Zambia - Livestock Development Project

CIRCULATING COPY TO BE RETURNED TO REPORTS PEK Document of The World Bank FOR OFFICIAL USE ONLY Report No. 1362 Project Performance Audit Report ZAMBIA LIVESTOCK DEVELOPMENT PROJECT (Loan 627-ZA) November 24, 1976 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. List of Abbreviations Goverment Agencies CSB - Cold Storage Board INDECO - Industrial Development Corporation MRD - Ministry of Rural Development (formerly Ministry of Agriculture) RDC - Rural Development Corporation (formerly ADC, Agriculture Develop- ment Corporation) ZCDL - Zambia Cattle Development Ltd. External Agencies ADS - IBRD Agricultural Development Service CP - FAO/IBRD Cooperative Program IBRD - International Bank for Reconstruction and Development LD - IBRD Livestock Division PMEA - IBRD Permanent Mission in Eastern Africa (whose name was later changed to Regional Mission in Eastern Africa) Exchange Rates 1968-1972: 1 Kwacha = US$1.400 1973-1976: 1 Kwacha = US$1.554 FOR OFFICIAL USE ONLY TABLE OF CONTENTS Page No. Preface Basic Data Sheet Highlights Summary and Conclusions j I. Project History A. Project Background 1 B. Project Preparation 2 C. The Project, as Agreed Between IBRD and Government 4 D. Project Implementation 6 E. The Final Stages of the Project 11 F. ZCDL, the Ranches and the Dairies After Loan Cancellation 14 II. Causes of the Failure to Complete Project Implementation A. Decisions Regarding the Project Objectives 19 B. Decisions Regarding the ZCDL Administration 22 C. Decisions Regarding the Project's Economic and Financial Viability 23 D. Decision to Reject the Second Livestock Project 23 E. Lack of Consistency Between the Project and the Government's Agricultural Production and Consumer Price Policies 25 III. Borrower Performance 26 IV. IBRD Performance 27 V. Concluding Remarks 31 Annexes 1. Schedule of Disbursements 2. Breakdown of IBRD Funds Disbursed, by Category 3. ZCDL: Breakdown of Ranch and Dairy Managers Between Zambians and Expatriates 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.  Project Performance Audit Report ZAMBIA LIVESTOCK DEVELOPMENT PROJECT (Loan 627-ZA) PREFACE Loan 627-ZA was approved in June 1969 to support a livestock develop- ment project on State-owned ranches and dairies. After two years of imple- mentation, the project came to a standstill. In 1972, both the Government and the Bank reappraised the project; the loan was eventually cancelled at the Government's request. This audit report reviews the design, implementation and achievements of the project. Since project implementation had to be suspended and the loan cancelled, the audit includes an unusually long historical description to pro- vide a framework within which both decisions may be analyzed. The audit was based, first, on information contained in the loan ap- plication, the preparation, appraisal and supervision reports, other material from Bank files, and several reports submitted by the Government, and secondly, on discussions with Bank staff in Washington, Nairobi, and Lusaka, staff of Govertnent ministries and agencies, and representatives of Barclays Bank (co- financer) in London and Lusaka, and Huntings Technical Services (consultant) in London. No completion report was ever prepared. The valuable assistance provided by all of these agencies in the prep- aration of this report is gratefully acknowledged.  BASIC DATA SHEET Zambia: Livestock Development Project (Loan 627-ZA) A. Amounts (in US$ mln) Original Disbursed Cancelled Repaid Loan 627-ZA 2. 0 0.83 1.67 0.83 B. Project Data Original Plan Actual First Mention in Bank Files 21 NOV 66 Government Application 25 MAR 68 Board Approval JUN 69 17 JUN 69 Loan Agreement JUN 69 30 JUN 69 Loan Effectiveness 30 SEP 69 10 OCT 69 Physical Completion DEC 74 Not Completed Percentage of Original Pro- 30-4o% /a ject actually completed Last Disbursement APR 72 Loan Closing NOV 7h Cancelled NOV 74 Total Costs US$5.8 US$3.26 /, Economic Rate of Return 16% C. Mission Data Sent Month, No. of No. of Date of by Year Persons Weeks Manweeks Report Identification PMEA/ADS/CP MAY 67 3 1 L 7 JUN 67 Preparation Qst visit) ADS Nov 67 2 1 2 16 JAN 68 Preparation (nd visit) ADS FEB 68 2 1 1 / 12 FEB 68 Follow-up Preparation ADS/CP/IBRD MAY 68 5 2 10 31 JUL 66 Appraisal (1st visit) IBRD OCT 68 3 3.5 l4 27 MAY 69 Appraisal (2nd visit) IBRD DEC 68 2 1 2 Subtotal 30 Supervision I IBRD SEP 69 1 0.8 0.8 8 OCT 69 Supervision II IBRD FEB 70 2 0.6 1.2 15 APR 70 Supervision IIIL IBRD SEP 70 4 0.5L 2 13 NOV 70 Supervision IV IBRD MAY 71 2 1 2 17 JUN 71 Supervision V PMEA NOV 71 2 1.5 3 20 JAN 72 Reappraisal IBRD APR 72 3 5 15 20 JUL 72 Supervision VI PMEA JUL 73 1 0.8 0.8 9 AUG 73 Subtotal 2 . D. Follow-on Project: Two would-be follow-on projects were prepared. The first (Livestock II) was rejected by the Bank in September 1970. The second (Mixed Farming) was withdrawn by the Government in April 1975. a 30% of disbursements out of the IBRD loan; 40% of the total project cost forecast at reappraisal. /b Due to inflation, at reappraisal the total project cost was forecast at about US$8 million. c These missions also discussed other projects. The time shown devoted to this project was estimated by OED. d This mission also appraised the Second livestock Project. OED assumed that it devoted half a week to supervising the first project and four weeks appraising the second project.  Project Performance Audit Report ZAMBIA LIVESTOCK DEVELOPMENT PROJECT (Loan 627-ZA) Highlights The audit reviews the progress under the livestock development proj- ect, partially financed by IBRD, which provided for the further development of a set of State-owned ranches and dairy farms. The project preparation docu- ment, by the Bank's Agricultural Development Service in Nairobi, and the Gov- erment's loan application, were originally criticized by IBRD, but a loan was eventually approved. Project implementation, which started rapidly, was ham- pered later by several problems (managerial, economic, technical) and eventu- ally suspended; the loan was then cancelled. Since cancellation, the Govern- ment has continued ranch/dairy development, though at a very slow pace. Most of the ranches are now making profits, reflecting increases in controlled prices. Both IBRD and the GoverMnent share the responsibility for the eventual failure in carrying out the project, though it must be mentioned that IBRD adopted the right position in each of three critical decisions. These related to approval of the first project, amendments at reappraisal, and rejection of the second project proposal. The following additional points may be of special interest: - Simple conditions to negotiate, sign and declare effective the loan agreement; staggered conditions to allow disbursements on each ranch or dairy farm (paras. 1.12 and 4.03) - Conflict between the project as a commercial endeavor and the Govern- ment's agricultural development policy (paras. 2.22 and 2.23) - IBRD's flexibility in waiving certain rules for procurement when the local and international conditions so required (para. 4.04) - IBRD's internal disagreements and their impact on the project (paras. 4.09 and 5.04) - Drawing the line between "overly optimistic" and'pessimistic" assump- tions in project preparation (paras. 4.10 and 5.03) - IBRD and the GoverMnent's differing views on what the "project" was, and the need to reach clear agreements (para. 5.02) - Risks of co-financing such a project with a commercial bank (para. 5.05) - The role of expatriates in project management (paras. 5.06 and 5.07)  Project Performance Audit Report ZAMBIA LIVESTOCK DEVELOPMENT PROJECT (Loan 627-ZA) SUMMARY AND CONCLUSIONS Project History i) Up to Loan Agreement 1. Between 1965 and 1967, the Ministry of Agriculture established several scattered ranches and dairies to preserve the national herd and to foster regional development. The Cold Storage Board (CSB) established other ranches to be used as holding grounds. In early 1967, the Government decided to run all of those State ranches and dairies on commercial terms under a para-statal organization (eventually, Zambia Cattle Development Ltd., ZCDL), and asked IBRD's Agricultural Development Service (ADS) to assess the feasi- bility of this operation and to program its implementation. 2. In September 1967, the Government decided not to proceed with the preparation of an entirely separate package program it had agreed on with IBRD because the program conflicted with its policy of giving priority to the less developed regions. Alternatives were sought, and four other proj- ects were identified as suitable for IBRD financing, including the State livestock plan. Since ADS was already preparing a report on the commercial development of the State ranches and dairies, and no work had been done on the other three possibilities, it was concluded that this livestock project was the only one that could be made ready in the near future. 3. The Government submitted a loan application based on the ADS report (March 1968). IBRD headquarters criticized both documents on technical and financial grounds, but decided, since the underlying developmental concept was attractive, to go ahead with the project, partly to support its rela- tions with the Government and partly to back up ADS. 4. The appraisal mission accepted all 10 ranches and 1 of the 2 dairy farms presented by the Government; in addition, it suggested that 2 other ranches and 4 other dairies be added later. Thus,the project ultimately called for the development of 12 ranches and 5 dairy farms. They would produce weaner steers for "emerging Zambian farmers" and would increase the production of beef and milk. A second, broader livestock project was to be prepared by the Govern- ment (plus a consulting engineer financed under the loan) shortly thereafter. The US$5.8 million first project was to be jointly financed by the Government (36%),Barclays Bank (21%),and IBRD (43%; Loan 627-ZA for US$2.5 million). IBRD and Barclays Bank disbursements on ranch and farm develop- ment expenses (Category IV) were contingent upon ZCDL's producing titles for each ranch or dairy farm. The Government agreed to consult with IBRD from time to time concerning the producer prices for beef and dairy products, and to refinance on a longer term CSB's Grazier Scheme for fattening cattle. - ii - ii) From Loan Agreement to Loan Cancellation 6. The Loan Agreement became effective on schedule (October 1969). ZCDL activities were already underway, and project implementation was able to begin immediately after effectiveness, a feature seldom achieved in other livestock projects supported in Africa or elsewhere by IBRD or IDA. 7. One year later, the Government decided to terminate the services of ZCDL's first General Manager, who had been seconded from ADS. Although satisfied with his performance, the Government felt that the ADS fee was too high,and complained that the General Manager could be identified with the pre-Independence school of thought, tried to implement the project too much as described in the appraisal report, and had a conflict of allegiance. 8. Another expatriate was appointed to replace him. During his as- signment, project implementation lagged behind schedule and ZCDL fell into a very serious financial position. The main factors leading to this situa- tion were managerial problems (inadequate performance of the new General Manager; lack of independence of ZCDL from Government; some ranch managers began to quit), economic and financial problems (increasing costs; price freezes; demand for weaners not increasing; prices and terms at which wean- ers and culls should be sold, and cattle from departing European farmers bought), and technical problems (ranch location; harassment by wild animals; grass and bush fires; animal diseases, etc.). 9. Eventually, project implementation had to be suspended; ZCDL be- came almost bankrupt. IBRD, as well as the Government, reviewed the project. rhe IBRD reappraisal mission found that ZCDL was indeed in a serious finan- cial position and faced critical management problems, but that most of the ranches and dairies had been substantially developed and that the herds were improving in quality. The mission recommended a more modest program. The Government Committee proposed a more drastic cutback and recommended that future investment be financed by the Government and, thus, that no further funds be withdrawn from IBRD, the remainder of the loan to be cancelled. 10. In April 1972, shortly after the reappraisal mission visited Zambia, Barclays Bank withheld disbursements to ZCDL, taking into account its lack of creditworthiness. Later on Barclays paid the outstanding claims, but the suspension had already dealt the "coup de grace" to the project. 