Document of The World Bank FILE COPY FOR OFFICIAL USE ONLY Report No. 2590 PROJECT PERFORMANCE AUDIT REPORT BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) June 29, 1979 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.  FOR OFFICIAL USE ONLY Project Performance Audit Report BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) TABLE OF CONTENTS Page Preface 111 Basic Data Sheet V Disbursement Table VI Highlights vii PROJECT PERFORMANCE AUDIT MEMORANDUM I. Summary 1 II. Main Issues 4 A. Interest Rates and Monetary Correction 4 B. Sub-loan Documentation 7 C. Project Accomplishments 8 D. Bank Performance 11 E. Conclusions 13 Annex I - Sub-loan Approvals Progress 15 Annex II - Selected Price Indexes 16 Annex III - Estimated Losses in Farm Stored Grain 17 Annex IV - Grain Handling Tariffs 18 Annex V - Borrower's Comments 19 PROJECT COMPLETION REPORT I. 'Introduction 23 II. The Project 24 III. The Principal Issue: Sub-loan Interest Rates 30 IV. Other Issues: Project Preparation and Public- Sector Enterprises 35 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.  Page Annex 1 - Comparison of Actual Investments with Estimates Resulting from Survey 41 Annex 2 - Comparison of Subproject Distribution Investment Cost and Installed Capacity of Subborrower Group 42 Annex 3 - Subloan Investment Numbers of Sub- projects and Storage Capacity 43 Annex 4 - Subloan Investment Numbers of Sub- projects and Storage Capacity 44 Annex 5 - Banco do Brasil: Summary of Loans Made 1966-1976 45 Annex 6 - Banco do Brasil: Numbers of Branches and Employees 1966-1976 46 Annex 7 - Banco do Brasil: Condensed Income and Expense Statements for Years 1966-1976 47 Annex 8 - Banco do Brasil: Condensed Balance Sheets for Years 1966-1976 48 Annex 9 - Number and Amount of Subloans under Successor Program 49 Annex 10 - Disposition of Subloan Approvals 50 Annex 11 - Subloan Recovery Experience 51 Annex 12 - Comparison of Project Cash Flow: Appraisal Estimate vs Estimate Based on Actual Transactions through 1976 52 Annex 13 - Grain Production in Project Area - 1975-1976 53 Annex 14 - Cost-Benefit Streams from Project as Implemented 59 Annex 15 - Estimated and Actual Completion Dates of CEAGESP Installations 61 Map  - 111 - Project Performance Audit Report BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) PREFACE This is a performance audit of the Grain Storage Project in Brazil for which Loan 857-BR was approved in September 1972 in the amount of US$30 million and was cancelled after US$18.2 million had been dis- bursed in September 1976. The audit report consists of an audit memorandum prepared by the Operations Evaluation Department (OED) and a Project Completion Report (PCR) dated June 1978 prepared by the Latin America and Caribbean Regional Office following a mission to the country in November 1977. The audit memorandum is based on a reading of the Appraisal Report (No. PA-135a) dated August 2, 1972, the President's Report (P-1115) of August 22, 1972 and the Loan Agreement dated September 27, 1972 and the PCR; correspon- dence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have been reviewed, and Bank staff who have been associated with the project have been interviewed. An OED mission visited Brazil in January 1979. The mission made field trips to visit some of the project supported storage facilities and had discussions with staff of Brazilian ministries and agencies con- cerned with the project, the Banco do Brasil and the Government Project Evaluation Office. The information obtained during that mission was used to test the validity of some of the conclusions of the PCR and is also reflected in the audit memorandum. A copy of the draft report was sent to the Borrower on April 4, 1979. The comments received from the Government (see Annex V) have been taken into account; noted corrections in the PPAM have been so incorporated. The audit finds the PCR comprehensive and accurate with respect to the project's principal achievements and shortcomings. Some of the points discussed by the audit mission have been selected to highlight the Bank's difficulties in dealing with interest rate indexing in economies characterized by high inflation.  - 1v - The valuable assistance provided by the Government of Brazil, the Banco do Brasil and their staff met during the preparation of this report is gratefully acknowledged.  PROJECT PERFORMANCE AUDIT REPORT BASIC DATA SHEET BRAZIL GRAIN STORAGE PROJECT (LOAN 857-BR) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$ million) 69.8 66.5 /1 Overrun (%) - n.a. /2 Loan Amount (US$ million) 30.0 30.0 Disbursed )- 18.2 Cancelled ) December 31, 1978 /3 - 11.8 Repaid to )- 1.9 Outstanding to )- 16.3 Date for Completion of Physical Components 05/78 n.a. /2 Proportion Completed by Appraisal Target Date (%) - 105 Proportion of Time Underrun (X) - 30 Economic Rate of Return (%) 25 20 /4 OTHER PROJECT DATA Original Actual or Item Plan Revisions Current Estimate First Mention in Files or Timetable - - 05/20/65 Negotiations 06/26/72 - 07/15/72 Board Approval 07/25/72 08/15/72 09/05/72 Loan Agreement Date 09/27/72 - 09/37/72 Effectiveness Date 12/29/72 - 01/19/73 Closing Date 12/31/79 - 04/02/76 Borrower Banco do Brasil Executing Agency Banco do Brasil Fiscal Year of Borrower January I - December 31 Follow-on Project Name none MISSION DATA Month, No. of No. of Date of Item Year Weeks Persons Manweeks Report Preparation I 08-09/65 1 3 3 09/09/65 Preparation II 11/65 1 2 2 12/10/65 Preparation III 03/66 1 1 1 03/15/66 Preparation IV 06/66 1 2 2 06/29/66 Preparation V 01-02/67 2 1 2 02/10/67 Preparation VI 03-04-68 4 4.5 18 04/19/68 Preparation VII 08-09/70 1 2 2 09/24/70 Preparation VIII 03-04/71 4 3 12 04/07/71 Preparation IX 08-09-71 1 1 1 09/18/71 Appraisal 10-11-71 6 4 24 08/02/72 Total 22 67 Supervision I 01-02/73 1 2 2 02/13/73 Supervision II 07/73 1 2 2 08/22/73 Supervision III 01/74 2 2 4 03/12/74 Supervision IV 09-10/74 1 4 4 11/20/74 Supervision V. 02/75 2 2 4 04/30/75 Supervision VI 09-10/76 1 1 1 10/26/76 Completion 10-11/77 1 2 2 06/05/78 Total 9 19 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Cruzeiro (Cr$) Year: Appraisal Year Average Exchange Rate: US$I = Cr$ 5.505 Intervening Years Average US$1 = Cr$ 5.379 Completion Year Average US$1 = Cr$ 8.204 /1 Total cost without land cost. /2 Due to cancellation of loan not applicable. T Most recent month available. 74 The Region assumed a higher ERR (see PPAM paras. 33-36) while CPS questions both valuations on the grounds that some development would have taken place also without the project.  - vi - Project Performance Audit Report BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) DISBURSEMENT TABLE (US$ million, cumulative) Period Appraisal Actual Actual as % Ending Estimate Disbursement of Estimated 12/31/73 0.6 1.7 283 06/30/74 2.6 5.8 223 12/31/74 6.0 8.6 143 06/30/75 9.6 11.8 122 12/31/75 13.6 17.4 127 06/30/76 18.1 18.2 101 12/31/76 22.9 - - 06/30/77 27.6 12/31/77 30.0  - vii - Project Performance Audit Report BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) HIGHLIGHTS The loan, amounting to US$30 million, was the first in Brazil for grain storage development. The project was designed to improve grain processing and storage in the grain producing areas of Southern Brazil through constructing, equipping and installing of new grain storage facilities and the improvement and expansion of existing facili- ties. Total investment in storage facilities was expected to total US$75 million. In addition to the Bank loan, US$30 million would be financed by the Bank of Brazil (BB) and US$15 million by the sub-loanees. Eight hundred sub-projects were expected to be financed in nine states which would provide a total of around 1.9 million m tons of storage capacity, of which 200,000 m tons would be on farms and 1.7 million m tons of capacity in intermediate and large scale commercial warehouses and processing facilities. As with the First and Second Livestock Development Projects (Loan 516-BR and Loan 868-BR), disagreement developed with the Borrower over the interest rate indexation. Eventually the Borrower established a competing grain storage credit program at essentially negative interest rates - Programa Nacional de Armazenamento (PRONAZEM) - which led to the Bank's request that remaining funds amounting to US$6.26 million be can- celled. The Borrower agreed to this on October 28, 1975. The Borrower then requested a further cancellation of US$5.52 million on September 27, 1976 due to surplus fund in the project unit account as a result of pre- payment of sub-loans by farmers and other recipients. For the disbursed portion of the loan, US$16.3 million, the rate of return estimated by OED is 20% compared to 25% at appraisal. However, the PCR estimated the rate of return at 38%, mainly due to different judgments on the grain losses prevented by new or improved storage. Other points of special interest are: - flawed project design and substantial changes made during implementation (PPAM paras. 31 and 46); - initial sub-loan documentation was too complex for small farm operators and warehousemen, leading to slow disbursements and documentation revisions (PPAM para. 26 and PCR para. 2.09);  - viii - - the Bank may have been excessively flexible in accommodating itself to a clear breach of a major condition of the Loan Agreement (PPAM para. 41); - insufficient reporting by Borrower led to diffi- culties in analyzing project progress (PPAM para. 45); - lower costs than expected at appraisal permitted construction of additional storage volume (PPAM para. 30; PCR paras. 1.02, 2.05 and 2.06); - difficulties in reconciling different problems of farm storage:commercial storage in a single project (PPAM paras. 20, 24 and 26); and - weaknesses in determining economic benefits of grain storage (PPAM paras. 33 to 36; PCR paras. 2.18 to 2.20).  Project Performance Audit Memorandum BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) I. SUMMARY Objectives 1. The Grain Storage Project was part of the Government of Brazil's program for the development of the country's grain processing and storage industry. It was designed to provide funds for "the construction, equipping and installation of new, and improvement and expansion of existing grain storage facilities", including the acquisition and installa- tion of machinery and equipment for grain processing and storage. 2. In more specific terms the project aimed at expanding and moder- nizing the grain storage and processing industry in the private sector. In so doing it was hoped that a number of secondary objectives would be achieved: (a) grain flows would be smoothed out, thereby promoting more efficient transport services; (b) handling costs would be reduced through the intro- duction of bulk handling wherever feasible; (c) marketing efficiency would be improved as farmers achieved more adequate storage facilities to "hold" grain for the most advantageous prices. In support of the latter objective, the original intent was to give priority to small-scale borrowers. Sub-loans for public sector storage facilities were not excluded in the initial project proposals, but would be accorded a low priority, and would require prior Bank approval. 3. Quite apart from the above objectives of a domestic develop- mental nature, the Bank wished to see the project strengthen the recent establishment of positive real interest rates for agricultural credit in Brazil. The agreement, therefore, carried provisions for monetary correc- tion on sub-loans, to overcome negative real interest rates resulting from a high level of inflation. Identification and Appraisal 4. The need for expanded storage facilities to expedite corn and other agriculture exports was first identified in 1965. A project preparation mission visited Brazil, and a government working group was - 2 - established to elaborate a project in three separate areas: farm storage, intermediate storage and processing facilities and port facilities. The studies which were to have been carried out were subject to prolonged delays, or were inadequate for appraisal purposes, and by 1970 no real progress had been made in preparing a project, despite six separate Bank missions. 5. By early 1970, however, the shortage of grain storage facili- ties in Brazil was acute, preparation of a project was re-instituted, and was ready for appraisal by mid-1971. The loan agreement was signed on September 27, 1972, and became effective on January 16, 1973. The Project 6. The project as appraised called for a total investment in storage facilities of US$75 million, on a funding basis of US$30 million by the Bank; US$30 million by the Bank of Brazil, and US$15 million by the sub-loan recipients. It was intended to finance construction of approximately 1.9 million metric tons static storage capacity, of which 200,000 tons would be on-farm storage and 1,700,000 tons intermediate and large-scale commercial warehouse and processing facilities. It envis- aged approximately 800 sub-projects in nine states of south and south central Brazil. Project Execution 7. The executing agency for the project was the Bank of Brazil, operating through its many branches throughout the production area. Bank of Brazil established a specific line of credit called "PRODESAR" to accommodate sub-loans. In the early months of the program, demand for loans was strong, but the number of loan approvals was low (Annex 1). This was the result of overly complex application and documentation requirements, particularly for small-scale sub-borrowers. When these requirements were subsequently simplified, loan applications and approvals skyrocketed. Take up of sub-loans continued at a fast rate through mid- 1974. Earlier that year it had become evident that the internal rate of inflation was proceeding at a much faster rate than monetary correction under the loan agreement formula, and that real interest rates under the project were negative. Bank of Brazil was asked by the Bank to suspend further loan approvals after June 1, 1974, until a new monetary correc- tion formula could be agreed. This was not achieved until mid-September, and thereafter loan applications and approvals again slowed markedly. 8. Nevertheless, by mid-1974 the project was well ahead of appraisal schedule, and it was anticipated that loan funds would be fully committed by the end of 1975. Accordingly, consideration was given to a follow-up project, both by the Bank and the Government, and Bank of Brazil was preparing a feasibility study for presentation to the Bank. -3- 9. By the end of 1974 Bank of Brazil had approved 442 sub-loans, and had only 16 under review, including a large project by a state-owned company, CEAGESP. The latter was subsequently appraised separately by the Bank, and approved for inclusion in the project. It involved the construction of four warehousing units of 60,000 ton capacity each, at an overall cost of US$25 million - one-third of the total project fund. Termination of the Project 10. The new monetary correction formula had the effect of tying the sub-loan interest rate more or less directly to the domestic rate of inflation (general price index) and resulted in nominal interest rates approximating 40% per year. There was, nevertheless, still a strong demand for additional storage capacity, and the Government was under strong pressure to liberalize the terms under which development capital could be made available. 11. In early 1975, the Government instituted a new program - Programa Nacional de Armazenamento - "PRONAZEM" - to provide funds for on-farm and secondary grain storage facilities in areas where the need was most acute. Interest rates were initially set at 8% for farm storage loans and 15% for secondary storage, both without monetary correction. The immediate effect of the announcement of this program was that new applications for loans under PRODESAR abruptly ceased, and existing applications were withdrawn. The one exception was the outstanding Public Sector application from CEAGESP. After approval of this latter sub-project, the Bank requested that the remaining uncommitted funds under PRODESAR, amounting to US$6.26 million be cancelled and this was agreed with the borrower on October 28, 1975. 