Romania - Agricultural Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 13830 PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (LOAN 1876-RO) FOURTH LIVESTOCK PROJECT (LOAN 1937-RO) MOLDOVA AGRICULTURAL CREDIT PROJECT (LOAN 2077-RO) DECEMBER 28, 1994 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. ACRONYMS BA Agriculture Bank (Banca Agricola) BAFI Bank for Agriculture and Food Industry BBSP Buscani-Buzau-Siret-Prut Irrigation CPS Central Project Staff ERR Economic Internal Rate of Return MAFI Ministry of Agriculture and Food Industry MEBO Management-Employee Buy-Out System MOA Ministry of Agriculture SMA Stations for Agricultural Mechanization FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 28, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on ROMANIA - Orchards Project (Loan 1876-RO) Fourth Livestock Project (Loan 1937-RO) Moldova Agricultural Credit Project (Loan 2077-RO) Attached is the Performance Audit Report on the above three agricultural projects prepared by the Operations Evaluation Department. The projects were to expand capacity in their respective subsectors through subloans made by the Bank for Agriculture and Food Industry, mainly to state and cooperative farms. The projects were implemented as envisaged and closed as scheduled. Their outcome was considered at the time to have been satisfactory. The Performance Audit Report reaches different conclusions and revises the performance ratings of all three projects which had been based on "desk PCRs." Project design is assessed as faulty, capital intensive and distorted by centralized decision making. Technology transfer is judged to have been inhibited by government policy geared to securing free foreign exchange by ensuring that all bids would be won by domestic suppliers. The socialist structure of production made support to individual producers unacceptable to government and constrained effective marketing. Attachment 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.  FOR OFFICIAL USE ONLY PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (Loan 1876-RO) FOURTH LIVESTOCK PROJECT (Loan 1937-RO) MOLDOVA AGRICULTURAL CREDIT PROJECT (Loan 2077-RO) TABLE OF CONTENTS Page No. P reface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i B asic D ata Sheets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Evaluation Sum m ary .... ................................. . .. ix 1. BACKGROUND ... I 2. THE PROJECTS ............................................ 2 The O rchards Project ........................................ 2 The Fourth Livestock Project ................................... 3 The Moldova Agricultural Credit Project ........................... 4 3. DEVELOPMENTS SINCE LOAN CLOSING .......................... 5 4. PROJECT STATUS AT THE TIME OF AUDIT. .........................6 Orchards ................................................6 Cold Stores .............................................. .7 Dairy Farms, Beef Fattening and Heifer Raising Centers. ..................7 Improved Pastures........................................... 7 Erosion Control Works. ....................................... . 7 Vineyards................................................ 8 Agro-Industries............................................. 8 Farm Machinery............................................ .8 Laboratory Equipment....................... ......... ....6.....8 5. PROJECT RESULTS...............................................8 Orchards Project............................... ......7.... . 9 Fourth Livestock Project............................. ....7.... 10 Moldova Agricultural Credit Project............................... 1 Repayment of Subloans under the Three Projects ............................. 11 This audit was prepared by Josd Olivares, Task Manager; Charina Go provided administrative support. 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.I Page 6. ISSUES AND LESSONS I I The Bank's Contribution to the Projects .............................I I Financial Participation ......................... ........ 12 Technology Transfer............................. . . . . . . ..13 Support to Individual Producers........................... . ..15 The Borrower's Contribution to the Projects...................... . ..15 The Fourth Livestock Project and the Economic Analysis of Projects in the Bank . . . 16 Project Sustainability After the Revolution: Four Transitional Issues.. . . . . . ..18 Land Restitution.................................. . . . . ..18 Marketing........................................ . . ..19 Privatization....................................... . . ..20 Rural Finance and Credit.............................. . . . ..20 PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (Loan 1876-RO) FOURTH LIVESTOCK PROJECT (Loan 1937-RO) MOLDOVA AGRICULTURAL CREDIT PROJECT (Loan 2077-RO) PREFACE This is the Performance Audit Report (PAR) of three agricultural projects in Romania, for which Loans 1876-RO, 1937-RO and 2077-RO for a total amount of US$225.0 million were approved respectively in June 1980, December 1980 and April 1982. Loan 1876-RO was closed in June 1986, six months behind schedule; the two others were closed on schedule in August and December 1986. The loans were fully disbursed with the exception of Loan 1937-RO, for which a residual amount of US$120,000 was canceled. The PAR is based on the respective Staff Appraisal and President Reports, the Loan documents, project files, discussions with Bank staff, and the Project Completion Report (PCR) issued by the former Europe, Middle East and North Africa Region in April 1990. In 1987, Romania broke relations with the Bank, and paid back in full outstanding loans. Therefore, the PCR was a desk work, based on information contained in supervision and progress reports up to December 1986, the time of the last supervision reports. An OED mission visited Romania in May-June 1994, about eight years after loan closing. The mission discussed the effectiveness of the Bank assistance with officials of the Ministry of Agriculture, Agricultural Bank (Banca Agricola, BA), and provincial authorities. The mission also interviewed project farmers and a number of managers and staff of mechanization centers, research stations and commercial companies. Their kind cooperation and valuable assistance are gratefully acknowledged. The PAR elaborates on the importance and effectiveness of the Bank's financial, technical and social contribution to the projects, and the achievements and problems encountered during project execution. It also reviews the sustainability of project benefits after the sweeping political and economic changes of the early 1990s and draws lessons from the experience for investment projects in other countries in transition from centrally-planned to market economy. Copies of the draft PAR were sent to the Borrower for comments. None were received. u iii PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (Loan 1876-RO) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as % of Appraisal Estimated Appraisal Estimate Actual Estimate Total Project Cost (US$ million) 323.9 323.9 100.0 Loan Amount 50.0 50.0 100.0 Disbursed (US$ million) 50.0 50.0 100.0 Economic Rate of Return (%) 14 Institutional Performance good good Technical Performance good good CUMULATIVE DISBURSEMENTS FY81 FY82 FY83 FY84 FY85 FY86 Appraisal Estimate (US$ million) 2.2 9.8 29.6 43.6 50.0 50.0 Actual 1.4 8.7 20.3 30.9 47.9 50.0 Actual as % of Appraisal Estimate 64 89 69 71 96 100.0 Date of Final Disbursement: June 25, 1986 Principal Repaid as of Dec. 31, 1989 (US$ mil) 50.0 PROJECT TIMETABLE Item Date Planned Actual Date Board Approval 06/19/80 Loan Agreement Date 06/30/80 Credit Effectiveness 09/30/80 09/16/80 Date Physical Components Completed 12/31/87 06/25/86 Credit Closing 12/31/85 06/30/86 STAFF INPUTS (Staff weeks) FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 TOTAL Preappraisal 23.4 23.4 - - - - - - - 46.8 Appraisal - 52.9 - - - 52.9 Negotiations - 4 3 - - - 4 3 Supervision - - 7 1 17.5 11.8 7 7 2 9 13 1 62 3 TOTAL 23.4 80.6 7.1 17.5 11.8 7.7 2.9 13.1 2.2 166.2 iv MISSION DATA Mission Dates Persons Mandays in Field SpecializationsC StatuS/d Trend' Identification 10/78 4 84 " a,b,c,d - - Preparation 06/79 2 6" a,b - - Appraisal 11/79 4 8 a,b,c,d - - Supervision I 12/80 1 8 a 1 2 Supervision H 07/81 3 16 c,c 1 2 Supervision 111 04/82 2 18 "C a,c 2 3 Supervision IV 04/83 2 22 'o a,c 2 1 Supervision V 12/83 2 18 a,d 2 2 Supervision VI 12/84 1 5 c 2 1 Supervision VII 12/85 3 26' a,b,c 2 2 Supervision VIII 11/86 2 24 c a,b 2 2 OTHER PROJECT DATA Borrower Bank for Agriculture and Food Industry (BAFI) Guarantor The Socialist Republic of Romania Executing Agency Ministry of Agriculture and Food Industry (MAFI) Fiscal Year of Borrower January 1 to December 31 Currency Name (abbrev.) lei (lei) Currency Exchange Rate Appraisal Year Average US$ 1.00 = lei 18.0 Intervening Years Average $ 1.00 = lei 16.2 Completion Year Average $ 1.00 = lei 13.7 la Specialists: a = economist, b = engineer, c = agriculturist, d = financial analyst /b Status: 1 = problem free or minor problems, 2 = moderate problems, 3 = major problems. Trend: 1= improving, 2 = stationary, 3 = deteriorating. /c Includes mandays in field spend concurrently on supervision/preparation of other projects V PERFORMANCE AUDIT REPORT ROMANIA FOURTH LIVESTOCK PROJECT (Loan 1937-RO) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as % Appraisal Estimated of Estimate Actual Appraisal Estimate Total Project Cost (US$ million) 412.0 412.0 100.0 Loan Amount (US$ million) 80.0 80.0 100.0 Disbursed (US$ million) 80.0 80.0 100.0 Economic Rate of Return (%) 14 Institutional Performance good good Technical Performance good good CUMULATIVE DISBURSEMENTS FY82 FY83 FY84 FY85 FY86 Appraisal Estimate (US$ million) 13.3 39.2 66.7 80.8 80.0 Actual 9.9 4.4 55.5 77.0 79.9 Actual as % of Appraisal Estimate 74.0 103 83 95 100 Date of Final Disbursement: January 29, 1987 Principal Repaid as of December 31, 1989 (US$ mil.) 79.9 PROJECT TIMETABLE Item Date Planned Actual Date Board Approval 12/23/80 Loan Agreement Date 01/09/81 Credit Effectiveness 04/30/81 04/07/81 Date Physical Components Completed 12/31/85 12/31/85 Credit Closing 06/30/86 06/30/86 STAFF INPUTS (Staff weeks) FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 ToLi Preappraisal 38.4 - - - - - 38.4 Appraisal 46.8 9.6 - - - - - - 56.4 Negotiations - 3.1 - - - - - - 3.1 Supervision - 2.7 13.4 22.7 14.0 9.9 3.5 0.9 67.1 TOTAL 85.2 15.4 13.4 22.7 14.0 9.9 3.5 0.9 165.0 vi MISSION DATA Mission Dates Persons Mandays in Field Specializations" Status' Trend' Identification 05/79 2 18 '' a,b,c,d - - Preparation 12/79 2 12 a,b - - Appraisal 04/80 5 88 a,b,c,d - - Supervision I 03/81 1 5 a 1 2 Supervision II 12/81 4 32 c,d 1 3 Supervision m 04/82 1 7" c 1 1 Supervision TV 09/82 1 14" c 1 2 Supervision V 04/83 3 36 c,d 1 2 Supervision VI 10/83 2 24 b,c 1 2 Supervision 05/84 2 26 " a,c 2 2 VII Supervision 11/84 2 38 ' b,c 2 2 VM Supervision IX 12/85 2 20 1 a,c 1 1 OTHER PROJECT DATA Borrower Bank for Agriculture and Food Industry (BAFI) Guarantor The Socialist Republic of Romania Executing Agency Ministry of Agriculture and Food Industry (MAFI) Fiscal Year of Borrower January 1 to December 31 Currency Name (abbrev.) lei (lei) Currency Exchange Rate Appraisal Year Average US$ 1.00 = lei 18.0 Intervening Years Average $ 1.00 = Iei 16.4 Completion Year Average $ 1.00 = lei 15.3 /a Specialists: a = economist, b = engineer, c = agriculturist, d = financial analyst /b Status: 1 = problem free or minor problems, 2 = moderate problems, 3 = major problems; Trend: 1 = improving, 2 = stationary, 3 = deteriorating. /c Includes mandays in field spend concurrently on supervision/preparation of other projects. vii PERFORMANCE AUDIT REPORT ROMANIA MOLDOVA AGRICULTURAL CREDIT PROJECT (Loan 2077-RO) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as % of Appraisal Estimated Appraisal Estimate Actual Estimate Total Project Cost (US$ million) 290.2 290.2 100.0 Loan Amount (US$ million) 95.0 95.0 100.0 Disbursed (US$ million) 95.0 95.0 100.0 Economic Rate of Return (%) 14 Institutional Performance good good Technical Performance good good CUMULATIVE DISBURSEMENTS FY83 FY84 FY85 FY86 FY87 Appraisal Estimate (US$ million) 12.0 40.0 72.0 85.0 95.0 Actual 62.8 85.0 94.0 95.0 95.0 Actual as % of Appraisal Estimate 523 212 130 100 100 Date of Final Disbursement: July 10, 1986 Principal Repaid as of Dec. 31, 1989 (US$ million): 95.0 PROJECT TIMETABLE Item Date Planned Actual Date Board Approval 01/07/82 Loan Agreement Date 04/09/82 Credit Effectiveness 04/09/82 08/09/82 Date Physical Components Completed 06/30/86 06/30/86 Credit Closing 12/31/86 12/31/86 STAFF INPUTS (Staff weeks) FY80 FY81 FY82 FY83 FY84 FY85 FY86 FY87 TOTAL Preappraisal 12.7 13.3 6.6 - - - - - 32.6 Appraisal - 21.6 42.4 - - - - - 64.0 Negotiations - - 2.2 - - - - - 2.2 Supervision - - 0.3 13.8 8.0 1.5 1.3 4.3 29.2 TOTAL 12.7 34.9 51.5 13.8 8.0 1.5 1.3 4.3 128.0 Viii MISSION DATA Mission Dates Persons Mandays in Field