Uzbekistan: Second Small and Microfinance Development Project

Transcription

1 Completion Report Project Number: Loan Number: 2634 September 2015 Uzbekistan: Second Small and Microfinance Development Project This document is being disclosed to the public in accordance with ADB's Public Communications Policy 2011.

2 CURRENCY EQUIVALENTS Currency Unit sum At Appraisal At Project Completion (as of 1 March 2010) (as of 28 August 2015) 1.00 = $ $ $1.00 = SUM1, SUM2, ABBREVIATIONS ADB Asian Development Bank AFS audited financial statement APFS audited project financial statement BOTR back-to-office report CBU Central Bank of Uzbekistan DMF design and monitoring framework FIL financial intermediation loan GAP gender action plan MCO microcredit organization MOF Ministry of Finance MSE micro and small enterprise PAM project administration manual PCB participating commercial bank PIU project implementation unit SLA subsidiary loan agreement SMDP Small and Microfinance Development Project TA technical assistance NOTES (i) (ii) The fiscal year (FY) of the government, its agencies, and the participating banks ends on 31 December. FY before a calendar year denotes the year in which the fiscal year ends, e.g., FY2009 ends on 31 December In this report, "$" refers to US dollars. Vice-President W. Zhang, (Operations 1) Director General S. O Sullivan, Central and West Asia Department (CWRD) Director B. Wilkinson, Public Management, Financial Sector, and Trade Division, CWRD Team leader Team member L. Schou-Zibell, Principal Economist, CWRD F. Teves, Project Analyst, CWRD M.C. Yabut, Operations Assistant, CWRD In preparing any country program or strategy, financing any project, or by making any designation of or reference to a particular territory or geographic area in this document, the Asian Development Bank does not intend to make any judgments as to the legal or other status of any territory or area.

3 CONTENTS Page BASIC DATA I. PROJECT DESCRIPTION 1 II. EVALUATION OF DESIGN AND IMPLEMENTATION 2 A. Relevance of Design and Formulation 2 B. Project Outputs 3 C. Project Costs 4 D. Disbursements 4 E. Project Schedule 5 F. Implementation Arrangements 5 G. Conditions and Covenants 5 H. Related Technical Assistance 7 I. Consultant Recruitment and Procurement 8 J. Performance of the Borrower and the Executing Agency 8 K. Performance of the Asian Development Bank 9 III. EVALUATION OF PERFORMANCE 9 A. Relevance 9 B. Effectiveness in Achieving Outcome 10 C. Efficiency in Achieving Outcome and Outputs 11 D. Preliminary Assessment of Sustainability 11 E. Impact 11 IV. OVERALL ASSESSMENT AND RECOMMENDATIONS 12 A. Overall Assessment 12 B. Lessons 12 C. Recommendations 13 APPENDIXES 1. Design and Monitoring Framework Gender Action Plan: Implementation Status Status of Compliance with Project Covenants (Loan Agreement) Participating Commercial Banks Compliance with the Financial Ratios Contribution to the ADB Results Framework 41 i

4

5 BASIC DATA A. Loan Identification 1. Country 2. Loan Number 3. Project Title 4. Borrower 5. Executing Agency 6. Amount of Loan 7. Project Completion Report Number B. Loan Data 1. Appraisal Date Started Date Completed 2. Loan Negotiations Date Started Date Completed 3. Date of Board Approval 4. Date of Loan Agreement 5. Date of Loan Effectiveness In Loan Agreement Actual Number of Extensions 6. Closing Date In Loan Agreement Actual Number of Extensions 7. Terms of Loan Interest Rate Maturity Grace Period 8. Terms of Relending Interest Rate Maturity Grace Period Uzbekistan 2634 Second Small and Microfinance Development Project Republic of Uzbekistan Ministry of Finance $50,000,000 PCR:UZB January February March March April May June August 2010 none 31 December April 2014 none London Interbank Offered Rate (LIBOR)-based lending facility 20 years 5 years 0.5% 15 years 3 years

6 ii 9. Disbursements a. Dates Initial Disbursement 29 October 2010 Effective Date 3 August 2010 Final Disbursement 19 March 2014 Original Closing Date 31 December 2013 Time Interval 41 months Time Interval 41 months Subloan Open Joint Stock Commercial Bank Agrobank Open Joint Stock Commercial Bank Hamkorbank Open Joint Stock Innovation Commercial Bank Ipak Yuli b. Amount ($ 000) Original Allocation Last Revised Allocation Amount Added / (Canceled) Net Amount Available Amount Disbursed Undisbursed Balance 20,000 6,000 (14,000) 6,000 6, ,000 27,000 7,000 27,000 27, ,000 17,000 7,000 17,000 17, Local Costs (Financed) - Amount ($) Not applicable - Percent of Local Costs Not applicable - Percent of Total Cost Not applicable C. Project Data 1. Project Cost ($) Cost Appraisal Estimate Actual Foreign Exchange Cost 50,000,000 50,000,000 Local Currency Cost 92,850,000 92,850,000 Total 142,850, ,850,000