12. While no new sub-loan commitments were made after mid-1975, dis- bursements under approved loans continued, particularly to CEAGESP. On the other hand, the revised monetary correction formula, with resulting high interest rates, stimulated a major pre-payment program on the part of sub-borrowers. Bank of Brazil consequently found the PRODESAR account in surplus funds, relative to outstanding commitments, and requested a further cancellation of US$5.52 million on September 27, 1976. This was agreed, reducing the net Bank Loan to US$18.2 million, as compared with the original US$30 million. Closing of the Loan 13. With disbursements being made only for the CEAGESP sub-project, and no possibility of further sub-loans under PRODESAR, the Bank suggested that the Loan be closed as soon as these disbursements were completed. This was accomplished on October 4, 1976 - more than three years ahead of appraisal schedule. -4- II. MAIN ISSUES 14. The PRODESAR Project encountered early administrative problems which slowed implementation, but which were solved with time. The major issue, however, was a fundamental difference between the Bank and the Government in loan philosophy, and this eventually caused the collapse of the program with only 60% of the loan funds utilized, although the need for storage facilities remained acute. A. Interest Rates and Monetary Correction 15. Traditionally, agricultural loans in Brazil had enjoyed a degree of subsidy which successive governments had apparently felt necessary for the economic and social development of the sector. The Bank had been working toward arrangements which would guarantee positive real interest rates on sub-loans for agricultural development projects in Brazil and other South American countries. In the negotiation for the Grain Storage Project Loan, the Bank was insistent on a clause incorpo- rating monetary correction on sub-loans, on the same basis it had applied to a parallel Loan Agreement for the livestock development projects (Loans 516/868-BR).1/ The clause called for an annual interest rate on sub-loans of 7-1/2% plus a monetary correction representing "the rate of increase in the value of dollars in terms of the currency of the Guarantor between successive monetary correction dates of June 30th and December 31st in each year." The Loan Agreement further specified that the Borrower should immediately suspend approval and granting of sub-loans if, for any 12-month period preceding a monetary correction date, the average annual rate of increase in the general price index exceeded by 7-1/4% or more the average annual increase in the dollar/Cruzeiro exchange rate as computed by the Central Bank of Brazil. 16. During the first year of effectiveness of the Loan Agreement the dollar/Cruzeiro exchange rate remained almost constant, in spite of an internal rate of inflation of around 15.5%, and consequently the effect of monetary correction on the rate of interest paid by sub-borrowers was negligible. The supervision mission of January 1974, however, noted that domestic inflation was proceeding at a pace which resulted in a negative real rate of interest to sub-borrowers. Although the Agreement called for immediate suspension of approval of sub-loans by Bank of Brazil under such circumstances, unless the Bank otherwise agreed, the mission recommended, and the Bank agreed to let Bank of Brazil continue 1/ PPAR No. 2402. - 5 - sub-loan approvals on the same terms until June 1, 1974, so as not to disturb the flow of applications. This concession was predicated upon Bank of Brazil accepting, by that date, a new monetary correction formula tied more or less directly to any increase in the national treasury bond price index (ORTN) rather than in the exchange rate. 17. At the time of this agreement Bank of Brazil had approved 90 sub-projects. It was not until mid-September, 1974, that a new formula was agreed to by the Bank and Bank of Brazil, by which time sub-loan approvals stood at 327, representing 75% of the final total storage capacity under the project. The new formula reduced the nominal interest rate from 7-1/2% to 5%, plus the monetary correction tied to the percentage increase in the ORTN. 18. The new formula applied only to sub-loans granted after September 19, 1974, and reflected a rate of inflation of approximately 29% at the time. This was later to escalate to around 40%, with disas- trous results for some sub-borrowers. At the same time, however, the monetary correction under the original formula began to cause concern as the Central Bank reflected more directly and immediately the level of internal inflation in its foreign exchange rates. Whereas the dollar/ Cruzeiro rate had remained almost stationary in 1973, it began to advance sharply in 1974, with the Cruzeiro weakening, on average, about 12% for the year and over 33% by 1975. Sub-borrowers found themselves in a position of having to pay the equivalent of roughly 40% annual interest on their loans within a year of having completed construction of the facilities. 19. Sub-borrowers' resistance to the new indexing formula was quickly evident, particularly among the small-scale farmers for whom the program was primarily designed. The farmer felt he could not afford to take out PRODESAR loans, regardless of how urgently he required storage facilities, when he had no idea of the ultimate cost of such loans. The concept of indexing for domestic inflation did not in itself appear to be the major deterrent. Rather the fact that the cost of farm inputs tended to rise more rapidly than the general price index, and the price of farm products more slowly, accentuated the financing cost. (See Annex 2). In addition farmers had little control over output in the high risk operation of cereal production, and consequently income levels were uncertain. 20. The principle of positive interest rates is no doubt sound, but in the audit's opinion the terms of the PRODESAR project should perhaps have been tempered somewhat to achieve the major purpose for which the project was designed - the expansion and improvement of primary storage and handling facilities. In this respect, as in others (para. 45), - 6 - the scope of the project was perhaps too broad to permit a completely successful conclusion. Farm storage needs, in retrospect, should have been treated differently than commercial handling and processing facilities. As the program evolved the terms, conditions, procedures and engineering were too complicated, and too costly, to serve the requirements of relatively simple and unsophisticated farm storage. Nor were farmers with no direct control over their costs or prices, in a position to pay interest on capital investment which "doubled their cost" every 2-1/2 years. 21. Sub-loans did continue after the introduction of the new index- ing system, indicative of the continuing strong demand for storage facilities. The Government, however, was under strong pressure to introduce a program with more liberal lending terms, and they acceded to this demand in early 1975. A new program - Programa Nacional de Armazenamento 'PRONAZEM' - was funded by the Government with Cr$ 800 million, and became operational about mid-year. The funds were to be disbursed over a period of four years for the construction of grain storage and processing facilities, at 8% interest for farm and 15% interest for secondary facilities, both without monetary correction.l/ 22. The institution of this program effectively killed the PRODESAR project, except for the major sub-loan to CEAGESP for public sector storage construction, then under appraisal. The new program caused bitterness amongst sub-borrowers under PRODESAR who were paying far higher interest rates, since "switching" from one program to the other was definitely not allowed. As a result a very high proportion of sub-borrowers paid out their loans almost immediately, using available cash from crops and re-financing from other sources, or by diverting personal funds from programmed projects such as farm expansion, new machinery, seed improvement, etc. 23. The Government was clearly in breach of the Loan Agreement in introducing the PRONAZEM project. It was a decision taken on the grounds of political reality, recognizing the developmental needs of a sector of the agricultural economy. In this context subsequent events have, perhaps justified the decision - there has been a very rapid expansion of storage facilities, up to mid-1978 approximately 1.3 million tons static capacity of on-farm storage and 2.4 million tons of secondary storage under PRONAZEM. 1/ Subsequently, interest rates increased somewhat, to 10% for farm storage loans. -7- 24. There is no doubt that the fundamental disagreement in lending philosophy between the Government and Bank has had an adverse impact on subsequent loans for agricultural development. In the course of the OED mission for this audit, Government officials expressed a strong feeling that some degree of subsidization was necessary to get agricultural proj- cects moving, and that farmers and small processors particularly cannot cope with interest rate uncertainty because of the high degree of varia- bility in both production and price. The Bank has recognized this problem in other projects in Brazil involving direct assistance to small farmers, and has allowed interest subsidization.1/ This record of experience may point to the need to segregate out smaller farm development projects from commercial and industrial ones. 25. In the light of developments arising from the monetary correc- tion problem, and the abrogation of the loan terms by the Government, one wonders why the Bank did not seek to cancel the loan completely, with immediate full repayment when the PRONAZEM program was introduced. Under somewhat similar circumstances this was the Bank's attitude with respect to the livestock projects (Loans 516/868-BR), where an accelerated repay- ment schedule was negotiated, requiring cancellation of uncommitted funds plus full repayment of disbursed funds within two years. One wonders also why the Bank agreed to continue with the appraisal of the CEAGESP sub- project after the introduction of PRONAZEM was confirmed, except as a last attempt to "save" the project. There is a need for some consistency in Bank policy in dealing with instances where Loan Agreements are broken. B. Sub-Loan Documentation 26. The major administrative problem experienced in implementing the project has been related to the "Standing Instructions" which were pre- pared in accordance with Schedule 5 of the Loan Agreement, and distributed to the various branches of the Bank of Brazil within the project area to govern loan application and review. The documentation requirements of the instructions were simply too complex and detailed for the majority of the sub-borrowers which the program hoped to assist, the farmers and small- scale processors and warehousemen. The original standing instructions applied equally to all loan applicants, whether a small farmer seeking a US$1,000 loan for on-farm storage, or a commercial processor seeking US$1 million and more for plant and equipment. Not only did the potential recipients find the documentation confusing and expensive, the officials 1/ It is the Latin America and Caribbean Regional Office's contention that these cases represented the exception rather than the rule and that the agreement was given reluctantly. While it is true that operators of small farms are vulnerable to uncertainties related to production and prices, other means such as crop insurance, price supports, and repayment moratoriums could be more efficient than a special interest rate policy and make nominal (but positive real) interest rates less threatening to farmers. - g - of the Bank of Brazil branches also found the documentation excessive and too complex. As a result the program got underway slowly with very few loan approvals in the first six months of operation. 27. The problem was recognized early by the first supervision mission which visited Brazil within a month of the project's effective- ness. The mission recommended that the instructions be revised and simplified, and specific Bank of Brazil personnel trained in their use. The work was undertaken by Bank of Brazil, which revised the manual of instructions and submitted it for Bank approval in July 1973. It was not until March 1974, however, that the new manual was approved and put into operation, after which sub-loan applications and approvals increased dramatically. 28. The Bank records are sketchy in respect to the reason for the long delay in making an acceptable manual operational. Bank of Brazil officials maintain that the original standing instructions, although developed by their staff and consultants, was based on Bank requirements and of course, had been approved by the Bank. They also maintain the amount and complexity of the documentation required far exceeded ordi- nary commercial loan documentation. The revised manual of instructions simplified procedures and geared the application requirements to loan size. There appears to have been no subsequent problems with it, although Bank records repeatedly indicate that the small-scale farmer was still not being served by the project. Discussions with farmers indicated to the audit mission that the minimum structural requirements under the PRODESAR program were probably too sophisticated for average size farms of 200 to 250 hectares. 29. There appears to have been some weakness in the original prepa- ratory work for the project in setting up acceptable and workable terms of reference for sub-loan operating instructions. What was developed was aimed at meeting all financial and engineering requirements for the larger commercial sub-projects, and did not recognize that these were not applicable to simple and unsophisticated primary storage facilities. In view of the extensive investment by the Bank in money and man-hours of labor in identifying, preparing and appraising the project, there should have been a closer liaison with a new borrower in drafting operational procedures and instructions. This is particularly true where the design of the project cuts across several financial and structural levels of operation within an industry, as did the PRODESAR program. C. Project Accomplishments Physical Accomplishments 30. From the standpoint of physical accomplishment the project can be considered a partial success. It did finance the construction of - 9 - just over 2 million tons of badly needed storage capacity, at signifi- cantly lower cost than estimated at appraisal. This was no doubt because the appraisal had anticipated a greater number of small-scale farm facilities with a relatively high unit cost. The project has been more adapted to the size of facilities required by cooperatives and large- scale farms in the 2000 plus hectare range. Consequently these larger units reflected economies of scale. 31. The typical units viewed by the audit mission were well designed and constructed, efficient, low-cost storage facilities. In the final analysis, however, only 60% of the funds allocated by the Bank were utilized; and of these, less than half were directed toward the priority recipients, the farmers and small entrepreneurs. The project objectives were in effect altered during the course of loan disbursement, to accom- modate the changed circumstancep brought about by the Government's introduction of a subsidized interest program. As a result, a major portion of Bank funds was directed into public sector storage facilities which, while not excluded in the Loan Agreement, obviously carried a lower priority in terms of project objectives. 