Specializations' Statusd Trendd Identification 06/80 3 21" a,b,c - - Preparation 03/81 2 14" a,c - - Appraisal 06/81 6 126 a,b,c,d - - Supervision I 11/82 1 18 b 1 1 Supervision II 03/83 2 12/c b,d 1 1 Supervision 111 11/83 2 20 b,c 1 2 Supervision IV 05/84 2 38/c a,b 2 2 Supervision V 11/84 1 191C a,b 2 2 Supervision VI 11/85 2 26' b 2 2 Supervision VII 11/86 2 26 i a,b 2 1 OTHER PROJECT DATA Borrower Bank for Agriculture and Food Industry (BAFI) Guarantor The Socialist Republic of Romania Executing Agency Ministry of Agriculture and Food Industry (MAFI) Fiscal Year of Borrower January 1 to December 31 Currency Name (abbrev.) lei (lei) Currency Exchange Rate Appraisal Year Average US$ 1.00 = lei 15.0 Intervening Years Average $ 1.00 = lei 16.0 Completion Year Average $ 1.00 = lei 15.2 /a Specialists: a = economist, b = engineer, c = agriculturist, d = financial analyst /b Status: 1 = problem free or minor problems, 2 = moderate problems, 3 = major problems; Trend: 1= improving, 2 = stationary, 3 = deteriorating. /c Includes mandays in field spend concurrently on supervision/preparation of other projects. ix PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (Loan 1876-RO) FOURTH LIVESTOCK PROJECT (Loan 1937-RO) MOLDOVA AGRICULTURAL CREDIT PROJECT (Loan 2077-RO) EVALUATION SUMMARY INTRODUCTION It incorporated modem technologies such as high density planting with dwarfing rootstocks, and 1. In the early eighties, when the three controlled atmosphere to extend fruit storage loans under audit were approved, agricultural life. Its ERR was estimated at 27 percent. Two production was dominated by Agricultural hundred orchards were completed. The final Production Cooperatives and State Agricultural cost of the project was estimated by the PCR to Enterprises. Individual farms covered only 9 be the same as at appraisal. The project was percent of the agricultural area, mostly in the considered to have been successful. In contrast mountainous regions. The sector remained to appraisal design, however, most orchards undeveloped. Crop and livestock yields and were planted on marginal lands and steep slopes, labor productivity were lower than those and not equipped with irrigation; orchards on obtained in similar conditions in other countries. cooperative farms were not provided with There were serious shortages of machinery and spraying equipment. Project cold stores were inputs. State farms owned their agricultural not equipped with controlled atmosphere equipment, and Stations for Agricultural equipment or separate control devices for each Mechanization (SMAs) performed all mechanical chamber. These deficiencies resulted in lower work for cooperatives. Marketing was the yields than expected, marketing problems, poor responsibility of the Ministry for Agriculture and fruit quality, and the need to sell much of it for Food Industry (MAFI). processing at low prices. 2. For all three projects, the Bank for 4. The Fourth Livestock Project included Agriculture and Food Industry (BAFI) was the the construction or modernization of dairy executive agency. They were implemented as farms; establishment of breeding heifer farms envisaged, and closed in 1986 as scheduled. In and beef fattening units; improvement of 1987, the Government broke relations with the pastures; milk and beef processing and storage Bank and paid back all outstanding loans. The facilities; and technical services. Major PCR, a desk exercise, was not able to assess the technical innovations were to provide low cost results of the projects, but evidence available in feed mix instead of concentrate, and improved the files indicated that the projects had been cow housing and modern milking parlors. Total completed in time, within cost estimates, and project costs were estimated at US$412.0 achieved most objectives. million. The ERR was estimated at 22 percent. At closing, the project had achieved its physical 3. The Orchards Project included the goals at a cost estimated to be similar to planting of some 30,000 ha; construction of 20 appraisal. Project results would be less packing and storage facilities, and technical satisfactory than anticipated due to deficiencies assistance, at a total cost of US$323.9 million. in the application of fertilizers in pasture, the x low protein content of cattle feed, and generally trade deficit accounted for more than 50 percent poor nutritional condition of the cattle. of the overall trade deficit in 1992. But the reform process is starting to pay off, production 5. The Moldova Agricultural Credit Project has increased substantially, and there are financed farm mechanization; eight agro- favorable signs for a good 1995 crop. industries; soil erosion protection; pasture improvement; dairy farm modernization; 8. Orchards on state farms and research plantation and rehabilitation of vineyards; and station farms have generally been well credit to individual farmers. Total project costs maintained. Many orchards on cooperative were estimated at US$290.2 million and the lands, in contrast, were abandoned or uprooted ERR at 31 percent. Most investments were to restore croplands or pastures. Orchards implemented as envisaged. Farm machinery exploited individually are not adequately was mostly procured locally; its performance maintained. Better results have been obtained proved significantly lower than that of imported on orchards whose individual owners grouped equipment, mostly in terms of fuel consumption them into formal or informal associations. and grain losses. The number of subloans was Dairy farms allocated to state farms are still in more than twice appraisal estimates. However, operation but the size of their herds, which continuous shortages of fertilizer was a major never reached design capacity, has been problem. considerably reduced. Livestock facilities and equipment provided to cooperatives were PROJECT STATUS AT THE TIME OF AUDIT abandoned, and cattle redistributed among former owners and members. Cows are well 6. At the time of the audit mission, the maintained and milk production has remained cooperative farms had been abolished and their stable. Milk marketing is constrained by the lands restituted to 6.5 million previous land lack, or insufficient capacity, of collecting owners. State farms had been restructured as centers at the village level and the low official commercial companies, with 30 percent of its milk price. Private entrepreneurs have bought shares distributed to the original land owners. or rented a few of the project facilities for milk Government divestiture from agricultural production or beef fattening. Pasture lands on marketing and input supply has been limited, cooperative farms have been redistributed as and little privatization of public agencies has well. Pastures on state farms seem to be in taken place. The former price control and good condition and still maintained by subsidy system has been discontinued, except for Government. cereals and milk. High nominal interest rates and refusal of Banca Agricola (BA, the 9. Farm machinery, which proved useful Agricultural Bank, the commercially-oriented before land restitution, has now reached the end successor to BAFI) to accept preliminary land of its useful life. Privatization of machinery titles as collateral have limited farmers' access to stations has started, mostly under MEBO agricultural credit. arrangements. Project erosion control works ended up mostly in private hands; they are still 7. The disruption created by the land maintained by Government. Vineyards on restitution process, combined with shortages of former cooperative lands have generally been agricultural inputs and a sharp decrease in properly maintained and are profitable. Agro- irrigated area, resulted in a decline of industries are operating in difficult conditions; agricultural output of about 6.4 percent p.a. many of them are being privatized. Laboratory during the 1989-92 period. Export bans were equipment provided under the Livestock project placed on many products and the agricultural xi has proven useful. It is still in good condition since project completion. Project outcomes are and satisfactorily operated by research stations. thus rated as unsatisfactory but likely to be sustainable at their present level. PROJECT OUTCOME AND SUSTAINABILrry 13. Moldova Agricultural Credit Project. 10. If judged within the economic and Despite its shortcomings, the farm machinery institutional parameters of the regime then in component achieved its objectives. Agro- place, a satisfactory rating could be considered industries seem to have performed well. Most for all three projects. They would have been erosion control works are still effective. sustainable for as long as the regime sustained Vineyards have generally been successful. In itself. But after the 1989 Revolution and the summary, the project is rated as satisfactory, ensuing policy changes, both ratings seem less and likely to be sustainable. appropriate. ISSUES AND LESSONS 11. Orchards Project. The vast majority of new orchards were planted on slopes and The Bank's Contribution to the Projects marginal lands, thus doomed to low productivity and yields and high operating costs. With an 14. All three projects were characterized by overall orchard abandon rate guess estimated at very large total costs and a relatively low Bank 25 percent, yields lower than anticipated and contribution; unsuccessful technology transfer high post-harvest losses, fruit production at full and failure to support individual producers. development was estimated at audit at most about one-half of appraisal projections. 15. Financial Participation. Bank lending Farmgate prices of fruits have also been much for agriculture supported high-cost projects. In lower than anticipated. Thus, project returns per capita terms, the total cost of Romania's can be deemed to be below 10 percent and its Bank-supported agricultural program was by far outcome rated as unsatisfactory. The current the largest in the world, nine times as large as level of production, however, looks stable and the world average. This was due to four main the current outcome is deemed as sustainable. factors. First, agricultural projects appeared to Project experience demonstrates that the be successfully implemented. Second, Romania distribution of large and sophisticated orchards was an important showcase in Eastern Europe. to scores of inexperienced smallholders, each Third, they provided a convenient vehicle for owning a few rows, is likely to fail unless an local currency financing, eagerly sought by adequate technical management organization is Romania authorities. And fourth, the Bank was set up. unable to modify the capital-intensive approach preferred by Romania policy makers. In the 12. Fourth Livestock Project. Most project event, Bank support to such a large investment investments were abandoned; buildings have program resulted in a pace of investment which been destroyed or cannibalized; milking and may have exceeded the absorptive capacity of manure removal equipment is no longer used; the sector. While implementation of the three the benefits of improved pastures have not been projects under audit was not affected, four sustained. Cattle formerly owned by the concurrent irrigation projects were considerably cooperatives, however, are now better fed and delayed. better taken care of, and most likely are producing more milk and meat than before. On 16. The Bank's financial contribution to the state farms, the number of cattle and the volume projects, however, was small; on average, only of milk and meat production have decreased 21 percent of the projects' total costs. Such a xii low contribution was due to the maximum farm 'machinery was awarded to Romanian possible size of the Bank's lending program for manufacturers. Its quality and performance are Romania; the priority which the Bank ascribed now criticized by national authorities and local to agricultural lending; and the resistance of the officials on grounds of poor quality, limited Government to borrow to import goods and durability, high fuel consumption, and high services. grain losses. Given Romania's distorted pricing structure and its centralized decision-making, 17. Technology Transfer. The projects only the letter of the Bank's procurement offered an important opportunity to introduce policies was observed - not its spirit. Similar new technology. Both the Bank and the deficiencies were noted in the PAR on the four Borrower are to blame for the failure to actually concurrent irrigation projects mentioned above. effect such technological transfer. Therefore, the audit raises the issue of whether inadequate Bank review of the technical 18. The Orchards project had important specifications in the procurement process technological goals. But the Government's jeopardized the possibilities of technological pursuit of physical targets at the expense of transfer under both the Orchards and Moldova quality, combined with poor Bank supervision, projects. resulted in serious shortcomings. First, a