11. The Government requested that IBRD cancel the outstanding balance of the loan, of which only US$0.83 million had been disbursed (November 1973). IBRD agreed, but requested that the Government pre-pay the funds already withdrawn. The loan was cancelled in November 1974. iii) After Loan Cancellation 12. ZCDL still exists; the ranches and dairies are in operation and continue to produce meat and milk; since beef and milk prices have in- creased, many of the farms are now making a profit. Ranch and dairy develop- ment has been almost paralyzed, however. Technical and managerial problems still affect ZCDL, but some improvements have been achieved. ZCDL, with - iii - Government aid, is meeting its current operating deficits and gradually reducing its accumulated losses. Stock numbers on ranches and dairy farms are over pre-appraisal levels though under those forecast; the present technical coefficients are far from satisfactory. Causes of the Failure to Complete Project Implementation 13. The factors pointed out in paragraph 8 eventually led ZCDL to suspend project implementation. Those factors stemmed mainly from certain decisions adopted by IBRD, the Government, or both together. 14. First, four partially different sets of objectives were succes- sively associated with the project: (i) The ranches and dairies were originally established to preserve the national herd and to foster re- gional development. (ii) Then, the Government decided to devote them to producing beef and milk on a commercial basis. (iii) The appraisal mis- sion placed the main emphasis on producing weaners. (iv) Later on, the Government imposed socially oriented responsibilities on ZCDL. 15. These changes raised financial problems and conflicted with ranch location: (i) ZCDL had to find unbudgeted funds to buy more cattle from departing European farmers than were currently needed, but was not allowed to make extra cash selling the additional culls. Furthermore, the Government asked ZCDL to subsidize Zambian producers by reducing its sale price for wean- ers and for the culls it was allowed to sell. The cash flow situation of ZCDL worsened when the demand for weaners failed to increase and it had to fatten its own weaners. (ii) Some of the ranches chosen to meet the first mentioned objectives were not adequate for producing beef or weaners. When the new pro- duction objectives were established, these ranches should have been rejected by both IBRD and the Government. A rather smaller but more profitable and manageable projvet would have resulted. 16. In addition, the project's commercial objectives were inconsistent with the Government policies of subsidizing agricultural producers and the prices of consumer goods. 17. Secondly, the Government tried to carry out the project, involving US$6 million in investments, with an underpaid management team. IBRD did not object to this policy, or to the Government argument regarding the "excessive cost" of the first ZCDL General Manager. In addition, the Government did not give ZCDL the independence and authority it needed to formulate and im- plement its commercial policies. 18. Thirdly, the Government decisions to restrict imports from South Africa, Angola and Mozambique, increasing input prices, and to keep consumer prices low for beef and milk, jeopardized the economic and finan- !iai viability of ZCDL. - iv - 19. The second livestock project was expect-d to foster t'he demand for weaners produced under the first project. However, the loan application sub- mitted by the Government did not meet minimal TBRD standards and was rejected at appraisal. dealing a severe blow to the implementation of the first project. The main reasons behind IBRD's decision were: (i) Government policies con- flicted with IBRD's lending criteria; (ii) the Government representatives and the consultants understood that IBRD had approved the project preparation in- terim report and the strategy for livestock development they were working on; (iii) thus, the consultants did not warn the Government strongly enough about the IBRD standards; (iv) IBRD changed the staff in charge of the project be- tween the review of ti-e interim report and appraisal; and (1) IBRD requested "at the project be prepared in a very limited period of time. Conclusicrs 20. The project was partially successful in physical terms (ranches and dairies were developed. herds were built up. production was increased), but was disappointing in economic and financial terms (very low or even negative financial rates of return were computed for all but one of the ranches and dairies at the end of the project implementation period). The loan was can- celled. ZCDL is still in the red; substantial State subsidies have been transferred to it. 21. IBRD and the Government share the responsibility for the eventual failure in carrying out the project as they had agreed. The project design was too optimistic. The project objectives and the set of ranches accepted for its implementation were not fully compatible with each other. or with some other Government policies. IBRD went along with the proposal more on "area" or "country" grounds than on "project" grounds. and was not critical enough of the proposed ranch locations. The Government. the consultant, and IBRD share the responsibility for the events that led to the rejection of the second livestock project. The Government never consulted with IBRD concerning the oroducer Prices of beef and milk; IBRD did not reqjest that the Government do so. IBRD f.rgot to follow-up the refinancing of the Grazier Scheme. 22. On the positive side, the experience showed that IBRD can respond promptly to the request of a country, financing a simple project while a more complex one is prepared; that conditions to negotiate, sign and declare effective a loan agreement can be made simple; that a livestock project may have a rapid start-up; and that IBRD can be flexible in waiving certain rules for procurement when the local and international conditions so require. Moreover, some Govern- ment officials pointed out that IBRD had been very helpful to the country in setting uo the oro4ect. 23. In additior- this project sheds light on several issues. First, IBRD and governments at times have different views of what any particular project comprises. In this case. their different points of view led to some misunderstandings between the partners. Secondly, it is very diffi- cult to draw fairly the line between what can be questioned as "overly op- timistic" assumptions in project preparation. and what would be "pessimistic" ones. Thirdly. joint financing projects with a commercial bank involve some risks that the Zambia experience points to quite clearly. And finally, it is remarkable that the use of expatriates in the management of the company and the ranches and dairy farms was never an issue within the Government. -r between IBRD and the Government. I. PROJECT HISTORY A. Project Background 1.01 At the Government's request, an FAO/IBRD Cooperative Program (CP) mission visited Zambia in 1965 to analyze agricultural development priorities and identify investment possibilities in that sector. Thirteen specific proj- ects were listed. After discussions held in late 1966, IBRD proposed that the Government prepare a "package program of agricultural improvements, intensely applied in a specific area," as the most productive immediate effort toward agricultural development in the country. However, IBRD's proposal that the project be located in the most suitable areas was at variance with the Govern- ment policy to give primary attention to the outlying, more backward regions, away from the railroad line. After discussions with the Permanent Mission in Eastern Africa (PMEA), Agricultural Development Service (ADS),and CP, the Gov- ernment decided in September 1967 not to proceed with the package program. 1.02 During the same period (1965 to 1967), the Ministry of Agriculture had established and/or invested in several scattered ranches,1/ in order to pre- serve the national herd, threatened with decimation after independence, and to foster the development of several remote regions. The Cold Storage Board (CSB, a governmental meat marketing corporation), on its part, had established some holding grounds which in practice had evolved into fattening and breeding ranches. Since neither agency was operating its properties on an economic basis, the Government decided in 1967 that the Industrial Development Corpora- tion (INDECO), which had operational autonomy and a reputation for a relatively hard-headed commercial attitude, take over these State ranches under a new subsidiary company. INDECO had no staff competent in this field and was re- luctant to assume responsibility without adequate review. Thus,its General Manager asked ADS to: (i) assess the feasibility of INDECO taking over the State beef ranches and dairies, (ii) prepare a program for their operation, and (iii) help find an executive to head the proposed company (July 1967). No question of IBRD financing is discussed in the files for that period. 1.03 The Government policy to invest in the more undeveloped areas nar- rowed the field for IBRD financial support in agriculture to four possible projects'.2/ Of these, the project that could be prepared in the shortest time was the development of the State ranches and dairies: a report was 1/ Some of them on vacant State land; others totally or partially purchased from expatriate owners. 2/ None of these four projects was on the original list of thirteen. There was a relation in some cases, but the development of the State ranches was not among these. - 2 - already being prepared by ADS. Although PMEA and ADS felt that "the general atmosphere in the Government with respect to realistic agricultural develop- ment programs remained rather discouraging," they concluded that livestock development offered the most promise for finding a satisfactory project in the near future and that this project "looked like the only agricultural as- sistance that we could usefully and prudently render to the Zambia Govern- ment." Therefore, in November 1967, PMEA suggested to the Government, and the Government agreed, that IBRD could finance a broadened livestock develop- ment project based on the State ranches. B. Project Preparation 1.04 In January 1968, ADS presented its report on the development of the State ranches to IBRD and to the Government. At the time, beef prices in Zambia were very low, and so the rates of return for the proposed invest- ments turned out to be below the current rates of interest. In the report's covering letter to INDECO, ADS stressed the need to increase beef prices if the project were to be rendered profitable. In mid-February, beef prices were increased by 38%. In March, the Government submitted to IBRD a loan ap- plication based on the amended ADS report (see diagram 1). The project ex- ecuting agency would be a new State-owned livestock company (eventually, Zambia Cattle Development Ltd., ZCDL) instead of INDECO. 1.05 Both the ADS report and the Government loan application were criti- cized by IBRD on the grounds that: (i) no analysis had been made to determine whether this one was the best kind of project for the development of Zambia's livestock industry; (ii) the ranches were widely scattered throughout almost the entire length and width of Zambia, instead of taking advantage of features such as climate, soil type, contour, access to markets, etc., features which indicated a more concentrated pattern and would favor profitable beef produc- tion; (iii) the end borrower (formerly INDECO, now ZCDL) would be unable to serve the debt it would incur under the project, even when allowance was made for a disbursement period of 11 years, and a repayment period of 20 years at 8% with a 12-year grace period; (iv) the project would cover only a small part of the national deficit of beef and milk foreseen for 1985 and was likely to have very limited impact on either the development of the live- stock industry or the economy of the country as a whole; and (v) neither the commercial nor the traditional sectors were included. 1.06 In spite of the criticisms, IBRD decided to go ahead with the proj- ect on the grounds that: (i) "the Africa Department could not risk jeopardiz- ing the good relations which had recently been so carefully established with Zambia by a response to such a request which the Zambians would consider un- satisfactory"; and (ii) "it was important that because the ADS, a part of the Bank, had prepared the report on the State ranches that these ranches should be incorporated in the first livestock project." Moreover, IBRD and the Gov- ernment agreed that the State ranches project would be the first phase of a broader livestock development project to be prepared shortly thereafter by the Government with the help of ADS. They also agreed that the first phase should not be deferred until this second phase was ready. - 3 - Diagram 1 CHRONOGRAM OF THE PROJECT'S EARLY STAGES 9 4 7 1 9 6 8 1 9 9I II I II II I I I First Mention in IBRD files Project identified by PMEA/ADS/CP Mission TOR for Project Preparation by ADS Project Prepared (ADS Report) Government rejected the "package program" Loan Application PMEA suggested to Government that IBRD could LFollow-up Preparation Mission finance this project Appraisal Mission Negotiations Loan Agreement signed Loan Agreement effective 1.07 IBRD felt, however, that the project could not be appraised immedi- ately. A CP "follow-up preparation mission." later expanded to a PMEA-ADS/ CP/IBRD mission at PMEA-ADS insistence, was sent to Zambia in May 1968 to resolve the above-mentioned issues. 