32. One of the primary objectives of the project had been the modernization of the storage industry, specifically to encourage the switch from bagged to bulk handling of grain. Indeed, a major factor in the economic analysis was based on this conversion. Unfortunately, there are no statistics available to indicate the extent to which this objec- tive was achieved. Bulk handling, in varying degrees, was no doubt accommodated in all of the major construction works, but one can only make estimates of the extent to which handling costs have been reduced through this type of innovation. Financial and Economic Results 33. The audit mission had great difficulty in rationalizing some of the bases for calculating economic return in both the appraisal docu- mentation and the Project Completion Report. The appraisal report does a well-reasoned analysis of the projected economic rate of return, but the major weighting given to "savings in grain losses" does appear sus- pect. The source material used 1/ estimates grain losses due to deficient storage facilities to be 13.5% of total grain handled on farms, and 9.5% in non-farm storage. The project facilities were estimated to produce a saving in this respect of 6% and 4%, respectively. Under the most adverse conditions losses might approach this magnitude, if one were considering total weight loss in all post-harvest operations; field losses, transport, 1/ Study by University of Vicosa (Minas Gerais). - 10 - cleaning out foreign material, drying to normal moisture standards, processing as necessary (dehulling), insect damage, spoilage and storage losses. In commercial (non-farm) storage particularly, no operator could long survive with a 9.5% volume loss on operations. It must be understUUG that in calculating losses for rate of return, one must exclude losses due to cleaning and drying, for allowances are made for such losses in foreign material and moisture when grain is taken into commercial storage. Real losses in commercial storage are likely to run no higher than 1 to 2% of "purchased" volume, according to industry operators questioned - and this assumes traditional flat storage for bagged grain. Farm storage losses are much more difficult to assess because of the great variability in con- ditions. "Marketable" grain is normally not kept for any period of time on small-scale farms, but rather tends to be sold immediately after harvest. The percentage loss in farm storage is likely, therefore, to have been somewhat lower than that used in the appraisal calculations. This view is supported by a second study 1/ utilized in the appraisal report, which shows seed plus handling losses for farm-stored grain substantially lower than utilized for the rate of return. (See Annex 3). In the view of the audit mission rates of return based on half the savings in grain loss would be more realistic than the percentages utilized. 34. The PCR does not question this "savings" assumption in the rate of return analysis, and carries it through for the completion analysis, adjusted only for current prices and actual tonnage constructed under the project. However, where the appraisal calculations use a throughput rate of approximately 1.6 times capacity for commercial installations (one for farm storage), the PCR calculation is based on only one turn-over per year. In an area where double-cropping is normal, this appears an unrealistically low turn-over rate. 35. It is obvious that the high level of "savings on grain losses" utilized for the rate of return calculation gave such a favorable result that other legitimate factors were ignored in the PCR. Savings on bulk as opposed to bagged handling were not used in the calculation, nor was any allowance made for savings due to the reduced need for emergency storage facilities. Unfortunately no statistics are available on the actual proportion of the facilities built which utilize bulk handling. There is no doubt the proportion is high, and the 75% assumption used in the appraisal report seems realistic. Based on current warehousing rates, this would result in a net saving of 25% of the cost per ton on grain handled through commercial facilities (Annex 4). 36. The rate of return as calculated in the PCR does not, in the view of the audit mission, reflect an accurate evaluation of the project. 1/ Getulio Vargas Foundation. See Appraisal Report Annex 8, Tables 3 and 4. - 11 - If pre-project storage losses are brought to a more realistic level of 1.5%, but on the other side allowance is made for higher turn-over due to double-cropping and savings in handling costs the ERR is estimated at about 20%.1/ D. Bank Performance 37. In assessing Bank performance for the Grain Storage Project as a whole, consideration must be given to the frustrations associated with the protracted delays and failures in establishing a viable project in the years between 1965 and 1971. Only when the shortage of storage facilities became critical was there proper problem identification and program coordination by domestic authorities to produce a well-documented loan proposal. Whether or not the Bank could have provided greater gui- dance and direction during this early period is debatable, but the major expenditure in time and money in finalizing a project may well have pro- duced a strong determination on the part of officials to make every effort to achieve final success. 38. The Appraisal Mission produced a detailed and well-documented proposal for a storage program which would embrace all sectors of the industry except seaboard terminals. Indeed the weakness in the proposal was that the program was too broad in scope, the sub-projects too diverse in nature to be amenable to the same rules, regulations and administrative procedures. 39. During negotiations the Bank was firm on the inclusion of an acceptable monetary correction clause, while making concessions on other factors. Nevertheless, the Bank was less firm in insisting that the pro- visions of the clause be immediately instituted when the first supervision 1/ The Central Projects Department is not convinced that the project provided a positive economic rate of return from a with and without project viewpoint. "...that the Bank did not act as a lender of last resort is dramatically shown by the fact that the Brazilian government substituted a competing program and agreed to terminate the loan and by the premature repayments of loans of many private borrowers. (The PCR observes that, "...This reflects prosperity among soybean farmers in particular, which allows them to use their own profits to pay off indexed loans while concurrently taking up unindexed loans.") (PCR para. 2.15). Both the PCR and the audit should pay some attention to this question, particularly in light of the ex post evidence on the capacity of the private sector and the Government of taking care of acute storage shortages timely and adequately." - 12 - mission determined that it should be invoked. As a result of the eight months delay in establishing a new indexing basis, about 50% of the sub- loans actually were granted on a lower real interest rate than prescribed in the Loan Agreement. 40. The Bank acceded also to requests by the Borrower to liberalize the terms under which the Bank had a direct participation in loan approvals. The limit was increased from projects of US$1 million or more as specified in the Agreement, to US$2 million or more, to help minimize and expedite administrative procedures for Bank of Brazil. Similarly requests by the Borrower for relief from standard procurement rules and regulations were granted. 41. In each of these instances the Bank showed a willingness to coop- erate, and a marked degree of flexibility to expedite and facilitate project implementation. The Bank also approved, retroactively, a sub-loan made to a state storage enterprise by Bank of Brazil, which should first have been submitted for Bank approval on two counts; investment value and borrower identity. Finally, after the introduction of PRONAZEM, in clear breach of a major condition of the Loan Agreement, the Bank continued funding the Project for the CEAGESP facilities. Flexibility in adjusting administra- tive details during the course of project implementation is desirable; however, it would appear to be an equally valid proposition that termina- tion of a project may be preferable to retreat from agreement terms which involve matters of substantive principle. 42. The Bank finances a multiplicity of projects in individual developing countries, and this in itself allows some flexibility on the part of the borrower in accepting and conforming to certain Bank objec- tives and conditions. There is the ability to choose for Bank financing, those projects which are the least "politically" sensitive, and financing out of internal resources those where domestic considerations dictate more liberal terms and conditions. To minimize the impact of this on Bank policy objectives requires a high degree of coordination between the various project departments within the Bank's operational structure. 43. Project administration by Bank officials was well carried out for the most part. Both the initial appraisal report and that for the CEAGESP sub-project were detailed and well documented. The early super- vision missions also performed well, and put an early focus on problems. There was unfortunately no supervision from the time the PRONAZEM project was mentioned as being under consideration, February 1975, until September 1976, a period in which roughly one-third of the total sub-loans were approved and disbursed. A mission did visit Brazil in May 1975, to supervise the existing loan, appraise the CEAGESP sub-project and analyze prospects for a repeat project PRODESAR II. On receiving confirmation of the PRONAZEM program, the supervision and analysis functions of the - 13 - mission were abandoned. There is, unfortunately, a resulting gap in documentation of the project's progress, and in continuing analysis of number, size and distribution of sub-loans until the project was effectively completed. 44. In regard to the PCR, it is a matter of judgement whether the CEAGESP sub-project should be credited with the degree of success it has received as an integral part of the overall project and its contribution to institutional development. It is a success in physical terms, will no doubt prove financially viable under conditions of increased production and improved marketing and has been a distinct contribution to institu- tional development. The conditions and standards set by the appraisal have been carried out, or exceeded, and the corporation appears to be operating on a sound management foundation. 45. A review of Bank performance would be incomplete without men- tion of difficulties encountered in receiving full cooperation from the Borrower in reporting procedures required under the Agreement. For what- ever reason, reporting was invariably delayed, and was the subject of frequent communication between the Bank and Bank of Brazil so that moni- toring the project could be carried out effectively. This placed an undue burden on project officers engaged in supervision, because necessary material for analysis was frequently available only during country visits. Some strengthening in reporting commitments seems necessary at levels of senior responsibility. E. Conclusions 46. The main conclusion of the audit is that there was a major error in project design. It attempted to encompass far too broad a spectrum of the whole grain storage industry, ranging from simple and inexpensive farm storage structures to sophisticated and costly indus- trial installations. The credit, management and engineering requirements were too diverse to be compatible in a single program, and because of this and domestic political considerations, the project failed to achieve completely the original objectives. There is no question of the economic need for the project, as the initial demand and subsequent domestic programs have clearly demonstrated. Individual projects directed at farm storage on the one hand and commercial/industrial storage on the other, with somewhat different loan terms and conditions, seem a more objective approach to a successful program from both an administrative and a philosophical standpoint. 47. Broader lessons emerging from this project experience point to the need for a greater degree of coordination within the Bank on: (a) policy objectives in developing projects; and - 14 - (b) reaction to and action to be taken in instances where projects run into difficulty as a result of fundamental philosophical differences between the Bank and the Borrower or in instances of breach or default of agreement terms. There is evidence of considerable variation in treatment on these ques- tions, not only as between countries, 1/ but even between projects within the same economic sector in a country. 2/ The result must be frustrating to administrators on both sides. 1/ Different action was taken in the case of Zambia Livestock Development (Loan 627-ZA), PPAR No. 1362, November 23, 1976 and Venezuela Livestock Development (Loan 807-VE), PPAR No. 1343, November 9, 1976. 2/ Brazil Livestock Development (Loans 516-BR and 868-BR), PPAR No. 2402, April 5, 1979. ANNEX I BRAZIL - GRAIN STORAGE PROJECT Loan Agreement: September 27, 1972. Effective Date: January 16, 1973. Sub-Loan Approvals Progress SUPERVISION REPORT No. of LOANS APPROVED VALUE OF LOANS Accumulative For Period Accumulative (Million Cruzeiros) To: Jan 31/73 19 16.7 July 1/73 26 45 32.9 Dec 31/73 45 90 68.7 Sept 30/74 277 367 152.0 Dec 31/74 75 442 167.0 Sept 30/76 175 617 381.5 - 16 - ANNEX II BRAZIL - GRAIN STORAGE PROJECT SELECTED PRICE INDEXES Year General Price Index Wholesale Prices Prices Paid by Grains and Pulses Farmers 1) Index % Change Index % Change Index %Change 1970 234 132 172 1971 280 19.7 167 26.5 200 .16.3 1972 329 17.5 201 20.4 256 28.0 1973 380 15.5 207 3.0 327 27.7 1974 490 28.9 261 26.1 510 56.0 1975 625 27.6 326 24.9 693 35.9 1976 894 43.0 471 44.5 934 34.8 1977 1285 43.7 615 30.6 1308 40.0 1970-77 549.1 465.9 760.5 Average 28.0 25.1 34.1 SOURCE: National and Regional Economic Indexes Conjuntura Economica, Rio de Janeiro November, 1978 1) State of Rio Grande do Sul. - 17 - ANNEX III BRAZIL - GRAIN STORAGE PROJECT Estimated Losses in Farm Stored Grain (Project Area) Retained For" % of Production Grain Production Seed + Losses Retained for Seed plus Losses (000 tons) (000 tons) CORN 16,419 656.8 4.0 RICE 8,865 797.8 9.0 WHEAT 2,417 265.8 11.0 DRY BEANS 2,049 164.5 8.0 SOYBEANS 2,403 120.1 5.0 TOTAL! 32,153 2005.1 6.2 NOTE: Tonnage of Grain retained for seed not given, but will be a significant portion of Column 2. 1) Production Forecast for 1975 - Appraisal Report Annex 8 Table 3. 2) Estimated grain retained on farms for seed, plus losses. Appraisal Report Annex 8 Table 4. - 18 - ANNEX IV BRAZIL - GRAIN STORAGE PROJECT Grain Handling Tariffs - CEAGESP 1979 CR$ per Metric ton Bulk Handling Bagged Grain WEIGHING 20.00 DISCHARGING 28.00 15.06 PILING 22.34 STORAGE for 30 Days 17.36 34.24 CLEANING 9.10 9.10 DRYING-from 15 to 13% Humidity 72.57 72.57 FUMIGATION 20.00 11.27 LOADING OUT 22.34 15.06 TOTAL 211.71 157.30 Net Saving in bulk Handling = Cr$ 54.41 Source: CEAGESP, Revised Tariffs 1979 NOTE: The above tariffs would represent a minimum cost differential. In actual practise, if bagged grain were to be cleaned and dried, there would be additional charges for opening sacks and rebagging. ANNEX V - 19 - age SECRETARIA DE PLANEJAMENIO DA PRESIONCIA DA REPUBLICA INSTITUTO DE PLANEJAMENTO ECONOMICO E SOCIAL (IPEA) CIA/ASS/SUP-IPIAN/0 c//79 May 11, 1979 Ref. UAPI-5S /79 Mr. Shiv Kapur World Bank 1818 H Street, NW Washington DC 20433 U.S.A. Re: Ccments on the Draft PPAR of Loan 857-BR (Grain Storage) Dear Mr. Kapur, After discussion with Banco do Brasil and other related institutions, our con- ments on the draft PPAR of loan 857-BR are as follows: a) The report is considered well done and representative of the execution of the project. b) Discussion in Banco do Brasil led the staff of this Bank to recommend the following adjustnent in the figures of two pages: = Page (vi) of the Preface: According their figures, the dates of disbursement were a little diferent from yours. So,. they suggest to adjust the column "Actual disbursement" according the attached copy, and to recalculate accordingly the last column. = Page 10: According their data, there are suggested the changes indicated in the at- tached copy. Yours very trut> Frnando'Caldas ASSESSOR DA SUP Copy to Banco do Brasil: Dr. Antonio Chagas Meirelles, Diretor de Controle. FC/rmgsc. IPEA-1i - 20 - ANNEX V Page 2 BRAZIL GRAIN STORAGE PROJECT (LOAN 857-BR) Disbursement Table (US$ million, cumulative) Appraisal Actual Actual % of Period Ending Estimate Disbursement Estimated 12.31.73 0.6 1.7 06.30.74 2.6 5.8 12.31.74 6.0 8.6 06.30.75 9.6 11.8 12.31.75 13.6 17.4 06.30.76 18.1 18.2 12.31.76 22.9 - 06.30.77 27.6 09.30.77 30.0 - 21 - ANNEX V Page 3 /"..mid-year. The funds were to be disbursed over a period of four years for the construction of grain storage and processing facilities, at 8% interest for farm and 15% interest for secondary facilities, both without monetary correction. 