7 iii 2. Financing Plan ($) Cost Appraisal Estimate Actual Implementation Costs ADB Financed 50,000,000 50,000,000 Commercial Banks Funds 50,000,000 50,000,000 Subborrowers Equity Funds 42,850,000 42,850,000 Total $142,850,000 $142,850,000 IDC Costs Borrower Financed Not applicable Not applicable ADB Financed Not applicable Not applicable Other External Financing Not applicable Not applicable Total ADB = Asian Development Bank, IDC = interest during construction. 3. Cost Breakdown by Project Component ($) Component Appraisal Estimate Actual Loan 2634 Loans to Participating Commercial Banks 50,000,000 50,000,000 Agrobank 6,000,000 6,000,000 Hamkorbank 27,000,000 27,000,000 Ipak Yuli Bank 17,000,000 17,000,000 Participating Commercial Banks Funds 50,000,000 50,000,000 Agrobank 6,000,000 6,000,000 Hamkorbank 27,000,000 27,000,000 Ipak Yuli Bank 17,000,000 17,000,000 Subborrowers Equity Funds 42,850,000 42,850,000 Agrobank 5,142,000 5,142,000 Hamkorbank 23,139,000 23,139,000 Ipak Yuli Bank 14,569,000 14,569,000 Total 142,850, ,850, Project Schedule Item Appraisal Actual Estimate Loan 2634 October December 2010 Disbursements Not available Agrobank 8 Nov 2010 Hamkorbank 3 Nov 2010 Ipak Yuli Bank 29 Oct Dec 2010 January December 2011 Disbursements Not available Agrobank 31 Mar Jun 2011 Hamkorbank 25 Feb Dec 2011 Ipak Yuli Bank 25 Jan Dec 2011 January December 2012 Disbursements Not available Hamkorbank 23 Jan May 2012 Ipak Yuli Bank 13 Jan Jan 2012

8 iv Item Appraisal Actual Estimate January December 2013 Disbursements Not available Hamkorbank 21 Jan Oct 2013 Ipak Yuli Bank 21 Jan Oct Project Performance Report Ratings Implementation Period Development Objectives Ratings Implementation Progress From 3 August 2010 to 31 December 2010 Satisfactory Satisfactory From 1 January 2011 to 31 December 2013 On track a a From 1 January 2011 the project performance rating system was replaced with eoperations. D. Data on Asian Development Bank Missions Name of Mission Date No. of Persons No. of Person-Days Specialization of Members a Consultation Jan a Country Consultation Dec b, c, d, Loan Reconnaissance (Upgraded to Fact-finding) b 27 Jan 12 Feb b, c, e, f, g, h, i, j Inception 1 7 Jun b, e Review 19 Nov 2 Dec a Review 28 Nov 2 Dec k Midterm Review Nov k, l, m, n, o Review 9 12 Jul k a a = senior project management specialist, b = principal financial management specialist, c = young professional (economics), d = senior private sector development officer, e = assistant project analyst, f = counsel, g = finance and private sector development officer, h = economics officer, i = poverty reduction and gender specialist, j = microfinance specialist, k = senior economist, l = financial sector specialist, m = operations analyst, n = associate project analyst, o = senior social development specialist. b The Uzbekistan Resident Mission provided support to all missions in the field.

9 I. PROJECT DESCRIPTION 1. In November 2009, the Government of Uzbekistan requested that the Asian Development Bank (ADB) approve by 22 April 2010 (i) a loan of $50 million from ordinary capital resources for the Second Small and Microfinance Development Project, to be channeled through three participating commercial banks (PCBs); and (ii) an associated technical assistance (TA) grant for $600,000 to improve (a) their subborrowers' entrepreneurial skills, and (b) PCBs' micro and small enterprise (MSE) finance capacity. 1 To meet the government s request the loan processing was fast-tracked, with a loan reconnaissance mission to Uzbekistan from 27 January to 12 February 2010 (upgraded to a fact-finding mission on 5 March 2010). By April 2010 loan approval was obtained under summary procedure The project was to facilitate MSE access to credit across Uzbekistan. Its envisaged impact was increasing the number of people earning their livelihood in the formal private MSE sector. This was to be achieved through employment generation, with 30,000 new jobs to be created by the end of 2013 (with equal opportunities for women) through activities financed by the ADB project (through the provision of subloans to subborrowers) and attached TA grant (through capacity building). The project s intended outcome was expanded, viable, and sustainable MSE finance operations. Outstanding PCBs microfinance loans were expected to increase from $75 million at the end of 2009 to $360 million by the end of 2013, while keeping the PCBs delinquency and write-off ratio at no more than 5% at the end of each year with respect to their total loan portfolios. The project had two outputs: (i) (ii) Delivery of training to improve the entrepreneurial capacity of subborrowers, including women and microcredit organizations (MCOs). Of the 2,000 potential subborrowers trained, at least 1,000 should be men and 1,000 should be women, and at least 25% of the trainees should have access to micro and small business credit. An increase in microfinance loans to financially viable subborrowers (men and women). Of the 25,000 microfinance accounts to be opened by the end of 2013, at least 25% should be opened by women, and 10,000 should opened by new microfinance clients. (Appendix 1) 3 3. The project was justified based on the large unmet credit demand from MSEs. A 2013 World Bank survey indicated that only 16.1% of firms used banks to finance investment. 4 The large unmet demand was also implied by the continued demand for ADB financial intermediation loans (FILs) to Uzbekistan for micro, small, and medium-sized enterprises. 5 1 The three PCBs are Hamkorbank, Agrobank, and Ipak Yuli Bank. ADB Report and Recommendation of the President to the Board of Directors: Proposed Loan and Technical Assistance Grant to the Republic of Uzbekistan for the Second Small and Microfinance Development Project. Manila. 2 A country consultation mission fielded from 24 to 30 January 2009 explored on a preliminary basis a possible small and microfinance credit line for Uzbekistan for processing in BTOR L2634 UZB Country Consultation. 3 During the midterm review, the project team noted that several design and monitoring framework (DMF) indicators were not achievable; however, after internal discussions, the targets were not changed. BTOR L2634-UZB Midterm review (with attachments), Attachment 10, VIII. November World Bank Group Enterprise Surveys. Country Note. Uzbekistan. Firms in the Enterprise Survey are referred to small, medium and large. 5 ADB Report and Recommendation of the President to the Board of Directors: Proposed Loan and Technical Assistance Grant to the Republic of Uzbekistan for the Rural Enterprise Development Project. Manila; ADB Report and Recommendation of the President to the Board of Directors: Proposed Loan to the Republic of Uzbekistan for the Small and Medium Enterprise Development Project. Manila; ADB Report and