1.08 The mission found the proposed project to be sound technically, financially and economically. It concluded that: (i) Commercial and co- operative ranching by African cattle owners was in an embryonic stage, and commercial European ranchers expressed little desire to expand beef produc- tion; thus, the State ranches project provided the only immediate avenue for catalyzing the development of the Zambia livestock industry. (ii) During the initial development period, the proposed livestock company would be unable to cover all of its financial obligations from generated income; the mission proposed that Government finance this deficit through either an additional equity contribution from the parent corporation or an interest-free long- term advance from the same source. (iii) The corporate arrangements for run- ning the ranches had been satisfactorily made by the Government (see next paragraph). The mission made no comment on either the location of the ranches or the small contribution to the national demand. 1.09 At the same time, the Government established the proposed livestock company: Zambia Cattle Development Ltd. (ZCDL), as a subsidiary company of the newly created Agricultural Development Corporation of Zambia Ltd. (ADC, later transformed into the Rural Development Corporation, RDC) to take over the ranches and dairies and to run the project. Its main objective was to produce beef and milk for the local market. Other objectives mentioned were: to furnish, in due time, good young female breeding stock and good quality bulls to farmers; to meet the demand for trek oxen in provinces where these were in short supply; to act as a demonstrator of good ranch and veldt man- agement; and to train ranch managers. The Government asked PMEA to second an ADS staff member to manage ZCDL. C. The Project, as Agreed Between IBRD and Government 1.10 The project was appraised in October-November 1968. and an IBRD loan for US$2.5 million (Loan 627-ZA) was approved and signed in June 1969. The project's main aim was to complete the development of 12 beef ranches and 5 dairy farms 1/ and to organize them into an integrated, commercially viable operation under the management of ZCDL. In addition, the project pro- vided for the establishment of pilot pasture improvement trials and a legume seed multiplication plot, and the financing of technical services to assist the Government both in planning national livestock development and in pre- paring the second phase livestock project. 1/ Most of them were on-going ranches and dairies, with some level of development that had been financed by either the Government or their former owners. - 5 - 1.11 Until appraisal the main objective of the project had been the pro- duction of beef and milk; weaner steers and heifers would be produced too. The approved project still called for the production of both beef and milk, but the main emphasis was shifted to producing weaner steers for sale to emerging Zambian producers in both the commercial and traditional sectors.l/ The second phase project, to be prepared, would provide for the development of these emerging producers and, therefore, would foster the current demand for weaners. 1.12 The US$5.8 million project was to be jointly financed by Government (36%), IBRD (43%) and Barclays Bank (21%). The loan comprised five main cate- gories: (i) equipment - US$500,000; (ii) technical services - US$200,000; (iii) artificial insemination - US$100,000; (iv) ranch and farm development expenses - US$1,400,000; and (v) unallocated - US$300,000. It would finance 100% of the direct foreign exchange costs of the project (estimated at US$0.8 million) and 32% of the remaining ranch and dairy development expenditures (representing the estimated foreign exchange component of these expenditures). Barclays Bank's contribution would amount to 55% of whatever IBRD disbursed for ranch and farm development (category iv). In addition, Barclays Bank was to grant ZCDL overdraft facilities up to a maximum of US$0.4 million under a guarantee given by the Government to finance all the working capital needed by ZCDL (see table 1). IBRD disbursements from category iv on account of ex- penditures on any ranch or dairy farm were contingent upon the acquisition of "user's rights" by ZCDL to the whole parcel of land on which such ranch or farm was situated, and upon certification that ZCDL had made corresponding ap- plication to Barclays Bank. 1.13 Elements fundamental to the success of the project would be: (i) a high standard of management, mainly at the top level (general manager, senior beef ranch manager. and senior dairy farm manager), but also at tie ranch/farm level; (ii) good quality breeding stock, to obtain superior results and avoid high mortality and low productivity; (iii) fast and complete stocking programs, to maximize economy of scale effects; (iv) easy and fluent marketing of off-take assumed to be undertaken mainly by CSB and private butchers (beef cattle), CBS's Grazier Scheme (most of the weaners)2/, and the Dairy Produce Board (milk); (v) an adequate level of prices for beef and dairy products, to secure the profit- ability of the ranches, the dairy farms, and the project investment program as a whole (see paragraph 1.15); and (vi) the preparation, approval, and eventual implementation of the second stage livestock development project. 1/ According to the PMEA-ADS/CP/IBRD mission report (July 1968), only three out of the nine ranches then analyzed would specialize in producing weaners; some 5,000 weaners would be produced each year. The appraisal mission (October 1968) assigned nine out of the twelve ranches to producing approxi- mately 13,000 weaners a year. Consequently, only some 5,000 slaughter stock heads were going to be produced annually, instead of the 7,100 forecast in the previous documents. 2/ The Grazier Scheme purchased surplus cattle from commercial ranchers and the traditional sector for lending at commercial interest rates to qualified in- dividual ranchers and grazing associations. The CSB interest, fees and prin- cipal were recovered at slaughter with the residual profit (or loss) accru- ing to the borrower. - 6 - Table 1 FINANCING OF PROJECT COSTS (US$ Million Equivalent) Government ZCDL Barclays Bank IBED Total Amount Am = Amount - Amount k Amount a) Dev.elonTnt Exceno.iturc )i Equipment in- cluding vehicles, water pipes, fencing wire 1/ - - - - - - 0.5 100 0.5 100 iij Technic&l Services - - - - - 0.2 100 0.2 100 ii)Artifi-cial Insemiration - - - - - - 0.1 100 0.1 100 iv) Other Devel- opment Expen- ditures 1.5 35 0.6 l 0.8 19 1.h 32 4.3 100 Unallocated - - - - - - 0.3 IO 0.3 100 b) Working Capital 0.4 100 - - 0-.4 100 c) Total Project Costs 1.5 26 0.6 10 1.2 21 2.5 43 5.8 100 Source: Appraisal Report; Loan Agreement. 1/ This category conprised all goods to be purchased for the Project on which expenditure was 100% foreign exchange. - 7 - 1.14 The Government presented ten existing State-owned beef ranches, covering 602,000 acres, and two State-owned dairy farms, covering 10,400 acres, to be developed under the project. The appraisal mission accepted all ten ranches, but only one of the two dairy farms. One of the ranches was to be devoted to bull breeding (Monze), two others to breeding and fat- tening operations (Mbala and Chishinga), and the other seven to breeding and weaning operations (see table 2 and the map). The appraisal mission pro- posed, and the Government agreed, that the project be enlarged, including two additional breeding/weaning ranches and four additional dairies;l/ these would be identified and purchased later by ZCDL. 1.15 IBRD asked the Government to maintain producer prices for meat and dairy products at such a level as to provide adequate incentives to ZCDL to expand production during the entire life of the project. At negotiations, the Zambian delegation objected strongly to the wording of that proposal. It was agreed, as a compromise, that the Government "should consult with the Bank from time to time concerning the producer's prices for beef cattle and dairy products." Neither was the term "from time to time" precisely de- fined or otherwise specified, nor was it stipulated at whose initiative those consultations had to be made. 1.16 In addition, the Government agreed on re-financing the CBS's Grazier Scheme (later transformed into the Cattle Finance Company of Zambia, partly at the appraisal mission's instigation) with a loan whose term would appropri- ately match the length of the loans made to the participating ranchers, re- placing the current financing by revolving 90-day Treasury bills. D. Project Implementation 1.17 The Loan Agreement became effective in early October 1969, on schedule, after the necessary legal steps had been taken and the required evidence furnished to IBRD. 1.18 ZCDL, already established and with a General Manager seconded from ADS, was able to achieve a rapid start-up for the project after effectiveness. During the first year following effectiveness, ZCDL obtained land titles for seven ranches and the dairy farm, and tried to acquire the other two ranches and four dairy farms envisaged (see table 2). Ranch and dairy development, the implementation of the pilot pasture improvement program, and the legume seed multiplication plot, were started.2/ In the view of the first three 1/ Bringing the total number of ranches to twelve and the total number of dairy farms to five. 2/ In fact, project implementation had started before the date of effective- ness. Work on beef ranches and dairy farm development and cattle purchase costing in all approximately $300,000 had taken place since January 1, 1969 (when ZCDL took them over from the Government) and had been paid from Gov- erment sources. Of the total, 37% (approximately $110,000) were declared eligible for reimbursement under the loan. Table 2 ZCDL: TITLE DEEDS POSITION OF RANCHES AND DAIRY FARMS BY THE END OF THE FIRST TWO YEARS Existing Title Position Area Acquired Title Position Kind of Projected at as of as of as of Ranch Area Appraisal October 31, 1970 October 31, 1971 October 31, 1971 1. Ranches (Acres) (Acres) Chisamba Breeding/ 37,510 Yes Delayed due to 38,162 Yes Weaning Mortgage Monze Bull Breed- 20,000 Yes Yes 18,353 Yes ing Mugoto/Wolverton Breeding/ 20,000 Yes Being Purchased 20,219 Yes Weaning North Kalomo Breeding/ 80,000 Yes Being Purchased 73,184 No Weaning South Kalomo Breeding/ 55,000 No Being Purchased 30,467 No Weaning Chishinga Breeding/ 104,770 Yes Yes 109,000 Yes Fattening Mbala Breeding/ 109,000 Yes Yes 99,576 Yes Fattening Mkushi Breeding/ 46,250 Yes Yes 45,460 Yes oo Weaning Kitwe Breeding/ 70,000 Yes Yes 70,000 Yes Weaning Solwezi Breeding/ 106,770 Yes Yes 107,770 Yes Weaning Katete/Chinjara Breeding/ 48,260 Yes Katete: Yes 48,525 Yes Weaning Chinjara: No (Twelfth ranch) Breeding/ 55,000 No No Nil Never Bought Weaning 2. Dairy Farms Palm Grove 3,500 No Being Purchased Nil Expected Soon Davels 3,500 No No Nil No Gravetts 1,500 Yes Yes 1,593 Yes Bekkers 3,500 No Being Purchased 2,574 Yes Eldorado 3,500 No Being Purchased 1,583 Yes Source: Appraisal Report; Supervision Mission Reports. - 9 - supervision missions, the performance of ZCDL was very satisfactory. Since most of the livestock projects supported in Africa or elsewhere by IBRD or IDA have had a slow start-up, the quick implementation of this project, facilitated by the lack of cumbersome conditions of effectiveness and boosted by a good project manager, is pleasantly surprising and worth mentioning. 1.19 The satisfactory start-up was achieved in spite of some problems: (i) ZCDL recruited several competent ranch managers (some Zambians, some ex- patriates), but at salaries lower than usual in the civil service and in the private sector; it was expected that retaining them would be difficult; (ii) it was hard to find sufficiently large blocks of land in the central and south- ern provinces which could meet ZCDL requirements and which did not include land already in productive use; (iii) it proved an extremely slow process to obtain from the Government the Rights of Occupancy on Trust Lands and the leases on State lands: acquisition of the land that would make up most ranches was con- sequently delayed; and (iv) the Grazier Scheme did not develop its activities to the extent expected by the appraisal mission. 1.20 In late 1970, the Government decided to terminate the services of the ADS staff member seconded as General Manager of ZCDL. The Government said that it was satisfied with his performance but that: (i) ADS's fee (US$24,000 per annum) was too high for ZCDL; (ii) a General Manager was needed who, having the necessary academic background, would be able to give the Gov- ernment advice on broader issues regarding livestock planning and policy; and (iii) Zambian staff should be trained to take over from expatriates, by ex- patriates who were not identified with the old school of thought of pre-inde- pendence Africa. In addition, it was the Government's point of view that the General Manager (iv) was trying to implement the project too much as described in the appraisal report and the loan agreement, despite the Government view that ZCDL should progressively satisfy other social objectives (see paragraph 2.04), and (v) had a conflict of allegiance because he was an ADS staff member but had to answer t-- the Government. 