Document of The World Bank FOR OFFICIAL USE ONLY Report No. 11491 PERFORMANCE AUDIT REPORT MOROCCO FIFTH AND SIXTH AGRICULTURAL CREDIT PROJECTS (LOANS 2367-MOR AND 2731-HOR) DECEMBER 30, 1992 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. CURRENCY EQUIVALENTS Currency Unit - Moroccan Dirham (DH) (year end) 1982 US$1.00 = DH6.3 1983 = 8.1 1984 = 9.6 1985 = 9.6 1986 = 8.7 1987 = 7.8 1988 = 8.2 1989 = 8.1 1990 = 8.0 GLOSSARY ADB - African Development Bank BNDE - Banque Nationale de Developpement Economique CIH - Cr4dit Immobilier et H6telier (Housing & Tourism Bank) CLCA - Caisse Locale de Crddft Agricole CNCA - Caisse Nationale de Cr6dit Agricole CRCA - Caisse R6gionale de Cr6dit Agricole EDI - Economic Development Institute EDP - Electronic Data Processing EEC - European Economic Community ERR - Economic Rate of Return FADES - Arab Fund for Economic and Social Development FDA - Agricultural Development Fund HQ - CNCA Headquarters IRR - Internal Rate of Return KfW - Kreditanstat fur Wiederaufbau, Germany M&E - Monitoring and Evaluation MARA - Ministry of Agriculture and Agrarian Reform MIS - Management Information System OED - Operations Evaluation Department PAR - Performance Audit Report PC - Personal computer PCR - Project Completion Report SDI - Subsidy Dependence Index USAID - United States Agency for International Development FISCAL YEAR Before 1987 September 1 - August 31 After 1987 January 1 - December 31 Borrower Caisse Nationale de Cr6dit Agricole FOR OFFICIAL USE ONLY THE WORLD BANK Wshington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 30, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on MOROCCO Fifth and Sixth Agricultural Credit Projects (Loans 2367-MOR and 2731-MOR) Attached is a copy of the report entitled "Performance Audit Report on Morocco: Fifth and Sixth Agricultural Credit Projects (Loans 2367-MOR and 2731-MOR)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only In the performance of their offcLal duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY PERFORMANCE AUDIT REPORT MOROCCO FIFTH AND SIXTH AGRICULTURAL CREDIT PROJECTS (Loans 2367-MOR and 2731-MOR) TABLE OF CONTENTS Page No. Preface .......................................................... i Basic D ata Sheet ..................................................... iii Evaluation Summary .................................................. vii I. Background .................................................... 1* H. Design and Implementation ........................................ 2 A. Project Design ............................................... 2 B. Project Implementation ......................................... 3 El. Project Outcome ................................................ 6 A. Farm-Level Impact ............................................ 6 B. CNCA's Institutional Development ................................ 7 C. CNCA's Financial Performance ................................... 9 D. Impact on Sectoral Issues ....................................... 17 IV. Findings and Issues .............................................. 19 A. Overall Assessment of the Projects ................................ 19 B. Outstanding Issues ............................................. 21 Annexes: 1. Financial Indicators: Tables 1-4 ...................................... 25 2. CNCA Commitments, Repayments and Arrears: Tables 1-5 ................. 29 3. CNCA Lending by Category of Investment: Tables 1-2 .................... 34 Tables in Text: 1. Appraisal vs Actual Lending by CNCA ................................ 4 2. Internal Rates of Return Estimated by CNCA's 1984 Farm Survey and by the 1986 Appraisal Report .................................. 7 3. Construction and Rehabilitation of CNCA Branches ...................... 8 4. Number of CNCA Staff Members Receiving Training ..................... 8 5. In-service Training for CNCA Headquarters and Field Staff ................. 9 6. Estimated and Actual Number of Sub-loans Extended by CN CA ..................................................... 10 7. Net Disbursements to CNCA Borrowers, 1985-1990 ....................... . 10 8. Average Sub-Loan Amounts Extended by HQ/CRCAs and CLCAs ........... 11 9. CNCA's Financial Ratios ........................................... 13 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 No. 10. CNCA Loan Recoveries and Arrears .................................. 15 11. CRCA and CLCA Recovery Rates by Region, 1990 ....................... 15 12. CNCA Interest Rates Compared with Inflation and Market Rates ............ 18 Appendix: Comments from the Borrower ........................................... 37 Map IBRD 19570 -1- PERFORMANCE AUDIT REPORT MOROCCO FIFTH AND SIXTH AGRICULTURAL CREDIT PROJECTS (Loans 2367-MOR and 2731-MOR) PREFACE This is a Performance Audit Report (PAR) on the Fifth and Sixth Agricultural Credit Projects in Morocco, involving IBRD loans in the amount of US$115.4 million and US$120 million, respectively. The objective of both projects was to increase agricultural production by financing on- farm investments, to improve access to credit for small-scale farmers and farmers in rainfed areas, and to strengthen the Caisse Nationale de Cr6dit Agricole (CNCA) through support for institutional development. The Fifth Agricultural Credit Project (Loan 2367-MOR) was approved on December 13, 1983, became effective on May 11, 1984 and was fully disbursed on March 17, 1987, five months before the original closing date of August 31, 1987. The Sixth Agricultural Credit project (Loan 273 1-MOR) was approved on June 26, 1986, became effective on December 10, 1986 and was fully disbursed in February, 1990, one month before the original closing date of March 31, 1990. The PAR is based on the Project Completion Report (PCR) prepared by the Middle East and North Africa Region, Agriculture Operations Division in Country Department I, the Borrower's Project Completion Reports for the two projects (translated copies of which can be found in Regional files), the Staff Appraisal Reports, the President's Reports, the loan documents, the transcripts of the Executive Directors' meetings at which the projects were considered, on a study of project files, and on discussions with Bank staff. An OED mission visited Morocco in December, 1991, and discussed the effectiveness of the Bank's assistance with CNCA and other relevant Government agencies. Their kind cooperation and valuable assistance in the preparation of this report is gratefully acknowledged. The PCR provides a good account and assessment of the project experience, and discusses the performance of the Bank and the project executing agencies. The PAR elaborates on particular aspects such as the evolution of CNCA financial ratios over the course of the two projects (1984- 1990) and the persistent problem of arrears. Detailed and helpful comments on the draft PAR were received from the Borrower on October 1, 1992, and are attached as an Appendix.  - Ill - PERFORMANCE AUDIT REPORT MOROCCO FIFTH AGRICULTURAL CREDIT PROJECT (Loan 2367-MOR) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as Z of estimate estimated actual appraisal estimate Project Costs (DH an) 601.9 n.a. 108 Loan Amount (US$ mn) 115.4 115.4 100 Number of Sub-loans 580,000 965,000 166 Internal Rates of Return 122-44% n.a. 1 Actual compared with estimated CNCA disbursements 1984-1990. PROJECT DATES Activity Original Plan Revisions Actual Appraisal 03/83 Negotiations 10/18/83 Board Approval 12/13/83 Loan agreement 03/01/84 Loan effectiveness 06/01/84 05/11/84 05/11/84 Closing 08/31/87 08/31/87 Completion 08/31/87 03/87 CUMULATIVE DISBURSEMENTS FY84 FY85 FY86 FY87 FY88 Appraisal estimate (US$ mn) 10.0 38.0 78.0 113.0 115.4 Actual (US$ million) 15.4 47.0 106.0 115.4 Actual as % of estimate 154.0 124.0 136.0 102.0 Date of final disbursement March 17, 1987 STAFF INPUTS FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY91 TOTAL Preappraisal 2.2 0.7 6.5 9.4 Appraisal 34.3 12.8 47.0 Negotiation 8.8 8.8 Supervision 6.6 9.1 0.8 5.6 1.8 23.8 Other 0.1 0.7 3.6 9.8 14.2 Total 2.3 1.4 44.4 37.9 9.1 0.8 5.6 1.8 103.2 - iv - MISSION DATA No.of Mandays Specializations Performance Types of Mission Dates Persons in field represented I rating 1 Trend Problem Appraisal 3/83 3 60 E A FA Supervision 1 2/84 1 4 E 1 2 Supervision 2 7/84 1 9 E 1 2 Supervision 3 12/84 1 12 FA 1 1 Supervision 4 3/85 1 11 F Supervision 5 7/85 1 13 F Supervision 6 7/86 1 10 FA 1 Subtotal 59 TOTAL 119 COMPLIANCE WITH COVENANTS See PAR para 2.15; PCR paras 3.25-3.30. OTHER PROJECT DATA Borrower Caisse Nationale de Crddit Agricole (CNCA) Follow-up Project Name Sixth Agricultural Credit Project Loan number 2731-MOR Amount US$120 million Date Board Approval 06/26/86 11 E = Economist; FA = Financial Analyst; A = Agronomist. 1 = problem-free or minor problems; 2 = moderate problem; 3 = major problems. 1 Some of this time was spent in preparation of the Sixth Project. - V - PERFORMANCE AUDIT REPORT MOROCCO SIXTH AGRICULTURAL CREDIT PROJECT (Loan 2731-MOR) BASIC DATA SHEET KEY PROJECT DATA Appraisal Actual or Actual as I of estimate estimated actual appraisal estimate Project Costs (DH in) 720.3 n.a. 108 it Loan Amount (US$ mn) 120.0 120.0 100 Number of Sub-loans 263,000 655,000 249 Internal Rates of Return 15%-54% n.a. l Actual compared with estimated CNCA disbursements 1984-1990. PROJECT DATES Activity Original Plan Revisions Actual Appraisal 01/86 02/86 10/85 Negotiations 06/86 07/86 05/02/86 Board Approval 07/86 09/86 06/26/86 Loan agreement 09/17/86 Loan effectiveness 12/16/86 12/86 12/10/86 Loan amendment 03/06/90 03/06/90 Closing 03/31/90 03/31/90 Completion 09/30/89 02/90 CUMULATIVE DISBURSEMENTS FY87 FY88 FY89 FY90 Appraisal estimate (US$ mn) 15.0 50.0 95.0 120.0 Actual (US$ million) 32.2 68.2 112.7 120.0 Actual as % of estimate 213.0 136.0 118.0 100.0 Date of final disbursement February 7, 1990 STAFF INPUTS 1985 1986 1987 1988 1989 1990 1991 TOTAL Preappraisal 8.7 12.2 20.9 Appraisal 32.8 32.8 Negotiation 4.8 4.8 Supervision 0.6 2.2 4.3 2.0 7.6 7.3 24.1 Other 1.5 11.3 0.9 13.7 Total 10.2 61.7 3.1 4.3 2.0 7.6 7.3 96.2 - Vi MISSION DATA No.of Mandays SpecializationF Performance Types o Mission Dates persons in field represented rating ' Trend 11 Problem - Appraisal 11/85 5 125 1I E A FA Supervision 1 4/87 4 51 EN E FA 0 Supervision 2 11/87 1 8 FA Supervision 3 4/88 6 108 FA Al 0 Supervision 4 7/88 1 6 EN Supervision 5 8/89 3 17 FA EN 0 Supervision 6 2/90 5 9 LS FA Al 0 Subtotal 199 TOTAL 324 COMPLIANCE WITH COVENANTS See PAR para 2.15; PCR paras 3.25-3.30. OTHER PROJECT DATA Borrower Caisse National de Crddit Agricole (CNCA) Follow-up Project Name National Agricultural Credit Project Loan Number 3088-MOR Amount US$190 million Date Board Approval 06/14/89 E = Economist; FA = Financial Analyst; A = Agronomist; EN = Engineer; Al * Agro-Industry; LS = Livestock Specialist; 0 = Other. Form 590s were not available in files. Some of this time was spent on supervision of the Fifth Project. Some of this time was spent on supervision of the follow-up project. - vii - PERFORMANCE AUDIT REPORT MOROCCO FIFTH AND SIXTH AGRICULTURAL CREDIT PROJECTS (Loans 2367-MOR and 2731-MOR) EVALUATION SUMMARY Introduction nents were to support medium- and long-term farm credit, with an emphasis on smaller farmers 1. In the mid-1980s, the Government of and on rainfed areas. The institutional develop- Morocco embarked upon a program of adjust- ment components were to support upgrading of ment for the agricultural sector. The objective the management information system (MIS), of the program was to improve self-sufficiency in better electronic data processing (EDP), staff agricultural products and reduce social and training, expansion of the branch network, regional disparities. This was to be achieved by increased deposit mobilization and development redirecting public investment towards activities of medium-term banking strategies. with high returns and promoting private invest- 3. The Bank loan of US$115.4 million for the ment by adjusting agricultural prices and incen- fifth project was 19% of the total project costs tives, strengthening agricultural support services estimated at US$601.9 million. Co-financing by and providing more agricultural credit. The the African Development Bank (ADB), the Fifth Agricultural Credit Project (Loan 2367- Arab Fund for Economic and Social Devel- MOR) and the Sixth Agricultural Credit Project opment (FADES) and the German aid agency, (Loan 2731-MOR) supported these objectives by Kreditanstalt fur Wiederaufbau (KfW), account- providing financial and technical assistance to ed for another 14% of appraisal project costs. the Caisse Nationale de Cr6dit Agricole These funds were to support CNCA's operations (CNCA), which is the main channel through during the 1985 and 1986 fiscal years. The total which agricultural credit is delivered. The project costs for the sixth project were estimated Bank's co-operation with CNCA dates back to at US$720 million at appraisal. The Bank loan the First Agricultural Credit Project (Loan 433- of US$120 million accounted for 17% and the MOR), approved in 1965, and is currently con- co-financiers, ADB and KfW, financed an addi- tinuing under the National Agricultural Credit tional 14%. The sixth project was to support Projct (oan 088-OR).CNCA's operations during the 1987, 1988 and Project (Loan 3088-MOR).years. Objectives Implementation Exprience 2. The objectives of the two projects were to increase agricultural production and to strength- t Bth project became te sche en CNCA's management and financial position. if pr in M,8a the xth o These objectives were addressed by a credit delays at the start of the fifth project, due to the component and an institutional development drought of 1984/85, disbursements were general- component in each project. The credit compo- ly ahead of schedule. The fifth project was fully - viii - disbursed five months before the original closing system to process the data. In 1987, a more date of August 31, 1987 and the sixth one was dynamic management assumed control of CNCA fully disbursed one month before the original and greater efforts and progress were made on closing date of March 31, 1990. computerization of CNCA branches, training, deposit mobilization, planning and overall man- 5. Implementation of the credit components agement. The Bank provided valuable assistance went quite smoothly, although accumulated in its supervision missions by encouraging appro- arrears increased sharply during implementation priate reforms at CNCA. due to consecutive years of drought in the early 1980s. As a result of the arrears, CNCA became 8. Compliance with covenants was mixed. concerned about its liquidity and the Bank Heavy borrowing from abroad during the fifth opened a Special Account to reduce the time it project resulted in the debt-equity ratio exceed- took to re-finance CNCA loans. The account ing the agreed ceiling of 6:1, while, during the was kept in local currency, which is an exception sixth project, the need to increase provisions for to normal Bank practices, but it was used only bad debts resulted in a return on equity below once because CNCA obtained loans from other the agreed minimum of 12%. Although the external donors to address its liquidity problems. PCR contends that CNCA also failed to meet Although CNCA requested a Special Account the current ratio requirement of at least 1.2:1 for the sixth project as well, the account proved during the fifth project, the Audit concurs with unnecessary and was never opened. CNCA's argument (see Appendix) that the required ratio was met (the issue is one of how 6. While CNCA did not keep any figures on current assets and current liabilities are mea- project costs, CNCA commitments during the sured). two projects exceeded appraisal estimates by 8%. Lending for livestock was well above appraisal Results estimates, whereas lending for land improvement and irrigation was below expectations. This 9. As mentioned earlier, the M&E unit pro- probably reflects the fact that livestock invest- vided little information on which to base an ments provide a quick and steady return, which assessment of the farm-level impact of the is important in rainfed areas, as well as problems project. On the other hand, the duty of calculat- relating to insecure land tenure and fragmenta- ing ex-post economic rates of return (ERRs) was tion of landholdings, both of which discourage assigned to the M&E unit of the Ministry of fixed on-farm investments. Agriculture and Agrarian Reform (MARA) under the fifth project, and presumably also 7. Implementation of the institutional develop- under the sixth project, although the MARA ment components proved more challenging. unit is not mentioned in the staff appraisal While CNCA introduced a very successful train- report (SAR) for the sixth project. There is no ing program in 1987, construction and upgrading evidence in project files that MARA calculated of the branch network proceeded less rapidly any ERRs. IRRs estimated on the basis of a than expected, and delays in reaching a decision farm survey undertaken by CNCA in 1984 on the most suitable hardware and software for suggest returns of 15-27%. These figures are the EDP system, as well as on the information lower than SAR estimations for the sixth pro- structure for the MIS, held up the introduction ject, however the numbers are not strictly coi- of the new systems. The Monitoring and Evalu- parable. Investments generated sufficient re- ation (M&E) unit produced little analytical turns to permit a steady rise in the real volume output, although it collected a substantial of repayments per borrower, but were probably amount of farm-level data. The M&E unit insufficient to repay both current and accumulat- explained that it was waiting for the new EDP ed dues. dyamc-angeen-asuedcotrl-f-NC 10. Progress was made on some aspects of was overcome by soft loans and grants from the CNCA's institutional development. In spite of EEC and USAID, and CNCA was able to meet delays in the construction program, 29 branches its obligations, reschedule loans in arrears and (including rented facilities) were added to the continue to expand its loan portfolio. Although CNCA network and an additional 12 branches CNCA requested that a special fund be estab- were upgraded. The number of CNCA staff lished in the course of the fifth project to speed benefitting from the training program each year up Bank refinancing of CNCA disbursements, rose from 13% in 1987 to 41% in 1990. The the special fund was hardly used. Higher than success of the program has led the Bank's Eco- expected foreign borrowing pushed the debt- nomic Development Institute (EDI) to suggest equity ratio above the 6:1 ratio stipulated in a broader, regional training role for CNCA. The covenants for the fifth project, but the ratio first stage of a new MIS is to be introduced in remained below the 8:1 ceiling set under the June, 1992. After several years of delays and of sixth project. After 1988, a sharp increase in disagreements over the design of the MIS with provisions for bad debt lowered the return on both local and foreign technical consultants, CNCA's equity but placed CNCA on a sounder CNCA is reaching a decision on the structure of financial footing. CNCA currently lends at rates the information system and is writing its own that are positive in real terms, although they are programs to run the EDP system. In the mean- on balance slightly below market rates. It has time, personal computers have been introduced also begun to assume a greater portion of the in twenty regional branches (CRCAs). foreign exchange risk on its external borrowing. 