10 2 According to the Central Bank of Uzbekistan (CBU), the amount allocated to microcredit by commercial banks increased by about 100% in 2009 compared to 2008, and was targeted to increase by another 30% to reach $275.0 million in Based on historic trends, the MSE loan balances of the three PCBs were forecast to reach a total of $484.6 million by 2013, justifying a $50.0 million loan. The project was to be catalytic by raising an equal amount of loan funds from the PCBs, with 30% in equity being provided by the subborrowers. The project was to be demonstrational by making use of market-based principles in MSE lending, including use of market-based onlending rates, choice of currency, and the ability to draw down loans in cash (to effectively address existing difficulties in withdrawing cash from bank accounts). To improve outreach, PCBs were allowed to lend up to $20,000 equivalent to each MCO and up to $1,000 to each subborrower. To ensure wide distribution, 50% of the aggregate outstanding principal amounts lent by PCBs to subborrowers other than MCOs were to be for loan sizes up to $10,000. Lending by the three PCBs to women was to represent one-quarter to one-third of the total number of microcredit loans. TA was provided to help improve the PCBs capacities for credit appraisal of MSEs and gender sensitivity, as well as to improve the skills of subborrowers (particularly women) in preparing business plans and applying for credit from PCBs. 4. MSEs access to credit was constrained by the underdeveloped nature of the country s financial sector along with various market distortions. These included directed lending at subsidized interest rates leading to a misallocation of resources, and practical difficulties in cash withdrawals from bank accounts, leading to a premium for cash and a trust gap for banks in their ability to mobilize customer deposits. This was manifested in a low (individual and household customer) bank deposits gross domestic product ratio of 6.6% in 2009 based on CBU records. To mitigate the low deposit ratio, measures were put in place by the government to increase public confidence in the banking system and raise deposit mobilization. 6 II. EVALUATION OF DESIGN AND IMPLEMENTATION A. Relevance of Design and Formulation 5. The project remained relevant at appraisal and at completion. The project was consistent with the government s Welfare Improvement Strategy , which provided the strategic framework for the government s efforts to reduce poverty and improve welfare. 7 It supported the government s Microfinance Development Program, operational from 2007 to 2010, which sought to make microfinance a sustainable component of the country s financial sector by, among other initiatives, expanding the network and participation of nonbanks in the sector. Financial sector development and reform formed one of the key tasks of the Welfare Improvement Strategy. The project was aligned with ADB s country strategy and program, , which supported sustainable economic growth, poverty reduction, and gender and Recommendation of the President to the Board of Directors: Proposed Loan and Technical Assistance Grant to the Republic of Uzbekistan for the Small and Microfinance Development Project (SMDP). Manila. 6 The government (i) revoked the law allowing tax authorities to access customers bank deposits without a court order, (ii) ended legal restrictions on customers cash withdrawals from their bank accounts, (iii) introduced a deposit insurance scheme, and (iv) supported banks increased capitalization to improve their solvency. These measures, complemented by the banks own deposit mobilization efforts, measures that contributed to the consistent increase in (individual and household) bank deposits gross domestic product ratio from 0.9% (SUM33.6 billion) in 2000 to 6.6% (SUM2.9 trillion) in 2009 an average of 0.6 percentage points per year. Footnote 1 (Macroeconomic and Sectoral Analysis, accessible from the list of linked documents in Appendix 2). 7 International Monetary Fund Republic of Uzbekistan: Poverty Reduction Strategy Paper. Washington, DC; ADB. Financial Sector Development in Central and West Asia. Consultant s report (Uzbekistan Country Assessment, Sector Assessment [Summary]: Micro, Small, and Medium-size Enterprise Development). Manila (TA 7709-REG). Unpublished.

11 3 development. 8 The project was also aligned with ADB s microfinance development strategy. 9 At completion the project was aligned with ADBs Country Partnership Strategy , which supports access to credit for SMEs and for microfinance beneficiaries in rural and urban areas through new and ongoing assistance. 6. The project (i) encouraged forward and backward linkages in the agricultural sector, as mentioned in the country strategy and program ; (ii) supported the development of rural financial services; (iii) promoted more efficient intermediation functions by commercial banks; and (iv) took into account lessons learned from previous ADB FILs to PCBs in Uzbekistan, which had underestimated the demand for micro and small business credit. During design the ADB mission consulted stakeholders comprising MSEs, nongovernment organizations, women s organizations, MCOs, credit unions, commercial banks, the CBU, and other development partners. At project completion, the project remained consistent with government programs aimed at developing MSEs and with ADB s country strategy and program More time for project preparation could have contributed to better project design and formulation. Issues that could have been resolved during project appraisal included: (i) restrictions on banks preventing provision of dollar-denominated subloans to be used in subborrowers credit cards made for use outside Uzbekistan; (ii) requirements on banks to deposit their daily cash balances with the CBU (as part of its monetary policy), which limited the ability of banks to make sum-denominated subloans in cash to their subborrowers, as envisaged by the project; and (iii) CBU requirements for PCBs to present documents on every subloan (rather than on a batch of subloans) to be able to withdraw from the PCBs imprest accounts. Had these issues been resolved during project appraisal, the initial delay in declaring loan effectiveness and initial slow loan disbursement could have been avoided. 8. At project appraisal, Hamkorbank was allocated $20 million, Agrobank $20 million, and Ipak Yuli Bank $10 million. However, due to Agrobank s breach of financial covenants in 2011, the remaining unused allocation to Agrobank of $14 million was reallocated to Ipak Yuli Bank and Hamkorbank in equal amounts of $7 million. At project completion the final allocation was $27 million to Hamkorbank, $17 million to Ipak Yuli Bank, and $6 million to Agrobank. More time for project preparation could have contributed to a more diligent review of potential PCBs and ex-ante identification of weaknesses in Agrobank. B. Project Outputs 9. Under output 1, 11,254 potential subborrowers were trained in business planning and management, of which 7,500, or 67%, accessed microfinance loans under the project, exceeding the 25% target. Of these trainees 4,682 were women and 2,818 were men, exceeding the targets of 1,000 women and 1,000 men trained. Specific training on business plan and credit application was provided to women borrowers. The curriculum covered the business plan, marketing, production, inventory control, financial statements, and ratio analysis. A small handbook describing requirements for business establishment, registration, and implementation was produced and distributed to the trainees. 10. Under output 2, a total of 819 microcredit loan officers, or 86% of the total number of microcredit loan officers, were trained in microcredit appraisal, exceeding the design and 8 ADB Country Strategy and Program: Uzbekistan, Manila. 9 ADB Finance for the Poor: Microfinance Development Strategy. Manila.