22. The institution of this program effectively killed the PRODESAR project, except for the major sub-loan to CEAGESP for public sector storage construction, then under appraisal. The new program caused bitterness amongst sub-borrowers under PRODESAR who were paying far higher interest rates, since "switching" from one program to the other was definitely not allowed. As a result a very high proportion of sub-borrowers paid out their loans almost immediately, using available cash from crops and re-financing from other sources, or by diverting personal funds from programmed projects such as farm expansion, new machinery, seed improvement, etc. 23. The Government was clearly in breach of the Loan Agreement in introducing the PRONAZEM project. It was a decision taken on the grounds of political reality, recognizing the developmental needs of a sector of the agricultural economy. In this context subsequent events have, perhaps justified the decision - there has been a very rapid expansion of storage facilities, up to mid-1978 approximately 1.3 million tons static capacity of on-farm storage and 2.4 million tons of secondary storage under PRONAZEM." 1/ Borrower comments continued from page 19. These have reference to PPAM page 6, para numbers as above.  - 23 - Project Performance Audit Report BRAZIL GRAIN STCRAGE PROJECT (Loan 857-BR) PROJECT COMPLETION REPORT 1. Introduction 1.01 This completion report assesses the Grain Storage Project (loan 857-BR), which was prepared in 1970-71 by a study group under the leadership of the borrower, Banco do Brasil S.A.(BB), appraised in Novem- ber 1971, and executed for the most part in 1973-75. This report is based on analysis of documents related to the project,. including a comple- tion report by BB, and on the findings of a November 1977 mission which visited Brasilia and the two states which participated most heavily in the project - Parani and Rio Grande do Sul. 1.02 As detailed in chapter 2, the pr)ject was highly successful from a technical vewpoint. The project as a.cuted includcd a grcatcr volume of storage facilities at a lower avera;e cost than expected at appraisal; subloans were taken up faster than expected; and its rate of return was higher. In addition, the project iad an unexpected benefit of stimulating institutional improvements at the state-owned storage company in Sao Paulo. 1.03 The project's principal purpose (other than providing necessary storage facilities) was to continue a then-rezent trend towards establish- ment of non-subsidized interest rates in lending to medium-scale and large agricultural and agro-related enterprises. 17, project did not accomplish this purpose. On the contrary, as discussed 'n chapter 3, it was terminated abruptly in 1975 when the government instituted another grain-storage pro- gram with highly subsidized loans, with a number of unfortunate after- effects including: a. further polarization of Bank and Brazilian attitudes on the vital issue of interest rates in development lending; b. financial hardships for many of the subborrowers under the project, who now cannot compete with similar facilities built under the successor program; and c. the Bank's inability to provide further assistance to the public-sector storage companies, which might well have been an objective of a follow-on Bank-financed project. 1.04 The project as finally appraised and implemented is notably different from the project.originally identified in 1965. As discussed in chapter 4, the initial preparrtion was badly managed by all parties. It would he hoped that the Bank would avail itself of the periene gained from thi- project's preparation by identil-ving preparation difficulties early and correcting them. - 24 - 2. The Project Purposes 2.01 The project had two principal purposes: to overcome an acute shortage of grain-storage capacity which had developed in the nine-state project region, and to strengthen the establishment of real positive inte- rest rates in lcnding for agricultural and agro-related development. As summarized in the following paragraphs, it was very successful in achieving its physical purpose. However, it did not achieve its broader purpose, as detailed in chapter 3 on the interest-rate issue. 2.02 Three secondary objectives of the project were enumerated when the project was re-identified in mid-1970 (see chapter 4 for discussion of the initial and revised preparation): a. stimulation of more efficient (i.e. bulk instead of sack) storage and handling; b. making services of publicly-owned enterprises more acceptable; c. updating of then-existing laws which were said to inhibit grain handling by commercial concerns. While the appraisal report stated that "bulk handling of grain would be introducad under the projcct whenever feasible", no specific loan provisions to promote such introduction were recommended, nor were data collection pro- cedures employ:_i to monitor the portion of the loan committed for bulk storage. Instead of trying to improve public-sector enterpriscs, the pro- ject as appraised set assistance to private-sector grain storage as a prin- cipal aim (see chapter 4 for further discussion of this issue). Because of the extent of improvements which might have been necessary at some public sector companiej, the decision to concentrate on the private-sector companies was correct for an initial project in the subsector, leaving the more ambitious task of overall subsector imorovement to future projects. The then- unforeseen very large subproject or the state-controlled Companhia de Entre- postos e Armazens Gerais de Sao Paulo (CEAGESP) addressed both the bulk-hand- ling and public-sector issues successfully by providing for institutional improvements by means of a subloan for large bulk-storage facilities. The third possible objective,legal improvenents, was originally recommended by the appraisal mission but dropped prior to negotiations. This turned out to be a wise course of action, as project execution was not impeded by the legal frame-w9rk in effect. Physical Dimensions 2.03 The project was intended to provide subloans on a 50-50 basis with BB to finance 80% of about 800 grain-storage subprojects in nine states of southern and south-central Brazil. BB established a specific line of credit (Projeto de Desenvolvimento da Estrutura de Armaz.,nagem - PRODESAR) to handle subloans. The stated intention was to provide about 1.9 million - 25 - tons, or 30-40% of anticipated needs (5 to 6 million tons) through 1975. The project as implemented included about 600 subprojects with installed capa- city of about two million tons. 2.04 Annex 1 makes an actual-vs-appraisal comparison of subproject distribution by state, identity of beneficiary, grains to be handled and type and size of investment. There are some significant differences: more farmer participants and fewer cooperatives/grain processors; many more soybean-handling and fewer corn-handling facilities; fewer snall invest- ments and many fewer expansions cf existing facilitiCs resulted from the project than were estimated as a result of the BB survey. Actual distribu- tion of number of projects and storage capacity by state was significantly different from estimated: Parana was a much stronger participant, suggesting that shortage of grain-storage facilities there was even more acute than foreseen; and Sao Paulo had only 28 projects instead of the 181 foreseen, which suggests that the BB survey did not reflect the dominance of CEAGESP in that state's grain-handling subsector. Surveys of the nature undertaken at best provide only limited insight into the actual needs of whatever sub- sector is being surveyed, particularly when enormous areas are involved; despite the above-referred discrepancies, the BB survey offered a reasonable guide to overall storage needs in the project area. 2.05 As shown in annex 2, the actual cost distribution among civil works, machinery, engineering and studies was not significantly different from the estimates (exccpt for one group which constituted less than five per cent of the project). It is likely that vuch of the engineering costs were in fact absorbed in the reported cost of civil works, which would reduce the apparent differences in cost distributions. One interesting comparison is the actual cost of small (mostly on-farm) facilities - US$34/ ton capacity - with the estimated US$62. The average capacity of the small units actually built was much larger than estimated (about 1,200 tons vs. 300); the lowev cost/ton reflects economies of scale. Medium and large- scale farmers consider large on-farm storage to be absolutely essential, because the demand for transport during harvest season far exceeds the supply. Farmers with storage facilities can harvest their crops efficiently, i.e., without waiting for transport, and later move their crops to intermediate storage facilities when transportation is more readily available. 2.06 Even though the project consisted of three-quarters of the number of subprojects estimated, its total cost was only 6% below appraisal. estimates, excluding land (see annex 3). As noted above, the average sub- project size, particularly of on-farm units, was larger than expected, and the average cost/ton capacity was lower. 2.07 In surmary, the project accomplished its physical objective, and it did so ahead of schedule, as discussed in chapter 3. The differences in actual-vs-estimated distribution of subprojects reflect possible deficien- cies in the survey details, but a more accurate survey would not have had significantly different overall results. Proj nc- .:ccutico -,)c lc'c n d:L::i:u c:alized in early 1974. 2.10 As is standard practice for this type of project, the loan agreement included covenants requiring Bank approval of subloans for any large subprojects (in this case, for any subproject costing US$1 million or more). BB proposed to increase the "free limit" to US$5 million, and the Bank agreed to raise it to US$2 million and liitit its period of review to three weeks. Except for public-sector subprojects (Bank approval for which was required under differ,nt covenants - see para 4.14 et seq.), no further subloan approvals by the Bank were necessary under the higher free limit. 2.11 PerformancR of the bor-ower re the project. While DENOP was at t5mes overloaded with other wor., ics response to problems was satisfactory. Once the initial difficulties concerning the project manual were overcome, subproject preparation and approval met or exceeded appraisal schedule (para 3.01). Realizing that DENOP did not have the technical expertise neces- sary to evaluate large subprojects, BB fo=med another unit (COTEC) for this purpose. While BB forred a General Coordinating Committee (GCC), whose for- mation was required as a condition of effectiveness, the GCC did not coordi- nate the project as envisioned by the appraisal report. In fact, it barely functioned. While an effective GCC might have overcome project start-up - 27 - difficulties and facilitated training of bank managers, .DENOP managed for the most part to execute the project successfully, albeit with som2e delays. There is some question as to whether an already necessarily complex organiza- tional structure such as BB needed a GCC for one particular project, which was relatively small in relation to BBs overall size. 2.12 Overall nerfornance of the borrower. BB maintained its positi.on- as Brazil's pre-eminent agricultural lending institution during the project preparation and execution period. Year-by-year statistics of its lending volume (annex 5) and branches/employees (annex 6) indicate steady and signi- ficant expansion; its growth in percent is sumarized below: OVERALL GRC7TH IN PERCENT DURING: project execution total period period (1972-1976) (1967-1976) Number of branches 30 71 Number of employees 59 79 Number of loan! - crops 57 130 - livestock 48 232 - industrial 53 193 - total 55 146 Amount of lendi.ng (in 1970 Cr$) - crops 297 796 - livestoch 260 890 - industrial 304 1,033 - total 292 852 As measured in number of loans per employee per year (see annex 6), BB in- creased its efticiency notably from 1966 to 1572 (from 11 to 16 loans/employee) but efficiency has stabilized at that level since then. 2.13 BB's condensed income statements and balance sheets are shown as annexes 7 and 8. Because it operates a number of programs for the government, many of which are not subject to monetary correction or involve loan subsidies of some manner, and maintains a numbcr of government deposits and special funds - many of which are interest- and correction-free - ratios nornally used to measure banks' financial condition (e.g. loan/deposit, asset/capital, and profit/capital) are not entirely appropriate. Taken as a broad measure of pro- fitability, BB's operating profit (before provisions for bad debts and other contingencies) as a per cent of total assets has improved markedly - from 5.5% in 1967 to 15% in 1975, from which it declined to 10% in 1976. Project termination 2.14 As detailed in para 3.11, the government in early 1975 established its Programa Nacional de Armazenamento (?RONAZEM) for the pur"ose of stimulat- ing further construction of grain-storage units through subsidized, i.e. unindexed, subloans to farmers and intermediaries. DomaIId for PRODESAR subloans - 28 - disappeared, and BB cancelled about US$6.256 million of loan funds, represent- ing amounts uncommitted and not earmarked for the CEAGESP subproject. As shown in annex 9, take-up of subloans under the subsidized program has ex- ceeded that under PRODESAR. 2.15 The indexation of PRODESAR loans has caused two divergent effects: premature repayment of loans by an abnormally high proportion of beneficiaries, and recovery difficulties on many other subloans. As shown in annex 10, 22% of subloans under the project had been paid back by the end of 1976, and it has been reported that additional prepayments have been made subsequently. This reflects prosperity among soybean farmers in particular, which allows them to use their own profits to pay off indexed loans while concurrently taking up new unindexed loans. At the other end of the scale, commercial enterprises which have to compete with similar facilities built with subsi- dized financing are not generating enough operating income to service indexed loans. As a result, BB has been experiencing delays in collections, as indicated in annex 11. 2.16 Because of the premature subloan repayments, BB had accur=ulated excess funds in the PRCDESAR account and requested a second cancellation of US$5.52 million in September 1976, bringing the net loan amount to US$18.224 million. This results in major variations in the project cash flow: higher cash surpluses through 1976 because of premature subloan repayments; and conversely, lover cash surpluses for most of the rest of the subloan amorti- zation period (see annex 12). Project j!stification 2.17 The primary justification for the project was the deficit of grain-storage facilities in the project area. This deficit is summarized in annex 13. Year-by-year grain production is detailed by state on pages 1 and 2. Production in 1975, summarized by grain and state on pages 3 and 4, shows the dramatic growth of soybean production, which more than offset de- clines in other storable crops. Gross non-farm storage requirements as of 1975, summarized on page 5, show that production of corn, dry beans and rice was below estimates but that wheat/soybean production exceeded estimates, re- sulting in overall gross requirements about 700,000 tons or 4% higher than estimated, Following the same methodology as used in the appraisal report, the gross 1975 deficit in storage would be set at over 7 million tons (offset to some degree by apparent excess capacity of about 2 million tons in two states) if no new capacity had been installed after 1971, as shown on page 6. The 2 million tons of capacity provided by the project achieved the pur- pose of filling about 40% of the storage-capacity deficit. 