11. CNCA has also made some progress with Sustainability regard to its financial performance, although a number of issues remain to be tackled. While 13. n tainabioity hine. on ai CNCA successfully diversified its portfolio from declined from 67% in 1985 to 58% in 1990. In strictly agricultural credit to rural credit, it is not 1990, CLCAs recovered only 27% of overdue clear whether the goals of increasing penetration balances. Following the droughts of the early in rainfed areas and of providing more credit to 1980s, approximately 70% of CNCA's arrears small farmers were met. On the one hand, local were rescheduled in 1985 and another 22% of level branches (CLCAs), which only serve small dues in 1986. It is disconcerting that the rela- farmers, saw their clientele increase by 22% over tively good harvests of the late 1980s have not the period 1985-1990. On the other hand, permitted CNCA to contain the problem of CLCAs accounted for a decreasing share of total arrears. While the Audit is convinced that CNCA disbursements (32% in 1985, 20% in CNCA has made serious efforts to improve its 1990), and repayments exceeded disbursements recovery rate, it is concerned that, at this junc- at the CLCA level after 1988 whereas net dis- ture, a repeat of the droughts of the 1980s (as is bursements remained positive at the Headquar- occurring in the current season) would sharply ters (HQ) and CRCA levels, which serve largerand entail clients. Deposit mobilization accelerated sharply rat i oes asvithap es the inc1991, after 1988, rising by almost 30% per annum. Ca illve madeesignificant prgrs Nevertheless, the share of deposits in total CC ilhv aesgiiatpors Nevetheess,theshar ofdepoitsin ttal towards sustainability. An index of subsidization CNCA resources barely increased over the calculated for CNCA suggests that CNCA's period 1985-1990 and CNCA continues 'o have reliance on subsidies fell by almost 50% over the only one depositor for every seven borrowers. course of the two projects. While such progress at CNCA is encouraging, the broader issue of 12. As a result of a sharp increase in arrears developing sustainable rural financial markets in during the droughts of the early 1980s, CNCA Morocco has yet to be addressed. experienced a shortage of working capital, which - Findings and Lessons dix). If clients are unable to repay in spite of rescheduling and of fresh loans, then CNCA 14. Overall, both projects are assessed as may face considerable write-offs down the satisfactory. Although progress was slow on road. some of the objectives set out in the fifth and sixth projects, CNCA has developed into a Analysis by the M&E unit may also yield stronger financial institution and the farmers insights into how high CNCA's on-lending appear to have made good use of the loans. rates can rise without significant adverse The Government, CNCA and the Bank can effects on loan risk. This will be important if ensure that it is strengthened further by taking CNCA is to pass an increased share of the measures to ensure that CNCA's spread reflects the greater risks inherent in loans to the agricul- ign exchngsk tural sector, and by making further progress on the issues of arrears, deposit mobilization, intro- The Bank, which has provided one-quarter of duction of an improved MIS and more active CNCA's liabilities, must ensure that its lend- monitoring and evaluation. ing does not reduce CNCA's incentive to 15. The Audit draws the following lessons tackle the issue of deposit mobilization. This from the projects: consideration should be weighed in the deter- mination of loan amounts under future pro- * Since progress is likely to be slow on the jects. development of the new MIS and EDP systems, computerization of CNCA branches Insecu and data analysis by the M&E unit should be landholdings continue to constitute an imped- promoted even in the absence of an iment to agricultural investment and may ac- enhanced MIS. count for disbursements for irrigation and land improvement below appraisal expecta- * While the fifth project sought to prevent tions. duplication of efforts by MARA and CNCA on evaluation, stating that ERR calculations u T he ai nl supprin CNCA would be the duty of MARA's M&E unit, it uer thent A redit Pro- appears that neither MARA's nor CNCA's ject eharedit octrn unit has attempted to evaluate the degree to The proect's e a i oig which the cash flow projections that deter- CNaitomte bakigsect as ahigh mined CNCA loan commitments have indeed t , competi fulliank.aWhile materialized at the farm level and how the th i e o beninglrura finanialar- returns on farm investments have been used. ket a yet to be epreentatie plan Such information is essential for effective risk for a the curre ofrtioneeu. management and containment of the arrears T he re o t ned problem. Teeoe hl otne upr o problem.CNCA should be assured, further consider- * The problem of arrears calls for a more ation of the role that alternative institutions active role for CNCA's M&E unit in other can play in mobilizing savings and extending respects. In particular, it is important to credit in Morocco's rural financial markets is continue to monitor the success of the 1985 warranted at this juncture. rescheduling, as CNCA is doing (see Appen- PERFORMANCE AUDIT REPORT MOROCCO FIFFH AND SIXTH AGRICULTURAL CREDIT PROJECTS (Loans 2367-MOR and 2731-MOR) I. BACKGROUND 1.1 In the late 1970s, a sharp decline in the price of Morocco's main export -phosphates - coupled with the second oil shock placed considerable strain on the Moroccan economy. The trade imbalance worsened in the early 1980s as the US Dollar appreciated and a drought in 1981 entailed larger imports of foodstuffs. The fiscal deficit also grew as a result of significant producer and consumer subsidies, as well as the cost of the conflict in the Western Sahara. Structural adjustment began with IMF assistance in the late 1970s and in 1983 the World Bank began to support industrial and trade policy adjustment in Morocco. Nevertheless, per capita incomes have remained stagnant throughout the 1980s. 1.2 In 1985, the agricultural sector accounted for about 18% of GNP, 28% of exports and 40% of employment. It is characterized by a dual structure: the irrigated sub-sector covers about 14% of the 7.3 million hectares of cultivable land, 45% of total agricultural production (primarily fruits and vegetables) and 60% of agricultural exports, and has grown at an annual rate of about 5.5%. The rainfed sub-sector, which employs 75% of the rural population, has grown less rapidly than population. Since this sector accounts for most of Morocco's production of cereals, the need for imports of cereals has grown, rising from 25% of annual consumption in 1978 to 40% in 1984. This weak performance is attributable to four years of droughts in the early 1980s, as well as a concentration of government subsidies, investment and support services (including credit) on the irrigated sector, poor farming practices, fragmentation of land holdings and insecure land tenure. 1.3 In 1984 the Government of Morocco embarked upon an agricultural sector adjustment program which was supported by Bank sector loans in FY85 and FY88. It sought to improve self- sufficiency and reduce social and regional disparities by adjusting agricultural prices and incentives, strengthening agricultural support services, redirecting public investment towards activities with high returns and promoting private investment by providing more agricultural credit. 1.4 The Caisse Nationale de Cr6dit Agricole (CNCA) is the main channel through which agricultural credit is delivered, accounting for about 80% of all agricultural lending in Morocco. Established in 1961, CNCA has steadily extended its network of regional (CRCA) offices and local (CLCA) branches across the country. (The CLCAs service the smaller farmers - those with annual agricultural income less than about US$725.) It has also broadened its activities considerably over the past decade, opening guichets bancaires in urban areas that offer full banking services and mobilize urban savings, and diversifying lending in rural areas. Thus lending for agro-industries, rural housing, fisheries and rural non-farm activities accounts for a growing share of CNCA's portfolio. CNCA has developed into a stronger financial institution, lending at positive real rates of interest and making profits after adequate provisions for bad debts. Further progress is required on improving the accounting information system and on increasing loan recovery rates, which remain below the levels recorded before the years of drought. -2- 1.5 CNCA has benefitted from considerable support from the Bank and other external donors. The Bank has financed seven agricultural credit projects in Morocco since 1965, for a total of US$574.4 million. The first four sought to increase agricultural production and, beginning with the second project, to strengthen CNCA's financial position and provide modest support for institutional development. Lending expanded from medium- and large-scale farmers under the first two projects to cooperatives and small-scale farmers under the third project and agro-industries under the fourth one. Although performance was mixed under the first project, subsequent projects were considered satisfactory by OED. The fifth and sixth agricultural credit projects are the subject of this audit. A seventh project, the National Agricultural Credit project ([oan 3088-MOR), is currently under implementation. II. DESIGN AND IMPLEMENTATION A. Project Design 2.1 The Bank's growing confidence in CNCA is reflected in the fact that both projects were prepared by CNCA, with assistance from Bank supervision missions. The overall objectives of the two projects were to increase agricultural production and to strengthen CNCA's management and financial position. The PCR also notes that the projects were to be vehicles for removing subsidies to CNCA and its clients. These objectives were addressed by two components in each project: a) credit for on-farm investments, agro-industries and (in the sixth project) rural housing; b) institutional development through support for an improved management information system (MIS), better electronic data processing (EDP), staff training, expansion of the branch network, increased deposit mobilization, monitoring of financial indicators, maintenance of positive real rates of interest to sub- borrowers, development of medium-term banking strategies, and (in the sixth project) a redistribution of the foreign exchange risk towards CNCA and its clients. These objectives and components are described in greater detail in the PCR (paras 2.07-2.12). 2.2 The Bank loan of US$115.4 million for the fifth project was 19% of the total project costs estimated at US$601.9 million. Co-financing by the African Development Bank (ADB), the Arab Fund for Economic and Social Development (FADES) and the German aid agency, Kreditanstalt fur Wiederaufbau (KfW), accounted for another 14% of appraisal project costs. The total project costs for the sixth project were estimated at US$720 million at appraisal. The Bank loan of US$120 million accounted for 17% and the co-financiers, ADB and KfW, financed an additional 14%. 2.3 The fifth project calls for a more aggressive approach to reaching small-scale farmers with agricultural credit, increasing deposit mobilization, improving monitoring and evaluation and enhancing CNCA's financial viability. An average increase of nearly 2% in on-lending rates was secured in September, 1983, as a condition of Board presentation, and CNCA committed itself to reviewing the adequacy of these rates annually and to improving financial ratios. Specific targets were established for the current ratio (no less than 1.2), the debt-equity ratio (no greater than 6:1) and the return on equity (no less than 10% in fiscal 1984, 12% thereafter). Although several other policy issues were raised early in the course of project design, they were dropped by the time the project was presented to the Board. In particular, a component designed to support a cadastral survey to help tackle the problems of uncertain land tenure and fragmentation of land holdings was dropped at the Government's request. Other issues related to agricultural sector policy were postponed to -3- allow for further study and were addressed in the context of the 1985 Agricultural Sector Adjustment Loan (2590-MOR). 2.4 The sixth project's appraisal mission again focussed on reaching a larger number of farmers, particularly in rainfed areas, on strengthening CNCA's financial position and on increasing deposit mobilization. Targets were retained for financial ratios, although the debt-equity constraint was relaxed to 8:1 to allow for more external financing and the definition of current assets and liabilities in the current ratio was changed from maturities of less than twelve months to maturities of less than eighteen months.! The minutes of negotiations record a commitment by CNCA to increase deposit mobilization to 20% of total resources, although this requirement was not included in the loan agreement. A further commitment was obtained to introduce new lending appraisal procedures, which had been tested under Bank-supported agricultural development projects in specific regions of Morocco, in at least two more regions (paras 3.11-3.12). A new feature under the sixth project was cautious support for diversification of lending activities into rural housing. 2.5 Early in the course of project design, the Bank came to be concerned about the constraints placed on CNCA operations by the 1961 Dahir (law) that established CNCA. Before the project was presented to the Board, the goal of a thorough review of the 1961 Dahir and its possible replacement with a new charter was altered in favor of consideration of these issues by an upgraded management committee as a precondition for the next (seventh) project. Appointment of the management committee with adequate powers was made a condition of Board presentation of the sixth project. B. Proiect Implementation 2.6 General performance: Both projects became effective on schedule, the fifth project in May, 1984 and the sixth one in December, 1986. Project implementation seems to have gone quite smoothly for both projects, except for the introduction of the new MIS and EDP and the growing problem of arrears. After debating the merits of various information systems for several years, an agreement appears to have been reached on the implementation of a decentralized system to be introduced in stages, beginning in June, 1992 (paras 3.8, 4.8). With regard to arrears, the Audit mission was impressed by the seriousness of CNCA's efforts to monitor and improve the recovery rate, although these have yet to yield significant improvements in collections. The persistence and gravity of the arrears problem are discussed in paras 3.19-3.26 and 4.9. 2.7 Special Account: The arrears problem had important consequences for CNCA's liquidity, leading CNCA to request the opening of a special account under the fifth project in August, 1985. The Bank acceded to the request. The account was kept in local currency, which is an exception to normal Bank practices. Although it was used only once (in April, 1986) because funds became available from other external sources, the same arrangement was requested with higher ceilings for the sixth project. In the end, the special account for the sixth project was never opened and CNCA pre-financed all expenditures. 