IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-48040) ON A CREDIT IN THE AMOUNT OF SDR MILLION (US$57.4 MILLION EQUIVALENT) TO THE

Transcription

1 Public Disclosure Authorized Document of The World Bank Report No: ICR Public Disclosure Authorized IMPLEMENTATION COMPLETION AND RESULTS REPORT (IDA-48040) ON A CREDIT IN THE AMOUNT OF SDR MILLION (US$57.4 MILLION EQUIVALENT) Public Disclosure Authorized TO THE DEMOCRATIC SOCIALIST REPUBLIC OF SRI LANKA FOR A SMALL AND MEDIUM ENTERPRISE DEVELOPMENT FACILITY March 21, 2016 Public Disclosure Authorized Finance and Markets Global Practice South Asia Region

2 CURRENCY EQUIVALENTS (Exchange Rate Effective as of September 30, 2015) Currency Unit = LKR LKR = US$1 US$1 = XDR FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ADB AGO AWDR BPS BOC CBSL CCC CEA DFCC EA EIB EPL FA FM GDP GoSL ICR IRR ISR LOC M&E MoFP MTR NPV OM PAD PB PDO PFI PIU PSC RDB RSF SDR/XDR SIA Asian Development Bank Attorney General s Office Average Weighted Deposit Rate Basis Points Bank of Ceylon Central Bank of Sri Lanka Ceylon Chamber of Commerce Central Environmental Authority Development Finance Corporation of Ceylon Executing Agency European Investment Bank Environmental Protection License Financing Agreement Financial Management Gross Domestic Product Government of Sri Lanka Implementation Completion and Results Report Internal Rate of Return Implementation Status and Results Report Line of Credit Monitoring and Evaluation Ministry of Finance and Planning Midterm Review Net Present Value Operations Manual Project Appraisal Document People s Bank Project Development Objective Participating Financial Institution Project Implementation Unit Project Steering Committee Regional Development Bank (Pradeshiya Sanwardhana Bank) Risk Sharing Facility Special Drawing Rights Social Impact Assessment

3 SLIC SME SMEDeF SMF SMPB TA TTL Sri Lanka Insurance Company Small and Medium Enterprise Small and Medium Enterprise Development Facility Social Management Framework Sampath Bank Technical Assistance Task Team Leader Senior Global Practice Director: Gloria M. Grandolini Practice Manager: Niraj Verma Project Team Leader: Korotoumou Ouattara ICR Team Leader: Aminata Ndiaye

4 SRI LANKA Small and Medium Enterprise Development Facility CONTENTS A. Basic Information... i B. Key Dates... i C. Ratings Summary... i D. Sector and Theme Codes... ii E. Bank Staff... ii F. Results Framework Analysis... ii G. Ratings of Project Performance in ISRs... iv H. Restructuring (if any)... v I. Disbursement Profile... vi 1. Project Context, Development Objectives and Design Key Factors Affecting Implementation and Outcomes Assessment of Outcomes Assessment of Risk to Development Outcome Assessment of Bank and Borrower Performance Lessons Learned Comments on Issues Raised by Borrower/Implementing Agencies/Partners Annex 1. Project Costs and Financing Annex 2. Outputs by Component Annex 3. Economic and Financial Analysis Annex 4. Bank Lending and Implementation Support/Supervision Processes Annex 5. Beneficiary Survey Results Annex 6. Stakeholder Workshop Report and Results Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders Annex 9. List of Supporting Documents MAP

5 PROJECT DATA SHEET A. Basic Information Country: Sri Lanka Project Name: Small and Medium Enterprise Development Facility Project ID: P L/C/TF Number(s): IDA ICR Date: 03/21/2016 ICR Type: Core ICR Lending Instrument: FIL Borrower: Original Total Commitment: Revised Amount: Environmental Category: F REPUBLIC OF SRI LANKA XDR M Disbursed Amount: XDR M XDR M Implementing Agencies: Ministry of Finance and Planning Cofinanciers and Other External Partners: B. Key Dates Process Date Process Original Date Revised / Actual Date(s) Concept Review: 03/16/2010 Effectiveness: 01/05/2011 Appraisal: 06/29/2010 Restructuring(s): 06/26/2013 Approval: 09/07/2010 Mid-term Review: 06/10/ /11/2013 Closing: 03/31/ /30/2015 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Risk to Development Outcome: Bank Performance: Borrower Performance: Satisfactory Moderate Satisfactory Moderately Satisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Quality at Entry: Satisfactory Government: Moderately Satisfactory Quality of Supervision: Satisfactory Implementing Agency/Agencies: Moderately Satisfactory Overall Bank Performance: Satisfactory Overall Borrower Performance: Moderately Satisfactory i

6 C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments (if Indicators Performance any) Potential Problem Project at any time (Yes/No): No Problem Project at any time No (Yes/No): DO rating before Closing/Inactive status: Satisfactory Quality at Entry (QEA): Quality of Supervision (QSA): None None Rating D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Banking General industry and trade sector Public administration- Financial Sector 8 8 SME Finance Theme Code (as % of total Bank financing) Micro, Small and Medium Enterprise support E. Bank Staff Positions At ICR At Approval Vice President: Annette Dixon Isabel M. Guerrero Country Director: Francoise Clottes Naoko Ishii Practice Manager/Manager: Niraj Verma Ivan Rossignol Project Team Leader: Korotoumou Ouattara Cecile Thioro Niang ICR Team Leader: ICR Primary Author: Aminata Ndiaye Aminata Ndiaye F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The Project Development Objective is to improve access to finance, including term finance for Small and Medium Enterprises in Sri Lanka affected by the Global Financial Crisis in Sri Lanka. Revised Project Development Objectives (as approved by original approving authority) Not applicable. ii

7 (a) PDO Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) Formally Revised Target Values Actual Value Achieved at Completion or Target Years Indicator 1: Cumulative total number of active SME loans Value (quantitative or qualitative) , Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % achievement) Target 82% achieved. In total, 824 loans have been granted under the Small and Medium Enterprise Development Facility (SMEDeF) Project. Indicator 2: Cumulative number of active SME loans (>36 months maturity) Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % achievement) Target exceeded (134%). More than 80 percent of SMEDeF loans (by number) had a maturity greater than 3 years. Indicator 3: Cumulative total volume of loans to SMEs in US$, million Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Target fully achieved (100%). The actual amount disbursed is US$50.4 million, as Comments (incl. % achievement) per the Operations Portal. As per the Project Implementation Unit (PIU), the total equivalent amount disbursed in local currency is LKR 6.5 billion (all SME loans were granted in local currency). Indicator 4: Cumulative volume of loans to SMEs (>36 months maturity) in US$, million Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % achievement) Target exceeded (181%). In volume, 90% of SMEDeF loans had a maturity greater than 3 years. Indicator 5: PFI SMEDeF Project Portfolio Quality (NPL [% ] of more than 90 days) Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % achievement) Target exceeded. The amount of NPLs in the SMEDeF portfolio is 3%. iii

8 (b) Intermediate Outcome Indicator(s) Indicator Baseline Value Original Target Values (from approval documents) iv Formally Revised Target Values Actual Value Achieved at Completion or Target Years Indicator 1: Volume of Bank Support: Line of Credit - SME Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % Target fully achieved (100%). The line of credit was fully disbursed to SMEs. achievement) Indicator 2: Cumulative number of PFIs introducing processes/products/management systems specific to SMEs Value (quantitative or qualitative) Date achieved 09/07/ /31/ /30/ /30/2015 Target fully achieved (100%). All Participating Financial Institutions (PFIs) Comments (incl. % introduced new products, processes, or systems dedicated to their SME business (see achievement) core text). These include development of a management information system (MIS), credit scoring, upgrade of credit evaluation system, and so on. Indicator 3: Number of SMEs trained Value (quantitative or qualitative) 0 n/a 9,000 21,212 Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % achievement) Indicator 4: Target exceeded (236%). Through training sessions organized by the 8 PFIs or the Ceylon chamber of Commerce in collaboration with the PIU, 21,212 SMEs have been trained under the SMEDeF Project. Cumulative number of loan officers trained in SME banking procedures and credit risk management Value (quantitative or qualitative) ,000 12,652 Date achieved 09/07/ /31/ /30/ /30/2015 Comments (incl. % Target exceeded (158%). The project co-financed the training of 12,652 loan officers achievement) in SME banking and credit risk management. G. Ratings of Project Performance in ISRs No. Date ISR Actual Disbursements DO IP Archived (USD millions) 1 04/13/2011 Satisfactory Satisfactory /08/2012 Satisfactory Satisfactory /02/2012 Satisfactory Satisfactory /27/2013 Satisfactory Moderately Satisfactory /06/2013 Satisfactory Moderately Satisfactory /22/2014 Satisfactory Satisfactory /25/2014 Satisfactory Moderately Satisfactory /28/2015 Satisfactory Satisfactory /28/2015 Satisfactory Satisfactory 53.12

9 H. Restructuring (if any) Restructuring Date(s) Board Approved PDO Change ISR Ratings at Restructuring DO 06/26/2013 N S MS IP Amount Disbursed at Reason for Restructuring & Key Restructuring Changes Made in USD millions The main objective of the Level 2 restructuring was to accommodate the government s request that the risk sharing facility (RSF) be linked to the line of credit (LOC), as per the legal opinion of the Attorney General s Office (AGO). As a consequence, the proposed changes included (a) a reallocation of funds among project components and (b) an extension in the closing date to allow reallocated funds to be disbursed. Changes were also made to the Policy and Capacity Enhancement Component 2, which was scaled back and some activities dropped, as they were being undertaken by other donors. In addition, the project s results framework was amended to be coherent with the restructured components and activities, including the fact that funds for the RSF would entirely be made available from the government s resources. Disbursements under the project picked up after the restructuring. v

10 I. Disbursement Profile vi

11

12 1. Project Context, Development Objectives and Design 1.1. Context at Appraisal 1. At project appraisal, Sri Lanka was recovering from the tsunami, the global financial crisis, and a nearly three-decade-long conflict. The onset of the financial crisis resulted in a sudden outflow of foreign capital and a drain in the country s foreign exchange reserves, which were halved to US$1.2 billion between mid-september 2008 and mid-march The rapid rise in food and commodity prices during the first half of 2008 fueled consumer price inflation, which peaked in June 2008 to 28 percent (year on year). Armed confrontations in the north also intensified during 2008, peaking at the height of the global financial crisis, during the first half of The compounded effects of these crises threatened the country s achievements over the preceding years, in terms of economic growth and poverty reduction efforts. The end of the conflict in May 2009 brought about optimism but also signaled the many upcoming challenges. Large resettlement efforts in the north along with the global financial crisis strained public finances and dampened gross domestic product (GDP) growth from 6 to 3.5 percent between 2008 and Because the stimulus effect of reconstruction efforts and growing domestic demand were diffused, economic growth was starting to rebound in early 2010 as the project was being appraised; real GDP grew 7.1 percent in 2010 Q1 (year on year). 2. The Sri Lankan banking sector, although quite resilient through the different crises, could not fully play its role of financing economic recovery. The direct impact of the crises through the banking sector was limited because Sri Lankan banks were not exposed to toxic assets. However, with the decline of exports and growing pressure on the balance of payments, bank lending contracted significantly. The system remained overall well-capitalized with a capital adequacy ratio of 14.9 percent in end-december and profitable, although asset quality started to deteriorate in The nonperforming loan (NPL) ratio increased to 8.5 percent at the end of 2009 and was even higher for corporate and small and medium enterprise (SME) loans. Private sector credit to GDP fell from 33.3 percent in 2007 to 24.7 percent in Commercial bank lending volumes fell 3 percent in 2009 and SMEs were primarily affected because they are perceived to be riskier by the financial sector. Table 1. Banking Sector Financial Indicators % Gross NPL ratio Provisions/net NPLs ROA before tax ROE Loans/deposits Capital adequacy ratio Liquid assets ratio Source: Central Bank of Sri Lanka - Annual statistics. Note: ROA = Return on Asset; ROE = Return on Equity. 1

13 3. Despite their important role in the Sri Lankan economy and sustained demand for credit, SMEs were primarily constrained by access to and cost of financing after the global financial crisis. According to the 2008 Industry Survey, 91 percent of firms operating in Sri Lanka are SMEs, 1 employing a quarter of the labor force and representing 17 percent of the value-added generated. SMEs have been more adversely affected by the crisis than large firms (a) corporate lending rates increased to up to 21 percent in September 2008, translating into SME lending rates of up to percent; (b) with the global economic slowdown, many SMEs turned to subcontracting with large firms rather than directly exporting, which lowers their profit; and (c) banks shifted lending from SMEs to larger firms, which are a less risky segment. In a sample of seven large banks, the share of SME lending as a share of total lending decreased from 20 percent in 2008 to 18 percent in 2009, although total lending had increased by 2 percent over the same period. The financial and operational appraisal of a sample of banks estimated SME credit demand for in the range of US$1.5 billion, half of which was for long-term financing. SME lending by commercial banks was constrained by banks lack of technical capacity, lack of confidence in SMEs due to the unfavorable economic environment, and maturity mismatch preventing them from lending long term to SMEs. 4. This prompted an SME-friendly policy response, including a request to the World Bank for assistance in improving access to finance for SMEs. The Government of Sri Lanka (GoSL) introduced two stimulus packages of LKR16 billion in December 2008 and LKR 8 billion in May 2009, to support sectors affected by the global financial crisis, including exportoriented sectors. Sri Lankan authorities also requested a World Bank project that would provide access to medium and long-term funding to banks for on-lending to SMEs. The project was expected to increase the bank s ability to provide longer term maturities to SMEs for their productive investments and generate employment, in support of the country s recovery and reconstruction efforts. The project would target all provinces to ensure that SME credit and productivity gaps between advanced and lagging regions would not widen during the post-crisis period. 5. The operation was highly consistent with national priorities and the Bank s strategy and part of a comprehensive package of financing and technical assistance (TA) facilities provided by the Bank and other international organizations. The National Strategy for SME Sector Development published in 2002 defined broader access to medium-term finance a priority area for the government s SME policies, in an effort to make SMEs more competitive. Furthermore, the FY09 FY12 Country Assistance Strategy was built on the National Development Plans 2 and called for an expansion of economic opportunities in lagging regions. The project was financed through the Pilot Crisis Response Window Facility, which aims to assist eligible countries in alleviating the impact of the crisis. The project also built on the ongoing Renewable Energy for Rural Economic Development (RERED) Project approved in 2002 (with additional financing in 2008), and complemented credit lines for SMEs provided by the Central Bank of Sri Lanka (CBSL) and international organizations, such as the European 1 There was no single, uniform definition of SMEs in Sri Lanka at the time of appraisal. Annual turnover (usually less than LKR 300 million), total assets (less than LKR 100 million), and, in some cases, number of employees was used by commercial banks. This figure excludes firms of 1 5 employees. 2 The 2005 and 2010 National Development Plans of Sri Lanka, the Mahinda Chinthana, recognize the SME sector as the engine of growth for the national economy. 2

