69324 Fiji Microfinance Policy Review Final Draft October 2003 * This review was commissioned by the East Asia Financial Sector Group of the World Bank, Washington, D.C. The first draft of the policy review was prepared by Dr. John Conroy, Consultant to the World Bank, following a mission to Fiji in February 2003 which was led by Desiree Green of the World Bank Group. The final draft of the review has benefited from comments from the National Microfinance Unit of the Ministry of Commerce, Business Development and Investment; Robert J. Simms (SPPF); United Nations Development Fund (through the NMFU); and Desiree Green and Sameer Goyal of the World Bank. The mission team and those involved in the finalization of this report wish to express their appreciation to the Government of Fiji, in particular, the National Microfinance Unit, for its support and cooperation in this effort. Cooperation received from bilateral donor agencies, the Asian Development Bank, the United National Development Program, and the Microfinance institutions in Fiji is much appreciated and duly acknowledged. ACRONYMS AND ABBREVIATIONS ADB Asian Development Bank MIS Management Information System AMFU Aglow Microfinance Unit NCSMED National Center for Small and Microenterprise Development ANZ Australia and New Zealand Bank NFMU National Microfinance Unit ATM Automatic Teller Machine NGO Non-governmental Organizations CCSLA Cane Farmers’ Cooperative Savings and Loans Association NZODA New Zealand Official Development Assistance CGAP Consultative Group to Assist the Poorest PAR Portfolio at Risk CNB Colonial National Bank PIN Personal Identification Number CUFA Credit Union Foundation of Australia PPTA Project Preparation Technical Assistance (ADB) CU/CUL Credit Union/Credit Union League PSLP Pacific Sustainable Livelihoods F$ Fijian Dollar Program FCO Field Credit Officer RBF Reserve Bank of Fiji FCOSS Fiji Council of Social Services. S&Ls Savings and Loan Institutions FCUL Fiji Credit Union League SMILE Sustainable Microfinance & Livelihoods through Empowerment FDB Fiji Development Bank SEEDS Social and Economic Equity for the FNTC Fiji National Training Council Disadvantaged GoF Government of Fiji SME Small and Microenterprises GDP Gross Domestic Product (Government of Fiji usage) IDL Individual Lending Program (of T/A Technical Assistance NMFU) UNDP United Nations Development IHRDPEP Integrated Human Resource Program Development Program for UNOPS UN Office of Project Services Employment Promotion UNV United Nations Volunteer ILO International Labour Organization WB World Bank ME/MED Microenterprise/ Microenterprise Development WOSED Women’s Social and Economic Development MF Microfinance MFI Microfinance Institution EXCHANGE RATE $F = $0.5060 at February 2003 i Table of Contents EXECUTIVE SUMMARY ........................................................................................................... iv I. Introduction and Overview ......................................................................................................1 The Environment for Microfinance in Fiji ..................................................................................1 Microfinance in the Fijian Context ..............................................................................................2 Microfinance institutions in Fiji ..................................................................................................3 II. Sources of Support for Microfinance in Fiji ............................................................................6 Financial and Capacity-Building Assistance for Microfinance Institutions ................................9 Support for Improving the Policy and Regulatory Environment for Microfinance .......................10 III. Viability of Government Support and Conformity with Good Practice ............................12 A. Assessment of the National Microfinance Unit .................................................................12 NMFU’s “Individual Lending� Program ...............................................................................15 Village ‘Banks’ ......................................................................................................................16 Conclusion .............................................................................................................................16 B. Assessment of a government-supported MFI: Aglow, a Grameen Bank Replication .......17 Performance Summary ..........................................................................................................17 MIS and operational performance .........................................................................................18 IV. Other Microfinance Providers ...........................................................................................20 Commercial Banks .....................................................................................................................20 Fiji Development Bank ..............................................................................................................22 Savings and Loans Cooperatives ...............................................................................................23 Credit Unions .............................................................................................................................23 Informal Moneylenders .............................................................................................................24 Finance and Hire Purchase Companies .....................................................................................24 Registered Moneylenders ..........................................................................................................25 V. The Enabling Environment for Microfinance ........................................................................26 Legal Constraints on Microfinance Providers ...........................................................................26 Government Policies Relating to Microfinance.........................................................................28 Aspects of the Macroeconomic Environment Relevant to Microfinance ..................................29 Poverty in Fiji ............................................................................................................................29 Role of the Reserve Bank of Fiji ...............................................................................................30 National Centre for Small and Micro-Enterprise Development ................................................32 Conclusion .................................................................................................................................34 ii VI. Conclusions and Recommendations ..................................................................................36 Issue: Conformity of Government-Supported Microfinance Operations with Recognized Good Practice.............................................................................................................................36 Conclusions............................................................................................................................36 Recommendations: Conformity with Good Practice, and Sector Building ..........................38 Issue: Structure and Modalities of Government and Donor Support........................................40 Conclusions............................................................................................................................40 Agencies with Potential to Take Lead Roles in the Sector ....................................................40 Recommendations Concerning the Structure and Modalities of Support .............................41 Policy Framework for Microfinance in Fiji ...............................................................................42 Conclusions............................................................................................................................42 Recommendations: ................................................................................................................44 Certain positive initiatives may be taken to loosen constraints on Microfinance: ....................45 Conclusion: An Architecture for the Microfinance System in Fiji .......................................46 Tables Table 1.1: A taxonomy of Microfinance Providers in Fiji………………………………………. 4 Table 2.1: Microfinance in Fiji – Providers and Sources of Support…………………………… 7 Table 2.2: Financial Support to Microfinance Providers by NMFU…………………………… 9 Table 3.1: Capital (Loanable Funds) Budget Allocation for NMFU, 2000-2003……………… 14 Table 3.2: NMFU’s National Microfinance Outreach as at 12/31/2002………………………...15 Annexes Annex A: Basic Assessment Criteria for the NMFU: Key Indicators………………………… 48 Annex B: Microfinance Provider Mechanisms: Aglow, a Grameen Bank Replication………... 60 Annex C: Microfinance Provider Mechanisms: Individual Lending program of the NMFU…. 66 Annex D: Capacity-building Options for the Microfinance Sector…………………………….. 68 iii EXECUTIVE SUMMARY 1. The most appropriate definition of microfinance in Fiji is a broad and inclusive one: the provision of financial services to poor and low-income households that lack access to the services of formal financial institutions. For reasons of low population density and relatively high wage rates, the unit costs of microfinance services in Fiji are always likely to be high. This situation exists in a context of growing urbanization, open urban unemployment and underused resources in the rural economy. Overcoming the obstacles that impede access by poor and rural people to financial services is therefore a challenge for public policy and for private and community initiative. 2. Institutions in Fiji which provide microfinance services, or have the potential to do so, are found in the public, private and community sectors. They range from formal financial institutions (such as commercial banks and finance and hire-purchase companies regulated by the Reserve Bank of Fiji, the central bank) to relatively informal institutions not subject to financial regulation, such as voluntary sector microfinance institutions (MFIs), village community banks and unlicensed moneylenders. In an intermediate position are formal institutions not under the purview of the central bank, but otherwise regulated by legislation. These include the Fiji Development Bank, the National Centre for Small and Micro-Enterprise Development (NCSMED), Savings and Loans Cooperatives and Credit Unions, and licensed moneylenders. Standing somewhat outside this framework is the National Microfinance Unit, a Government of Fiji (GoF) initiative. 3. The microfinance ‘sector’ obtains support from a number of sources, but is by origin more a governmental than a voluntary sector initiative. The role of GoF agencies in providing funding and in-kind assistance has been crucial, with the support, primarily technical rather than financial, of external agencies. It has been more ‘supply-led’ than ‘demand-driven’. At the same time, other GoF programs providing grants or soft loans have proved unhelpful to the growth and sustainability of microfinance. Currently, the National Microfinance Unit (NMFU) is the lynchpin of direct government support to microfinance, and provides substantial resources for lending. The Ministry of Women was the pioneer in the field. From 1993 it supported WOSED, the first internationally-influenced MFI in Fiji. A number of others have emerged, including Aglow and FCOSS which are currently the most significant operators, having received substantial funding and technical support from NMFU. The NMFU is also supporting initiatives in community-based ‘village banking’ in some relatively remote locations. 