12 4 monitoring framework (DMF) target of 30%. However, 18% of the trained microcredit loan officers were women, which was below the gender action plan (GAP) target of at least 25% women microcredit loan officers trained. Training focused on (i) financial and nonfinancial analysis, and (ii) risk assessment in lending to MSEs. This output resulted in (i) the training of 17 PCB trainers in capacity building for gender mainstreaming, (ii) the establishment by Hamkorbank and Ipak Yuli Bank of gender focal points to support gender-sensitizing capacity, and (iii) the training of 113 male and 10 female PCB staff members in gender and development mainstreaming. By completion 18,646 microfinance loan accounts were opened (short of the targeted 25,000 accounts), of which 37% were opened by women (exceeding the 25% target). The loan was fully disbursed within the implementation period, thus meeting the target of $50 million disbursed by end It was noted during the midterm review that several targets were not likely to be attained (number of jobs created and outstanding microfinance loans), although these indicators were not changed. The midterm review mission could not establish the basis for the DMF targets. 10 The deviations from targets (paras. 9 and 10) are not considered to have affected project costs, time schedules, or expected benefits. C. Project Costs 12. The project cost of $ million comprises the ADB loan of $50 million, the counterpart funds from the PCBs that match the ADB loan of $50 million, and the 30% equity contribution by subborrowers. 11 At appraisal, up to $20 million of the ADB loan was allocated to Hamkorbank, up to $20 million was allocated to Agrobank, and up to $10 million was allocated to Ipak Yuli Bank. D. Disbursements 13. Loan proceeds were withdrawn in accordance with ADB s Loan Disbursement Handbook. The disbursements for the loan were through imprest accounts under the statement of expenditure procedure. The imprest fund was highly useful in project implementation and was effectively utilized, with statements of expenditures regularly submitted by the PCBs to ensure adequate fund flow. Loan disbursement was slow at the start, prompting ADB intervention, especially with regard to subloans issued by Agrobank. 12 While the appraisal disbursement schedule was realistic, it was not anticipated that there would be (i) a suspension of disbursement of funds to Agrobank in September 2011, and (ii) a delay in effecting the reallocation of the unused allocation of $14 million from Agrobank to Hamkorbank ($7 million) and to Ipak Yuli Bank ($7 million) by the Ministry of Finance (MOF) in July Nonetheless, the full amount of $50 million was disbursed and liquidated within the project implementation period without the need to extend the loan. Initial disbursements were made to Ipak Yuli Bank on 29 October 2010, to Hamkorbank on 3 November 2010, and to Agrobank on 8 November 2010; final disbursements were made to Agrobank on 29 June 2011, and to Hamkorbank and Ipak Yuli Bank on 31 October A summary of the project s appraisal and actual financing plans and costs are provided in the Basic Data section of this report. The time interval between the initial and final disbursements was 41 months, the same time interval between the actual effective date and original loan closing date. 10 Project team leadership was changed in November Schedule three, para 8 of the loan agreement requires the subborrower to meet and maintain a debt to equity ratio of 70:30 for the project, which the subloan is financing. 12 A loan review mission was fielded in December 2010 at which time loan disbursement was reviewed.

