: Derived from data provided by TBZ Economist. 2. Project Period: Actual costs relate to period 1973/74 - 1978/79, while appraisal data relates to period 1973 - 1977 inclusive. - 74 - ZAMBIA Table 6 INTEGRATED FAMILY FARMING PROJECT Completion Report SCHEDULE OF CUMULATIVE DISBURSEMENTS Appraisal Actual as % of Estimate Actual Appraisal Estimate ---------------- US$ Million ---------------------- 1973/74 December 31 2.8 0 June 30 4.1 0 1974/75 December 31 5.0 0 June 30 6.5 1.1 17 1975/76 December 31 7.4 2.9 39 June 30 8.7 2.9 33 1976/77 December 31 9.6 3.9 41 June 30 10.9 3.9 36 1977/78 December 31 11.5 3.9 June 30 - 4.4 38 1978/79 December 31 11.5 4.4 38 June 30 2/ 11.5 4.4 38 1979/80 December 31 11.5 5.1 44 January 31 11.5 7.2 63 March 24 3/ 11.5 7.73 67 1/ Appraisal Completion Date. 2/ Appraisal Closing Date. 3/ Final closing date was March 24, 1980. Balance of funds cancelled. Although closing date extended by six months to December 31, 1979, actual payments were made after December on (a) disbursement applications received by December 31 and (b) disbursements to the auditors who drew up dis- hursement claims anl were onlr PhlA to submit fees after December 31. - 75 - Table 7 ZAMBIA INTEGRATED FAMILY FARMING PROJECT Completion Report Returns to Tobacco and Maize (1978/79) Tobacco 1/ Hybrid Maize 2/ Local Maize I ha 1 ha 1 ha Yield (Kg) 850 3,000 1,080 Price (Kg) 3/ 1.0 .10 .10 Gross Return (K) 850 300 108 Variable Costs 3/ (K) Tractor Rental 4/ 42 33 - - ploughing (23) - discing ( 1) - ridging (10) - transport ( 8) Planting Material 62 9 4 Fertilizer, Insecticides 55 50 - Other Materials 28 6 3 Insurance 17 - 1 Interest (12% p.a. - 10 month ) 24 12 - TOTAL VARIABLE COSTS 226 110 8 Gross Margin 624 190 100 Man-days 780-620 90 85 Gross Margin/Man-day 0.8-1.0 2.1 1.2 1/ Costs are based on analysis of credit record of farmers from Zemba, Sinda- Misale, Munkonge, Kabile, Mkushi West, Koama and Nyawa schemes. Costs reflect actual cash (credit) costs paid by farmers to TBZ for goods and services. Does not include labor costs. 2/ As footnote 1, but adjustment made to tractor-rental cost to reflect cost that would be incurred if TBZ actually ploughed farmers' maize land. In practice, TBZ only -plough about 15% of area sown to maize. 3/ Constant 1978/79 Kwacha. 4/- TBZ charges K 22.5/ha for ploughing,'K 11.2/ha for discing and K 11.2/ha for ridging. Thus only 10% of farmers' tobacco land is disced and 85% ridged. Source: Data Provided to March 1979 Supervision Mission by TBZ economist. - 76 - Table 8 ZAMBIA INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) COMPLETION REPORT Tobacco Purchased by TBZ ('000 Kg) Grower Category 1973 1974 1975 1976 1977 1978 1979 Commercial 2,350 2,000 2,000 1,100 1,400 764 975 Tenants 1,735 1,810 1,574 1,745 1,506 945 1,206 Assistant Tenants 743 857 1,029 1,350 1,299 797 680 FFP 501 642 1,062 1,071 1,245 1,047 1,410 Other 980 980 1,020 1,200 300 153 319 Total 6,308 6,239 6,683 6,466 5,750 3,707 4,590 - 77 - Appendix 1 Page 1 ZAMBIA INTEGRATED FAMILY FARMING PROJECT Completion Report TOBACCO BOARD OF ZAMBIA REPORT OF THE AUDITORS FOR YEAR ENDED 31ST DECEMBER, 1977 1. Our report on the accounts for the year ended 31st December 1976 stated that:- (a) There was insufficient control over the day to day transactions of the board; (b) We had been unable to: (i) Verify the net book value of K7,826,248 of the Tobacco Production Department fixed assets, and satisfy ourselves as to the fair statement of the depreciation charge thereon of Kl,002,474; (ii) Satisfy ourselves as to the recovera- bility of advances of K2,863,h34 to tenants; (iii) Satisfy ourselves as to the complete- ness and accuracy of scheme creditors of K519,554. (c) Subject to the reservations in paragraphs (a) and (b) above, the accounts gave as true and fair a view of the Board's affairs at 31st December, 1976 and of its loss for the year ended on that date as is possible from the information available. 