14 Investment Bank (EIB), Kreditanstalt für Wiederaufbau (KfW), and United States Agency for International Development (USAID) Original Project Development Objectives (PDO) and Key Indicators (as approved) 6. The PDO is to improve access to finance, including term finance, for SMEs in Sri Lanka affected by the global financial crisis. To measure progress toward the achievement of the PDO, the following key performance indicators were defined: (a) Number and volume of loans to SMEs (b) Number and volume of long-term loans to SMEs (c) Portfolio quality as measured by the percentage of NPLs (defined as being at least ninety [90] days past due) in the portfolio supported by the project 7. A discrepancy between the Financing Agreement (FA) and the Project Appraisal Document (PAD) in defining term lending through indicators to be monitored by the PIU has been noted. The FA defines term lending as having a maturity of at least five years (including after project restructuring), while the PAD, restructuring paper, and results framework defined term finance as lending maturity of more than three years. In addition, the initial Implementation Status and Results Report (ISR) framework immediately started monitoring loans with maturity greater than three years. Finally, the Operations Manual (OM) defined working capital loans as loans with a maturity of three years or less and investment loans as those with a maturity greater than three years. The five-year term in the FA, thus, appeared to be a mistake because it was not captured in the minutes as a negotiation point. Ultimately, loans with a maturity greater than three years were systematically tracked throughout project implementation as an outcome indicator for term finance. Such a definition (three years) is also consistent with typical definitions of long-term financing in SME projects and so technically valid. Data for five years was also not captured, and hence, the feasible option for the team was to report this discrepancy while keeping to the data that was available and the de facto definition that was applied during the project life. 8. Other key indicators that have been indicated in the PAD and FA to capture the outcomes of the TA components of the product include the number of participating financial institutions (PFIs) that introduced processes, products, or systems specific to SMEs, the percentage of SMEs that received a loan from a PFI after attending a training, and the availability of recommendations of a diagnostic study on addressing SME constraints. The PAD also anticipated that the project s higher-level objectives such as the impact of the project on productive activities (as measured by sales and employment) will be measured through impact evaluations. 9. SMEs affected by the global crisis in effect referred to all SMEs in the country because the impact of the global crisis cut across SME sectors and was all-pervasive. The global financial crisis, in addition to the Sri Lanka internal civil war, had hit SMEs in all sectors harder than large firms. Because they were perceived as higher risks, banks shifted lending away from SMEs in general toward large firms, resulting in the share of SME lending as a share of 3

15 total lending decreasing from 20 percent in 2008 to 18 percent in 2009, although total lending had increased by 2 percent over the same period Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification 10. The PDO was not revised during project implementation. The end-project targets of PDO indicators were revised at the time of project restructuring in 2013 to reflect the increased amount of the line of credit (LOC). 3 In addition, three intermediate outcome indicators were dropped. The indicator for Subcomponent 2b on Support to strengthen the enabling environment for SME banking was dropped because the subcomponent itself was eliminated due to concomitant activities from other donors related to SME development. The indicator on the relationship between the LOC and training (the percentage of SMEs that received a loan from a PFI after attending a training) was also dropped. The third indicator dropped was the number of loans provided through the RSF Main Beneficiaries 11. The PAD identified participating financial institutions (PFIs) as the project s direct beneficiaries. Targeted PFIs were the state-owned, private commercial and licensed specialized banks with a national branch network. Targeted PFIs would benefit from long-term funding and training for their employees. 12. The project s beneficiaries were identified as new and existing SMEs in all sectors of the economy in need of financing because their businesses suffered from the effect of the crises (including the global financial crisis and the internal civil war conflict) as a way to support SME growth and economic recovery. Beneficiary SMEs are those that receive financing from targeted PFIs and/or training through the project Original Components (as approved) 13. The project comprised two main components: Component 1: Financing and Risk Sharing Facility (US$41 million equivalent) consisted of two subcomponents: o o Subcomponent 1a: Line of Credit (LOC) to PFIs to refinance eligible short- and long-term loans to SMEs (US$28 million equivalent) Subcomponent 1b: Risk Sharing Facility (RSF) to reduce the banks risk of lending to SME borrowers (US$13 million equivalent). The Sri Lanka Insurance Company (SLIC) was intended to provide a limited risk coverage on a 50/50 pari passu, portfolio basis with PFIs that will elect to participate. 3 Two PDO indicators related to the number and volume of SME loans with a maturity of less than three years were replaced by indicators tracking the total number and volume of SME loans. The number and volume of loans with a maturity greater than three years was kept, with targets revised to reflect the new amount of the LOC. 4

16 Component 2: Policy and Capacity Enhancement for SME Banking (US$12 million equivalent) consisted of three subcomponents: o o o Subcomponent 2a: Capacity enhancement for lending institutions and SMEs (US$7 million equivalent). The project will provide TA on a matching basis to the commercial banks benefitting from the LOC or the RSF. The project will also include capacity enhancement and TA to SME clients, to be provided by the PFIs at the branch level. Subcomponent 2b: Support to strengthen the enabling environment for SME financing (US$1 million equivalent). Subcomponent 2c: Implementation support (US$4 million equivalent). 14. The project also included unallocated funding of US$4.4 million equivalent, which would be available for additional funding for any of the above components Revised Components 15. The project was restructured in June 2013, at which time the technical design of Component 1 was amended. Subcomponent 1b for the RSF was redesigned, after the RSF could not be operationalized in the first two years of the project in the absence of the legal agreement of the Sri Lankan authorities. Indeed, contrary to what was originally envisaged, the Attorney General of Sri Lanka issued an opinion in late 2012 that the RSF should be linked to the LOC, and only loans refinanced under the LOC would be eligible for coverage under the RSF. As a result, Subcomponent 1b was revised and renamed to show that the RSF will be offered to PFIs alongside a second credit line (LOC2). The RSF would this time be managed by the PIU (and not the SLIC), and offered on a loan-by-loan basis to cover only loans funded through LOC2. In addition, to ensure that project disbursements would not be halted by the mechanism of the RSF, the GoSL decided that the RSF would be fully funded from its own funds the total subcomponent amount would go toward LOC2. For LOC2, the interest rate charged by the GoSL to PFIs was changed from the six-month average weighted deposit rate (AWDR) to the one-month AWDR to allow the LOC to adjust more quickly to the lower interest rate regime toward the latter part of the project. 16. At the time of restructuring, Component 2 on Policy and Capacity Enhancement for SME Banking was also scaled back. In particular, activities that aim to strengthen the enabling environment for SME financing were dropped after the GoSL informed the team that they were already being funded by other donors. Original activities included (a) assessment to identify the legal and regulatory issues that constrain SME banking; (b) establish an out-of-court mechanism for rehabilitation of troubled SMEs; and (c) improve access to finance through setting up a movable collateral registry. However, during implementation, IFC signed an agreement with the Credit Information Bureau to provide technical support to enhance the Movable Asset Registry and German Agency for International Cooperation (GIZ) provided grant funds through its SME program to the GoSL to develop an SME strategy. Subcomponent 2b was hence dropped. Activities for Subcomponent 2a - Capacity enhancement for lending institutions 5

17 and SMEs, which were important for project sustainability, were kept to ensure that PFIs continue to undertake training on a number of activities that included specialized TA to improve their processes, as well as training to SMEs on relevant subject areas. Revisions to Subcomponents 2a and 2c, thus, only consisted of funding reallocations, and no activities were dropped (see section below) Other significant changes 17. Reallocations of project funds undertaken at the time of restructuring are summarized in table 2. Project funds were reallocated from Component 2 and unallocated funds toward a second LOC, in response to high demand for financing by SMEs. The second LOC was possible because of the following cost savings: (a) cost of TA for PFIs (Subcomponent 2a) was lower than budgeted; (b) Subcomponent 2b was dropped; (c) the operating cost of the PIU was lower than budgeted; and (d) unallocated funds allowed for some flexibility during project implementation. Table 2. Project Costs per Component (US$, millions) Components/Activities Original Post Restructuring 1 - Financing and Risk Sharing Facility a - Line of credit to PFIs (LOC1) b - Line of credit and Risk Sharing Facility (LOC2 + RSF)* ** 2 - Policy and Capacity Enhancement for SME Banking a - Capacity enhancement for lending institutions and SMEs b - Support to strengthen the enabling environment for SME banking c - Implementation support and monitoring Unallocated Total Note: *The component name changed at the time of restructuring. The original name was Risk Sharing Facility to reduce the banks risk of lending to SME borrowers ; **The full amount is dedicated to LOC For coherence with the restructured components, indicators associated with both Components 1 and 2 were revised. Indicators related to the RSF and Subcomponent 2b were dropped, while targets were revised for those associated with LOC1. Annex 1 presents the fully updated results framework. 2. Key Factors Affecting Implementation and Outcomes 2.1. Project Preparation, Design and Quality at Entry 19. The project s design was built on the Bank s experience with credit line operations in Sri Lanka 4 and other countries and analytical work on SME development. Given the rapid response required to the crisis, the project was approved by the Board within six (6) months of concept note approval. 20. The following lessons learned and prior guidance were incorporated: 4 The Bank s series of four Sri Lanka Small and Medium Industries projects (SMI I-IV) was reviewed by the IEG. 6

18 (a) Sound, yet flexible, project design is necessary. The project had no restrictions on eligible sectors 5 or type of SME loans to be financed (working capital or investment loans). There were also no regional restrictions. In April 2013, in its restructuring request, the government sought to exclude SMEs from the western province, given their high share in the SMEDeF portfolio (21 percent), with the objective of directing funds toward the more financially constrained provinces. Project design remained flexible after restructuring, balancing the need to meet demand for financing and to ensure equitable access to financing for all SMEs independently of their location. The project design was hence in line with conclusions of the World Bank Guidance Note on Using Line of Credit Projects to Support SMEs during the Financial Crisis, which explicitly cautioned against any restrictions in the case of crisis response LOCs. Finally, the selection of PFIs for the original LOC followed an open tender process. This ensured not only that participation was demand-driven but also the participation of sound financial intermediaries with a strong interest in developing SME finance. (b) Lending to SMEs should be demand-driven and market-oriented. The project did not impose the on-lending rate to SMEs but rather left to PFIs to lend at competitive rates. Eligibility criteria for subprojects were limited, and PFIs were responsible for selecting and carrying the credit risk of eligible SMEs. Qualitative criteria, set out in the OM and monitored throughout implementation, helped ensure that choices were made on a sound and objective basis. In addition, the foreign exchange risk of the project was borne by the GoSL. The government was able to integrate losses in the market on-lending rate that it charged PFIs for the LOC. PFIs were able to borrow and lend in local currency only, which shielded them from dollar volatility. This is a major lesson learned from prior credit lines. Finally, contrary to other credit lines, the project did not impose additional financial management (FM) and procurement requirements on SMEs record-keeping responsibilities were fully assumed by PFIs. This is coherent with the project objective to finance SMEs that have been affected by the crisis. Additional project requirements could deter SMEs from seeking financing through the project, even at better terms, as discussed in recent project completion reports. (c) PIU should demonstrate strong management and monitoring capacity. The evaluation of previous credit lines systematically point to weak accountability and inadequate management capacity as key determinants of the failure of LOCs. The PIU was staffed with civil servants and contractual staff from the Ministry of Finance and Planning (MoFP), and implementation arrangements were well defined. The project also anticipated the recruitment of a technical (SME banking) expert before project effectiveness to strengthen the PIU in its capacity to implement the project. In addition, a Project Steering Committee (PSC), comprising GoSL and private sector representatives, was a key project requirement and was formed before 5 It is noteworthy that targeted beneficiaries during project preparation were initially exporting SMEs. This restriction was lifted before Board approval because Sri Lankan SMEs had been affected by the Global Financial Crisis across sectors and across the country. 7