13 5 E. Project Schedule 14. The project was approved on 21 April 2010 under summary procedure. The loan agreement was signed on 1 May The loan became effective on 3 August The original schedule for loan effectiveness was 15 June 2010; the delay was caused by the drafting and signing of the subsidiary loan agreements (SLAs) between the government and the three PCBs, and the obtaining of legal opinions on the loan, which were to be provided ADB by the executing agency. At appraisal, the project completion date was 30 June 2013, and the loan closing date was 31 December The actual project completion date was 31 December 2013, and the loan was closed on 28 April F. Implementation Arrangements 15. Implementation arrangements to deliver project outputs and achieve the project purpose were adequate. The MOF was the executing agency for the project. The loan was onlent by the government to three PCBs under subsidiary loan agreements. Subloans by the PCBs to subborrowers were covered under separate project agreements. Each PCB established project implementation units (PIUs) for this purpose. A project administration manual (PAM) was prepared detailing arrangements for procurement, disbursement, accounting, audit, monitoring, and evaluation, including consulting services for the TA. 16. A minor change in implementation arrangement for the loan was approved on 25 January 2013, allowing the reallocation of the $14 million unused allocation by Agrobank to Hamkorbank ($7 million) and Ipak Yuli Bank ($7 million) (para. 31). The minor change in implementation arrangements also increased the subloan ceiling from $20,000 to $40,000. Agrobank disbursed $6 million of its original allocation of $20 million. G. Conditions and Covenants 17. The PCBs sum-denominated subloan interest rates ranged from around 16% to 36% per year depending on whether the loan was in cash or noncash, the loan maturity, and the use of funds. The interest rates included the weighted average cost of capital and a risk premium that reflected the average annual rate of currency depreciation (10%). The CBU refinancing rate was 12.0%, and the inflation rate for 2012 was 12.9% according to International Monetary Fund estimates, and about 7.0% during the same period according to the CBU methodology. 18. At loan processing the subloan limit was set at $20,000. In response to requests made by the PCBs during the midterm review mission in November 2012, the subloan limit was raised from $20,000 to $40,000. This was in line with the law on microfinance that limits the maximum size of microcredit to less than or equal to 1,000 times the minimum wage (SUM79,590 in November 2012). The higher subloan limit was also in line with market demand for a higher limit, as it was viewed to encourage business growth, development, expansion, and competition. The project agreement required that at least 50% of all subloans made by each PCB should be for loan amounts not exceeding $10,000 equivalent each, with the exception of subloans to MCOs, which could be for up to $20,000 each. Based on a screening of all subloans issued during the project implementation period, 67% were for $1 to $10,000, and 33% were above $10, The PCBs did not extend subloans to MCOs. MCOs are allowed to receive external funds from donors, but the process is cumbersome due to conflicting regulations. In the typical case of a donor-funded onlending program in which commercial banks receive funds for onlending to either MSE clients or MCOs directly banks are not allowed to lend without a

14 6 purpose, which leads to direct loans to the clients of the MCOs, cutting the MCOs out. There is no possibility to extend a standing credit line to MCOs, which they can use at their own risk for their clients. Accepted and approved practice is that when MCOs borrow from banks and onlend to their clients, banks must open separate loan accounts for each client of a MCO and transfer the amount to their accounts. 20. Of Hamkorbank s and Ipak Yuli Bank s total outstanding loan portfolio, subloans were distributed as follows: trade, 80%; manufacturing, 10%; services, 6%; agriculture, 3%; and other, 1%. Of their total outstanding loan portfolio, 36% were corporate loans, 28% were SME loans, 21% were microfinance loans, and 15% were individual loans. 21. As set out in the project agreement, for subloans the following criteria were to be maintained by qualified subborrowers: (i) a debt equity ratio not higher than 70:30 for the subproject that the subloan is financing, (ii) a debt service coverage ratio not lower than 1.25 times, and (iii) a proven track record (i.e., no past defaults). Subloan documentation was reviewed during project review missions to ensure that qualified subborrowers met these criteria. The results indicated widespread compliance, with one exception. A review mission in July 2012 established that not all subloans issued by Ipak Yuli Bank were in compliance with subloan requirements. As a result, a full compliance review was undertaken for all subloans issued by Ipak Yuli Bank from 10 May 2013 to 10 July The review confirmed that 97.5% of the subloans were in compliance. Those that did not meet the criteria were not credited by ADB. 22. Subloans were issued in domestic currency due to restrictions as set out in para Financial covenants were achieved by Hamkorbank and Ipak Yuli Bank throughout the project implementation period (Appendix 4). Each PCB was to maintain a capital adequacy ratio of not less than 12%, a return on assets of not less than 1%, a loan deposit ratio of less than 100%, an intermediation cost ratio of less than 15%, and a loan delinquency and write-off ratio of no more than 5% of its total loan portfolio. 13 Disbursement of funds to Agrobank was suspended by ADB in September 2011 after Agrobank breached its loan covenants by (i) failing to meet the return on assets ratio of 1% as stipulated in the loan and project agreements for 2010 and 2011, and (ii) failing to provide either its 2010 audited financial statement (AFS) within the stipulated period or an auditor s report on the use of loan proceeds. 14 In addition, the misappropriation of funds by former employees of Agrobank from February 2009 to February 2010, which was reflected in the 2010 and 2011 audit, necessitated a restatement of Agrobank s 2009 AFS. The misappropriated funds, in a departure from International Financial Reporting Standards, were reported as receivables from former employees rather than being written off the books. Agrobank is majority-owned by the government, with the MOF holding about 25.1% of its shares as of Qualified subprojects for which the PCBs provided subloans to qualified subborrowers were for activities that (i) were excluded from the list of prohibited investment activities, and (ii) 13 Definitions of requirements: capital adequacy ratio: ratio of regulatory capital to risk assets as a percentage; return on assets: net profit divided by the sum of total assets in the year at issue and total assets at the end of the previous year; loan deposit ratio: total net loans to customers divided by customer deposits; intermediation cost ratio: noninterest expenses divided by average total assets; loan delinquency and write-off ratio: aggregate of loans classified as a loss and loans written in a year as a percentage of total loan portfolio. 14 In a letter dated 8 September 2011, ADB informed Agrobank it would cease disbursing funds to Agrobank due to Agrobank s failure to submit (i) its 2010 AFS within the prescribed deadline, and (ii) quarterly progress reports in The 2010 AFS, which was subsequently submitted to ADB on 21 December 2011, showed noncompliance with the financial covenant that the return on assets be no less than 1%. As a result, ADB discontinued the allocation of the remaining $14 million to Agrobank.