1.21 A new expatriate, coming from outside of Africa, was selected by the Government and, after Bank approval, appointed as the second General Manager of ZCDL. During his assignment, project implementation began to move more slowly and encountered more serious problems: some new, others stemming from those mentioned above. Ranch and dairy development lagged behind sched- ule; IBRD disbursements were made at a rate about half the revised estimates (see annex 1); ZCDL ran into a stringent financial position, with a huge cumu- lative cash deficit (US$1.6 million equivalent by the end of 1971) and, in the early part of 1972, it was unable to meet its current liabilities. 1.22 Several causes of different nature worked together to bring about this abrupt change in the pace of project implementation: (i) Managerial Problems (a) The new General Manager did not perform well; when the Government dis- missed him in mid-1972, the General Manager of RDC (ZCDL's parent corporation) told IBRD that ZCDL was in a "mess" as a "result of 18 months of neglect." - 10 - (b) ZCDL management did not enjoy good relations with RDC; the ZCDL Board of Directors was not given the independence and authority necessary to formu- late and implement the company's commercial policies. (c) Some of the ranch managers began to quit because of their low salaries; they had to be replaced by less qualified people. (d) Some of the ranches were too isolated; communication with them was dif- ficult. (ii) Economic and Financial Problems (a) The Government policy to restrict imports from South Africa and Angola, and through Mozambique, led to marked increases in import prices and resulted in acute shortages in some equipment and supplies. Labor costs increased as well. At the same time, the Government froze meat and milk prices at the levels established in 1968, thereby shrinking the profit margins of ranches and dairies and threatening the financial viability of ZCDL. The pricing policy applied to everyone, not just ZCDL. It did, however, affect ZCDL more than other producers because ZCDL was going through a stage of heavy develop- ment expenditure (which few other producers were doing). (b) Government regulations forced ZCDL to sell beef cattle only to CSB, and only at the official prices.l/ (c) The expected demand for weaners did not materialize: actual demand was negligible and prices unattractive; the Grazier Scheme did not provide the expected outlet for them; ZCDL had to retain the weaners for fattening and cash revenues were delayed. (d) The Government requested that ZCDL sell weaners and culls to emerging Zambian farms at lower prices than projected in the appraisal report, and purchase cattle in bulk from departing European farms at prescribed prices. This restriction prevented ZCDL from buying the best of the Europeans' cat- tle, which was sold in limited groups at higher prices, and forced it to acquire a large number of inferior animals, many of which could not be culled because of stringent Government regulations. 1/ The appraisal mission had assumed implicitly that ZCDL would sell its beef cattle to either CSB or the private butchers. At appraisal, CSB gave a kind of guaranteed minimum price, and the butchers offered as much as 10-20% above it. - 11 - (iii) Technical Problems/ (a) Some of the ranches were located in very difficult terrain. Some were heavily wooded. In one (Solwezi), the area tends to favor bracken in- stead of grass when cleared. (b) Harassment and killings of stock by wild animals, mainly lions, wild dogs,and snakes. (c) Grass and bush fires. (d) Other technical problems, such as infertility, malnutrition and disease. E. The Final Stages of the Project 1.23 As recommended by the fourth supervision mission in May 1971, an econ- omist from the Ministry of Rural Development (the former Ministry of Agricul- ture) was seconded by the Government to ZCDL to review its operations and to determine its immediate and long-term financial position. His report, produced in late 1971, forecast a bleak financial future for ZCDL: an increasing cumu- lative negative cash balance during the next seven years; negative financial rates of return for eight of the ranches and slightly above zero for all but one of the others; a mere 4% financial rate of return for the dairy farms, etc. 1.24 During the first quarter of 1972, the development of the ranches and dairies was almost paralyzed, and ZCDL became almost bankrupt. 1.25 As a result of the MRD economist's report and other obvious signs of trouble, IBRD, as well as the Government, called for a review of the project. IBRD sent a reappraisal mission headed by a financial analyst in March/April 1972. The Government set up a Committee, including representatives of MRD and the Ministry of Finance, to analyze the project and recommend measures to render ZCDL and the project financially viable (see diagram 2). 1.26 The IBRD reappraisal mission reported: "ZCDL is in a serious financial position: by the end of 1971, its accumulated losses were some US$1.6 million equivalent; in the early part of this year (1972), it was unable to meet its current liabilities. It also faces critical management problems. On the other hand, most of the ranches and dairies have been substantially developed and the herds are improving in quality. Government recently approved increases in the minimum producer's price of beef and milk. Given a period of consolidation and reorganization, the company should earn profits and made a substantial con- tribution to Zambian beef and milk production." The mission recommended a more modest project which included the disposal of three ranches (Kitwe, Solwezi and 1/ One should also mention that the technical standards and speed of development assumed in the project design - against which the actual achievements were compared - were too ambitious. - 12 - Diagram 2 CHRONOGRAM OF THE PROJECT'S LAST STAGES 7'I I 97 2. ' 4 7 3 I 9 7 I I I I I I I I I I MRD economist's report Government Committee established Reappraisal Mission in Zambia Barclays Bank rejected two claims Barclays Bank made last payment Reappraisal Mission Report Government Committee Report Government requested cancellation IBRD requested pre-payment Loan Agreement cancelled - 13 - Katete/Chinjara), the discontinuation of maize growing on a fourth (North Kalomo), and a substantial reduction in future ranch development under the project. It also recommended that the ZCDL Board be given greater responsi- bility and authority for formulating and supervising the company's commercial policies, and that the company should be allowed to operate on a fully com- mercial basis.l/ On such a basis, additional investment of US$760,000 equiva- lent would be required, reducing the total project cost from US$5.8 million 2/ to US$4 million, and the amount financed by IBRD from US$2.5 million to US$1.7 million. The reappraisal mission report acknowledged that Government had contributed much more, both absolutely and proportionately, than orig- inally envisaged. 1.27 The Government Committee agreed that a number of ranches with little potential should be disposed of and that further development of other ranches should be reduced. But it proposed a far more drastic cutback in ranch de- velopment than that suggested by the IBRD reappraisal mission and recommended that this limited investment be financed by Government. Thus, no further funds would be withdrawn from IBRD, the remainder of the loan would be can- celled. The main reason to support the latter recommendation was to give Government more freedom to choose the objectives of the project and the ways to carry it out without IBRD interference. 1.28 In April 1972, shortly after the reappraisal mission visit to Zambia, Barclays Bank withheld disbursements to ZCDL.3/ This decision was adopted by Barclays taking into account the "present crisis of the company, the lack of evidence that it will be able to service its debt, and the fact that we are being called upon to disburse on properties that may well be disposed of in a matter of months, probably at a very low realization value." Barclays felt that "the whole basis of the project oad)been undermined."4/ IBRD, based on the preliminary report of the reappraisal mission, let Barclays Bank know that it considered that no grounds existed for holding up disbursements. Barclays, although it did not share IBRD's "optimism for the continuation of the company's business," paid the ZCDL retained claims (this was the last dis- bursement ever made by Barclays, May 1972). However, Barclays withholding had already dealt the "coup de grace" to the project because it left ZCDL without working capital. 1/ For example, that ZCDL be allowed to sell cattle to the buyers offering it the best prices (i.e., private butchers and traders) instead of sell- ing only to CSB; similarly, ZCDL should have the right to dispose of any farm or ranch which proved to be unprofitable. 2/ Because of inflation, the reappraisal mission estimated that the original project cost would then be about US$8 million equivalent (the US$4 million estimate was expressed in 1972 prices). 3/ Barclays Bank had already sent both IBRD and ZCDL four letters strongly criti- cizing the slowed pace of project implementation (February, April, July and December 1971). In the last one, Barclays asked ZCDL "to refrain from making any further claims under (their) loan agreement for the time being pending reappraisal of the project and (their)agreement to that reappraisal" because they did "not wish to make matters worse by withholding disbursenents. 4/ In addition, Barclays Bank's decision might also reflect the extremely tight liquidity position which all the commercial banks in Zambia faced in the early part of 1972. Barclays was affected particularly badly. - 14 - 1.29 The Government of Zambia, after about 15 months of consideration, re- quested in November 1973 that IBRD cancel the outstanding balance of the US$2.5 million loan, of which only US$0.83 million had been disbursed (see annex 2), and asked for a revised amortization schedule. IBRD agreed, but requested that the Government pre-pay the funds already withdrawn,on the grounds that mere cancellation did not relieve the Bank of the responsibility to monitor and report on the progress of the project and, as long as a part of the loan agreement remained in effect, the Bank's agreement would be needed for any changes in the method of carrying out the project, a Bank role incon- sistent with the Government's intentions. It took eight months to negoti- ate the prepayment, and another four months for IBRD to let the Government know the amounts due. The Government of Zambia paid them in full; the loan was cancelled immediately thereafter (November 1974). 1.30 ZCDL had borrowed K. 325,000 (US$455,000) from Barclays Bank. Both agencies agreed on a revised repayment schedule which, after a delay in serving the first installment, has been met on schedule. F. ZCDL, the Ranches and the Dairies After Loan Cancellation 1.31 ZCDL still exists, the ranches and dairies are working, and meat and milk production continues on them. Ranch and dairy development has remained almost paralyzed; only minor investments have been made since loan cancella- tion. About K. 2.5 million (US$3.5 million) was invested from 1969 to 1973; only K. 152,000 (US$236,000) was invested in 1974.1/Most of the investments financed under the project and thereafter were on buildings (37%), tractors, equipment and vehicles (20%), and fencing and handling units (18%) (see table 3). Technical and management problems still affect both ZCDL and the opera- tion of its ranches and dairies; however, some improvements were noticed. 1.32 Only one of the three ranches the reappraisal mission recommended that ZCDL give away was actually disposed of: Katete/Chinjara, which was transferred to the Zambia Tobacco Board. The other two, Kitwe and Solwezi, were to be transferred to the Projects Division of the Ministry of Rural De- velopment. but the idea was dropped after failure to agree on the valuation of assets. Provincial pressures have prevented ZCDL from disposing of these two ranches in a different way. 1.33 Several ranches are making profits now, as a result of the increase in meat prices recently authorized by Government, the tacit authorization given to ZCDL to sell part of its cattle to private butchers, and the slight but persistent improvement in the administration of ZCDL and the ranches. After an improvement in 1974, the situation of the dairy farms worsened again in 1975, in spite of the increase in milk prices (see table 4). 1.34 The company as a whole had been able to reduce its annual losses from about K. 0.6 million (US$0.83 million) in 1971 and 1972, to K. 0.25 million (US$0.4 million) in 1974, but the situation worsened again in 1975 when the company lost K. 0.56 million (US$0.87 million). The Government has been back- ing ZCDL, allocating K. 1 million in 1973, K. 525,000 in 1974, and K. 0.6 mil- lion in 1975 (US$1.55 million, US$0.82 million, and US$0.93 million, respec- tively) to help it meet it current operating deficits as well as gradually reduce its accumulated losses (see table 4). The opportune disposal of the two other ranches, as recommended by the IBRD appraisal mission, would have substantially improved the current ZCDL financial position. 