4. Multilateral assistance, particularly that of UNDP in the context of regional ‘livelihood’ programs, has been of particular significance. Much of this work has focused on the policy and regulatory environment, and on provision of savings services. ILO has incorporated microenterprise development into its technical assistance for Fiji, contributing to the emergence of NCSMED. Bilateral development assistance has been directed to microfinance providers, including AusAID support for the Credit Union League (via the Credit Union Foundation of iv Australia, which provides periodic technical assistance). Korea has provided technical assistance for the major cooperative credit body and New Zealand aid has underwritten small loans provided by the Fiji Development Bank (FDB). Finally, there has been external non- governmental assistance in the NGO sector, support from a German foundation for the credit union movement, and some Peace Corps participation in the creation of WOSED’s microfinance program. Among international financial institutions, ADB has proposed a major program of assistance for credit unions and savings and loans cooperatives as well as for micro-lending via the Fiji Development Bank, all in the context of structural adjustment in the sugar industry. 5. The Review examines the viability of government programs of support for microfinance and their conformity with good practice principles (taken in this context to mean performance standards defined by CGAP, the Consultative Group to Assist the Poorest). It concludes that the NMFU has allowed short-term operational issues to overshadow its sector-building objectives. NMFU has been unable to disburse to its client institutions most of the funding provided by Government for microfinance lending. Nor has it made enough effort to break the capacity constraints which underlie this limited absorptive capacity. Its own retail lending activities, besides diverting attention from more fundamental objectives, have proved unsustainable and demonstrated that retail microfinance should not be conducted by government agencies which are not formal financial institutions. The experience of MFIs engaged in group-based lending and supported by NMFU illustrates the difficulty of achieving sustainable levels of operation given Fijian demography and costs. However, experience does suggest the potential for savings as an element in microfinance service provision. NMFU has also initiated a promising initiative in rural, community-based microfinance, commencing with savings mobilisation. 6. In macroeconomic terms, the Fijian economy has experienced a decline in investment as a proportion of GDP, and particularly in private investment, since the 1980s. Banks are highly liquid but perceive a dearth of good lending opportunities, leading to a decline in interest rates paid on deposits, although lending margins have been maintained and bank profitability is high. Commercial banks do not in general make ‘micro’ loans and, with the exception of one bank in which the GoF retains a minority stake, are not concerned with mobilizing the savings of small passbook account holders. High fees and negative real rates of interest on deposits act as a disincentive to micro-savings. In 1999 an official Committee of Inquiry into Financial Services drew attention to the limited availability and high cost of financial services for low-income and rural people, and this observation appeared still to be apt at the time of the Mission, early in 2003. Circumstances of excess bank liquidity and negative real interest rates can only be overcome by macroeconomic measures; the revival of economic growth would stimulate the demand for credit and increase competition among financial institutions for savings, including those of poor and rural people. This would test the proposition that Fijians lack a ‘savings culture’. 7. The Reserve Bank has proposed an extension of basic rural banking services. This would require banks to contribute to a banking infrastructure development fund, and to design products for small rural savers, including those organized in self-help groups by MFIs, together with the installation of shared ATM facilities in rural locations. Discussions with the banks on these proposals were continuing during 2003. The policy and regulatory environment for other institutional models of microfinance service delivery, including the credit unions and savings and loan cooperatives, is to be reviewed by the GoF in the period 2003-2005. Modern credit union legislation appears to be an urgent need. The ADB has proposed a program of assistance for v these non-bank institutions, and for the micro-lending operations of the Fiji Development Bank. The FDB itself has been reviewed and the GoF is to respond with a statement of policy for the bank’s future direction. It is clear that microfinance is intended to play a role in the strategic development planning of the GoF. The plan for the NMFU to be absorbed into the NCSMED could help harness microfinance in the service of broader aims of microenterprise development and income generation. However, the appropriateness of some proposals, such as for interest rate controls to prevent ‘exploitation’ of borrowers, or for compulsory savings schemes for particular segments of the population, or for comprehensive microfinance legislation, can be questioned. The case for comprehensive microfinance legislation is not persuasive at the present level of development of microfinance, while interest rate caps are antipathetic to the sustainability of microfinance operations. The absence of a ‘savings culture’ among Fijians, if indeed it is lacking, would be better approached from the standpoint of incentives and improved access to savings services than by compulsion. 8. The Review suggests the outlines of an architecture for microfinance in Fiji in which the NCSMED, as envisioned by the GoF, is the primary source of funding and technical assistance for MFIs. But to assure a sound financial sector building approach to policy it is desirable that the RBF also have substantial input. The Review also discusses the capacity-building needs of the sector and proposes that in addition to the role of NCSMED as the domestic focal point for capacity-building, the World Bank could be an appropriate external international agency to support the National Centre in that role. vi I. INTRODUCTION AND OVERVIEW This review is structured around four main themes, as follows:  the sources of support for Microfinance in Fiji, past, present and future;  a discussion of the viability, and conformity with good practice, of existing government support to Microfinance:  a discussion of actual or potential suppliers of Microfinance services:  an assessment of the enabling environment for Microfinance and of the potential for expansion of Microfinance services, including savings, payment services and credit, leading to conclusions and recommendations. These conclusions and recommendations are related to:  the conformity of existing government-supported Microfinance operations with recognized good practice, and with the wider aim of sector-building;  how government and donor Microfinance support could be improved or restructured;  what a conducive policy framework for Microfinance might look like in Fiji: and,  how donors and the Government of Fiji could better coordinate Microfinance support activities. However, before addressing the main issues in the report, some preliminary comments are necessary. These will set the scene for a discussion of Microfinance in the particular economic and social environment of Fiji and consider what definition of Microfinance is appropriate in that environment. THE ENVIRONMENT FOR MICROFINANCE IN FIJI Distinct features of Fiji’s demography and geography decisively influence the nature of and scope for Microfinance (MF) in the country. In terms of demography, both absolute population and population density in Fiji are substantially lower than in many countries, including the most-cited example, Bangladesh, where various models of MF have flourished. More than half of the population of Fiji is located in rural areas and on smaller, dispersed, islands where all services, not least MF, are difficult and expensive to provide. MF provision is a labor- intensive activity, and higher population density, other things being equal, permits higher client- loads per field officer and/or lower time and transport costs per client. (These two variables are virtually the reciprocal of one another.) 1 Short of technological, perhaps electronic, advances applicable in the Fijian context, the unit costs of delivering conventional MF services will remain relatively high in Fiji. This impediment is not compensated by a wage-cost advantage in Fiji, vis-à-vis countries where MF has thrived. Although the point is obvious, it should be noted when considering the performance of MF in Fiji in a comparative perspective. This review of MF in Fiji – as well as the consideration of achievable goals to improve financial services in Fiji – has taken these constraints into account. MICROFINANCE IN THE FIJIAN CONTEXT This review proceeds on the assumption that, in the Fijian context, the appropriate definition of MF is the provision of financial services to poor and low income households that lack access to the services of formal financial institutions. That access to formal financial services is insufficient is attested by the official Committee of Inquiry into Financial Services of 1999: “There appears to be an insufficient number of products and services for lower income earners who have difficulty meeting opening or minimum balance requirements. Nor are they able to access certain loan products that may suit their needs, as they lack security for them.�1 The limitations of access may flow from non-provision of services, from conditions attached to services (e.g. collateral requirements) that consumers are unable to satisfy, or from high user transaction costs of services where they are nominally available. In Fiji, it is the case that the great bulk of the population lacks access to formal credit services, but that access to the passbook savings services of regulated financial institutions is in principle widely available. However, transaction costs for savers are very high in relation to typical savings balances. Incorporation of remittances into MF would benefit many low-income households. Poor and low-income people in Fiji need access to convenient, liquid and deposits that are protected against inflation by positive real rates of interest. Savings help individuals smooth consumption expenditures in the face of uncertain income streams and protect against catastrophic events such as cyclones which would otherwise force the vulnerable to divest productive assets. Similarly, the poor and low-income that make a living outside the formal sector of the economy need access to credit to increase productivity in self-employment or to free them from exploitative financial relationships. Fijian households have relatives working away from home, in urban centers or overseas, and receive periodic remittances. Transactions costs of remittances for recipients are high. Although there is a growing group of urban residents whose connections with their rural origins have become tenuous over time, it makes less sense in the case of Fiji than in many other countries to think in terms of a significant category of landless poor. Many urban and rural residents employed outside agriculture retain access to rural customary land, but judge it too remote from markets or amenities. In their view, the balance of advantage between rural and 1 Ministry of Finance (1999). Report of the Committee of Inquiry into Financial Services: Suva: Parliament of Fiji, Parliamentary Paper No. 19 of 1999, p.xx. 2 urban life is in favor of the towns. Accordingly, they have voted with their feet. At the same time, many rural households do not exploit their resources of agricultural land fully. Concern within the Government of Fiji (GoF) over rapid urbanization is reflected in rural development policies designed to slow the pace of urban growth and develop regional infrastructure. Urban unemployment levels are high and rising,2 suggesting the need for MF to support microenterprise (ME) development in rural areas. Policies for MF development have to be considered in that broader framework as well as in the context of macroeconomic policy, which is a major influence on development of a sustainable system of MF in Fiji. MICROFINANCE INSTITUTIONS IN FIJI To complete the overview of Microfinance in Fiji, it is useful to describe the major institutions operating in the sector and to categorize them in terms of the legal and regulatory environment in which they operate. Initially, however, some brief comments on the broader financial system will help to place the microfinance institutions of Fiji in their sectoral context. Regulation and supervision of the system are performed by the Reserve Bank of Fiji, which is also responsible for other central banking