large number of orchards were established on steep 21. Support to Individual Producers. At the slopes and marginal lands without irrigation; and time of audit, there was no evidence that the peach and apricot orchards, which required "private sector" orientation of the Livestock and irrigation, were not planted at all. Second, Moldova projects had ever been accepted by the cooperative farms were not provided with farm Government, let alone implemented. "Individual machinery and spraying equipment. Third, the producers" receiving sub-loans under the cold storage facilities were not equipped with projects turned out to have been mostly controlled-atmosphere equipment or separate cooperative members. chamber control devices. And fourth, all cold stores were located on state farms and research The Borrower's Contribution to the Projects stations, and for their exclusive use. These developments remained unnoticed by the Bank 22. The Borrower must be credited with and were not mentioned in the supervision effective and timely project implementation. But reports. the audit revealed a different view of the factors behind such performance. All details of every 19. The Livestock project introduced project had to be personally reviewed and improved design of dairy units, modern milking approved by the President. Once approved, equipment, changes in animal nutrition, and there was no problem with procrastination or better use of pastures. The training and research non-compliance. But this method for decision- components were well designed and proved making is to blame for most of the projects' useful. The advice provided by the Bank was worst design features as well. It was the much appreciated by the Borrower. But the President who forbade planting high-density change from large-scale dairy farms to family orchards on flat, irrigable lands, and forbade units made many such innovations technically culling to increase the size of the national herd. irrelevant. It was also his decision to utilize foreign exchange provided by the loans to finance 20. The Moldova Agricultural Credit project investments in other, more "strategic" sectors missed the opportunity to bring in technical and, later on, to prematurely repay the country's innovations. Almost every ICB contract for foreign debt. Therefore, project managers were xiii under explicit instructions not to incur expenses Project Sustainability after the Revolution. in foreign exchange. Most loan proceeds were Four Transitional Issues disbursed against local expenditure, thereby generating "free" foreign exchange for the 25. Land Restitution. In 1990, the Government. Romanian Government decided to restore land to its former owners. But most of these are old 23. Project experience demonstrates the people, who left the region after land limits of Bank's influence in a centrally-planned expropriation in the late forties, resided in urban country where the decision-making process was areas, lost experience in agriculture, and lack concentrated basically on a single individual. financial resources; only some 40 percent of Compounded by the Bank's limited leverage, the them are active farmers. Further, each family projects' designs were distorted and supervision was eager to recover the same parcels of land had a negligible impact. owned prior to expropriation, resulting in a high degree of fragmentation. In contrast, China's The Fourth Livestock Project and the redistribution program (1979-83) distributed Economic Analysis of Projects in the Bank responsibility for farming collective lands to former cooperative members and state farm 24. The appraisal process which eventually employees, who quickly adopted a common led to the Fourth Livestock project was to play cropping pattern. This contributed to a a major role in reinstating Bank policy on considerable production increase during the economic project analysis. In 1978, the team years following land distribution. The question appraising a possible fourth livestock project can thus be raised as to whether a dual system, (Dairy Farms) realized that official domestic offering cash compensation as an alternative to milk prices bore no resemblance to economic land restitution, as in Albania, would be a more values, and decided to apply border price effective solution. equivalent, but as a result the project proposals submitted by Government would have a negative 26. Marketing. Romania's experience economic rate of return. The Romanians did not illustrates the difficulty to change and modernize accept the milk valuation criterion and strongly the marketing systems for horticultural and dairy protested to the Bank. Bank middle managers products, and to adapt them to a different farm were concerned as well, as they were committed structure. In spite of a legal framework favoring to deliver the loan. But Central Projects Staff private initiative, the establishment of private (CPS) and, later on, the Loan Committee marketing channels has been slow. Besides confirmed the appraisal team's position as the facing a shortage of working capital and difficult only proper one. The project proposal was access to credit, the markets for fruits and milk aborted. The new project prepared in its stead are characterized by high demand by consumers eventually became the project under audit while heavy losses or inappropriate use of (Fourth Livestock: Cattle). Having stopped the products at the farm and marketing levels occur. Dairy project, Central Project Staff were now Often, the marketing infrastructure is not suited able to stop other projects every year on similar to the new land tenancy structure. More small grounds without the same kind of hassle. And milk collecting centers at the village level are middle managers were put on notice that lending needed, and silos and grain stores are no longer programs and pressure to lend would not be suited to the large number of small grain accepted as a substitute for proper economic traders. For dairy products, low official price analysis. has not induced private traders to develop the fresh milk collection and distribution market. xiv 27. The lesson is that establishing a legal of financial resources has been a constraint for framework is not a sufficient step for private private entrepreneurs in terms of purchasing or marketing channels to develop. There is a need renting project-built facilities (e.g. stables) or for adequate financial resources and for specific investments (e.g. orchards), as well as fruit training programs on management skills, market traders and milk collectors at the village level. knowledge, and handling, transporting and marketing perishable products. Joint ventures 29. Rural Finance and Credit. The banks with experienced partners appear to be another are very security-conscious. They do not accept route for modernizing both the fruit and milk land titles or storage warrants as collateral, and processing industries, are reluctant to lend to borrowers who cannot guarantee that their products will be sold. 28. Privatization. Privatization has been Limited access to credit has been one of the slow due to, inter alia, an asymmetric and main constraints to project success and burdensome taxation system; an unequitable sustainability. But most existing banks are not allocation of responsibility for repaying past interested in lending to the agricultural sector debt; and lack of financial resources for and use cumbersome procedures to discourage investment and working capital. Individuals or potential agricultural borrowers. Low-interest firms acquiring divested state companies or sub-loans from external sources (World Bank- establishing new companies benefit from tax supported credit programs, among others) are exemptions and incentives. But commercial or particularly attractive for potential borrowers, "inter-coop" associations formed to continue but only a few have been able to meet the strict operating former cooperative assets must pay all criteria of these special credit programs. taxes and assume the remaining debt. And lack PERFORMANCE AUDIT REPORT ROMANIA ORCHARDS PROJECT (Loan 1876-RO) FOURTH LIVESTOCK PROJECT (Loan 1937-RO) MOLDOVA AGRICULTURAL CREDIT PROJECT (Loan 2077-RO) 1. BACKGROUND 1.1 At the end of 1972, Romania became the second Eastern European country (after Yugoslavia) to join the World Bank. By that time, its national income was growing at an annual rate of about 9 percent p.a., and investments absorbed 35 percent of it. Over the previous two decades, Romania had pursued a strategy to industrialize a primarily agrarian economy. At the end of 1970, industry had become the leading sector of the economy, accounting for about 50 percent of GDP and 34 percent of employment. 1.2 In the early eighties, when the three loans under audit were approved by the Board, agriculture still played an important role, accounting for about 14 percent of GDP. Despite rapid industrialization, more than half of the population still lived in the rural areas and one third of employment was provided by agriculture. About 15 million hectares, or 63 percent of Romania's land area, was used for agriculture. Crop production, mostly cereals and oilseeds, accounted for 56 percent of total agricultural product, and livestock accounted for 44 percent. 1.3 Following land collectivization after the second world war, agricultural production was dominated by Agricultural Production Cooperatives and State Agricultural Enterprises. In 1980, the area under cooperative farms was 9.1 million hectares, or 61 percent of the total agricultural land. State farms farmed 2.0 million hectares, equivalent to 14 percent of agricultural land. The average size of cooperatives was 2,080 ha, with 530 members, mostly former land owners. The average size of state farms was 5,100 ha, with a labor force of 550. Other state agricultural units, including research stations, covered 2.5 million ha, or 16 percent of agricultural land. Individual farms covered only 9 percent of the total agricultural area, and mostly in the most mountainous regions. 1.4 While state farms owned their agricultural equipment, Stations for Agricultural Mechanization (SMAs) performed all mechanical work for cooperatives, as well as some work for private farmers. Marketing of agricultural inputs and outputs was under the responsibility of the Ministry for Agriculture and Food Industry (MAFI). Government agencies had a monopoly on foreign trade and agro-processing. 1.5 Although the performance of Romanian agriculture had been impressive over the previous three decades (average annual growth of 4.5 percent) the sector remained relatively undeveloped in 1980. Crop and livestock yields were significantly lower than those obtained in similar conditions in other countries, and labor productivity was low. Serious shortages of machinery and inputs were experienced, aggravated by foreign exchange policies, as the Government sometimes exported fertilizers and restricted input imports as a result of foreign exchange shortages. In addition, Government policy had given greater emphasis to developing the state farms rather than cooperatives. In 1979, the level of investment per hectare was more than five times larger in the former than in the latter. 2 1.6 Concerned about such problems, the Government declared agriculture a priority sector for the 1981-85 Plan period, during which the projects being audited were designed and implemented. The Plan target was to increase agricultural production through the use of more efficient technology, large investments in irrigation, increased use of agricultural inputs, and changes in financial flows between enterprises and state budget to allow enterprises to retain a greater part of their profits for distribution between investment, working capital, and profit-sharing. 2. THE PROJECTS 2.1 Bank assistance to the agricultural sector from 1975 to 1986 included sixteen operations totalling US$1.07 billion. The three projects under this audit were the twelfth, thirteenth, and sixteenth of the series. They were implemented and disbursed as envisaged, and closed in 1986 as scheduled. In 1987, the Government broke relations with the Bank and paid back all outstanding loans in full. No information was thus available after the November 1986 supervision missions as to the projects' economic and social impacts. Therefore, a single PCR for the three projects was prepared by the Bank in April 1990 without a field visit. The PCR was not able to assess the results of the projects, but evidence available in the files indicated that the projects had been completed in time and within cost estimates, and had achieved most of their objectives. 