1/ ZCDL did not provide for the audit data for 1975. - 15 - Table 3 ZCDL: TOTAL INVESTMENT AS OF DECEMBER 31, 1974 ('000 Kwaches) /a A. -Tvestment in Rinch/Dairy Development, at their Original Cost Value Percentage Buildings and Furniture 935.6 36.8 Tractors, Equipment and Vehicles 502.2 19.8 Fencing and Handling Units 465.3 18.3 Water Assets and Equipment 239.6 9.h Pasture Improvement 108.2 4.3 Roads and Bridges 66.8 2.7 Milking Parlors Equipment 38.7 1.6 Airstrips, Firebreakers, Silage Pits, etc. 30.8 1.2 Free Hold Land 150.5 5.9 Total 2,537.7 100.0 B. Total Investment, at Current Prices Depreciated Value of Investments in Ranch/Dairy Development Listed Above 1,56o.o h1.0 Livestock, at Current Prices 2,240.0 59.0 Total Value of Assets 3,800.0 100.0 Less Accumulated Losses - 847.6 Net Value of Assets 2,952.4 /a Exchange rate: 1 Kwacka = US$1.554. Source: ZCDL, Directors Report for the year ending December 31, 1974. - 16 - Table 4 ZCDL FINANCIAL RESULTSa ('000 kwackas) /b 1968/69 1970 1971 1972 1973 1974 1975 1. Ranches Chisamba - 1.8 - 11.2 3.7 - 4.5 8.8 52.3 0.8 Monze 10.4 - 14.6 - 17.7 - 8.1 16.8 17.4 18.6 Mugoto/Wolverton 8.7 - 19.4 2.3 18.1 6.9 48.7 45.0 North Kalomo - - 0.8 - 7.5 - 60.4 57.9 77.0 38.2 South Kalomo - - 1.1 - 15.4 - 1.3 2.8 39.5 29.4 Chisinga - 31.6 - 21.2 - 22.1 1.1 - 30.6 3.6 - 16.1 Mbala - 12.8 2.9 - 35.0 - 48.7 - 13.1 - 6.8 - 86.4 Mkushi - 17.2 - 28.9 - 4.3 - 9.1 - 6.3 - 8.9 - 17.4 Kitwe/- - 9.8 - 32.4 9.2 - 13.2 - 20.8 - 16.5 13.7 Solwezi-/c - 17.9 - 17.3 17.3 - 19.6 - 26.9 - 45.1 - 27.6 Katete/Chinjarac - 14.9 - 29.6 - 49.0 - 7.0 - - Total Ranches - 86.9 -173.6 -118.5 -152.7 - 4.5 161.3 - 1.8 2. Dairy Farms Palm Grove - - 5.2 14.3 - 22.8 - 11.1 9.4 - 28.4 Davels - - 13.1 - 21.2 - 3.3 1.8 - 22.7 Gravetts - 27.7 - 15.9 11.6 - 9.7 - 17.4 - 2.8 - 13.4 Bekkers - - - 21.1 - 32.5 - 34.3 - 38.8 - 58.4 Eldorado - - 13.0 - 22.2 - 48.9 - 35.0 - 40.1 - 33.0 Total Dairy Farms - 27.7 - 34.1 - 4.4 -135.1 -101.1 - 70.6 -155.9 3. Other Expenses (Bank Charges, Inter- est Paid, Commitment Fees, Headquarters, Workshop, etc.) - 21.8 - 218.0 -169.3 -303.3 - 277.7 -347.3 -400.3 4. Total Annual Losses -136.4 -425.7 -592.2 - 591.1 -383.3 - 256.6 -588.0 5. Government Subsidy 1000.0 525.0 600.0 6. Annual Losses and Surplus After Subsidy -136.4 -425.7 -592.2 -591.1 616.7 268.4 42.0 '. Cumulative Losses -136.4 - 352.1 -f154.3 -1745.4 -112&7 -860.3 -818.3 /a This table includes all sources of income, all current operating costs and the an- nual depreciation of the capital investments made before, during and after the loan. /b 1968-1972: 1 kwacha = US$1.400; 1973-1975: 1 kwacha . US$1.554. /c Ranches that the reappraisal mission recommended disposing of in 1972. Source: ZCDL, Directors Reports for the years ending December 31, 1974 and 1975. - 17 - 1.35 Stock numbers on ranches (28,750 beef animals in 1974 and 25,910 in 1975, excluding calves, see table 5) are over pre-appraisal level (23,400 beef animals altogether; 17,800 excluding calves) but well under the figures forecast at ap- praisal for the project's fifth year (78,000 and 56,800, respectively). They also seem to be under the stock reported by the last supervision mission (35,300 by mid-1973, but no information was given as to whether this figure included calves). On the dairy farms, Gravetts (239 and 183 cows in 1974 and 1975, re- spectively) is in between the pre-appraisal level (72) and the figure forecast at full development assumed to be reached in the first year (400). The other four dairies are under the amount forecast for their first year (920 cows in 1974 against 1060) and hold 40% of the milking herd forecast at full develop- ment, assumed to be reached in the fourth year. 1.36 The present herds' technical coefficients are far from satisfactory (see table 5). According to the ZCDL Directors' Report for 1974, calving rates were above the pre-appraisal level on seven ranches and over the envisaged full development level on six of them in 1973; but they dropped sharply in 1974, with six ranches at or under 40%, two at only 20%. Mortality rates were under both pre-appraisal and full development levels on only four ranches in 1974. Calving rates were under those forecast for the dairy farms (one with only 39%); mortality rates were higher than forecast on three of the four dairies for which information was reported by ZCDL; two of these showed 10% mortality. According to the report for 1975, calving rates improved in 1975, but mortality rates increased. Milk reported as being produced daily per milk- ing cow seems to be satisfactory on most of the dairies, but milk produc- tion related to the total milking herd is low on two dairies because only half of their milking herd was reported as being in milk.1/ 1.37 By the end of 1974, the depreciated value of the ranch and dairy developments plus the current value of the livestock (K. 3.8 million : US$5.9 million) was slightly greater (2.7%) than the value of ZCDL's share capital (K. 3.7 million - US$5.75 million). As its accumulated losses were some K. 850,000 (US$1.32 million), the net value of ZCDL's assets was about 80% of its share capital. Had it not been for the Government subsidy, by the end of 1974 the cumulative losses would have reached almost K. 2.4 million (US$3.7 million), and the net value of ZCDL's assets would have been about 38% of ZCDL's share capital (see tables 3 and 4). 1/ The information regarding milk production presented in the Directors' Report shows some inconsistencies and overstatements; thus more definite conclusions cannot be drawn. - 18 - Table 5 ZCDL: STOCK NUMBERS AND CALVING AND MORTALITY RATES Heads (Except Calves) Calving Rates/C Mortality Rates/d 1974 1975 1973 1974 1975 1973 1974 1975 1. Ranches Chisamba 3,061 2,808 69 40 58 n.a. 2 3 Monze 2,175 1,997 79 63 87 2 3 4 Mugoto/Wolverton 2,459 2,225 72 71 69 3 1 2 North Kalomo 4,342 4,615 69 61 76 n.a. 1 2 South Kalomo 2,480 2,641 79 92 80 n.a. 1 2 Chishinga 3,689 2,787 56 42 65 n.a. 6 7 Mbala 3,477 2,407 46 40 59 13 9 22 Mkushi 3,003 2,262 62 22 55 6 8 5 Kitwe /a 2,130 2,528 74 37 46 4.5 11 6 Solwezi /a 1,929 1,640 48 23 66 7.6 14 10 Katete/Chinjara /a /b - - - - - - Cows in the 2. Dairy Farms Milking Herd Palm Grove 283 298 83 81 89 10 6 9 Davels 183 212 79 39 64 6 5 10 Gravetts 239 183 63 53 73 2 2 8 Bekkers 125 n.a. n.a. n.a. n.a. n.a. n.a. 9 Eldorado 328 359 63 57 61 10 10 9 1,158 n.a. /a Ranches recommended to be disposed of by 1972 IBRD reappraisal mission. /b Ranch transferred to Zambia Tobacco Board. /c Pre-appraisal level assumed to be 77% for Monze, 60% for the other ranches, and 88% for the dairies; full development level forecast at 85% (4th year) for Monze, 70% (3rd year) for the other ranches, and 88% for the dairies. (Appraisal Report). /d Pre-appraisal and full development level assumed to be 2% for Monze and the dairy farms; pre-appraisal level assumed to be 5% for the other ranches, decreasing to 3% at full development (5th year). (Appraisal Report). Source: ZCDL, Directors Reports for the years ending December 31, 1974 and 1975. - 19 - II. CAUSES OF THE FAILURE TO COMPLETE PROJECT IMPLEMENTATION 2.01 In physical terms, the project was partially successful: most of the ranches and dairies were substantially developed; the herds increased in quantity and improved in quality; more beef and milk was made available to the Zambian consumers. In addition, the project contributed towards saving a part of the national beef breeding herd - which would have been slaughtered in the aftermath of the exodus of European farmers - and developing a number of backward areas. 2.02 Project implementation had to be suspended, however, and was not com- pleted; this shortfall has to be interpreted as a failure of both partners (IBRD and the Government) in carrying out the project they had agreed on. Several factors, already pointed out in paragraph 1.22, hampered the work of ZCDL and eventually led to a paralysis in ranch development and, there- fore, in project implementation. It is necessary to find the underlying reasons which brought about this failure. 2.03 Certain decisions taken by IBRD, by the Government, or by both to- gether, provide most of the explanation. Some of them were directly related to the project: (i) the decisions by IBRD and the Government that several times changed the project's objective; which brought about financial problems and con- flicted with the location of the original set of ranches; and (ii) some of the Govern- ment decisions regarding the ZCDL administration, which resulted in bad man- agement of the company, the ranches and the dairies. Other decisions were not strictly related to the project but affected it nevertheless: (iii) the Government decisions to divert imports away from South Africa and the then Portuguese Territories, and to hold down beef and milk prices, which combined to jeopardize the economic and financial viability of ZCDL; and (iv) the IBRD decision to reject, at appraisal, the second livestock project, which pre- vented the expected increase in the demand for weaners that the new project would have generated. In addition, the commercial objectives of the first project were rather inconsistent with the Government policy of subsidizing agricultural production and consumer goods prices. A. Decisions Regarding the Project Objectives 2.04 The project's objectives were partially or totally changed three times. As a result, four different sets of objectives can be associated with different stages in the life of the project: (i) Originally, the Ministry of Agriculture had established some ranches and dairies to try to preserve the national herd and to foster regional development; the Cold Storage Board had other ranches to serve as holding grounds (see paragraph 1.02). (ii) Later, the Government assigned those ranches and dairies the task of producing beef and milk on a commercial basis to help satisfy the growing national demand; a project and a loan application were prepared along those lines. (iii) With- out eliminating the production of beef and milk as a primary objective, IBRD shifted the main emphasis to the production of weaner steers for sale to the "emerging Zambian farmers;" the loan was made for these expanded purposes (see paragraph 1.11). (iv) During project implementation, the Government progress- ively imposed socially oriented responsibilities on ZCDL, such as purchasing - 20 - cattle in bulk from departing European farmers at prescribed prices to prevent depletion of the national herd, and subsidizing the "emerging farmers" by sell- ing them weaners and culls at prices lower than projected at appraisal (in fact, at a loss). As a result of these changes, ZCDL encountered serious un- anticipated financial problems, and the set of ranches selected between 1965 and 1967 by the Ministry of Agriculture and CSB to meet the original set of objectives became partially inadequate to meet the new objectives. 2.05 The original objectives were to be pursued with allocations from the Government budget. Additional money had to be invested to further develop and to stock the ranches and dairies when the objectives were changed to com- mercial production. IBRD and Barclays loans, plus additional Government equity, were called on to finance them. The fixed investments and working capital needed were calculated; the respective cash flows were forecast for the ranches, the dairies and the company as a whole; measures were recom- mended and adopted to fulfill their requirements during implementation. But when the Government imposed the above-mentioned new social responsibilities on ZCDL, no change was made in the financial arrangements. ZCDL had to find unbudgeted funds to buy cattle, in numbers beyond the ranches' current need, from departing European farmers.l/ Meanwhile, the stringent Government reg- ulations prevented ZCDL from making cash selling the extra stock that would otherwise have been culled. Furthermore, the financial situation of an enter- prise like ZCDL, established as a commercial unit and expected to operate as such, suffered even more when its objectives were shifted from the commercial area into activities involving the transfer of subsidies to certain groups of farmers, such as the "emerging farmers." ZCDL could have been efficiently run as a mechanism to transfer Government funds into other hands, but not to transfer as subsidies part of its own equity or working capital. 