functions including the conduct of monetary policy, issuance of currency and management of external reserves. Commercial banking services are provided by five institutions, all with total or majority private, foreign ownership. The sector is highly concentrated, with two of these banks, ANZ and Westpac, accounting some 60 to 70 % of total bank assets and liabilities. Both are wholly-owned by Australian interests. The Government of Fiji is minority shareholder of the third-ranking institution, Colonial National Bank. CNB has the most pronounced retail focus of all Fijian commercial banks and correspondingly the most widespread network of branches and agencies. In broad terms, the commercial banks account for almost a third of total assets and liabilities of the financial sector, while the central bank holds some 15% or 16% and non-bank financial institutions some 45% to 46%. The non-bank sector consists typically of just one or a few institutions in each category. The single, state-owned, development bank accounts for some 5% or 6% of assets and liabilities. The Fiji National Provident Fund is the largest financial institution, with about a third of total assets and liabilities. Other institutions hold some 10% to 12%, including insurance companies with around 6% to 7%. Some features of the banking system are relevant to the concerns of microfinance; there is little price competition between commercial banks and they are highly profitable, as well as being highly liquid. Lending is constrained by a perceived lack of bankable projects rather than by shortage of loanable funds and commercial banks are not moved to engage in ‘bottom-end’ lending. Banks do not provide microcredit in Fiji. On the other hand, the Fijian banking system 2 A household survey in July 2002, covering some 35,000 or 42% of households in all urban areas of Fiji, found more than 23,000 people were actively looking for work. This was an urban unemployment rate of 14.1%. However, reflecting a point made in the text above, concerning the continuing access of many Fijians to land, the survey also found that a third of the unemployed were engaged in subsistence activity. Fiji Times, 2/21/03, letter from T. I. Bainimarama, Government Statistician. 3 held some 560,000 savings deposit accounts in 1997 (for a national population of less than 800,000) and are potentially an important source of deposit services for the poor. As against that, access to banking services has deteriorated in recent years (thus, the number of persons per bank branch or agency rose from around 3,800 people in 1994 to about 6000 in 1997, reflecting difficulties experienced by the (then) government commercial bank and the contraction of its services) and there is a marked disparity in access between urban and rural areas. Banks appear to discourage passbook savings accounts, the preferred savings product of low-income people. There is negligible overlap between the formal financial sector whose institutions are regulated by the Reserve Bank of Fiji and institutions providing microfinance services. Following is a summary of all MF providers in Fiji, including all institutions which have actual or in-principle involvement with the sector. The institutions concerned are discussed in detail elsewhere in this Review. A taxonomy of providers is constructed in Table 1.1. It distinguishes institutions as public, private or community sectors, and by the institution’s relationship to the regulatory or legislative environment. Table 1.1: A taxonomy of Microfinance Providers in Fiji Categories Regulated by Reserve Otherwise Regulated or Not Subject to Financial Other of Bank of Fiji Legislated Regulation Provider Public Fiji Development Bank National Micro- Sector National Centre for Finance Unit SMED (NMFU) Private Commercial Banks (eg, Moneylenders Sector CNB, Westpac, ANZ) Finance & Hire Purchase Companies Community Savings and Loans (S & MFIs (eg, WOSED, Sector L) Cooperatives (eg, Aglow, FCOSS, SEEDS) CCSLA, farmer cooperatives) Village and Tikina Banks Credit Unions (FCUL & indep-endent CUs) Informal Moneylenders As discussed above, there is the set of service providers which are subject to central bank regulation. These are the commercial banks and finance companies, all in the private sector. No community sector providers are subject to RBF regulation at present, although certain actions, such as accepting deposits from the public, would immediately place any non-regulated MF provider under the scrutiny of the central bank. Then there is the category of providers which are otherwise regulated or legislated for financial service provision, including the Fiji Development Bank, incorporated under its own legislation, reporting to the Ministry of Finance, and required 4 to consult the RBF about its lending guidelines. The NCSMED (National Centre for Small and Micro-Enterprise Development), for example, was incorporated under its own Act and given credit functions under that legislation. In the private sector are registered moneylenders whose activities are subject to the Moneylenders Act. Finally, in the community sector are cooperatives and credit unions. The former are authorized under the Cooperatives Act to provide services to their members, which may include savings and loans as authorized by by-laws agreed by members. They are supervised by the Department of Cooperatives. Credit Unions and established under their own legislation, the Credit Unions Act, for the specific purpose of promoting thrift and providing financial services (savings and loans) to a group of members who share some common bond. There is a category of MF providers in Fiji not currently subject to any financial legislation, although they may be subject to processes of registration and approval (for example, to determine their charitable status) under non-financial legislation. These registered institutions consist of the MF institutions (MFIs) set up by NGOs, notably FCOSS and Aglow. A somewhat different category comprises the embryonic institutions, village or Tikina banks. These have the sanction of being established by local-level administration with the assistance of a government agency, the NMFU, but do not appear to be mandated by legislation. Finally, there is the category of moneylenders in the informal sector. They operate outside the boundaries set by the Moneylenders Act and their informal methods set them apart from regulated moneylenders. In this taxonomy, they have been placed in the community, rather than the private sector, because of the nature of their activities (i.e. informal) and their embedment in the communities. The category ‘other’ has been reserved for the National Microfinance Unit (NMFU), which occupies a unique place in the sector. It is a pilot project of the Ministry of Commerce, Business Development and Investment, operating both as a wholesaler and retailer of microcredit and supporting MFIs and village banks in mobilizing the savings of their members. It is discussed at length in section III of this Review. 5 II. SOURCES OF SUPPORT FOR MICROFINANCE IN FIJI Several GoF agencies, and both official and non-official donors, support MF in Fiji. Table 2.1 provides a summary of the principal MF providers in Fiji, the forms of support, the activities (technical assistance for capacity-building; support for improving the policy and legal environment), and plans for future involvement in the sector. It also provides an introduction to some of the principal domestic organizations in the sector. The table is not exhaustive; not included is a number of small and ephemeral NGO- or community-based initiatives that failed in the difficult circumstances of the early 2000s. In addition to MF, other GoF programs have involved (and continue to involve) the direct disbursement of funds, whether by grant or soft loan, to particular population sub-groups for ME activities. Some of these activities are on the margin of MF; the beneficiaries and their economic activities are potential candidates for MF services. Such programs are competitive with genuine MF for resources and clients. A number of GoF agencies been active in MF both as financier to service providers and as provider through government staff and resources. GoF has provided resources for capacity- building of service providers, and has also employed government staff and other resources for direct capacity-building. And it has devoted attention to the policy and legal framework for MF in Fiji, an area where external and voluntary agencies have also contributed. International donors have had limited involvement in MF in Fiji, with NZODA perhaps the most active of the bilaterals and UNDP among the multilaterals. The coup of May 2000 led most bilaterals to scale back their programs, and important donors such as Australia and New Zealand have only now redeveloped comprehensive assistance programs. MF is not a significant component of these programs. UNDP, however, has been a significant source of technical assistance, and plans to continue involvement in the sector. 6 Table 2.1: Microfinance in Fiji – Providers and Sources of Support MF Provider Government Other Domestic Official Develop Other Forms of Comments Assistance Assistance Assistance ODA External Assistance Assistance NMFU Dept of UNDP, under the NMFU: grants Placement of UN (National MF Commerce, UNDP/GoF MF Services and TA to MFIs Volunteer within Unit) (Pilot Business & Partnership arrangement. and village unit, 2001-2. Project within Inv. banks. Ministry) from 1999 UNDP: T/A National Center Dept of ILO T/A (see IHRDPEP, NCSMED can NCSMED Created, for Small and Commerce, below). provide grants, inter alia, to absorb Microenterprise Business & loans and T/A to NMFU and to Development Investment MFIs and assume credit (NCSMED) from 2003 second-tier responsibilities. (statutory body) institutions, & also directly to SMEs. WOSED Ministry of US Peace Corps, 1993-95 Gov’t grants for Operations (Hybrid/MFI) Women and NZODA loan funds, suspended by Culture, from recurr- ent costs government, pro 1993. UNDP T/A. & T/A. tempore Peace Corps: T/A Anandpur (MFI) Ministry of Ba Rotary Club UNDP T/A Grants for Now inactive Women/Cultu lending, T/A. re, Ministry National Planning, 1997- 2000. SEEDS (MFI) NMFU UNDP (T/A) NMFU: grants Inactive, though (National MF for loanable some components Unit) 1999- funds & T/A. continue 2001. UNDP: T/A. FCOSS MFI NMFU, from Fiji Council of UNDP (T/A) NMFU grants A distinct unit within (MFI) 1999. Social Services. for loanable FCOSS, NGO apex funds & T/A. agency. CNB provides client passbooks. Aglow MFI NMFU from Aglow Fiji, UNDP Aglow Inter- Grants for (MFI) 2000 social & national loanable funds religious NGO. (USA) & T/A from CNB provides NMFU. client passbooks UNDP: T/A Village Banks NMFU, NMFU grants Community-based (MFIs) Provincial for recurrent MFIs and Tikina costs, loanable administration funds, also T/A s, from 2002 7 MF Provider Government Other Domestic Official Develop Other Forms of Comments Assistance Assistance Assistance ODA External Assistance Assistance Fiji Credit AusAID has supported Hanns Seidel T/A, Apex body for Credit Union League CUFA to provide T/A. Foundation institutional Unions (FCUL) ADB PPTA (2003) (Germany) strengthening, proposes support to Mid-1990s. supported by strengthen FCUL and AusAID, Hanns create at least 30 new CUs Seidel Credit Union Foundation over five years, with line Foundation of of credit for on-lending. Australia (CUFA) (regional T/A, intermittent) Cane Farmers’ Department Republic of Korea T/A, with Apex body for Co-operative of volunteers on Savings and Loans Savings and Cooperatives attachment. Cooperatives Loans From 1994 and Association continuing. (CCSLA) ADB PPTA (2003) proposes capacity- building and greater outreach for CCSLA ,with line of credit for on- lending. FDB (Fiji Bank reports NZODA has supported a Grants for Role of FDB now Development to Ministry Small Loans (revolving loanable Funds under review by Bank). Small of Finance. fund) Scheme since 1989. government Loans Scheme, Gov’t ADB PPTA (2003) from 1989. approves proposes line of credit for FDB bond ‘larger’ MF loans for raisings. ‘alternative livelihoods’ project IHRDPEP Inter- ILO T/A. (See Integrated HRD Policy departmental above: Program for development, collaboration NCSMED Employment sectoral From 2000. emerged from a Promotion employment and component of HRD programs. this project) SMILE Program Participation by UNDP