2.18 In assessing the project's rate of return, the appraisal compared the installation cost of the subprojects with the following benefits: a. reductions in grain-storage losses, which had been estimatod at 4% and 6% respectively of non-farm and on-farm volumes of grain, based on a then-recent study of actual losses done by the University of Vicosa (Minas Gerais) and on experience - 29 - with well-managed storage units; b. reductions in operating costs resulting from bulk handling as opposed to sack storage, assuming that 75% of the project's capacity would be in bulk; c. savings from not having to build or provide makeshift storage facilities. Under the most-likely assumptions, the estimated rate of return was shown to be 25%; under separate assumptions of a 25% reduction in benefits and a 15% increase in costs, the rates of return were 18% and 21% respectively. 2.19 A number of factors contributed to increasing the rate of return substantially for the project as implemented. Among these factors were: a. much higher prices for all the commodities to be stored by facilities constructed with project funds - prices, which are detailed on page 2 of annex 14, were at least double those es- timated at appraisal; b. lower investment cost per ton capacity, as whown in annex 2 (S$33 vs. US$37/ton); c. a higher proportion of on-farm storage, where loss reductions are 50% higher than for non-farm storage; d. faster-than-expected project implementation, which accelerated the benefits. Because the extent of bulk-storage-facility construction could not be estimatnd readily (para 2.02), no benefits resulting from lower bulk-storage operating costs were included in assessing the rate of return of the project as imple- mented. Even without this series of benefits, the project's rate of return is over 100%. Estimated benefits from not having to construct makeshift facit- ities, while real, are difficult to quantify; the appraisal report related such savings to CIBRAZEM's sack-storage rates without explaining the relation- ship. If those were also eliminated from the comparison, the only remaining benefit would be the reduction of grain losses. Based on this comparison alone, tne project's rate of return (using 1974-75 prices for grain) was 38%. Cost and benefit streams and grain loss savings are detailed in annex 14. 2.20 The rate of return computed for the project as implemented should be regarded as a minimum rate of return, because it does not: a. include the two significant benefits which are difficult to quantify, as discussed above; b. allow for double-cropping, i.e. the double use many facilities undergo in wheat-soybean and corn-dry bean cropping patterns. Such allowance would of course increase the benefits from reductions - 30 - in grain losses; c. shadow-price foreign exchange. However, if the grain-loss benefits were reduced, the project's rate of return would still be acceptable, as the following summary indicates: Reduction Project in project rate of benefits return 10% 33% 20% 28% 30% 23% 50% 14% In summary, the project as implemented was economically beneficial to the country. 3. The Principal Issue: Subloan Interest Rates Introduction 3.01 Mcre than enough has already been written (especially from the macro viewpoint) about the fundamental difference between Bank and Brazilian attitudes concerning interest rates to cormercial farmers, which was the principal project issue and cause of the lcan's premature termination. In support of its subsidization policies, Brazilians have noted spectacular achieve- ment in increased agricultural production and exports purportedly as a result of stimulation through interest-rate subsidy. Bank papers, noting that growt'i in agricultural credit has far exceeded growth in agricultural production, have reached the inescapable conclusion that stbstantial substitution has occurred. 3.02 Much less has been written from the micro viewpoint, i.e. that of the individual farmer (in this case, the grain farmer). While he under- stands the concept of the Bank's attitude concerning interest rates, he also is keenly aware of factors which he uses to rationalize interest subsidies: that certain costs which affect the competitiveness of his products are higher in Brazil than in other grain-exporting countries; and that the government tends to impose additional financial burdens on him which may limit his ability to pay full monetary correction. At the same time, government price supports are very low. All of these factors result to some degree from government - policy: high costs of inputs, at least partially from protection of Brazilian manufacturers; high costs of transportation, at least partially from deficien- cies in the public-sector rail system and consequent dependence on trucks. Of even greater concern to the farmer are those issues directly between him and the government: low minimum prices, tax increases, special levies and lending terms. For example, during the project execution period, the government doubled its tax (for FUNRURAL) on gross receipts of farmers and in 1977 esta- blished an export tax on soybeans and derivatives which varied between 7 and - 31 - 12% of FOB values; this tax was passed back to the farmer. The Bank might well re-examine the financial situation of representative commuercial farmers to see if the combination of increased taxes, special levies and non-subsi- dized lending would not impose an undue financial burden - especially when prices are high - while the present low price suports do not provide adequate protection during periods of low prices. If this were found to be the case, the Bank, in any future commercial-farmer lending operations in Brazil, might consider allowing commercial farmers to reduce their monetary-correction/ interest obligations by any special levies or increased taxes unforeseen at the time of appraisal. 3.03 With respect to the grain storage project, the following para- graphs trace the issue's development and analyze the rationale for the action taken. Development of the Issue 3.04 When the project was initially identified in 1965, agricultural credit bore an interest rate of 7.5% p.a. when price inflation was 90% p.a.; cumbersome administrative devices had to be used to ration credit. At the time, Banco Central (BACEN).officials expressed some hope to begin introducing some form of indexation in all loans, recognizing the need for differentia- tion between small and large beneficiaries' capability to service indexed loans, especially when commodity-price increases did not match general price- level indices. 3.05 Little more was discussed concerning interest during the initial preparation because of more fundamental problenos concerning preparation it- self. The Brazilian working group for the revised project proposed prefixed interest rates in early 1971, but the Bank declined to respond to the pro- posal until the working group had discussed it with the Minister of Finance, who understood the importance the Bank attached to positive interest rates for commercial agriculture. By the time the working group had finished most of its work, the Bank, in its penultimate preparation mission, was apparently able to reach agreement in principle with the working group that loan indexation to achieve a real rate of interest of about 6% for subloans under the project would be feasible. This was further defined by the final preparation mission,which obtained the working group's tentative agreement with its proposed formula (indexation tied to the foreign exchange rate plus 6% p.a.). In a survey of prospective subborowers, the mission found that half would accept the formula, half would consider it, but none rejected it out of.hand. Some Brazilian officials were to continue suggesting a pre- fixed interest rate, however, in conversations with other Bank personnel. 3.06 Within the Bank, considerable discussion of the formula to be used for indexation took place. Bearing in-mind: the difficulties which had initially been encountered in the Livestock Development Project (Loan 516-BR) because subborroLers were slow to accept the only monetary-correction formula available; and the significant increase in subborrower participation when greater flexibility was introduced, the agriculture projects staff proposed giving subborrowers a choice of four indices to revalue the principal loan amounts due: exchange rate between Brazilian and US currency; weighted whole- sale price index of the commodities involved; general price index; or general - 32 - agricultural wholesale price index. Because of a lack of a firm Bank position, the appraisal mission was not able to discuss the issue with Brazilian author- ities, during field appraisal. By the yellow-cover appraisal report stage, indexation based on US/Brazilian exchange rate changes (following the Brazilian preference) was recommended as the Bank negotiating position; but with an escape clause, viz that if there should be an increase in the domestic whole- sale price index which for any appreciable period (later defined as 12 months preceding a Monetary Correction Date) exceeded the rise in the US$-Cr$ ex- change rate by more than 7.25%, the Bank and BB would consult with a view to modifying the indexation system to ensure that the real re-lending interest rate was not negative, and that subloan commitments would be suspended until such modified indexation system had been agreed upon. 3.07 Discussion within the Bank continued to focus on the interest- rate formula and escape clause. Project advisory personnel in particular feared that the real interest rate under the formula could be as low as zero percent, which would encourage subborrowers to draw funds from subloans under the project and concurrently invest their own funds in treasury bonds, whose indexation followed price inflation more clQsely. On the other hand, country p m9ams personnel objected to the escape clause - as being potentially provocative because it implied that the government would change its exchange- rate policy and because there was no alternative formula in case the escape clause were invoked. 3.03 The pioposed formula was accepted by the Brazilians during negc- tiations; BB representatives welcomed it as good banking practice and.in fact suggested that the relending rate be raised from the 7.25% p.a. proposed to 7.5% p.a. to which the Bank agreed. 3.09 Necessity for considering application of the escape clause became evident to a January 1974 supervision mission. During 1973 the in- crease in general price levels in Brazil was 15.7%, while that of US dollars in terms of cruzeiros was nil. This was caused to some degree by weakness of the US dollar in international monetary markets; Brazil's cuZrency was affectively devalued by about 9% vis-a-vis the aggregate currencies of her trading partners. Because the recently-installed government had had little time to focus on such complex issues as indexation and real interest rates and because the agreed-on formula might work except when the US dollar was under pressure, the Bank suggested that the indexation formula be changed to one based on national treasury bond (ORTN) prices (used in subloans under the agro- industries credit project - Loan 924-BR) but allowed commitments to be made under the existing indexing system through June 30, 1974 and left the door open to further analysis and discussion of the matter. 3.10 In July 1974, however, it was evident that the relationship be-- teengeneral price increases and exchange rate variations had not improved: Period ending June 30,1974 Six-month Twelve-month Increase in general price levels 22% 31% Increase in value of US dollar 9.6% 11.7% - 33 - The Bank then urged more strongly on-BB the adoption of in dexing based on ORTN prices plus 5% p.a. interest, and BB had little alternative but to accept. Since BB had a number of subprojects already approved and required time to notify its branches of the changes,the Bank agreed to _xtend the old-formula indexation to subloans approved by BB through September 19,1974. 3.11 As the first grain storage project was progressing smoothly, BB's president requested a second grain-storage loan, and the Bank agreed in October 1974 to appraise a second project; a November 1974 supervision mission outlined feasibility-study requirements and indicated possible appraisal in early 1975. However, a February 1975 supervision mission learned from BACEN that the government was intending to establish a new program to promote construction of grain-storage units where the need was moct acute, viz wes- tern Parana, southern Mato Grosso and Goigs. Brazilian representatives also raised the question of whether the Bank would consider limiting subloan monetary correction. In a meeting with the Bank in Brasilia on May 1, 1975, BB's president indicated that it would be difficult to continue making grain- storage loans under the revised interest-rate formula (5% p.a. plus full ORTN correction) but requested that the Bank reconsider returning to the exchange-rate formula for the proposed second loan. By the time a Bank mission(4ppraisal) arrived in mid-May, however, a new government program - PRONAZEM - had been established to make loans for on-farm grain storage at 8. p.a. and for intermediate storage at 15% p.a., both without indexation and theref3re bearing highl) negative interest rates; BB and the government confirmed tha: they were no longer interested in a Bank-financed seccnd project. 3.12 Fxcept for the very large CEAGESP subproject, (which, being public-sector, did not then qualify for PRONAZEM funding) demand for subloans under PRODESAR evaporated because of PRONAZEM's much lower re-lending interest rates. The Bank requested BB to cancel loan funds which were then uncommitted after providing for the CEAGESP subproject. BB, after soliciting reconsider- ation of this request, complied. Analysis of the issue 3.13 After considerable dialogue with Brazilian authorities concerning interest rates for agricultural lending (which had generally been negative), the Bank was hopeful that the loan would serve to expand the practice of making agro-related loans at real positive interest rates. The project's pre- lature demise and the government's reversion to highly negative interest rates in the successcrprogram negated a principal underlying purpose of making the loan: fortifying the trend towards establishment of real positive interest rates in agricultural lending. The differences between the two programs are also causing distortions among Brazilian grain-storage enterprises; in the case of a typical commercial PRODESAR beneficiary, interest and monetary correction under the project subloan exceed operating profits of the enterprise. To compete with other storage enterprises (whose facilities were built with subsidized funds and whose revenue requirements to cover operating and financial expenses are therefore lower), PRODESAR-funded enterprises have little choice but to set their rates at levels which are too low to cover all their financial expenses. - 34 - 3.14 To a degree, the Lank did accommodate the Brazilians by allowing commitments under the exchange-rate formula from January through mid-September 1974, when it could have cut them off in January. But the question of whether a change in the indexation formula was necessary at all remains. 