Document of The World Bank

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Document of The World Bank IMPLEMENTATION COMPLETION REPORT (TF TF IDA PPFI-P8390) ON A CREDIT IN THE AMOUNT OF US$ 15 MILLION TO THE KYRGYZ REPUBLIC FOR A PRIVATE ENTERPRISE SUPPORT PROJECT MARCH 30, 2004 Report No: 28378

2 CURRENCY EQUIVALENTS (Exchange Rate Effective March 30, 2004) Currency Unit = KGS KGS 1 = US$ US$ 1 = KGS FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS CAS DITAC EBRD ERIS FERD FINSAC Goskominvest IFC KPPA NBKR NPRS PAU PCB PDO PESAC PESP PHRD ROE SAR SSN USAID WTO Country Assistance Strategy Directorate for Investment and Technical Assistance Coordination European Bank for Reconstruction and Development Entrepreneurial Referral and Information Service Fund for Enterprise Restructuring and Development Financial Sector Structural Adjustment Credit State Commission on Foreign Investment and Economic Assistance International Finance Corporation Kyrgyz Project Promotion Agency National Bank of the Kyrgyz Republic National Poverty Reduction Strategy Project Administration Unit Participating commercial bank Project development objective Privatization and Enterprise Sector Adjustment Credit Private Enterprise Support Project Policy and Human Resources Development Return on equity Staff Appraisal Report Social Safety Net Credit United States Agency for International Development World Trade Organization Vice President: Country Director Sector Manager Task Team Leader/Task Manager: Shigeo Katsu Dennis de Tray Tunc Uyanik Alexandra Gross

3 KYRGYZ REPUBLIC Private Enterprise Support Project CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 1 4. Achievement of Objective and Outputs 6 5. Major Factors Affecting Implementation and Outcome 9 6. Sustainability Bank and Borrower Performance Lessons Learned Partner Comments Additional Information 18 Annex 1. Key Performance Indicators/Log Frame Matrix 19 Annex 2. Project Costs and Financing 22 Annex 3. Economic Costs and Benefits 25 Annex 4. Bank Inputs 26 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 28 Annex 6. Ratings of Bank and Borrower Performance 29 Annex 7. List of Supporting Documents 30 Annex 8. Beneficiary Survey Results 31 Annex 9. Stakeholder Workshop Results 36

4 Project ID: P Team Leader: Alexandra L. Gross Project Name: Private Enterprise Support TL Unit: ECSPF ICR Type: Intensive Learning Model (ILM) of ICR Report Date: March 30, Project Data Name: Private Enterprise Support L/C/TF Number: TF-22161; TF-22169; IDA-27170; PPFI-P8390 Country/Department: KYRGYZ REPUBLIC Region: Europe and Central Asia Region Sector/subsector: Other industry (95%); Banking (5%) Theme: Other financial and private sector development (P); Regulation and competition policy (P) KEY DATES Original Revised/Actual PCD: 08/20/1993 Effective: 08/14/ /17/1996 Appraisal: 06/15/1994 MTR: 03/27/ /10/1997 Approval: 05/04/1995 Closing: 06/30/ /30/2003 Borrower/Implementing Agency: Other Partners: KYRGYZ REPUBLIC/GOSKOMINVEST Japanese Government STAFF Current At Appraisal Vice President: Shigeo Katsu Johannes F. Linn Country Director: Dennis de Tray Michael A. Gould Sector Director: Anil Sood Kadir T. Yurukoglu Team Leader at ICR: Alexandra Gross Mohamadou Diop ICR Primary Author: J. Matthew Mitchell; Carlo Segni 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: Sustainability: Institutional Development Impact: Bank Performance: Borrower Performance: U L M U S QAG (if available) Quality at Entry: Project at Risk at Any Time: Yes ICR U 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The original objective of the Private Enterprise Support Project (PESP) was to provide an integrated