19 Board approval. With regard to the selection and monitoring of PFIs, arrangements were described in the project s OM. (d) The development of the SME sector requires a more comprehensive approach than just providing long-term funding. The project was innovative in three ways: (i) an RSF was included to test the appetite for leveraging project funds to substantially increase SME financing in Sri Lanka; (ii) TA was provided to both banks and SMEs to build capacity on both the demand and supply sides; and (iii) the project aimed to review the environment for SME financing and alleviate related constraints to SME financing. Prior analytical work and previous experience of IFC s TA program in Sri Lanka (both for banks and SMEs) have been considered in seeking to achieve the highest impact for sustainably improving access to finance by SMEs. 21. The project document included a candid discussion of several risks to the project and associated mitigation measures. The project s risks included (a) macroeconomic risk because of the global economic slowdown; (b) credit risk of SMEs; (c) appetite for SME financing by banks; and (d) implementation risk in terms of PIU capacity and political interference in credit decisions. These risks were adequately mitigated by the resilience of the Sri Lankan economy to the global crisis and to project-specific measures described in the PAD and the OM. 22. The risk framework, however, failed to discuss a major risk that materialized during project implementation, that is, the operationalization of the RSF. While the PAD discussed the risk of a lack of interest from the private sector in the proposed instrument and the investment risk of the RSF for GoSL, the discussion of a possible low uptake of the RSF was not included. This was not corrected at project midterm and restructuring Implementation 23. Several factors affected project implementation. Implementation of the RSF: 24. The RSF was not used by banks, as envisioned for two main reasons: (a) The original operational mechanism of the RSF was not approved by the Sri Lankan authorities. The Attorney General s Office (AGO) legal opinion on the RSF (precondition for its operationalization) was substantially delayed and did not come until late 2012, two years into implementation. The AGO did not have adequate understanding of the new financial instrument (RSF), which followed current best practice in its design, and the AGO thought that the RSF would lead to contingent liability for the GoSL beyond the explicit sovereign guarantee in the RSF agreement. In addition, as noted in a June 2012 aide memoire, there were changes of the official in the AGO s department who was handling the RSF. After a protracted series of discussions between the AGO and the MoFP, the AGO rejected the original project design, and its legal opinion was that the RSF should be restricted to loans provided under the SMEDeF LOC. The team had no choice but to realize in January 8

20 2013 that the component as designed would never take off. A project restructuring was undertaken in June 2013 to revise the design of the RSF as approved by the AGO. (b) The confidence and appetite of the banking sector in the RSF was limited. As noted in several aide memoires, as well as the project restructuring paper, banks had a negative experience with the previous RSF that was operated by the CBSL. In particular, the claim processing delays and difficulty to trigger the guarantee remained a fresh memory for banks. To accommodate the fear of the banks, the RSF was, thus, to be implemented by an entity (SLIC) other than the CBSL. Although the technical underpinning of the RSF being offered as an innovative instrument were strong, in the end, despite high commitment from the authorities and some banks expressing a particular interest in participating in the RSF, they did not make use of the facility on grounds that SME borrowers were not willing to pay the premium of having their loans backed by the RSF (banks would pass the RSF fee of 100 basis points (bps) onto borrowers, in exchange for lower quality/value of collateral). PIU Capacity and Effectiveness of the PSC 25. Two key implementation arrangements, which have been defined as legal requirements in the FA, were not fully effective. (a) Unavailability of SME banking expert in PIU throughout implementation. The decision to strengthen the management and implementation capacity of the PIU with the recruitment of an SME banking expert aimed to mitigate the risk of project implementation delays because of the PIU s lack of familiarity with banking practices. Despite the requirement to maintain this position throughout project implementation, the SME banking expert position was only filled between mid-2011 and mid The most direct impact of this absence was felt in the PIU s proactivity and in the implementation of the TA component (Subcomponent 2a), which had a slow start, was cut down at the time of restructuring, and was less quality focused than anticipated. In fact, the proactivity of the overall PIU was variable as the project director and the deputy project director kept other full-time positions within the MoFP. (b) Effective Project Steering Committee (PSC). The PSC met only once during project implementation, in August 2015, that is, a month before project closing. That was despite repeated reminders from the Bank team as revealed in project supervision documents. The inactivity of the PSC was most forcefully felt in the inability to receive a legal opinion on the RSF until The objective of the PSC was, owing to its composition, to be able to rapidly escalate implementation issues that required high-level attention and decisions. It was anticipated that the Committee would meet regularly throughout implementation to play that role and accelerate implementation. 9

21 Line of Credit a) Credit demand. The project included a crisis response LOC, which did not take off immediately; the project was launched in April 2011, approved in September 2010, and became effective in January The absence of a national clear definition of SMEs also contributed to the slow start as the PFIs had to adjust to a new definition and adapt their processes accordingly. The initial delay was rapidly overcome because of overwhelming demand and financing shortage. The demand for LOC1 exceeded almost four (4) times the size of the LOC. By June 2012, 75 percent of LOC1 was committed and 44 percent disbursed. By June 2013, up to 95 percent of LOC1 had been committed and 77 percent disbursed. This disbursement lag is best explained by (a) the lag between issuance of non-objection on an SME loan application (by the PIU) and the full disbursement of the loan to the SME (by the PFI for example construction loans were disbursed in tranches) ; (b) the interest rate differential (see paragraph below), which meant that banks sometimes used their own resources to disburse loans from the SMEDeF pipeline; and (c) lag for SMEs to obtain proof of compliance with environmental regulation (see section 2.4). More generally, for LOC1 and LOC2, the disbursement of the LOCs was also affected by the uncertainty of the business environment. b) Interest rate. The project has been affected by the country s high interest rate regime. The interest rate charged by the MoFP to PFIs under the LOC was variable, pegged against the AWDR, which is used as a basis for commercial lending rate. With the AWDR increase by 260 bps between June 2012 and June 2013, SME lending rates reportedly varied from 12 to 19 percent as some banks kept their spread as high as 9 percent. Overall, prime commercial lending rates increased from 9 percent in June 2011 to 13 percent in mid-2013, before substantially decreasing to 6 percent in September To keep the lending rate to SMEs as close to market rates as possible, the lending rate to PFIs was revised from the six-month AWDR to a one-month AWDR at the time of restructuring. This was especially helpful to sustain the demand for the SMEDeF LOC when prime lending rates started to fall as the six-month AWDR was slow to adjust to market conditions. Several banks used their own resources to lend to SMEs before this pricing revision, to remain competitive and retain clients. Some SMEs were indeed aware that lower rates were offered through other donors lending programs (for example, the EIB), although eligibility criteria differed between donor programs and they did not necessarily qualify for these loans. c) Participating Financial Institutions. The SMEDeF LOC was made available to licensed commercial banks for SME loans, with pricing and maturity in line with ongoing commercial practices to ensure that no market distortions are created. Given the project s ambition to strengthen SME financing, it was agreed that TA would be mandatory for all PFIs participating in the LOC. LOC1 agreements were signed with eight PFIs that were selected in March 2011 following a national competitive process. The OM for the LOC provided the eligibility criteria for PFIs (both at the time of selection and during project implementation). LOC2 agreements were signed with PFIs selected for LOC1 that were subsequently interested in 10

22 participating in LOC2.6 Table 3 summarizes the LOC allocations and usage for the eight PFIs. As evidenced in table 3, three banks Sampath Bank (SMPB), People s Bank (PB), and the Regional Development Bank (RDB) increased their appetite in the SMEDeF LOCs between the two rounds. Further details are provided in section 3. Type Table 3. Allocation of the LOCs by PFI LOC1 Allocation (LKR, million) LOC2 Allocation (LKR, million) Total Allocation (LKR, million) Share of Total PFI Allocation (percentage) Disbursement as Percentage of Allocation BOC Public PB Public , HNB Private Commercial Bank Private DFCC Bank Private NDB Private SMPB Private , , RDB Public Total 3, , , Source: PIU Monitoring and Evaluation (M&E) Design, Implementation and Utilization 26. At approval, the project s M&E framework was comprehensive and well-designed. The M&E framework designed PDO indicators in line with the project s objectives. The original indicators were in line with the project s activities and adequately measured the project s impact. The PAD also provided detailed M&E arrangements and reporting requirements, which were further elaborated in the project s OM. The OM provided templates for loan applications and some reporting for consistency and reliability purposes. The frequency and scope of monitoring reports were explicit. In addition, baselines and targets were indicated for all indicators. Indicators were also effectively monitored during project supervision. Almost all aide memoires and ISRs provided updated information for the results framework. 27. At the time of the project restructuring, M&E arrangements were amended for consistency with the revised project design. First, the targets for the results were revised to take into account the new LOC and the implementation delays related to the RSF (the associated indicator was dropped). Second, the OM was revised to incorporate LOC2 and the new operational mechanism of the RSF. Third, three indicators were dropped, one of which aimed to measure the impact of the TA to SMEs on their ability to receive financing. The quality and content of the training provided to SMEs were difficult to monitor for the PIU, and monitoring efforts were thus limited to the number of SMEs trained. Possible indicators for the TA to SMEs could have been related to the content of classes offered and reported use of training by SME beneficiaries in a beneficiary survey. 6 All but one PFI also signed up for LOC2. 11

23 28. The PAD anticipated that impact evaluations would be undertaken to measure the impact of the project, against the higher-level objectives of increased employment and sales. Surveys of SME beneficiaries were expected to be carried out and also administered to nonbeneficiary SMEs. Although a rigorous impact evaluation at the beneficiary level was not undertaken, the Bank team recruited a consultant who followed up individually with the PFIs to collect such impact data directly from their clients. In addition, the project carried out a Social Impact Assessment (SIA) in 2014, which provides useful insight on the spillover effects from the project. 29. The PIU s M&E capacity was limited. Although the PIU was able to provide the necessary information on the LOC for the results framework 15 days after the end of every quarter, the system was unable to provide the updated, real-time disbursement status of the LOC within that reporting period. This is because of the PIU s dependency on the CBSL s system to obtain disbursement data to PFIs 7. The PIU s M&E system only provided information on the noobjections given to PFIs on loan applications submitted (that is, loan applications that have been sent by PFIs and that are eligible for financing, based on the LOC s eligibility criteria; disbursement to the PFIs occurs afterwards). This has no effect on the reliability of data provided for the results framework but only limits (a) the frequency at which the results framework could be updated and (b) the wealth of data available Safeguard and Fiduciary Compliance 30. Fiduciary and safeguards requirements were detailed in the project s Operations Manual (OM). This sections presents key arrangements and highlights issues raised during project implementation. Procurement 31. Procurement arrangements for the project were described in the PAD and further developed in the OM, with clear delineation of roles and responsibilities. According to the PAD, and based on an assessment carried out at the time of appraisal, procurement under the LOC facility and the capacity enhancement activities was allowed to be carried out considering economy and efficiency. The arrangements followed local established commercial practices acceptable to the Bank. 32. The overall performance of the project throughout implementation is Satisfactory. Semiannual post reviews were carried out by the Bank procurement specialist at the PIU and PFI levels. A random review joint procurement and financial management (FM) review covering a sample of beneficiary SMEs was undertaken in the field in 2014 to ensure compliance of procurement requirements by SMEs. In addition, procurement plans were submitted on time. No issues were raised during implementation. 7 The Project s Designated Account was held at the Central Bank. This meant that all project s disbursement data had to wait for central bank data to be generated. As a result, the PIU did not have realtime access to the project s financial data. 12

24 Financial Management 33. Financial management arrangements were provided in the PAD and OM, including detailed reporting and internal systems requirements. An assessment of the FM capacity and control systems of the PIU was carried out during appraisal and judged to be satisfactory. PFIs receiving funds from the LOC were responsible for disbursing them to the SMEs as well as for maintaining the necessary books and records relating to such disbursements. 34. The overall performance of the project in terms of FM has been Satisfactory. Semimonthly implementation support review missions were carried out by the Bank s FM specialist at the PIU and PFI levels, complemented by the random review of SMEs in Elements including budgeting, financial reporting, external auditing arrangements, accounting and FM staffing in the PIU, funds flow, and internal controls have not experienced any major issue throughout project implementation. The PIU has generally submitted accurate and timely interim unaudited financial reports, as well as acceptable annual audit reports, thereby complying with the two main FM covenants in the project FA. No serious observations and accountability issues were contained in the project s external audit reports. No audit observations have required further follow-up action from the Bank. Environmental and Social Safeguards 35. Project documents provided very detailed information and practical guidance to the PIU s and PFI s lack of prior experience. At appraisal, the project was assigned Category FI in accordance with Bank safeguard policy OP/BP/GP Environmental Assessment, and it was agreed that the project would not finance an identified set of SMEs/sectors (see also the OM). An Environmental Risk Management Framework was prepared to help minimize potential risks associated with beneficiary SMEs financed under the LOC. The PIU was required to be staffed with an environmental specialist. In the OM, the section related to environmental safeguards was comprehensive and included an environmental screening tool and guidelines for actions for the PIU and PFIs. Roles and responsibilities were clearly outlined (PIU, PFIs, SMEs, and IDA). It was agreed before approval that beneficiary SMEs would be required to provide evidence of compliance with environmental standards including the Environmental Protection License (EPL) to be issued by the Central Environmental Authority (CEA). In addition, annual environmental audits would be undertaken to assess field-level compliance with environmental safeguards. 36. The overall environmental safeguard compliance rating of the project throughout the life of the project is Moderately Satisfactory. At the onset of the project, the PIU prepared environmental guidelines for SMEs in Sri Lanka, which were prepared to serve as a technical guide for the financial sector and SMEs to increase their awareness and understanding of environmental standards. However, during the first years of implementation, there were delays in ensuring full compliance with the related project requirements. Annual environmental audits, which covered a sample of 10 percent of beneficiary SMEs, were initially delayed and reporting by the PIU was slow. Implementation improved subsequently, and the project maintained a satisfactory rating over its lifetime. According to the routine reviews conducted by the Bank s environmental specialist, screening procedures and monitoring of implementation by PFIs have been satisfactory. No major environmental issues have arisen over the lifetime of the project. 13