2.8 Project Costs: Since the Borrower has not provided figures for actual project costs, comparisons with appraisal estimates are not possible. Nevertheless, figures are provided below in .1 In spite of the new definition (Article IV, Section 4.04 of the Loan Agreement for the sixth project), CNCA has continued to include only assets and liabilities with maturities of less than 12 months in its measure of the current ratio (see Appendix). -4- Table 1 and in Annex 3 of the Audit which exclude the borrowers' contributions and Government subsidies, and compare sub-loan amounts estimated at appraisal with actual CNCA disbursements over the period September, 1984, to December, 1990, by category of expenditure. 2.9 CNCA lending exceeded appraisal estimates by 9.4% over the life of the two projects. While both appraisal missions were careful to note that the demand for credit could not be estimated with a great degree of accuracy, it turns out that their overall projections were quite accurate. An item by item comparison, however, reveals significant differences between expected and actual distributions of loans (Annex 3). Demand for irrigation was consistently well below expectations, whereas demand for rural housing loans under the sixth project greatly exceeded appraisal estimates. The most notable differences are with respect to livestock and greenhouses, both of which yield quick returns to farmers. Lending for both activities was more than twice the appraisal estimates during the fifth project and demand for livestock continued to be strong in the sixth project, although demand for greenhouses declined as a result of marketing difficulties (para 3.25). Lending for livestock accounted for 61% of all lending by CLCAs under the fifth project, rising to 70% of all CLCA lending under the sixth project. This is explained by the widespread pattern of cereal and livestock production in rainfed areas, with livestock providing a steady return as well as some insurance against a poor harvest. Medium-term financing for livestock thus proved attractive, and the Audit mission suspects that more of the credit demand for livestock could have been met with short-term rather than medium-term loans (para 4.7). Overall lending through CLCAs declined, as is explained in para 3.10. Table 1: APPRAISAL VS. ACTUAL LENDING BY CNCA (DH Millions) Actual ------ Appraisal ------- Actual as 2 of CNCA V CNCA VI Total Total Appraisal On-farm investments Constructions 390.2 874.0 1,264.2 641.5 51% Land improvement 44.4 205.5 249.9 144.4 58% Irrigation 302.3 734.9 1,037.2 450.1 43% Farm equipment 844.7 874.5 1,719.2 1,615.0 94% Draft animals 509.4 356.1 865.5 624.9 72% Livestock 957.5 1,234.6 2,192.1 3,473.8 158% Plantations 110.8 178.5 289.3 245.9 85% Greenhouses 62.5 111.9 174.4 395.8 227% Others 396.3 Sub-total 3,221.8 4,570.0 7,791.8 7,987.7 103% of which CLCAs: 1,463.2 2,248.0 3,711.2 3,540.1 95% Non-farm investments Agro-industry 149.3 190.0 339.3 444.7 131% Rural housing 148.0 148.0 449.5 304% Fisheries 38.9 Forestry .3 Handicrafts 15.6 Total 3,371.1 4,908.0 8,279.1 8,936.7 108% -5- 2.10 Disbursements: With the exception of delays at the start of the fifth project due to the drought of 1984/85, disbursements were generally ahead of schedule. In May, 1985, the Bank agreed to increase disbursement percentages for small-, medium- and large-scale farmers (loan categories A and C) because demand was growing and co-financiers' funds were being exhausted more rapidly than expected. The project was fully disbursed on March 17, 1987, five months before the original closing date of August 31, 1987. The Sixth Agricultural Credit project was also fully disbursed ahead of schedule, with the final disbursement made in February, 1990, one month before the original closing date of March 31, 1990. Lending for rural housing, which was a new activity, started up slowly under the sixth project but CNCA initiatives (including lower interest rates) sped up disbursements under this category. 2.11 Disbursements under both projects were broadly in line with appraisal estimates (PCR annex 2), with the exception of subloans to agrarian reform co-operatives under the fifth project (due to arrears and consequent ineligibility for further subloans) and the institutional development components of both projects. Eighty percent of Bank funds for institutional development in the fifth project and 96% in the sixth project were reallocated to the credit component. Since 10 CRCAs and 19 CLCAs were added with minimal Bank support over the life of the two projects to the 1985 network of 38 CRCAs and 99 CLCAs, and since CNCA's computer department states that it purchased 60 personal computers (out of 150 expected under the sixth project) without Bank funding, it appears that funding for the hardware elements of the institutional development components was not necessary. 2.12 Management Committee: As required in the loan agreement, a management committee was set up under the sixth project to help the board of directors develop a medium term strategy, but it did not prove to be useful. Soon after a new general manager was appointed at CNCA in 1987, a Bank supervision mission found a new dynamism characterized CNCA's approach to computerization, training, deposit mobilization and overall management, and the management committee became superfluous. The committee was disbanded with the Bank's approval (in favor of a stronger board of directors) and the loan agreement was amended accordingly. 2.13 Monitoring and Evaluation (M&E): Although an M&E unit had been set up in the course of the third project, it was disbanded in 1980 because it was not serving CNCA's needs. A new M&E unit was established in 1984 to assist management by obtaining data on selected parameters during sub-loan appraisals in order to provide a characterization of CRCA and CLCA clients, as well as to ensure proper application of lending guidelines at the branch level, improve sub-loan monitoring and investigate reasons for defaults. The appraisal report for the fifth project clearly states that measuring on-farm impact would not be the responsibility of CNCA's M&E unit, but rather that of an M&E unit in the Ministry of Agriculture and Agrarian Reform (MARA). There is no indication in project files that MARA's unit carried out this task. 2.14 Once a year, trained staff working in CRCA and CLCA branches survey 5% of all borrowers, in accordance with agreements reached under the fifth project. However the information gathered from these surveys, as well as from sub-loan appraisals, has yet to be exploited. The M&E unit contends that it has been waiting for the arrival of the new EDP system to analyze the data, but the Audit believes that CNCA is already sufficiently well equipped to conduct statistical analyses of the data and could have begun this process some time ago. The M&E obstacles are reflected in the uneven quality of the Borrower's version of the PCR. CNCA has indicated that M&E activities should gather momentum following the installation of the new MIS system in June, 1992. In the -6- meantime, findings in this Audit on farm-level impact are based on a farm survey conducted by CNCA in 1984 and made available to the Bank in 1986. These findings are discussed in paras 3.1-3.2. 2.15 Compliance with Covenants: There were several shortfalls with respect to financial covenants, reflecting liquidity problems at CNCA, the need to borrow more from external donors, the imposition of corporate income tax on CNCA beginning in 1988 and changes in accounting procedures resulting in greater provisions for bad debts and for foreign exchange risks. Although the provisions and the tax entailed a lower return on equity after 1987, with ratios below the 12% mandated by the loan agreement, the lower returns largely reflect more prudent financial policies rather than a worsening of CNCA's financial situation (paras 3.14-3.18). Thus, the PCR rightly concluded that financial covenants do not in and of themselves ensure sound financial practices (PCR para 4.03). The compliance of the Borrower and the Government with other loan covenants was mixed. Further details are provided in the PCR. 2.16 Bank Supervision: Biannual supervision missions consistently gave the fifth and sixth projects a rating of "1" (problem-free or minor problems). Supervision missions identified no major problems with the credit component and raised only minor concerns with the institutional development component. Arguably difficulties with the latter component were underestimated or understated by the supervision missions. For example, there were significant problems with the development of the MIS, including a bitter disagreement in 1985 between CNCA and consultants hired to develop the system. Although a supervision mission reported that differences had been reconciled in June, 1985, the contract with the consultants was severed less than two weeks later. Another issue, which is flagged in the PCR, is the lower appraisal estimate of the number of borrowers under the sixth project than under the fifth project. This reduction, together with the decline in disbursements through CLCAs after 1986, merited closer investigation by the Bank, as they may point to underlying problems such as increasing arrears. Nonetheless, Bank supervision missions have consistently encouraged reforms aimed at strengthening CNCA and have thus made an important contribution to CNCA's development. III. PROJECT OUTCOME A. Farm-level Impact 3.1 For reasons discussed above, no current data are available on the farm-level impact of the two projects. However, a 1984 survey of 472 farms covering 16,314 hectares in northwestern Morocco provides data on basic economic indicators for each farm: size, labor force, investments, activities, production, yields and marketing data. Based on these data, CNCA calculated internal rates of return (IRRs) for various investments - tractors, stables and cows, and harvesting equipment - and farms of various sizes operating under irrigated or rainfed conditions. Although the assumptions differ from those made in farm models developed at appraisal, vitiating direct comparisons between the two, figures from the farm survey and from the sixth appraisal report are presented below in Table 2. -7- Table 2: INTERNAL RATES OF RETURN ESTIMATED BY CNCA'S 1984 FARM SURVEY AND BY THE 1986 APPRAISAL REPORT FOR THE SIXTH AGRICULTURAL CREDIT PROJECT (percentage) -CNCA Farm Survey- Sixth Project Investment Farm Size Rainfed Irrigated Appraisal Tractor 20-50 ha 20 25 15.1 50-100 ha 19 21 >100 ha 25 27 Motor-pump 20-50 ha 20 50.4 & wells >50 ha 21 Stables 5-10 ha 25 10-20 ha 19 21 Combine >100 ha 26 harvesters Cows 10-50 ha 15-26 53.5 3.2 The lower rates calculated on the basis of CNCA's survey cannot be used to conclude that actual rates of return (IRRs) during the fifth and sixth projects were lower than appraisal estimates. What is certain, given the high rates of arrears in the early 1980s, is that under the fourth project successive droughts resulted in IRRs below appraisal expectations. Under the fifth and sixth projects, rates of return were sufficiently high to permit a 36% real increase in repayments per borrower over the period 1985-1990. But they were probably insufficient to permit full repayment of both current dues and accumulated dues. Since willful default was largely mitigated by CNCA's serious efforts to collect its loans, the more plausible explanation for continuing delinquency would seem to be the income factor. Field visits by the Audit mission in the center of the country indicated that farmers were aware of the relative merits of various production patterns and chose what appeared to be the best ones. B. CNCA's Institutional Development 3.3 Construction: Both the fifth and the sixth projects included support for the construction and/or rehabilitation of CNCA branches (Table 3 below). Loan disbursements for this purpose were well below appraisal estimates, although CNCA's expenditure on construction over the life of the two projects - about US$7 million - was close to appraisal figures. There were considerable delays in the implementation of the construction program, which the PCR attributes to administrative and procurement problems, as well as to an overly ambitious schedule. Nonetheless, CNCA added 10 CRCAs and 19 CLCAs to its branch network over the period 1984-90 through a combination of rented facilities and new constructions. In addition, CNCA has increased its outreach in remote areas at little expense by opening over 150 seasonal banking windows (guichets saisonniers) in MARA facilities. These banking windows channel loan applications and disbursements between farmers and the nearest CLCA branch. Both CNCA's Auditors and the Bank underlined the importance of upgrading CNCA facilities and the Bank urged CNCA to establish a task force to focus on expanding and upgrading CNCA's network. Provisions have been made under the follow-on project for construction and/or rehabilitation at a cost of US$3.5 million, although CNCA has indicated that the current program is still quite ambitious and may experience delays. -8- Table 3: CONSTRUCTION AND REHABILITATION OF CNCA BRANCHES Completed, Carried over Additional under 5th Project To 6th Project 6th Project Total New CNCA branches 8 9 12 29 Expanded branches 1 4 0 5 Rehabilitated branches 1 6 0 7 3.4 The most significant delays to date have been with respect to the construction of the new headquarters. With CNCA offices spread across 11 buildings in Rabat, there is an acute need for an appropriate facility. A prestigious new site has been selected and the project has been supervised closely by the Royal Palace, which is expected to authorize the start of construction work in 1992. 3.5 Training: As noted in the PCR, CNCA has paid great attention to training its staff. The Audit reviewed the training program and discussed training issues with staff at headquarters and in the field. Training has reached an increasing proportion of CNCA staff (Tables 4 and 5 below) and the feedback received by the Audit has been very positive. Table 4: NUMBER OF CNCA STAFF MEMBERS RECEIVING TRAINING 1987 1988 1989 1990 Basic training 99 47 104 78 In-Service training in Morocco 285 783 920 1087 abroad 35 45 32 150 Specialized training 52 78 116 139 Total 471 953 1172 1454 3.6 In the course of the fifth project, Bank-financed consultants prepared a training program that benefitted 430 professionals. CNCA subsequently elaborated a medium-term strategy to address the training needs identified in a survey of CNCA management and staff. A training department was established in 1987 and has been offering introductory training courses to new staff, in-service training both in Morocco and abroad to upgrade skills, .and specialized training for selected staff members. The specialized training leads to a degree from a university or training institute and is paid for by CNCA if the staff member graduates (otherwise the staff member must pay for the training). Feedback is obtained regularly from management and staff to ensure that training is of good quality and addresses CNCA's needs. Initially the program focussed on staff at headquarters, but the number of field staff receiving training has grown steadily. -9- Table 5: IN-SERVICE TRAINING FOR CNCA HEADQUARTERS AND FIELD STAFF Total Number of Staff Staff Trained as CNCA Staff Trained each Year Percent of Total Mg Field Total M Field Total RQ Field Total 1987 808 1666 2474 320 320 40% 13% 1988 935 1797 2732 475 353 828 50% 20% 30% 1989 961 1930 2891 380 572 952 4 '% 30% 33% 1990 845 2159 3004 323 904 1237 38% 42% 41% 3.7 The training strategy was revised in 1991 to focus on ways to improve CNCA's internal co- ordination and to introduce new training programs as CNCA's lending activities become more diversified. As CNCA's program has grown in strength, it has begun to offer training to staff from other institutions in Morocco and trainees from abroad, particularly from the Middle East. The Bank's Economic Development Institute (EDI) has noted CNCA's potential as a training center for credit officers from Francophone countries in Africa, and CNCA has expressed an interest in providing such training, although additional funding would be required for these activities. 