5 program of technical assistance and loans for working capital and investment to private enterprises to promote their successful adaptation to a market-based system. This objective was an important element of the country s economic reform program, which aimed to address the major macroeconomic instability that characterized the first years of Kyrgyz independence, and promoted the transition from the centrally-planned system of the Former Soviet Union to a market-based economy. In response to these challenges, the Bank developed a comprehensive series of rehabilitation, adjustment and investment operations designed to stabilize the economy, create a conducive environment for private sector development, and finance viable investment projects and technical assistance. The 1993 Country Assistance Strategy (CAS) set the strategic context for the PESP and three other operations: the Rehabilitation Credit in 1993; the Privatization and Enterprise Sector Adjustment Credit (PESAC) in 1994; and the Social Safety Net Credit (SSN) in The Rehabilitation Credit provided financing for critical imports to sustain the economy while adjustment and investment operations were under preparation. The PESAC concentrated on improving the pace and quality of privatization and enterprise sector adjustment, emphasizing greater transparency in the process. The SSN Credit supported capacity-building to deal with the social aspects of economic transition, including poverty reduction and the deployment of displaced workers. PESP was the first investment operation to target the enterprise and banking sectors. In hindsight, the development objectives were too broad and difficult to achieve within the timeframe of the project. In addition, the project design and implementation arrangements were overly complex, involving (i) the establishment of a new, state-owned financial vehicle, the Fund for Enterprise Restructuring and Development (FERD) (ii) a Project Administration Unit (PAU); and (iii) a foreign consultant to manage the operations of the credit line and enter into a twinning operation with a local bank. As a result of these deficiencies, the project did not achieve its original objectives and was restructured in 1998 to channel resources instead through local commercial banks. It continued to perform poorly, however, and in 2001 was further modified to target small and micro enterprises. Subsequently, performance and disbursements began to improve slightly. 3.2 Revised Objective: By 1998, only one sub-project for US$0.5 million had been financed under the credit line. As a result, the Bank and the Borrower agreed to restructure the project to concentrate on the Borrower's need to reform the enterprise and financial sectors, and strengthen the capacity of the commercial banking system. The project became a credit line providing financing to private enterprises through local commercial banks. The FERD was dissolved and responsibility for implementation of the credit line was transferred to the National Bank of the Kyrgyz Republic (NBKR) which established an Apex Unit for the project within the Directorate for Investment and Technical Assistance Coordination (DITAC). The Investment Promotion Unit at Gaskominvest assumed responsibilities for the technical assistance component. The restructured project included new performance indicators. The key indicators were: (i) improved risk management and management information systems combined with better appraisal capability in commercial banks; (ii) enhanced risk management systems of the NBKR; (iii) improved infrastructure in the PAU for providing project preparation guidance to private enterprises; and (iv) improved financial management and project preparation. 3.3 Original Components: Lending Component: The lending component consisted of US$14.2 million for investment and working - 2 -

6 capital sub-loans through the FERD to eligible enterprises. Eligible enterprises included viable and sound private enterprises and state-owned enterprises at an advanced stage of privatization. A private enterprise was defined as having at least 51 percent control by private investors. The project included lending for incremental working capital sub-loans up to US$1.0 million equivalent and free-standing working capital sub-loans up to US$200,000 with terms between one and eight years. In order to receive project funds, eligible enterprises had to produce financial statements and a feasible business plan. The project required enterprises to assume the foreign exchange risk and pay market-based interest rates. Beneficiaries also had to meet minimum financial and economic requirements, such as debt servicing capacity in foreign exchange, satisfactory environmental reviews, and have a financial and an economic rate of return for their project of at least 15 percent. The design of the lending program was based on a survey of 50 enterprises during project preparation. This survey resulted in a pipeline of 27 eligible sub-projects with total estimated investment cost of US$49 million and a related credit demand of US$34 million. The preparatory work also included a review of other donor-financed lending products to ensure there was no overlap of donor resources. Among other donors, for example, the European Bank for Reconstruction and Development (EBRD) had established a credit line for large enterprises and had an estimated pipeline of US$20.4 million. Technical Assistance Components: The project included two technical assistance components: (i) The technical assistance for enterprises (US$620,000) consisted of consulting services to prepare diagnostic studies and business plans of individual firms and provide technical, managerial, and financial assistance to help implement the proposed business plans. Foreign firms primarily provided these services after having signed partnership agreements with local firms. Eligible firms were required to cofinance the contracts. A Japanese PHRD grant also supported this component with funding for diagnostic and pre-feasibility studies of sample enterprises and established regional centers to provide local enterprises with access to market information. (ii) The technical assistance for the financial sector complemented other donor activities in the sector and included: (i) a diagnostic audit of Savings Bank (US$200,000); (ii) a review of the legal and regulatory framework of insurance companies (US$100,000); and (iii) a trade finance study (US$50,000). Implementation Arrangements: During project preparation, the Bank and the Borrower determined that local banks had neither the operational capability nor the capital to engage in lending activities or manage a credit line. Thus, in order to bypass the weak banking system, the Fund for Enterprise Restructuring and Development (FERD) was established under the project to channel funds to eligible enterprises in the form of sub-loans. The FERD was a government-owned financial intermediary vehicle. It had no staff, minimum capital, physical assets or structure of its own. Rather, it was materialized by its statutes, operating procedures and a bank account at the NBKR. Its operations were fully managed by an international bank that was selected as the Fund Manager financed under a Japanese PHRD grant and entered into a twinning arrangement with a local bank. The Fund Manager was responsible both for the lending and technical assistance components of the project. The Bank retained the right to approve all sub-projects approved by the FERD. The goal was for the Fund Manager to train staff of the local bank, the Orient Bank, to progressively take over some of the management and lending responsibilities of the FERD. At the same time, as the banking sector developed, other local banks would take over the FERD's portfolio