25 37. The implementation of the environmental safeguards raised two issues for follow-up. First, during supervision missions and the Implementation Completion and Results Report (ICR) mission, PFIs raised the delay in obtaining the required EPL as a key hindrance to LOC disbursements. Several PFIs informed the Bank team of the long time taken by the CEA or the local authority to process EPL applications and grant operating licenses to their clients. In response, the PIU granted a 120-day grace period for PFIs to submit the EPL for SMEDeF loan applications. Still, the most common cause of rejection of loan applications submitted by PFIs was the absence of the EPL. Second, a related issue is the absence of follow-up on sustained compliance with environmental regulations for SMEs that have received an EPL. During field visits (supervision and ICR), it was apparent that some SMEs that had obtained an EPL did not take action to fully adhere to the requirements. This is primarily because of the lack of monitoring and random visits by the concerned authorities, primarily the CEA. The two issues point to a need to strengthen the regulatory, supervisory, and monitoring capacity of relevant authorities. 38. Within the scope of the project, two initiatives were taken to strengthen compliance with environmental regulation in a sustainable manner: (a) capacity building of SMEs through the TA component; training sessions on environmental standards were provided to SMEs through the Ceylon Chamber of Commerce with support from the PIU s environmental specialist and (b) monitoring of and TA to PFIs. Some PFIs have conducted related training for their branch officer, cofinanced by the project under Component 2 on policy and capacity enhancement for SME banking. As a result of the project s support, PFIs have started to incorporate environmental safeguards mechanisms into their credit appraisal process, with some banks now requiring EPLs, although with added flexibility compared to project requirements (six-month grace period to submit EPL after loan approval). 39. Social safeguards requirements were adequately presented in the PAD and OM. No social issues were foreseen under the project. In keeping with the Bank s safeguard policies, the PIU prepared a Social Management Framework (SMF) for the overall project that identified the potential social risks associated with SMEs to be supported under the SME lending program and the need for ensuring compliance with the labor laws and regulations of the country and the Bank s social safeguard policies. By identifying potential risks of the project, the SMF spelled out mitigation measures that were necessary for minimizing risks through improving access to finance and outreach in different provinces. Among these mitigation measures was the requirement to conduct an annual SIA. This sought to assess the extent to which concerns of unequal access to project resources for SMEs (especially for SMEs operating in the northern and eastern provinces) as well as risk of violation of national labor laws by beneficiary SMEs would materialize. 40. Overall implementation of social safeguards is rated Moderately Satisfactory throughout the life of the project. Although the launch of the SIA was required in the FA within one (1) year of project effectiveness, the Bank team advised the GoSL to undertake the SIA in 2012, to cover a larger number of loans disbursed under the credit line. In the end, only in May 2014, the PIU submitted the project s first SIA, which had a built-in mechanism to assess the broader social impact of the project and identify any foreseen negative social impact or obstacles to fair access by SMEs. The SIA also aimed to ascertain that SMEs were abiding by the country s labor laws, the monitoring of which was not undertaken by the PIU but rather by the 14

26 Labor Department. The absence of a focal person dedicated to social safeguards within the PIU was identified as a reason for the delay. In addition, an impact evaluation according to the SMF was not prepared at project closing. There were, however, no major issues raised during the project s life. Most of the impacts of Component 1 on the LOC were related to improved access, submission, review and approval of loan applications, and the impacts of SMEs on the regions and then on wider society, once the loans were granted. Therefore, most of the mitigation strategies suggested in the SMF were related to creating awareness among potential borrowers, ensuring improved access to finance, and so on. While the PIU can be commended for the efforts taken to raise awareness among potential beneficiaries, the project team should have taken a leaf out of the awareness raising strategy adopted by the Bank-funded Sustainable Tourism Development Project where the regional chambers were put to maximum use Post-completion Operation/Next Phase 41. The PIU has developed project management capacity and experience with fiduciary, safeguards, and reporting requirements that will be useful for future projects, including the recently approved credit line facility financed by the Asian Development Bank (ADB). This was the first LOC operation managed by a PIU located within the MoFP. Almost all PIU staff were civil servants and all staff have been trained during project implementation on Bank requirements. The experience of the SMEDeF Project prepares the PIU well for future projects that would fall under the leadership of the MoFP, either to support SMEs in general or more specialized SMEs. Until now, credit lines have been managed either by the CBSL or directly by a few banks (for example, DFCC Bank). Having a PIU within the MoFP presents the distinct advantage of being able to cater to a wide range of financial institutions and of obtaining direct feedback for the formulation of policies to improve SME finance. The ADB has approved a US$100 million credit line that will also be managed by the Department of Development Finance, which was responsible for the implementation of the SMEDeF Project. 42. Demand for SME credit remains high, as well as requests for capacity building for both PFIs and SMEs. SMEs continue to receive financial support from other donors such as the EIB and the ADB. The particularity of the SMEDeF Project was highly appreciated by beneficiaries in that it included both financial assistance and TA, while other donors provide one type of assistance or the other. During the ICR mission, all project stakeholders insisted on the need to combine both forms of assistance to ensure higher-level impact on SME credit (both demand- and supply-side issues) and ultimately on competiveness of the private sector. In addition, PFIs confirmed the shortage of long-term funding to allow them to lend to SMEs at long maturities. Future credit lines should thus also aim to provide long-term financing to PFIs, in exchange for the requirement to continue revolving project funds toward SMEs at least once after project completion for increased leverage of project funds. 43. More generally, access to finance remains a major constraint to SMEs in urban and rural areas, both in terms of equity and debt. In its 2016 budget speech, the Government announced the creation of a risk-sharing facility in 2016 that would be set up with government capital and participation of selected financial institutions. Under the credit guarantee scheme, SME loans will be covered at 75 percent (principal), with the goal of reducing financing costs and lengthening maturities. The government also signaled its intention to help address the financial sector s reluctance to lend to innovative SMEs and its high reliance on collateral for 15

27 financing. Efforts of Sri Lankan authorities to improve the business environment, enhance the credit infrastructure, reduce financing gaps between developed and lagging areas, and develop alternative sources of funding for SMEs (including expanded access to capital markets) ought to be encouraged and supported. 3. Assessment of Outcomes 3.1. Relevance of Objectives, Design and Implementation Rating: Substantial 44. The project remains highly relevant to Sri Lanka s development priorities. The project is fully consistent with the FY13 16 Country Partnership Strategy and the Government s vision the Mahinda Chintana Vision that aimed for sustained domestic investment for high economic growth as the country transitioned to middle-income status. In addition, as outlined in the government s 2016 budget speech, access to long-term finance for SMEs remains high on the agenda. The announced creation of a new RSF in 2016 and the acknowledged need to develop the capacity of SMEs (core business skills, marketing, and so on) also confirms the relevance of project design. The project is also fully aligned with the National SME Strategy published in October The project design reflects the identified constraints facing SMEs at preparation and during implementation. An enterprise perception survey of Sri Lankan firms undertaken in 2011 revealed that access to finance remained among the major constraints for firms. In the context of the global economic crisis, developed and emerging economies have sought to provide substantial support to SMEs, which were most affected. In Sri Lanka, the effect of the crisis was compounded by the country s sociopolitical environment. The project thus provided essential, countercyclical funding, at competitive rates. 46. The relevance of project design is high because of the following features: (a) the project was demand-driven it did not set rigid pre-allocations between PFIs, sectors, or geographic areas; (b) the designed intervention sought to address both demand- and supply-side constraints SMEs and PFIs received an assistance package of financing and training to allow for a sustainable development of SME finance; and (c) the project was innovative in the diversity of instruments (LOC and RSF). 47. Moreover, the objective of establishing an RSF that could be sustained beyond the life of the project is still relevant. While its operationalization failed to take off due to a combination of lack of diligence from the GoSL and confidence by PFIs, the renewed commitment to establish a national RSF in the 2016 budget speech is evidence of the saliency of SME credit risk in the banking sector. 48. The project restructuring undertaken in June 2013 revised the design of the RSF while increasing the amount of the credit line. The ability to have a second LOC, a de facto additional financing, was enabled by the GoSL s decision to fund the RSF s amount from its own resources. This implementation decision led to a higher reach of the project than would have been achieved if the project sought to finance both a second LOC and the RSF with the initial US$22.4 million allocation. However, while this meant increased financing for PFIs for on- 16

28 lending to SMEs, the PFIs concerns related to the RSF were not assuaged. An increased implementation role of the GoSL in the RSF may have brought forward memories from the previous RSF and thus contributed to the failed launch of the restructured RSF. 49. The implementation of the capacity enhancement Component 2 became, however, mechanical and slightly less impact driven than envisioned at project design. This is best explained by (a) the unclear communication of eligible activities for TA to PFIs (including SME process enhancements and not only training); (b) the absence of the SME banking expert in the PIU in the last two years of the project, once the LOC was in cruising mode; and (c) the results framework s emphasis on the number of people trained after the only indicator linking the capacity building of SMEs to their ability to receive financing was dropped. Toward the last year of implementation, the PIU and Bank team put greater emphasis on tailoring training sessions for SMEs to the development of entrepreneurial, managerial, and professional skills. 50. The M&E arrangements, although adequate for compliance with reporting requirements, did not allow for additional, granular-level information on SME borrowers and loans. As such, information required for the results framework was readily available for regular reports required by the Bank and for the ICR, but further data analysis was constrained. For example, information on individual SMEDeF borrowers and their participation in any training sessions was unavailable at the PIU level. Similarly, the average maturity of loans or interest rate per province was not available. This also meant that the number and volume of loans with a maturity greater than five years (to reflect the FA s definition of term finance) could not be provided by the PIU because it was not tracked in any of their systems. It is noteworthy that all other project documents (PAD, OM, restructuring paper, and their respective results framework) referred to long-term SME loans as those with a maturity of more than three years. This was, thus, the indicator related to long-term finance that was tracked during project implementation Achievement of Project Development Objectives Rating: Substantial 51. The Project Development Objective is to improve access to finance, including term finance, for SMEs in Sri Lanka affected by the global financial crisis. 52. Banks could positively respond to crisis-affected SME borrower demand because of the availability of the SMEDeF project funding. SMEs had been more adversely affected by the global crisis than large firms: (a) corporate lending rates increased to up to 21 percent in September 2008, translating into SME lending rates of up to percent; (b) with the global economic slowdown, many SMEs turned to subcontracting with large firms rather than directly exporting, which lowers their profit; and (c) banks shifted lending from SMEs to larger firms with increased demand after the end of the war, which are a less risky segment. Given the mismatch of maturities in the lending and deposits, relatively smaller banks such as SB, DFCC, RDB and NDB had access to long-term credit under the SMEDeF facility and they could positively respond to the demand from crisis-affected SMEs in all sectors of the economy. 53. Improved access to finance was measured by the number and volume of loans to SMEs. In total, 824 SME loans (82 percent of revised target achieved) of an aggregate amount of 17

29 LKR 6.5 billion (US$50.4 million) 8 have been disbursed during the project life. The majority of loans, 57 percent (in number), were below LKR 5 million (US$38,500). The average loan amount in the SMEDeF portfolio is LKR 7.9 million (US$61,000). The maximum loan amount was LKR 60 million (US$462,000). The median loan amount is LKR 15 million (US$115,500). 54. The provision of term finance, defined as financing with maturity of three (3) years or more, was successfully achieved in the project. In total, 671 SME loans (134 percent target achieved) totaling LKR 5.9 billion (US$45.6 million) had a maturity greater than three years. Within the SMEDeF portfolio, long-term loans represented 81 percent of total loans by number and 90 percent by volume. Thus, the project targets for total and volume of long-term lending were exceeded (134 percent and 181 percent, respectively). The average size of long-term SMEDeF loans was LKR 8.8 million (US$68,000). By comparison, the average size of shortterm loans (maturity below three years) was about LKR 4 million (US$31,500). 55. The sustainable increase in access to finance requires an acceptable level of NPLs in the portfolio supported by the project. Portfolio quality in the SMEDeF portfolio is higher than targeted. The share of nonperforming SMEDeF loans was 3.37 percent (compared to target of 8 percent fully achieved). The NPL ratio ranged from 0 percent to 8.2 percent for the eight participating banks. 56. Performance against the targets associated with the original LOC before restructuring (that is, LOC1) is also satisfactory. Under LOC1, 409 (82 percent target) SME loans have been disbursed, more than 75 percent of which had a maturity greater than three years (against target of 50 percent). The NPL ratio was also substantially lower than the target of 8 percent (unchanged during implementation period). 57. The project s impact far exceeds the amount of financing provided through the project, as revealed in the PFI survey. First, the project helped banks reach out to new SME clients: about a third of the loans have been granted to new SMEs (by number and volume). Second, SMEDeF loans helped finance 107 new startup enterprises. Third, PFIs reported that for every LKR 1 million loan disbursed to an SME in the portfolio, they have contributed through their own funds, an additional LKR 350,000 (US$2,700) in additional loans or other financial assistance to the same SME. PFIs further estimated that the total volume of loans disbursed because of the SMEDeF Project (project and PFI resources combined) is close to LKR 10 billion (US$77 million). Fourth, with more than 80 percent of loans classified as investment loans (three years or more), the SMEDeF contributed to the higher-level impact of increased investment and employment. This suggests that the SMEDeF LOCs helped address the maturity mismatch of funds in the banking system, thus helping banks lend at longer maturities. The PFI survey also revealed that with LKR 6.5 billion of loans received, SME beneficiaries have actually financed projects of LKR 13.9 billion (2.2 leverage), contributed LKR 5.8 billion in fresh equity in their business, and created 6,500 additional jobs. 8 Loans were provided by the PIU to PFIs in local currency. We thus rely on the total volume of loans provided by PFIs to SMEs, as per the PIU disbursement data (LKR 6, ) throughout the report. In the Bank Operations Portal, the total amount disbursed for LOC1 and LOC2 is US$50.4 million. The corresponding exchange rate of LKR per US$ will thus be used throughout the report as the average exchange rate during the life of the project. The average monthly exchange rate during project implementation varied quite significantly from to in September