2.2 The projects were to expand production capacity through funding of physical assets and equipment, and to address the prevailing problems of production instability and low productivity through the introduction of new technologies and strengthening of research institutions. The Bank's emphasis on the private sector and the role it could play in achieving food production objectives led to the introduction of subloans to individual producers, especially in the Livestock and Agricultural Credit projects. THE ORCHARDS PROJECT 2.3 In the late seventies, the level of fruit consumption in Romania was low compared with other countries at a similar stage of development. This was attributed by the Government to physical factors, i.e., short supply and inadequate storage facilities. 2.4 Project components included: the planting of some 30,000 ha in 120 orchards of 250 ha each, all on collective farms (70 cooperative farms and 50 state farms) and most of then with irrigation; the construction of 20 packing and storage facilities, to be located close to the main fruit producing areas; and technical assistance. Total project costs were estimated at US$323.9 million (of which 89 percent for the new plantings) to be spent over five years. Project design incorporated modern technologies such as high density planting systems with dwarfing rootstocks, controlled atmosphere to extend fruit storage life, and scientific equipment for research laboratories. The Bank for Agriculture and Food Industry (BAFI), the executive agency, was to lend 70 percent of investment costs to cooperatives and 100 percent to state farms, to be repaid in 10 years after a 10-year grace period. Incremental production at full development was expected to reach 528,000 tons, the market for which was assumed to exist. The project's ERR was estimated at 27 percent. 2.5 Project plantings started soon after loan effectiveness and 206 orchards (equally divided between cooperatives and state farms) were completed by Spring 1984, about six months behind 3 schedule. In contrast to appraisal expectations, most orchards were planted on marginal lands and steep slopes, and were not equipped with irrigation (details in paras. 6.7-6.9). Moreover, orchards on cooperative farms were not provided with pesticide spraying equipment They had no access to project-built cold stores and depend on conventional (and usually outmoded) storage facilities. These deficiencies have resulted in lower yields (about 40 percent) than estimated at appraisal, poor fruit quality, marketing problems, and the need to sell a significant part of the fruit production for processing (most generally for distillation) at a low price. 2.6 Packing and storage facilities were designed by the Government's Institute for Fruit Marketing, the only agency in Romania then authorized to design agro-industrial projects. Delays were experienced in construction due to budgetary constraints and changes in design. Contrary to appraisal specifications, project cold stores were equipped neither with controlled atmosphere nor with separate automatic temperature control devices for each chamber, which thus cannot be operated individually, resulting in high operation costs and wastage of energy. At loan closing, in June 1986, 8 out of 20 facilities had been completed and were operational. Works were progressing satisfactorily on the remaining 12, which were expected to be completed by June 1987. All specialized equipment (except those for controlled atmosphere) had been purchased, and training provided by the suppliers. The final cost of the project was estimated by the PCR at US$323.9 million, the same as at appraisal. The project was considered to have been successful at completion. THE Fouirm LIVEsTOCK PROJECT 2.7 Romania's climate and resource base are well suited to cattle, with a long history of animal husbandry. Although past growth rates had been satisfactory, in the late seventies cattle productivity was still low. The Government was aware of the need for improved technology, particularly in the area of cow housing and milking. Project objectives were to increase the production and improve processing of beef and milk, and generate foreign exchange earnings. 2.8 Project components included the construction of 92 new dairy farms of 922 cows each; modernization of 32 dairy farms; establishment of 17 breeding heifer farms and 19 beef fattening units; improvement of pastures (109,000 ha); milk and beef processing and storage facilities; support to producers, technical services and research. Subloans were to be made by BAFI to state farms and cooperatives. About 80 percent of the dairy investments and 44 percent of those on beef fattening were to be directed to cooperatives; state farms would implement the remaining 20 percent and 56 percent. About 90 percent of pasture improvement would be for cooperative members and individual farmers. The major technical innovations of the project were to provide high quality, low cost feed mix as a substitute for the expensive concentrate-based system; and to introduce improved cow housing and modern milking parlors. Total project costs were estimated at US$412.0 million over five years. The ERR was estimated at 22 percent. 2.9 Soon after start of implementation, emphasis moved from new construction to modernization of existing dairy farms, and from heifer rearing farms to cattle fattening units. The pasture improvement program was significantly expanded. The proposed technical innovations were successfully tested and well accepted by cooperative and state farms. Milking equipment was procured as planned and the performance of agro-industries was satisfactory. Implementation of the research component was satisfactory. 4 2.10 At loan closing, the project had achieved its physical goals at a cost estimated to be similar to appraisal. Its economic benefits could not be re-estimated. There were indications, however, that project results would be less satisfactory than anticipated due to deficiencies in the application of fertilizers in pasture, the low protein content of cattle feed, and generally poor nutritional condition of the cattle. THE MOLDOVA AGRICULTURAL CREDIT PROJECT 2.11 The project area covered the greatest part of the Moldova region, located in the northeastern part of Romania and representing one-seventh of the country's total land and population. This was one of Romania's lesser developed regions in spite a significant potential for agricultural production. The project objective was to address the main constraints affecting agriculture in the region, particularly the low use of fertilizers, lack of farm mechanization, and soil erosion, through the provision of agricultural credit to cooperatives, state farms and individual farmers. 2.12 Project components consisted of farm mechanization (mostly tractors and harvesters) to be operated by 97 SMAs; the construction of eight agro-industries; soil erosion protection for 142,000 ha; pasture improvement on 44,000 ha; dairy farm modernization; plantation and rehabilitation of vineyards; and credit to individual farmers mostly for the purchase of heifers and sheep. Total project costs over five years were estimated at US$290.2 million, of which 47 percent were for farm mechanization, 18 percent for agro-industries, 15 percent for soil erosion control, and 20 percent for the other project components. The ERR was estimated at 31 percent. 2.13 Project implementation started quickly after effectiveness; the credit component was fully disbursed 15 months ahead of schedule. Most of the project investments were implemented or procured as planned at appraisal, except for dairy farm modernization, being financed under the Fourth Livestock Project, and the sugar mill, found not viable as a result of the deterioration of the world sugar price. Loan funds for these components were reallocated to mechanization, the most successful project component. Farm machinery was mostly procured locally and only a few combine harvesters were procured from Western Europe. Even though the country was a net exporter of tractors at the time of project appraisal, the performance of locally-made machinery proved significantly lower than that of imported equipment, mostly in terms of fuel consumption for tractors and harvesters, and grain losses for the latter.- More modernization of vineyards and less new planting than anticipated were noted. The number of subloans (23,678) was more than twice appraisal estimates but their amount was correspondingly smaller. 2.14 Project cost at completion was estimated to have been as anticipated in US$ terms, and slightly higher (106 percent) in lei. Although project benefits could not be fully estimated, the PCR concluded that the most evident sign of project success was the positive impact of mechanization on agricultural production, due to reduction of grain losses, more efficient and timely land preparation, better plant protection, and improved silage operations. However, continuous shortages of fertilizers at the national level was considered a major problem, adversely affecting the pasture improvement component. 1. Figures of 10 percent of grain lost for domestic harvesters, compared to 2.5 percent for imported ones, were mentioned to the credit mission. Even larger differences in fuel consumption per ton of grain harvested were mentioned as well. 5 3. DEVELOPMENTS SINCE LOAN CLOSING 3.1 The Government which took office after the 1990 elections introduced sweeping changes in the economic system. Regarding agriculture, it decided to restitute to their former owners about 15 million ha of agricultural land; to divest Government from most agricultural services and privatize agricultural input and output marketing; and to liberalize agricultural prices. 3.2 At the time of the audit mission, the cooperative production system had been abolished and lands in cooperative farms restituted to 6.5 million previous land owners (or their heirs) and former cooperative members. The maximum size of individual farms has been set at 10 ha, but the actual average land holding is about 1.7 ha, often divided into 4 or 5 parcels. State farms, which were excluded from the land restitution program, had been restructured as commercial companies, owning 1,500 to 2,000 ha each with 70 percent of its shares being retained by the state and 30 percent distributed to the original land owners. 3.3 About 57 percent of the land owners are above the age of 60, and 43 percent live in urban areas. Most of them received a temporary land certificate confirming the extent of their land- holdings, but only few have received the formal land title, defining the location of the land and parcels. Various kinds of associations for joint agricultural production or management have emerged spontaneously among land owners: Agricultural Societies, with legal recognition comprising on average 200 members and about 500 ha; and Farmer Associations, which have no formal status and comprise on average about 63 members and 156 ha each. 3.4 Government divestiture from agricultural marketing and input supply has been limited, and little privatization of public agencies has taken place. Although some private traders have emerged, state firms continue to dominate agricultural product and input markets. Their performance and financial position have continued to deteriorate. The provision of mechanization services for private farms has remained mainly in the hands of SMAs (renamed AGROMECs), many of which are being privatized under a management-employee buy-out (MEBO) system. Emerging private farm machinery firms now compete with AGROMECs. 3.5 The former price control and subsidy system has been discontinued, except for cereals and milk. High inflation interferes with marketing and private producers concentrate on self-sufficiency, or tend to store rather than sell their products. High nominal interest rates (more than 100 percent p.a.) and refusal of Banca Agricola (BA, the Agricultural Bank, a commercially-oriented successor to BAFI) to accept preliminary land titles as collateral have limited farmers' access to agricultural credit, resulting in low investment and lack of working capital. 3.6 The disruption created by the land restitution process, combined with shortages of agricultural inputs and a sharp decrease in irrigated area, resulted in a decline of agricultural output of about 6.4 percent p.a. during the 1989-92 period. Export bans were placed on many agricultural products and the agricultural trade deficit accounted for more than 50 percent of the overall balance of trade deficit in 1992. 3.7 But the reform process is starting to pay off. Agricultural production increased substantially in 1994 and a senior official in the Ministry of Agriculture stated that there are favorable signs for 6 a good 1995 crop. As a result, Romania would be producing food surpluses again. Agriculture is expected to contribute 25 percent of GDP this year, compared with 22 percent in 1993. 