7 The FERD was supervised by a Board of Directors composed of the Chairman of the NBKR, the Minister of Finance, and the Vice Chairman of the State Commission on Foreign Investment and Economic Assistance (Goskominvest). The FERD was intended to be an interim financial vehicle, and thus, the project included a sunset provision for it of 10 years. A Project Administration Unit (PAU) was established in Goskominvest to monitor the execution of the management contract for the FERD. Other responsibilities of the PAU included (i) coordination between the Bank and the Government; (ii) ex-post review of sub-loans; (iii) supervision of consultants hired under the technical assistance components; and (iv) implementation of technical assistance under the Japanese grant. The original implementation arrangements for the project proved to be overly complex and ineffective. The responsibilities of the PAU, the FERD's manager and the Board of the FERD overlapped and resulted in a number of conflicts during project implementation. Furthermore, project implementation was affected by the poor performance of the Fund Manager, and the fact that its local counterpart bank went bankrupt during an early stage of implementation. The project s design allowed for changes in the implementation framework, with qualified commercial banks taking over the FERD's lending portfolio, but it was too open-ended and dependent on the development of the banking sector. Lessons from Past Projects: The Bank incorporated several important lessons from experience with financial intermediary lending operations in designing the project. Specifically, the preparation team focused on how such operations can be hampered by unstable macroeconomic environments, the absence of market determined interest rates, political interference in lending decisions, and weak and unsound financial institutions. Thus, the project contained an explicit provision to require a market-based interest rate structure. In addition, to minimize political interference and, at the same time, address weaknesses in the private banking sector, the project designated a private international bank to manage the lending program instead of creating a new government institution. Benefit/Risk Evaluation: The project s evaluation of benefits in the Staff Appraisal Report (SAR) was inadequate. The SAR states the project would support successful post-privatization performance through improvements in management efficiency and productivity, marketing, and profitability. It adds that these improvements would result in tangible benefits for investors, workers and the country as a whole, without defining them. The SAR also notes that the project would: (i) develop the institutional foundation for financing projects with high financial and economic returns and technical assistance to enterprises; and (ii) contribute to capacity-building for institutional reform in the financial sector, which a follow-up financial sector loan would support. However, the SAR does not sufficiently explain the linkage between these expected results and the detailed indicators of project achievements provided in Annex 5 of the report. In this context, it does not fully assess the transition process for building local commercial bank capacity to take over the management of the FERD from the international bank and how this process relates to the timeline for transition and project performance indicators. 3.4 Revised Components: In 1998 the Borrower and the Bank restructured the project in an effort to improve performance. The restructured project redesigned both lending and technical assistance components and introduced adjustments to take into account emerging market conditions and to reflect a more realistic view of credit demand

8 Under the new implementation arrangements, funds were channeled through qualified participating commercial banks (PCBs) directly rather than through the FERD. The FERD was dismantled. The NBKR took responsibility for the management and supervision of the credit line. As a result, DITAC, established by the NBKR in 1995, became the Apex Unit of the project. Monitoring financial soundness of the PCBs relied on the general banking sector supervision activities of the NBKR. The application/approval procedures for sub-loans remained complex, but more clearly defined. The PCB became solely responsible for credit decisions, while the Apex Unit assumed responsibility only for supervision and monitoring. The Bank retained the right of final approval of each sub-loan, with the understanding that, as PCBs progressed in terms of credit and risk management capacity, these arrangements would be modified to fully entrust PCBs to approve sub-projects introducing with only a post-review by the Apex Unit. The Bank also imposed strict eligibility requirements for PCBs in order to select the best performing banks in the country. These requirements included sound financial condition and management and annual IAS-based audits by auditing firms satisfactory to the Bank (throughout the life of the project, none of the local auditing firms were qualified by the Bank to conduct IAS-based audit of banks). The terms and conditions for the sub-loans did not change substantially. Sub-loans up to a maximum of US$1.0 million were extended by PCBs to eligible private and privatized enterprises for their medium- and long-term investment and working capital needs. The maximum maturity was seven years for investment sub-loans and two years for working capital sub-loans. Interest rates to PCBs were set at US$ six months LIBOR plus 2 percent to cover operating costs. PCBs were given the option to charge sub-borrowers a 7 percent margin over their funding cost. The PCBs assumed the foreign exchange risk as funding was provided in US$ and PCBs could extend sub-loans in both US$ and Soms. The project beneficiaries were required to meet the same financial and economic requirements set forth in 1995 Credit Agreement. The technical assistance component was revised to include provisions for consultancy services to: (i) help enterprises prepare and implement business plans and investment projects; and (ii) train commercial bank officers with a view to strengthening their capacity for project appraisal, risk management and loan delivery. Additionally, the PHRD grant available for the project implementation provided funding for: (i) the development of sub-project appraisal software; (ii) the development of local audit capacity through training in international accounting standards; and (iii) joint training sessions for private enterprises and commercial banks in preparing and assessing investment projects and business plans. The Apex Unit became responsible for technical assistance to PCBs and the PAU was restructured into an Investment Promotion Unit at Goskominvest and named PIU. The PIU then became responsible for technical assistance to enterprises. Despite the changes to the project design, the demand for credit funds and technical assistance remained sluggish. To stimulate demand, there were two further changes to the project design. In December 2001, the Bank and the Borrower agreed to introduce a micro-lending component. Under this component, approval procedures for sub-loans of less than US$5,000 with a maturity of less than six months were streamlined, requiring only post-review by the Apex Unit, and no prior review by the Bank. The PHRD grant financed consultants (i) to help the PCBs develop and implement new crediting and risk management procedures and train credit officers on micro-lending, and (ii) to help medium-size enterprises (above US$5,000 sub-projects) to prepare business plans. In 2002, an amendment to the Credit Agreement enabled: (i) individuals to become project beneficiaries; and (ii) the Ministry of Finance to bear the foreign exchange risk. The amendment also reduced the amount of the credit line to US$5.5 million and extended the closing date of the project to June This extension was granted with the idea that experience with the microfinance component would contribute to the preparation of a new credit line. The Bank agreed with the Borrower that the new credit line would take - 5 -