30 58. Through support to PFIs, the project also contributed to strengthening their capacity to increase their lending activities. The TA to PFIs contributed to (a) the training of 12,652 loan officers (158 percent of target) and (b) process/products/systems improvement in all eight PFIs (100 percent of target). The TA for PFIs, mandatory for participation in the LOC, was cofinanced by banks on a matching basis and aimed to increase the banks capacity to lend to SMEs. The following enhancements can be directly attributed to the project: integration of environmental and social safeguards policies in the credit appraisal process; development of e- learning modules (as alternatives to face-to-face training); preparation of the bank s SME strategy; establishment of SME centers (business development centers dedicated to SMEs, clients, and nonclients); and development of refinancing, risk management, credit risk rating, and M&E systems for SME finance. For some banks, the impact of the project was even greater than for others (see Box 1). For example, the RDB and PB had a vision to lend to SMEs and used the project to launch their SME lending activities. Therefore, the banks request for strengthening their SME lending capacity at the appraisal stage was achieved through the PFI staff training provided. 59. For SMEs, the performance is more nuanced. A total of 21,212 SMEs (236 percent of revised target achieved) have attended training sessions organized by PFIs and the Ceylon Chamber of Commerce. The delivery of these training sessions was heterogeneous: some PFIs reached out to regional SME associations, a few outsourced the training to IFC s authorized Business Edge trainers, and others delegated it to their branches. The content, focus, and length of the training, as well as the characteristics of SME participants (gender-focused, existing versus potential clients) varied substantially. Because the M&E framework had no qualitative measure related to the training, the extent to which these training outputs translated into meaningful outcomes cannot be fully evidenced. Additional data collected through a PFI survey for the purpose of the ICR yields the following insights: (a) more than 20 percent of SMEs trained by PFIs were not clients of the PFI at the time of the first training and (b) 4 percent of SMEs trained by PFIs were able to secure a loan after the training. The heterogeneity was confirmed because one of the PFIs, which had a very focused training curriculum and delivery mechanism, achieved 50 percent on both indicators (see box 1). At the time of appraisal, the banks did not have much appetite to fund new SMEs given that they considered the SME segment to be riskier. Therefore, the training of SMEs was requested by the banks to make their potential clients more bankable. This was demonstrated by the fact that 705 SMEs, who were trained under the TA facility, had borrowed after the training (see annex 3 for details). 19

31 Box 1. Beyond the Results Framework Regional Development Bank: Establishing Itself as an SME Bank At the launch of the project, the merger of six regional banks into the RDB was being completed and the late integration of their IT systems resulted in a delayed start of the RDB in the project. The RDB used the opportunity of the LOC and TA to develop its SME lending capacity and established 19 centers dedicated to SMEs in The six regional banks initially had clients below the SME level, and training for staff aimed to develop their appraisal skills for this new type of clientele. With regard to SMEs, the RDB used the training sessions to build long-term relationships with SMEs: providing support for loan applications, reviewing feasibility studies, providing tailored assistance to startups, or helping SMEs with environmental standards (for example, guidance on obtaining an EPL license). People s Bank: Integrating the SMEDeF Project with the Bank s Strategy SMEs were clearly recognized as a priority in the PB Strategic Plan, the implementation of which the project supported. In particular, the PB set up a dedicated SME unit at the time that the project was implemented. The project then financed the development of an SME strategy and the development of in-house training modules (including credit appraisal and NPL management). A total of 23 SME centers have been established since 2011, some with direct financing from the project. SME centers were defined by the PB as dedicated business management offices located in all their zonal and regional offices to provide business development services, financing facilities, and business advice to SME customers. SME centers were given, among others, the following functions: identifying entrepreneurs that belong to the SME category in the area covered, providing entrepreneurial skills required for SMEs before the disbursement of any loans, regularly following up after disbursement of any financial facility and addressing issues encountered by SMEs if they are within the bank s purview or directing SMEs to institutions that are better able to assist (for example, Chamber of Commerce and Industrial Development Board). Since June 2011, SME centers have granted more than 3,000 loans for a total amount of LKR 10 billion (US$77 million) while keeping NPL levels low. Sampath Bank (SMPB): Integrating TA and LOC In line with its strategy of strengthening its SME portfolio, SMPB was among the few PFIs that seized the opportunity of the SME training to attract a new SME clientele and improve the skills of its existing SME clients. SMPB thus partnered with specialized associations such as the National Enterprise Development Authority or the women Chamber of Commerce in Colombo to provide training. SMPB had a very specific gender focus for the training (about 40 percent of entrepreneurs trained are women) and sought to design the curriculum of womenspecific training sessions to address their specific challenges with the goal of increasing financial inclusion for women. By the end of the project, SMPB had trained more than 650 SMEs, about half of which were not clients at the time of the first training session attended. Also, half of the SME beneficiaries received a loan under the SMEDeF Project after completing some training delivered through SMPB. In general, the SME training sessions were quite focused and included topics such as business plan preparation, financial record-keeping, foreign currency training, accounts opening, leasing and factoring, e-remittance, and trade-related businesses. 60. By the time of project restructuring, more than 4,000 SMEs and 4,000 bank staff had been trained. The target of 400 loan officers trained on SME banking procedures and credit risk management was, thus, far exceeded. At the time, there was no target on the number of SMEs trained. The original indicator related to SME training (measuring the impact of training for SMEs in terms of the percentage of SMEs that get a loan afterwards) was replaced at the time of restructuring by a new indicator that tracked the number of SMEs attending training. The actual value of the indicator was not updated and remained 0 percent (against a target of 50 percent). The monitoring of this indicator was found to be difficult for the PFI. It is worth noting that the restructuring team felt that the quality and effectiveness of the training needed to be strengthened. It was, then, recommended that the current capacity-building mechanism be 20

32 refocused to ensure that PFIs continue to undertake TA on a more narrow scope of activities that will include specialized TA to improve their processes as well as training to SMEs on relevant subject areas. In addition, the PIU will select suitable service providers who can directly train the SMEs in areas of need. That was successfully achieved through the Ceylon Chamber of Commerce. 61. The project also aimed to have a long-term impact on SME lending in the country through the creation of the RSF and support to improve the business environment for SMEs. By project closing, the RSF was still not operational. Given the crisis-response nature of the project, preparation time was fast-tracked (that is, concept to approval of the project took only six months) and legal authorities on the client s side may not have had adequate time to carry out an in-depth and comprehensive assessment of the innovative RSF instrument proposed. This situation might have been aggravated by the frequent changes of officers handling the RSF legal assessment. The minor subcomponent related to support to strengthen the enabling environment for SME banking was also dropped because the same activity was being undertaken by other donors. So, no result indicator was associated with either activity after project restructuring. Before restructuring, it was anticipated that 1,000 new SME borrowers would get access to finance through the RSF (0 percent achieved) and that recommendations of a diagnostic study of constraints to SMEs would be shared with relevant authorities (not achieved). These activities represented only a quarter of total project funds before restructuring and 0 percent afterwards. 62. The project provided essential term funding to SMEs to help them weather the crisis. SMEs were affected by the crisis in several ways, either through higher rates, higher rejection rates, or lower loan sizes, because of their worsened credit profile or liquidity constraint. The project aimed to provide funding to viable firms at reasonable cost and through the provision of term finance, allow them to make productive investments. The project left it to the PFIs to identify firms that have been affected by the crisis but whose growth prospects were solid. The Bank team further indicated to the PIU and PFIs that the project funds could be used to grant new loans to SMEs that had their loans restructured during the crisis Efficiency Rating: High 63. The project s efficiency is rated High based on (a) the comparison of the expected net present value (NPV) and internal rate of return (IRR) for the project at approval and at closing; (b) the actual versus expected employment generation; and (c) the substantial efficiency gains realized during project implementation (see annex 3 for further details). Benefits of the LOC and TA for PFIs. A quantitative costs and benefits analysis was possible for the LOC component and its directly associated TA to PFIs. With an overall LOC of US$50 million and TA of about US$500,000 in the updated model, the project generates a 30 percent IRR and an NPV of US$191,000. Given the high fluctuation in interest rates during project implementation (with SME rates up to 19 percent midway through implementation), the IRR would have been even higher if the model had assumed an average on-lending interest rate higher than 15 percent (higher spreads for PFIs). 21

33 Benefits of the LOC and TA to SMEs o o Outreach. Based on the analysis in the PAD, the project would have been expected, at the time of the ICR, to have 918 new SME borrower accounts owing to the US$50 million disbursed, using the actual average loan size of US$61,000. There were already 824 new SME loan accounts through the LOC, and the PFI survey reveals that 705 SMEs got loans after their participation in a PFI training. While it is not known which share of those 705 borrowers had been already counted in the 824, one can assume that most of them were firsttime clients. Assuming that share to be 50 percent, one can estimate that more than 1,150 SME loan accounts were opened as a consequence of the project. The total number of SMEs reached through financing in TA was 22,036, of which 275 were new SME clients and almost 4,000 were SMEs with which PFIs built new relationships through training (705 of these received loans). Employment generation. The PAD assumed that an average of 22 additional jobs per SME would be created because of the project. By project closing, based on limited data available in the PFI survey, the total number of new jobs created by SME loan beneficiaries was found to be 6,456 for 824 loans disbursed (different from number of SMEs, possibly much higher, if there are repeat borrowers). The only inference one can make is that for every loan disbursed, the SME created, on average, 8 new jobs. Unfortunately, data availability issues do not allow for a full comparison between appraisal estimates and project closing actuals. Benefits of the RSF. The RSF was never operational during project implementation. No benefits and costs were generated. Operational efficiency. The actual project implementation cost was US$0.4 million, against an estimate of US$4 million at approval, and US$3 million at project restructuring. These costs refer to the PIU s operational costs to implement both the LOC and TA components. In addition, the unit cost of delivery of the TA was LKR 5,360 (US$41) per staff trained and LKR 1,330 (US$10) per SME. At the time of restructuring, the estimates were US$250 per staff (US$2 million for 8,000) and US$222 per SME (US$2 million per 9,000). Significant efficiency gains have thus been achieved Justification of Overall Outcome Rating Rating: Satisfactory 64. Despite a few implementation issues, the project was able to exceed the majority of its targets. It achieved a noticeable impact on improving access to finance for SMEs through capacity building of PFIs and SMEs Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development 22

34 65. The project ensured equitable and fair access to finance for all eligible SMEs operating in Sri Lanka. Up to 68 percent of total loans by number and 59 percent by volume were disbursed to SMEs operating outside the western province. More specifically, 5 percent of loans by volume have been disbursed in the conflict-affected northern and eastern provinces. In addition, the PFI survey shows that 10 percent of loans (87) were granted to women entrepreneurs. Finally, through its requirements, the project also raised awareness on safeguards issues and labor regulations. (b) Institutional Change/Strengthening 66. As described in section 3.2, including in box 1, the long-term capacity enhancement and institutional development of PFIs is uncontested. In addition, all banks have substantially grown their own SME portfolio during the life of the project while being able to decrease their NPL ratio (see table 7.2 in annex 3 related to the borrower s ICR report). This suggests an improved capacity to lend to SMEs. In 2015, the RDB signed a credit line with the EIB, which was fully committed in just six months, because of the experience of the SMEDeF credit line. Several PFIs have expressed their interest in participating in the ADB credit line operation, based on their experience in the SMEDeF Project. PFIs have also acquired a good understanding of their business needs, SME financing needs and skill gaps, and the continuous training required to sustainably increase their lending portfolio. (c) Other Unintended Outcomes and Impacts (positive or negative) 67. The SIA report published in 2014 provided the following insights into the project s unintended outcomes: (a) the SMEDeF beneficiaries ranged from startups to more established SMEs, located throughout the island to finance various business needs and (b) the focus of the project on social impact, environmental safeguards, and labor regulation was strongly felt by beneficiaries, which were, on average, more informed about such responsibilities than nonbeneficiaries. However, a few project beneficiaries received financing at variable interest rate without understanding its consequences and anticipating the impact of higher rates of their future loan repayments. This showed the importance of providing financial education training to SMEs, which was subsequently undertaken by the PIU through the Chamber of Commerce Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops Not applicable. 4. Assessment of Risk to Development Outcome Rating: Moderate 68. With the country s more stable political and economic environment, the GoSL s commitment to SME development and finance, and the profound institutional changes achieved in most PFIs, the likelihood that development outcomes may be reversed is Moderate. PFIs have also informed the ICR team that they will continue engaging with SMEs through focused, short, but regular, training sessions. With the US$100 million ADB credit line and other initiatives by the GoSL, SME finance will likely continue to develop. Unfortunately, none of the ongoing projects will pursue the agenda of building the capacity of banks and other financial institutions to better serve SMEs. 23

35 5. Assessment of Bank and Borrower Performance 5.1. Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Satisfactory 69. The project was highly consistent with country priorities and the Bank strategy at the time. The design of the project responded to the crisis by providing much-needed financing for SMEs to sustain and develop their activities during the crisis. Project preparation was short (Concept Note review meeting in March and Bank approval in June) but did not compromise on the design to address SME constraints. At the design stage, the RSF was rightly considered because it was an innovative and potentially useful instrument that could have further strengthened SME financing. The design of the RSF followed best practice at the time for such facilities. The demand for the RSF during the preparation phase was broad based, receiving considerable interest from the potential partner financial institutions as well as high government commitment. The concerns from financial institutions expressed during project preparation were taken into account by entrusting the management of the facility not to the CBSL but to an independent agency (SLIC). The flexibility of design proved useful during implementation, when demand for a second LOC could be accommodated. The decision to require all PFIs to sign TA agreements as a condition for participation in the LOC has substantially contributed to increasing the impact of the project on SME finance in Sri Lanka despite the limited size of the project. In addition, the risks to project implementation that have been identified during project preparation were appropriately mitigated by the team, including through the use of legal covenants. 70. There were minor shortcomings in design. The definition of term finance was inconsistent between the PAD and FA. The low demand for the RSF was not clearly identified as a risk despite the known experience of the previous RSF (the first aide memoire, before project launch, already mentioned the possible limited interest of the RSF as an issue). While the M&E framework was well defined and connected to project outcomes and components, M&E arrangements were limited in both the PAD and OM. This did not, however, affect the reliability of data for the results framework. (b) Quality of Supervision Rating: Satisfactory 71. The continuity of the project team, and its composition (field presence) were significant assets for project implementation. Supervision missions (7) were regularly undertaken between September 2010 and September Given the lack of prior experience of the PIU and almost all PFIs in Bank project requirements, including environmental safeguards requirements, the Bank team had regular informal sessions to address implementation bottlenecks. To address M&E shortcomings, the Bank team recruited an experienced consultant who worked with the PIU and followed up with PFIs to obtain additional information on the projects, including through the PFI survey. The Bank s decision to rely on country-based commercial practices for fiduciary and safeguards policies is positive because it imposes less burden on beneficiaries. The team was also proactive in addressing most implementation issues (including adjusting interest 24