3.8 Management Information System: The PCR describes the difficulties encountered in designing and putting into place a new MIS and an appropriate electronic data processing (EDP) system. These included considerable internal debate within CNCA on such issues as decentralization, the launching of new banking services, and the appropriate hardware and software for the EDP system. In 1985, irreconcilable differences led to the termination of a contract with foreign consultants working on a decentralized MIS. In 1988, the findings of a second team of consultants were rejected as incomplete. Thereafter, CNCA began to write its own programs for a limited number of banking applications for use on personal computers (PCs). Twenty CRCAs (i.e. about 40%) have been equipped with PCs on a test basis, and demand by other CRCAs for computerization appears to be growing. In the meantime, CNCA is gradually moving towards a consensus on the information structure of the MIS to be established and is designing its own computer programs to run the EDP system. The Bank has objected to this approach, favoring the purchase of existing software from a reputable vendor to speed up modernization. CNCA affirms that it has carefully reviewed the software applications used in other similar institutions but has not found any programs that meets its needs. The Audit believes that CNCA has presented some fair arguments for proceeding independently, but shares the Bank's concern that it could ultimately result in higher costs and a longer development period. C. CNCA's Financial Performance 3.9 Lending Policies and the Pattern of Lending: Under the fifth and sixth projects, CNCA was to increase lending to small farmers, as well as to introduce new lending criteria that would enable CNCA to make larger loans. While the second goal has been achieved, the objective of extending more credit to smallholders met with only partial success. 3.10 The number of sub-loans under both projects greatly exceeded appraisal estimates, as shown in Table 6 below. Furthermore, the number of CRCA clients in the active portfolio rose by 13% from 60,000 to 67,000 during implementation of the two projects and the number of CLCA clients - 10 - rose by 22% from 364,000 to 444,500 over the same six-year period. CNCA currently reaches about 70% of potential CRCA clients and 33% of potential CLCA clients. The number of active loans (dossiers) increased by 17% for CRCAs and by 38% for CLCAs over the period, indicating a rise in the number of loans per client, and the average indebtedness of CRCA and CLCA clients rose by 77% and 62%, respectively. While total outstanding balances doubled for both CRCAs and CLCAs between 1985 and 1990 (Annex 2), new disbursements have not kept pace with repayments of interest and principal, particularly at the CLCA level, and net annual disbursements to the small-farmer clientele served by CLCAs have become negative, as shown in Table 7 below and in Annex 2. Whereas CLCAs used to account for one-third of total CNCA disbursements in 1985, they accounted for only one-fifth in 1990. Table 6: ESTIMATED AND ACTUAL NUMBER OF SUB-LOANS EXTENDED BY CNCA -- Fifth Project-- -- Sixth Project-- Appraisal Actual Appraisal Actual Loans through HQ/CRCAs: 70,000 210,000 63,000 145,000 Loans through CLCAs: 510,000 755,000 200,000 510,000 Table 7: NET DISBURSEMENTS TO CNCA BORROWERS, 1985-1990 (DH Millions) 1985 1986 1987 1988 1989 1990 Headquarters & CRCAs: (1) Disbursements 1800.5 2108.3 2284.4 2573.7 2627.3 3187.9 (2) Repayments 1201.0 1742.4 2033.2 2410.7 2521.3 3046.1 (3) Net Disbursements 599.5 365.9 251.2 163.0 106.0 141.8 (4) Ratio of (1)/(2) 150% 121% 112% 107% 104% 105% CLCAs: (1) Disbursements 837.5 979.7 945.1 743.6 691.4 817.9 (2) Repayments 525.5 696.5 791.5 957.0 1045.3 822.2 (3) Net Disbursements 312.0 283.2 153.6 -213.5 -353.9 -4.9 (4) Ratio of (1)/(2) 159% 141% 119% 78% 66% 99% Ratio of CLCA Disbursements to Total Disbursements: 32% 32% 29% 22% 21% 20% 3.11 Lending policies were generally considered adequate at appraisal of the fifth project, although agreements were reached to revise them by extending the small farmer credit policies developed in the Bank-supported Fes-Karia-Tissa Agricultural Development Project (Loan 1602- MOR) to other regions of Morocco. The revised policies and procedures are explained in detail in special manuals published for CRCA and CLCA staff. - 11 - 3.12 CNCA has traditionally established credit ceilings linked to the fiscal incomes of its clients, as determined by the fiscal authorities. Since these have tended to underestimate the earnings potential of farmers, this policy has constrained farmers' access to credit. Following the 1984 declaration of a tax holiday on agricultural incomes until the year 2000 A.D. by the King of Morocco, fiscal authorities stopped recalculating fiscal incomes for farmers. As a result of these factors, CNCA agreed to review the results of the Fes-Karia-Tissa experiments, as well as the results of a pilot lending scheme under the Oulmes Rommani Agricultural Development Project (Loan 2217-MOR), to determine whether the new credit policies, which are based on direct assessment of farmers' income, should be extended to other regions. The new policies apply only to landowners with proof of ownership. They increase credit ceilings to 50% of projected annual gross income or 70%-90% of the value of CNCA-financed investments, provided that total yearly repayments do not exceed 25% of annual farm receipts. Seasonal credit for inputs is not included in these figures. For those who are tenant farmers or lack proof of ownership, i.e. most CLCA clients, the old system of credit limits based on fiscal incomes continues to apply. The appraisal report for the sixth project states that application of the new procedures resulted in the extension of more medium- and long-term credit as well as improved recovery rates in the experimental areas. Under the sixth project, CNCA agreed to introduce the new system in the provinces of Meknbs (1987) and Settat (1988). The new methods have not been extended beyond these two provinces. Overall, average loan amounts have increased over time, but not to the levels anticipated at appraisal (Table 8). Table 8: AVERAGE SUB-LOAN AMOUNTS EXTENDED BY HQ/CRCAS AND CLCAS (in Dirham) -------------Appraisal-------------- -----------Actual------------ 5th Project 6th Project Change (%) 1984/85 1989/90 Change (2) HQ/CRCAs 22,080 40,200 82% 24,566 70,986 126% CLCAs 9,327 11,240 21% 3,256 4,621 42% 3.13 Savings Mobilization: Both the fifth and the sixth projects placed emphasis on savings mobilization in order to reduce CNCA's dependence on external sources of funds. This objective has not been achieved. In spite of a 134% increase in the volume of deposits and a 39% increase in the number of depositors between August, 1985 and December, 1990, deposits as a share of total CNCA resources fell from 14.3% in 1985 to 11.9% in 1988 before climbing to 15.6% in 1990, somewhat below the sixth project's target of 20% (Table 9 below). Meanwhile, external liabilities grew more rapidly than domestic sources of funds, rising from 58% to 63% of total liabilities. The PCR states that CNCA's deposit mobilization has been constrained because it has not been allowed to mobilize deposits in urban areas, since these are the domain of commercial banks. This prohibition was not mentioned in earlier project documents, perhaps because it is the result of a strict interpretation by the Ministry of Finance of CNCA's role as an agricultural bank, rather than an explicit prohibition in the 1961 Dahir (law) that established CNCA. This strict interpretation (and the resulting ?J CNCA's figures are somewhat higher for 1988 (12.4%) and 1990 (16.4%), for reasons explained in para. 3.14. CNCA also pointed to an improvement in the ratio to 18.1% in 1991 (see Appendix). 2/ CNCA has pointed out that the figures for external funds are revalued annually at year-end exchange rates and thus include a potential exchange rate shortfall for which the Treasury would be liable, rather than CNCA (see Appendix). - 12 - prohibition) gave way to a more active banking role for CNCA in 1987, when the King of Morocco designated CNCA as a "model bank". Since then, CNCA has adopted a more aggressive approach to savings mobilization in urban areas, opening bank branches and collecting savings repatriated by Moroccans working abroad. On the other hand, savings mobilization in rural areas has met with less success. Deposits at the CLCA level accounted for less than 2% of all CNCA deposits in 1990. In spite of the introduction of super carnet vert passbook savings, the Audit mission found that CLCA clients continue to be reluctant to deposit their savings with CNCA, preferring to invest in real assets such as livestock. CLCA staff also pointed out that deposits with CLCAs were highly seasonal, making them a poor source of funds for agricultural lending. 3.14 CNCA's Financial Ratios: Most of the ratios presented in Table 9 were calculated by the Audit, based on a balance sheet presentation that differs slightly from CNCA's presentation in the Appendix.:y In spite of the differences in presentation, both the Audit's and CNCA's calculations reflect the fact that CNCA experienced difficulties in meeting some financial covenants (Table 9 below, Annex 1 Table 4, Appendix, PCR paras 3.18-3.20, 3.28-3.30 and PCR Annex 3). One covenant that CNCA did fulfil is the requirement that the ratio of current assets to current liabilities exceed 1.2. Although data in the PCR reveal a shortfall when current maturities on medium/long term instruments (i.e. maturities of less than 12 months) are excluded from the ratio, the Loan Agreements for both projects allow for the inclusion of such maturities. Thus CNCA calculations show that it met the current ratio requirement (Table 9). Furthermore, although the Loan Agreement revised the definition of current maturities from the standard 12 months to permit the inclusion of maturities of up to 18 months, CNCA has noted that it continued to base its calculations on assets and liabilities with maturities of 12 months or less. It should be noted that the current ratio calculation does not take into account the recovery rate on risk assets. Thus the fact that CNCA met the current ratio requirement does not mean that it did not experience liquidity problems. 3.15 During the fifth project, external borrowing was heavier than expected and the debt:equity ratio rose above the 6:1 ratio stipulated in the fifth project's loan covenant. The other striking shortfall with respect to financial covenants was in the return on equity. Beginning in 1988, CNCA's income became taxable at a rate of 40%. That same year, provisions for bad debts rose from 2% to 2.8% of total assets (Table 9). The Audit mission was informed that this was a long-overdue attempt to improve CNCAs bad debt reserves in the light of mounting arrears, rather than a direct response to the new tax, particularly since CNCA is in any case a Government-owned agency. The PCR argues that the lower (pre-tax) return on equity resulting from the higher bad debt provisions (and higher foreign exchange risk provisions - para 3.28) belie the sounder financial structure that CNCA has developed. The Audit believes that the strength of the institution was previously overestimated, as the cost of establishing adequate reserves for bad debt was simply postponed. It concurs with the PCR that accounting practices need to be reformed so that interest accrued on non-performing loans is not included in current-period earnings. Furthermore, with the return on equity at 3.9% and inflation at 6.7% in 1990, further increases in interest rates may be necessary in order to avoid erosion of the real value of CNCA's capital. CNCA has set itself the goal of achieving a positive real Y For example, the Audit's figures include the CLCA support fund and the KfW equipment subsidy under liabilities, whereas CNCA has included them under quasi-equity. More importantly, in 1988 CNCA changed its accounting procedures, presenting its risk assets net of provisions, whereas the Audit continued to present total loans on the asset side and provisions on the liabilities side of the balance sheet, in order to provide comparable figures for the entire period 1985-90. The reader is invited to refer to both sets of financial ratios. - 13 - return on equity in order to avoid decapitalization, although it believes that the return need not equal market rates. Table 9: CNCA'S FINANCIAL RATIOS 1985 1986 1987 1988* 1989 1990 Balance sheet ratios Current Assets / Current Liabilities** - Appraisal: 1.5 1.5 1.2 1.2 1.5 1.3 - Covenant: 1.2 1.2 1.2 1.2 1.2 1.2 - Actual: 1.4 1.5 1.7 1.6 1.7 1.8 Long-term Debt / Equity - Appraisal: 3.9 4.4 4.9 5.4 8.2 8.1 - Covenant: 6.0 6.0 6.0 8.0 8.0 8.0 - Actual: 5.1 6.1 6.5 6.3 6.5 7.2 Total Debt / Equity - Appraisal: 7.5 8.1 8.5 8.7 12.6 13.0 - Actual: 8.4 9.8 9.7 10.0 10.7 11.4 Deposits / (Total - Appraisal: 15.5% 15.5% 18.3% 20.3% 13.6% 16.1% Liabilities + Equity) - Target: - - - 20.0% 20.0% 20.0% - Actual: 14.2% 13.6% 12.8% 11.9% 14.1% 15.6% Cumul. Provisions / - Appraisal: 3.2% 2.6% 4.5% 5.0% 6.4% 6.5% Risk Assets - Actual: 4.9% 5.1% 6.6% 8.1% 10.2% 11.0% Provisions / Arrears - Actual: 25.1% 21.3% 24.4% 32.0% 35.0% 39.0% Income & Expenditure Ratios Operating Income / Risk Assets - Appraisal: 13.3% 12.7% 12.8% 13.2% 12.0% 12.3% - Actual: 12.9% 12.0% 12.5% 12.8% 12.1% 12.0% Interest Expense / Liabilities - Appraisal: 6.0% 6.3% 5.5% 5.8% 5.0% 5.1% - Actual: 4.7% 5.0% 5.3% 4.9% 5.4% 5.5% Operating Expenses / Risk Assets - Appraisal: 4.3% 4.1% 3.0% 3.2% 3.9% 4.1% - Actual: 3.1% 2.8% 2.7% 3.6% 3.4% 3.7% Annual Provisions / Risk Assets - Appraisal: 0.7% 0.5% 2.22 2.4% 1.4% 1.6% - Actual: 2.3% 1.5% 2.0% 2.8% 2.8% 2.2% Net Return before Tax / Risk Assets - Appraisal: 1.7% 1.5% 1.7% 1.5% 0.6% 1.0% - Actual: 1.9% 1.6% 1.4% 0.7% 0.8% 0.7% Net Return after Tax / Risk Assets - Appraisal: 1.7% 1.5% 1.7% 1.5% 0.6% 1.0% - Actual: 1.9% 1.6% 1.4% 0.3% 0.5% 0.4% Net Return / Equity - Appraisal: 11.3% 11.3% 12.7% 12.7% 5.4% 10.7% - Covenant: 12.0% 12.0% 12.0% 12.0% 12.0% 12.0% - Actual: 12.1% 11.5% 11.1% 2.6% 4.0% 3.9% * The CNCA fiscal year was changed in 1987 from Sep.1-Aug.31 to Jan.1-Dec.31 (the figures for 1985-1987 end on Aug.31, those for 1988-1990 on Dec.31). ** The Appraisal measure of the current ratio counts only short term assets and liabilities, whereas the Actual measure, provided by CNCA, includes the current portion of medium/long term assets. Note: Appraisal figures are derived from projected balance sheets in the appraisal reports for the Fifth Project (1985- 86), Sixth Project (1987-88) and National Project (1989-90). Breaches of loan covenants are underlined. - 14 - 3.16 Subsidization of CNCA: One of the objectives of the projects was to build up CNCA's profits and thereby eliminate the need for Government operating subsidies and share capital contributions. CNCA has indeed benefitted from a number of privileges and subventions that commercial banks in Morocco have not enjoyed. The most important of these is access to relatively inexpensive external loans for which the Government has acted as guarantor and has (until recently) assumed most of the foreign exchange risk (para 3.29). Other privileges include exemption from reserve requirements, exemption from income tax until 1988, and access to over 150 MARA offices staffed by MARA agents at no cost to CNCA. The offices, which are used as seasonal banking windows, have virtually doubled CNCA's branch network. Although the financial benefit of the free banking windows has not been quantified, a Subsidy Dependence Index (SDI) has been constructed which incorporates most of the other subsidies that CNCA has received (Annex 1). Used in conjunction with standard ratio analysis, the SDI offers a truer picture of the costs associated with CNCA's operations.!' The index calculates the subsidy that CNCA receives from access to funds at below-market rates of interest, adds the implicit subsidy resulting from exemption from reserve requirements and imputes a market rate of return on equity in order to reflect the opportunity cost of capital. CNCA's profits are subtracted from this total to obtain the net subsidy. The net subsidy is then measured against interest income in order to establish how much interest income would need to rise for CNCA to earn a market rate of return on equity without any subsidies.Y 3.17 Although CNCA continues to benefit from circa DH600 million in implicit subsidies per year, primarily through access to loans at about half the market cost of funds, the SDI has decreased significantly in the course of the two projects. In 1986, average on-lending interest rates would have had to increase by 69% (from 11.7% to 19.7%) in order to raise sufficient revenue to offset the subsidies. In 1990, an increase of 36% (from 11.8% to 16.1%) would have sufficed.2' Whether such increases are feasible or desirable in practice is another question. Nonetheless, measurement of the subsidies that CNCA receives should constitute an integral part of its financial analysis, with a view to gradually reducing CNCA's reliance on subsidies and making it increasingly self-sustainable. 3.18 CNCA pointed out to the Audit mission that the subsidies were accompanied by concomitant costs that commercial banks do not bear. These include costs arising from the need to keep open unprofitable CLCA branches or direct credit to a particular clientele, insufficient remuneration for CNCA's management of Agricultural Development Fund (FDA) operations, and delays in Government reimbursements for such expenses as rural housing subsidies or foreign exchange risksy No attempt has been made to quantify these costs or assess the extent to which they offset the subsidies that CNCA receives. 