Regional T/A T/A provided by for Sustainable Bank of Hawaii (1998-2002) UNDP to NMFU, to MF & In ‘ComeSave’ all NGO-based MFIs Livelihoods & ‘WorkSave’ and to Village Banks through group savings under regional Empowerment schemes in Suva programs district. PSLP: Pacific UNDP Regional T/A Incorporates SMILE Sustainable (from 2002 and and other related Livelihoods continuing) programs Program 8 FINANCIAL AND CAPACITY-BUILDING ASSISTANCE FOR MICROFINANCE INSTITUTIONS Beyond the longer-established S&L cooperatives and credit unions, the first Fijian MF institution (MFI) based on international models of MF was the Women’s Social and Economic Development (WOSED). WOSED was established in 1993 by the Ministry of Women and Culture, and the Ministry has provided financial and in-kind support to WOSED since its inception. This has involved the provision of loan funds and hands-on management of credit operations by Ministry field staff, assisted by volunteer workers. WOSED benefited from the full-time assistance of a US Peace Corps volunteer for its first two years together with funding support from NZODA. It has subsequently received technical assistance from UNDP under regional programs. WOSED is a hybrid organization, neither strictly government nor NGO; issues of legal identity and ownership remain to be determined after the promised review. Operations have been suspended pending the results of an external review. A number of entities in the MF sector sprang directly from the National Conference on Microfinance, organized by the Ministry of Finance in 1998. The most significant outcome of the conference was the creation of the National Microfinance Unit (NMFU), currently located within the Ministry of Commerce, Business Development and Investment. While acting primarily as an apex funding institution and source of technical assistance for MFIs, NMFU carries out a direct individual lending program. More recently, NMFU has begun trialing village banks, informal rural financial institutions organized on group principles. These are community entities, organized at the village or Tikina (a group of related villages) level. NMFU has funded establishment costs and salaries of local-level staff, and invested considerable staff time in the launch of two such institutions. With this range of activities, NMFU is currently the most significant source of financial support for the sector (table 2.2); though it has been unable to lend the full allocation of onlending resources made available by the GoF for the purpose. Starting with an initial allocation for loanable funds of $F3.0 million in 1999, the NMFU has been allocated $F1.0 million in each subsequent year, 2000-2003. Table 2.2: Financial Support to Microfinance Providers by NMFU Cumulative, Calendar 2000-2002 ($F ) FCOSS Aglow Aglow IDL Village Tikina Other Total West Labasa 420,166 431,402 190,000 199,000 24,408 55,632 150,000 1,470,248 Two important MFIs emerged from the 1998 conference: FCOSS MF Institution, set up by the Fiji Council of Social Services, the umbrella group for Fijian NGOs, and Aglow MF Unit (AMFU). These two MFIs have been the principal beneficiaries of the NMFU, each receiving more than $F0.4 million in grants over the three years 2000-2002 for recurrent costs and loan funds. The NMFU has worked closely with each of these organizations, providing technical assistance in the form of operational packages. The packages provided guidance on methods of organization, accounting and management information systems. NMFU also provides supervision services. UNDP has rendered technical assistance to both MFIs through the agency of UN Volunteer (UNV) staff under its regional Pacific SMILE (Sustainable Microenterprise and 9 Livelihoods) program and has involved their staff in a range of regional training exercises. Assistance has included on-site visits, training and support in gauging the impact of operations on clients. UNDP, via its Suva-based South Pacific Regional Office, has provided substantial technical assistance to the MF sector. Over the past five years, the program has been served by a succession of UNV advisors and has benefited from a regional program director with a strong interest in the field. Aside from technical assistance to the NMFU and individual MFIs, SMILE has been active in Fiji in nurturing sustainable and self-managed savings models tailored to the Fiji context. These occur in factories and other workplaces, schools and community-groups. The Bank of Hawaii facilitated these savings programs before its takeover by a major commercial bank. Fijian staff of MFIs and credit unions have benefited from training and exposure to related institutions in the region, in workshops. UNDP’s assistance to the NMFU is formalized under a UNDP/Fiji MF Services Partnership Arrangement. In the MF sector, two major categories of community credit institutions are governed by legislation. The first category is comprised by credit unions, which are both urban and rural and generally under the umbrella of the Fiji Credit Union League. The second category, thrift/savings and loans cooperatives, are active in the rural and agricultural sectors. The most significant sub-group is the Canefarmers’ S&Ls, organized under the Canegrowers’ Cooperative Savings and Loan Association (CCSLA). Both have received technical assistance from overseas agencies. A recent Project Preparatory Technical Assistance document from the ADB 3 proposes extensive assistance for these two entities and for the Fiji Development Bank. This is to total $F18.5 million over five years and would take the form, inter alia, of lines of credit for on- lending and support for capacity building. The latter would involve the creation of at 30 new rural savings and credit unions over five years, and would require strengthening the Fiji Credit Union League and the CCSLA, to assure the institutional sustainability of the project’s initiatives after it is completed. SUPPORT FOR IMPROVING THE POLICY AND REGULATORY ENVIRONMENT FOR MICROFINANCE The UN system, through its regional programs, has been the most consistent supporter of the GoF in efforts to secure a supportive policy and regulatory environment for MF. In 1994, UNDP’s Regional Bureau for Asia and the Pacific commissioned consultations on MF in several of the Pacific Islands, including Fiji, under its Regional Poverty Alleviation Program. 4 This permitted exchanges of experience between practitioners and policy-makers in the region. In preparation for the Bank Poor ’96 meeting in Kuala Lumpur, the Regional Bureau also commissioned a study of MF in the Pacific, including Fiji, on the policy and regulatory 3 ADB (2003). Alternative Livelihoods project: Mid-Term Report, TA No 3887-FIJ, prepared by Lincoln International. 4 See FDC (1996). Banking with the Poor in the South Pacific. Brisbane: The Foundation for Development Cooperation. 10 environment.5 The UNDP SMILE program co-sponsored the 1998 National Conference on MF with the Ministry of Finance, and in advance of the event, prepared a review of Fijian laws and regulations.6 More recently, UN ESCAP Pacific Operations Center commissioned a study of the need for specific legislation for MFIs, at the request of the National MF Unit.7 The International Labor Organization (ILO) was the driving force behind the interdepartmental Integrated Human Resource Development Program for Employment Promotion (IHRDPEP). Since 2001, this has assisted the GoF to develop policy frameworks, including one for Small and Microenterprises, of which a by-product is the National Center for Small and Microenterprise Development (NCSMED). It is established under its own legislation, the Small and Microenterprises Development Act (No. 1 of 2002) to play a central role in micro and small enterprise development. Among the functions of the Act are “to provide or source loans or grants to small and microenterprises� and “institute safeguards for the protection and stability of the provision of credit to small and microenterprises.� NCSMED’s Strategic Plan 2003-2005 proposes to incorporate the NMFU within itself, and lists the provision of loans as a medium term objective of the NCSMED. There have been recent domestic initiatives as well. At the request of the GoF, the Reserve Bank of Fiji (RBF) has undertaken a study of options to improve the access of the rural population of Fiji to banking services. This is discussed in some detail in section V, below. The Fiji Credit Union League has a technical assistance relationship with the Credit Union Foundation of Australia and also received assistance from the German credit union movement in the 1990s. The proposals for ADB technical assistance in the Alternative Livelihoods Project arise from the need for GoF to address an impending economic and social crisis. It is expected that there will be a displacement of a significant number of small cane growers and sugar mill workers as a result of structural adjustment in the Fijian sugar industry. The project proposes that, in addition to providing lines of credit and institutional strengthening, the ADB assist GoF in establishing and implementing supervision arrangements for credit unions and S&Ls. It also proposes that ADB arrange for Fijian policy-makers and regulators to visit countries where comprehensive frameworks for the regulation and supervision of MFIs are in place and successful apex institutions are in operation. The document canvasses the desirability of Government’s taking the initiative to establish a wholesale facility for MF as a necessary complement to ADB assistance to the sector. No doubt the NCSMED could be seen in that light. 5 McGuire, PB (1997). Microfinance in Pacific Island Countries. Brisbane: the Foundation for Development Cooperation. 6 UNDP/UNOPS (1998). A review of Fiji laws and regulations affecting Microfinance institutions. Suva. 7 Serge Belloni (2001). About the opportunity to enact a specific law for micro-finance institutions. Port Vila: UN Economic and Social Commission for Asia and the Pacific, Pacific Operations Centre. 11 III. VIABILITY OF GOVERNMENT SUPPORT AND CONFORMITY WITH GOOD PRACTICE This section deals with conscious GoF efforts to establish a sustainable MF sector, and specifically with the National Microfinance Unit. This is seen as a ‘supply-leading’ initiative in the field. The availability of resources from the NMFU called into existence a number of “MFIs� or MF institutions in the conventional sense, which have their origins in the voluntary sector rather than the financial sector. Two of these were active at the time of the Review and were the principal clients of the NMFU. In addition the NMFU is trialing an innovative grass- roots approach to “village banking.� The notion of ‘good practice’ is derived from microfinance industry performance standards which have been developed internationally over more than a decade. The definition and dissemination of ‘best practice’ standards was stimulated by the creation of CGAP, the Consultative Group for Assistance to the Poorest, in 1995. CGAP is a consortium of donor agencies, with a secretariat in Washington DC provided by the World Bank . The Consultative Group has facilitated exchanges between donors and practitioners which have produced definitive statements on a wide range of issues touching on the efficiency, effectiveness and sustainability of microfinance services. A useful summary of these ‘best practice’ performance standards is contained within a CGAP document, the Format for Appraisal of Microfinance Institutions, at http://www.cgap.org/html/p_technical_guides04.html. A simplified version of this ‘Appraisal Format’ would be a useful tool for the NMFU in assessing the performance of Fijian microfinance institutions at the current stage of the industry’s development. A draft format for the purpose is provided with this Country Review, as Annex A. A. ASSESSMENT OF THE NATIONAL MICROFINANCE UNIT Government financial support for MF is disbursed primarily via NMFU. The viability of government support for MF in recent years, and its conformity with good practice, is examined in that context. The Unit’s corporate plan for 2003 states as its vision: “The National Micro Finance Unit will be the leading institution in the development of a sustainable micro finance sector in an enabling environment that will actively contribute to the improvement in the quality of life for the disadvantaged in Fiji.