36 rate to reflect market conditions and reorienting SME training toward the end of the project), was extremely candid in all supervision mission documents, was prompt to identify bottlenecks, and provided comprehensive updates on project outcomes. The Bank team also raised in the aide memoires, management letters, and ISRs important implementation issues such as meetings of the committee, absence of an advisor, and delay in the preparation of the SIA. 72. A project restructuring was initiated once the final opinion of the AGO on the RSF was received, in response to the significant delays in the implementation of the RSF since approval. The restructuring was done in June 2013, based solely on the need to review the design of the RSF. The project s midterm review (MTR) was undertaken in September 2013, that is, three months after restructuring and 33 months after project effectiveness (as opposed to 18 months as required in the FA). The Bank team could have seized the opportunity of the MTR to comprehensively review the design of all components if the MTR had been carried out in early 2013 when the need to restructure the project was already agreed upon. This could have influenced the project restructuring by emphasizing the complementarity of TA (with the new LOC and RSF in the restructured project) or by holding consultations to find incentives for the utilization of the RSF. Despite this minor shortcoming, the MTR was instrumental in putting more focus on the implementation of the TA component and the need to ensure maximum impact for the use of funds. PFIs reacted to this by undertaking different delivery models for the training of SMEs, some more effective than others. (c) Justification of Rating for Overall Bank Performance Rating: Satisfactory 73. Based on the ICR guidelines, the appropriate combined rating is Satisfactory Borrower Performance (a) Government Performance Rating: Moderately Satisfactory 74. The GoSL, throughout implementation, has set the development of the SME sector as a key priority. This commitment culminated with the preparation of the national SME strategy and the recent budgetary announcements on SME finance issues. At the time of project restructuring, the government also agreed to finance the RSF from its own resources and fully transfer the responsibility of its implementation from the PIU to the MoFP at project completion. The MoFP demonstrated the Government s highest commitment by obtaining special cabinet approval to allocate dedicated funds for the RSF liability beyond the project period (Cabinet Paper No. 10\2143\404\116 of September 8, 2010) to develop a financial instrument to mitigate future risk in SME lending. Finally, the government assigned qualified civil servants to the PIU. 75. There were, however, important shortcomings that affected the implementation of the RSF. First, the AGO s legal opinion on the RSF was severely delayed and hindered the launch of the RSF for two full years, which threatened project outcomes. The Bank team could not have anticipated that the final decision of the AGO, rendered at the end of 2012, would go against project design as negotiated and approved by the Board. Second, the PSC, although created on paper, never met until August 2015, one month from project closing. Changes in the committee 25

37 membership were also not promptly undertaken; thus, at times, the committee did not have full membership. Almost all mission aide memoires raised the lack of formal committee meetings, as required in the FA. The committee during project implementation, especially in the first years of implementation, could have played an essential role in an early resolution of key issues such as the RSF. 76. While these shortcomings were major, they only affected the RSF implementation, which represented only one of three project activities. The rating of the component is thus Moderately Satisfactory. (b) Implementing Agency or Agencies Performance Rating: Moderately Satisfactory 77. Staffing. The PIU, housed by the MoFP, comprised an extended team of specialists to implement the project. Key positions for the implementation of fiduciary and environmental requirements were filled on time and adequately. The project director and deputy project director were assisted by an SME banking advisor for technical guidance during the first two years of the project. Unfortunately, both the project director and his deputy still held concomitant full-time positions, which interfered with their proactivity under the project. The PAD originally required a part-time director and full-time deputy. By mid-term review, as the advisor s term ended, the SME advisor position was left vacant. His absence adversely affected the PIU s ability to implement a focused and effective TA component. The absence of a social development officer, a position that was vacant since January 2012, also explains the substantial delay in the preparation of the SIA. The SIA, an annual requirement, was prepared only once during the life of the project, in Implementation readiness. The project had a slow start despite its crisis-response nature. After the project was approved in September 2010, the selection process for PFIs was started with a request for proposals in mid-december and completed in March The project was then formally launched in April M&E. The PIU carefully monitored project outcomes as required for the results framework. The system put in place was however limited in its ability to provide timely information on project disbursements. The PIU depended on quarterly CBSL data to be able to provide disbursement data for the LOC and thus could only provide no-objection information on loan applications submitted at the PIU. 80. The rating is Moderately Satisfactory because the shortcomings identified had a moderate effect on the implementation of the overall project. (c) Justification of Rating for Overall Borrower Performance Rating: Moderately Satisfactory 81. Based on the ICR guidelines and the overall satisfactory outcome, the overall borrower performance is rated Moderately Satisfactory. 26

38 6. Lessons Learned 82. Credit line operations can achieve a higher, more sustainable impact on improving access to finance when TA and financing go hand in hand. By addressing demand- and supply-side constraints to SME finance, the project ensured that SMEs will continue to receive finance for their growth. After project completion, given that funds were not revolving and banks already made quarterly repayments to the GoSL, the longer-lasting effect of the project was the institutional changes in PFIs as well as the newly acquired skills that resulted from the project. 83. In introducing an RSF, clarification of all legal ramifications is crucial. In most countries, the legal authorities may lack specialized financial knowledge for assessing new financial instruments such as RSF if the legal implications are not very clear from the outset. Thus, extensive consultations with all stakeholders, especially the legal authorities, at the design stage is critical because misinterpretation of legal ramifications and, at times, experience may result in a negative response. It is, therefore, vital that the technicalities are clearly discussed and agreed upon before negotiations. 84. Flexibility in project implementation and design parameters for a financial intermediary loan can be key for successful outcome. The project s decision to remain flexible and adjust the parameters of the LOC to prevailing market conditions was critical to maintain a high demand from PFIs and SMEs. At the same time, too much flexibility in the delivery of the training to SMEs yielded mixed results. In the SMEDeF Project, the quality and focus of training sessions has been unequal because the number of entrepreneurs trained was easier to monitor. Clear communication of project objectives should ensure that quality is not compromised. 85. A key success factor for the implementation of RSFs is the ownership of PFIs. RSFs are attractive financial instruments for the provision of relatively cheaper and longer-term financing. As experienced in the project, before and after project restructuring, getting the design right is an essential but not sufficient condition for successful implementation of the RSF. While PFIs were initially interested in the instrument, their previous unsuccessful experience meant that the cost of trying it again outweighed the potential benefits. It is essential for future operations with RSFs to clearly communicate mechanisms, criteria, and conditions; provide incentives as appropriate to outweigh past negative experiences; and engage stakeholders on a regular basis. 86. Steering committees have an important role to play in project implementation. Steering committees are not required for all projects, but they should be deeply rooted in implementation arrangements for projects that require high-level decisions to be taken. It is then essential that project requirements are effective and do not become mere formalities. In the SMEDeF Project, the existence of the steering committee was never put to use, although the condition of having a steering committee was met. An effective steering committee could have helped escalate issues such as the RSF that required urgent attention at a higher level. For effectiveness in action, a steering committee composition should be diversified to ensure that project stakeholders are represented at any point in time. 27

39 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies Borrower s performance was rated satisfactory in the Borrower s ICR (see Annex 7) and moderately satisfactory (MS) in the Bank s ICR. The Borrower s self-assessment put an emphasis on the outputs achieved which are acknowledged and captured in this document. However, the MS rating in the Bank s ICR is justified by: (a) the project s relatively delayed start, (b) the delayed opinion of the AGO which hindered the implementation of the RSF (although the design of the RSF was in the negotiated documents), (c) the understaffed PIU which reduced its efficiency (the Deputy Project Director was only part-time rather than full time as originally expected; the SME Banking expert position was only filled during half of the project s life); and the ineffective project steering committee. (b) Cofinanciers Not applicable. (c) Other partners and stakeholders Not applicable. 28

40 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Components Appraisal Estimate Actual/Latest Estimate Percentage of (USD millions) (USD millions) Appraisal 1 - Financing and Risk Sharing Facility Policy and Capacity Enhancement for SME Banking Unallocated Total Baseline Cost Physical Contingencies Price Contingencies Total Project Costs Project Preparation Fund Front-end fee IBRD Total Financing Required (b) Financing Source of Funds Type of Cofinancing Appraisal Estimate (USD millions) Actual/Latest Estimate (USD millions) Percentage of Appraisal Borrower International Development Association

41 PDO Level Results Indicators Cumulative total number of active SME loans Cumulative number of active SME loans (>36 months maturity) Cumulative total volume of loans to SMEs in US$, million Cumulative volume of loans to SMEs (>36 months maturity) in US$, million PFI SMEDeF project portfolio quality (NPL [%] of more than 90 days) Annex 2. Outputs by Component Table 2.1. Results Framework Baseline Original Target Revised Target Value at End of project , Intermediate Result (Component 1): Component 1: Financing and Risk Sharing Facility Volume of Bank Support: LOC - SME Number of new SME borrowers under RSF portfolio (cumulative)* ,000 Intermediate Result (Component 2): Component 2: Policy and Capacity Enhancement for SME Banking Cumulative number of PFIs introducing processes/products/management systems specific to SMEs Cumulative percentage of SMEs getting loan after training by PFIs* Number of SMEs trained** 0 n.a. 9,000 21,212 Cumulative number of loan officers trained in SME banking procedures and credit risk management Completion of findings of the diagnostic affecting SMEs and recommendation to address them made available* ,000 12,652 0 Recommendations drafted and shared with relevant authorities for action Note: *Indicators dropped at restructuring and no longer tracked. Latest value achieved was 0 ; **Indicator introduced at the time of restructuring; All PFIs introduced new products, processes, or systems dedicated to their SME business (see core text). These include the development of a management information system (MIS) and credit scoring, upgrade of credit evaluation system to integrate environmental and social standards, development of e- learning modules, and development of an M&E system. 30

42 PFI Allocation Table 2.2. Outputs of Component 1 - LOC (LOC1 + LOC2) Commitment Commitment as a Share of Allocation Disbursement Disbursement as a Share of Commitment LKR, LKR, Number LKR, Number of % million million of Loans million Loans % BOC COM DFCC HNB NDB PB 1, , , RDB SMPB 1, , , TOTAL 6, , , Sector Total Disbursement Number of Loans Value (LKR, million) Agriculture Agro processing 139 1, Apparel and garments Beauty culture/dressmaking Carpentry, furniture, and lathwork Construction Educational activities Electricity and water supply Fisheries Floriculture Hotels and restaurants and tourism 157 1, High-profile seeds development IT/software development Jewelry and gem Livestock and dairy Manufacturing Manufacturing - gift items and craft Mining and quarrying Ornamental fish Packing and packing industry Services Spices and minor export crops Transport and communications Wholesale and retail trade Total 824 6,

43 Disbursed Loans by Size PFI Loan Size (LKR, million) BOC COM DFCC HNB NDB PB RDB SMPB Total Value (LKR, million) 1, , , , District Total Disbursement Number of Loans Value (LKR, million) Central Province Kandy Matale Nuwara Eliya Eastern Province Ampara Baticaloa Trincomalee North Central Province Anuradhapura Polonnaruwa North Western Province Kurunagala Puttalam Northern Province Jaffna Kilinochchi Vavuniya Sabaragamuwa Province Kegalle Ratnapura Southern Province Galle Hambantota Matara Uva Province Badulla Monaragala Western Province Colombo 109 1, Gampaha Kalutara Total 824 6,

44 Component 2: Policy and Capacity Enhancement for SME Banking TA Financed by the Project for PFIs (excluding Training) This is a list of outputs cofinanced by the project for participating banks: Development of an MIS for the SME Department Development of a refinancing and M&E system for the SME unit Strengthening the capacity of the SME unit Preparation of an SME strategy Improving risk management tools Development of e-learning modules Table 2.3. Training for PFI and SMEs PFI Allocation Amount Reimbursed (LKR, million) Number of Trainees US$ PFI training SME training Total PFI training SME training Total BOC 84, ,268 3,222 5,490 COM 19, ,420 2,852 DFCC 114, ,008 2,768 5,776 HNB 40, ,341 1,560 NDB 193, ,804 4,899 7,703 PB 156, ,206 1,716 2,922 RDB 175, ,837 1,568 3,405 SMPB 44, ,532 Total 828, ,652 18,588 31,240 Administered by Amount Reimbursed (LKR, million) Number of Trainees PIU ,624 33