3.19 Arrears: Both the PCR and the supervision reports understate the gravity of the arrears situation. At the end of 1990, the volume in arrears stood at DH2,823 million, compared to DH866 -' For details on the SDI, see J. Yaron: "Successful Rural Financial Institutions", World Bank Discussion Paper No. 150, 1992. V Although dependence on subsidies can be reduced by other methods, e.g. cutting administrative costs, increasing interest income is used for convenience. Z CNCA believes that the SDI calculation should be tailored more specifically to the case of CNCA and therefore reserves its judgment on the figures presented by the Audit (see Appendix). tv The FDA channels MARA grants to farmers. - 15 - million at the end of fiscal 1985. Thus, while CNCA assets were increasing at an annual rate of 16.4%, the volume in arrears was rising at an annual rate of 27.6%. Furthermore, these figures apply after massive loan rescheduling: CNCA rescheduled 70% of all arrears in 1985, following several years of drought, and 22% of total dues in 1986.2 In spite of more favorable weather in the late 1980s, arrears have continued to grow rapidly. 3.20 The Audit's figures on recovery rates are considerably lower than the figures cited in the PCR for 1988-1990 (Tables 10 and 11 below, Annex 2 and PCR paras 3.21-3.22). The reason for this is that the PCR's series for 1988-1990 reflect collections on current dues only, rather than the overall collection rate on both current dues and overdues. As a result, the Audit does not concur with the PCR's finding that CNCA's performance compares favorably with that of agricultural credit banks in other countries. Furthermore, Audit data show a steady decline in the recovery rate from 67% in 1985 to 58% in 1990.L Table 10: CNCA LOAN RECOVERIES AND ARREARS 1985 1986 1987 1988 1989 1990 Collection Ratio CLCAs 61% 57% 52% 51% 51% 42% CRCAs 67% 66% 65% 62% 59% 53% Headquarters 78% 75% 70% 95% 82% 96% CNCA Average 6W 64T 62T M M 58T Figures cited in the PCR 65-70% <65% <65% 76% 75% 76% Arrears as Percent of Loan Portfolio CLCAs 24% 28% 32% 33% 38% 41% Headquarters + CRCAs 18% 22% 24% 21% 25% 23% CNCA Average 2 O NT M M M! Table 11: CRCA AND CLCA RECOVERY RATES BY REGION, 1990 ------ CRCAa ------- ------ CLCAs ------- Recovery % of CRCA Recovery % of CLCA Rate Loans Due Rate Loans Due Region North-West 63% 15% 45% 23% North Central 60% 8% 37% 15% Central 67% 31% 50% 26% South Central 68% 9% 49% 10% Southern 42% 12% 67% 2% Eastern 18% 12% 21% 6% Tensift 36% 13% 30% 18% Average for Morocco Yff T"o rn TW 97 Concessionaire assistance from the EEC and USAID was obtained to reschedule the 1985 arrears over 10 years and replenish CNCA's working capital. L9 In 1987, CNCA changed its fiscal year to a calendar year to harmonize it with general banking practice in Morocco. Although this was also expected to result in improved figures on recovery rates, since the collection ratio generally improves between August and December, Audit data suggest that no improvement occurred. - 16 - 3.21 The figures suggest that a large number of CNCAs borrowers may be over-indebted and therefore unable to repay on a timely basis. Indeed, as noted earlier (para 3.10), average outstanding loan amounts per CRCA and CLCA borrower rose by 77% and 62%, respectively, over the life of the two projects. While annual collections at the CLCA level were 56% higher in 1990 than in 1985, loans due were 128% higher. In spite of a 154% increase in annual repayments at the Headquarters and CRCA levels, loans due increased more rapidly at this level too. Currently, annual recoveries are almost sufficient to service current dues but are well below what is required to repay both current dues and overdues. 3.22 Over the past five years, the age of CNCAs arrears has risen rapidly. Between the appraisal of the fifth and sixth projects, the share of arrears that was more that 15 months overdue rose from 19% to 27% (and in a 1991 Bank memorandum on CNCA's policy on provisions, it was reported that less than 30% of arrears were more than 18 months overdue). The aging of arrears is partly attributable to CNCA's accounting practices, as CNCAs figures on overdues include arrears that have accumulated since CNCA's inception and have not been written off, even though a portion of them are uncollectible. In the first debt cancellations since operations began in 1961, CNCA obtained permission to write off DH1.4 million in 1990 and DH15 million in 1991. One questionable accounting practice is the accrual of interest on principal that is up to three years overdue. Nevertheless provisions are made against this interest, as well as 100% provisions against all principal that is more than 36 months overdue. 3.23 In 1990, total arrears equalled 28% of total risk assets. Noting that about 30% of CNCA's arrears are more than 18 months overdue, and assuming that half of these are never recovered, (an assumption which is consistent with CNCA provisions of 50% to 100% against arrears of 18 to over 36 months and with the fact that a portion of the arrears has been on the books since the 1960s), loan losses would amount to 4.2% of CNCA's risk assets. This estimate does not include potential defaults on loans that are less than 18 months in arrears. 3.24 The breakdown of arrears by regions in Table 11 suggests that recovery rates are higher in regions with more irrigated land (the center of the Kingdom). The arid Eastern region has the lowest recovery rates.L' The current MIS does not permit a breakdown of arrears by lending activity. Furthermore, since it is organized on a loan basis rather than a client basis, it is not possible to distinguish between borrowers who have not received loans recently because they have overdue loans, and borrowers who have received new loans following the rescheduling of their earlier overdues. It would be useful to establish which group is performing better. On the one hand, the policy of not granting new loans to those with overdues may be hampering investment on their farms and therefore repayment of loans. On the other hand, those who have received fresh loans may not be earning sufficient amounts to pay off both the new loans and the rescheduled loans. CNCA has commented that, generally speaking, the results of the 1985 rescheduling have been satisfactory in regions that have subsequently enjoyed successful crop years (see Appendix). Nevertheless, this issue deserves further attention from CNCA's M&E unit. There can be little doubt, however, that the farmers who were hardest hit by the droughts have gained from rescheduling and continued access to loans. ' Although CNCA has commented that three regions (the East, Haouz and the South) were afflicted by adverse climatic conditions throughout the 1980s and account for 70% of all arrears (see Appendix), more recent data suggest that the figure is closer to 45% and falling. - 17 - 3.25 A further issue that the Monitoring and Evaluation Unit should have addressed is a high level of arrears among borrowers who are potentially CNCA's best clients. The Audit mission visited Safi, which lies along the central coastline of the Kingdom, and was surprised to learn that greenhouse farmers, in spite of production under an irrigated, controlled environment, accounted for a large portion of the arrears. They mainly produce tomatoes for the European market. The Audit mission was informed that, as a result of stricter regulations in the European Community and increased competition resulting from the liberalization of Morocco's export sector, farmers have experienced marketing, rather than production, difficulties. Those that have not met exacting timing and quality standards have been particularly affected. Since this problem has affected a number of farmers for three or more consecutive years, they have been unable to reduce their debt burdens. The fact that clients employing the most advanced production techniques are experiencing difficulties in repaying their loans should be a cause for concern at CNCA. 3.26 The Audit mission was impressed by the serious efforts that CNCA has made to address the problem of arrears. Notifications of amounts due are sent to all loaners well before the due date and CNCA staff are mobilized to meet the farmers at the markets (souk) every week in June through August or, in some cases, to collect loans at the farms. Strong measures are taken against delinquent borrowers within two months of the due date, unless the latter present a good case for rescheduling of their loans. Local authorities co-operate with CNCA in recovering loans from delinquent borrowers. CNCA's Auditors believe that provisions for bad debts, which rose from 25% to 39% of total arrears during the two projects, were more than adequate (for the first time) in 1990.2 In the light of the drought experienced during the 1992 crop year, CNCA has taken further measures to ease farmers' debt burdens, including the cancellation of late payment charges for clients in arrears who are making current payments on time, as well as rescheduling loans for clients who pay off a portion of their overdues. D. Impact on Sectoral Issues 3.27 Interest Rates: Both the fifth and the sixth projects required CNCA to review its interest rates annually to ensure that they remained positive in real terms. CNCA increased its nominal rates by an average of 1.9 percentage points before Board presentation of the fifth project. Since then, its average on-lending rate has been positive in real terms, largely as a result of low rates of inflation. Controls on most of CNCA's on-lending rates were lifted in 1990 and 1991 as part of a financial liberalization program supported by the Bank. CNCA's rates have, however, tended to be slightly below market rates (see Table 12 below). For some activities, the rate to sub-borrowers has been lower than the rate received by CNCA, due to Government subsidies. In particular, loans for rural housing bear a nominal rate of 12-13%, but include a subsidy of 4-5%. This subsidy used to be 6-7%, but was reduced at the Bank's insistence. A 2% subsidy on CNCA loans for fisheries is allowing CNCA to undercut competition from the commercial banking sector. This transfer of clients from commercial banks to CNCA does not promote additional investment or contribute to the development of financial markets. EY According to CNCA's data (see Appendix), provisions declined to 23% instead of rising to 39%. The Audit's figures are higher as they include a reserve for credit risk constituted in 1988 (which appears as quasi-equity in CNCA's presentation). - 18 - Table 12: CNCA INTEREST RATES COMPARED WITH INFLATION AND MARKET RATES 1985 1986 1987 1988 1989 1990 CNCA Interest Rate Earned On Average Outstanding Loans Short-term: CLCAa 10.1% 10.8% 10.8% 12.7% 9.7% 10.3% CRCAa 10.3% 10.3% 11.0% 10.4% 10.2% 10.0% Long-term: CLCAs 10.3% 10.5% 10.5% 11.1% 10.6% 10.8% CRCAs 11.0% 10.6% 11.3% 11.8% 11.6% 11.1% Inflation Rate and Average Market Rates Inflation 7.7% 8.8% 2.8% 2.3% 3.1% 6.7% Market Reference Rate 11.5% 11.5% 11.5% 11.5% 11.5% 10.5% Comercial Bank Paper 10.3% 10.5% 10.5% 10.6% 10.6% 10.6% Comercial Bank Medium-Term Rates 14.5% 14.5% 14.5% 12.8% 12.8% 12.8% Connercial Bank Long-Term Rates 16.0% 16.0% 16.0% 14.0% 14.0% 14.0% Bank Al-Maghrib Rediscount Rate* 8.5% 8.5% 8.5% 8.0% 8.0% 7.0% Treasury Bills - One Year 12.0% 12.0% 12.0% 11.0% 11.0% 11.0% BNDE Medium-Term Loans 15.0% 15.5% 14.0% 12.0% 12.0% 12.0% CIH Housing Loans 15.0% 15.0% 13.5% 12.0% 12.0% 12.0% Note: Bank Al-Maghrib is the Central Bank of Morocco 3.28 CNCA's "Solidarity" Bonds: Following the drought of 1980/81, commercial banks were required to invest 3.5% of their deposits in one-year CNCA bonds bearing an interest rate of 3%. The aim was to channel more domestic resources into agriculture. During negotiations on the sixth project, CNCA persuaded the Bank to change its proposals for phasing out this privilege over seven years to requesting a study on ways to phase out CNCA's reliance on the bonds. Since CNCA experienced difficulties mobilizing alternative domestic resources, it did not comply with the covenant eliciting the study by June, 1987. It did comply with the covenant requiring that DH 1 out of every DH 100 raised through "Solidarity" bonds be set aside in a new reserve for natural calamities, although the covenant was dropped in an amendment to the loan agreement in March, 1990. The Bank has not pursued the issue under the follow-on project, but it appears that CNCA's access to these resources is being contained. Commercial banks are now obliged to invest 2%, rather than 3.5%, of their deposits in these bonds and the interest rate on the bonds has risen to 4.25%. Furthermore, the requirement is not strictly enforced. The bonds accounted for 8% of CNCA's total resources in 1990. 3.29 Foreign Exchange Risk: The question of foreign exchange risk was addressed for the first time under the sixth project. Under the fifth and earlier projects, the Government assumed the foreign exchange risk, which constituted a significant subsidy to CNCA and its borrowers.2 For example, during implementation of the fourth project, the value of the Dirham declined by approximately 50%, (from DH4 to DH7.5 per US Dollar), entailing substantial liabilities for the Government. It was decided under the sixth project that CNCA would bear the risk of any exchange rate variation up to 2%, and would maintain a foreign risk fund financed by the surplus earned on LY Although a foreign exchange risk mechanism was set up as early as 1973 at the World Bank's request and CNCA argues that the Government assumed only a portion of the exchange risk (see Appendix), the SAR for the fifth project (para. 6.13) states: The Government would guarantee the debt service of the Loan and bear the foreign exchange risk". Similar statements appear in earlier SARs. - 19 - subloans refinanced with concessionaire borrowing as well as a 1% commission on all new medium- or long-term loans except those to small farmers, and 1% of the outstanding principal on CNCA's 3% bonds. This formula was established under the second Industrial Trade and Policy Adjustment Loan (2604-MOR). By 1989, CNCA's foreign exchange risk provision had risen from zero to DH108 million, or 1.7% of long-term external borrowing. The exchange risk provision has been strengthened under the follow-up project to pass more of the exchange risk to CNCA and its borrowers. The foreign exchange risk issue is also being addressed under the recently approved Financial Sector Development Project (Loan 3365-MOR). This project goes even further in that it aims to ensure that financial intermediaries are indifferent between foreign and local resources by requiring contributions to a foreign exchange risk fund whenever the interest rate on foreign borrowing is below the domestic, market-determined cost of funds. The Audit is concerned that, in the light of CNCA's heavy reliance on external sources of funds and of CNCA's relatively small profit margins, these provisions may entail significant increases in lending rates to a debt-laden clientele. For example, with market rates of 10.5% in 1990, CNCA would have had to lend at over 18% (12% in real terms) to cover its costs and ensure a market return on equity.' IV. FINDINGS AND ISSUES A. Overall Assessment of the Projects 4.1 General Performance: Project implementation went quite smoothly for both projects. Bank funds were fully disbursed ahead of schedule in both cases and CNCA disbursements to target groups were generally in line with appraisal estimates, with the notable exception of loans to agrarian reform co-operatives under the fifth project. With a 22% increase in the number of CLCA clients, it is likely that the goal of increasing penetration in rainfed areas was achieved, but the Audit cannot confirm this. Due to the lack of ex-post calculations of economic and financial rates of return, the Audit is also unable to confirm that appraisal rates of return were obtained. However, the technologies that were financed appear to be well adapted to farmers' needs and climatic conditions under both projects have generally been favorable, which may explain a reported increase in collection rates for 1991. Overall, both projects are assessed as satisfactory. 4.2 Diversification of CNCA Lending: Although Ministry of Finance regulations still constrain the degree to which CNCA can lend for non-agricultural purposes, the diversification of CNCA's lending from agricultural credit to rural credit has progressed satisfactorily. The projects contributed by supporting lending for agro-industries and (in the sixth project) rural housing. CNCA now finances fisheries, forestry development, veterinarians, artisans and other rural clients and has launched campaigns targeted at women in particular. These campaigns have been very successful, with repayment rates near 100% as well as significant deposit mobilization. 4.3 CNCA's Financial Performance: Although CNCA's return on equity has declined, the decline in the SDI indicates that CNCA has developed into a stronger financial institution. CNCA's provisions for bad debt have been built up to levels that its Auditors consider satisfactory, and it has lent at interest rates that are positive in real terms and only slightly below market rates. CNCA .4 See CNCA's comments on the new exchange risk coverage system (Appendix). - 20 - intends to maintain a positive real average on-lending rate, which should be easier in the light of the flexibility accorded to CNCA under the recent financial liberalization. As a result of foreign exchange risk provisions, CNCA's overall cost of funds is also somewhat nearer market rates. Arguably, the risks involved in lending to the agricultural sector are greater than for many activities served by the commercial banking sector, so the Bank, the Government and CNCA should take measures to ensure that an appropriately higher spread for CNCA is maintained. This suggests rates to agriculture slightly above market levels. CNCA believes that a more important measure for strengthening its financial structure would be the establishment of a guarantee fund to insure against natural disasters (see Appendix). 