� This vision requires certain objectives to be achieved:  To develop, promote and initiate MF programs that will help foster the growth of smalland microenterprises  To work towards the development of an integrated framework through government and non-government bodies that will assist in the promotion, development and sustenance of a healthy micro finance sector; 12  To conceive, test and operate mechanisms that will continuously improve the technical competence, creativity and effectiveness of staff; and,  To evaluate the Pilot Phase Program and organize the merge with the NCSMED. Originally conceived as a pilot program to run through 2001, NMFU was extended to the end of 2003. The original motivations for the choice of MF as an instrument of poverty eradication policy through ME development are defined in a 1998 Cabinet Memorandum 8 which proposed the National Conference on Microfinance. The document maintained the need for access to small credit, mentioning Grameen Bank as a model, and stressed the need for a supportive policy and legislative environment for MF. It recognized the need for a coordinated MF delivery system and recognized the potential of delivering MF within institutions such as the Fiji Development Bank, the S&Ls and credit unions. Proposed themes for the conference included outreach, targeting, sustainability, policy environment, and resource mobilization. Thus, an emphasis on systemic issues underlay the creation of the NMFU, and these elements remain substantially reflected in the its current corporate plan. However, for most of the intervening period, the NMFU neglected systemic issues in favor of a preoccupation with short-term operational objectives. These objectives were determined by the political decision to allocate substantial amounts of loan capital to the NMFU ($F3.0 M in its first year of operation, and an additional $F1.0 M in each subsequent year) and to set a short time-span (less than 3 years) for the initial program. These decisions suggested the urgency of government to quickly launch a program, and with an almost exclusive emphasis on microcredit. NMFU concentrated, at least initially, on encouraging the creation of Grameen Bank replications. It was slow to consider assistance to existing financial institutions. The irony is that, in neglecting systemic issues such as capacity-building and the creation of an enabling environment, the NMFU has been unable to establish conditions in which its initiatives could flourish. At its establishment, NMFU was placed within the Ministry of Finance. Subsequently, in 2000, with the election of a new Government, NMFU was transferred to the Ministry of Commerce, Business Development and Investment. The Ministry of Finance had arranged overseas MF training for staff of NMFU, but these staff remained with Finance at the time of the transfer. Subsequently, it appears that most staff of NMFU received inadequate preparation for their tasks, which in the initial stages included the selection and preparation of NGOs to conduct MF programs. Two NGOs were the initial recipients of NMFU funding: the FCOSS MF Unit, active in the Central division around Suva, and the Aglow MF Unit. Both were created within existing NGOs and were to use the Grameen Bank model of MF, employing group methods of organization. In the first quarter of 2001, the NMFU launched its own retail MF operations. It moved away from the group to an individual lending program in which loan sizes were larger than those administered by the NGOs. This was an attempt to trial a different mode of lending but may also have been influenced by concern about the slow disbursement of available loan capital. The individual lending program (IDL) was administered by an internal unit formed within the 8 Cabinet Memorandum, No. CP (98) 76, March 31, 1998, Micro-Finance Development in Fiji. 13 NMFU. By venturing into retail MF, NMFU blurred the focus of its operations, taking them further away from systemic issues. This and other difficulties appear to have been the result of erratic management in the first three years of operation.9 An external review of NMFU operations ending in March 200210 cited NMFU’s hands-on micro-management style and tensions with the NGOs forming its client base. This led to interruptions and delays in the disbursements of NMFU onlending capital to its NGO partners. Clearly, the financial resources allocated to NMFU for lending far exceeded its capacity for disbursement, indicating the shortage of Fijian organizations with credible credentials to carry out conventional MF based on the Grameen model, as well as shortcomings of the NMFU itself. The focus of GoF policy was on credit and on short-term results. The reputation of the Grameen model had imprinted itself upon the consciousness of policy-makers and the early management of the NMFU. The preoccupation with Grameen required the creation of new institutions and no real consideration was given at that time to using existing institutions as vehicles for expanding financial services. The Grameen model requires extensive capacity- building of institutions and staff and painstaking preparation of borrowers before lending is commenced. However, budget allocations for capacity-building, either within client organizations or NMFU itself, appear to have been inadequate by comparison with the capital available for lending. In each of its first three years, NMFU returned substantial unspent loan funds to the Ministry of Finance (table 3.1) although disbursement performance improved as client programs matured. Table 3.1: Capital (Loanable Funds) Budget Allocation for NMFU, 2000-2003 Expenditure 1999 2000 2001 2002 2003 Budget $F3,000,000 $F1,000,000 $F1,000,000 $F851,500 $F1,000,000 Actual $F 200,000 $F 215,000 $F 366,000 $F689,248 N/A % Used 6.66% 21.5% 36.6% 81% N/A Source: adapted from Garcia (2002) with additional data for 2002 from Director, NMFU. Note: The capital budget for 2002 was reduced by $F148,500 to cover operating expenses of NMFU. Table 3.2 provides details of the main programs to which NMFU has disbursed grants for on-lending over the period to end-2002 and the collection performance of each. Data reflect an expanding portfolio, with a number of institutional clients in the early, pre-lending, stages of their development. The NMFU has now come through experience to an appreciation of the value of the “savings first, lending later� approach to MF. 9 The current Director of the NMFU was appointed to the post at the time of the external review and has a background of experience as a loans officer with the Fiji Development Bank. 10 Garcia, Willie M. (2002), Program Evaluation Report: An in-depth assessment of the operations of the NMFU and of the micro finance programs implemented through local NGOs. Suva: UNDP/UNV. 14 A number of MF models are being supported. These include Grameen replications (FCOSS and Aglow, together with a new institution affiliated with Aglow), for which NMFU acts a “wholesaler,� supplying loanable funds and providing supervision. The Aglow program is discussed in some detail in the second part of this section, and in Annex B. NMFU also operated its own individual lending operation, the IDL, for a brief period. This is also discussed briefly below and in Annex C. Finally, NMFU supports two variants of a rural, village-based, MF model. Both village models employ local-level systems of decision-making and administration, and continue a recent trend within the NMFU towards emphasizing the value of savings. They appear to be a most interesting innovation and their progress deserves to be followed closely. Some details of the village banks are given below. Table 3.2: NMFU’s National Microfinance Outreach as at 12/31/2002 ($F) Aglow Aglow Individual Village Bank Tikina Indicator FCOSS West, North, lending by Burenitu Bank TOTALS Labasa NMFU Nakuailava Lautoka Grants 420,166 431,402 190,000 199,000 24,048 55,632 1,320,248 Loans disbursed 189,467 280,071 Not yet 248,166 Not yet Not yet 717,704 lending lending lending Amount collected 214,339 287,674 N/A 153,226 N/A N/A 655,239 Accounts: loans 3,051 & savings 803 1,624 65 284 88 187 Savings 38,794 42,848 1,721 25,397 6,207 2,411 117,378 Collection 95.2% 93.88% N/A 61.74% N/A N/A 85.27% Source: NMFU. Notes: 1.There is a discrepancy between the total grants shown here ($F1.32M) and total disbursements of $F1.47 shown in table 2.2, above. This appears to reflect grants made to a number of discontinued MF programs, including SEED Foundation, totaling some $F150,000. 2. There are substantial discrepancies between data reported by NMFU for Aglow Lautoka and data collected from Aglow records, reported below in the description of Aglow MFI. 3. Amounts collected appear to include interest due, and the collection rate shown is calculated in relation to payments having fallen due as at end-2002. NMFU’s “Individual Lending� Program As mentioned above, NMFU also operated a retail MF program for a short period, lending to individual clients. This is discussed in greater detail in the case study reproduced as Annex C. In summary, the outreach of the Unit’s ‘Individual Lending Program’ (IDL) during the year of its active operation was quite limited, with 238 clients of whom only 46 took second or subsequent loans, despite the short loan terms involved. Portfolio quality was quickly established as poor (with an on-time collection rate of little better than 70% after only nine months of operation) and continued to deteriorate. Given this collection performance and the high level of operating costs, the IDL could not be considered a sustainable operation. It is not 15 clear that higher levels of activity with more clients could overcome the inherent disabilities associated with trying to run a direct lending operation out of a government office that has a range of other functions. The NMFU has suspended all new lending under this program. This is appropriate since there is a good case for deciding that public sector entities should not be part of the financial sector and cannot become sustainable long-term providers of financial services. Village ‘Banks’ A novel approach to financial service provision is being trialed in two relatively remote areas of Northern Viti Levu, the main island, under the guidance of the NMFU. One involves a cluster of related villages, a Tikina, and is based on traditional authority structures as employed by the GoF for local administration. Locally recruited field officers, young educated people, are to be paid modest stipends for the part-time work involved in administering the scheme at this stage. The other is organized along similar lines, except that it is being trialed in a single village and is accordingly smaller in scale. With the benefit of prior experience, the NMFU has started these schemes with savings, and levies a one-off fee of three cents in the dollar on deposits as a service charge. The same procedure is employed in the Aglow MFI, where savings performance has been very encouraging. This is the more so when one remembers that these savings, as explained in the discussion above of commercial banks, earn no interest from the bank in which they are deposited. The weekly collections of savings are deposited in a single master account in the name of the collecting institution. Each individual saver has his or her transactions recorded in a passbook donated to the MFIs and village banks by CNB. The village banks were scheduled to commence lending to a small number of members of the village banks in March 2003, financed by grants of loanable funds from the NMFU. These borrowers were to have been chosen on the score of excellent savings records. It is hoped that the short period remaining for the activities of NMFU as a pilot project have not impelled a premature release of loan funds, but in any case this is an interesting experiment, worthy of close observation. The NMFU hopes to commence similar operations in the more remote Maritime division of Fiji before the end of 2003. Conclusion In terms of GoF support for MF and its conformity with good practice, NMFU has been found wanting in a number of respects. On the issue of political influence on NMFU operations, there is no evidence of any crude political influence in operations of the NMFU on issues such as the allocation of funds to organizations or to individuals. However, in a more general sense the initial impetus for the establishment of NMFU reflected a political decision that microcredit offered a