2.06 The financial situation of ZCDL worsened when the increase in the demand for weaners, whose appearance had been assumed when the third set of objectives was established, failed to materialize. ZCDL had to keep and fat- ten its own weaners, and its financial structure and cash flows were greatly disrupted as a result of the deferred sales.2/ 1/ In addition, the Government requested that ZCDL buy them in bulk and at a pre-fixed price. Therefore, ZCDL could not either acquire the best animals, previously sold at a better price, or reject the inferior ones. 2/ The appraisal report stated that "if demand for weaner steers and heifers slackened for any reason in the foreseeable future, ZCDL would have no dif- ficulty, in light of suitable empty veldt available near its ranches, in grass fattening these weaners to slaughter age." This statement, though correct on technical grounds, overlooked the financial implications of such a change in the timing of sales (which may be delayed for two or three years)on ZCDL's cash flow. - 21 - 2.07 The Ministry of Agriculture had scattered its ranch acquisitions all over the country (at least one in each province; most in remote places, some in areas whose inhabitants had never seen a herd of cattle before) to meet the first set of objectives mentioned above: to satisfy regional (and so, limited) demands for beef, milk, weaners, heifers, bulls and oxen; to introduce cattle raising in heretofore ungrazed parts of the country, and to increase regional income in the non-commercial, socially oriented way the Government wanted it to be at that time.1/ However, to produce beef and milk on a commercial basis, as the Government decided later, differ- ent ranches and dairies should have been selezted, with adequate ecologi- cal conditions and close proximity to roads or the railway, and to the mar- kets. Some of the existing ranches and dairies were adquate indeed for both purposes, but others were not. Later on, the appraisal mission shifted the Enphasis from beef and milk to weaner steers. The best ecological condi- tions for producing beef and milk differ from those for producing weaner steers, and so does the spatial location of the firms or individuals demand- ing each of them. Therefore, if a critical review of the proposed ranches and dairies was required after the first change, it was absolutely unavoidable after the second change - but it was not done. One of the proposed dairy farms (Kafubu) was in fact rejected by the appraisal mission because it was found to be unsuitable for milk production; Barclays Bank's Agricultural Development Officer, IBRD's supervision missions, and ultimately the re- appraisal mission, were to point out later the unsuitability of some of the other ranches for the project's current objectives. 2.08 It is rather difficult to avoid the conclusion that MID, as well as the Government, made a wrong decision in basing the project o:. a set of ranches which included some inadequately suited for producing either beef and milk or weaners. Better project preparation could have prevented such a situation. Later on, IBRD should have rejected some ranches as part of the project to be financed with the loan (the appraisal mission must be credited with the rejection of the proposed Kafubu dairy farm, but it added six addi- tional ranches and dairies), and should have restricted it to those ranches which lent themselves to the kind of production the project tried to promote - as IBRD itself had to request three years later. A rather smaller,2/ but more manageable, more profitable and perhaps eventually successful project might have resulted had the obviously unsuitable ranches 3/ not been accepted by IBRD; the financial position of ZCDL would have been substantially better, even al- lowing for the detrimental effects of the managerial problems (described 1/ It can be questioned whether better locations might have been found in each province at that time, but this query does not change the main point, that is, that the ranches were selected looking for objectives different from those chosen later. 2/ Perhaps too small to qualify as a Bank-financed project or to be inter- esting enough to the Government. However, it must be kept in mind that this project had been conceived as a first phase leading to a broader proj- ect to be prepared and approved shortly thereafter. 3/ Those later mentioned by IBRD after the reappraisal mission findings: Kitwe, Solwezi and Katete/Chinjara. - 22 - later on).l/ 2.09 By accepting all but one of the ranches and dairies already in the hands of ZCDL, IBRD implicitly linked the fate of the project to the viability of this "new, untried company." When ZCDL became almost bank- rupt, project implementation had to be suspended, despite the fact that ranches were being developed (though more slowly than expected), herds were increasing and improving, and some of the ranches were already making prof- its (see table 4). B. Decisions RZgarding the ZCDL Administration 2.10 After the successful start-up period, the absence of adequate management to run both the company and the ranches and dairies was detri- mental to the project. Two Government decisions affecting the ZCDL admin- istration contributed to this situation: (i) the decision to pay managers modestly; and (ii) the decision to limit ZCDL's right to define and imple- ment its own commercial policies. 2.11 Big investment and cheap management do not mix. The Government set up ZCDL to invest US$5.8 million over a 5-year period. The Government should not have decided that a General Manager costing US$24,000 was too ex- pensive, nor have allowed ZCDL to hire ranch and dairy managers at sal- aries under those prevailing in the civil service. But it did both. IBRD strongly stated that "the management of ZCDL, a new, untried company, is likely to be crucial to the success of the project," and required that ZCDL "have as General Manager a person whose qualifications, duties and powers, as well as the terms and conditions of his employment, are accept- able to the IBRD." However, it did not challenge RDC's argument regarding the "excessive cost" of ZCDL's first General Manager. In retrospect, it is clear that from the strict, narrow point of view of the project, his dismissal (on these and other grounds) was detrimental to its implementa- tion. In the case of the ranch and dairy managers, many of the first group of them quit their jobs; some of them had to be replaced by less qualified people willing to work for the low salaries paid by ZCDL. 2.12 It was already mentioned that the Government did not give ZCDL the independence and authority it needed to formulate and implement its own commercial policies.2/ IBRD has always tried to prevent political inter- ference in the management of the projects supported by loans or credits. Several times it has insisted that a special project unit or an independent agency be set up for this purpose. In this particular case, it was not 1/ The same ZCDL management could have been more effective had it had fewer ranches to look after. 2/ It is quite clear that this decision was linked with the previous one: the Government might have given ZCDL more freedom had it been able to find better managers to run ZCDL. - 23 - enough. IBRD should have required that the Government allow ZCDL to operate on a fully commercial basis long before the reappraisal mission pointed it out as one of the basic requisites to further project implementation. C. Decisions Regarding the Project's Economic and Financial Viability 2.13 Other Government decisions proved to have a detrimental effect on the financial viability of ZCDL and, therefore, on the project, namely: (i) to direct imports away from South Africa and the then Portuguese Territories, and (ii) to keep down beef and milk prices. 2.14 The decision to divert imports, whenever possible, toward countries the Government considered more friendly to its policies led to marked in- creases in import prices and acute shortages of some equipment and supplies procured by ZCDL.l/ 2.15 Since Independence in 1964, the Government as a matter of national policy had kept consumer prices low, allowing increases only when it was ab- solutely necessary. To implement this policy, payments to some producers were made by the Government.2/ Beef and milk prices were kept low by the Cold Storage Board, however, without any compensatory payment ever being made to the producers. Thus, ZCDL had to transfer part of its expected profits (as it made no profits, of its capital) to consumers, without com- pensation. D. Decision to Reject the Second Livestock Project 2.16 The mission appraising the first project overestimated the current demand for weaner steers by the emerging Zambian farmers, but was right in assuming that the second livestock project, which was to be prepared as soon as possible, would foster it. The rejection of the second project at ap- praisal dealt a severe blow to the first one. Therefore, it is relevant to analyze why that project was rejected. 2.17 As required by IBRD, the project was jointly prepared by a Govern- ment committee and a consultant firm, financed under the first loan. The terms of reference and the plan of operations for project preparation were 1/ IBRD's supervision missions realized that this situation would weaken the financial and economic viability of the project. However, a para- graph drafted by one of the supervision missions, asking the Government to compensate ZCDL for the additional costs incurred because of the new import policy, was not included in the letter sent by IBRD to the Gov- ernment after the mission returned. 2/ Subsidies had to be reduced by Government in January 1976, due to the drastic cutback in the revenues from the copper industry. -24 - written by the Government with the help of the members of the first super- vision mission to Loan 627-ZA. Project preparation started in late 1969. A mid-term review of the draft project document (Interim Report) was held in IBRD headquarters in Washington in March 1970. The formal application was presented by Government in August 1970 and appraised in September- October 1970. 2.18 The appraisal mission found the project as proposed by Government unsuitable for IBRD financing. The envisaged collective ranches were found to be not financially viable and to require a Government subsidy through- out the life of the project; the tenant ranching and tenant dairy systems would involve making loans to uncreditworthy sub-borrowers with inadequate training and experience; the large scale Zambian commercial farmers, who it was envisaged would participate in the project, did not exist; and the tech- nical coefficients assumed in the project were described as unobtainable. 2.19 A second, broader livestock project was needed both for helping Zambia to develop its livestock sector and for increasing the demand for the weaners to be produced under the first project. The actual application sub- mitted by the Government, however, did not meet even the minimal IBRD stand- ards of financial and economic viability; the appraisal mission's decision to reject the application was. in the view of the audit, justified. 2.20 How could it happen that the Government and the consultant sub- mitted a project proposal so different from what IBRD could accept, espe- cially when a mid-term review had been made on project preparation only five months before? Several factors brought this situation about: (i) Government policies conflicted with IBRD's lending criteria. The Govern- ment was trying to develop cooperative and family farms, even if that might mean heavy subsidization, and de-emphasize large privately owned commercial farms. This resulted in ranching schemes that were not self- supporting. (ii) The four reasons. reflecting those Government policies, for which the project was eventually rejected, were present in the Interim Report that was discussed with IBRD in March 1970. Both the Government representatives and the consultants left that difficult meeting under the impression that IBRD had approved the progressive strategy of livestock development proposed by the Government delegation and the interim docu- ments.l/ Unfortunately, no formal agreement or formal minutes were written and approved by the participants. Neither the Government nor the consultants sent a letter of understanding to IBRD after the meeting; nor did IBRD send one to them.2/ Therefore. the joint Goverument-consultant team proceeded 1/ Both the Government and the consultants still think that the second proj- ect provided the correct approach to Zambia's livestock development, even though it was submitted in a form that IBRD could not accept. 2/ The participants kept their own minutes, recording their own recollec- tions. Later on. each participant's recollections proved to be different from those of the others. - 25 - to prepare the project along the lines they had understood had been approved. (iii) The consultants did not warn the Government strongly enough about IBRD standards of viability. They were under the impression that IBRD had agreed to the program's overall strategy and would not, therefore, be rigid on lend- ing criteria. In addition, the consultants have said that in most cases work- ing in joint teams with governments inhibits independent recommendations to some extent. They prefer to be given an independent role, though they under- stand that a joint role may sometimes be necessary. They think that since IBRD had insisted, in this case, on the joint team approach, it should also have appreciated its limitations. (iv) None of the members of the Bank team appraising the project had been at the March mid-term review meeting. The change in staff resulted in at least one change in IBRD's position: accord- ing to the consultants. they had altered certain coefficients (notably milk yields) because IBRD found those in the Interim Report to be too low; later the appraisal mission found the amended coefficients too high. (v) IBRD pres- sured both the Government and the consultant to prepare the second livestock project in a limited period of time, a period the consultant argues was much too short. 