4. PROJECT STATUS AT THE TIME OF AUDIT 4.1 Political and economic changes since project completion have had a considerable impact on most project components. Changes in cooperatives' land tenure status affected about half of the fruit plantings supported by the Orchards project and most of the dairy farms and pastures financed by the Livestock and Moldova projects. AGROMECs and small agro-industries supported by the Moldova project are in the process of privatization. State farms owning project-funded orchards, vineyards and cold stores are being transformed into commercial companies. Problems of marketing, agricultural credit and shortages of capital have jeopardized the profitability of project investments. Production and consumption of fruits declined by about 17 percent and 26 percent, respectively, between 1989 and 1992. A more detailed description and analysis per type of investment follows. ORCHARDS 4.2 Many orchards planted on cooperative lands restituted to individual farmers were abandoned or uprooted in order to restore croplands or pastures; low fruit prices and profitability (para. 5.5) also contributed. Although the abandon rate of cooperative orchards is not available at the national level, it is estimated by local MOA authorities in the Suceava Province at about 50 percent. Surviving orchards are being exploited under either of two modalities: (i) individually; and (ii) in voluntary associations. Some have been rented out to private entrepreneurs. Generally, orchards exploited by their individual owners are not adequately maintained for lack of expertise and working capital; pruning, spraying and fertilization have often been neglected or discontinued. In contrast, orchards whose individual owners grouped them into formal or informal associations (para. 3.3) under the management of agronomists or fruit specialists have fared better. Some of these associations have been profitable and able to provide employment and pay dividends to their members during the past years.' 4.3 Orchards on state farms and research station farms, whose managers and technicians have been kept in their previous positions, have generally been adequately maintained. In some cases, participation of former land owners in the capital of the new commercial companies has been successful, resulting in distribution of dividends and employment opportunities within the company. In other cases, former land owners illegally took over their land and occasionally uprooted the new plantings. The rate of abandon or uprooting of orchards planted on state farms is estimated at 10 percent in the Suceava Province. 2. In 1993, a successful and profitable association has been able to pay the equivalent of US$ 220 to its members. Another orchard association distributed the equivalent of 700 kg of apples to each. In both cases, members have the possibility to work as laborers or technicians in the orchards as well. 7 COLD STORES 4.4 Research stations and state farms which were allocated project-funded cold stores are still operating them but only for their own fruit production. Shortages of working capital prevent other public and private cold store enterprises to buy and store fruits from producers, resulting in both unused cold store capacity in most provinces and post-harvest losses on former cooperative orchards. DAIRY FARMS, BEEF FATTENING AND HEIFER RAISING CENTERS 4.5 Livestock facilities and equipment provided by the Livestock and Moldova projects to cooperative farms were abandoned. Cattle were redistributed among former land owners and members, each family now keeping two or three head. Cows are well maintained and milk production has remained relatively stable. Milk marketing is constrained by the lack, or insufficient capacity, of collecting centers at the village level and the low official price of fresh milk (para. 3.5). As a result, a substantial part of milk production is now used for feeding calves and piglets. 4.6 Private entrepreneurs have bought or rented a few of the project facilities for milk production or beef fattening. Such a trend is constrained, however, by the lack of capital of potential buyers and high interest rate of loans. 4.7 Dairy farms allocated to state farms are still in operation but the size of their herds, which almost never reached design capacity, has been considerably reduced by the need to repay existing debts. Former land owners have become shareholders of the newly established commercial companies. In some cases, former land owners illegally occupied a substantial fraction of the state dairy farm lands and established crops on them. IMPROVED PASTURES 4.8 Pasture lands on cooperative farms have been restituted as well. Most of them would seem to be used by villagers as traditional communal pastures, resulting in overgrazing, poor maintenance, and lack of fertilization. Project pastures on state farms (10 percent of the total) seem to be in good condition and still adequately maintained by provincial Government agencies funded from the budget. However, intrusion of private livestock in state pastures is said to be frequent and difficult to control. EROSION CONTROL WORKS 4.9 Project erosion control works, consisting mainly of terraces and contour lines, ended up mostly (85 percent) in private hands since land restitution. Since they are considered as works of national interest, they are still maintained by provincial Government agencies under specific budget allocations. Some of them, however, were destroyed when cooperative land was restituted to former land owners. Moreover, the fact that land was generally redistributed (and is being cultivated) as narrow strips across the slopes (perpendicular to road alignments, providing each farmer with a sampler of all land qualities in the former cooperative farm) and not along the contour lines has significantly increased the risk of soil erosion since land restitution. 8 VINEYARDS 4.10 Vineyards on former cooperative lands have generally been better maintained than orchards, mostly because they remained financially attractive for producers. A substantial part of the project on-lending for vineyards went to large research stations, and to cooperative farms, now restituted to former land owners and cooperative members (para. 3.2). Although these companies face serious financial difficulties, they are still profitable and many have been able to pay dividends to their shareholders.' AGRO-INDUSTRIES 4.11 Agro-industries, including flour mills, grain silos, slaughterhouses, and milk processing plants, are operating in difficult conditions. Many of them are being privatized, sometimes with the participation of foreign investors. ROMCEREAL, the large state-owned grain marketing company which benefitted from subloans of the Moldova project, is losing its grain marketing monopoly as private grain traders have emerged since the Revolution. FARM MACHINERY 4.12 Farm machinery, including 6,500 tractors and 1,500 combine harvesters, financed under the Moldova project has been used by AGROMECs for preparation of cooperatives' land and harvesting their crops. It was useful before land redistribution and has now reached the end of its useful life. But local machinery, designed for large farms, is not well suited to the small plots resulting from the land restoration exercise, performs poorly, and lacks durability (para. 2.13). 4.13 Privatization of AGROMECs has started in the Moldova region, mostly under MEBO arrangements (para. 3.4). The future of the new companies is hampered by old and outmoded equipment, competition from private machinery operators (now owning about 40 percent of the country's farm machinery), and charges and taxes higher than those set for their competitors (para. 6.37). LABORATORY EQUIPMENT 4.14 Laboratory equipment provided under the Fourth Livestock project has proven useful. It is still in good condition and satisfactorily operated by research stations. 5. PROJECT RESULTS 5.1 The projects were approved in support of a centrally-planned economic regime. Scale of investments was to be large; inputs were assured, as was marketing of outputs; funding was never challenged; and costs and profitability were never an issue. Domestic technology was paramount, and all important (and many not so important) technical and managerial decisions were issued from Bucharest. With some exceptions, the three projects largely delivered what they were meant to 3. Up to US$80 equivalent for each shareholder of a large company owning 2,000 ha in the Foceani province. 9 deliver, i.e., their efficacy is not in doubt. Some signs of the projects' weaknesses, however, had already been starting to appear; as the economy was drawing towards bankruptcy and funds for purchasing subsidized inputs (animal feed, gasoline, pesticides) were no longer available, projects (and particularly the Livestock one) began to falter. Nonetheless, if judged within the economic and institutional parameters of the regime then in place, although lacking in quantitative data, a satisfactory rating might have been considered for all three projects. They would have been sustainable for as long as the regime sustained itself. 5.2 Up to December 1989, that is. As described above, after the Revolution cooperative farms were repossessed by their previous owners. Land was split in accordance with the area each previously owned; project-funded investments, if mobile or movable, were taken away; if not, abandoned or cannibalized. State farms were reorganized as commercial companies, wherein the State retained 70 percent of their equity and former land owners shared in the other 30 percent (again in proportion to their erstwhile land holdings). Input supply and output marketing are being privatized, no longer assured to the individual farm. Costs and profitability, as well as commercial skills to procure inputs, obtain capital (labor seldom being a problem) and sell outputs, now matter. Competition is emerging. Thus, a reassessment of the projects' ratings is required since the relevance of the original project objectives and design is gone. 5.3 The following paragraphs assess each project's outcome and sustainability on the basis of evidence collected by the audit mission, and of comparisons with projections and sensitivity analyses made at their appraisals. It must be noted, however, that most of the data which would have been required to produce robust estimates is lacking as cooperatives, the main beneficiaries of the projects, have been dismantled and no records have been kept since land restitution. ORCHARDS PROJECT 5.4 New plantings, of which more than half were on cooperative farms, represented about 89 percent of total project costs. The vast majority were planted on slopes and marginal lands, being thus doomed to low productivity and yields, and high operating costs. The SAR calculated that the project ERR, estimated then at 27 percent, would fall to 9 percent if overall benefits would decrease by 44 percent. With an overall orchard abandon rate guessestimated at 25 percent, yields lower than anticipated and high post-harvest losses, the project fruit production at full development (1992) could be estimated at audit at most about one-half of appraisal projections. 5.5 Farmgate prices of fruits have also been much lower than anticipated due to lack of exports (expected to account for 15 percent of total output) and the need to sell a significant portion of the production for processing or distillation. With such combination of low production and low prices, the project returns can be deemed to be below 10 percent, and its outcome rated as unsatisfactory. The current level of production, however, looks stable; input and output marketing is likely to improve as privatization progresses; and prices should firm up as marketing improves, national incomes grow, and economic recovery in Western Europe leads to enhanced export opportunities. The current level of benefits, albeit unsatisfactory, is deemed as sustainable. 5.6 Project experience demonstrates that the distribution of large and sophisticated orchards to scores of inexperienced smallholders, each owning a few rows, will fail in the absence of an organization for sound technical management. Successful pest and disease control, in particular, 10 requires ownership of sprayers as well as discipline and coordination among farmers, conditions which have not been met in most redistributed orchards. In this regard, the success of associations formed among owners of fragments of erstwhile orchards, managed by competent agronomists or fruit specialists, has demonstrated that orchards can remain a major source of employment and income for land owners after land restitution. FOURTH LIVESTOCK PROJECT 5.7 More than half of project expenditures went for the construction and modernization of dairy farms and procurement of milking equipment. The purchase of cattle accounted only for 41 percent of investments in new dairy farms and for 10 percent in farm modernization costs. About 11 percent of total project costs were for pasture improvement. Cooperative farms were the main project beneficiaries with 80 percent of the project dairy farms and 90 percent of improved pastures. Project ERR was estimated at 22 percent at appraisal, less sensitive to changes in capital costs than to variations in revenues. 