4.4 Construction: On the institutional side, there have been some positive developments. In spite of some difficulties with construction, the network has grown at a steady pace. The Bank should continue to provide technical and financial support for the upgrading of CNCA facilities, particularly since slow progress on deposit mobilization has been attributed in part to poor facilities. The Bank's promotion of a CNCA task force to give the construction program increased priority is a valuable contribution. Another development that the Bank should support is more rapid computerization of CRCA branches. Lessons may be drawn in this respect from the rapid and successful computerization of the branch network of the National Agricultural Bank of Tunisia. 4.5 Training: CNCA's training program has been particularly successful. The number of staff trained has grown steadily and participants say they have benefitted from the courses. CNCA is currently renovating its training facilities to improve its service, particularly in computer training. EDI has pointed to CNCA as a possible regional training center and Bank staff could promote such a development by providing co-ordination and support for training of credit officials, particularly from Francophone African countries, at CNCA. 4.6 New Loan Appraisal Procedures: The Audit believes that CNCA loans, even small ones, are granted on the basis of sound technical criteria. An increasing share of CNCA loans have been approved on the basis of new loan appraisal techniques. These have reduced CNCA's reliance on loan criteria tied to fiscal incomes and have allowed for somewhat larger loans to be extended, although the fiscal income criteria continue to apply to tenants and those without proof of land ownership. There is no evidence as yet to suggest that the new techniques are tied to lower collection rates, although CNCA's monitoring and evaluation team should continue to verify this and ensure that loans granted under the new criteria do not exceed the debt-carrying capacity of the clients. 4.7 Allocation of Resources by CNCA Clients: Although CNCA's interest rates have been slightly below market rates, there is no evidence to suggest that misallocation of resources by CNCA clients has been a serious problem. While there is a risk of some diversion occurring because regular clients obtain their short term credits for seasonal inputs in cash, documentation is required for disbursements for items financed with longer-term loans, such as new equipment. The risk of diversion is further mitigated for crops, such as sugar, that are processed and marketed by state- owned enterprises, because farmers' sales are more easily monitored. The likelihood of substitution is greater than that of diversion. In particular, medium-term loans for livestock may have been used to increase farmers' working capital, since the animals purchased with the credit can be sold quite quickly at a higher price and the proceeds can be reinvested before the loan is due. Although the problem of crowding out of smaller farmers by wealthier clients has beset agricultural credit schemes in many other countries, it has not been an issue in Morocco, because of the way that CNCA is - 21 - structured. CNCA has solved the problem by directing CLCA loans exclusively to small farmers. Nevertheless, the share of total annual CNCA disbursements made through CLCAs fell from 32% in 1985 to 20% in 1990. This shift in emphasis may be explained by the greater importance of new, non-agricultural lending activities, which are concentrated at the CRCA level (and, for the largest loans, headquarters). But the decline in the share of CLCA disbursements has also been accompanied by a decline in the real value of annual disbursements to CLCA clients. This could reflect the higher rate of arrears on CLCA loans, which makes a growing number of the CLCA clients ineligible. The issues merit further investigation by the Bank. B. Outstanding Issues 4.8 Management Information System: The importance of developing a functional accounting and management information system cannot be overstated. This point is illustrated by discrepancies in the figures on arrears, which suggest that consistency in the presentation of data needs to be enhanced. For example, the seventh appraisal report provides figures on arrears for 1982/3 to 1984/5 which are lower than those reported in the sixth project's appraisal report. CNCA's current goal of developing a client-oriented MIS rather than a loan-oriented MIS should provide much more information on patterns of credit risk by type of activity, by region and so on. These should help to reduce CNCA's future exposure to such risk. The Audit is concerned that CNCA's decision to develop the improved MIS on its own may result in considerable delays in implementation. In the past, delays on MIS development have contributed to slower computerization of CNCA branches and have discouraged data analysis by the M&E unit. These two activities should be promoted actively even in the absence of an enhanced MIS. 4.9 Arrears: In spite of several years of favorable climatic conditions, and of impressive collection drives by CNCA, the recovery rate has been declining. Thus the problem of arrears is arguably the most pressing problem that CNCA faces. It calls for increased diversification of CNCA's portfolio, the continuation of vigorous collection drives, and the establishment of a properly financed and rigorously controlled natural calamity fund to offset a portion of the lending risk. It also calls for a more active role for the M&E unit. The M&E unit can contribute to CNCA's management of the arrears problem (and reduction of future loan risk) by using the existing MIS and client surveys to provide a detailed breakdown of arrears by lending activity and by other characteristics of the clientele. The concentration of arrears in certain regions has already been noted. If many clients in those regions are simply not viable borrowers, then providing agricultural loans is not the best way to help these clients - either the Government can provide more direct support to them or CNCA can emphasize non-farm credit in these regions. In addition, CNCA needs to monitor the outcome of its 1985 rescheduling of 70% of all arrears. Thus far CNCA believes that the operation has been successful in certain cases (see Appendix). However, more detailed information would be valuable on whether and under what circumstances the repayment performance of clients who benefitted from additional loans has been superior to that of borrowers who did not receive fresh loans after the droughts. Such an analysis may help to guide future rescheduling operations and preempt considerable write-offs down the road. 4.10 Deposit Mobilization: The Bank has attempted to support increased mobilization of domestic resources over a series of projects with little success. The third project provided funds for consultants to assist CNCA in developing a banking program with particular emphasis on deposit mobilization. The PAR and the PCR for the third project do not mention deposit mobilization. Savings mobilization was not addressed in the fourth project, but funds were provided in the fifth - 22 - project for consultants to develop another medium-term banking plan, and again in the sixth project to develop a savings mobilization strategy. The follow-on project has also included strategic planning and upgrading of facilities to promote savings mobilization. To date, the planning appears not to have produced significant results. In 1985, CNCA mobilized only 2.3% of all deposits in the Moroccan banking system. By 1989, this figure had barely risen, to 2.7%. Poor performance has been attributed to factors as wide-ranging as delays on the MIS, poor quality facilities, lack of authority to compete in urban areas (before 1987), too much competition in urban areas and clients' preferences. On the other hand, CNCA has a vast network of branches, which, unlike other banks', are open on Saturdays, and it is the only bank that pays interest on current accounts. CNCA needs to launch a serious deposit mobilization drive, supported by technical assistance from commercial banking experts. There are indications that a determined effort can succeed: deposits increased by more than 50% in real terms during 1988-1990, rising from 11.9% to 15.6% of CNCA resources.' However, the Bank, which has provided one-quarter of CNCA's liabilities, (and other external donors, who have provided an additional 36%), must ensure that access to external funds does not reduce CNCA's incentive to tackle the issue. This consideration should be weighed in the determination of loan amounts under future projects. 4.11 Foreign Exchange Risk: The Government's original intention, as stated in the appraisal report for the sixth project, was eventually to make CNCA's borrowers fully responsible for the foreign exchange risk. This objective, which was inaugurated under the Second Industrial and Trade Policy Adjustment Loan (2604-MOR) approved in July, 1985, is a long way from being achieved. Since CNCA's foreign exchange reserves cover only a small fraction of total external liabilities, and since any losses from larger variations in the exchange rate would have to be borne by the Government, very little risk has been transferred from the Treasury to CNCA. Furthermore, CNCA, which is itself a Government agency, is financing about 20% of the risk fund out of regular income. The rest comes from one-time fees on farmers. Thus far, only commercial farmers receiving medium- or long-term loans have been required to pay the fee; smaller farmers have been exempted. The new foreign exchange risk policy introduced under the Financial Sector Development Project (Loan 3365- MOR) is likely to be a catalyst for increased mobilization of domestic resources, but it is not clear that a sudden change to the new policy is warranted. Aside from CNCA's arguments that the domestic market rates may be distorted and that CNCA has no direct access to the foreign currency that it borrows, the Audit is concerned that immediate implementation of the new policy may have deleterious effects on loan recovery if interest rates rise sharply, particularly under the present adverse climatic conditions. The Bank furthermore needs to address the criticism that its policy on foreign exchange risk has not been applied consistently across countries, and that even within Morocco, the new policy differs from CNCA's contractual obligations under the National Agricultural Credit project (Loan 3088-MOR). Nevertheless, the Audit does believe that CNCA clients should share in the foreign exchange risk on their loans. 4.12 Legal Constraints on CNCA Operations: The limitations that the 1961 Dahir, (the law that created CNCA), placed on CNCA's banking activities were a major issue during preparation of the sixth project. For example, the Issues Paper noted that "[t]his legal structure limits CNCA operational autonomy" and proposed a comit6 de r6flexion or management committee to study the possibility of incorporation of CNCA. The issue appears to have receded following the King's expression of support for an enhanced banking role for CNCA in 1987 and the dynamism injected LY According to CNCA's figures, the increase was from 14.3% in 1988 to 16.4% in 1990 (see Appendix). - 23 - into CNCA operations by its current chief executive. The 1961 Dahir continues to govern CNCA activities, albeit with a broader understanding of CNCA's role. However, a banking charter for CNCA or suitable amendments to the Dahir may prove necessary if CNCA's deposit mobilization an other activities continue to be restricted in ways suggested by the PCR (para 3.13, footnote 5). 4.13 Agricultural Subsidies: One possible impediment to CNCA's development is MARA's intention to introduce subsidized loans in addition to grants as part of the FDA program. In Tunisia, the combination of grants and cheap loans has had a deleterious effect on credit operations and the Bank has vigorously opposed it. The Audit is concerned that such a development may prove equally harmful to repayment discipline and collection rates in Morocco. 4.14 Land Reform: Insecure land tenure and fragmentation of landholdings continue to be obstacles to agricultural development. Although the Bank intended to address these issues under the fifth project by supporting a cadastral survey, the component was dropped at the Government's request. The Audit mission found that even some of the smallest farms consisted of several parcels of land. Furthermore, the majority of CLCA clients do not have formal titles or adequate alternative legal claims to the land that they farm. Acquisition of titles continues to be a lengthy and costly process, and although CNCA has introduced loans to help farmers acquire land titles, demand has been low. Land tenure difficulties help to explain why disbursements for construction, irrigation and land development were on balance below appraisal expectations, whereas loans for livestock and farm equipment were on balance above appraisal expectations. An effort was made to tackle the issues of land tenure and fragmentation under the Second Agricultural Sector Adjustment Loan (2885- MOR), as well as under earlier area development projects. These efforts deserve further support under future loans affecting the agricultural sector. 4.15 The Future of Agricultural Credit in Morocco: Over the past two decades, investment and recurrent expenditures on agriculture have declined as a share of total Government expenditures. Producer subsidies were cut in the mid-1980s as part of a Bank-supported sectoral adjustment program, and while the Bank is currently supporting improvements in agricultural support services under an Agricultural Research and Extension Loan (3036-MOR), many of the benefits have yet to be realized. As a result, a well-functioning agricultural credit system has become even more important for the development of Moroccan agriculture. This raises two major issues, namely whether the farm-level investments financed by CNCA loans have proven to be remunerative, and what impact the Bank's projects have had on the development of sustainable rural financial markets that can provide long-term support to agriculture. Regarding the first issue, the Audit is dismayed that no attempt appears to have been made to evaluate the degree to which the cash flow projections that determined CNCA loan commitments have indeed materialized at the farm level and how the returns on farm investments have been used. Such information is essential for effective risk management and containment of the arrears problem. The MARA monitoring and evaluation unit, which was to conduct impact evaluations according to the Staff Appraisal Report of the fifth project, is not mentioned in the appraisal of the sixth project or indeed of the follow-on project. There is no evidence in project files that the Bank has pursued the issue. 4.16 The Bank has supported CNCA through seven agricultural credit projects. Currently the Bank is playing an important role in promoting the integration of CNCA into a competitive banking sector as an increasingly autonomous, self-sustaining, full-service bank with an emphasis on agriculture. Fulfillment of this objective should prove highly beneficial for the Moroccan economy. Nevertheless, the issue of broadening rural financial markets has not been adequately addressed. For - 24 - example, while other Moroccan banks have received Bank and IFC support, (most recently under the current Financial Sector Development Project), there is little indication that the World Bank Group is actively encouraging greater participation by financial institutions other than CNCA in rural finance. The Bank could, for example, encourage commercial banks to compete for CNCA's most creditworthy clients by ensuring that CNCA does not offer below-market rates to these clients, and by supporting technical assistance for commercial banks to increase their expertise in agricultural lending or encouraging their use of CNCA training. At the other end of the market, innovative, group-based or cooperative schemes independent of CNCA may be necessary to reach smaller clients with less collateral at low cost and draw them into the financial system. The rewards from targeting female rural artisans have already been demonstrated. 