2.21 IBRD asked the Government for further preparation and the intro- duction of several modifications. It proved difficult to modify the pro- posed second livestock project so that it would become economically viable. Instead, the Government and IBRD agreed to prepare a new combined livestock and crop project. After several exchanges back and forth within the Govenm*t, this mixed farming project loan application was finally submitted to IBRD in December 1972 and appraised by it in March 1973. In February 1974. IBRD invited the Government to negotiate the loan, but negotiations were put off five or six times. Eventually, in April 1975, the Government with- drew the project from IBRD, three years after the implementation of the first livestock project had been suspended. E. Lack of Consistency Between the Project and the Government's Agricultural Production and Consumer Price Policies 2.22 As a part of its development policy. the Government of Zambia has subsidized agricultural production since Independence. At the same time, it has subsidized consumer goods prices. A project such as the First Livestock Project, which called for commercially oriented production of beef and milk in State-owned ranches and dairies, was rather inconsistent with both. (This inconsistency was made expl.cit when the second live- stock project was prepared, presented to IBRD, and rejected by it.) The Government should have either prepared the first project along very dif- ferent lines (and perhaps financed it without IBRD assistance) or changed its overall policy on subsidization. The inconsistency between the policy and the project contributed significantly to the weakening of ZCDL as a company financially able to carry out the project. - 26 - 2.23 IBRD shares the blame for not having noticed this inconsistency and realized its consequences. The relevant IBRD relationships with the Government were handled mainly through ADS, a technical assistance body that, in addition, happened to be very strongly committed to the project. Area and projects staff, who were in a much better position to analyze sectoral policies and detect such inconsistencies, had only minor participation in that relationship. - 27 - III. BORROWER PERFORMANCE 3.01 This was the first IBRD-financed project ever carried out by Zambia in the agricultural sector, and it was carried out by a newly estab- lished State-owned company under a parent State-owned corporation which was also newly formed. The loan was made only five years after Independence. The following comments must be interpreted within that framework. 3.02 It has already been mentioned that: (i) the Government undertook to carry out a project rather inconsistent with its agricultural production and consumer price policies (see paragraph 2.22); (ii) upon accepting the loan, the Government committed itself "to cause the project to be carried out with due diligence and efficiency and in conformity with sound admin- istrative. . . practices. . . ."; the Government did not fully meet this covenant (see paragraphs 1.19(i), 1.20, 1.22(i), and 2.10-2.12); (iii) the Government and ZCDL must be praised for the rapid start-up of project im- plementation after effectiveness, a feature seldom achieved in other live- stock projects, in Africa or elsewhere; (iv) IBRD cannot dispute the Gov- ernment's right to decide where imports should come from or through, and the way in which copper revenues should be allocated between current public expenditures, long-term investments, and subsidies. However, once the Gov- ernment agreed with IBRD to carry out a commercial production project, the Government should have subsidized ZCDL for the losses the company suffered as a result of the implementation of the above-mentioned decisions. 3.03 The Government never consulted with IBRD concerning the producer prices for beef cattle and dairy products, as it had agreed to do "from time to time" and, in fact, did not review those prices during project im- plementation. The Loan Agreement did not stipulate, however, at whose in- itiative those consultations were to be made. 3.04 During h! first two years, the presentation of reimbursement claims by the Government to IBRD was delayed because of (i) the long delays ZCDL confronted in obtaining land titles for its ranches and dairies from the Government; and (ii) the lack of clear understanding between IBRD and ZCDL as to the precise requirements in the field of documentation and other detailed matters required by IBRD to process claims. IBRD had to send two detailed explicative letters to solve the problem. - 28 - IV. IBRD PERFORMANCE 4.01 The overall IBRD performance was not completely satisfactory. In most :ases IBRD helped considerably, but better decisions could have beenr adopted at certain times. 4.02 LoRD was correct in giving the country a prompt response, by sup- porting a limited (despite some expansion at IBRD1s initiative) first phase livestock project, while allowing that a second broader and more comprehensive phase be prepared. It would not have made sense to delay the development of the State ranches while waiting for the more complex sector-wide project. 4.03 The conditions of effectiveness were extremely simple: mainly some legal steps to be taken and some evidence to be furnished to IBRD. No technical or organizational requirements were set up as conditions of effectiveness. The provision of land titles for the ranches and dairy farms involved in the project was made a condition of reimbursement for the expenditures actually made on each ranch or farm but not a condition of effectiveness for the set of ranches or the project as a whole. This approach differed from the one followed in some other IBRD/IDA projects,in which several technical, policy, or organizational matters were set up as conditions of effectiveness, delaying unnecessarily the implementation of the project as a whole or of some of the project components, which im- plementation did not require that such conditions be met.l/ 4.04 IBRD was saiisfactorily flexible in handling a Governme. t request to change the rules for frocurement. ZCDL requested in March 1971 that the Bank waive its requirement for international and local tenderi,g. Re- garding international tendering, ZCDL's annual requirements were relatively small and even when bulked did not attract international interest. There had been little response to the tender invitations issued in late 1970 through the full procedure; in addition, suppliers were reluctant to quote prices f.o.b. Lusaka because of the chaotic conditions of transportation from the coast. Regarding local tendering, ZCDL argued that the time re- quired for tendering might cause severe delays in the implementation schedule of the project: there were only a couple of potential local suppliers; these did not carry stocks of some essential items; orders placed would lead to delivery 9 or 10 months later. On the basis, on the one hand, of the very special situation Zambia was in at that time, and, on the other hand, of the need to maintain a consistent Bank policy and to avoid creating a pos- 1/ Cf. OED Kenya First Livestock Project (Credit 129-KE) - Project Performance Audit Report, Report No. 1317, October 19, 1976. - 29 - sible preceden, IBRD agreed that full international competitive bidding was not required if procurement were to continue to be undertaken on an annual basis, which did not seem to result in contracts exceeding the US$25,000 min- imum stipulated in the Loan Agreement. In the case of local tendering, IBRD did not grant a formal waiver, but agreed to consider individual cases of "shopping around" whenever local tendering did not result in any satisfactory bids. 4.05 The recommendations of the reappraisal mission were appropriate in terms of the objectives agreed at appraisal, and,had they been imple- mented, would have put the project back on better footing. 4.06 The IBRD decision to reject the second livestock project as sub- mitted by Government was right, though IBRD had some responsibility for the events that led to its rejection. 4.07 In addition, some Government officers pointed out that IBRD had been very helpful to the country in setting up the project; it acted like a catalyst that helped to put together resources already in the country. 4.08 IBRD's main shortcomings were related to pro*ect conception and appraisal: (i) all the ranches presented by the Government were accepted by IBRD; Barclays Bank and the IBRD supervision and reappraisal missions were to say later that some of these ranches were not suitable for the pro- duction objectives envisaged; (ii) IBRD did not realize that the project, as proposed, was rather inconsistent with some overall Government policies; (iii) IBRD accepted that the loan covenant regarding minimum prices for beef and milk be worded in such a way as to make its enforcement impossi- ble; and (iv) the earlier IBRD criticism regarding the financial prob- lems which ZCDL would incur was overcome once the Government gave as- surance that it would contribute additional funes as required to wipe out ZCDL's expected operating deficits. 4.09 Two reasons might explain IBRD's actions during project prepara- tion and appraisal. First, at the time the project was being prepared and appraised, IBRD was carefully establishing good relationships with the Gov- erLnent of Zambia and tried to avoid any decision that might jeopardize those relationships. However, a more critical review of the project and the loan application would have done less harm to the relationships between IBRD and Zambia than the set of problems that arose during implementation and cancel- lation.l/ Secondly, three different IBRD units confronted each other during that period: (i) the Africa Department, which wanted IBRD to become involved in financing agricultural projects in Zambia as soon as possible; (ii) ADS, which wanted IBRD to accept the Government's loan application (based on ADS's report) as it was presented; and (iii) the Livestock Division (LD) at head- quarters, which analyzed the project like any other livestock project, making 1/ Both IBRD and the Government describe their present relations regarding agricultural projects as normal. - 30 - the criticisms already mentioned (see paragraph 1.05). It may be assumed that IBRD's position regarding the loan application would have been more critical had LD had a stronger say prior to appraisal. 4.10 The project design was too optimistic: the targeted technical coefficients were set at levels almost impossible to reach during the project life (only one or two ranches have achieved them); too much was expected from the emerging Zambian farmers, a group whose existence was more a hope than a reality; too much was expected from ZCDL, a "new, un- tried company." IBRD is to blame for the project's overly optimistic design, although the Government is not flawless in this respect, for it did not say clearly what it wanted and needed, and accepted trustingly what IBRD technicians proposed. Too little was known at that time about the technical parameters of several ranches (some of them were situated in regions in which no cattle had ever been raised before) and about the sectoral and national constraints on designing and implementing such a project. Most of the actual problems and constraints were discovered when ZCDL began working on each ranch. A thorough sub-sectoral diagnosis might have avoided most of the above-mentioned design flaws. 4.11 The appraisal mission's introduction of weaner production as a project objective, with greater emphasis than that on beef and milk pro- duction, was based mainly on the mission's interviews with knowledgeable Government officer; and with villagers who expressed interest in buying steers for fattening. However, no study had hitherto been made of the demand for weaners (current sources of supply, number traded each year, prices, age and weight, kind of ranches or mixed farms demanding them, location, etc.), nor was it made by the mission. Later on, it was dis- covered that such a demand did not exist in the amount expected by the mission. 4.12 The appraisal mission had correctly attributed great importance to the CSB Grazier Scheme, for the latter was crucial to the marketing of project output and its failure to perform as expected would have been criti- cal. Since the existing financial set-up was very weak (90-day Treasury Bills), IBRD required at negotiations that the Government (and the Govern- ment agreed to) "make arrangements satisfactory to the Bank for refinancing the Grazier Scheme. . . so as to reflect adequately in the loans by which the Scheme is financed the length of the contracts granted to ranchers parti- cipating in the scheme" (see paragraph 1.16). All of the supervision missions overlooked the implementation of this covenant; no reference to its fulfillment was made by any supervision report or even by the reap- praisal mission report. IBRD could not have known whether this covenant had been accomplished and, if not, how important this shortfall was in ex- plaining the problems encountered by ZCDL in marketing its output. - 31 - 4.13 During project implementation, IBRD did not request that the Government: (i) increase beef and milk prices (or even consult with it "from time to time" about the necessity and timing of such revisions); and (ii) compensate ZCDL for the additional costs incurred because of the new import policy. Through such requests, IBRD might have succeeded in having the Government take either or both actions, thereby improving ZCDL's economic and financial position and making it possible to continue implementing the project. No definite explanation for these oversights was obtained during the auditing process. Perhaps IBRD thought that pre- serving its relations with Zambia was worth overlooking its dissatisfac- tion with these particular Government shortcomings. - 32 - V. CONCLUDING REMARKS 5.01 Some additional comments may be offered regarding the different way in which the country and IBRD look at a project, the proper balance between optimistic and pessimistic assumptions in project preparation, the role of both the IBRD area and projects departments in project gen- eration, the decision to finance jointly this particular project with a commercial bank, and the role of expatriates in project management. 