5.8 With the dismantling of cooperatives, most project investments have been abandoned. Buildings have been destroyed or cannibalized; milking and manure removal equipment is no longer used; the benefits of improved pastures have not been sustained. Cattle formerly owned by the cooperatives, however, are now better fed and better taken care of, and most likely are producing more milk and meat than before, but such production increase is not attributable to the project. On state farms, the number of cattle and the volume of milk and meat production have decreased since project completion. The project outcome has undoubtedly been unsatisfactory. 5.9 The transfer of cattle ownership did not result in productivity losses such as those experienced in the orchards project. Quite the opposite. The situation of the cattle herds in the last years of the old regime was very poor: poorly fed, sick with tuberculosis and mastitis, high mortality rates and very low yields, compounded by an instruction from the Presidency to retain all animals in order to "enlarge" the national herd. Although most buildings and equipment were lost or remain unused, milk production has not declined after restitution of cows to individual farmers, who have demonstrated their ability to take care of their animals. Health control by public veterinaries is available, and private veterinaries are starting to offer their services. The main constraints affecting livestock activities are low official prices, poor marketing, and an inappropriate milk collection system, no longer suited to the new livestock ownership. Project outcomes are thus rated as sustainable at their present level. MOLDOvA AGRICULTURAL CREDIT PROJECT 5.10 At project completion, farm machinery accounted for 65 percent of total project costs (47 percent estimated at appraisal), followed by erosion control works (about 12 percent) and agro- industries (about 10 percent). More vineyards than expected were planted. At appraisal, the overall project ERR was estimated at 31 percent (32 percent for farm machinery, 13 percent for agro- industries and 16 percent for erosion control works). 5.11 Despite its shortcomings, the farm machinery component achieved its objectives of land preparation and crop harvesting. There are indications that project agro-industries have performed reasonably well. Most erosion control works are still effective. Vineyards have generally been 11 successful. In summary, the Moldova project has achieved most of its objectives and its ERR is likely to be not significantly lower than its appraisal estimates. It is thus rated as satisfactory, and likely to be sustainable. REPAYMENT OF SUBLOANS UNDER THE THREE PROJECTS 5.12 The three projects were administered by the Bank for Agriculture and Food Industry (BAFI) under long-term and low interest rate lending policies and procedures. For orchards, BAFI on-lent to cooperative farms up to 70 percent of investment costs at terms of up to 20 years including 10 years of grace. For state farms, it was up to 100 percent of investment costs, with 17 years and 7 years of grace. Lending terms and conditions were about the same for investments in livestock and farm machinery, with shorter repayment periods. 5.13 BAFI, and its commercially-oriented successor Banca Agricola (BA), have recovered most of the project subloans, even though the sense of the word "recover" in a centrally planned economy is not necessarily the same as in a market economy. Loans made before 1984 and not repaid by 1985 were included in BAFI's balance sheet as non-performing loans; the same was done again in 1990. Ninety percent of the amounts involved in such loans were repaid by the Government, and the remaining 10 percent out of BAFI's Risk Fund. Each type of sub-borrower displayed a different repayment behavior. State farms, on average, managed to repay more than 50 percent of their sub- loans. SMAs and AGROMECs are profitable companies and have repaid their sub-loans before and after the Revolution without problems. Individuals fully repaid their sub-loans. In contrast, a portion of the sub-loans outstanding for cooperatives was canceled by the Government in 1989. The remaining debt is being treated in two different ways: if commercial or "inter-coop" associations have been formed to continue operating the former cooperative assets, they assume the remaining debt and repay BA out of their income. If nobody took over the former cooperative assets, these were repossessed by BA and will be sold through bidding; proceeds would accrue to BA's Risk Fund account. In short, neither BAFI nor BA suffered financially from operating the projects' on-lending activities. 6. ISSUES AND LESSONS 6.1 Although project results have been considerably affected by political and economic changes since project completion, some issues related to the respective contribution, support and performance of the Bank and the Borrower are worth reviewing. In addition, the project that would have been Fourth Livestock project played a unique role in reaffirming the Bank's reliance on sound economic project analysis as a necessary step in project processing, which is worth bringing to the fore. Finally, several lessons can be drawn about the main factors affecting project sustainability for other countries in transition from centrally-planned to market economies. THE BANK'S CONTRIBUTION TO THE PROJECTS 6.2 All three projects were characterized by very large total costs and a relatively low Bank contribution; unsuccessful technology transfer; and failure to support individual producers outside the socialist sector. 12 Financial Participation 6.3 Bank lending to the Romanian agricultural sector in the late seventies and early eighties supported a large number of high total cost investment projects. Over three years, from March 1979 to January 1982, nine projects with a total cost of US$3.35 billion were approved by the Board. In per capita terms, the total cost of Romania's Bank-supported investment program (at US$142 per capita) was by far the largest in the world (sole exception: Western Samoa, with 150,000 inhabitants and US$200 per capita), and nine times as large as the world average (US$16). When compared with other medium-sized countries, the Bank supported agricultural investments per capita almost three times larger than the average of such well-known Bank clients as Malaysia, Mexico, Morocco, Portugal, Turkey and Yugoslavia (average for the six: US$53 per capita). (See Graph 1). This ambitious program contrasted with the country's slowing economy and the need, recognized at the time by both the Government and the Bank, for lower growth and investment rates and better allocation of resources between consumption and investment. 6.4 The Bank's hyperactive lending to Romanian agriculture was due to four main factors. First of all, there were the twin perceptions that agricultural development was very important, and that agricultural projects, measured in input terms, appeared to be successfully implemented and Bank loans quickly disbursed. Second, in the early eighties Romania was for the Bank an important show case in Eastern Europe, an opportunity to influence development policy in a socialist country which had adopted an independent foreign policy stance. Third, they provided a convenient vehicle for local currency financing, eagerly sought by Romania authorities. And fourth, the Government had adopted a capital-intensive approach to agricultural development. Bank support was financially useful as it provided Romania with foreign exchange. This convergence of interests resulted in a lending program and a pace of investment which exceeded the absorptive capacity of the sector. While implementation of the three projects under audit was not affected, implementation of four concurrent irrigation projects was considerably delayed and then suspended due to budgetary constraints.4 6.5 Another factor common to all agricultural projects in Romania was the low Bank financial contribution to their financial plan. The nine loans mentioned above amounted to US$700 million, only 21 percent of the projects' total costs. Such a low relative contribution was due to the limitations imposed by the maximum possible size the Bank's aggregate lending program for Romania; by the priority which the Bank ascribed to agricultural lending; by the small foreign exchange component of the projects; and by the resistance of the Government to borrow to import equipment or send staff abroad for training (more on this in the next Section). At the time of their Board approval, significant parts of the financing plans of the Orchards and Fourth Livestock projects were still to be filled by unidentified cofinanciers. As no cofinancier was found, almost 80 percent of project costs was borne by the Government. 4. Performance Audit Report: ROMANIA - Mostistea and Calmatui Irrigation and Drainage Project, Covurlui Irrigation, Buscani- Buzau-Siret-Prut (BBSP) Irrigation, Caracal-Titu Irrigation (Loans 1670, 1795, 1938, 1971-RO), OED Report No. 10224, dated December 30, 1991. 13 Technology Transfer 6.6 One of the reasons which prompted Romania to join the Bank in 1972 was the acquisition of more advanced technologies than those then available in the Eastern block. Bank-supported projects offered an important opportunity to introduce new technology. But both the Bank and the Borrower (see next Section) are to blame for the failure to actually effect such technological transfer. In some cases, the Bank properly introduced in the project design new technologies but then failed to check and challenge during supervision the deviations introduced by Government. In other cases, the Bank missed the opportunity for technical transfer. 6.7 In the case of the Orchards Project, modern technologies including dwarfing rootstocks and high planting densities had already been introduced in Romania, but their adoption had been haphazard and, often, they had been uncritically introduced onto unsuitable soils and topographical conditions. Besides, technical, commercial and economic aspects of fruit storage had been neglected. To avoid repeating these mistakes, the Bank insisted that the most intensive high density planting system, accounting for about 70 percent of all new plantings, be used only in the most favorable areas, where soils were good and irrigation water was available.! The remaining 30 percent would be less intensive orchards established on poorer soils and planted at lower density. Farm machinery, particularly sprayers, would be provided under the project to state farms and cooperatives. Fruit packing and cold stores would use prefabricated panels for insulation, controlled atmosphere to ensure longer-term conservation, and separate devices to control the temperature and atmospheric gas composition of each cold chamber individually. 6.8 Orchards project, as appraised, had important technological goals. The Government's pursuit of physical targets at the expense of quality, combined with poor Bank supervision, resulted in serious shortcomings. First, as already noted (para. 2.5), the audit found that a large number of project orchards had been established on steep slopes and marginal lands, and without irrigation, resulting in higher operating costs and lower yields than forecast at appraisal. By the same token, peach and apricot orchards, which require good irrigated lands in the Danube plain, were not planted at all. According to authorities and staff in the Ministry of Agriculture, BAFI and research centers involved in the fruit subsector during project implementation, the decision was made by the President himself, who strongly opposed fruit planting on good, flat and irrigated soils which, in his view, should be reserved for grains. Second, and contrary to the appraised project design, cooperative farms were not provided with farm machinery and spraying equipment; the Government decided instead that all machinery would be concentrated in the SMAs, a major impediment to effective orchard management, where timely pest and disease control is essential. Third, the project's cold storage facilities were designed by force account by a Government-owned Institute, without tendering and without specialized technical support or control. The Institute used the technologies with which it was most familiar and, in order to minimize use of foreign exchange, did not provide for the chambers of the cold stores to be equipped with controlled-atmosphere equipment, nor with the proposed separate devices, preventing individual (and less expensive) operation of each chamber. And fourth, all cold stores, which at appraisal were meant to be located close to the main fruit producing areas, were located on state farms and fruit research stations, and for their exclusive use, thus leaving cooperatives without project-built facilities to store their fruit. 5. See SAR of the Orchards Project; Report No. 2907-RO, May 28, 1980, paras 7.02 and 7.03, and Annex 1. 14 6.9 These developments remained unnoticed by the Bank and were not mentioned in any supervision report. This was due to the fact that supervision missions were less frequent than Bank average (except for 1983, there was only one mission per year), seldom included fruit specialists, and were discouraged from spot checks at field visits by Romania authorities. 