9 place only if disbursement and the overall quality of the sub-projects improved during the last year of project implementation. 3.5 Quality at Entry: Project appraisal took place before the establishment of the Bank s formal quality assurance review process. In general, the project objectives and design: (i) responded to the priority needs of the country; (ii) reflected the Bank s overall country assistance strategy; (iii) fit into a logical framework of ongoing projects; and (iv) took into account of some lessons learned from past projects. However, because of substantial deficiencies in risk analysis, overly complex institutional arrangements, and lack of adequate market analysis, overall quality at entry is rated as unsatisfactory. The Bank did not adequately assess the risks of creating a new state-owned institution for credit delivery, relying heavily on a foreign contractor (the Fund Manager) to ensure effectiveness. Moreover, the absence of clear division of responsibilities between the FERD Board, the PAU and the Fund Manager created overlaps and conflicts, thus undermining the overall project implementation. While the project s restructuring improved the framework for credit delivery, the complex lending procedures potentially could have been avoided with better preparation. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The overall project outcome is rated as unsatisfactory. This rating is based on original project objectives and design because the revised objectives resulted primarily from deficiencies in the original project design and other external factors. The lending component performed poorly and showed only modest improvements following the restructuring in Following the amendments in 2001, there was a slight increase in demand for sub-loans, especially for micro-loans, and the participating commercial banks gained some experience in credit risk and management and lending to small enterprises. Technical assistance had no significant impact on the lending component before project restructuring in 1998, mainly due to the reluctance of enterprises to borrow for technical assistance, and the unsatisfactory performance of the Fund Manager. After restructuring, the execution of a smaller, more clearly focused technical assistance component helped improve disbursements and the project outcome. However, actual financial resources delivered to the private sector over the expanded seven-year life of the project amounted to only 24 percent of the original level and about 30 percent of the revised target at 1998 restructuring. 4.2 Outputs by components: Line of Credit Component: There were essentially three phases in the implementation of the lending component. During the first phase, from the Credit s effectiveness to 1998 restructuring, performance was highly unsatisfactory. The lack of disbursement activity resulted in large part from the lack of demand for credit, largely due to complex application procedures, poor economic conditions, perceived high interest rates, excessive collateral requirements, and an unwillingness of enterprises to borrow for technical assistance to prepare sub-project applications. During the second phase, from 1998 to December 2001, disbursements increased once the funds were channeled directly through commercial banks. While the sub-loan approval process remained somewhat complex with a three-step approval process involving the PCB, the Apex Unit and the Bank, the new - 6 -

10 implementation arrangements were a positive development. Under the new arrangements, the Apex Unit was in charge only of the supervision of PCBs' activities. Furthermore, the grant-funded technical assistance helped banks improve their subproject appraisal and risk management skills. As a result, the project financed 18 sub-projects totaling about US$2 million. This level was still far below the target of US$12 million for the restructured project, and the project had only one more year of operation. However, during this period the Bank rated the performance as satisfactory due to improvements in the quality of sub-projects and the delivery of technical assistance to enterprises. The third phase began in December 2001 with the shift in focus to micro-enterprises. The result was an increase in the number of sub-projects financed. Within one year, the project had extended 72 sub-loans totaling US$318,000. The Bank extended the project s closing date from June 2002 to June 2003 to allow for continued progress in the lending component. There was also an amendment to allow individual entrepreneurs--in addition to legal entities--to become sub-borrowers. The Ministry of Finance also agreed to assume the foreign exchange risk, enabling commercial banks to receive the proceeds of the project either in local currency or US dollars. At the end of the project, the lending component had financed a total of 293 sub-loans, of which 266 were small credits. Technical Assistance Components: Prior to 1998, there were no disbursements for technical assistance funds to support enterprises. Most of the activities financed from the Japanese grant (about US$3.3 million) did not have a clear development impact. This grant funding helped set up: (i) an Entrepreneurial Referral and Information Service (ERIS) to assist entrepreneurs in developing small and medium-scale businesses; and (ii) the Kyrgyz Project Promotion Agency (KPPA), responsible for implementing the ERIS system. KPPA was to contract with the FERD to provide technical assistance to enterprises and help build a pipeline of projects for an annual fee. Both local staff of the FERD and ERIS received intense training. However, the impact of the agencies was negligible, as enterprises made little of their services. From 1998 to June 2003, the technical assistance focused on training programs and seminars for both enterprises and commercial banks in business plan preparation, project appraisal and risk/portfolio management. The project also provided two foreign resident advisors to the credit departments of participating banks. In addition, the project developed software on project planning and analysis for dissemination to entrepreneurs and financial institutions. Goskominvest produced promotional materials and a television campaign to promote the project. Furthermore, it developed a website to promote the development of Kyrgyz enterprises, provide information to potential foreign investors, and offer a common forum for various government agencies to promote the Kyrgyz enterprises in world markets. Intensive Learning Survey and Workshop: At the end of the project, the Bank sponsored an Intensive Learning Survey and Workshop to assess the achievements of the project and extract lessons (see Annex 8). The workshop was held in Bishkek in June 2003 to discuss the findings of a survey of project stakeholders. About 60 enterprises and 20 officials and practitioners involved in the project responded to the survey. Despite the poor assessment of the project overall, the survey pointed to a number of positive outcomes resulting from the final changes to project design and the new emphasis on micro loans. The survey also indicated that the project helped PCBs improve their credit policies and risk management. In particular, the project s micro lending component helped the PCBs gain a better understanding of the benefits and constraints in downscaling the credit portfolio and focusing on small/retail banking. The Apex Unit gained experience in financial management under Bank-financed projects. The benefits for enterprises were: (i) longer-term financing at competitive rates, especially for medium sized enterprises; and (ii) simplified and more cost-effective approval procedures under the micro-lending component. In general participating - 7 -