15 7 had minimal or no adverse environmental or social risks. The PCBs and their subborrowers adhered to these requirements. 25. The project was classified as category C for the environment, involuntary resettlement, and indigenous peoples. As such, the PCBs were not required to implement an environmental and social management system that complies with ADB s Safeguard Policy Statement (2009). ADB staff and a consultant reviewed the current status of the existing environmental and social management system and provided advisory support to PCBs where gaps were identified. Hamkorbank and Ipak Yuli Bank have their own environmental and social management systems that (i) contain requirements on environmental screening, conducting environmental and social due diligence, and environmental impact assessment and (ii) include environmental monitoring and reporting procedures. The banks have designated safeguards officers who are responsible for environmental policy implementation. Responsible environmental and social management was confirmed during review missions and on site visits to subborrowers. 26. Ipak Yuli Bank and Hamkorbank (i) prepared their financial statements (balance sheet, statement of income and expenses, and related statements) in accordance with International Financial Reporting Standards; (ii) each have an internal audit unit with an independent reporting structure; (iii) arranged for their financial statements to be audited by reputable, internationally affiliated audit firms; and (iv) have formal risk management units. 27. Ipak Yuli Bank and Hamkorbank were delayed by an average of 16.9 months in submitting their annual audited project financial statements (APFSs) and related audit reports. In 2013, both PCBs submitted their consolidated APFSs for FY2010 FY2012, including audit opinion on use of loan proceeds and compliance with financial covenants. The delay in submitting APFS was because the project agreement was not clear on reporting requirements, and it did not make specific reference to APFSs. Both Ipak Yuli Bank and Hamkorbank were required to submit consolidated APFSs, while Agrobank has not provided any APFSs. The project team encountered difficulties in providing the PCBs with clear guidance on reporting requirements related to APFSs. This issue is being resolved internally with more clarity on project agreements with PCBs, loan agreements, and ADB requirements for audit terms of reference. 28. Ipak Yuli Bank and Hamkorbank submitted AFSs on a timely basis. Agrobank was informed that ADB would stop disbursing funds since Agrobank had failed to submit its 2010 AFS within the prescribed deadline. The 2010 AFS was submitted to ADB on 21 December 2011, at which time ADB discontinued the allocation of the remaining $14 million to Agrobank, as its AFS showed noncompliance with the financial covenant to maintain a return on assets of not less than 1%. H. Related Technical Assistance 29. The TA was rated successful. 15 The TA of $615,000 (with $15,000 contributed by the government in kind) was attached to the loan for capacity development to strengthen (a) their subborrowers' entrepreneurial skills, and (b) PCBs' micro and small enterprise (MSE) finance capacity. The TA became effective on 17 May In November 2011, a minor change in implementation arrangements for the TA was made, extending it for another year to continue to help PCB staff strengthen their capacity in credit appraisal; accounting reporting and standards; 15 ADB Technical Assistance Completion Report: Capacity Building for Microfinance Development in Uzbekistan. Manila (TA 7523-UZB).

16 8 identify key constraints to the development of women MSEs, products available for women entrepreneurs; improve production and dissemination of public information materials on women s entrepreneurship; and review progress made in the implementation of the gender action plan (GAP). This was made possible through available uncommitted funds of $245,000. Accordingly, the TA completion date was moved from 30 June 2011 to 30 June Of the $600,000 allocated for the TA grant, $456, was disbursed, leaving 23.9% of TA funds unutilized. TA resources were used efficiently and at lower cost than originally anticipated. TA targets were achieved or exceeded. I. Consultant Recruitment and Procurement 30. To undertake the TA, the TA envisaged recruitment of a consulting firm comprising a team leader with expertise in small and microfinance enterprise credit, a micro and small enterprise credit and financial reporting specialist, and a gender specialist. Maxwell Stamp was selected as the consulting firm. In addition, six individual consultants were also engaged. They reviewed and provided recommendations on: (i) current conditions relating to bank lending to MSEs; (ii) key constraints to the development of women MSEs, (iii) products available for women entrepreneurs; (iv) production and dissemination of public information materials on women s entrepreneurship; and (v) progress made in the implementation of the gender action plan (GAP). Additional training was provided in credit appraisal skills and gender sensitization of PCB staff, and the GAP was further developed. J. Performance of the Borrower and the Executing Agency 31. The performance of the executing agency for the loan is rated partly satisfactory. The project was delayed at the start. Loan effectiveness could not be declared, as the SLAs and legal opinions, as well as the notices of PIU establishment by Agrobank and Hamkorbank, were not ready. Loan disbursement was initially slow, necessitating an ADB mission to follow up in December The delay was traced to inconsistencies in CBU restrictions and the project design (para. 7). The November 2011 review mission established that there was no CBU policy or procedure allowing the use of cash to pay for US dollar-denominated subloans; as a result this was not allowed. The project was designed to ensure that subborrowers might have access to foreign currency loans to be able to purchase equipment and materials outside Uzbekistan as a means of raising productivity and employment for the benefit of Uzbek consumers. As part of project design, the ADB loan agreement specified that the subloan carry interest at an appropriate rate that takes into account the cost of funds, administrative costs and credit, and foreign exchange risks. 32. In June 2012, ADB recommended that MOF reallocate Agrobank s unused allocation to the other PCBs. The reallocation was effected in July 2013 when the MOF countersigned the SLAs, providing Ipak Yuli Bank and Hamkorbank with only 5 months to fully disburse their new allocations of $7 million each. The MOF did not provide ADB with a project completion report. Following a review mission during October 2014, the project team requested a project completion report to be provided. This was documented in an aide memoire dated 28 November The performance of the executing agency for the TA is rated satisfactory. Facilities provided were good, and the PCBs were very obliging in meeting with the TA team, allowing access to their branches and personnel, and explaining their methods and tools. They also involved the TA team in resolving their problems by asking for advice and possible solutions. Attendance at trainings was good, but it would have been preferable if the venues had been far