6.10 Bank technical contributions to the Livestock Project included improved design of dairy units, modern milking equipment, changes in animal nutrition, and better use of pastures. Despite initial reluctance on the part of the Romanian authorities (who argued that the proposed technologies were not suited to Romanian livestock), Bank missions succeeded in convincing project staff to adopt them. The training and research components were well designed and proved useful. The animal feed issue, which figured so prominently in the previous three livestock projects (paras. 6.21 and 6.22) and were to hurt the preparation of the fifth project (in the event, not completed), does not seem to have been a major problem in the project under audit. Bank supervision was adequate (2 missions p.a.) and included livestock specialists who developed good relations with their Romanian counterparts. Both during implementation and at the time of audit, there was substantial evidence that the advice provided by the Bank was much appreciated by the Borrower. The size of the dairy farms (700 to 2,000 cows each; 922 on average), however, was too large for efficient management under public sector enterprises. Nowadays, the change from large-scale dairy farms to family units made many innovations introduced under the project technically irrelevant. 6.11 Contrary to the Orchards and Livestock projects, the Moldova Agricultural Credit project missed the opportunity to introduce more advanced or efficient farm machinery. With the exception of a few tractors and combine harvesters, all ICB contracts for farm machinery were awarded to Romanian manufacturers. The audit found, however, that the quality and performance of locally- made farm machinery are now severely criticized by authorities and officials of the Ministry of Agriculture and AGROMECs. The main complains refer not only to poor quality when compared to their imported equivalents, but to limited durability, higher fuel consumption, and, in the case of harvesters, high grain losses (para. 2.13). A similar problem was found by the 1990 performance audit report on the four concurrent irrigation projects mentioned above (para 6.46), in which inefficient, fuel-gulping motors and pumps of obsolete design were locally procured. 6.12 Given Romania's distorted pricing structure and its centralized decision-making, only the letter of the Bank's procurement policies was observed - not its spirit. In all cases (irrigation; cold stores; farm machinery), procurement technical specifications were prepared in Romania by Government agencies or institutes, and at a time when a strict Government "procure locally" policy was in effect (para. 6.17). Therefore, these deficiencies raise the question of whether the procurement process was handled in such a way that it gave an inappropriate advantage to technically inferior domestic machinery, i.e., whether the technical specifications and performance/efficiency factors in the tender documents were discriminating enough to ensure that machinery of different technical standards and performance would not be compared directly as if they were equivalent. Inadequate Bank review of the technical specifications in the procurement process may have thus jeopardized the possibilities of technological transfer under both the Orchards and Moldova projects. 6. OED, Repon No. 10224, op. cit. 15 6.13 These technical procurement issues, however, appear to have been subordinated to Romania's insistence on participating in Bank-financed procurement so as to secure free foreign exchange. This lasted through most of the projects' implementation period. Once Regional Projects management realized the kind and magnitude of the distortions being introduced into the procurement process, it insisted on a more straightforward approach to project financing and procurement. As a result, Romanian authorities stopped to request Bank loans. Support to Individual Producers 6.14 The Bank was concerned that the small-scale private producers in the mountainous areas of the country, left outside the socialist sector, had been neglected by Government policy. As a result, both the Livestock and Moldova projects provided for credit to individual farmers. At project completion, data available indicated that subloans to private farmers had notably exceeded SAR projections. At the time of audit, however, there was no evidence that this "private sector" orientation of the projects had ever been accepted by the Government, let alone implemented. "Individual producers" receiving sub-loans under the projects turned out to have been mostly cooperative members, who were allowed to borrow (or even instructed to borrow, see para. 6.15) to develop the individual plot they were allowed to farm within the cooperative. THE BoRRowE's CONTRIBUTION TO THE PROJECTS 6.15 The Borrower must be credited with effective and timely project implementation. Project works were done on time; procurement proceeded unchallenged (see paras. 6.12 and 6.17); disbursements were timely. At the time, when portfolio performance was assessed largely through input indicators, Romania enjoyed a high prestige in the Bank on precisely these grounds. A more detailed exploration of the implementation process, including interviews with individuals who were involved in project implementation at different levels (top executives, middle managers, provincial Ministry and BAFI representatives, managers of state farms and cooperatives, and members thereof) revealed a different view of the factors behind such a picture-perfect performance. All details of every project had to be personally reviewed and approved by the President." Once approved, there was no problem with procrastination or non-compliance. Direct investments were included in the budget and dutifully implemented. "Credit" was never what Bank staff, trained in market economics, thought it was. In Romania, it was a centralized procedure for resource allocation, parallel to the budget. Firms and individuals did not freely approach the BAFI for a loan for a seemingly worthwhile investment; they were rather instructed to get funding for such and such centrally-decided investments from the BAFI, where the money was - of course - already earmarked for them. Again, a picture-perfect on-schedule credit-taking by "beneficiaries". Some targets were even exceeded. 6.16 But at the same time, this highly-centralized, non-challengeable method for decision-making is to blame for most of the projects' worst design features. It was the President who decided that high-density orchards would not be planted on flat, irrigable lands - actually, the only lands where the 7. Sometimes, to the Bank's embarrassment, like once when Bank staff believed they had the Government's agreement on a certain document only to discover later on that there was a page attached to the supposed agreement indicating the "must change' items for the document to be acceptable to Romania. 16 full advantages of these orchards could be reaped. He also forbade culling and instructed all agricultural producers to retain all available animals alive to increase the size of the national herd.' 6.17 It was also the President's decision to utilize foreign exchange provided by the loans to finance investments in other, more "strategic" sectors rather than to use it for the intended project- related purposes. Later on, when he made the decision to prematurely repay the country's foreign debt, foreign exchange provided by the loans was used for this purpose. Therefore, project managers were under explicit instructions not to incur expenses in foreign exchange. As a result, very few persons were sent to study abroad, and those, only in the initial project years. Technical assistance funding was never used. The final engineering design of the cold stores did not include any equipment which could not be manufactured locally (e.g. controlled-atmosphere equipment) and most tenders for farm machinery were awarded to domestic producers. And no bid was ever called for the design jobs; all engineering designs and technical specifications for tenders were done on force account by Government agencies and institutes. 6.18 Most loan proceeds were disbursed against local expenditure, thereby generating "free" foreign exchange for the Government. Farm machinery tenders were almost invariably awarded to inferior domestic products. Project works were completed with imported equipment lacking. Training abroad took place only to a minimal extent. Even the so-called "private sector" lending proved to have been a fagade to convey financial resources to the cooperative sector. Functionnaires taking decisions or drafting the designs or specifications could not but go along with the central directives.' In short, the same Government and governing methods to be credited for an expedient project implementation must as well be blamed for the lost opportunities in introducing modern technology into Romania. 6.19 Project experience demonstrates the limits of Bank's influence in a centrally-planned country where the decision-making process was concentrated basically on a single individual; political considerations prevailed over technical aspects; and Bank requirements for information or for detailed field-checks during supervision were perceived as interference in internal affairs. Under such conditions, compounded by the Bank's limited leverage (para. 6.5), the projects' designs were distorted and supervision had a negligible impact. THE FOURTH LIVESTOCK PROJECT AND THE EcONOMic ANALYSIS OF PROJECTS IN THE BANK 6.20 The appraisal process which eventually led to the Fourth Livestock project was to play a major role in reinstating Bank policy on economic project analysis. 6.21 While both the Orchards and Moldova projects were the first ones in their series, the Livestock project was the fourth one. The First and Second Livestock Projects (Pigs I and II; July 8. The dismissal of this old-regime "non-culling" instruction resulted in a reduction by one-third of the national cattle herd in the couple of years after the Revolution, as unproductive or barely productive animals were removed from the national herd. Milk and meat production did not fall and the former may actually have increased substantially, at a lower cost in both financial and pasture/feed consumption terms. 9. The audit mission received frequent evidence about the impossibility (only at personal risk!) of challenging or even discussing directives coming from the President and, by association, from the central staff in the State or Party. 17 1977-June 198210, and March 1979-June 1984, respectively) made credit available through BAFI to state farms and agricultural cooperatives for investments in pig production and processing facilities to increase the supply of meat and meat products for domestic consumption and export, and to increase the industry's efficiency and productivity. In spite of longer implementation periods, project objectives were generally achieved and their rates of return, albeit lower than appraisal estimates, were rated as satisfactory. Together they financed 93 percent of all investments in pig production and processing facilities in Romania during their implementation period. The projects were soundly conceived and the technologies involved were well understood in Romania; innovations included agricultural credit to cooperative members and research on pig nutrition. But the lack of foreign exchange to import protein feed and Romania's inability to significantly increase domestic protein feed production, two factors which the Government refused to acknowledge, seriously affected the industry's bio-economic and production efficiency parameters. 6.22 The Third Livestock Project (Poultry; October 1979-December 1985) brought to the fore some of the technical change issues which were to surface so strongly in the three projects under audit. It was also administered by BAFI, which made funds available and supervised the implementation of sub-projects carried out by cooperatives and state farms for breeding of high quality broiler parent stock; expansion of broiler growing units; provision of new and modernized slaughterhouses; research; and quality and disease control. Physical implementation was satisfactory. Procurement and disbursement proceeded without significant delays. Despite the improvement in the genetic quality of the broiler population by the importation of grandparent stock, broiler performance deteriorated as a result of low protein content in the feed. This had been envisaged at appraisal as a critical factor in the success of the project, but the Government refused any suggestions by the Bank to include in the project provisions for supply of protein feed. At the insistence of the Romanian negotiators, a Bank-demanded covenant regarding protein feed importation was not included. The rate of return for the broiler complexes fell to 0 percent (compared to 17 percent at appraisal) and the overall rate of return for the project to 4 percent (compared to 12 percent). 