3.15 There can be little disagreement that the ORTN device is concep- tually preferable to an exchange-rate formula; the ORTN follows price infla- tion more closely, reduces possibilities of substitution and is theoretically less subject to exogenous influences, i.e. external currency fluctuation and governmental policies. The proper time for establishing the ORTN formula (or trying to, anyway),however, would have been during the appraisal and negotiation process, not midway through project execution. 3.16 In the internal dialogue concerning indexation, it had been pointed out that the escape clause could be invoked in the event (then considered unlikely) that the US inflation rate exceeded 7.25% p.a. even if Brazil main- tained its crawling-peg exchange-rate policy. As a consequence, the Bank recorded in a post-negotiations memo its intention not to suspend commitments under the loan if the escape clause became applicable "owing to acceleration of inflation ii the USA". While this memo has no official standing, it must be presumed that the Bank's intcntion was coamunicated orally to the Brazilians negotiating the loan. In its analysis of the reasons for the non-devaluation in 1973 (para 3.09), however, the Bank noted that weakness in the dollar in international uarkets (which reflected inter alia US inflation rates vis-a-vis those of other countries) was a major reason for the stability in the US$-Cr$ exchange rate. As shown below, the unexpected event of accelerating US inflatio did in fact occur: Annual inflation rate as Year measured by average CPI 1971 4.3 1972 3.3 1973 6.2 1974 11.0 1975 9.1 1976 5.8 There is no record to show that the Bank considered its above-quoted intention during its deliberations on invoking the escape clause, nor is there any evidence that BB reminded the Bank of that intention. 3.17 Retrospectively, the increase in the ORTN rate has been only marginally higher than that in the exchange rate since 1974. As shown below, both the exchange rate and ORTN adjustments were within 7.25% of general price increases in 1975 but would have fallen outside the prescribed range in 1976- 1977. - 35 - Year-end to year-end change in Cruzeiro General price value of ORTN index for US$ coefficient domestic supply 1974 17.5 33.3 34.5 1975 22.0 24.2 29.2 1976 36.1 37.2 46.4 1977 29.9 30.1 39.3 3.18 Speculation on the probable course of action of the Brazilian authorities if the Bank had been more flexible about invoking the escape clause and insisting on substitution of the ORTN formula is difficult, but the record does show that they attempted to propose indexation formulas which would fall short of full monetary correction but still continue the concept of ex post indexation. It is quite possible, of course, that those attempts were merely ruses: that if the Bank had been more flexible, the Brazilians might have tried to erode a less rigid Bank position even further; or that Brazilian authorities had reached a policy decision to revert to non-indexa- tion in any event. 4. Other Issues: Project Preparation and Public-Sector Enterprises Project Idantification and Prevaration: Initial Effort 4.01 A storage project was first identified by an economic mission ir May 1965 which found that inadequate storage and handling facilities hampered the export of agricultural commodities, particularly corn. At the time, it was proposed to channel funds through the recently established Banco Central (BACEN), which was undertaking to create an agricultural credit department. 4.02 The Bank's first preparation mission (August 16 - September 4, 1965), which included a commodities expert from FAO, resulted in the establish- ment of a working group, with representatives from the national development bank (BNDE), the federal supply (SUNAB) and storage (CIBRAZEM) agencies, the Bank of Brazil's (BB) rural credit department, and the ministries of planning and foreign affairs. By November 1965 it was established that BNDE would nave primary responsibility for the project, which was to include: a. farm-storage facilities to be financed principally by BB and commercial banks; b. intermediate storage facilities to be executed by state-owned and private companies and cooperatives, with financing respon- sibility to be defined; c. port facilities to be executed by port authorities, state- owned or private companies,and to be financed by BNDE. - 36 - 4.03 Terms of reference for separate feasibility studies for each of the three components were drafted with Bank assistance during the Bank's second preparation mission (November 6 - 19, 1965). The intention at that time was for the economic departments at BNDE and BB's rural credit department jointly to conduct the farm-storage study, for a local firm to study the inter- mediate storage needs and for a foreign firm to do the terminal study (including preliminary engineering), which by that time had been limited to the port of Santos. 4.04 It soon became apparent to the Bank (in particular, to the Area Department) that project preparation was not proceeding as quickly as originally foreseen. BNDE, whose experience and interest lay with industrial development, was not in fact coordinating project preparation: a year after the definitive meetings in November 1965, BNDE had not appointed a full-time coordinator for the project. Little or no substantive output occurred between Bank (or Bank/ FAO) preparation missions, the next three of which visited Brazil in March and June 1966 and January/February 1967. 4.05 Farm-storage component. The proposed joint BNDE-BB study of farm storage never started. In mid-1966 the Ministe: of Agriculture established a special office, consisting of staff members of BNDE, CIBRA1EM and the planning ministry, to carry out the farm-storage survey., but there is no record of this working group ever having produced a study. Early in 1967 it was decided to have the farm-storage study undertaken by the Iural Agricultural Institute of the University of Minas Gerais. However, no f:rm deadline was established fo. the study's completion, and even by late 1967 INDE was not able to predict when it would be ready and so proposed to limit the project to intermediate and port storage. 4.06 Intermediate storage component. Delays in starting the interme- diate storage study were almost as long as the farm-storage one. A contract with a local consulting firm (Servicos de Planejamento-SPL) was not signed until September 1966, and despite early hopes for quick completion, the SPL study was not available in English until January 1968. Worse than the delay was the quality of the report itself. While it did amass considerable infor- mation concerning production of storable products in the project area, its determination of storage requirements was faulty on several counts (failure to measure the impact on intermediate storage of developing on-farm and terminal storage, failure to consider the probable availability of coffee-storage faci- lities or to assess inter-regional transfer), it contained no specific invest- ment program for storage facilities and provided no economic justification for intermediate storage. In short, the SPL study did not provide the basis for a Bank appraisal. Despite subsequent explanations that the study was intended to be an interim measure (to be used in,project definition after review of its findings), the Bank in 1968 informed the Brazilians that the need for intermediate storage facilities was not demonstrated by the study and sus- pended consideration of that project component (ironically, at the very time when the real need for such facilities was becoming acute). 4.07 Port Storage Development. The feasibility study for the Port of Santos terminal was signed with the U.S. firm Weitz-Hettelsater and its Brazilian associate Geotecnica in February 1967, more than a year after project - 37 - definition. The Weitz study, completed in May 1968, was found by a Bank consultant (Cargill of Minneapolis) to have a number of technical defi- ciencies. To make matters worse, the site of the proposed terminal within the Santos port area remained in contention between different Brazilian agencies. These differences were not completely resolved when :he site was appraised as a subproject within the Port of Santos project. Thus ended the initial preparation stage of the project. Six Bank or Bank/FAQ prepa- ration missions, a number of Area Department missions, man-months of Bank effort and probably man-yearsof Brazilian effort had resulted in a relatively small and still incompletely-defined subproject within a port project. 4.08 Project Preparation: Rebirth of a Project. The project was resuscitated in May 1970 as a result of a visit by Brazil's Finance Minister to the Bank. By that time, the shortage of storage facilities had become acute; harvested crops were being stored in the open,protected at most by plastic covers. The Ministry of Finance then recruited a CIBRAZEM official (who had been in charge of salvaging the SPL study) to organize an inter- agency group to formulate an intermediate-storage project, consisting of a credit program aimed mainly at the private sector to construct storage facil- ities; the first phase was to cover 20 - 30% of the storage requirements identified in the SPL study. 4.09 By early 1971, the working group (which consisted, except for the Chairman and a BACEN representative, of BB personnel) had completed a survey in the 9 states (Rio Crandc do Sul, Santa Catarina, Parani, Sao Paulo, Rio de Janeiro, Espirito Santo, Minas Gerais, Goigs and Mato Grosso) which had accounted for 95% of storage-facility loans in 1969-1970 and were therefore defined as the project area. 4.10 By March 1971, except for the preparation of models of typical facilities expected to be financed and calculations of economic justifica- tion of the project, the preparation was complete. BB, the obvious choice because of its pre-eminence in agricultural lending throughout Brazil, was selected as the executing agency. By June 1971, the project was practically ready for appraisal; field appraisal took place on October 28 - November 23, 1971. Observations on Project Preparation 4.11 With the benefit of hindsight, a number of observations con- cerning the project's preparation are evident. Reasons for the selection of BNDE as initial project-preparation coordinator are unclear, but the results of that selection were predictable since that institution had very little experience in agricultural or agro-related lending. BB, on the other hand (which by virtue of its pre-eminent role in agricultural lending in Brazil and its consequent familiarity with agro-related needs should have had a major role in preparation), participated only to a very limited extent in the initial project preparation. Once BB's.experience was properly utilized (i.e. in the revived project preparation in 1970-1971), a well-documented and thorough feasibility study emerged as the basis for project appraisal. 4.12 For its part, the Bank - especially the projects personnel - failed to control project preparation ir the early stages. Terms of reference, which the Bank helped draft, were admittedly incomplete, but no follow-through was undertaken. The Bank established no written schedule for study completion with BNDE, nor did it urge prompter execution of prep- aration on the coordinating agency in writing. Once the project was revived, no control of this nature was necessary. 4.13 To avoid repeating the experience, the Bank should: ensure that BB is an active and enthusiastic participant in (although not necessar- ily in charge of) any project involving agricultural er rural credit; establish detailed programs, including timetables for study completion, of any project, and be more objective in assessing the capability of the agency(ies) involved in project preparation, particularly the projects which the government merely finds acceptable (as opposed to those for which the government is especially anxious, as was obviously the case for the revived project). Public Sector Storage Enterprises 4.14 In the project's initial preprat:.on, including the SPL study, private-sector participation was encouraged, and suggestions were made that public-sector investmeat in the indstry shouLd be limited. Much later, the appraisal mission found that many public-sectcr storage units had been under- utilized and were in financial difficulties, principally because of substan- dard management, i.e. poor marketing, lack of perception of grain-flow mecha- nisms and producer needs, slow response to changing patters of grain produc- tion and transportation, inappropriate locati-n of new units. While the stated aim of the project was to assist private-sector storage developments, the loan agreement did provide for financing of piblic-sector companies, subject to Bank approval. 4.15 Subprojects of three state-owned companies were submitted for Bank review. In the first case, Bank approval of a subloan to the Minas Gerais storage company (CASEMG) was delayed by about ten months because the subborrower did not provide necessary information. In that case, the Bank could have de- clined to participate because BB had made the subloan commitment prior to loan. signing. The second request, for a subloan to the Rio Grande do Sul Company (CESA), was rejected because the company had already obtained financing from other sources. The third subproject was that of Companhia de Entrepostos e Armazens Gerais de Sao Paulo (CEAGESP), which was proposed in June 1974 to install new capacity of 320,000 tons at a cost of US$26 million, of which about US$10 million was to come from the Bank. 4.16 Because of the size of the CEAGESP subproject, two Bank missions visited Brazil principally to review and appraise it. While the findings of the first mission were negative, principally because the company's tariffs were too low and financial condition unsatisfactory, the door was left open for project modifications and tariff increases. As a result, CEAGESP scaled down the project somewhat (to 240,000 tons), and the Bank sent a full-scale mission to appraise the revised subproject in May 1975. The mission, while approving technical features of the four units (which were exclusively for bulk grain storage) cstablished the following subloan conditions: an imme- diate 50% tariff increase; subsequent annual tariff increases based on indices, an immediate contribution by the state to the company's capital, and adoption of a salary structure suitable to attract and retain competent staff. CEAGESP, which installed a new top-management group during Bank consideration of the subproject, complied with or exceeded the conditions. 4.17 As indicated in annex 15, page 1, the construction of all the new CEAGESP units ran behind schedule, principally because of bad weather and poor subsurface conditions. These delays and a number of contract exten- sions or additional works (many of which might have been anticipated when the subproject was prepared and appraised) caused a cost overrun of 37% in nominal cruzeiros as shown in annex 15, page 2, or somewhat over 20% in real terms. 4.18 Grain flows through CEAGESP's bulk facilities were approximately the same in 1976, when the subproject was substantially completed but still in its break-in period, as in 1974 (annex 15, page 3). Evidence from future years will be needed to verify if the -subproject's purpose - more efficient grain handling through the use of large bulk facilities - was achieved. 4.19 The tariff increases put into effr:ct at the beginning of 1976 were sufficient to provide the company with operating profits and net in- come for the year, in contrast to the previous year, when substantial opera- ting losses were sustained(annex 15, page 4. However, the company's financial rate of return (net income related to year-end equity capital of Cr$ 673 million) was less than 1%, which signifies some subsidization. 4.20 By improving CEAGESP's financial situation (which, with more responsible management and staff, would hopefully be continued in future years) and by calling the company's attenticn to the need for astute analysis of grain flows and rotation to assess the reed for new installations, the Bank was able to make a significant institutional contribution through the project. 