11 enterprises reported improvements in revenues and product diversification. 4.3 Net Present Value/Economic rate of return: Not applicable to this project. 4.4 Financial rate of return: Not applicable to this project. 4.5 Institutional development impact: The project originally aimed to develop the institutional foundation for providing financing and technical assistance to private enterprises to develop and implement business plans. To do so, the project focused on building the capability of the FERD and the PAU. However, the developmental impact was low and these institutions were ultimately dismantled. Following the restructuring in 1998, institutional development focused on commercial banks. Overall, during this period, project supervision reports indicate that the PCBs improved the quality of credit appraisals and became more careful in evaluating business plans and collateral. The Intensive Learning Survey also indicated that PCBs improved their skills in credit risk evaluation and management and built capacity to lend to small borrowers. The high repayment rate on sub-loans under the project (95 percent) reflects these improvements. During project implementation, nine banks applied for participation, but only four were accepted due to tight eligibility criteria (DosCredo Bank, AKB Kyrgyzstan, IneximBank and Energo Bank). These four banks accounted for more than 25 percent of the assets in the Kyrgyz banking system. However, DosCredo was suspended in 2001 because it was undercapitalized. The bank did not manage to restore its capital position, and the NBKR limited its license about nine months after it was suspended from the project. Further, AKB Kyrgyzstan was suspended in 2002 because of the audits prepared by a local auditing firm, contrary to the project requirements. In 2003 the two remaining banks (IneximBank and Energo Bank) accounted for about 9.7 percent of banking sector in terms of assets. Section 10 (Additional Information) of this report provides background and more detailed information on performance of the individual PCBs. In the final years of the project, the PESP portfolio achieved high quality and steady growth. As of end-june 2003, about 95 percent of the sub-borrowers were repaying their loans, with problem loans amounting to less than 10 percent of the PESP portfolio. PCBs disbursed US$1.1 million in small sub-loans, and US$2.5 million in medium sub-loans, thus contributing to supporting private sector development. In aggregate, at end-2002, PESP sub-loans accounted for about 35 percent and 6 percent of the foreign currency denominated long-term (over one year) and short-term (less than one year) credit portfolio of the whole banking sector. Overall, the participating commercial banks (excluding DosCredo) have been profitable with capitalization improving to 25 percent on average as of the end-year 2002, against 22 percent for the whole sector. Return on operating assets during the same period was positive at 2.3 percent on average, compared to 1.6 for the whole banking sector. Asset quality is good compared with other Kyrgyz banks, with non-performing loans of about 12 percent on average for the three banks. The three PCBs are also among the best banks in terms of intermediation functions. On average, at end-2002, the three banks maintained a deposits-to-assets ratio of 50 percent, a loans-to-deposits ratio of 136 percent, and loans-to-total assets of 51 percent. This compared favorably to the whole sector for which the deposits-to-assets ratio was 45 percent, loans-to-deposit 61 percent, and loans-to-total assets 23 percent. Furthermore, the three banks - 8 -