17 9 from the participants workplaces due to frequent disruptions arising from business calls and work exigencies to attend other matters. 16 K. Performance of the Asian Development Bank 34. The performance of ADB is rated satisfactory. There were nine missions carried out throughout the project period. The timely interventions by ADB helped sustain the project at its various stages: (i) at inception to meet loan effectiveness conditions, (ii) when disbursements lagged behind schedule, (iii) during the suspension of disbursement for the Agrobank loan and its prompt reallocation to the other PCBs, and (iv) during the raising of the subloan limit from $20,000 to $40,000 to encourage expansion of lending into more business areas and to better meet higher demand by the PCBs For the TA, ADB performance is rated satisfactory. Project review missions were fielded in December 2010 when TA consultants were guided regarding their terms of reference, and in November 2011 when the TA implementation period was extended after review. A midterm review mission was fielded in November 2012, which included a review of performance targets of the DMF and the GAP. Field visits were conducted during which subborrowers were interviewed for feedback regarding their training and other services provided by the PCBs. III. EVALUATION OF PERFORMANCE A. Relevance 36. The project design was relevant. It was consistent with the country s development priorities as enunciated in its Welfare Improvement Strategy, , which outlined pillars of its economic growth strategy, including financial sector development and reform; and the acceleration of the development of the private sector, especially small and medium-sized enterprises (SMEs) and microfinance. The project was aligned with ADB s country strategy and program Small and micro-businesses was one of the sectors selected under the updated country operations business plan The project was justified on the bases of market and coordination failures, and the filling in of market gaps. Access to finance for MSEs has been constrained by an underdeveloped financial sector and policy distortions that abound in the sector. Consequently, there is large unmet demand for credit from MSEs based on CBU forecast indicators as shown in the report and recommendation of the President. 19 The project was also justified on the basis of employment generation and its contribution to expanded, viable, and sustainable MSE finance operations. 38. The country strategy and program, for Uzbekistan stated that the Operations Evaluation Department found that credit lines were provided before ensuring that conditions were in place to allow credit to work effectively. 20 The validation report for the Small and Microfinance Development Project (SMDP) observed that heavy reliance on the use of credit 16 Consultant Report Technical Assistance 7523-UZB Final Report. Maxwell Stamp Ltd. June. Bangladesh. 17 Inception Mission, 14 June 2010, BTOR L2634 UZB Loan; Review Mission, 10 December 2010, BTOR L2634 UZB Loan; Review Mission,12 January 2012, BTOR L2634 UZB Loan. 18 ADB Country Operations Business Plan: Uzbekistan, Manila. pp. 19 Footnote 1, Para. 4 in the report and recommendation of the President discuss the unmet demand for small and microcredit. 20 Footnote 8, para 58.

18 10 lines through commercial banks was inadequate in meeting the intended long-term development needs and constraints of the microfinance sector. 21 The intended impact and outcome of the project were therefore incongruent with the policy gaps in the financial and MSE sectors. The Presidential Resolution No. 1438, issued in 2010, consisted of an action plan focused on (i) priority areas for further reforms, (ii) sustained improvement of the country s finance and banking system, and (iii) measures aimed at development of the microfinance and nonbank financial sectors and to achieve high ratings by international rating agencies. 39. Presidential Decree No was issued in 2011 to further strengthen MSME development. The measures outlined in the decree aim to improve the enabling environment for MSMEs by reducing bureaucratic bottlenecks, rationalize MSMEs registration and inspection requirements to lower transaction costs, and increase the competitiveness of MSME exports. While the policy environment may not have been ideal, sector reforms had been initiated and the overall direction of reforms was set. The SMDP disbursed 98.32% of the original loan amount. Outstanding MSE subloans reached $211.9 million at loan closing, which was attributed to economic growth and increasing demand for MSE loans. 22 In the absence of conditionalities, experience shows that advisory TA can be effective in the right circumstances. 23 To this end, the Second SMDP included an attached TA in its design. 40. The decision during the midterm review to raise the subloan limit from $20,000 to $40,000 contributed to meet market demand for larger subloans. B. Effectiveness in Achieving Outcome 41. The project was less than effective in achieving its expected outcome. PCBs outstanding microfinance loans reached $127 million (from a baseline figure of $75 million) against the original target of $360 million. The indicator of PCBs having a delinquency and write-off ratio not exceeding 5% was achieved (Appendix 4). The outcome target for outstanding microfinance loans established during project processing was considered overly ambitious; this was also highlighted following the midterm review. 42. At completion, $142 million was disbursed under the project. The excess amount was due to recycling of repaid subloans. The PCBs disbursed $600 million of subloans: Agrobank disbursed $30 million, Hamkorbank disbursed $432 million, and Ipak Yuli Bank disbursed $138 million. 43. Achieving the intended outcome of expanded, viable, and sustainable MSE finance operations was hampered by a lack of clarity with regard to CBU restrictions on subloans disbursed in cash and in US dollars, which resulted in banks lending only in local currency. Furthermore PCBs did not lend to MCOs as originally envisioned under the project, as microloans under the law on microfinance could not be given for entrepreneurial activities and microcredits could not be given in cash under CBU regulations The interest rate is not considered to have impacted the disbursement performance of the PCBs (para. 18). 21 ADB Validation Report: Small and Microfinance Development Project. Manila. 22 ADB Completion Report: Uzbekistan: Small and Microfinance Development Project. Manila. 23 ADB Country Assistance Program Evaluation for Uzbekistan. Manila. p. vi. 24 Government of Uzbekistan Law of the Republic of Uzbekistan on Microfinance.