2. In our opinion the Board's system of internal control was not adequate during the year under review to minimise the risks of error and fraud, safeguard the Board's assets and ensure the accuracy, reliability and completeness of the accounting records. 3. We have been unable to: (a) Satisfy ourselves as to the correctness of the accounting treatment of the inter departmental differences, amounting to K1,173,678 which has been written off as an exceptional item, due to the non-availability of information to identify the exact nature of these differences; - 78 - Appendix 1 Page 2 (b) Satisfy ourselves as to the correctness of miscellaneous income of K772,037 credited as an exceptional item due to the non availability of adequate details and the information required by us for its verification; (c) Verify the net book value of fixed assets included in the accounts at an amount of K8,424,128 and the depreciation charge thereon of Kl,109,078 because records detailing cost and accumulated depreciation of individual assets have not been maintained. Furthermore, the results of the physical verification of the Tobacco Production Department fixed assets included in the above figure at their net book value of K6,983,249 disclosed significant differences as detailed in note 4(c); (d) Verify the recoverability of advances of K2,483,789 to tenants included in debtors and prepayments of K6,351,680 after provisions for doubtful recovery thereon of K984,877 in the absence of specific confirmation from the Government of the Republic of Zambia that they accept liability under guarantees detailed in note 4. (e) Verify the completeness and accuracy of scheme creditors amounting to K618,477 included in creditors and accruals of Kh,160,472 as the information required by us was not made available; (f) Determine the Board's liability to lending institutions, and the contingent liability arising from guarantees given on behalf of farmers; (g) Satisfy ourselves as to the accuracy of cash and bank balances of K72,319 due to material differences between the Board's and the banker's records not being reconciled; (h) Verify the existence and valuation of stocks amounting to K1,667,597 due to the material differences between the physical verification carried out by the Board's staff and the stock records not being explained and the information required by us to establish the accuracy of the pricings of stocks not being made available. - 79 - Appendix 1 Page 3 4. For these reasons, we are unable to report that:- (a) We have received all the information and explanations; (b) The books have been properly kept; (c) Proper returns adequate for the purposes of our audit have been received from branches not visited; (d) The accounts give a true and fair view of the state of the Board's affairs at 31st December, 1977 and of its deficit for the year ended on that date, and comply with the Tobacco Act, 1967. LUSAKA, 20th March, 1979. u a e - 81 - Appendix 2 ZAMIA Page 1 INTEGRATED FAMILY FARMING PROJECT (LOAN 882-ZA) COMPLETION REPORT 'Ron or only Partial Compliance with Major Covenants Loan Agreement 3.01(a) Borrower to ensure TBZ nerforms Both MOF and MOA failed to in accordance with Project Agree- control effectively the ment and Project Financing Agree- operations of TB'L, and in ment. Borrower to provide funds particular, to comply with necessary to enable TBZ to per- conditions of BA. Severe form its obligations, budgetary cuts hampered operations of TBZ. 3.02 Prior to January 1 in each year No programs ever submitted. a program for the development (See Section 2.03 of Project of project schemes, sections Agreement). and pilot sections to be agreed between Borrower, TBZ and Bank (programs shall include, inter alia, proposed locations of project sections to any project schemes and estimated financial requirements). 3.03 Borrower shall maintain and Until after BCR mission, operate schools and dispensaries TBZ failed to provide Bank constructed under project. with number of schools and dispensaries constructed under the project. Missions unable therefore, to check detailed compliance. 