solution to problems of poverty and unemployment and that the Grameen Bank model should be the preferred option for the rapid multiplication of MF services. That a high level of resources was made available to NMFU over a short time-span supports this judgment. Since its establishment, the choice of activities of the NMFU has largely reflected that narrow definition of the NMFU mandate, thus its failure to take a more strategic (or systemic) approach has reflected unhelpful political influence. In this connection, the early transfer of NMFU from the Ministry of Finance to the Ministry of Commerce, Business Development and Investment may have reinforced the focus of 16 NMFU on microcredit for ME development, as opposed to a financial systems development approach, which would have considered the need for a range of micro-financial services, notably savings. A more pronounced financial systems orientation on the part of NMFU might have permitted attention to strengthening existing institutions (such as S&Ls and credit unions) in addition to creating new ones. As a result of its experience of direct and indirect lending programs, the NMFU has more recently come to a greater appreciation of the importance of savings, and has attempted to find ways of cooperating with established institutions, including the credit union movement, to extend the outreach of MF. The current NMFU director regards this last matter as unfinished business from its pilot stage, along with testing its village bank methods in more remote locations. In terms of efficiency, NMFU experienced severe management difficulties in its first three years of operation. If the major indicator of an agency charged with the disbursement of funds is the rate at which it succeeds in disbursing them, then NMFU must be accounted inefficient. Of course, the inability to disburse was the result of limited absorptive capacity in the NGO sector where NMFU concentrated its efforts in the first three years. But the way to break this constraint would be to find more resources for capacity-building in the sector. It is not evident that NMFU has argued for the diversion of any loan funds to capacity-building, or that management was conscious of the possibility of addressing the capacity constraint in this manner. NMFU’s future appears to lie within the more-recently created National Centre for Small and Micro Enterprises. The implications of this for microfinance are considered below in section V, devoted to discussing the ‘enabling environment’ for microfinance in Fiji. B. ASSESSMENT OF A GOVERNMENT-SUPPORTED MFI: AGLOW, A GRAMEEN BANK REPLICATION Aglow is a faith-based NGO. When the GoF called for expressions of interest from community organizations to conduct MF in 1999, Aglow applied and was accepted into the NMFU pilot program. It is one of two such institutions supported by the NMFU, and it is examined here as being representative of this embryonic sector. The Aglow MF unit is a Grameen replication exercise, operating in and around Lautoka, in Western Viti Levu. Detailed operational data for Aglow are given in Annex B. Aglow set up its MF unit as a separate entity and received its initial grant early in 2000, commencing lending operations just before the coup of May 2000. Amid the unstable political environment, the MF Unit experienced substantial collection problems in that early period. Performance Summary In January 2003 AMFU had slightly more than 1000 clients. It had opened more than 1650 loan accounts since its inauguration, more than 1200 of which had been paid off in full. First loans totaled 1,012, with 427 second cycle loans and 169 third cycle loans over the period of less than three years of AMFU’s operations. In February 2003, arrears of the program totaled around $F18,000, after rather less than three years of operation and a cumulative loan disbursement of some $F360,000. 17 In February 2003, net savings balances held in the names of Aglow clients totaled $104,000. AMFU holds member savings in a single master account with a commercial bank, keeping individual records for each saver and charging a one-off fee of three cents in the dollar on deposits to cover its transaction costs. Around $F75,000 was held as the voluntary savings of borrowers. Aglow clients clearly depart from the common Grameen pattern in the substantial degree to which their voluntary savings exceed the contractual savings required. This is evidence of an unsatisfied demand for safe, accessible and affordable savings services among low-income people in Fiji. Aglow’s savings mobilization performance is the more striking in that Aglow pays no interest on savings balances. It succeeds in retaining former borrowers as savers and has a significant proportion of members who regard the savings facility as the primary advantage of membership. It is quite opportunistic in regard to savings mobilization, visiting market-places and collecting directly from vendor-members in situ. Market activity reaches a peak after paydays, usually on Thursday, with high turnover in the period to the end of the weekend. At this time, of course, the banks are closed. Market-vendors appreciate being able to put aside money quickly after its receipt. MIS and operational performance Aglow has a computerized management information system (MIS) which permits tracking savings, lending and repayment performance of each of its three sub-units, as well as aggregate performance. Reports are prepared weekly, and are normally available within a fortnight. There is a single loan product, a 24-week loan with weekly repayments incorporating a 25 % interest charge. The data permit the calculation of Portfolio at Risk (PAR) and a cumulative repayment rate, reported as 31% and 96%, respectively in January 2003. The detailed discussion of Aglow’s operational performance in Annex B canvasses the usefulness of these measures as indicators of performance. It suggests the need for Aglow to collect data on the age-profile of arrears and to calculate measures of current rather than cumulative performance, such as the ‘on-time’ collection rate. It recommends Aglow should ‘bite the bullet’, by writing off the arrears remaining on its balance sheet from the early unfortunate experience of default. These inflate the PAR measure to a degree which misrepresents current portfolio experience and quality. From an inspection of the Aglow Microfinance Unit’s operational data and some limited access to accounting statements, certain conclusions emerged concerning Aglow’s portfolio quality, outreach and sustainability. These are documented in greater detail in Annex B. Briefly: Concerning sustainability, it was obvious that operating expenses exceed income by a considerable margin. For instance, personnel outlays for 2002 amounted to some $F84,000, with total operating expenses in the order of $F112,000, as against interest income of less than $F33,000. The NMFU reports lending costs for the Nadi operation at 28 cents in the dollar at end-2002, and at 39 cents for Lautoka (although it is not known whether or how headquarters costs were allocated in making this calculation). Costs for the newly-established Ba sub-unit were much higher at $1.19 per dollar. There is scope for greater efficiencies and for greater scale economies, but by any measure the operation is expensive. As to portfolio quality, the Aglow operation appeared to be much superior to the IDL program of the NMFU in terms of standard measures of arrears. The cumulative repayment rate 18 is 96%, not a particularly good result for a program with a relatively short-term loan portfolio. Portfolio at risk, using a very strict definition, was calculated at 31% at the time of the preparation of this report, although on examination it appeared that by writing off the overhang of pioneer arrears this figure could be reduced to a more respectable 18% (or even less if Aglow were to adopt a less rigorous definition). Aglow should in addition calculate current collection measures of arrears, such as the on-time collection rate to give a better guide to portfolio quality. The savings performance of Aglow members is impressive. The value to members of this service and the prospects for it to be scaled-up as a part of future expansion should be considered. The accumulated savings balances of some $F104,000 for about 1,000 clients are high, in relation to the size of loan portfolio. Aglow has a client base of slightly more than 1000 clients. It has opened more than 1650 loan accounts since its inauguration, more than 1200 of which had been paid off in full. First loans totaled 1,012, with 427 second cycle loans and 169 third cycle loans over the period of less than three years of AMFU’s operations. A single client had entered her seventh loan cycle in that period. Taking into account its good savings performance, this is a respectable level of outreach with strong evidence of repeat business. 19 IV. OTHER MICROFINANCE PROVIDERS The preceding section was concerned with the NMFU as the agency set up by GoF in a conscious, supply-leading, initiative to establish a sustainable MF sector. In that context the Review considered the NMFU’s own MF activities as well as those of the MFIs and ‘village banks’ created in response to the availability of NMFU’s resources. This section surveys other institutions operating in Fiji which are actually or potentially suppliers of MF services. The institutions concerned are a mix of public and private, market-oriented and socially-based, formal and informal, and their characteristics are summarized in the taxonomy presented in table 1.1, above. COMMERCIAL BANKS As previously described, there are five full-service commercial banks operating in Fiji, all of which are private and foreign-owned. The GoF retains a minority stake in the former state commercial bank (formerly the National Bank of Fiji, or NBF, and now the Colonial National Bank, or CNB). There is now no commercial bank with a commitment to serving non- commercial market segments such as low-income and rural populations. This should be qualified to the extent that CNB sees retail banking as a priority market and that it appears more receptive than its major competitors to trialing new approaches to achieving rural outreach. As a legacy of its former status as the state commercial bank, it has greater representation in rural areas than the other banks, with an arrangement with Post Fiji for basic passbook services to be conducted via rural postal agencies. However, the introduction of private franchise operations in Post Fiji has seen many rural service points closed. Hence, rural outreach has been reduced in recent years. More generally, the number of all banks branches and agencies has been falling since the 1990s, with the introduction of electronic banking To appreciate the approach of commercial banks in Fiji to the commercial opportunities presented by MF, it is relevant to consider some findings on the industry in the report of the 1999 Committee of Inquiry into Financial Services.11 The evidence cited here is relevant to the challenges and opportunities faced by the broad category of non-bank financial institutions (including MFIs, credit unions, FDB and others) which serve lower-income groups. Judgments cited by the Committee appear broadly correct. First, the Committee characterized the market as an oligopoly, dominated by two major foreign banks. It noted that all the banks enjoy high profits and healthy returns on equity and that these returns exceed those earned on the 11 Ministry of Finance (1999). Report of the Committee of Inquiry into Financial Services: Suva: Parliament of Fiji,Parliamentary Paper No. 19 of 1999. 20 consolidated (i.e. foreign) operations of the banks. The observed high levels of profitability reflect the privileged position and the lack of competition of the banks, which ensures them a captive market. High profits, in the Committee’s view reflect the high interest rate spreads and margins and the high levels of fees and charges in the industry. With regard to the impact of this situation on the poor, the report concluded that there “appears to be an insufficient number of products and services for lower income earners who have difficulty meeting opening or minimum balance requirements. Nor are they able to access certain loan products that may suit their needs, as they lack security for them. Further, the Committee expressed concern at the apparent trend of the reduction of branches and agencies across the country. This is causing hardship to many people in rural areas as they have to spend considerable time and money in accessing banking services.