19 11 C. Efficiency in Achieving Outcome and Outputs 45. The project was efficient in achieving outcome and outputs. Efficiency was impacted by (i) a delay in the declaration of loan effectiveness due to the late submission of SLAs, legal opinions, and the notices of PIU establishment by Agrobank and Hamkorbank; (ii) restrictions on disbursement (para. 6), which continued mainly via funds transfers; (iii) lack of clarity with regard to submitting loan withdrawal applications (which was resolved after an ADB review mission that guided PCBs on the methodology and procedures for submitting loan withdrawal applications, and clarified issues on subloan disbursements); (iv) the problems with Agrobank, and the reallocation of its remaining loan balance to the other PCBs; 25 and (v) the setting of overly ambitious performance targets. D. Preliminary Assessment of Sustainability 46. The project is likely to be sustainable. MSE operations have strong support from the government and have been viewed as a priority as part of accelerated private sector development and inclusive growth since Special programs have been initiated for SMEs. Microcreditbank was set up in 2006 with government involvement. Laws on MCOs and microfinancing, adopted in September 2006, create the foundations for the emergence of a new microfinance segment of the financial market, aside from credit unions, which were closed nationwide in In June 2007, the government adopted the Microfinance Development Program. 26 The government declared 2011 the year of MSE development. With regard to the Second SMDP, the PCBs find the operations profitable (with manageable risk), and the sector will continue to grow. 27 The Second SMDP, based on feedback from subborrowers and other stakeholders during review missions, (i) helped provide access to cash; (ii) improved access to medium-term financing of up to 3 years; (iii) supported women-owned and rural businesses; and (iv) helped improve borrowers capacity in management skills, financial literacy, and awareness of financial products. 28 E. Impact 47. The impact of the project is likely to be significant. Under the project, 21,968 jobs were created through PCB subloans, 37% of which were filled by women, well above the GAP target of 25%. While the total number of jobs created fell short of the targeted 30,000 jobs as indicated in the DMF, that target established during loan processing can be considered overly ambitious, and the results are substantial. 48. The impact on gender is likely to be significant. As indicated in para. 8, capacity building objectives were achieved or exceeded based on the number of loan officers and subborrowers trained. The GAP targets were substantially achieved, with more than 25% women loan officers and about 38% of women subborrowers trained. About 37% of individual microfinance loans opened were accounted for by women. The following were carried out by the TA consultants: (i) data collection on gender impact monitoring data from the PCBs; (ii) review of public information materials for women and enterprises; (iii) review of credit products, policies, and procedures, and existing financial products and services, with the aim of recommending modifications and improvements in products and delivery mechanisms to enhance outreach to women micro and 25 PricewaterhouseCoopers Uzbekistan, the external auditor, issued an unqualified opinion on the 2009 financial statements of Agrobank. 26 International Monetary Fund Republic of Uzbekistan: Poverty Reduction Strategy Paper. Washington, DC. 27 Discussions with the government point toward a third SMDP. 28 Midterm review (with attachments), Attachment 2, para.3. November BTOR L2634-UZB Loan.

20 12 small entrepreneurs; (iv) preparation of gender-sensitive MSE credit appraisal guidelines; and (v) a train-the-trainers workshop on gender-sensitive microfinance practices for overall financial inclusion for the three PCBs. 49. The impact on the PCBs is likely to be significant, as the TA concluded that there is a high need and demand for training even on a basic level, including for subborrowers. There is also a need to intensify training in selected core areas when there is limited training. 29 The PCBs have implemented environmental and social management systems (para. 25). The PCBs received several trainings from ADB staff and a consultant on environmental and social safeguards standards, their environmental and social management systems (ESMS) were reviewed and they were provided recommendations on how to strengthen their ESMS further. IV. OVERALL ASSESSMENT AND RECOMMENDATIONS A. Overall Assessment 50. The project was successful. Given the feedback from subborrowers, the project (i) made positive contributions in meeting requirements as indicated in para. 35 (except the provision of subloans in US dollars and in cash due to the CBU restriction); (ii) helped fill the credit gap for MSE loans in Uzbekistan; (iii) helped generate employment (although not to the extent envisioned at appraisal); and (iii) made positive contributions to capacity building and institutional development, and introduced requirements for greater and more intensive support in these areas in the future. The PCBs have gained practical experience in project implementation techniques and procedures as well as project monitoring. 51. The project was not implemented as planned, as disbursements to Agrobank were discontinued and reallocated to Ipak Yuli Bank and Hamkorbank. There was a substantial delay before the reallocation of unused funds became effective due to the delay in countersigning the SLAs. 52. While the DMF was generally well designed, three performance targets were overly ambitious; this was also true of some of the GAP targets. Both the DMF and the GAP would have benefited from a more rigorous analysis of macroeconomic, financial sector, and gender developments. Despite three DMF targets not being met, other targets were exceeded. B. Lessons 53. Promoting the success of MSE operations and helping PCBs fulfill their greater developmental functions would address existing constraints. Access to finance is just one constraint hindering the development of MSEs. In a 2003 survey of SMEs in Uzbekistan, the International Finance Corporation identified some of the negative factors hindering the operation of enterprises. These include administrative procedures, currency conversion, access to finance, taxation, and import and export restrictions (footnote 8). More effective consultation and dialogue among various market participants and government agencies regarding proposed reforms can limit inconsistencies between government policies and the formulation and implementation of rules and regulations. 54. Monitoring and analysis of the MSE sector would benefit from estimates of the credit gap to confirm additionality of ADB FILs. This could also help ensure that private sector funds are 29 ADB Capacity Building for Microfinance Development. Consultant s report. Manila (TA 7523-UZB).