3.04 and Borrower shall maintain roads See comment under 3.03 3.03(b) of and water supplies constructed (above). Shortage of funds Project under project. resulted in poor maintenance Agreement of roads. 3.05 Borrower shall ensure that far- Project farmers enjoyed mers under project shall not subsidies in same way as benefit from subsidies accorded other growers. to other growers. 4.02 Borrower shall (i) submit a Proposals never submitted detailed proposal to the Bank to Bank. by June 30, 1973, for the re- capitalLzation of TBZ to establish TBZ on a sound financial footing and (ii) implement or cause TBZ to implement such proposals - 82 - Appendix 2 Page 2 Loan Agreement (cont'd) 4.02 (cont'd) according to a schedule to be agreed with the Bank. 5.02(a) Borrower shall furnish to the Not complied with. See Bank all such information Sections 2.06(b), 2.06(c), on administration, operations 4.01 and 4.02 of Project and financial conditions, Agreement. resources and expenditures of TBZ as Bank shall reasonably request. Project Agreement 2.01 TBZ shall carry out project TBZ administrative (e.g., with due diligence and efficiency record keeping) and financial in conformity with appropriate practices not in conformity administrative, financial and with appropriate practices. agricultural practices. 2.02(a) and TBZ shall lend to farmers in Most requirements of this Schedule 1 accordance with Schedule 1 covenant violated. which covers eligibility of farmers, seasonal and medium- term loan amounts and conditions, documentation and methods of loan recovery. 2.02(b) TBZ shall establish a revolving The revolving credit fund credit fund in which TBZ will has never been properly deposit all repayments of established or maintained. principal and interest. 2.03 TBZ shall (i) each year, pre- Development plans never pare and submit to the Bank submitted to the Bank. for their agreement a program for development of Project Schemes and Sections including estimated financial require- ments and (ii) implement agreed programs. 2.06(b) TBZ shall (i) maintain records Financial and physical adequate to record progress record keeping in TBZ of the Project (including totally inadequate to project costs) and to identify record progress of project. the goods and services Information requested -83- Appendix 2 Page ' Project Agreement (cont'd) 4.01 TBZ shall maintain adequate TBZ Auditors note that records to reflect in accor- they were unable to report dance with consistently that (i) bzoks were pro- applied accounting practices perly kept and (ii) ac- its operations and financial counts gave a true and condition. fair view of TBZ's affairs. In sum,/ financial records not adequately kept. 4.02 TBZ shall furnish to Borrower Accounts and audit notes and Bank not later than six always in arrears. months after end of each fiscal year (i) certified copies of financial statements and (ii) audit report. 4.03 TBZ shall annually review Bank not informed in direct costs of marketing advance. Zambian auctions of tobacco and shall inform no longer held. Borrower and Bank prior to opening of auction floors on the proposed level of marketing charges for that season. Project Financing Agreement 3.04 Funds made available to TBZ Equipment, vehicles and shall be applied solely to staff provided under the the project. project were used on other projects within TBZ. 3.05 TBZ to submit to Government TBZ frequently over 2 years within one month following in arrears in submission of month in which expenditure disbursement claims. TBZ was incurred, documents to also was unable to dis- enable GRZ to apply for tinguish between seasonal reimbursement; documentation and medium-term loans. to distinguish between seasonal and medium-term loans. Appendix 2 Page 3 Project Agreement (cont'd) 2.06(b) (cont'd) financed under the project by Bank supervision mis- and (iii) furnish to the sions was not supplied by Borrower and Bank all such TBZ on: project expendi- information as shall be tures, physical infrastruc- reasonably requested ture completed, and equip- ment purchased and on site. 2.06(c) TBZ shall prepare annual Evaluation reports not project evaluation and pro- prepared. Progress reports gress reports and shall, not prepared systematically before January 31, each year and never by deadline. Super- submit report of previous vision missions were, how- year to Borrower and Bank. ever, usually handed brief progress report (more qualitative than quantitative) covering period since last mission. 