12 enjoy depositors confidence, which has been a major weakness in the Kyrgyz banking sector. Their share of deposits to total banking sector deposits in fact grew from 18.8 percent in 1999 to 21.4 percent in IneximBank had the largest portfolio of the participating commercial banks and disbursed about 48 percent and 78 percent of medium-size sub-loans and small sub-loans, respectively. AKB Kyrgyzstan disbursed 41 percent of the total PESP medium-size portfolio. The bank did not disburse micro sub-loans, as it was suspended before the new component was developed. Energo Bank disbursed 22 percent of the micro sub-loans and 10 percent of the medium size sub-loans. DosCredo Bank disbursed only 1 percent of the medium size sub-loans. Project supervision reports also indicate that the PESP helped support the institutional development of the Apex Unit. During the final phase of the project, the Apex Unit improved its capability to supervise Bank-financed projects, including monitoring procurement and special accounts, as well as increased its knowledge of credit risk and portfolio management. This experience helped it gain skills that will benefit the unit in the future when it serves as the PIU for the Bank-financed Payments and Banking System Modernization Project and the PHRD grant available for the preparation of this project. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: Several external factors had a negative impact on the project and contributed to the projects poor performance. The main factors were: (i) The Russian financial crisis in The Russian financial crisis had a substantial adverse effect on macroeconomic performance in the Kyrgyz Republic. Growing external and fiscal imbalances led to a rapid depreciation of the national currency. This depreciation increased the cost of capital for both enterprises and banks. As a result, it became difficult for some banking system customers to pay back loans, especially those denominated in dollars, which accounted for about 40 percent of the banking system s assets. The share of non-performing loans in the banking system s portfolio increased significantly and banks became reluctant to lend in foreign currency. (ii) Weakness of the commercial banking system. The weakness in the banking sector was a main reason for the original project design which bypassed the local banking system and relied on an international bank to manage the FERD. It was also reflected in the fact that, following the restructuring in 1998, nine banks applied to participate in the credit line, however, only four met the eligibility criteria and only two ended up participating in the end. The small number of PCBs limited the potential impact of the project on the overall banking system and in terms of volume of sub-loans delivered to the private sector. (iii) The poor performance of the international contractor managing the FERD and the collapse of the contractor's local counterpart. The original project design relied on an international bank to serve as the Fund Manager and manage both the lending and the enterprise technical assistance components of the project. It also required the international bank to enter into a twinning arrangement with a local counterpart bank. This arrangement turned out to be unsuccessful due to the very poor performance of the international bank and the collapse of the local counterpart, the Orient Bank. The NBKR withdrew the license of Orient Bank and appointed an administrator to resolve conflicts between management and shareholders. This move undermined the basic framework of the project and contributed to the need for - 9 -

13 its restructuring. 5.2 Factors generally subject to government control: The improved stability of the economy during the latter part of the project ( ) had a substantial positive effect on the project by increasing the demand for credit, though at a much lower level than anticipated. The government's economic reform efforts most likely contributed to this result. The latest Country Assistance Strategy has shown that the Kyrgyz Republic has made considerable progress in restoring macroeconomic stability since the Russian financial crisis of During , real GDP grew at about five percent on an average annual basis. Consumer prices have declined from nearly 39 (1999) percent to only about 3.7 percent in 2001 and the foreign exchange rate remained relatively stable. The government's efforts to strengthen the commercial banking sector with support of the Financial Sector Structural Adjustment Credit (FINSAC, ) and the FINSAC Technical Assistance, also had a positive effect on the project. The FINSAC was a comprehensive reform program aimed at improving the efficiency and profitability of the financial system. The strengthening of commercial banking sector allowed the project to use commercial banks to channel the PESP funds following the restructuring in The persistence of a poor business environment in the country appears to have had a negative impact on the project. While the government made some progress in this area, it did not go as far as it could have in eliminating legal and administrative barriers for micro and small enterprises. By joining the World Trade Organization (WTO) in 1998, the Kyrgyz Republic liberalized foreign trade and investment and streamlined registration procedures for micro and small enterprises. However, such enterprises still faced the constraints of excessive administrative barriers and bureaucracy as well as limited access to finance and trade facilities. The excessive bureaucracy continues to increase transaction costs, push some businesses into the shadow economy, and create conditions for corruption. This is likely to become a priority area of reform in the future. 5.3 Factors generally subject to implementing agency control: The main factors subject to the control of the project implementing agencies were the management of the project, and the use of technical assistance, project promotion and project monitoring. During the first phase of the project from 1996 to 1998, there was a lack of effective unified project management. The FERD was responsible for generating marketable projects and assisting enterprises in developing them. The PAU was in charge of monitoring the performance of the Fund Manager and for managing the project s financial sector studies. However, project records indicate that the PAU consistently and unduly interfered in the activities of FERD to an extent that threatened the project s viability. Furthermore, the director of the PAU was also the head of the KPPA, which created a potential conflict of interest since KPPA was competing with the FERD in providing technical assistance to enterprises. During the second phase of the project, the Apex Unit began to supervise the credit line. As a result, the promotion of PESP improved. Staff from the unit prepared and distributed various promotional materials in various regions of the country and established a comprehensive internet site. All of these new implementation measures had at least a modest impact on the increase in lending activity from 1999 until the end of the project