4.21 Technical assistance and institution-building of this nature could have been extended more systematically to the grain-storage subsector in general and the publicly-owned portion of it (about 45% of the total) in particular through a second project. At the time a second project was under initial consideration (in late 1974), CIBRAZM's operations and functions were under review, and a closer collaboration among CIBRAZE4, BB and Bank staff in evaluating future grain production and storage needs was envisaged. Possibilities of institution-building of specific publicly-owned companies were not explored in the short time the second project was under considera- tion, but they would have been difficult to ignore during the appraisal process. - 40 - 4.22 In summary, the Bank made a positive contribution towards insti- tution building with respect to the public-sector portion of the industry, which was an unanticipated project benefit and fully justified the covenant requiring Bank approval of public-sector projects, but greater involvement in subsector rationalization was precluded by the project's abrupt ter- mination and lack of Bank participation in a follow-on project. .，二J一j讓奮q個卜讓一I讓•1：】1,hl》甲；唔 &..”。鳥｀馴：&&，叩’&&,&&,&&,jlu”。∥，”•萬“0打叫．r”。規∥。”.’。，&&.u&&j.&&1&”。’&,’亡”、吧。，&&&，。化“八 口一－---－一一一－－一一一一一一一一－一一一一一一一一一一－一一一一－ 藝 ::::：∥訕：&:l,：訕：.．一：斗’；二！訕：認！訕一：訕：.,:：么訕：么一訕。＿一桌翼．。 茄下闐t面不間1.&I「蒲－盧t二。祇～，讓〔。訂電c嗎一砍、．[t一需01蘇署“1一了不一，‘必i，。石～肥。•佻一疋可必t、‘痲蒲于“之I二•‘ 不二蒲7斗「1鬥不州r一蒲1界痲斤－不不－兀，:-，二，?1，︰一1一－？二1兀＝1-1鬥lres lee褶we不，r一鬥r一1－訂＝衣一r．以叩M。‘”二闖 ，〔‘膩c名．91〕1.,．。，,,t&g。·蓄1．可，·－一《，·1認，。‘。〔，.t&-.&.1&I‘。凶。。‘l間001“禺 瀉二。l,.01，電Ir中“鬨1,.“寥。“一。開t一I‘·電。t·《。乙。zoc，〔‘,+.&,z‘必ct，以膩闐11切乞“。 ,&,&,,,&&&,，一”&,,,,,，·，&,,,‘一”,&‘一‘，一‘,&,，一’&，。“&&&&&,”一’&&&&&&,，。一，黑共鍔慧．，。 面了闐I一配1闐，,.1不萬ee〞，,．。蒲r,t‘。訂Ic，一瑪可奮［[i一n 01一訂：t&1·啊「一．9之．[．方r eo記oc·汗乙記I《·蘇不－&．邑［,，。1 不1一之一不r館•斤不萬r－不不？r一r-T;r萬勿r~-?？「r？可一二甚r一了一荻，r一斤，鬥可～了一于「二寥r一汗，”一風。＇州 ”一〔。‘1&z,,&l&t&c&I，必91&，一《t‘開g&c。、一亂斗·、”09,．。“.1.(,,‘遞甚一【‘1.,&,.,,d，。”．綢、•1一” &&&&,&,&&,&,&&&&&，·’&,&&,&&&&&&,&&&，一’&‘開”,&&,&,&,&&&,&&&,&,&,”了：:：縱．,, 而．，畝二．,‘。乙萬～‘。，。。『，:：。“、．，一視：.,，一石．。．訪。．。一下八－。：:::．『：。：。一了：:，一不了一。‘:.,,．。 n斤丁蘇丁斤而不衣T一斤「r「市rr才？一？二鬥，付r，了「一r「m；『m兀，「『乏付啊→一了“二〝、“ 、寬‘必‘一‘1,09·實疋，•‘·［1·－-－二‘·1‘·‘妝‘［·。t可‘。I一01·“一l&,&&J〝 t 00,&1蓄‘呢右，。一。，。，·，一，-－·‘·－．一，一‘,，二•：It《叫二，‘』a ‘車居奮．1&．覦”•邊爭l龍［‘蘊‘。，電l妝自。佻•2疋‘一g〔不鳥·遛I記弋一醜一1〔奮I。‘o電口·讓亂，。c奮“鳥eog.,l& 1，…，,，。‘.：一，.:,，·一：·．-.，…：、·．．一，&,，一‘,,,,，一，&,，·，&’一”’。．。．，二二蠶｀,, 啊啊狗：討萬州討’‘一誹討計萬叩丫〕并〕一無（.) 糁間二哉不「不二不不萬萬下斤討閑， ,_&’黔．兀．&:.:‘斗‘吃．&：江‘斗‘亡，&::’斗’二”.’斗：”荒‘;&::&:，二‘汗‘訌。‘媒‘徐’:::,&:z:’訕‘;[.’斗’鴃’：斗‘~j么、‘&&‘。一，“一 －一·．．一二，-.，二。。界。，．。．斤．n。而州“。nr一以〔m.,.r∥”一轟一“,l1．馮，&‘巾”&,－一l,.，同”,&，二l-& 一，二下．,，二「，：二。：一一二：：寶一，·n：。。、一『，一『：。一汗。一·一一一；糞汰箕遝＃ “編，•轟蘇，一｀一浩」－奮」L一計必二r才‘一合一一－一7一一一一不一－一二－一一不「－--，→ &”】勾戶而區劉闐圍「圖n，付斤作扇可，訂弱不間「不序？刀石”下•觀〕開”寥網獨n顫函讓 （口一侈．開劇n) 訪面而r訪斤面nl馬口 朋奮碩口 BRAZI1, GRAIN STO!ý-(ýL PRO-ECT (1.- 57-1!R) C~parf.on oj oubproSect distrib-ton, Inv-rment coste and ut.b-~er Vý,~ - -L-l Group 3 and 4 - -d i- el zad ..ck Group I - fa- atarage. ~ t Group 2 - a~lt to emedlum-alzed nd b. l , w.,. h _... nd . il..1 ~rehouse. and proces.lng unita söck or bulk warelic-uoea large bulk warehouseg and allo&. st.. of I-e.t.-t (US$.10 3 Ig, ,, 50 St) t. 100 0-, !00 T.t.I Act-I E.tlmst,:,1- Di ffer-ce AcI-I Eýtlmstcd* Di ffe-P,e. Art-I Etitl-ted* DIffe-c. Act..1 E.tf-C,d* DIflerence A. of ..:,prol,cta Subp-l.cC. - -.1,- 417 604 - 187 100 79 + 21 100 130 - 30 617 813 - 196 per cent of tocal 67.6 74.3 - 6.7 16.2 9.7 * 6.5 16.2 16.0 + 0.2 100 100 - Invesc-nt excl.land per cent 26.1 17.9 + 8.2 4.6 7.2 - 2.6 69.3 74.9 5,6 100 100 .1 Lv:,1 5. Prulect co3C ( 55 x 103 CLýil -k. 10,969 7.744 + 3,225 1.947 3,735 -1.788 33.314 32.999 315 46,230 M-Iitikury &tid electrical 44.478 + 1.752 workg 6,266 4.506 + 1.760 1,103 1,030 + 73 12,572 16,513 3,941 19.941 22,049 - 2,108 E.6ý-:ing 80 250 - 170 10 242 - 232 67 2.477 2,410 157 2,969 - 2.812 F..xIblItty t.dfe. 35 25 + -- lo 10 18 - 8 139 258 119 184 301 - M T.tsl -1. 14nd 17,350 72,-525 4,EF 3,070 -3,025 -1,955 5,092 U- _ý2, 2 4 Y 6»US 12 -V9 . 79 7 - 3,285 L-J NÅ - NÅ 1,3s0 NÅ 3,853 MA 5 2203 T.t.I 12,525 6.375 56,100 75,000 C. A-,1y,t. f c.at 63.2 61.9 + 1.3 63.4 74.4 - 11.0 72.3 63.2 + 9.1 69.5 63.7 5.8 .1-t,tc.I -k. 36.1 35.9 + 0.2 36.0 20.5 + 15.5 27.3 31.6 - 4.3 30.0 31.6 - 1.6 0.5 2.0 1.5 0.3 4.3 4.5 0.1 4.7 - 4.6 0,2 4.3 - 4.1 (),2 0.2 0.3 0.3 0.3 0-5 - 0.2 0.3 T.i,1 -1. !a.. - - -- - - - - - 0. ýL 041 100.0 In0.0 100.0 100.0 100.0 100.0 100.0 100.0 D. C.1,city D- C.p-Lty 1..t.tt.d (t- x 10 3 507 200 + 307 163 280 117 1.343 1,420 77 2.013 1,900 + 113 cost per ton 34 62 - 28 19 18 1 34 37 3 33 37 - 4 c.i.-tty (-l- l.", 1. 115$) fk,ydlcal and price contingenetes reapectively calculnted gt 101 of baoe co.t and 101 at Löse cust plus phytiical contingency, have been prorated to cost c~ponenta for £1,1. -p.rl.on BRAZYL. - GRAIN STORACE PROJECT (Loan 857-BR) Compartaon of Act,sal with Etimare' Ph1sing of Investsmt. Subloan Investment, Numbere of St.bprojcts and Storage C.pacity (Thouaauda ot Year I (1973) Yar 2 (1974) Yar 3 (1975) - Ya. 4 (1976) Total 1. S,bloon Inve.tent US$) Actual Estimated Difference Actual Euti~ted Difference Actual E.timated Difference Estimated only * Actual Estimated Dåfference 1. Land H 0.648 HA 1,075 mA 1,290 1.287 NA 4,300 2. Civil Work@ 8.947 5,539 14,734 9.190 22.549 11,027 11,002 46,230 36,758 3. Mchinery; (-) Supply 2,164 ( 3,591 ( 4,309 4,298 ( 14.362 (b) Installation 3,512 0.298 6.151 (0,495 10,278 ( 0,594 0,593 19,941 ( 1,980 4. ElectricaI Work. 0,283 <0,470 <0.564 0,562 ( 1,879 5. EninerIng 0,082 0.370 0.044 0,6l4 0,031 0,736 0.734 0,157 2,454 6. F-åibillty Studies 0.090 M,038 0,039 0,062 0,055 0.075 0 0 7 i 0,184 M9 Subtotal 9,340 15,497 18,595 18,550 61,9d2 Phy.fcal CntLingency (101) 934 1 550 1,859 1,855 6,93 Subtotal 10,274 17,047 20,454 20,405 68,1t0 Price Cotingency (101) 0,260 1,000 2.96 3,364 6.820 Total 1024 18,047 27650 23 69 75 > T.tal c-land 12,631 9,886 + 2.745 20,968 16.972 + 3.996 32,913 21,360 + 11.553 22.482 66.512 70,700 - 4.188 iTotal 19 14 + 5 32 24 + 49 30 + 19 32 100 100 .II,.umbeta of Sub-Loans 112 112 - 319 192 + 127 186 240 - 54 256 617 800 - 183 111. Approx-ate S eorage Capacity 684 285 + 399 803 475 + 328 526 570 - 44 570 2,013 19)0 + 113 (tono . 10 ) CumulatIv- 684 285 + 399 1,487 760 + 727 2.013 1,330 + 683 1,900 * No con-ntaents were made after 1975 M BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) Comparison of actual and estimated loan disbursements (in thousands of US$) ,Quarterly disbursements Cumulative DiFlursements BRD scal Actual as % ear Quarter Actual Estimated Difference Actual Estimated Difference of Estimated 973 1 2 - 3 - 4 - 564 - 564 - 564 - 564 0 .974 1 - 752 - 752 - 1,316 - 1,316 0 2 1,699 1,316 + 383 1,699 2,632 - 933 65 3 1,647 1,583 + 64 3,346 4,215 - 869 79 4 2,416 1,752 + 664 5,762 5,967 - 205 97 L975 1 1,708 1,802 - 94 7,470 7,769 - 299 96 1 2 1,154 1,852 - 698 8,624 9,621 - 997 90 3 1,224 1,912 - 688 9,848 11,533 - 1,685 85 4 1,945 2,076 - 131 11.793 13,609 - 1,816 87 976 1 1,442 2,215 - 773 13,235 15,824 - 2,589 84 2 4,138 2,300 +1,838 17,373 18,124 - 751 96 3 - 2,369 -2,369 17,373 20,493 - 3,120 85 4 851 2,374 -1,523 18,224 22,867 - 4,643 80 L977 1 - 2,374 -2,374 18?224 25,241 - 7,017 72 2 - 2,374 -2,374 18,224 27,615 - 9,391 66 3 - _2385 -2,385 18,224 30,000 -11,776 61 TOTAL 18,224 30,000 - 11,776 DRAZIL - CRAIN STORAGE PROJCt...Loan 857-R) 5nc. d, Br. a sl:- S-~nry of Lons made 1966-1976 (alle 14 12 12 19 18 0.6 46 0.6 48 0.6 71 0.6 97 1.0 219 0.7 129 Others 428 449 724 _ 901 904 8.1 1.103 8.6 1,204 8.4 1.473 17.2 4.212 13.5 4916 13.9 4.769 Subtotal 4.7 442 4.9 461 6.0 736 7.4 920 9.0 922 8.7 1,149 9.2 1.252 9.0 _544 17.8 4,3r9 14.5 5,135 14.6 4,898 321.8 1,953 353.0 2.451 363.2 3,036 376.9 4,040 396.8 3.944 448.1 4.623 492.4 5.686 515.1 7,518 578.7 12,717 737.2 18.826 771.3 19,106 1. Agrculture Far Pdrodctlo 445 460 554 548 731 138.7 1,002 149.7 1,614 167.5 1,926 169.2 2.198 200.6 3,576 195.7 3,491 Storage 3 6 8 5 11 3.2 35 5.0 51 5.7 52 8.9 87 14.6 112 6.5 156 Ln C-ain MIIIIng 5 5 7 10 8 3.8 18 6.8 68 7.2 69 8.4 149 6.3 209 7.1 76 Sub.taI 75.2 453 75.6 471 93.0 569 90.0 563 105.6 750 145.7 1.055 161.5 1.733 180.4 2.047 186.5 2,434 221,5 3,897 209.3 3.723 2. LIvoa..,ck . 51.0 338 53.3 338 77.3 475 87.5 530 91.7 562 115.7 874 119.9 1,132 125.7 1,710 120.9 1.815 143.1 2.532 122.6 3.090 3. Indutry Crai.. 0il, and Fe.d M11 i 1 2 3 0.3 15 0.3 16 0.3 20 0.3 33 0.3 27 0.3 22 Oth er 54 122 152 - 204 _ 224 2.5 229 2.8 323 3.3 835 3.0 866 3,3 1,779 2.7 709 Subtoal i.3 55 1.6 123 __2.1 154 2.3 ?o-' 2.3 227 2.8 244 3.1. 339 3.6 855 3.3 899 3.6 L806 3.0 731 TOZ.l 127.5 846 130.5 932 172.4 9 1798 1,300 199.6 1 F539 264.2 2173 284.5 - 309.7 4 310.7 5.148 368.2 8 235 3 III. A1.L ti.jas 1. AgrIculture 376.5 1,872 406.9 2,358 424.6 2,680 428.9 3.411 463.3 3.553 550.4 4,221 599.6 5.842 643.3 7,565 694.0 10.243 786.8 15,422 866.7 16.7b4 2. Livestck 66.8 430 70.1 440 102.9 666 118.1 801 121.8 782 150,4 1,182 165.0 1,457' 168.9 2,166 174.3 2.414 300.. 4,698 221.9 4,257 3. Indu.try 6.0 497 6.5 584 8.1 890 9-7 1_1.2 11.3 1,149 11,5 1.393 12.3 1 ål 12.6 2,399 21.1 5,208 18.1 6,941 _1_.6 5,629 Total 449.3 2_199 483.5 3382 535.6_ ,236 556.7 5_139 596.4 _84 712.3 6,796 776.9 8.890 824.8 1? 889.4 1865 1,105.4 27 04106. 1 250 BRAZIL - GRAIN STORAGE PROJECT (Loan 857-BR) Banco do Brasil S.A.:Number of Branches and Employees 1966-76 1966 1967 1968 1969 19701jS 192 23 197 15A 12JI6 I.TOTAL BRANCHES 645 1A NA NA 751 797 814 835 878 923 1,068 Branches in Project Area Espirito Santo 10 10 10 12 12 12 13 13 16 16 19 Rio de Janeiro 22 22 22 23 23 63 64 67 69 77 32 Sao Paulo i29 13u 1 2 144 14 165 171 173 175 183 211 Parana 44 46 47 51 53 57 58 59 64 65 77 Santa Catarina 26 26 27 29 29 32 32 34 38 39 47 Rio Crand do Sul 69 70 78 83 88 97 98 101 107 107 117 Minas Gerais 102 102 102 105 107 110 111 113 118 121 138 Goias 34 35 35 46 47 39 40 40 43 43 53 Mazo Grosso 19 20 21 21 22 27 27 27 29 29 36 Total 455 461 474 515 535 602 614 627 659 680 780 % Increase - 1.3 4.1 12.9 17.5 32.3 34.9 37.8 44.8 49.4 71.4 ]LSTAFF Y OCCUPATIONS(in thousands) ikng Services Accounting(professional) 25.7 26.3 25.9 25.7 29.0 28.5 28.7 30.0 30.5 34.4 34.4 AccounLing(auxiliary and junior) 4.8 5.1 5.5 6.0 4.4 7.9 10.7 13.0 15.3 17.7 24.4 Other Office Staff 0.4 0.4 0.5 0.4 0.4 0.5 0.5 0.5 0.5 0.5 0.4 Trea--ury 1.2 1.3 1.3 1.1 0.8 0.6 0.5 0.5 0.4 0.3 0.2 RurLl Credit Field Staff 0.6 0.6 0.6 0.6 0., 0.5 0.5 0.5 0.5 0.4 0.4 Attendants, Porters,Messengers,etc. 6.3 7.1 7.0 6.9 6.7 6.4 6.4 6.7 8.5 10.9 11.5 Subtotal 39.5 40.8 40.8 40.7 41.8 44.4 47.3 51.2 55.7 64.2 71.3 supporting Services Professional and Specialists 0.5 0.4 0.4 0.4 0.6 0.6 0.6 0.6 0.6 0.6 0.6 Artisans and others 0.4 0.5 0.4 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Subtotal 0.9 0.9 0.8 0.9 1.1 1.1 1.1 1.1 1.1 1.1 1.1 Total 40.4 41.7 41.6 41.6 42.9 45.5 48.4 52.3 56.8 65.3 72.4 % Increase - 3.2 3.0 5O 6.2 12.6 19.8 29.5 40.6 61.6 79.2 OM. NUMBER OF LOANS PER EMPLOYEE 11.1 11.6 12.9 13.4 13.9 15.7 16.1 15.8 15.7 16.9 15.3 ICY% DRAZUL CRAIN STORACE PROJECT (Loan 857-BR) Banco do Braall S.A. - C-ndenved Inccome and E90 1229 L17§_ L- B_08 _ 50 i 55 850 2.079 596 3 35 4507 RAZI. GRAIN STORAGE PROJECT (Loan 857-ER) B-.no do Brail S.A. - Conen-d Blance Shed, for Ye.r 1966-1976 (in billiond of 1970 Cruziros) 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 ASSETS C..h 0.23 0.25 0.17 0.22 0.24 0.18 0.19 0.22 0.57 0.27 0.20 Prod.ction 4.94 4.12 4.77 ..b. !.41 12.09 15.27 19.11 25.81 36.02 37.27 C - .r 0.69 0.e1 4.44 5.35 4.27 4.51 3.18 3.54 5.26 5.19 7.01 Trsury 7.92 7.88 4.97 4.08 3.40 2.85 2.46 2.13 1.58 1,22 0.84 oher. 0.09 0.18 0.65 1.20 1.35 1.70 2.06 2.94 4.69 5.72 9.66 13.64 12.99 14.83 16.47 18.43 21.15 22.97 27.72 37.34 48.15 54.78 AJvan-- and Credits: Tre.suy - 0.83 1.b7 2.99 3.80 3.22 1.71 1.52 0.32 1.37 0.79 o-e~* 10.49 8.75 7.96 9.96 10.20 7.98 9.97 8.83 10.94 9.53 9.49 Int-rnI - - 0.84 1.24 1.67 2.63 0.96 0.39 - 0.76 - Ccher. 4.23 4.46 4.45 3.74 3.74 3.63 5.74 8.65 9.05 10.83 21.75 14.72 14.04 14.92 17.93 19.41 17.46 18.38 19.39 20.31 22.49 32.03 S-gettable securitie- 0.02 0.05 0.07 0.49 0.61 0.63 0.85 1.47 1.61 1.91 2.10 Fi-l A,aets (less depreciaton) 0.10 0.12 0.06 0.08 0.09 0.16 0.17 0.14 0.14 0.31 0.71 0.07 0.13 1.01 0.10 0.01 0.00 0.00 0.00 0.00 0.00 0.00 28.78 27.58 31 06 35.29 38.79 39 58 42.56 48.94 59.97 73.13 89.82 LIAh!LITIES CapItal an, Reervs~ C.plC.I 0.05 0.11 0.08 0.29 0.72 0.90 1.30 1.80 2.35 4.14 4.25 0.62 0.74 0.99 1.07 1.01 1.36 1.91 2.01 2.52 3.20 3.59 0.67 0.85 1.07 1.36 1.73 2.26 3.21 3.81 4.87 7.34 7.84 peo c.(sho~-tterm and aight): PublIc 1.78 2.20 2.78 3.32 3.8u 4.14 4.83 6.40 6.80 7.91 7.33 bank anJ Flinance In.titution. 3.07 4.59 1.94 2.60 2.66 3.79 2.11 2.20 2.14 2.19 1.75 Treasury 6.72 6.07 5.47 5.41 5.88 5.49 5.12 6.84 9.17 10.52 6.48 Ceucal bank and other Cov. entities 5.31 2.18 4.68 4.26 3.91 4.10 4.39 4.99 5.24 3.57 3.65 16.88 15.04 14.87 15.59 16.25 17.52 16.45 20.43 23.35 24.19 19.21 U>eposit (.nJuei-term) 0.07 0.12 0.12 0.12 0.12 0.34 0.39 0.53 1.23 1.69 0.21 Dep-etis (de-and and »pecial) 9.91 10.24 12.72 14.84 16.45 13.75 15.35 12.79 14.59 20.01 27.53 SPe1l F.njs - - 0.81 1.65 2.15 3.21 4.60 8.56 13.45 18.01 33.12 1.25 1.33 1.47 1.73 2.09 2.50 - . 2.56 2,82 2.48 1.89 1.91 28.78 21 3529 3.9 942556 48.9 59.97 _7.13 89 2 OparaIng pfit Averaba LotaI ...et (Lar cent) 5.5 6.7 6.0 6.2 7.2 7.9 9.0 13.0 15.0 9.9 BRAZIL - GRAIN STORAGE PROJECT (Loan 857-BR) Number and Amount of Subloans Under Successor Program .(Programa Nacional de Armazenagem - Pronazem) uuna in thuu,andz of Cruzoiro2) Farm-storage Intermediate Total program subprojects and Lerminal sturage Subprojects Total value Total Total Number Av.Value Value Number Av. Value Value Number Per Cent Amount Per Cent Rio Grande do Sul 286 746 213,263 27 9,176 247,760 313 35.7 461,023 45.7 Santa Catarina 8 714 5,708 1 2,276 2,276 9 1.0 7,984 0.8 Parang 285 750 213,648 9 2,755 24,793 294 33.6 238,441 23.7 Sao Paulo 61 721 43,988 4 2,038 8,150 65 7.4 52,138 5.2 1 Rio de Janeiro - - - 3 7,577 22,730 3 0.3 22,730 2.2 Minas Gerais 4 425 1,698 5 2,949 14,747 9 1.0 16,445 1.6 Coigs 34 500 17,010 5 1,860 9,301 39 4.5 26,311 2.6 Mato Grosso 128 768 98,257 10 7,772 77,720 138 15.8 175,977 17.5 Distrito Federal 4 695 2,780 - - - 4 0.5 2,780 C.3 Balkia - - - 1 3,200 3,200 1 0.1 3,200 0.3 Maranhao 1 590 590 - - - 1 0.1 590 0.1 Total 811 736 596,942 65 6,318 41C,677 876 100.1 1,007,619 100.0 BRAZIL GRAIN STORAGE PROJECT - LOAN 857-BR Disposition of Subloan Approvals Total subloans approved by Banco do Brasil 617 381,459 Less: subloans dropped from project before disbursement 9 18 533 Subloans financed by project 608 362,926 Less: subloans completely prepaid - 1975 23 10,595 1976 114 36,739 Total 137 47,334 Remainder 471 315,592 Less: partial amortization of subloans, much of which was ahead of schedule - 22,918 Subloans remaining at December 31, 1976 471 292,674 - - AMlNEX 11 ERAZIL GRAI STC1AG. PROJECT (Loan 857-BR) Subloan F7ecovery E e.ience (A) (Anounts in thousands of Cr$) Loan repaymcer_ts: Interest: Year 1975 Year 1976 Number Amount Number Amount Overdue at beginning of year 75 1,437 498 13,330 Receivable during year 553 11,279 471 13,034 Total receivable 628 12,716 969 26,364 Received during year 394 7,824 358 (B) 14,268 Overdue at end of year 234 4,829 611 12,096 Recovery percentage(received - receivable) 61.5 54.1 Analysis of year-end overdues by duration Less than 3 months 192 3,752) ) 447 8,589 3 - 6 months 16 192) 6 - 12 months 22 765 114 2,100 12 - 24 months 4 183 44 1,209 More than 24 months _ _ 6 198 Total 234 4,892 611 12,096 Notes A: Due to a misunderstanding, inappropriate data concerning subloan repayments were provided in 1976. Data concerning interest overdues were not available until 1976. B: By difference. Source data are inconsistent. - 52 - mila 80 o1 0 ~l 0 - - 4 nt n -Lfl4 - 8 - - 2 f 1 t Z i 4; flO 0fl fl <0 0 fl o 2 .ii t l4 -f f - 4 2 . ~..2, .. . 2 :K fl 4. f40 . * f1 J0 fJ fl o må- dni - 0 f ~ 0 -- - f . fl fl fl f 1 - - ;0. .0 , 040'lfl fl 0f fl~~ 0' '' ~ .0" Of CO l l 0 0 0 l lf fl4 '4 lf 0 OA fl 0 f 2 .-. - .. -- - C D fiflOfl fl 0l4 .. flf .' 