45 Annex 3. Economic and Financial Analysis 1. This annex follows the same structure as the one prepared in the PAD, for ease of reference. Benefits of the LOC and TA to PFIs 2. Main assumptions and analysis are provided as follows: Key Assumptions PAD There is a 66-month repayment from the PFI to the GoSL, made at the beginning of each year. ICR Repayments are made by the PFI on a quarterly basis (both principal and interest). However, the PFIs begin and finish repaying the LOC to the GoSL at the same frequency as they receive loan repayments from SMEs. Thus, for model simplification purpose, we assume that the same rate is charged to all projects with an average maturity of five years and a one-year grace period. 9% discount rate based on the CBSL s estimate Unchanged (for comparison purpose) 6% discount rate for the LOC cost to the PFI based on the current Average Weighted Deposit Rate(AWDR) of 6% Credit losses of 3% are based on discussions with banks at pre-appraisal mission. Minimum 1% provision costs are based on the CBSL requirements. Overhead costs are based on an assessment of the banking system conducted during project preparation. The total value of TA to the PFI of US$3 million equivalent is based on project Subcomponent 2b, which has allocated US$7 million for TA for both the LOC and RSF. Over , the average AWDR was 7.78% (Source: CBSL - Economic Developments). The SMEDEF portfolio NPL was 3.37%. Unchanged. (for comparison purpose) Unchanged. (for comparison purpose) A total amount of LKR 68 million (US$0.52 million) was reimbursed to the PFIs for TA. The model was adjusted to reflect that the project only contributes 50% of TA needs. Year-end exchange rates have been used to convert actual project expenditures. Actual LOC and TA disbursements have been used as well. In an improvement to the original model, an additional benefit to PFIs was quantified; the additional profit that the PFIs generate owing to participation in the project. Based on the PFI survey, we estimated that 35% new business is generated from the LOC under management. The margin on that new business is lower than through the LOC but estimated at 200 bps above the cost of funding (from the PFI s own resources). A quantitative costs and benefits analysis is possible for the LOC component and its directly associated TA to the PFIs. With an overall LOC of US$50 million and TA of about US$520,000 34

46 in the updated model, the project generates a 30 percent IRR and an NPV of US$191,000. This is compared to an IRR of 21 percent and an NPV of US$492,000 that was expected at the time of project approval (LOC of US$28 million and TA of US$7 million) and an IRR of 25 percent and an NPV of US$1 million at the time of project restructuring (total LOC of US$50.4 million and TA of US$5 million). For the model in the ICR, the relatively lower NPV is because of (a) the difference between the actual and estimated annual flow of funds to PFIs and (b) the repayment method of the LOC for PFIs (which means that the funds are returned to the government instead of revolving in their books for further lending to SMEs). Given the high fluctuation in interest rates with SME rates up to 19 percent midway through implementation, the IRR would have been even higher if the model had assumed an average on-lending interest rate higher than 15 percent (higher spreads for PFIs) applied to the model. Figure 3.1. Economic analysis: model and results Item (values in,000s USD) Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year LOC On Lending (Avg Balance) 4,377 14,419 4,680 9,937 16,207 outstanding balance at year end 4,377 17,921 18,841 24,083 33,607 23,683 14,635 8,470 3,241 Overhead Costs 2.00% (88) (358) (377) (482) (672) (474) (293) (169) (65) 0 Cost to PFIs of TA (193) (201) (140) (351) (167) LOC Cost of Funds to PFIs 7.78% (341) (1394) (1466) (1874) (2615) (1843) (1139) (659) (252) 0 Avg Rate on LOC On Lending 15.00% 657 2,688 2,826 3,612 5,041 3,552 2,195 1, Minimum 1% Provision Costs 1.00% (44) (179) (188) (241) (336) (237) (146) (85) (32) 0 Credit Losses On LOC On Lending 3.37% (148) (604) (635) (812) (1,133) (798) (493) (285) (109) 0 Profit stemming frm new business generated through partification in LOC 2.00% Net Benefit/ (Cost) (138) 9 39 (106) Discount Rate 9.00% NPV 191 IRR 30% Benefits of the LOC and TA to the SMEs The benefits identified in the PAD for the SMEs were increased outreach and employment generation. Outreach. The analysis in the PAD estimated that the original US$35million (US$28 million LOC and US$7 million total TA) will help the PFIs increase their outreach to SMEs an incremental 1,400 active SME borrower accounts will be added to their portfolio, assuming an average loan size of US$50,000. If the same analysis had been undertaken at the time of project restructuring, the estimated number of new borrower accounts would have been 2,176 for US$54.4 million disbursed. Using the same ratio, at the time of the ICR, we would have expected 918 new SME borrower accounts attracted by the PFIs owing to the US$50.4 million disbursed and the actual average loan size of US$61,000 (numbers were actually calculated based on local currency amounts to eliminate exchange rate effects). There were already 824 new SME loan accounts through the LOC, and the PFI 35

47 survey reveals that 705 SMEs got loans after their participation in a PFI training. While we do not know which share of those 705 borrowers were counted in the 824, we can assume that most of them were first-time clients. Assuming that share to be 50 percent, we can estimate that more than 1,150 SME loan accounts were opened as a consequence of the project. The total number of SMEs reached through financing in TA was 22,036, of which 275 were new SME clients and almost 4,000 were SMEs with which PFIs built new relationships through training (705 of these received loans). Employment generation. The PAD assumed that an average of 22 additional jobs will be created by the firm. By project closing, based on limited data available in the PFI survey, we found that the total number of new jobs created by the SME loan beneficiaries is 6,456 for 824 loans disbursed (different from the number of SMEs, possibly much higher, if there are repeat borrowers). The only inference we can make is that for every loan disbursed, the SME created, on average, eight new jobs. Unfortunately, data availability issues do not allow for a full comparison between appraisal estimates and project closing actuals. Benefits of the RSF The RSF was never operational during project implementation. No benefits and costs were generated. Operational Efficiency The actual project implementation costs were US$0.4 million, against an estimate of US$4 million at approval and US$3 million at project restructuring. These costs refer to the PIU s operational costs to implement both the LOC and TA components. The unit cost of delivery of the TA was LKR 5,360 (US$41) per staff trained and LKR 1,590 (US$10) per SME. At the time of restructuring, the estimates were US$250 per staff (US$2 million for 8,000) and US$222 per SME (US$2 million per 9,000). Significant efficiency gains have thus been achieved. 36

48 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team Members Names Title Unit Responsibility/ Specialty Lending/Grant Preparation Cecile Thioro Niang Program Leader GFMDR Task Team Leader (TTL) Ann Christine Rennie Consultant GFMDR Asta Olesen Senior Social Development Specialist SFGSP Aza Rashid Program Assistant GFMDR Chau-Ching Shen Senior Finance Officer CTRFC Darshani De Silva Senior Environmental Specialist GENDR Ina Hoxha Extended Term Consultant Joyce Miriam Denise Witana Procurement Specialist GGODR Lohita Karunasekera Private Sector Development Specialist GTCDR Minneh M. Kane LEGES Miriam Witana Procurement Specialist Nadeera Rajapakse Consultant GENDR Sameena Dost Senior Counsel LEGES Sashikala Jeyaraj Program Assistant GTCDR Sriyani M. Hulugalle Senior Economist GTCDR Sunil Chandrasiri Consultant Supul Chamikara Wijesinghe Financial Management Specialist GGODR Susan Maslen Senior Counsel LEGOP Thao Le Nguyen Senior Finance Officer W.D. Premachandra Consultant Supervision/ICR Cecile Thioro Niang Program Leader GFMDR TTL Korotoumou Ouattara Senior Financial Economist GFMDR TTL Aminata Ndiaye Young Professional GFMDR ICR TTL Aza A. Rashid Program Assistant GFMDR Bernadeen Enoka Wijegunawardene Senior Financial Management Specialist GGODR Darshani De Silva Senior Environmental Specialist GENDR Anjali U. Perera Vitharanage Procurement Specialist GGODR Henry Bagazonzya Head GFMDR John P. Byamukama Financial Analyst GFMDR Lashantha H. Jayawardhana Local Consultant ST GENDR Lohitha Karunasekera Private Sector Development Specialist GTCDR Minneh Mary Kane Lead Counsel LEGES Mokshana Nerandika Wijeyeratne Environmental Specialist GENDR Sandya Kumari Salgado Sr External Affairs Off Sashikala Krishani Jeyaraj Program Assistant GTCDR Sriyani M. Hulugalle Senior Economist GTCDR Susrutha Pradeep Goonesekera Consultant GSURR Wellana Gamage Thungasiri Local Consultant ST GFMDR 37

49 (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle US$, thousands (including Travel Number of Staff Weeks and Consultant Costs) Lending FY , FY , Total: , Supervision/ICR FY , FY , FY , FY , FY , FY , Total: ,

50 Annex 5. Beneficiary Survey Results Not applicable. 39

51 Annex 6. Stakeholder Workshop Report and Results Not applicable. 40

52 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR 1. Objective of the Project: The general Project Development Objective of the Small Medium Enterprise Development Facility Project (SMEDef) is to improve access to finance (including term finance) for SMEs affected by the Global Financial Crisis in Sri Lanka and new entrepreneurs who are willing to doing business by utilizing new economic opportunities created in the economy as a result of the termination of the thirty-year long conflict. The specific objectives of this project are (a) provide medium to long term credit for SMEs; (b) reduce the banks risk of lending to SME borrowers by implementing Risk Sharing Facility (RSF)(c) enhance the capacity of lending institutions and SMEs and (d)support to strengthen the enabling environment for small and medium enterprise financing. 2. Components of the Project: The project consists two components before restructuring the project. Component 1: Financing and Risk Sharing Facility - (US$41 million equivalent) There are two sub-components in the component 1 i.e. the line of credit and risk sharing facility (a) Line of credit - (US$28 million equivalent): World Bank provides a credit line to the Government of Sri Lanka for on-lending to participating licensed commercial and specialized banks on competitive interest rates to refinance short term and long term SME loans (US$28 million equivalent): This refinancing facility will provide loans with a maximum maturity of 10 years including a maximum grace period of up to 2 years. (b) Risk Sharing Facility (RSF) - (US$13 million equivalent): The purpose of the RSF is reducing the banks risk of lending to SME borrowers (US$13 million equivalent). This facility make an additional incentive for commercial banks to lend to SMEs cover the commercial lenders losses in cases of defaults by SME borrowers. It was planned to issue the RSF by Sri Lanka Insurance Company (SLIC) on a 50/50 pari-passu basis with PFIs. However, RSF was not implemented due to the legal barriers. Therefore, the project was restructured and reallocated that amount to the credit line on 2 nd August Component 2: Policy and Capacity Enhancement for SME Banking (US$12 million) (a) Capacity enhancement for lending institutions and SMEs (US$7 million equivalent): In partnership with the IFC and other development agencies, the project provides technical assistance on a matching basis to the commercial banks benefitting from the credit line or RSF, in order to i) help build their capacity to provide lending services to SMEs more efficiently and ii) ensure that SMEs have the capacity to effectively utilize loans from the banks for the growth of their businesses. 41

53 (b) Support to strengthen the enabling environment for small and medium enterprise financing (US$1 million equivalent): This includes undertaking a diagnostic of constraints and measures addressing such constraints. (c) Implementation support for the Project Implementation Unit (PIU) (US$4 million equivalent): This helps to building expertise to manage the project, compensating the Facility Agent to manage the RSF, undertaking a detailed impact evaluation framework for assessing the impact of the project on the targeted clients and other implementation support activities. The project was restructured on 2nd August Accordingly LOC was operated in two stages i.e. LOC1& LOC2. The LOC2 is combined with an optional RSF Facility with a funding of US$ 22.4Mn. Therefore, the project components was changed and the amount that have been allocated to some components was reallocated to other components as follows. Component 1: Financing and Risk Sharing Facility US$53.6 million (a) Line of Credit Facility (LOC1) - US$ 28 million (b) Line of Credit Facility (LOC2) and Risk Sharing Facility - US$25.60 million LOC facility combined with optional Risk Sharing Facility (RSF) to refinance short term and long term SME loans Component 2: Policy and Capacity Enhancement for SME Banking: US$3.8Million (a) Capacity enhancement for lending institutions and SMEs - US$2.6 million (b) Support to strengthen the enabling environment for SMEs - US$ 0 million (c) Support the Project Implementation Unit - US$ 1.2 million 3. Project Cost: After restructuring the project, the total project investment cost was estimated at US$57.40 million comprising the US$28 million of line of credit and US $ optional risk sharing facility. 4. Implementation Arrangements: The Ministry of Finance and Planning (MOFP) was the Executing Agency (EA) which responsible for coordinating the project activities with WB and PCBs. The Project Implementing Unit (PIU) was established under the supervision of the Department of Development Finance that responsible to implement the project. Project preparatory activities along with preparation of Project Appraisal Document were done by the PIU with the assistance from the World Bank and finalized on 08th November 2010 and the Finance Agreement between the GOSL and World Bank was signed. The effective date of the project was 6th January According to the agreement, MOFP was established the PSC comprising 8 members with the chairmanship of the 42

54 Secretary to the Treasury. Only one PSC meetings has been held on 11 th August 2015 at the later part of the project. 5. Disbursements: Government of Sri Lanka was the borrower and the disbursed through the selected eight PFIs. In December 2010, Request for Proposals (RFP) from 26 commercial and licensed specialized banks were called. 12 Banks responded for the Line of Credit (LOC) and 01 Bank opted for the Risk Sharing Facility. Out of 12, 8 banks were selected as Participating Financial Institutes (PFIs) for the LOC which fulfilled the selection criteria. The Participatory Banks were: Bank of Ceylon (BOC), People's Bank (PB), Hatton National Bank PLC (HNB), Commercial Bank of Ceylon PLC (COM), Development Finance Corporation of Ceylon PLC (DFCC), National Development Bank PLC (NDB), Sampath Bank PLC (SMPB), Pradeshiya Sanwardhana Bank (RDB). Eight Line of Credit Agreements (LOC) were signed between the GOSL and PFIs by 25th March The project was restructured and its restructuring was finalized in June 2013.The restructuring increased the amount that was allocated to the LOC and link the LOC with the RSF. Accordingly LOC was operated in two stages i.e. LOC1& LOC2. The LOC2 is combined with an optional RSF Facility with a funding of US$ 22.4Mn. The Risk Sharing facility was not implemented due to unavoidable legal circumstances. The Government made the WB loan available to the PFIs at the Average Weighted Deposit Rate (AWDR). The total loan amount of US$ million were allocated among PFIs based on their performance and approximately 98.3 out of that were disbursed by PFIs to end borrower during the 5 years period ( ). Initially, it has been planned to disburse within the 3 and half years period but the project period was extended another 18 months due to the restructuring of the project. The impress account was established for this Loan in December 5 th 2010 and the first tranche of US$ 29.8 million received from WB on 30 th June Performance of the stakeholders: The overall performance of the stakeholders can be rated as above satisfactory level since the most of targets were achieved as expected PFIs: The PFIs performance can be rated as above satisfactory. The actions taken place to ensure compliance with the anti-money laundering framework and corporate governance standards in Sri Lanka, fulfillment of the disbursement criteria and loan evaluation process were highly satisfactory. However, the progress of disbursement, the effort to create more awareness about credit line were not as expected in time. Initially, the project was scheduled for three and half year period from 2011 to mid However, it was extended up to September 30, 2015since increasing the allocated amount of the LOC due to the reallocation of funds under the restructuring of the project. At the beginning, the slow progress was achieved due to poor attention to create awareness, high interest rate, un-availability of clear demarcation about SME definition. However, the PFIs recorded a remarkable progress in latter part of the project and 43