14 5.4 Costs and financing: The original project cost was US$20.3 million. This included US$15.0 million from the IDA credit, US$2.75 million from the Japanese PHRD grants, and US$2.5 million in local co-financing from sub-borrowers, commercial banks and Orient Bank. Of the US$20.3 million, 80 percent was for the credit line component and the remaining 20 percent was for the technical assistance component. The estimated actual cost of the project is US$7.3 million or about 35 percent of the original project costs. A total of US$4.6 million was disbursed from the IDA credit line and US$2.75 million was from the two Japanese PHRD grants. In June 2002, US$7.14 million was cancelled from the IDA credit line and the remaining undisbursed balance of US$1.85 was cancelled at closing. Project costs were overestimated at both appraisal and during the 1998 restructuring. This was compounded by weaknesses in project implementation arrangements and other external factors that reduced demand. Only US$4.6 million (30 percent of the IDA credit) was disbursed to fund sub-loans, the PAU, the FERD and project audits. None of the co-financing from local sources materialized. There were no disbursements from the IDA credit for the technical assistance component, as all technical assistance activities were funded by the Japanese PHRD grants. 6. Sustainability 6.1 Rationale for sustainability rating: Despite the unsatisfactory rating for the project outcome, the sustainability is likely of the project's main achievements--strengthening the PCBs and supporting small and micro-enterprises following the 1998 restructuring. This rating is primarily due to the fact that external financing will continue to be available to the participating commercial banks for small and micro lending under the Kyrgyz Micro and Small Enterprise Finance Facility (see section 6.2 below for more details). In addition, the staff of the participating commercial banks that were trained under the project have continued in their roles even after the project has finished, and the banks have continued to lend using the lending methodologies developed under the PESP. 6.2 Transition arrangement to regular operations: Follow-Up Projects to Support Enterprise and Financial Sector Development: The Bank will not to support a PESP follow-up project as was originally envisioned. This is due in large part to the fact that external financing will continue to be available to private commercial banks through the US$15.3 million Kyrgyz Micro and Small Enterprise Finance Facility set up by the EBRD, IFC, USAID, and the Government of Switzerland. The facility aims to provide small and medium enterprises with a long-term source of financing through commercial banks. The credit line builds directly upon the recent achievements of the PESP and works through two of the PCBs (among others) and uses the same lending methodologies. The credit line also provides grant funding for technical assistance, mainly to train credit officers on credit appraisal and risk management using best practices. The facility focuses on providing loans of less then US$50,000 with a maximum term of three years. The Bank is supporting two other projects that will also help to address a number of issues that were identified under the PESP as weaknesses in the financial and private sectors: (i) the Payments and Banking System Modernization Project (approved in March 2004) which provides technical assistance to the

15 Government in support of priority reforms to improve financial sector infrastructure, and (ii) the Enhancement of Business Environment Project, which is currently under preparation. This project will help improve the business climate and competitiveness through eliminating administrative barriers to businesses and improving the technical regulatory framework and quality standards. 7. Bank and Borrower Performance Bank 7.1 Lending: During project preparation and appraisal, there were a number of positive aspects of the Bank s performance including: (i) the identification of the need to mobilize financing for a fledgling private sector; and (ii) the integration of the project into a comprehensive economic rehabilitation and adjustment support framework. Furthermore, the Bank incorporated reasonable lessons focused on the need for macroeconomic stability and strong government commitment to reform. The staffing of Bank preparation and appraisal missions was appropriate, consisting of specialists in economics, finance, investment analysis, and business management. However, the negative impact of design deficiencies was substantial, resulting in a complete restructuring of the project two years after it became effective. The main reasons for this were attributable to: (i) the lack of a realistic market study at appraisal, which resulted in a large overestimation of size of the credit line relative to country conditions; (ii) the reliance on a new, state-owned fund as the main vehicle for credit delivery; (iii) the unclear relationship between the FERD and the local banks; and (iv) the overlapping roles and responsibilities of the PAU and the FERD, which resulted in ambiguous reporting relationships and became a continuing source of conflict. These shortcomings could have been avoided. Therefore, on balance, the Bank's lending performance is rated as unsatisfactory. 7.2 Supervision: While the Bank's performance during supervision varied substantially throughout implementation, on balance, project supervision is rated as unsatisfactory. There were several positive aspects of the supervision during the last three years of operation. The first was the Bank s flexibility in responding to the government s request to keep the line of credit open and focus on lending to micro enterprises. The second was the cautious approach to appraisal of sub-projects and the qualification of participating banks, which resulted in high quality of the portfolio and PCBs. The third was the decision to prepare an intensive-learning ICR to gain a deeper insight in the project s implementation experience and help define a strategic future direction for supporting private enterprise in the Kyrgyz Republic. Unfortunately, these positive elements came late in the project cycle and there were several negative elements that overshadowed the overall supervision performance: (i) the US$12 million lending component under the restructured project was unrealistically large; (ii) the Bank s satisfactory project implementation ratings between 1998 and 2001, which did not accurately reflect the low level of activity of the lending component; (iii) the frequent turnover of task managers--seven during the eight-year life of the project--resulting in a lack of continuity and historical memory; and (iv) the lack of consistent monitoring of agreed performance indicators in project status reports. 7.3 Overall Bank performance: The overall performance of the Bank is rated as unsatisfactory