� The report deplored the move by banks to “user pays� provision of services (said by it to be forced on banks in other countries by greater competition in the markets of those countries, but not justified by the situation in Fiji) on the grounds that the charges levied disadvantage on lower-income people, being regressive in their incidence. Indeed, while moves to user pays are usually defended in terms of some presumed need to abandon cross-subsidies of benefit to lower- income groups under the pressures of competition, the Committee decided that cross- subsidization not only continued to operate in Fijian banking but was actually regressive in its effect. The report concluded that the evidence available to it was open to the interpretation that high transaction costs imposed by banks on passbook savings accounts (the preferred financial product of the great majority of Fijians) were determined quite arbitrarily because the banks were incapable of costing the various joint-products they offer, and that it was their policy to discourage some persons from being involved in banking. Concerning the financial products offered by the banks which could have the character of MF, in terms of lending, it appears that the commercial banks did not, other than in exceptional circumstances, lend in amounts or for purposes that could be considered microcredit in the Fijian context. Certainly, public submissions to the 1999 Committee of Inquiry referred to difficulties experienced in obtaining small personal loans and a reluctance to lend to small business and agriculture. On the side of deposits, the Committee of Inquiry reported that in 1997 there were more than 560,000 savings deposit accounts in Fiji and the Committee suggested it would be reasonable to infer that most residents of Fiji held a savings account. However transaction costs and minimum balance requirements are an issue for account holders, the great majority of whom hold small balances in passbook accounts. Moreover, interest rates on passbook accounts have been very low, and in the current monetary conditions of Fiji they actually carry a zero rate of interest. The highly liquid state of bank balance sheets and the sluggish market for loans (or at least the lending which interests banks) means that commercial banks are indifferent to the small deposits of low-income Fijians. As to fees and other conditions, CNB requires an account-opening deposit of $F100 and a minimum operating balance of $F200 to avoid a monthly F$2.00 charge. Depositors pay $F2.00 per withdrawal if they take out money more than twice a month, and incur a fee of $F30 if their account is dormant more than 12 months. A replacement passbook costs $F25. Westpac, one of the two dominant commercial banks, requires a minimum monthly balance of $F1,000 to avoid monthly charges, allows one free withdrawal per month and then charges $F2.00 per withdrawal thereafter. Remembering that the Aglow clientele has a mean balance of only around $F100 per 21 member, it can be seen how such transaction costs would eat into passbook accounts of such size. It seems clear that, as the Committee contended, banks are trying to discourage passbook accounts in favor of statement accounts which can be accessed via ATMs (and for which charges are substantially less). But these alternative accounts are not feasible for many rural people because of distance from ATMs, difficulties with the delivery of mailed statements and the problem of keeping PINs confidential. For a variety of reasons, many Fijians have to operate a bank account, even though their balances are low for most of the time. In consequence, they find fees diminishing their meager balances. On the issue of charges for dormant or inactive accounts, the Committee of Inquiry was quite critical, noting that a charging policy which actually reduces the balance of a passbook account is a discouragement of savings. This outcome is even more likely under the conditions of 2003, when deposit rates have fallen to zero. In more isolated areas, Post Fiji offers basic deposit and withdrawal services for passbook account holders of CNB. It also accepts insurance premium payments and bank loan repayments. However, new accounts cannot be opened by Post Fiji. In terms of coverage, it had 51 Post Offices and 104 postal agencies in 2002. The services were available at 32 of the 51 post offices and at 20 of the 83 postal agencies classified as cash accounting agencies. More than 40 cash accounting agencies had been suspended from offering the service, suggesting that the process encounters difficulties in isolated locations. FIJI DEVELOPMENT BANK FDB is the primary development financing institution in Fiji and operates under its own legislation. It held around 6% of financial sector assets and liabilities in the second half of the 1990s although it was responsible for about a quarter of total financial sector lending and was the largest lender to agriculture.12 It has 10 branches in main provincial centers, giving it greater geographic outreach than any of the commercial banks other than CNB. A recent review of the FDB by ADB13 describes FDB as having had “minimal profitability in recent years due to lack of equity funding by the government as owner, high funding costs because of reliance on government-guaranteed bonds, high loan provisions due to past loan losses, and high operational costs.� The review also notes that FDB’s loan portfolio continues to decline from its peak of $363 million in 1999 to the current (end-March 2002) level of $303 million. FDB’s customer base has fallen from more than 10,000 in the early 1990s to about 6,000 in 2002. It is apparent that FDB has its own problems. The ADB review describes government’s goals for the FDB as “having a locally owned development bank that can be an efficient, sustainable, well-managed distribution channel for government financial assistance by way of grants, interest rate subsidies, and guarantees for micro, small, and medium enterprises, 12 Asian Development Bank (2001) Financial Sector Development in the Pacific Developing Member Countries, Vol 2, Country Studies. 13 Asian Development Bank (2002), Review of the Fiji Development Bank. Technical Assistance No. 3436, July 2002. 22 particularly those that commercial banks consider an unacceptable risk.� This is a set of goals that, as the ADB review notes, involves “the conflicting goals of development, social responsibility, and the need to achieve profitability to ensure a sustainable bank.� The review suggested that the government and the FDB board more clearly define the parameters within which the bank is to operate – what it can do and what it cannot in terms of markets, loan sizes, products, and expected levels of profitability. SAVINGS AND LOANS COOPERATIVES S&Ls are registered under the Cooperatives Act and administered by the Department of Cooperatives. In rural areas they fall into two broad groups: those in sugarcane areas and those serving other farmers. The former are financially stronger by virtue of the fact that their peak body, the CCSLA, has the right to take liens over the crops, land and housing of their members. Enforcement of this system is facilitated by the production contracts entered into by growers with the sole buyer, Fiji Sugar Corporation. Its lending rate is capped at 13.5%, plus fees, for loans within the range of $F500 to $F10,000. Loans are for production purposes, but also for consumption and provident purposes. CCSLA regards itself as capital-constrained and claims it cannot finance all the bankable proposals it receives. It provides interlending services for its 97 member cooperatives, most of which are very small, localized groups of farmers. CCSLA produces a consolidated balance sheet for the movement. Deposits are accepted, on which 2.0% per annum interest is paid. Term deposits attract 0.5% above bank rates. The CCSLA is recognized in the legislation and hence is in a much stronger position than the Credit Union League. CCSLA also has some 4,000 individual farmer members, not associated with primary units, who transact business directly at the HQ in Lautoka. It accepts deposits from non-members, townsfolk who also transact directly at the head office. The minimum initial deposit is only $F20, and so long as this minimum balances is maintained there is no lower limit on withdrawals. In comparison with those of banks operating in the town, these are attractive terms. CREDIT UNIONS The Credit Unions (CUs) have been in decline, especially in rural areas, and the predominance within the CU movement of unions based on salaried urban workers has increased. This trend reduces the current relevance of the CUs for MF, as defined in the Fijian context. The apex body, the Credit Union League (CUL) has been unable to maintain the interlending mechanism for CUs and is substantially in debt to its member institutions. A small number of substantial CUs are not members of the CUL, depriving the movement of resources while, within the fold, membership and assets are skewed to a relatively small number of larger CUs. The number of CUs associated with the CUL has declined from close to 180 in 1991 to fewer than 50 in 2002. In the 1990’s, members peaked at more than 27,000 in 1993, but was little more than 10,000 in 2002. In that year, member shareholdings amounted to almost $F14 million, loans to members were some $F15 million and assets totaled $F18 million. The CUL is attempting to recover ground in rural areas, starting savings clubs as pre-qualification training for 23 CU status, an activity which has attracted the attention of the NMFU and has been discussed by the two organizations as a possible area for cooperation. The NMFU stated it now is prepared to consider assisting existing rural CUs where they satisfy NMFU’s criteria. However, detailed proposals from the CUL to the NMFU in 2001 for assistance with capacity-building and a capital grant for the revival of interlending had no results. The CUL will continue to seek capacity- building of its own staff in order to better assist members. CUs have their own legislation dating from 1954 which is administered by the Department of Justice. The CUL (for whose existence or activities no provision is made in the Act) regards the responsible department as lacking expertise to support the development of the credit union movement, and the legislation itself as no longer relevant to the needs of CUs in the contemporary Fijian economy (mentioning, inter alia, the constraint of interest rate caps on its capacity to self-finance growth). It would welcome closer association with the NMFU, claiming never to have received any financial assistance from government and making the point that the CU movement has been providing financial services to a broad stratum of Fijians since the 1940s. The Committee of Inquiry into the Financial System urged in 1999 that new legislation be brought forward and the CUL has prepared a draft bill to advance the debate. The Government’s Strategic Development Plan 2003-2005, discussed in section IV, below, sets 2004 as the target date for legislation of a new CU Act. INFORMAL MONEYLENDERS This is an opaque area, as in most societies. There is a long tradition of money lending in both rural and urban areas, and especially in the cane growing regions (so that one of the motives for the establishment of the CCSLA was to free members from their obligations to moneylenders). Despite this and other initiatives, informal money lending at high interest rates persists, often in the form of payday lending. This can involve the surrender of passbooks to the moneylender with prepaid withdrawal slips, or ATM cards with PIN numbers. This acts as security in the case of workers whose pay is automatically credited to bank accounts and enables the moneylender to operate the accounts. According to the Committee of Inquiry, some banks levied special charges on moneylenders who presented multiple passbooks for withdrawals. Moneylenders passed on this charge to borrowers. FINANCE AND HIRE PURCHASE COMPANIES Finance companies are represented by a leasing company, a home finance lender and a merchant bank. None of these is active in small lending. Purchases of private vehicle and consumer durables are