4.17 While CNCA has made considerable advances, it is not clear that any single institution can meet all the needs of the rural financial sector, and some of the policies pursued through CNCA may indeed have hampered the development of rural financial markets. For example, insufficient attention to deposit mobilization in the past has resulted in CNCA having roughly seven times more borrowers than depositors, although the inverse is more customary for most commercial banks. On the lending side, CNCA has offered loans below market rates even though agricultural ventures tend to be more risky than other investments. This has undoubtedly contributed to crowding out of other intermediaries, so that in 1987, CNCA's agricultural lending accounted for 79% of short-term and 83% of long-term institutional credit for Moroccan agriculture. Tentative plans for a follow-up to the current operation suggest that the Bank is aware of the need to broaden rural financial markets. The Audit concludes that, while continued support for CNCA should be assured in order to achieve the objectives for CNCA that are discussed above, further consideration of the role that alternative institutions can play in mobilizing savings and extending credit in Morocco's rural financial markets is warranted at this juncture. - CNCA has commented at length on this section of the PAR (see Appendix). It notes that, since 1991, it has already faced stiffer competition and that greater participation of commercial banks in rural financial markets should only be encouraged once certain prerequisites have been met, including revision of the 1961 Dahir, the implementation of administrative and financial reforms within CNCA and the establishment of a calamity fund. CNCA also believes that due recognition should be given to the public service nature of its operations. -25- Financial indicators: Tables 1-4 ANNEX 1 Page 1 of 4 CNCA BALANCE SHEET FY THROUGH AUGUST FY THROUGH DECEMBER ASSETS 1985 1986 1987 1988 1989 1990 30RO- LIQUID ASSETS CASH AND BANKS: 285.2 583.9 788.7 861.1 834.8 1,077.9 TREASURY BILLS: 500.0 600.0 350.0 180.0 1,000.0 500.0 785.2 1,183.9 1,138.7 1,041.1 1,834.8 1,577.9 SHORT TERM RISK ASSETS BILLS DISCOUNTED: 27.9 3.1 1.3 3.3 26.8 33.6 HQ & CRCAs: 814.6 1,169.2 1,431.5 2,083.8 2,349.6 2,872.4 CLCAs: 105.6 177.5 256.5 383.6 378.4 445.7 948.1 1,349.8 1,689.3 2,470.7 2,754.8 3,351.7 MEDIUM/LONG TERM RISK ASSETS HQ & CRCAs: 2,156.4 2,605.4 2,841.1 3,238.3 3,647.8 4,310.6 CLCAs: 1,320.5 1,713.3 1,999.6 2,366.6 2,315.6 2,371.2 3,476.9 4,318.7 4,840.7 5,604.9 5,963.4 6,681.8 FIXED INVESTMENTS FINANCIAL INVESTMENTS: 5.8 5.8 6.0 249.6 214.6 185.7 FIXED ASSETS: 42.1 53.8 61.8 77.9 95.2 153.6 OTHER ASSETS: 1,195.7 1,498.4 1,685.8 1,649.3 1,515.7 1,817.4 TOTAL ASSETS: 6,453.8 8,410.4 9,422.3 11,093.5 12,378.5 13,768.1 LIABILITIES SHORT-TERM LIABILITIES DEPOSITS: 921.6 1,143.3 1,201.8 1,316.7 1,751.2 2,152.5 ONE-YEAR BONDS: 900.1 1,130.3 1,103.5 1,307.0 1,447.1 1,077.6 OTHER: 261.1 362.1 309.9 728.6 808.2 924.5 2,082.8 2,635.7 2,615.2 3,352.3 4,006.5 4,154.6 LONG-TERM BORROWING IBRD: 1,386.5 2,133.7 2,426.0 2,689.8 2,626.7 3,082.4 OTHER EXTERNAL: 1,852.0 2,340.9 2,860.0 3,190.8 3,611.8 4,153.8 INTERNAL: 155.8 111.1 91.3 76.7 64.0 53.3 3,394.3 4,585.7 5,377.3 5,957.3 6,302.5 7,289.5 SPECIAL FUNDS (CLCA): 94.0 94.0 94.0 94.0 94.0 94.0 KFW EQUIPMENT SUBSIDY: 0.0 0.0 0.0 2.0 3.3 4.6 TOTAL LIABILITIES: 5,571.1 7,315.4 8,086.5 9,405.6 10,406.3 11,542.7 PROVISIONS EXCHANGE RISK PROVISION: 0.0 57.3 73.4 97.0 107.6 109.2 OTHER PROVISIONS: 2.8 5.6 19.4 39.9 42.7 45.5 PROVISIONS FOR CREDIT RISKS Short-term Loans: 68.5 98.9 143.4 219.4 300.9 348.0 Medium/Long-term Loans: 146.4 186.7 265.8 393.1 543.9 706.5 217.7 348.5 502.0 749.4 995.1 1,209.2 EQUITY FUNDS EQUITY: 425.6 425.6 425.6 425.6 425.6 425.6 GENERAL RESERVES: 167.9 239.5 320.8 489.8 513.0 551.6 CURRENT PERIOD EARNINGS: 71.6 81.4 87.5 23.1 38.6 39.1 TOTAL EQUITY FUNDS: 665.1 746.5 833.9 938.5 977.2 1,016.3 TOTAL LIABILITIES & CAPITAL: 6,453.9 8,410.4 9,422.4 11,093.5 12,378.6 13,768.2 - 26 - ANNEX 1 Page 2 of 4 CNCA*S INCOME STATEMENT FY THROUGH AUGUST FY THROUGH DECEMBER 1985 1986 1987 1988 1989 1990 INCOME =MRWz INTEREST INCOME SHORT-TERM LOANS: 115.5 139.1 179.9 210.3 243.7 330.3 MEDIUM/LONG-TERM LOANS: 331.1 450.0 523.9 600.8 645.2 694.4 OTHER INTEREST INCOME: 33.6 63.4 90.6 140.9 184.5 173.4 480.2 652.5 794.4 952.0 1,073.4 1,198.1 OTHER INCOME: 18.1 21.7 25.2 34.8 44.2 45.6 TOTAL INCOME: 498.3 674.2 819.6 986.8 1,117.6 1,243.7 EXPENSES INTEREST EXPENSES: 217.4 319.3 404.8 426.2 538.3 606.8 ADMINISTRATIVE EXPENSES STAFF EXPENSES: 93.5 110.4 128.9 207.8 225.2 270.5 DEPRECIATION: 5.4 6.9 9.0 9.0 10.4 14.1 OTHER EXPENSES: 21.1 25.7 28.5 42.7 48.5 61.3 120.0 143.0 166.4 259.5 284.1 345.9 TOTAL EXPENSES: 337.4 462.3 571.2 685.7 822.4 952.7 =zc======= ======= = ===== ====s== z=====mZ NET OPERATING INCOME: 160.9 211.9 248.4 301.1 295.2 291.0 PROVISIONS FOR BAD DEBT: 90.0 73.5 123.5 203.2 232.4 209.7 PROVISIONS FOR EXCHANGE RISK: 0.0 57.2 17.9 13.9 14.6 15.9 SPECIAL PROVISIONS: 0.0 0.0 24.8 38.9 2.8 2.8 INCOME NET OF PROVISIONS: 70.9 81.2 82.2 45.1 45.4 62.6 PRIOR PERIOD ADJUSTMENTS: (1.9) (0.3) (2.9) (0.3) 21.8 5.8 EXCEPTIONAL/OTHER INCOME: 2.7 0.5 8.2 4.1 (1.7) (0.6) NET INCOME BEFORE TAXES: 71.7 81.4 87.5 48.9 65.5 67.8 INCOME TAX: 0.0 0.0 0.0 25.7 26.9 28.8 NET INCOME: 71.7 81.4 87.5 23.2 38.6 39.0 DISTRIBUTED AS DIVIDENDS: 0.0 0.0 0.0 0.0 0.0 0.0 FREE RESERVES: 71.7 81.4 87.5 23.2 38.6 39.0 - 27 - ANNEX 1 Page 3 of 4 CALCULATION OF THE SUBSIDY DEPENDENCE INDEX2' FY THROUGH AUGUST FY THROUGH DECEMBER 1985 1986 1987 1988 1989 1990 A: MARKET RATE OF INTEREST: - 11.5% 11.5% 11.5% 11.5% 10.5% B: AVG. COST OF CONCESSIONAL FUNDS: - 5.3% 5.7% 5.2% 6.0% 6.1% C: AVERAGE CONCESSIONAL LIABILITIES: - 4,965.8 6,091.2 6,883.6 7,533.2 8,097.7 D: SUBSIDY ON LIABILITIES: - 307.2 353.9 431.5 416.2 356.5 E: AVERAGE EQUITY: - 705.8 790.2 886.2 957.9 996.8 F: SUBSIDY ON EQUITY: - 81.2 90.9 101.9 110.2 104.7 G: CASH RESERVE RATIO: - 2.2% 4.2% 5.5% 8.8% 12.5% H: RATE ON RESERVES WITH BOJ: - 0.0% 0.0% 0.0% 0.0% 0.0% I: SUBSIDY ON RESERVE EXEMPTION: - 14.7 34.7 52.0 94.2 136.4 J: TOTAL SUBSIDY: - 403.0 479.5 585.4 620.6 597.5 K: ADJUSTED PROFIT: - (49.7) (71.0) (11.8) 52.4 166.8 = NET INCOME AFTER TAXES 81.4 87.5 23.2 38.6 39.0 + EXCHANGE ADJUSTMENT 57.2 28.9 27.0 14.6 15.9 + LOAN LOSS ADJUSTMENT (188.3) (187.4) (61.9) (0.8) 112.0 L: INTEREST INCOME: - 652.5 794.4 952.0 1,073.4 1,198.1 M: SUBSIDY INDEX: - 69.4% 69.3% 62.7% 52.9% 35.9% N: CURRENT AVERAGE ON-LENDING RATE: - 11.7% 12.1% 12.6% 11.9% 11.8% 0: INCREASE IN ON-LENDING RATE REQUIRED TO ELIMINATE SUBSIDY DEPENDENCE: - 8.1% 8.4% 7.9% 6.3% 4.3% P: LOWEST SUSTAINABLE SUBSIDY-FREE RATE: - 19.7% 20.5% 20.5% 18.3% 16.1% NOTES ON THE SUBSIDY INDEX: A: THE MARKET RATE IS ESTIMATED BY ADDING A 1% MARGIN TO THE SIX-MONTH T-BILL RATE. D: THE SUBSIDY ON LIABILITIES COMPUTES THE BENEFIT OF ACCESS TO FUNDS AT RATES BELOW MARKET RATES (ONE-YEAR BONDS, EXTERNAL FUNDS, CLCA SPECIAL FUNDS & KFW EQUIPMENT SUBSIDY). IT IS CALCULATED AS THE DIFFERENCE BETWEEN THE MARKET RATE AND THE AVERAGE INTEREST RATE PAID ON CONCESSIONAL FUNDS, TIMES THE AVERAGE ANNUAL VOLUME OF CONCESSIONAL FUNDS, I.E. AS [A - B] x C. F: THE SUBSIDY ON EQUITY ESTIMATES THE OPPORTUNITY COST OF CAPITAL. IT IS CALCULATED AS [A x E]. 1: THE SUBSIDY ON RESERVE EXEMPTION COMPUTES THE COST THAT CNCA WOULD HAVE INCURRED IF IT HAD NOT BEEN EXEMPTED AND WANTED TO DISPOSE OF THE SAME LEVEL OF ASSETS. THE SUBSIDY FROM THIS EXEMPTION IS CALCULATED AS [C + El x EG / (1 - G)] x [A - H3. J: THE TOTAL SUBSIDY IS CALCULATED AS [D + F + I]. K: CNCA'S AFTER-TAX PROFIT IS ADJUSTED UPWARDS BY THE AMOUNT OF CNCA'S FOREIGN EXCHANGE RISK CONTRIBUTION, (SINCE THIS RISK IS ALREADY CAPTURED IN THE MARKET RATE). A SECOND ADJUSTMENT IS MADE TO REFLECT THE FACT THAT PROFITS MAY BE OVERSTATED (UNDERSTATED) IF PROVISIONS ARE INADEQUATE (EXCESSIVE). THE DETERMINATION OF AN ADEQUATE LEVEL OF PROVISIONS IS BASED ON CNCA AUDITORS' 1990 REPORT. M: THE SUBSIDY INDEX MEASURES THE INCREASE IN INCOME REQUIRED TO OFFSET NET SUBSIDIES AFTER SUBTRACTING PROFITS. IT IS CALCULATED AS [(J - K) / LI. P: LOANS AT THIS INTEREST RATE, CETERIS PARIBUS, WOULD PERMIT CNCA TO EARN A MARKET RATE OF RETURN ON ITS EQUITY WITHOUT RELYING ON SUBSIDIES. IT IS CALCULATED BY AUGMENTING THE CURRENT ON-LENDING RATE BY A FACTOR EQUAL TO THE SUBSIDY INDEX, I.E. [N + (Mx N)]. For details see: J. Yaron. Successful Rural Finance Institutions. World Bank Discussion Paper No.150. 1992. -28- ANNEX 1 Page 4 of 4 KEY RNANCIAL INDICATORS A. CNCA'S PROFITABILITY 1985 1986 1987 1988 1989 1990 NOMINAL GROWTH RATE OF CNCA'S EQUITY: - 12.2% 11.7% 12.5% 4.1% 4.0% RATE OF INFLATION: 7.7% 8.8% 2.8% 2.3% 3.1% 6.7% REAL GROWTH RATE OF CNCA'S EQUITY: 3.2% 8.7% 10.0% 1.0% -2.5% PRE-TAX RETURN ON AVERAGE EQUITY: - 11.5% 11.1% 5.5% 6.8% 6.8% AFTER-TAX RETURN ON AVERAGE EQUITY: - 11.5% 11.1% 2.6% 4.0% 3.9% RETURN ON EQUITY AFTER ADJUSTING PROFITS: - -7.0% -9.0% -1.3% 5.5% 16.7% PRE-TAX RETURN ON RISK ASSETS: - 1.6% 1.4% 0.7% 0.8% 0.7% AFTER-TAX RETURN ON RISK ASSETS: - 1.6% 1.4% 0.3% 0.5% 0.4% PRE-TAX RETURN ON TOTAL ASSETS: - 1.1% 1.0% 0.5% 0.6% 0.5% AFTER-TAX RETURN ON TOTAL ASSETS: - 1.1% 1.0% 0.2% 0.3% 0.3% NET PROFIT MARGIN: 14.4% 12.1% 10.7% 2.4% 3.5% 3.1% B. BALANCE SHEET RATIOS CURRENT ASSETS / CURRENT LIABILITIES 0.83 0.96 1.08 1.05 1.15 1.19 LONG TERM DEBT / EQUITY 5.10 6.14 6.45 6.35 6.45 7.17 TOTAL DEBT / EQUITY 8.38 9.80 9.70 10.02 10.65 11.36 DEPOSITS / TOTAL RESOURCES 14.3% 13.6% 12.8% 11.9% 14.1% 15.6% DEPOSITS / RISK ASSETS 20.8% 20.2% 18.4% 16.3% 20.1% 21.5% DEPOSITS / LIABILITIES 16.5% 15.6% 14.9% 14.0% 16.8% 18.6% RISK ASSETS / LIABILITIES 79.4% 77.5% 80.8% 85.9% 83.8% 86.9% EXTERNAL FUNDS / LIABILITIES 58.1% 61.2% 65.4% 62.5% 59.9% 62.7% PROVISIONS / RISK ASSETS 4.9% 5.1% 6.6% 8.1% 10.2% 11.0% SHORT TERM - 7.3% 8.5% 8.9% 10.9% 10.4% MEDIUM/LONG TERM - 4.3% 5.5% 7.0% 9.1% 10.6% PROVISIONS / ARREARS 25.1% 17.1% 20.2% 27.2% 30.7% 35.1% C. OPERATING ACCOUNT RATIOS GROSS REVENUES / RISK ASSETS: - 12.0% 12.5% 12.8% 12.1% 12.0% INTEREST EXPENSE / AVG. LIABILITIES: - 6.4% 6.6% 6.2% 7.1% 7.5% FINANCIAL MARGIN: - 5.6% 5.8% 6.6% 5.0% 4.6% NON-INTEREST EXPENSES / RISK ASSETS: - 2.8% 2.7% 3.6% 3.4% 3.7% NET OPERATING INCOME / RISK ASSETS: - 4.2% 4.1% 4.1% 3.5% 3.1% ANNUAL PROVISIONS / RISK ASSETS: - 2.6% 2.7% 3.5% 3.0% 2.4% D. PROVISIONS AND ARREARS OUTSTANDING ARREARS AT END OF YEAR: 866.0 1,370.0 1,760.0 2,041.0 2,538.0 2,823.0 CUMULATIVE BAD DEBT PROVISIONS: 217.7 291.2 428.6 652.4 887.5 1,100.0 ADEQUATE PROVISIONS, BASED ON LEVEL APPROVED BY CNCA AUDITORS IN 1990: 303.1 479.5 616.0 714.4 888.3 988.1 SHORTFALL OR EXCESS: (85.4) (188.3) (187.4) (61.9) (0.8) 112.0 - 29 - CNCA Commitments, Repayments and Arrears: Tables 1-5 ANNEX 2 Page 1 Of 5 CNCA OUTSTANDING BALANCES FY THROUGH AUGUST FY THROUGH DECEMBER A. CNCA OUTSTANDING BALANCES 1985 1986 1987 1988 1989 1990 OUTSTANDING BALANCES HEADQUARTERS AND CRCAs: 2,971.0 3,774.6 4,367.7 4,882.4 5,215.9 5,908.6 CLCAs: 1,426.1 1,890.8 2,699.0 2,751.7 2,694.0 2,828.2 4,397.1 5,665.4 7,066.7 7,634.1 7,909.9 8,736.8 NUMBER OF LOANS GRANTED HEADQUARTERS AND CRCAs: 227,302 301,676 202,483 258,745 218,722 265,000 CLCAs: 826,053 854,337 905,818 1,084,469 975,252 1,139,439 1,053,355 1,156,013 1,108,301 1,343,214 1,193,974 1,404,439 NUMBER OF BORROWERS HEADQUARTERS AND CRCAs: 59,871 64,154 59,869 62,921 62,999 67,387 CLCAs: 363,999 401,073 420,712 "4,247 435,889 444,545 423,870 465,227 480,581 507,168 498,888 511,932 AVERAGE OUTSTANDING LOAN AMOUNT (000s) HEADQUARTERS AND CRCAs: 13.1 12.5 21.6 18.9 23.8 22.3 CLCAs: 1.7 2.2 3.0 2.5 2.8 2.5 4.2 4.9 6.4 5.7 6.6 6.2 BORROWERS' AVERAGE INDEBTEDNESS ('000s) HEADQUARTERS AND CRCAs: 49.6 58.8 73.0 77.6 82.8 87.7 CLCAs: 3.9 4.7 6.4 6.2 6.2 6.4 10.4 12.2 14.7 15.1 15.9 17.1 B. PERCENT VARIATION RELATIVE TO 1985 OUTSTANDING BALANCES HEADQUARTERS AND CRCAs: 27.0% 47.0% 64.3% 75.6% 98.9% CLCAs: 32.6% 89.3% 93.0% 88.9% 98.3% 28.8% 60.7% 73.6% 79.9% 98.7% NUMBER OF LOANS GRANTED HEADQUARTERS AND CRCAs: 32.7% -10.9% 13.8% -3.8% 16.6% CLCAs: 3.4% 9.7% 31.3% 18.1% 37.9% 9.7% 5.2% 27.5% 13.3% 33.3% NUMBER OF BORROWERS HEADQUARTERS AND CRCAs: 7.2% -0.0% 5.1% 5.2% 12.6% CLCAs: 10.2% 15.6% 22.0% 19.8% 22.1% 9.8% 13.4% 19.7% 17.7% 20.8% AVERAGE OUTSTANDING LOAN AMOUNT ('D00s) HEADQUARTERS AND CRCAs: -4.3% 65.0% 44.4% 82.4% 70.6% CLCAs: 28.2% 72.6% 47.0% 60.0% 43.8% 17.4% 52.7% 36.2% 58.7% 49.0% BORROWERS' AVERAGE INDEBTEDNESS ('OOOs) HEADQUARTERS AND CRCAs: 18.6% 47.0% 56.4% 66.8% 76.7% CLCAs: 20.3% 63.7% 58.1% 57.8% 62.4% 17.4% 41.7% 45.1% 52.8% 64.5% - 30 - ANNEX 2 Page 2 of 5 CNCA'S ANNUAL DISBURSEMENTS AND REPAYMENTS 1985 1986 1987 1988 1989 1990 TOTAL DISBURSMENTS HQ and CRCAs 1,800.5 2,108.3 2,284.4 2,573.7 2,627.3 3,187.9 CLCAs 837.5 979.7 945.1 743.6 691.4 817.9 2,638.0 3,088.0 3,229.5 3,317.3 3,318.7 4,005.8 DISBURSEMENT OF SHORT-TERM LOANS Hg and CRCAs 949.2 1,271.9 1,531.4 1,602.3 1,622.6 1,906.6 CLCAs 159.4 256.1 286.0 242.8 237.4 316.9 1,108.5 1,527.9 1,817.4 1,845.2 1,860.0 2,223.5 DISBURSEMENT OF MEDIUM/LONG-TERM LOANS HQ and CRCAs 821.3 798.8 719.7 855.7 943.6 1,257.1 CLCAs 678.1 723.6 659.1 500.7 454.0 525.2 1,499.5 1,522.5 1,378.8 1,356.4 1,397.6 1,782.3 LOAN REPAYMENTS HQ and CRCAs 1,201.0 1,742.4 2,033.2 2,410.7 2,521.3 3,046.1 CLCAs 525.5 696.5 791.5 957.0 1,045.3 822.2 1,726.5 2,438.9 2,824.7 3,367.7 3,566.6 3,868.3 NET DISBURSMENTS HQ and CRCAs 599.5 365.9 251.2 163.0 106.0 141.8 CLCAs 312.0 283.2 153.6 (213.5) (353.9) (4.3) 911.5 649.1 404.8 (50.4) (247.9) 137.5 RATIO OF DISBURSEMENTS TO REPAYMENTS HQ and CRCAs 1.5 1.2 1.1 1.1 1.0 1.0 CLCAs 1.6 1.4 1.2 0.8 0.7 1.0 1.5 1.3 1.1 1.0 0.9 1.0 - 31 - ANNEX 2 Page 3 of 5 CNCAS RECOVERY RATE FY THROUGH AUGUST FY THROUGH DECEMBER 1985 1986 1987 1988 1989 1990 LOAN PORTFOLIO HEADQUARTERS & CRCAs 2,971.0 3,774.6 4,272.6 5,322.1 5,997.4 7,183.0 CLCAs 1,426.1 1,890.8 2,256.1 2,750.2 2,694.0 2,816.9 4,397.1 5,665.4 6,528.7 8,072.3 8,691.4 9,999.9 BALANCES NOT YET DUE HEADQUARTERS & CRCAs 1,242.3 1,190.7 1,210.4 1,773.6 1,961.6 2,463.7 CLCAs 562.0 666.3 733.8 890.6 635.0 844.9 1,804.3 1,857.0 1,944.2 2,664.2 2,596.6 3,308.6 BALANCES DUE HEADQUARTERS & CRCAs 1,728.7 2,583.9 3,062.2 3,548.5 4,035.8 4,719.3 CLCAs 864.1 1,224.5 1,522.3 1,859.6 2,059.0 1,972.0 2,592.8 3,808.4 4,584.5 5,408.1 6,094.8 6,691.3 REPAYMENTS HEADQUARTERS & CRCAs 1,201.0 1,742.4 2,033.2 2,410.7 2,511.3 3,046.1 CLCAs 525.5 696.5 791.5 957.0 1,045.3 822.2 1,726.5 2,438.9 2,824.7 3,367.7 3,556.6 3,868.3 OVERDUE BALANCES HEADQUARTERS & CRCAs 527.7 841.5 1,029.0 1,137.8 1,524.5 1,673.2 CLCAs 338.6 528.0 730.8 902.6 1,013.7 1,149.8 866.3 1,369.5 1,759.8 2,040.4 2,538.2 2,823.0 RECOVERY RATES HEADQUARTERS & CRCAs 69.5% 67.4% 66.4% 67.9% 62.2% 64.5% CLCAs 60.8% 56.9% 52.0% 51.5% 50.8% 41.7% 66.6% 64.0% 61.6% 62.3% 58.4% 57.8% OVERDUES AS PERCENT OF LOAN PORTFOLIO HEADQUARTERS & CRCAs 17.8% 22.3% 24.1% 21.4% 25.4% 23.3% CLCAs 23.7% 27.9% 32.4% 32.8% 37.6% 40.8% 19.7% 24.2% 27.0% 25.3% 29.2% 28.2% - 32 - ANNEX 2 Page 4 of 5 RECOVERY RATES ON CURRENT DUES AND ON OVERDUES FY THROUGH AUGUST FY THROUGH DECEMBER 1985 1986 1987 1988 1989 1990 OVERDUES FROM EARLIER FISCAL YEARS HEADOUARTERS & CRCAS 368.3 542.5 853.1 1,001.2 1,567.0 1,567.6 CLCAs 214.2 429.4 548.7 811.0 1,024.4 1,004.2 582.5 971.9 1,401.8 1,812.2 2,591.4 2,571.8 REPAYMENTS ON OVERDUE BALANCES HEADQUARTERS & CRCAs 178.0 311.5 436.8 406.0 752.9 470.6 CLCAs 119.6 266.7 343.3 368.0 436.3 242.6 297.6 578.2 780.1 774.0 1,189.2 713.2 RECOVERY RATE ON OVERDUES HEADQUARTERS & CRCAs 48.3% 57.4% 51.2% 40.6% 48.0% 30.0% CLCAs 55.8% 62.1% 62.6% 45.4% 42.6% 24.2% 51.1% 59.5% 55.6% 42.7% 45.9% 27.7% LOANS MATURING IN CURRENT FISCAL YEAR HEADQUARTERS & CRCAs 1,360.4 2,041.4 2,209.1 2,547.3 2,468.8 3,151.7 CLCAs 649.9 795.1 973.6 1,048.6 1,034.6 967.8 2,010.3 2,836.5 3,182.7 3,595.9 3,503.4 4,119.5 REPAYMENTS ON CURRENT DUES HEADQUARTERS & CRCAs 1,023.0 1,430.9 1,596.4 2,004.7 1,758.4 2,575.5 CLCAs 405.9 429.8 448.2 589.0 609.0 579.6 1,428.9 1,860.7 2,044.6 2,593.7 2,367.4 3,155.1 RECOVERY RATE ON CURRENT DUES HEADQUARTERS & CRCAs 75.2% 70.1% 72.3% 78.7% 71.2% 81.7% CLCAs 62.5% 54.1% 46.0% 56.2% 58.9% 59.9% 71.1% 65.6% 64.2% 72.1% 67.6% 76.6% TOTAL REPAYMENTS AS % OF CURRENT DUES ONLY HEADQUARTERS & CRCAs 88.3% 85.4% 92.0% 94.6% 101.7% 96.6% CLCAs 80.9% 87.6% 81.3% 91.3% 101.0% 85.0% 85.9% 86.0% 88.8% 93.7% 101.5% 93.9% - 33 - ANNEX 2 Page 5 of 5 RECOVERY RATES BY REGION, 1990 -- BALANCES DUE -----RECOVERIES --- RECOVERY AMOUNT % OF TOTAL AMOUNT % OF TOTAL RATE NORTH-WEST REGION CRCAs 521.6 15.0% 331.0 17.9% 63.5% CLCAs 456.3 23.1% 204.4 24.9% 44.8% 977.9 18.0% 535.4 20.0% 54.7% NORTH CENTRAL REGION CRCAs 270.0 7.8% 162.5 8.8% 60.2% CLCAs 301.1 15.3% 111.0 13.5% 36.9% 571.1 10.5% 273.5 10.2% 47.9% CENTRAL REGION CRCAs 1,086.9 31.3% 730.5 39.5% 67.2% CLCAs 508.4 25.8% 254.1 30.9% 50.0% 1,595.3 29.3% 984.6 36.9% 61.7% SOUTH CENTRAL REGION CRCAs 295.6 8.5% 201.5 10.9% 68.2% CLCAs 198.3 10.1% 97.1 11.8% 49.0% 493.9 9.1% 298.6 11.2% 60.5% SOUTHERN REGION CRCAs 433.3 12.5% 181.7 9.8% 41.9% CLCAs 38.8 2.0% 25.9 3.1% 66.8% 472.1 8.7% 207.6 7.8% 44.0% EASTERN REGION CRCAs 407.6 11.7% 74.1 4.0% 18.2% CLCAs 121.8 6.2% 25.3 3.1% 20.8% 529.4 9.7% 99.4 3.7% 18.8% TENSIFT REGION CRCAs 457.4 13.2% 166.8 9.0% 36.5% CLCAs 347.3 17.6% 104.6 12.7% 30.1% 804.7 14.8% 271.4 10.2% 33.7% KINGDOM OF MOROCCO CRCAs 3,472.4 100.0% 1,848.1 100.0% 53.2% CLCAs 1,972.0 100.0% 822.4 100.0% 41.7% 5,444.4 100.0% 2,670.5 100.0% 49.1% - 34 - CNCA Lending by Category of Investment. Tables 1-2 ANNEX 3 Page 1 of 2 APPRAISAL vs ACTUAL COMMITMENTS FIFTH PROJECT SIXTH PROJECT TOTAL FOR BOTH PROJECTS (1) (2) (1) (2) (1) (2) APPRAISAL ACTUAL (2)/(1) APPRAISAL ACTUAL (2)/{1) APPRAISAL ACTUAL (2)/(1) ON-FARM INVESTMENTS CONSTRUCTION 390.2 422.8 108.4% 874.0 218.7 25.0% 1,264.2 641.5 50.7% LAND IMPROVEMENT 44.4 19.3 43.5% 205.5 125.1 60.9% 249.9 144.4 57.8% IRRIGATION 302.3 172.3 57.0% 734.9 277.8 37.8% 1,037.2 450.1 43.4% FARM EQUIPMENT 844.7 707.7 83.8% 614.9 907.3 147.6% 1,459.6 1,615.0 110.6% DRAFT ANIMALS 509.4 432.1 84.8% 356.1 192.8 54.1% 865.5 624.9 72.2% LIVESTOCK 957.5 2,025.4 211.5% 1,234.6 1,448.4 117.3% 2,192.1 3,473.8 158.5% PLANTATIONS 110.8 112.7 101.7% 178.5 133.2 74.6% 289.3 245.9 85.0% GREENHOUSES 62.5 163.8 262.1% 259.6 232.0 89.4% 322.1 395.8 122.9% OTHER 64.7 331.6 396.3 SUB-TOTAL 3,221.8 4,120.8 127.9% 4,458.1 3,866.9 86.7% 7,679.9 7,987.7 