3.01(a) TBZ shall conduct its opera- Several senior management tions under competent and positions remained unfilled experienced management in for prolonged periods. TBZ accordance with appropriate was not an effectively administrative and financial managed institution and, practices. in particular, its financial affairs were not conducted in accordance with normally accepted practices. 3.01(b) TBZ shall recruit qualified Staff changes rarely staff for the positions of intimated to Bank. Poor Production Manager, Manager performance at some Schemes of Family Farming Schemes and Provinces would indicate and Provincial Managers. that several Scheme and TBZ shall also inform Bank Provincial Managers were of proposed appointments to not well qualified. these positions to enable Bank to comment. 3.03(a) TBZ shall at all times operate, TBZ's equipment, including maintain and repair its faci- vehicles, was poorly lities, equipment and machinery maintained. A study, in accordance with sound requested by Bank supervi- engineering and management sion missions, to identify practices. requirements of spares for vehicles was never completed. $$* 30' -K ZAMBIA _-- TANGANYIKA T A N Z A NI A TOBACCO BOARD OF ZAMBIA SCHEMES RESERVE Mpulungu Mbaa * C*j*ý PROPOSED PROJECT AREAS (LOAN 882-ZA) Nchelenge > ~ 1,050 ASSISTED TENANT SCHEME (A.T. SCHEME) ,Nko,d [ TENANT FARMING SCHEME (.ESCHEME) \ K-wabw N O R T H E N STAFF TRAINING SCHEME (5 ISCHEME) -AMILY FARMING SCHEME (FILE CURED VIRGINIA TOBACCO) -_~wmgu FAMILY FARMING SCHEME (BURLEY TOBACCO) Kasama Isoko M' fup. ChPnsl. - 138 M.... a.uA I n43a5r Samya 4 Mwnung T. Lubumbesh, LUav luw ZI • nMp,ku 2r A N GO L As 40 -- Bancro 1 N Chingo[o Kaluishi l Z D20 Luanshya. N O R T H 39 W E H T E R N C OPP E R B ELT Bolovale Ka mp. - - K.bompo 10m Mksh27 r 62E Ch-ft ..,, Musma, - - 9 -- rK-Kapr- ZEMBA 24 VUBWE 2 ) 46 'Katele 23E 51MIRCA INDE/AMSILS 47[ s 55 1Petauke 25 EMSL A KF UE 35Kubwe e5 -- - - -. s~~'-- .IV/S--IF---" ---- --- AIANTOIALRUNA5 Mankoya CKAOMA ,HlS MAP 13.lAE ON -D 39S, A 1 1977 Mumb-3 R, Ex~sNG SCHEMES HAS BEEN UPDa c ee AE R N M O Z A M B l Q U E \A,~~ on8 ,,. 4 4 Z l M B A B W E S O U UT H E R N 17 52 13 Pemb Korib. uai SIACHITEr -, S~ny-gw RAllWAYS sehee boM mb RIVERS AND LAKES Mata- -PROVINCIAL BOUNDARIES A R C -SmgsA?) IERNAIONAL BOUNOARIES AIRPORTSz B O T S W A N A 2 \MIE  TOBACCO BOARD OF ZAMBIA SCHEMES KEY TO MAP IBRD 3935, December 1981 1 SERENJE F.F. SCHEME 46 KATETE F.F. SCHEME 2 MKUSHI WEST F.F. SCHEME 47 MULUNGUSHI F.F. SCHEME 3 MKUSHI T.F. SCHEME 48 CHIKANTA F.F. SCHEME (GAMELA) 5 MPIMA A.T. SCHEME 49 SIPATUNYANA F.F. SCHEME (GAMELA) 6 MWINUNA A.T. SCHEME 50 CHANGANDA FARM DEVELOPMENT CO. 7 MWOMBOSHI T.F. SCHEME 51 CHIBWE F.F. SCHEME 8 CHIBWE A.T. SCHEME 52 POPOTA F.F. SCHEME (Green Leaf Prod.) 9 MUKONCHI A.T. SCHEME 53 CHOMA F.F. SCHEME 10 MIKATA (MPONGWE) F.F. SCHEME 54 MUKONCHI S.T. SCHEME 11 SINGANI (GAMELA) F.F. SCHEME 55 MUKONCHI T.F. SCHEME 12 SIACHITEMA (GAMELA) F.F. SCHEME 13 CHOMA T.F. SCHEME 14 KALOMO T.F. SCHEME 15 MUKWELA A.T. SCHEME 16 TARA A.T. SCHEME 17 POPOTA S.T. SCHEME 18 KACOMBA F.F. SCHEME 19 MAPANZA F,F. SCHEME 20 LUMEZI F.F. SCHEME 21 LUWERZI F.F. SCHEME 22 SINDE-MISALE F..F. SCHEME 23 VUMBWE F.F. SCHEME 24 ZEMBA F.F. SCHEME 25 SINDA F.F. SCHEME 26 PETAUKE F.F. SCHEME 27 KAPARA F.F. SCHEME 28 CHIPANGALI F.F. SCHEME 29 KAOMA F.F. SCHEME 30 KA0MA A.T. SCHEME 31 KABOMPO F.F. SCHEME 32 MUNKONGE F.F. SCHEME 33 NYAWA F.F. SCHEME 34 MANSA F.F. SCHEME 35 KAINDU F.F. SCHEME 36 CHALIMBANA F.F. SCHEME 37 CHINSALI F.F. SCHEME 38 KAPAMBA F.F. SCHEME 39 KASHIMA F.F. SCHEME 40 MAHEBA F.F. SCHEME 41 CHIVUMA F.F. SCHEME 42 NTEME F.F. SCHEME 43 CHAMA F.F. SCHEME 44 CHIKOMENI F.F. SCHEME 45 CHIPATA F.F. SCHEME