16 Borrower 7.4 Preparation: The Borrower s performance during project preparation is rated as satisfactory. The government was effective in establishing a coherent and comprehensive economic reform program, focusing on privatization, stabilization and structural adjustment ahead of other former Soviet Union countries. These actions helped to create a supportive operating environment for PESP. 7.5 Government implementation performance: The government's main role was to support a stable macroeconomic environment for private sector development. The most recent CAS indicates that the Kyrgyz Republic made good progress in liberalizing prices and exchange rates as well as initiating difficult reforms. As a result, the country is beginning to see a reduction in poverty rates. Performance audits of the two Bank-financed projects (the Rehabilitation Credit and the PESAC) found that government made progress in establishing an effective policy environment for the implementation of PESP and improving the legal framework for support private sector development. Regarding the PESP, the Government took a proactive stance in trying to improve the project s performance, especially supporting the re-focusing of the project on small enterprise lending through commercial banks. It maintained a reasonable level of commitment to the project throughout implementation. As a result, the Government s implementation performance is rated as satisfactory. 7.6 Implementing Agency: Implementing agency performance is rated unsatisfactory as a result of the unsatisfactory performance of the original implementation agencies--the FERD and the PAU. During the first two years of the operation of these two institutions, the project made negligible progress toward meeting its objectives. Project files indicate that management of the FERD by the international bank was unsatisfactory. The records also indicate that the PAU unduly interfered in the operations of the FERD, mainly in the evaluation of sub-loans, resulting in a number of conflicts that hindered progress. The restructuring of PESP in 1998 clarified the project s implementation structure and the performance of both the NBKR and the PIU (formerly PAU) of Goskominvest was satisfactory. These institutions were more effective in managing both the line of credit and technical assistance components. Project status reports consistently noted improvements of the Apex Unit in screening business plans of prospective sub-borrowers. The main deficiency was identified in the partial implementation of the amendment of 2002, and especially the provision to allow the Ministry of Finance to bear the foreign exchange risk. This depended on external factors, such as the required and lengthy process of ratification by the Parliament of this specific provision of the 2002 amendment. 7.7 Overall Borrower performance: On balance, the overall performance of the Borrower is rated as satisfactory. 8. Lessons Learned Lessons of Broad Applicability: (i) Demand surveys are critical elements of project preparation. The demand survey for the PESP resulted in an overestimation of credit demand and did not go far enough in identifying demand for specific lending products. Evaluations of demand for enterprise finance should rely not only on surveys

17 and technical studies of specific enterprises, but also on a realistic market studies, the existing financial infrastructure, macroeconomic development prospects, and the overall size of the credit market. (ii) Private enterprises will not pay for technical assistance when there is a perception that donor funds are also available for such support. The PESP required partial payment from enterprises for technical assistance. However, this component was not successful as enterprises, generally were not interested such assistance, perceiving the support either as a waste or available free of charge through grant financing. In the end, the project did not disburse any funds for this component. (iii) Implementation arrangements should be straightforward and lines of authority clear. It is important to make sure that the implementation framework is clear from the outset of the project and that lines of authority are unambiguous. Because the lines of authority and limit of roles were not fully clear under the project, the PAU often interfered with the day-to-day operations of the FERD, including the credit review process. (iv) Line of credit projects can build in mechanisms to support business environment policy reforms. While the government has taken a number of important steps to strengthen the business environment, PESP enterprise surveys indicated a number of perceived deficiencies in the business environment continue to adversely affect private enterprise development. These include administrative procedures governing businesses; taxation policy; corruption and crime; and adequacy of the legal framework. In the future, it would be a useful if the project could identify mechanism to provide feedback to the government on many of these issues to help support the policy dialogue on business environment. (v) The Bank should minimize changes in task managers to ensure continuity in the project. PESP had seven different task managers over an eight-year period. Such changes substantially disrupted project supervision, reduced institutional memory, and led to frequent shifts in approaches and strategies. This was a problem particularly for a project with a history of implementation problems. (vi) Adequate attention needs to be given to strengthening the local auditing profession in underdeveloped financial markets. Auditing remains a major issues for SME support projects in underdeveloped financial and banking sectors such as the Kyrgyz Republic. In the PESP, several banks were disqualified from the project due to the lack of an audit by an auditing firm, satisfactory to the Bank. None of the local auditors met the Bank-approval standard, and most banks could not afford an international audit. Key Lessons from the Intensive Learning Survey and Workshop (i) Project preparation should include detailed demand surveys to provide insight into the demand for specific financial products and services. The size of the project and the features of the sub-loans should be determined as accurately as possible, through a consultation process involving surveys and focus groups with entrepreneurs and practitioners during project preparation. The Intensive Learning Survey indicated that enterprises were particularly interested in the following products: leasing (94 percent); capital financing (78 percent); mortgages (67 percent); and insurance products (61 percent). None of these products were offered through the project, however, this was in part due to the rudimentary development of the financial sector at the time of project preparation. (ii) Project design should be as flexible as possible so that it can be adapted to changes in the enabling environment during implementation. In this regard, surveys and focus groups could provide