6.23 In 1978, appraisal of a possible fourth project (Cattle Development Project1), containing dairy development and beef fattening activities, was already well advanced. The appraisal team realized that officially set domestic prices for milk and milk products bore no resemblance to economic values. In a break with then accepted Bank practice of using domestic milk prices in economic analyses, the team used border equivalent prices (based on nonfat milk powder and butter oil of West Europe or New Zealand origin), but as a result the project proposal submitted by Government had a negative economic rate of return. Two post-appraisal missions, by marginally improving project" design and adopting domestic prices for a certain proportion of incremental milk production, managed to get a positive rate of return, albeit were below 10 percent. The Romanians 10. Approval date, and Closing date, respectively. 11. The project originally conceived in 1077-78 was to be a Cattle and Poultry Development Project. The poultry component was dropped from the project's design during the first appraisal mission for severe lack of preparation. Poultry eventually became the focus of the Third Livestock Project. 12. Romanian officials adamantly refused to accept significant changes to technical aspects of the project, although many would have been possible to cut cost and raise efficiencies. 18 did not accept the milk valuation criterion and strongly protested to the Bank. Bank middle managers were concerned as well, as they were committed to deliver the loan. Central Projects Staff (CPS) and, later on, the Loan Committee confirmed the appraisal team's economic assessment of the project as the only proper one. The project proposal was aborted. The new project prepared in its stead, taking advantage of the findings reflected in the economic evaluation of its predecessor, eventually became the project under audit (Fourth Livestock: Cattle). 6.24 Having stopped the earlier Cattle Development project in 1978, Central Project Staff were better able to stop other projects every year on similar grounds without the same kind of hassle. And middle managers were put on notice that lending programs and pressure to lend would not be accepted as a substitute for proper economic analysis. PROJECT SusTmNABILmy AFrER THE REVOLUTION: FOUR TRANSriONAL ISSUES 6.25 Any large-scale economic change implies major transitional problems. Everything cannot be changed concurrently, and problems or delays encountered in some of the reforms may weaken or jeopardize others. In the particular case of Romania, the policy reforms introduced after the Revolution are expected in the longer term to facilitate economic growth and stability in agriculture. In the short term, however, some reforms or the lack thereof, including the land restitution process, the limited changes to the marketing system, the slow rate of privatization, and the access to financial resources, have had negative effect on project success and sustainability. Land Restitution 6.26 Within countries moving from centrally-planned to market economies, Romania, China and Albania were among the first to undertake a large-scale, rapid land privatization program. While such programs have generally been successful and well accepted by the public, they generated a number of problems and created a number of rigidities which offer useful lessons for other countries. 6.27 In 1990, the Romanian Government decided to restore land, within a 10 ha limit, to its former owners or their heirs (paras. 3.1-3.3). But most of these are old people, who left the region after land expropriation in the late forties, resided in urban areas for many years, lost experience in agriculture, and lack financial resources to make a farm work. As a result, only some 40 percent of them are active farmers and many of those concentrate on self-sufficiency. The creation of voluntary Farmers' Associations and Agricultural Societies has partly compensated for lack of participation and lost experience of the land owners. However, with some exceptions, the operating system of these organizations is not much different from that of former socialist cooperatives, with the same weaknesses and negative connotations. These factors largely explain the initial decline in agricultural production which followed land redistribution; a substantial recovery is now being experienced (para. 3.7). 6.28 Further, each family was eager to recover the same parcels of land owned prior to expropriation, resulting in a high degree of farm fragmentation. About twenty million parcels of land were restored, a difficult process made worse by inadequate records and subsequent land disputes. In the future, the development of a land market should permit successful farmers to consolidate into larger and more efficient farm units, but this may take years to materialize. 19 6.29 In contrast, China's land redistribution program (1979-83) distributed responsibility for farming collective lands to former cooperative members and state farm employees, who quickly adopted a common cropping pattern. This contributed to a considerable production increase during the years following land distribution. The small size and fragmentation of holdings, prevailing in China as in Romania, was not an issue. The entrepreneurial spirit of farmers resulted in diversification and intensification of crops. Small farm machinery, well suited to the prevalent farm size, was promoted. 6.30 From these two experiences, the question can be raised as to whether restitution of land to inexperienced people - however rightful - should be avoided in the first place. Perhaps a dual system, offering a choice between restitution of land or compensation in cash, as currently done in Albania1, is a more effective and less risky solution for countries undertaking land redistribution to former owners. Marketing 6.31 Romania's experience illustrates the difficulty to change and modernize the marketing systems for horticultural and dairy products, and to adapt them to a different farm structure. In spite of a legal framework favoring private initiative, the establishment of private marketing channels has been slow. The markets for fruits and milk are characterized by high demand by consumers while heavy losses or inappropriate use of products at the farm and marketing levels occur. 6.32 Besides facing a shortage of working capital and difficult access to credit (discussed below), private marketing channels are weak and poorly organized. The country lacks experienced, specialized private fruit traders, and cold stores and refrigerated transportation. Most fruit trading companies are too small to buy the total production of a typical project orchard, or do not have enough working capital to buy fruit from producers at harvest time and keep it in cold stores. Contract farming, an effective option available in many western and Asian countries, is virtually non- existent. Most fruit processing industries are still state-owned and have old and inefficient equipment. Hence the difficulty encountered by fruit producers to sell their products, and the massive deviation of otherwise good products to processing. Even this has proved difficult; since domestic industries lack the working capital to acquire raw materials, fruits such as sour cherries have had to be sold to Italian manufacturers. 6.33 In many cases, the existing marketing infrastructure is not suited to the new land tenancy structure. For example, with the redistribution of project cattle among individual farmers, there is a need for more small milk collecting centers at the village level. Similarly, the project silos and grain stores are no longer suited to the large number of small grain traders who need smaller warehouses. 13. Impact Evaluation Report: CHINA - North China Plain Agriculture Project, Credit 1261-CHA, OED Report No. 13243, dated June 29, 1994 and CHINA - Agriculture to the Year 2000, A World Bank Country Study, The World Bank, Washington, D.C., 1985. 14. Performance Audit Report: ALBANIA - Rural Poverty Alleviation Pilot Project, draft OED Report; and ALBANIA - Law on Compensation in Value of Forner Owners of Agricultural Land, Tirana, dated April 21, 1993. 20 6.34 For dairy products, the demand is high for fresh milk, but its low official price (still controlled by the Government) has not induced private traders to develop the fresh milk collection and distribution market. As a result, milk is mostly consumed locally or used by farmers as animal feed; some surpluses are bought by small traders to make cream and cheese, not subject to price control. 6.35 The lesson is that establishing a legal framework is necessary but not a sufficient step for private marketing channels to develop. There is a need for adequate financial resources, and for specific training programs on management skills, market knowledge, and handling, transporting and marketing perishable products. Joint ventures with experienced partners appear to be another route for modernizing both the fruit and milk processing industries. PKvatization 6.36 Full privatization of all enterprises in agriculture and agro-industry is one of Government's objectives. But privatization has been slow due to, inter alia, an asymmetric and burdensome taxation system; an unequitable allocation of responsibility for repaying past debt; and lack of financial resources for investment and working capital. 6.37 On taxation, individuals or firms acquiring divested state companies or establishing new companies benefit from tax exemptions and incentives. In contrast, former state-owned enterprises being privatized under the MEBO system are heavily hit by the standard wage, profit and value-added taxes. This has hindered or slowed down the privatization of AGROMECs, agro-industries, cold stores, and other facilities supported by the projects. Since the MEBO method is central to the Government privatization strategy, it needs to be made more attractive from a fiscal perspective. 6.38 On allocation of responsibility for past debt, if commercial or "inter-coop" associations are formed to continue operating former cooperative assets, they must assume the remaining debt and repay BA out of their income (para 5.13). This encumbers such new enterprises with debt, sometimes heavy, which had actually been incurred by the former cooperative; repayment jeopardizes the financial profitability of the new enterprises and eats away whichever working capital they had. 6.39 Financial resources have been a constraint for private entrepreneurs in terms of purchasing or renting project-built facilities (e.g. stables) or investments (e.g. orchards), as well as for fruit traders and milk collectors at the village level. As bluntly summarized by the Livestock Director of the Ministry of Agriculture, "we have lands, empty stables, good specialists, cheap labor, but no money." Rural Finance and Credit 6.40 The Romanian experience shows that limited access to credit may become a major constraint to project success and sustainability. Under a massive land redistribution and firm privatization program, the need for adequate financial resources at all levels is considerable. But the banks, however, including BA, are very security-conscious and refuse to lend unless they get ironclad guarantees. For example, they do not accept land titles as collateral on the grounds that there is no land market in Romania and thus land has no market value and land accepted as collateral cannot be sold in case of default. Similarly, BA is reluctant to lend to farmer associations, agro-industries and private traders unless they can guarantee that their products will be sold (in practice, unless they have a selling contract with a state commercial agency). Storage warrants are generally not accepted 21 as collateral either. They also use cumbersome procedures to discourage potential agricultural borrowers. Sub-loans from external sources (World Bank-supported credit programs, among others) and subsidized loans are particularly attractive for potential borrowers, but only a few have been able to meet the strict criteria of special credit programs." 6.41 The Government is considering the creation of a more active, dynamic and decentralized rural credit bank. The Bank is not in favor of such an initiative on the grounds that it would focus mainly on subsidized credit and therefore would rely on public budget allocations for its business. But the fact remains that traditional agricultural and rural finance institutions and mechanisms are not suited to the sweeping economic and social changes Romanian agriculture is facing. Institutional reform with respect to land rights and the financial sector has a very high priority. 15. The slow disbursement of the Bank-supported Agricultural Credit Project (Cr. 3486-RO) is mostly due to difficulties for potential borrowers to meet the project eligibility criteria.