55 finally approximately 98.5percent of the total allocation was disbursed among SMEs covering all districts (excluding Mannar & Mullaithive) of the Island. Table 7.1. Progress of LOC by PFIs Bank Allocation Disbursement (LKR, million) (LKR, million) (% Disbursed) Bank of Ceylon People s Bank 1, , HNB Commercial Bank DFCC Bank NDB Sampath Bank 1, RDB Total 6, , Source: Progress Report, 4 th Quarter 2015 People s Bank developed the SME Strategy and Hatton National Bank developed the Environmental Strategy while Sampath Bank introduced the Credit Rating module for the bank through the technical assistance component. All the PFIs have introduced at least one new product or system under the Technical Assistance, which will ensure the sustainability. Under the project, total of 12,652 staff of 08 PFIs have trained which has exceeded targeted number of 8,000 at the project designing. In addition to the staff training PFIs started and continued SME Entrepreneur Training in Financial Record Keeping, Marketing & Product Quality, Human Resource & Labour, Business Planning and environment Awareness. The total number of SME entrepreneurs trained is 18,588 while the targeted number was 9,000. PFIs have reported the success in SME training and upgrading of the SME knowledge and improvement in the SME banking. As indicated that below table 03, the all most all eight partner banks have successfully increased their SME loan portfolio during the period of All private banks except Sampath Bank has able to manage the mandatory requirement of 10 percent SMEs loan portfolio out of their total loan portfolio but the state banks failed to maintain this mandatory requirement also. Almost all banks have managed their NPL ratio less than 5 percent and further improved their NPL ration during the period by streamlining their loan recovery process. The loan recovery of the SMEDeF project is done successfully by the PFIs and 6 banks have able to mage the lower ratio than mandatory ration of 5 percent. DFCC, NDB and SMPB have able to recover 100 percent of project loan during the period. PB BOC Bank Table 7.2. Disbursement Performance of the PFIs ( ) Year Total Loan Portfolio (LKR, million) SME Loan Portfolio (LKR, million) NPL of the Bank (%) NPL of the Project (%) , , , , , , , , , ,

56 , , , , HNB , , , , , , COM , , , , , , DFCC , , , , , , NDB , , , , , , SMPB , , , , , , RDB , , , , Source: DFCC, SMPB, COM, NDB, RDB, HNB, PB, and BOC Executing Agency: In macro point of view, the Executing Agency s (EA s) performance can be ranked as above satisfactory level. EA s performance with regards to preliminary steps such as establishment of PSC, opening the impress account, issuing no objection for refinancing, releasing allocations, preparing progress reports and giving guidance to PCBs was highly satisfactory. However, the field monitoring and evaluation responsibilities that have to be performed to assess the grassroots level progress were not done timely and sufficiently. As an example, only 135 sub- projects out of total 824 sub-projects were investigated in island wide by the team deployed by the EA during the entire project period. Project Implementation Unit (PIU) which was established under the Executing Agency conducted 29 training programs in all 25 districts of the country in collaboration with the Ceylon Chamber of Commerce (CCC). This training included improvement of financial literacy, best environmental practices, cash flow management, costing, incorporating and upgrading of banking knowledge and banking habits of the SME entrepreneurs. The total number of SMEs trained were 2,624 and feedback received was positive and highly supported to sustainability of the project. 6.3 World Bank: The performance of the WB during the implementation period of the project is rated as satisfactory. Around seven (7) review missions were fielded by the WB between May 2010 and December The missions were very effective in providing technical inputs, monitoring implementation progress, and addressing specific implementation issues that arose at the implementation level. Also processing of Government s requests regarding changers to the implementation arrangements, disbursement matters and approvals for procurements, and recruiting project staff were generally done timely. However, the payments for reclaims were not timely released and average 3 or 4 replenishment claims were released per annum reclaims, since more than 3 months was taken for clearance of one reclaim. So it also caused to make slow progress of the project. 45

57 7. Output of the Project: The Sri Lankan credit market suffers from a death of medium to long-term funds. The banks and financial institutions are unwilling to lend to SMEs due to the limited availability of the matching term funds. Therefore, this project aims to fill this critical credit supply gap in the financial market by providing medium to long-term credit for SMEs, enhancing the capacity of SME banking and SMEs. The project was implemented through two main components i.e. line of credit and risk sharing facility; and policy and capacity enhancement for SME banking in order to achieve the above aims Loan Disbursement The disbursement targets of the LOC1 and LOC2 has been successfully achieved by the PFIs and approximately 98.5 percent of the credit line has been disbursed as at 30 th September Progress by PFIs: As mentioned earlier, the eight commercial banks were selected as implementing agencies of the project. As of November 2015, a total amount of US$ 53.6 million has been disbursed by the eight partner financial institutions (PFIs) to 824 SMEs. Table 7.3. Overall Progress by PFIs ( ) PFIs Number % Amount % Average Loan Size (LKR, million) Bank of Ceylon People s Bank , HNB Commercial Bank DFCC Bank NDB Sampath Bank , RDB Total , Source: Progress Report, 4 th Quarter 2015 In terms of number of loans, RDB has played a vital role by granting 209 and followed by the People s Bank and Sampath Bank by providing 166 and 128 loans respectively. In terms loan amount disbursed, Sampath Bank has played the leading role by providing a loan amount of Rs million and followed by People s Bank and Regional Development Bank. Lowest performance in both number of loan and amount granted was recorded by the DFCC Bank providing 26 number of loans and Rs mn respectively. Sampath Bank granted around 23 percent of the total amount disbursed and more than 25 percent out of total number of loans has been granted by RDB proving their ability to cater real SMEs. In terms of the average loan size, private commercial banks such as ADB, DFCC, HNB and Sampath provided large loans compare to the government bank i.e. RDB, Bank of Ceylon and People s Bank which is recorded small average loan size of 3.7 mn, 6.6 mn and 7.1 respectively. 46

58 Progress by Sectors: The entrepreneurs who are engaging business related to eligible sub sectors which approved by the project only qualified to enjoy the benefit of the credit line. 24 sub-sectors were approved for the LOC1 and LOC2. The Hotels and restaurants & Tourism dominates the both loan amount disbursed and number of loans provided and followed by the Agro Processing, Manufacturing, and services sectors respectively. These four sectors have shared around 70 percent out of total disbursement and around 64 percent out of total numbers of loans granted under the project. The balance 30 percent disbursed among other sectors Progress by District: Colombo districts the highest number of loans were given to entrepreneurs who are doing their business in the Colombo district and followed by Gampaha, Kurunegala and Kaluthra respectively. The three districts which is belong to Western Province have shared around 46 percent of total amount disbursed and around 32 percent of number of loans granted under the project. The entrepreneurs who are doing their business in the far regions are highly affected by the access to finance issue compare to the businessman in the Western Province. Therefore, it is needed to be more favorable for the regional entrepreneurs, but the very less amount out of total amount disbursed under this project granted to the far regions such as northern, Eastern and Uva Provinces Employment Generation: In generally, the employment generation is one of the main objectives of any credit line project but it is not classified as development objective of this project. As an example the SMERDP project which was implemented under ADB assistance from mid-2008 to mid-2011 was expected to generate approximately 50,000 job opportunities throughout the projects. Since the employment creation not classified as a development objective of the project, the system was not developed to collect such data by the PFIs. Therefore, the impact on the employment generation of this project can t be estimated Training Banking Staff and SMEs. The project provides technical assistance on a matching basis to the PFIs to build their capacity to provide lending services to SMEs and ensure that SMEs have the capacity to effectively utilize loans from the banks for the growth of their business by training their staff and SMEs who have obtained loans through this project. Under the project total of 12,652 staff of 08 PFIs have trained which has exceeded targeted number of 8,000 at the project designing. In addition to that 18,588 SME entrepreneurs were trained by the PFIs regarding Financial Record Keeping, Marketing & Product Quality, Human Resource &Labour, Business Planning and environment Awareness. The targeted number to be trained under this project was 9000.Accordingly, 17,000 were targeted to train and over 31,000 have been trained under this project. It is nearly 200 percent achievement against the target.. 47

59 In addition to that the Project Implementation Unit (PIU) implemented a new mechanism for training SMEs in coordination with the Ceylon Chamber of Commerce (CCC) and conducted 29 training programs in all 25 districts of the country. The total number of SMEs trained were 2,624 throughout these 29 training programs. 8. Overall Assessment: The objectives of the project was mostly parallel with the desire development targets of the Sri Lankan government since it declared the well-functioning SME sector as a back born of the economy and could take the lead in generating jobs and economic growth while raising income levels for a large segment of the population. The regional inequalities are one of the major development challenges in Sri Lanka. The per capita gross regional domestic product of the most develop Western Province is more than four time those of other poorest regions and also youth unemployment is a critical problem mostly outside the Western province. The SMEs played vital role to establish an equitable regional growth since the regional SMEs comprise 75 percent of the total number of registered SMEs in the country. The programme loan component of this project encourages the regional SMEs by providing around 60 percent of disbursed amount for SMEs who are doing their business out of the Western Province. The contribution of the western province to the GDP was 45.1 percent in 2010 and it has reduced up to 42.0 percent in year 2013 and those contribution of other province has been increased comparatively. However, it is difficult to determine the influence made by the SMEDeF project for such regional distribution since there were similar other credit schemes and several government attempts with a view to reduce the regional disparities. Simultaneously, there was a huge impact to create direct and indirect jobs throughout the project for especially for regional unemployed people. However, the exact impact of the project on employment generation also cannot be estimated since the mechanism was not developed to collect the required information within the system. In generally, every Rs. 10 million injected to finance SME sector has the capacity to create approximately 10 new formal employment opportunities. According to that estimation, approximately 7000 new employment opportunities may create through this project and the indirect jobs also may create more than that. The most of banks especially state banks still practice the traditional banking instead of the development banking which is really needed to boom the SME sector in the country. The technical assistance provided under the project for partner PFIs on a matching basis assisted to improve their capacity of staff and existing systems to move towards the development banking. On the other hand 18,588 SME entrepreneurs were trained through the project regarding Financial Record Keeping, Marketing & Product Quality, Human Resource & Labour, Business Planning and environment Awareness which is really needed to sustain in the business environment. The most critical part of the access to finance issue is unavailability of sufficient collateral that is demanded by the banks to provide a loan. In recent past, the several credit lines were implemented by the various PFIs in collaboration with the Government and using their own funds. Therefore, credit facilities were available in the market compare to the 10 years back. However, it was observed that the most of SME entrepreneurs was struggling to access the finance even though the enough credit facilities are available in the market due to this collateral 48

60 issue. Considering these facts, the Risk Sharing Facility was introduced under this project first time in the history of the credit line projects but unfortunately it was not operationalized during the project period due to some legal barriers. As a whole, the disbursement were recorded the slow progress at the beginning of the project and very boost progress has shown at the last two quarters. The data in the quarterly report of the 4 th quarter 2015 is clearly shows that around 20 percent of the total disbursement of the project was disbursed during the last quarter of the project. The reasons i.e. the lack awareness regarding the loans scheme among the entrepreneurs, interest rate that was not affordable for SMEs, and uncertainty of the business and political environment ware caused to show a low progress of disbursement at the beginning of the project. 9. Lessons Learned and Recommendations: RSF is a facility that enhancing the demand for SME lending since it can be applied as alternative solution for collateral issue and it reduce the risk of PFIs losses in case of defaults by SME borrowers. There is a huge demand for such facility from the borrower s side since it makes a comfortable platform to obtain a loan. On the other hand banks also willing to implement such facility but it was not operationalized due to regulation barriers. Therefore, it is needed to make necessary actions to remove that regulation barriers that against to implementation of RSF facility. The government general expectation of implementing a credit line project is distributing the amount available as early as possible to qualified SMEs in order to boost the economy. However, the PFIs usually given low priority at the beginning of the project for disbursement and show a rapid progress at the end of the project since they know their allocated quota at the beginning of the project and there is no competition among PFIs to obtain the allocation. Therefore, it is recommended to apply a new competitive mechanism at future to allocate the funds among PFIs instead of non-competitive quota system applied currently. 49

61 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders N/A 50

62 Annex 9. List of Supporting Documents 1. Project Preparation (i) PAD (ii) Project Concept Note Review, Quality Enhancement Review-, and Decision Meeting: Decision and Comments (iii) FA (iv) OM (LOC and RSF) 2. Project Supervision (i) Project Aide Memoires and ISRs (ii) Restructuring Paper and Amended FA (iii) SIA 07/14 and Final PFI Survey (iv) PIU Progress Reports ( ) (v) Subsidiary Loan Agreements (GoSL and PFIs) 3. Steering Committee Minutes 4. Country Partnership Strategies, World Bank 5. National SME Strategy, GoSL (2015) 6. CBSL Annual and Recent Economic Development Reports 7. Sri Lanka Economic Update, World Bank (2009) 51

63 MAP 52