104.0% NON-FARM INVESTMENTS AGRO-INDUSTRIES 149.3 155.8 104.4% 190.0 288.9 152.1% 339.3 444.7 131.1% RURAL HOUSING 124.1 148.0 325.4 219.9% 148.0 449.5 303.7% FISHERIES 38.9 38.9 FORESTRY 0.3 0.3 HANDICRAFTS 15.6 TOTAL 3,371.1 4,400.7 130.5% 4,796.1 4,520.4 94.3% 8,167.2 8,936.7 109.4% - 35 - ANNEX 3 Page 2 of 2 ACTUAL COMMITMENTS BY FISCAL YEAR A. HQ & CRCAs 1985 1986 1987 1988 1989 1990 TOTAL CONSTRUCTION 47.1 43.5 35.6 23.2 15.2 15.8 180.4 LAND IMPROVEMENT 1.5 1.3 0.9 2.1 4.1 103.9 113.8 IRRIGATION 19.4 14.9 9.5 19.2 26.8 153.8 243.6 FARM EQUIPMENT 218.6 238.7 250.2 346.0 313.5 236.4 1,603.4 DRAFT ANIMALS 20.2 22.8 19.6 12.5 8.3 7.6 91.0 LIVESTOCK 316.8 231.9 229.3 171.0 116.9 123.2 1,189.1 PLANTATIONS 26.4 41.5 43.5 36.0 45.6 50.8 243.8 GREENHOUSES 48.6 64.3 50.9 90.0 77.6 64.4 395.8 OTHER 14.0 28.9 20.3 33.1 120.2 174.4 390.9 AGRO-INDUSTRIES 48.9 63.1 43.8 87.8 73.9 127.2 444.7 RURAL HOUSING 60.0 48.0 16.1 34.9 141.5 144.8 445.3 FISHERIES 38.9 38.9 FORESTRY 0.3 0.3 HANDICRAFTS 15.6 15.6 SUB-TOTAL 821.5 798.9 719.7 855.8 943.6 1,257.1 5,396.6 B. CLCAs 1985 1986 1987 1988 1989 1990 TOTAL CONSTRUCTION 87.5 106.1 103.0 59.7 54.3 50.5 461.1 LAND IMPROVEMENT 5.2 5.7 4.7 1.9 4.8 8.3 30.6 IRRIGATION 37.1 49.6 41.8 22.5 24.0 31.5 206.5 FARM EQUIPMENT 0.2 4.6 6.2 0.6 11.6 DRAFT ANIMALS 138.0 132.0 99.5 54.4 51.5 58.5 533.9 LIVESTOCK 409.2 429.4 408.8 355.6 310.8 370.9 2,284.7 PLANTATIONS 0.4 0.4 0.5 0.4 0.2 0.2 2.1 GREENHOUSES 0.0 OTHER 0.6 0.3 0.6 1.5 2.1 0.3 5.4 AGRO-INDUSTRIES 0.0 RURAL HOUSING 4.2 4.2 SUB-TOTAL 678.0 723.5 659.1 500.6 453.9 525.0 3,540.1 CNCA: ACTUAL TOTAL 1,499.5 1,522.4 1,378.8 1,356.4 1,397.5 1,782.1 8,936.7 APPRAISAL ESTIMATES FIFTH PROJECT 1,168.9 1,390.0 2,558.9 SIXTH PROJECT 1,895.1 2,299.6 2,767.2 6,961.9 SEVENTH PROJECT 1,600.0 2,010.0 2,819.0 6,429.0  - 37 - APPENDIX TRANSLATION KINGDOM OF MOROCCO CREDIT AGRICOLE General Director No 547/92/DG Rabat, October 1, 1992 Mr. Graham Donaldson, Chief Operations Evaluation Department International Bank for Reconstruction and Development 1818 H Street N.W. Washington DC 20433 U.S.A. Dear Mr. Donaldson: In reply to your letter of July 22, 1922 enclosing the preliminary evaluation report on the fifth and sixth Agricultural Credit projects, I am sending you herewith my remarks on the contents of that document. I should like first of all to congratulate the team of experts who produced the evaluation in question on the fine quality of the work and the distinct highlighting of the development prospects for Credit Agricole in spite of the particularly difficult situation which beset the 1991-1992 crop year. The first group of my comments has to do with the congruence of the figures: that subject is addressed in a detailed note attached to this letter. In examining the conclusions of the report, I shall confine my remarks to the topics of financial results (4.3), the settlement of arrears (4.9), the exchange risk (4.11) and, finally, the future of Credit Agricole (4.15) 1. Financial Performance Before tackling this subject, I think it is important to note that the CNCA had, at the appropriate time, already conveyed to the experts responsible for the evaluation the limitations of the Jacob Yaron model for calculating the degree of dependence on subsidies. Since very little time was available for the experts to apply the model to Credit Agricole, it had been decided to take this question up with the model's author himself in the course of July 1992. While awaiting the adaptation of the Subsidy Dependence Index (SDI) to the specific case of CNCA (since Mr. Jacob Yaron was unable to keep the July appointment because of other - 38 - engagements), judgment on the subsidy amounts estimated by the evaluation mission (see paragraph 3.17) should be reserved. Moreover, as rioted in the report, Credit Agricole today is in a more solid financial position than in the past, thanks to a stronger policy for setting up reserves for bad debts--one that is better able to cope with the risks facing the agricultural sector. It should be noted in this connection that our external auditors have--for the first time--stated that the reserves set up in 1990 were adequate, taking into account the risks inherent in the agricultural sector. In my opinion, however, strengthening of this financial structure is not achieved solely by raising interest rates, but also--and most importantly--by the establishment of a guarantee fund as insurance against natural disasters in order to protect the institution from other than banking risks. 2. The Settlement of Past Due Accounts The mission was able to observe the importance the CNCA assigns to the settlement of overdue financial obligations and the impressive efforts expended by Credit Agricole to resolve that situation. It was also in this context, and due to the drought experienced in the 1991-92 crop year-- the adverse impact of which had been noted by the experts--that steps were taken to clean up the arrears situation and ease the farmers' debt burden. To that end, it was decided to cancel the late payment charges for debtors who pay the amounts owed within the allotted period, and to grant rescheduled loans to the farmers who pay off part of their dues. I should like to point out, along the same lines, that the statistics presented by the CNCA reflect the account ledgers; and that inconsistencies noted by the evaluation mission stem from a confusion between the recovery rate on current maturities for the fiscal year and the rate on all accounts payable, including those for earlier collection periods. Generally speaking, the results posted as a result of the 1985 rescheduling have been satisfactory, when the region in question has experienced successful crop years in the wake of that operation. The analysis of previous overdue accounts should have taken into account the regional concentration of such arrears. Three regions (the east, Haouz and the south) were chronically beset by disasters throughout the 1980s and at the start of the 90s, and they account for more than 70% of the entire CNCA arrearages. This is why the development of Credit Agricole in those regions must include more widely diversified activities and risks, as well as the establishment of a guarantee fund for protection against natural disasters. 3. System for Coverage of Exchange Risk According to the simulation exercises conducted, application of the new exchange risk coverage system to external loans contracted by the CNCA will boost the interest rate by three to four points. This increase would jeopardize the profitability of the projects financed, and it will undoubtedly have an adverse effect on loan recoveries. - 39- While awaiting the implementation of specific financial mechanisms to cover exchange risks, it is eminently desirable to continue using the exchange coverage system negotiated in May 1989, (which was approved by the World Bank), for the new lines of credit that the CNCA hopes to mobilize in 1993. 4. The Future of Credit Agricole Originally a financial institution specializing in funding for agriculture, the CNCA has now begun to diversify its activities so that it will gradually become a universal bank for service to the rural community and, eventually, a bank that is able to provide its customers with all categories of banking services. That long and difficult process calls for an evolution strategy tailored to the country's specific needs as well as to its economic and social situation. The efforts deployed since 1987 are aimed at strengthening the institution's financial structure, cleaning up previous arrears, diversifying financing to spread the risks; actively attract savings deposits; strengthen the branch network; enhance the reliability of management control tools; and invest in the training of Credit Agricole's managers, professional staff and agents. This new strategy also calls for instituting basic reforms needed to meet the competition, which has become increasingly evident since the departitioning decreed by the monetary authorities in 1991. What are those necessary reforms? * First and foremost, revision of the text of the law governing Credit Agricole to adapt it to the reality of the nationwide financial sector reform and the changes in the national economy. * In addition, the establishment of a Guarantee Fund to cover Natural Disasters. * And finally, completion of the reforms currently in progress, covering administrative, accounting and data processing procedures, implementation of a management information system, and upgrading of the services offered by the Credit Agricole. Only when all of these prerequisites have been met and Credit Agricole's public service role as the source of financing for small farmers has been established--with all the associated implications in regard to the mobilization of stable concessional resources--only then can we contemplate the participation of commercial banks in the development of rural financial markets. There you have some observations I wanted to contribute to the important evaluation report which reviews almost eight years of Cred'i Agricole activity (1983-1990). To conclude, I would like once more to assure you of the need to take into account and support the strategy now being implemented to mobilize the means required for its complete success, without undue haste, in the knowledge that the most important mission of Credit Agricole will always be to make an effective contribution to improve living conditions for farmers and the rural population by sponsoring productive and profitable investments. - 40 - Thank you again for your cooperation and for the World Bank's ongoing assistance to Credit Agricole. Please accept, Sir, the assurances of my highest consideration. M. Rachid Haddaoui, Director General - 41 - Observations of the Caisse Nationale de Credit Agricole (CNCA) on the Contents of the Performance Audit Report on the Fifth and Sixth Lines of Agricultural Credit 22 July 1992 Page viii--5 - The special account in the fifth line of credit (IBRD 2367) was used once: the CNCA obtained an advance of US$5.9 million in the context of this account on April 25, 1986. - When calculated in accordance with the provisions of the IBRD loan agreement No. 2367- MOR (Article V, Section 5.07), the liquidity ratio is higher than 1.2 throughout the entire period considered (see the corrected table: financial position, sheet attached) - The long-term debt to equity ratio, set at 6:1 in IBRD loan contract 2367- MOR, rose from 5.36 on August 31 1985 to 6.75 on August 31, 1987. It remained well below the 8:1 figure established in IBRD loan contract 2731-MOR (see the corrected table: financial position sheet attached) Page x--paragraph 11 - See the foregoing comment on the liquidity ratio. Page 3--paragraph 2.4 - The definition of the liquidity ratio was never changed, and current assets include only those with maturities of 12 months or less. Page 5--paragraph 2.9 - CNCA loans exceeded the evaluation's forecast by 9.4% (see Annex 3, Table 1 of the July 1992 performance audit report) Page 13 The average indebtedness of CRCA and CLCA customers has risen by 77 per cent and 62 per cent, respectively. Page 15--Table 8 Average amount of subloans granted by the CRCAs 1984-85: 24,566 instead of 7,981, which represents the average size of CRCA and CLCA loans 1985-90: 70,986 instead of 18,051, which represents the average size of CRCA and CLCA loans Page 15--paragraph 3.13 - paragraph 3.13: deposits as a share of total CNCA resources fell from 14.3 per cent in 1985 to 12.4 per cent in 1988 before rising to 16.4 per cent in 1990 and 18.1 per cent in 1991 - line 10: during that period, external obligations increased faster than internal financing, rising from 50.2 per cent to 55.1 per cent of total liabilities. - 42 - It should be noted, however, that the balance outstanding on external loans is reevaluated at the year-end exchange rate, and it includes the portion of potential exchange shortfall for which the Treasury is responsible. In terms of external resources without the reevaluation, external borrowings as a share of total resources account for only 38.6 per cent in 1985 and 45.1 per cent in 1990. Page 15--paragraph 5.14 - Liquidity ratio (see preceding remarks) - The special account was used in the context of the fifth line of credit. Page 16--paragraph 3-14 - The liquidity ratio (idem) - The special account (idem) Page 17--Table 9 Some of the financial ratios are erroneous. Please see the attached table, which includes the financial ratios agreed upon with the IBRD. Page 20--Table No. 10 The collection rates need to be corrected. 1988 1989 1990 C.L.C.A. 51% 51% 42% C.R.C.A. 62% 59% 53% Headquarters 95% 82% 96% CNCA Average 62% 58% 58% Arrears as a Percentage of the Loan Portfolio 1987 C.L.C.A 27 Headquarters + CRCA 24 CNCA Average 25 Page 20--paragraph 3-21 Despite an increase of 154 per cent instead of 150 percent. Page 22--paragraph 3-26 The reserves for bad debts declined from 25 per cent to 23 per cent instead of 39 per cent. Page 24--paragraph 3-29 It is not correct that the State assumed the entire exchange risk in connection with the fifth project and earlier projects: the State assumed only a portion of that risk. Please refer to paragraph 1.4, section II, part one of the study on the Second National Agricultural Credit Project (May 1992), which traces the historical apportionment of the exchange risk. Reserves for exchange risk were set up in 1986 and amounted to 108 million dirhams in 1989. - 43 - Paragraph 28--paragraph 4.10 Deposits increased by more than 50 per cent in real terms between 1988 and 1990, rising from 14.3 per cent to 16.4 per cent of CNCA's resources. Statistical Annexes Annex 1, page 1: - The balance sheets include reserves for bad debts starting in 1988, in contrast to the CNCA figures, in which outstanding loans are net of reserves. - The loan loss reserve was not included in the equity capital. - The CLCA support fund and KFW equipment subsidy should be included in the quasi-equity figures to be in accord with the CNCA presentation. This difference in presentation results in differences in the ratios calculated on page 52, Annex 1 4/4. -44- FINANCIAL RATIO PROFILE 31-8-85 31-8-86 31-8-87 31-12-88 31-12-89 31-12- TOTAL ASSETS 6,453.9 8,410.2 9,422.4 10,890.2 11,942.7 13,122.7 TOTAL RISK ASSETS 4,425.0 5,668.4 6,530.0 8,075.6 8,718.1 10,033.5 CURRENT ASSETS 3,063.4 4,152.70 4,955.70 5,850.6 7,257.8 8,077.1 CURRENT LIABILITIES 2,238.0 2,780.20 2,877.80 3,624.0 4,305.9 4,507.7 RISK ASSETS (M<) 3,476.9 4,318.7 4,840.8 5,604.8 5,963.4 6,681.8 FIXED ASSETS 42.1 53.8 61.8 77.9 95.2 153.6 INVESTMENTS 5.8 5.8 6.0 249.6 214.6 185.7 LOAN PORTFOLIO 4,377.1 5,665.3 6,528.7 8,072.3 8,691.4 9,999.9 PROVISIONS FOR BAD DEBTS 214.9 285.6 409.1 203.2 435.6 645.3 EQUITY CAPITAL 593.5 665.1 746.4 1,324.5 1,347.7 1,386.3 QUASI-EQUITY CAPITAL 311.7 385.2 113.4 135.9 140.0 144.1 TOTAL MED.& LONG-TERM DEBTS 3,179.9 4,370.2 5,034.5 5,580.6 5,877.6 6,824.4 TOTAL DEPOSITS 921.6 1,143.3 1,201.8 1,316.7 1,751.2 2,152.5 CURRENT PERIOD NET EARNINGS 71.6 81.4 87.5 23.1 38.6 39.1 1. OPERATING ACCOUNT RATIOS Return on Risk Assets % 1.62 1.44 1.34 0.28 0.44 0.39 Return on Total Assets % 1.11 0.97 0.93 0.21 0.32 0.30 Return on Equity Capital % 12.06 12.23 11.72 4.09 2.86 2.82 II. BALANCE SHEET RATIOS (%) Total Loans/Deposits 474.95 495.52 543.24 613.07 496.31 464.57 Liquidity Ratio 1.37 1.49 1.72 1.61 1.69 1.79 Loan Loss Reserve as % of Risk Assets 4.86 5.04 6.27 2.51 5.00 6.43 M & LT Debts as % of Equity Capital 5.36 6.57 6.74 5.30 4.36 4.92 Fixed Assets as % of Equity Capital 7.10 8.09 8.28 8.51 7.06 11.08 Investments+ Fixed Assets as % of Eq.Cap. 8.08 8.96 9.09 9.45 22.98 24.48 Total Equity as % of Risk Assets 20.46 18.53 13.17 11.33 17.06 15.25 Total Equity/Assets at Risk + Fixed & Investment Assets 20.24 18.33 13.03 11.32 16.48 14.75 Total Equity as % of Total Assets 54.10 56.54 54.63 8.41 50.39 53.10 Total Equity as % of Med.& Long Term Debt 28.47 24.03 17.08 18.87 25.31 22.42 M & LT Risk Assets as % of Total Risk Assets 78.57 76.19 74.13 69.36 68.40 66.60 Growth Rate of Deposits 24.06 5.12 27.57 33.00 22.92 Growth Rate of Loans 29.43 15.24 10.37 7.67 15.05 Growth Rate of Total Assets 30.31 12.04 13.27 9.67 9.88 Growth Rate of Equity 12.06 12.23 9.77 1.75 2.86 Note: Total balance considered up to 31 Aug 87 includes provision for bad debt. - The 1988 growth rates are calculated relative to 31 Dec 87. - Equity capital = Initial Capital + General Reserves + Special Reserves. - MLT (medium and long-term) Debts = MLT Debts - Maturities of less than one year - Total risk assets = Gross loan portfolio + securities portfolio. IBRD 19570 ir -r MOROCCO/MARéC: 2 SIXTH AGRICULTURAL CRÉDITPROJEMirne:n SIXIÉME PROJET DE CRÉDIT AGRI L BRANCH NETWORKOF THÉ NATIÔNAL AGRId.ULTi REOI AN RÉSEAU DE l A CA ISSE NATIONAL E DE °v"l F CNCA NE REUE-CC- 5.dg. - - C R CA "I -L E I *CILCAPI CRCA CLCA Cj. A - - -f~ JM -.- J RAWNWALL/PLUVO TRIE -srå. ~ Isohyes ,gn .. eter ^R øAfe /soyes ew,n &-merne . • A.ogol,anl fall. - - O n- Mo~mnne a""è d"eoépn u.n 800+ 600 -800 200 - 00 0-200 ~~OM.M k • A*g . i 9 Nor.oal capia a..hrA iera.onal bounda,.es - - - F,rd.e,'e5 d aar --. f - o B•. EI 0 elaa-*-- ind.ca'es the .er dory of the Iormer >Errock d.a Span.ih Sa.haa (Western Sahara s,. ir.qe le terrt-a de G-aceen - Che I. . P mek Oihør zaa - o Ooso-a TT.~ 3r 3r ""3. '3MAUR'ÉTAN1A %6O~ET%RM RCH 19E0 00 5 a 0-- ? ~~øE P0n -43.S 00. FC'..., ALGERIA N rARTAI I I MAhI- Då.~~~MAC -I--------