5.02 According to Government officials, the "project" succeeded in in- creasing the national meat and milk production and in training some Zambi- ans in cattle raising and in ranch management. Further, the "project" is still being implemented, though with far fewer resources and less techni- cal advice than before loan cancellation. From IBRD's point of view, "project" implementation was never completed, its goals were not achieved, and the "project" had to be discontinued. Although in most projects fi- nanced by IBRD/IDA the two viewpoints seem to be the same and are treated as such, this experience makes it clear that the Government's objectives and views can indeed be different from those of IBRD: (i) The Govern- ment's "project" was to further develop certain sets of ranches and dairies to increase their production; IBRD was called upon to help with the design and temporary financing of this development in order to achieve it earlier than would have been possible relying only on national resources. The Government always looked at the "project" as its own, and felt it had an exclusive right to decide the ultimate objectives to be pursued; IBRD was looked at as an agency that could temporarily help the Government in this effort. (ii) IBRD's "project" was to develop those ranches and dairies, starting at a certain level (pre-appraisal) and bringing them to another, higher level (full development) in a certain span of time (disbursement period - see diagram 3). IBRD asked the Government to make some changes in the project design and to adopt certain policies or decisions during project implementation and on the grounds that they were needed to ensure project success. The Government agreed in writing to these changes (Loan Agreement, Project Agreement) but later felt itself unduly tied in their (i.e., Government's) project implementation by these covenants. They felt that IBRD's insistence on their fulfilling these covenants constituted in- terference in Government affairs. The difference between the legitimate objectives of both IBRD and a government in setting up a particular project must be kept in mind to avoid misunderstandings between the two partners. If government objectives later shift, or crystalize around a different rank- ing of priorities, extreme action (cancellation) may be the only solution. 5.03 On occasion the objectives or assumptions of a project have been described retrospectively as over-optimistic. Even when correct, this kind of characterization sometimes can be dangerous and unfair; for the opposite "sin" is equally serious, if not more so. Any project is an endeavor, a kind of quest, that tries to achieve certain goals within certain sets of constraints. The right balance between possibilities and constraints is very difficult to achieve and even to define. If too much consideration - 33 - Diagram 3 THE "PROJECT" AS SEEN BY THE GOVERNMENT AND IBRD Government Perspective (shaded area) Level of ranch and dairy development Target :r Actual achievement with the help of IBRD and Barclays Pre-development Hypothetical level that would have been achieved by Govern- ment alone Year 0 for Period of accelerated development the country thanks to the help of IBRD and Barclays IBRD Perspective (shaded area) Target Shortfall Actual . achievement Pre-development . . "Ex-post" level without level 'the project Year 0 Year 5 for IBRD Closing Date Project Implementation suspended - 34 - is given to the actual and potential constraints, usually very noticeable in any developing country, the resulting project will be pessimistic and, probably, no effort will be made to improve the present situation. The situation in Zambia, as reflected in a large number of other programs then in implementation difficulties, was not especially encouraging. The Gov- ernment lookedto the Bank for support; the Bank was eager to help. What is the relative role of the "optimists and the "pessimists" under such condi- tions? 5.04 Any particular project should stem from the combined effort of both IBRD "area" and "projects" departments. Though both of them were in- volved, initial IBRD support derived mainly from the "area" department's work; IBRD's acceptance and financing was based more on "area" policies and priorities than on "project" priorities. The "area" department had been seeking an agricultural loan to Zambia for a long time; after the re- jection of other alternatives, the further development of the State ranches was the only option left. The "projects" department criticized the proj- ect as prepared by ADS, and explicitly stated that there was no basis for ascertaining whether this project was the best kind of project for the development of Zambia's livestock sector. This criticism was correct. Originally the State ranches report was not intended to lead to a Bank project; it became one when the other possibilities failed to materialize. It was not based, as it should have been, on a diagnosis and strategy for the Zambia livestock sector. Neither existed at that time. When the de- cision to lend for livestock development was made, the Livestock Division, which represented IBRD in the "follow-up preparation" mission, had to agree that the State ranches project was the only livestock project immediately bankable available in Zambia. Whether such "area-sponsored" projects are generally less well identified and prepared is a question beyond the scope of the audit. 5.05 The joint financing of development projects by a development fi- nance agency and a commercial bank has worked well in many cases. However, it has limitations and involves risks on which this case sheds light. The problem may well surface more often when the financial situation of the ex- ecuting agency is jeopardized, as in this case. IBRD, as a development agency, came out of the reappraisal exercise in an optimistic mood: most of the ranches and dairies had been substantially developed, herds were im- proving in quality, the project was making some contribution to the country's livestock development and balance of payments. Barclays Bank, as a commer- cial bank, with its eye on different indicators, verified that the executing agency (ZCDL) was almost on the verge of bankruptcy, and was not going to be able to serve its debts out of its income. IBRD, in similar cases, usu- ally tries to help the Borrower overcome such problems - by proposing cor- rective measures, discussing new implementation and disbursement schedules, modifying project design, etc., as long as it thinks the project can still be adequately carried out and contribute to the country's development. Barclays Bank's decision to suspend disbursement doubtless gave more em- phasis than did IBRD to the risk of continued losses. No one can blame it for doing so, although it seriously impaired continued project implementa- tion. - 35 - 5.06 The use of expatriates in line positions has raised important is- sues in livestock projects in other African countries. This was not the case in Zambia. The first three ZCDL General Managers were expatriates,but only minor frictions with the Government can be attributed to this fact (see paragraph 1.20). Eighty percent of the ranch and dairy managers were ex- patriates at the project start, and fifty percent when project implementa- tion was suspended; Zambian managers took over from expatriates on seven ranches and dairies, but the reverse also occurred at certain times on two of them (see annex 3). However, no reference to the expatriate ranch/ dairy managers as a group or to their performance can be found in either the Government's reports and correspondence, or the supervision mission re- ports; nor can the actual success or problems in particular sub-sets of ranches or dairies be explained by the existence, absence, or replacement of that group, or individuals in it. Usually IBRD/IDA requires that an expatriate be hired as Project Manager; in this particular project, how- ever, it was the Government who asked for one.l/ 5.07 Nevertheless, some problems can be associated with the use of ex- patriates. The appointee as ZCDL's first General Manager was a hard-working man, extremely dedicated, and had very good relationships with his sub- ordinates. But he worked alone, separate from Government, trying to im- plement the project as agreed between IBRD and the Government. This led to two apparently contradictory situations: (i) his leadership allowed proj- ect implementation to start quickly and proceed at a fast pace, which was noticed by both the Government and IBRD; and (ii) his behavior alienated Government support and eventually led to his dismissal. The Government had agreed with IBRD to carry out a project rather inconsistent with some of its own fundamental social policies. When the first General Manager in- sisted on implementing the project as it had been agreed to, the Government concluded that he did not cooperate with the Government, that he had a con- flict of allegiance, and that he felt more responsible to IBRD than to the Government; they looked at him as an IBRD officer. The expatriate who re- placed him got along with the Government, but he was ineffective as ZCDL manager. Both the first and the second were unable to train a Zambian of- ficer to replace them.2/ Besides technical and managerial expertise, an expatriate serving in line positions must have very special capability to serve as broker between the partners while avoiding identification with an "expatriate" position. 1/ Perhaps IBRD did not ask for an expatriate because the Government had already asked for an expatriate seconded from ADS as General Manager of ZCDL before the project was appraised. Though IBRD required that "any appointment to the position of ZCDL's General Manager shall be made only after agreement with the Bank" (Project Agreement, Section 2.01 b), no specification was made regarding his nationality. 2/ The Government was also to blame for this shortcoming, for it did not appoint any Zambian as their deputy to be trained on the job.  ANNEX 1 SCHEDULE OF DISBURSEMENTS (US$ million equivalent) Latest Fiscal Year Appraisal Revised Actual and Quarter Estimates Estimate Disbursements 1969/70 1st 2nd 3rd hth 0.65- 1970/71 1st - - 2nd 0.40 0.30 3rd 0.70 0.46 hth 1.30 1.00 0.50 1971/72 1st - 1.30 (?) 2nd 1.70 0.83 3rd - 2.10 hth 1.87 2.20 1972/73 1st - - 2nd 3rd hth 2.25 2.25 1973/74 1st 2nd 3rd - hth 2.50 (?) ANNEX 2 Breakdown of IBRD Funds Disbursed, by Category (US$ equivalent) Amount Loan Disbursed Category I Equipment 500,000 Nil Category II Technical Services 200,000 46,770 Category III Artificial Inseminption 100,000 Nil Category IV Other Dev. Expenses 1,400,000 78,792 Category V Unallocated 300,000 Nil Total 2,500,000 832,562 ANNEX 3 ZCDL: Breakdown of Ranch and Dairy Managers Between Zambians (Z) and Expatriates (X) July Oct. Mar. June Sept. Dec. Mar. June Sept. Dec. Ranch/Dairy Farm 1969 1969 1970 1970 1970 1970 1971 1971 1971 1971 Mbala X X X X X Z Z Z Z Z Chisinga X X X Z Z Z Z Z Z Z Solwezi X X X X X X X Z Z Z Kitwe X X X X X X X X X X Chisamba Z Z Z X X X Z Z Z Z Mugoto/Wolverton X Z Z Z Z Z Z Z Z Z Mkushi X X X X X X X X X X Monze X X X X X X X X X X Katete/Chinjara Z Z Z Z Z Z Z Z Z Z Gravets X X X X X X X X X X Kalomo North X X X X X X X Eldorado X X X Z z x Palm Grove X X X X X X Kalomo South x x x x Bekkers X X z Davels x x z Percentage of Zambians 20 30 30 27 23 31 36 44 44 50 Source: ZCDL, Quarterly Reports submitted to IBRD. s ZAMBIA LAKE TANGAN1KA LIVESTOCK DEVELOPMENT PROJECT T A N Z A N l A PROJECT RANCHES AND DAIRY FARMS LAKE mwåHU Project ranches and dairy farms RANCH Areas ftom uhich cattle -ovements are restricted by ve r Ynary eg- THIS MAP IS BASED ON IBRO 2468. MAY 1969 ulatin' ONLY PORTIONS OF THE PROJECT HAVE BEEN CHISHINGA X../ Main road, UPDATED AN) SEVERAL NAME CHANGES KAWAmeWAI RANCH -- --- . $ odary rods 'LU¥nNG AAA$K (' REUBL IC f 5 -7 O- -II TN f - FORT ROSEDER tNI LNGA - A N O L A mOLWZ -- ) SOLWEZZ1; RANCH CHIN ÖL KITWE ITE 1 RANCHNOL LUJANS YA C IJR ALOVALE KASEMPAL SERENJEiRAC MKUJSHI TTE RANCH Kap,,, MpcIsht MKUSHIr RANCH Ktt KABWL l'MNKOA GRAVE TTS HRAMBA MAKOAOAIRY FARM RANC MONGUI ,MUMB1WA Ch,somb. WOLVERTONý DAVELS RANCH ''°U ADAIRY FARM M O Z A M B l Q U E N MWALoplU AA Amsel MON7F17 abk Kul.b. RANCH PALM GRVB EB N ---- ~ ~ 'NORTH KALOM 0! MUGOTO ELOAODRYF S [ RANCH _RANCH Iulobe . - K-n K5om \ A F R I C A --SOUTH KALOM OZAMBI - -- RANCH- -~~ LIVINGSTONE · 2 s 75 o - - LES1