0'4la.1 0Z --C fl fl. g 4 4 l e fl g0<- e afe fl ....,. - .. . .. .. .. 4 '.- ,. .,. . fl f l f 3 °* 0 0"ta ~ ~ 1 3 % i 2 i i ' - . .. . .a . . .~ • 9 04 0" ,g* 04U M2 1 - 0.1 as 0 - 4' - * eO f-l *4Q M Q 0 5 *f * MO0 • 0 f.. .3 . = - - .-I=- u 3 - d - - - 53 - ANNEX 13 BRAZIL GRAIN STORAGE PROJECT (Loan 857-BR) Grain Production in Project Area:1955-76 Page 1 of 6 (in thousands of tons) RIO GRAMDE DO SUL SANTA CATARINA Dry Dry Year Corn Rice Wheat Beans Soybeans Corn Rice Wheat Beans Sovbears 1955 - - - - - * 96 - - * 1956 1,444 790 683 126 107 495 130 92 72 - 1957 1,476 724 614 121 109 507 157 94 71 2.8 1958 1,484 805 407 139 118 548 134 97 70 4.0 1959 1,605 754 420 156 140 571 155 109 74 3.6 1960 1,582 889 532 144 188 597 136 109 79 3.8 1961 1,765 1,090 398 151 253 615 156 79 85 4.0 1962 1,871 1,170 521 163 321 . 624 137 105 86 4.1 1963 1,948 1,275 263 174 295 670 160 71. 92 4.3 1964 1,774 1,181 478 173 276 684 179 96 96 4.7 1965 2,244 1,304 421 .230 -463 748 178 85 102 5.1 1966 2,281 1,168 461 213 483 699 172 63 93 7.6 1967 2,331 1,281 482 234 551 898 212 68 120 9.2 1968 1,1 1,286 665 228 433 792 213 72 113 14.8 1969 2,234 1,354 1,066 219 74 990 211 82 104 31.6 1970 2,387 1,543 1,448 245 977 1,082 214 92 100. 53.0 1971 2,370 1,519 1,552 204 1,393 1,228 208 78 93 77.7 1972 2,235 1,613 f23 172 2,174 1,232 217 62 105 98.8 1973 2,101 1,43d 1,536 177 2,872 1,560 222 63 92 253.5 1974 2,236 1,550 1,.:30 153 3,870 2,218 231 81 128 431.5 1975 2,367 1,700 1,234 156 4,688 2,127 293 30 169 467.2 1976 2,443 1,850 1,809 140 5,107 2,453 318 26 29 403.9 Average Annual Increase 3.0% 4.8% 14.1.. 1.2. 23. 7w 9.0% 6.6% -2. 7% 3.1% 35.7% P A R A N X S A 0 P A U L 0 Dry D r Year Corn Rice Wheat Beans Sovbeans Corn Rice Wheat Beans Sovbeans 1955 - 152 - - - - - 2.9 210 2.5 1956 1,302 183 75 227 1.5 1 277 751 4.6 198 2.8 1957 1,193 241 67 238 4.5 1,371 970 5.7 221 3.4 1958 1,153 210 78 304 3.6 1,404 832 6.0 201 4.2 1959 1,221 273 77 307 4.9 1,388 799 4.6 166 2.6 1960 1,297 268 67 299 7.4 1,582 919 4.0 178 3.1 1961 1,340 309 61 323 9.0 1,575 916 5. 1 179 4.7 1962 1,478 335 71 339 13.9 1,722 865 6.9 158 4*6 1963 1,683 392 56 473 17.9 1,961 910 1.0 178 4.1 1964 1,765 416 63 539 18.5 1,299 792 4.8 135 3.8 1965 2,183 546 73 595 44.0 2,145 1, 095 5.3 162 8.9 1966 2,037 466 78 557 83.0 2,110 777 4.8 157 18.2 1967 2,229 491 75 559 113.0 2,273 1,077 3.5 152 38-2 1968 2,497 334 114 528 163.0 2,444 815 3.5 138 39.3 1969 2,712 432 221 470 213.0 2,115 774 5.1 128 61.0 1970 3,559 590 283 730 368.0 2,676 1,053 19.3 161 90.0 1971 3,655 599 335 757 462 2,177 641 38.7 150 86 1972 3,829 674 257 818 688 2,635 901 30.6 156 175 1973 , 3,006 . 661 385 472 1,325 2,630 603 34.9 148 330 1974 3,553 672 915 562 2,589 2,528 582 170.0 131 522 1975 3,813 851 - 444 608 3,425 2,100 510 70.5 108 678 1976 4,823 1,089 1,161 588 4,500 2,724 840 195.0 140 765 Average knnual Increase 7.4. 11.4t 23.3 6.7% 55.6% 5.8% 4.2% 55.7% -1.0% 38.7% Cont bNEX 13* Page 2 of 6 Crain Production in Proct Area:1955-76 (in thousands of tons) (cont.) Rio de Jan!iro Esoirito Santo Minas Gerais Dry7 Dry Dry Year Corn Rice Beans Corn Rice Beans Corn Rice Beans Sovbeans 1955 83 49 11.5 65 25 24 - - 298 - 1956 80 57 11.9 87 24 31 1,295 591 283 - 1957 108 82 13.0 136 33 34 1,720 746 338 - 1958 100 94 13.0 131 36 33 1 ,60 729 331 - 1959 90 76 11.0 122 34 29 1,533 703 264 - 1960 129 117 12.0 149 41 39 1,841 940 351 * 1961 137 119 10.5 150 46 38 1,846 929 324 - 1962 121 107 7.5 155 48 38 1,868 921 .244 - 1953 128 109 7.3 138 53 29 1,832 741 218 - 1964 114 116 7.9 152 64 41 1,640 957 252 - 1965 135 134 8.6 177 69 38 2,167 1,248 305 - 1966 102 97 7.9 132 50 26 1,960 890 259 - 1967 105 124 8.2 245 58 50 2,131 975 288 - 1968 81 118 10.4 273 64 47 2,173 1,039 273 - 1969 55 111 6.8 181 64 43 1,989 1,015 259 - 1970 93 113 7.0 299 20 42 ?,302 1,166 288 1.8 1971 62 60 6.3 218 77 41 i,776 752 263 1.4 1972 84 141 7.5 254 82 48 !,111 1,055 316 8.9 1973 60 118 6.5 234 93 44 1,994 828 282 36.3 1974 74 71 5.2 246 69 45 3,293 479 419 57.6 1975 55 77 6.3 198 65 44 ,322 773 234 87.4 1976 49 69 7.2 153 58 30 ?,340 962 266 105.5 Ave rage Annual Increase 0.9. 6.7, -0.9 6.5% 5.3% 4.0% 2.4% . q% 1. 3% 15& 7% MAT 0 GR 0 SS 0 G 0 1 A S Year Corn Rice Wheat Beans Sovbeans Corn Rice Beans Sovbeans 1955 70 100 - 28 - - - * - 1956 84 122 - 24 - 228 371 74 - 1957 88 139 - 37 - 242 542 65 - 1958 90 163 - 39 - 259 412 66 - 1959 105 212 - 46 - 268 474 46 - 1960 112 221 - 49 - 322 724 77 - 1961 111 262 - 51 - 395 762 76 - 1962 162 255 - 47 - 403 838 70 - 1963 220 296 - 68 - 516 831 85 - 1964 152 417 - 49 - 559 1,272 87 - 1965 175 491 - 64 - 669 1,470 126 - 1966 166 331 - 61 - 596 963 116 - 1967 141 354 - 52 - 639 1,165 128 - 1968 148 325 - 46 - 669 1,250 134 - 1969 204 390 - 44 - 561 915 96 - 1970 228 617 0.9 60 9.0 657 1, 218 115 9. 8 1971 228 512 7.5 63 16.2 654 973 133 41.0 1972 249 694 10.7 51 27.9 663 1,183 131 49.9 1973 245 782 12.9 51 103.2 704 1,166 74 89.7 1974 280 813 19.8 62 307.0 1,084 959 95 99.0 1975 382 1,003 9.0 45 272.i 1,229 868 112 73.4 1976 353 1,627 30.2 57 290.4 1,274 1,319 107 48.7 Average Annual Increase 9.65 16.3% 171.8 5.8% 102.5% 9 9% 9.9% 4 .8% 61.8% BRAZIL - GRAIN STORAGE PROJECT (Loan 857-BR) Comparison of Actual with Estimated Project Area Production in 1975 (in thousands of tons) Corn Rice Wheat Actual Forecast Difference Actual Forecast Difference Actual Forecast Difference Rio Grande do Sul 2,367 2,943 - 576 1,700 1,984 - 284 1,234 2,074 - 840 Santa Catarina 2,127 1,239 888 293 273 20 30 160 - 1.30 Parang 3,813 4,157 - 344 851 832 19 .. 444 179 265 Sao Paulo 2,100 3,492 - 1,392 510 974 - 464 71 4 67 Minas Gerais 2,322 2,636 - 314 773 1,531 - 758 - - - Goigs 1,229 1,243 - 14 868 2,122 - 1,254 - - - Mato Grosso . 382 303 79 1,003 858 145 9 - 9 Espirito Santo 198 313 - 115 65 106 - 41 - - - Rio de Janeiro 55 93 - 38 77 185 - 108 - - - Total 14,593 16,419 - 1,826 6,140 8,865 - 2,725 1,788 2,417 - 629 Dry Beans Soybeans Total. Actual Forecast Difference Actual Forecast Difference Actual Forecast Difference Rio Grande do Sul 156 325 - 169 4,688 1,862 2,826 10,145 9,188 957 Santa Catarina 169 141 28 467 36 431 3,086 1,849 1,237 Paran5 608 928 - 320 3,625 388 3,237 9,341 6,484 2,857 0ao Faulo 108 110 - 2 678 112 566 3,467 4,692 - 1,225 Minas Gerais 284 240 44 87 - 87 3,466 4,407 941 Goi5s 112 168 - 5 7i 1 72 2,282 3,534 - 1,252 Xato Grosso 45 79 - 34 273 4 269 1,712 1,244 468 Espirito Santo 44 52 - 08 - - - 307 471 - 164 Rio de Janeiro 7 6 1 - - - 139 284 - 145 Total 1,533 2,049 - 516 9,891 2,403 7,488 33,945 32,153 1,792 Source:Fundacao IBGE (D o ) BRAZIL - GRAIN STORAGE PROJECT LOAN 857-BR) Comparison of Actual with Forecast Project Area Marketable Balances in 1975 (in thousands of tons) MAIZE RICE . lFAT DRY BEAN SOYBEAN TOTAL Retained Marketable Retained Farketable Retained Marketable Retained Marketable Retained Marketable Retained Marketable STATE Prod.on farm Balance Prod. on farm Balance Prod. on farm Balance Prod. on farm Balance Prod. on farm Balance Prod. on far Balance Rio Grande do Sul 2.367 306 2,061 1,700 120 1,580 1,234 258 976 156 15 141 4.688 345 4,343 10,145 1,044 9.101 Santa Catarina 2,127 469 1,658 293 28 265 30 5 25 169 13 156 467 35 432 3,086 550 2,536 Parani 3,813 308 3.505 851 101 750 444 123 321 608 59 549 3,625 238 3,387 9,341 829 8,512 Sao PAUlo 2,100 158 1,942 510 105 405 71 17 54 108 16 92 678 45 633 3,467 341 3.126 I Mins Geraja 2,322 375 1,947 773 149 624 - - - 284 40 244 87 7 80 3,466 571 2,895 0N Coiis 1,229 167 1,062 868 187 681 - - - 112 16 96 73 4 69 2,282 374 1,908 M.to Crosso 382 39 343 1,003 218 9 5 4 45 5 40 273 20 253 1,712 287 1,425 Lpirito Santo 198 14 184 65 9 56 - - - 44 6 38 - - - 307 29 278 Rio de Janeiro 55 4 51 77 8 69 - - - 7 I 6 - - - 139 13 126 TOTAL 14,393 1.840 12,753 6 925 5 1 408 8 !a5i 171 1,362 9,691 694 9,19 3 4,038 290 Forecast Data 16.419 1,891 14,528 8.865 1,037 7,828 2,417 353 2,064 2,049 195 1,854 2,403 120 2,283 32.153 3.596 28.557 Diterence Amount - 1,826 - 51 - 1,775 2,725 - 112 - 2,613 629 - 55 - 684 516 - 24 - 492 7,488 574 6,914 1,792 442 1,350 Percent - 12 - 33 - 33 - 27 + 303 + 5 - 57- Page 5 oi 6 BRAZIL - GRAIN STORAGE PROJECT (Loan 857-BR) Gross Non-farm Storage Requirement in 1975: Comparison of Estimates Based on Actual and Forecast Production (in thousands of tons) Wheat & Actual Data by State Corn and Dr'V Beans Rice Sovbeans TOTAL Rio Grande do Sul 1,376.2 855.2 2,769.0 5,000.4 Santa Catarina 986.8 154.6 305.4 1,446.8 Parana 2,367.9 437.5 2,220.0 5,025.4 Sao Paulo 1,353.4 270.0 389.1 2,012.5 Rio de Janeiro 38.0 34.5 - 72.5 Espirito Santo 140.5 28.0 - 168.5 Minas Gerais 1,412.8 312.0 48.0 1,772.8 Goigs 651.5 340.5 41.4 1,033.4 Mato Grosso 197.6 392.> 147.8 737.9 TOTAL 8,524.7 2,824.E 5,920.7 17,270.2 Forecast 10,034.1 4,208.P_ 2,334.8 16,577.8 Difference Amount - 1,509.4 - 1,384.1 3,585.9 692.4 Percent - 15 - 33 + 154 + 4 ANINEX 13 I"AZIL -1 1R0T STOLAGv PROJECT (Loan S57-BR) Page6 of 6 Estimatod Proinct-area Storage Needs in 1975 (Thousands of tons) Gross Existing 1975 Capacity Actual Data by State Requirement end 1971 Deficit SurDlus Rio Grande do Sul 5,000.4 2,624.1 2,376.3 Santa Catarina 1,446.8 325.7 1,121.1 Parang 5,025.4 3,810.2 1,215.2 Sao Paulo 2,012.5 4,049.0 --2,036.5 Rio de Janeiro 71.4 240.5 Espirito Santo 168.5 168.0 0.5 Minas Cerais 1,772.8 506.3 1,266.5 Goigs 1,033.4 358.4 675.0 Mato Grosso 737.9 145.8 592.1 Total 17,269.1 12,228.0 7,246.7 2,205.6 Forecast data 16,577.8 12,288.0 5,994.9 1,645.1 Difference 691.3 - 1,251.8 560.5 Annex 14 BRAZIL - CRF.IN STOVCG PPROJECT (Loan 857-BR) Page 1 of 2 Cost and benefit strea1s tron r ct as implemented (In US$ thousands;) Cost of emergency storage without the project Total Costs Reduction in grain- Converted Planks and years storage losses buildings sheets 1973 3,297 3,404 392 12,631 j974 9,890 10,211 888 20,968 1975 16,482 17,017 1,378 32,913 1976 19,779 20,421 1,619 - 1977 989 - 1978 945 - 1979 1,081 - 1980 1,497 - 1981 1,631 - 19 8 2 1,013 - 1983 920 3,512 198N4 1,107 6,151 1985 1,471 10,278 186 1,657 - 1987 989 - 19 8i 945 - 198), 9 1,081 - 1990 1,497 - 1991 1,631 - I 92 1,013 - 1093 920 3,512 19Q4 V 675 6,151 905 16,482 17,017 86 10,278 006 9,890 10,211 92 - 3,297 3,404 - 234 t enefits from reduction in grain-storage losses and savings from not converting existing buildings have been phased in as follows: 1973 - one sixth of total benefits; 1974 - one-half; 1975 - five-sixths; 1976 on - all savings from not erecting planks and sheets have been repeated from the appraisal report. BRAZIL - GRAIN STORAGE PROJECT(Loan 857-BR) Annex 14 Estimated reduction in grain storage losses resulting from project Page 2 of 2 Non-farm (4%) Farm(6%) Total Price - Price Crop Tons Per Ton Value Tons Per Ton Value Tons Value Corn 4,198 95 399 2,120 85 180 6,318 579 Rice 26,459 230 6,086 13,361 210 2,806 39,820 8,892 Dry Beans 390 630 246 197 580 114 587 360 Wheat 4,979 250 1,245 2,514 225 566 7,493 1,811 Soy beans 24,213 230 5,569 12,227 210 2,568 36,440 8,137 Total 60,239 13,545 30,419 6,234 90,658 19,779 Note: prices per ton are in US$, based on average 1974-1975 prices in Cr$ converted at average exchange rates. value of reductions in losses are in US$x103 The CEAGESP Subproject Estimated and Actual Completion Dates of CEAGESP Installations Original Contract Contract Extensions Contract Months to Completion Number of Number of Completion Total Months Date Completion Date Extensions Months Date to Completion Silos: Palmital June 75 10 April 76 1 5 Sept.76 15 Tatui June 75 10 April 76 2 7 Nov.76 17 Sao Joaquim da Barra July 75 10 May 76 2 6 Nov.76 16 Araraquara July 75 15 Oct 76 2 3 Jan.77 18 Granaries: Palmital Dec.74 11 Nov. 75 2 10 Sept. 76 21 Tatui " " 11 " " 2 12 Nov. 76 23 Sao Joaquim da Barra " " 11 " " 2 12 Nov. 76 23 Araraquara " " 11 " 14 Jan. 77 25 o rz o The CEAGESP Subproject Final vs. Appraisal Cost of CEACESP Installations (Cr$ x 10 ) Sao Joaquim Palmital Tatul da Earra Araraquara Total F E F E F E F E F E D I. Contracts placed before September 1975 - at stated prices: A. Granaries 37.3 32.0 37.3 32.0 37.3 32.0 47.4 39.0 159.3 135.0 24.3 B. Silos 16.9 15.4 16.4 13.4 18.0 15.1 18.9 15.0 70.2 58.9 11.3 Subtotal 1 54.2 47.4 53.7 45.4 55.3 47.1 66.3 54.0 229.5 193.9 35.6 II. Additional works contracted for after September 1976: A. Cranary temperature control 0.8 0.J 1.2 1.3 4.1 4.1 B. Granary air circulation system 0.5 0. 0.5 0.5 2.0 2.0 C. Railroad siding 0.2 0.4 0.2 0.2 1.0 1.0 D. Paving of year 4.2 6.6 4.2 4.6 19.6 19.6 E. Access road 0.5 0.6 0.7 0.6 2.4 2.4 F. Other 0.5 0.8 3.8 2.1 7.2 7.2 Subtotal II 6.7 9 10.6 9.3 36.3 36.3 Total 60.9 47.4 63.4 45.4 65.9 47.1 75.6 54.0 265.8 193.9 71.9 F=Final E=Appraisal estimate D=Difference IIQ o0i ANNEX 15 Page 3 of 4 The CFAGPS Subnro Flow of Lulk Gr.iin t:ronC C17LAT Facilities (thousands of :2Lric tons) Beginning End Average of month of month Stock for Stock Inflow Outflow Stock Month January 1975 83.7 4.2 28.4 9.5 71.6 February 59.5 7.0 19.8 46.7 53.1 March " 46.7 15.6 16.4 45.9 46.3 April " 45.9 36.8 24.9 57.8 51.9 May " 57.8 62.4 12.6 107.6 82.7 June " 107.6 25.3 11.8 121.1 114.3 July " 121.1 12.3 10.6 122.8 122.0 August " 122.8 3.7 9.6 116.9 119.8 September " 116.9 2.5 15.3 104.1 110.5 October " 104.1 5.1 23.5 85.7 94.9 November " 85.7 2.0 23.8 63.9 74.8 December " 63.9 3.7 23.6 44.0 54.0 January 1976 44.0 0.8 19.0 25.8 34.9 February 25.8 3.5 12.8 16.5 21.1 March " 16.5 13.1 8.1 21.5 19.0 April 21.5 30.0 8.5 43.0 32.3 May 43.0 39.7 6.8 75.9 59.4 June " 75.9 32.2 9.4 98.7 87.3 July 98.7 24.2 11.0. 111.9 105.3 August " 11.1.9 13.6 6.0 119.5 115.7 September " 119.5 27.2 9.6 137.1 128.3 October " 137.1 23.8 14.0 146.9 142.0 November " 146.9 5.1 18.0 134.0 140.4 December " 134.0 2.8 17.8 119.0 128.5 Year 12 months Year 1974 1975 ended July '76 1976 A. Total inflow of grain for 12 months 214.5 180.6 16).6 21E.0 .. Total outflow of grain for 12 months 201.5 220.3 171.4 141.0 C. Static capacity 142.0 142.0 142.0 382.0 * D. Rotation rate (A or B C) 1.5 1.5 1.2 0.6 * E. Sum of average stocks for 12 months 1,077.5 995.9 813.4 1,012.2 F. Average monthly stock 89.8 83.0 67.8 84.4 G. Utilization capacity (F C) - percent 63.2 58.4 47.7 22.1 * H. Average storage time in months (E 1 A) 5.0 5.5 5.1 4.7 * The addition of 240,000 tons static capacity in late 1976 distorts the rotation rate and utilization ratio because the new units were still in their testing and breaking- in periods. The ratios would be 1.5 and 59.4% respectively if the previous capacity - 142 thousand tons - were used. AINNE7 15_ - 64. - Page 4 of 4 The CEAGESP Subproject CEAGESP SuTmmary Income Statements Years 1974-1976 (Cr$ x 106 1974 1975 1976 iZver.uest Storage 80.4 141.3 Other services rendered 33.8 66.1 Merchandise Sales 8.0 8.9 Other revenues 5.1 3.2 Total 75.5 127.3 219.5 Operating Expenses: Costs of goods sold and 19.5 21.0 99.5 service rende7ed Provision for lepreciation 14.5 16.2 26.6 General administration ?1.2 35.3 52.3 Other 25.4 68.7 26.8 Total 80.6 141.2 205.2 Operating profits - 5.1 - 13.9 14.3 Non-operating income 3.2 3.8 3.4 Exchange Losses ar.d inventory - - 19.7 - 11.5 corrections - net. (see note) Miscellaneous non-operating - 1.2 0.1 - 0.1 Total non-operating 2.0 - 15.8 - 8.2 Net for year - 3.1 - 29.7 6.1 NOTE: Until 1975 the Company had accounted for exchange losses and monetary corrections on the balance sheet, i.e. by reflecting the increased value of its indebtedness in the asset accounts. In compliance with accounting regulations, however, it showed such increases as non-operational provi- sions against income beginning in 1975. __ _ IBRD 3856 MAY 1972 V E N E Z U E L. A SURNAM RENCH GUIANA/ C OL O0MSB A..../c sov GUYANA GR A lN STORAGE PROJECT CORNý RIC E a WH E AT (S OYBE AN) ADISTRIBUTION LFERTAMEZA ~I s R/o1GRNDEN6 1 -, I A LI * RECIFE -4~0 MrE ,NAIALBUC --- --------uc RIO BRANCOO MACEIO I s s o t• es. v 'o MM0kNOA E T 4 R-~AN A"'- - . Corn '4"f"" 'ANO. 0 Approximote north boundary of Project Area . State boundaries ••w -- I-.ernationl boundaries q ~~~~ALEGRE h omuistanu hsmpd o à un9i Cendorsementor*accep neeNAyAth -pl b, Ciiohn,Tcns 0 200 420 600•wtn -30- KILOMETERS U R U G U A Y3 7. . one.n-tisol bondorie \ *.. 0 00511' >l,IC d,,,