commonly financed by supplier credit or hire purchase companies, incorporated with the Registrar of Companies for that purpose. Although in some countries in the region (e.g., Indonesia) finance companies provide a variety of lending products for quite small purchases which are competitive with microcredit, this is not the case in Fiji. 24 REGISTERED MONEYLENDERS These are governed by the Moneylenders Act and registered with the Registrar of Moneylenders. The legislation caps their interest rate at 12% per annum. The Committee of Inquiry called for a review of this 1938 Act on the basis that its social utility and enforceability were both doubtful. The Act does not apply to MFIs, but the cap has probably prevented the emergence of a more transparent and competitive process of commercial money lending, such as is seen, for example, in the Philippines, where pawnbrokers and other small registered commercial lenders offer services comparable and competitive with those of MFIs. 25 V. THE ENABLING ENVIRONMENT FOR MICROFINANCE LEGAL CONSTRAINTS ON MICROFINANCE PROVIDERS Two earlier surveys of the legal issues have been conducted in Fiji.14 These surveys were complemented with a review of the existing statutes and an opportunity to discuss relevant legislation with officers of a number of institutions. The main point to be made about the legal/regulatory environment for MF in Fiji is that the law does not explicitly take account of activities such as those conducted by the newly-introduced MFIs such as FCOSS and Aglow, which originate in the voluntary sector and have charitable purposes, yet carry out functions analogous to some of those performed by regulated financial institutions. The Strategic Development Plan 2003-200515 states a commitment to rationalize the financial services regulatory regime, to include establishing MF regulations and a revised credit union act by 2004. In the case of MF, there are grounds for questioning how far this is necessary. While it is possible to read obstructions into various existing pieces of legislation, which might conceivably pose difficulties for the credit activities of MFIs such as FCOSS and Aglow, common sense is called for. It is essential to remember that the thrust of official policy in Fiji is strongly in favor of sustainable MF. That alone is likely to discourage zealotry on the part of officials seeking to draw MFIs into the web of some inappropriate statute for whose administration they happen to be responsible. Turning to the implications of existing legislation for MF providers, consider the case of MFIs operated by NGOs. It is appropriate to start with the Banking Act (1995), which defines the primary institutions regulated by the RBF. In Fiji, institutions such as the Aglow and FCOSS MFIs extend credit. They also accept deposits, but only as agents for their clients, who authorize the MFIs to deposit funds on their behalf in a regulated deposit-taking institution, namely a commercial bank. As in most countries, it is the acceptance of deposits by an entity (other than one which has authority to accept deposits under some other legislation, such as the Cooperative or Credit Union Acts) which defines a banking operation and which requires approval and licensing by the central bank. But MFIs, which are not set up to mobilize and to hold deposits, are not performing banking operations in terms of the Act. Nor does the act of extending credit bring them within the Banking Act. The same would appear to apply in the case of community- based organizations, such as the village and tikina banks, which the NMFU is piloting. Secondly, group lending based on joint guarantee, or other modes of lending without physical collateral, are common to most NGO MF operations, though some element of 14 See the studies by UNDP/UNOPS and Belloni, referenced in footnotes 6 and 7, above 15 Page 89. 26 compulsory saving may serve as a guarantee (usually only partial) of repayment. However, the collateral requirements of banks in Fiji are a matter of commercial practice, not of the Banking Act, so the lending operations of MFIs are not constrained on that score. The central bank’s own legislation, the Reserve Bank Act 1985, empowers it to specify minimum security to be held against bank lending, or that of other credit institutions or individuals as necessary, but none is currently in force. Similarly, under its Act, the Reserve Bank may set interest rate ceilings and other controls on credit by banks (or other lenders, as above in the case of security) but since the late 1980s interest rates in Fiji have been deregulated and MFIs have no more reason to fear a reversal of this policy than any other lending entity. Exceptions exist: certain lending institutions operate under specific legislation; relevant cases are discussed below. There may be a difficulty in regard to interest rates charged in the direct lending of GoF funds by NMFU, as a government entity, or by WOSED (now inactive, as it happens) as a hybrid public/private entity. Thus, there is a question whether NMFU’s individual lending program should be subject to interest rate ceilings specified in the Finance Act 1981. But the conclusion of this Review is that individual lending such as that of NMFU is not in any case good practice MF. Further, NMFU itself is set to be absorbed within the NCSMED, which has its own legislated authority to lend, should it decide to do so. The issue does not arise in the case of funding provided to MFIs by NMFU, since these have been grants. If the Banking Act imposes no constraint on the lending of MFIs, and if the Reserve Bank appears unlikely to do so, it is necessary to consider whether other legislation constrains their lending in any way. MFIs originating from the NGO movement are registered legal entities mandated under the Charitable Trusts Act 1945 to serve charitable purposes. Eleven such charitable purposes are listed in the Act, of which two seem relevant to MF. The first is the encouragement of skill, industry and frugality, a purpose supported elsewhere in the Act by a reference to the furtherance of thrift and the improvements of standards of living by education and advice in matters of self-help. Should this fail as a sufficient justification for engaging in MF, there is a second option. An activity may be justified by such other purpose as may be declared by the Minster to be a charitable purpose. If Government is sufficiently supportive of MF as a matter of policy it should be possible to put this matter beyond question. If this is among the measures in the minds of the drafters of the Strategic Development Plan then the plan is to be welcomed. In practice, the right of NGOs to conduct MFIs for charitable purposes has not been challenged. Interest rate caps apply in the case of certain institutions with their own legislation. Thus, credit unions are restricted to charging 1% a month on the actual balance, as are licensed moneylenders (though in the case of the latter there is doubt as to whether this is being enforced or is enforceable). The 20 Year Development Plan16 for indigenous Fijians refers to plans to take action to regulate interest rates charged by informal moneylenders in rural areas. This is a difficult issue, posing a danger that the activities of MFIs might become caught up in an unintended and counterproductive way. So far as income tax is concerned, NGOs as charitable entities are exempt, as are the credit unions and S & Ls, each by virtue of its legislation. 16 Pages 250, 257. 27 In the case of credit unions, the Credit Union League (CUL) has criticized the Credit Unions Act 1954 as inappropriate to current conditions. According to CUL, the current legislation does not encourage the CUs to act as efficient financial intermediaries, nor does it set appropriate prudential standards for capital reserves. CUL believes that deposit and loan rates need to be determined by the membership of each CU in accordance with current market conditions. The Act makes no provision for CUs to have a peak body (unlike the situation in the cooperative movement) and the registration and supervision functions for primary unions are inappropriately located, by historical accident, in the Department of Justice. The CUL has drafted proposed legislation providing for regulation and supervision of the movement by a specialized unit located within an appropriate ministry, to increase the professionalism of the CUs, and to enable them to perform the function of financial intermediation more effectively in the interests of their members. Government is committed to revision of the Act by 2004. By contrast with the situation of the credit unions, the Cooperatives Act was revised extensively in 1996, entrenches the functions of peak bodies, and is regarded as more suitable for current conditions and the objectives of the member cooperatives. Aside from the case of the credit unions, whose legislation has been before parliament without resolution in various forms since the late 1990s, there does not seem to be an urgent case for regulatory reform. Drafting comprehensive regulations to cover the circumstances of MFIs may create more problems than it is designed to solve, especially when a few key determinations or regulatory decisions made on pragmatic bases may be enough to clarify the status and permissible activities of MFIs for the time being. The strongest argument against any attempt at comprehensive regulation is that its premature imposition on an embryonic industry may well stifle the creativity and experimentation necessary for workable Fijian models of MF to emerge. GOVERNMENT POLICIES RELATING TO MICROFINANCE A review of GoF policies in support of the MF sector in Fiji must necessarily encompass ME development. MF can be seen as instrumental in building a financial sector capable of serving the needs of all members of a population, but ME development (for which MF services are an essential input) provides the opportunity for sustainable livelihoods for lower-income people outside the formal sector of the economy. Both MF and ME development have received a great deal of attention in recent policy statements by the GoF and the two are intertwined in official thinking. Further, a review of government policies for MF cannot confine itself to narrowly- defined MF institutions (MFIs). This is because the only financial service available to the great majority of poor and rural Fijians is provided by the commercial banks, through their passbook savings accounts (the great majority of which have genuinely micro balances). For this reason it is necessary to consider government policies concerned with rural banking services and also to consider a range of other institutional possibilities for provision of financial services to the poor and low-income. 28 ASPECTS OF THE MACROECONOMIC ENVIRONMENT RELEVANT TO MICROFINANCE A long-term trend in the macro economy is of particular significance to the MF sector. This is the persistent decline in investment as a proportion of GDP. This decline slowed employment growth and thus sparked GoF interest in ME development as a solution to rising poverty and unemployment, as described below. The Reserve Bank noted that, during the period 1966-1987, economic growth was led by investment averaging around 22% of GDP. However, over the period 1988-1999, investment averaged only 13% of GDP. The decline in private sector investment has been particularly marked, down from 13.6% in 1985, to 7.5% in 1988 and to 3.5% in 2000.17 The period since the first coup in 1987 has been one of political and investment uncertainty, which has greatly influenced the lending environment for the banking system. For some time, banks have perceived a dearth of good lending opportunities in the corporate sector and have been highly liquid. The trend of interest rates was in general downward during the 1990s, but as the Committee of Inquiry into the Banking System noted, without affecting gross margins between borrowing and lending rates for the system, which remained highly profitable. Currently, high liquidity in the system is exacerbated by the return, at GoF direction in response to the coup of 2000, of the overseas investment funds of the Fiji National Provident Fund. This has impacted deposit rates to the point where the weighted average of rates on savings accounts in the commercial banks was 0.56% in November 2002 18 and the rate was actually zero on passbook savings of small depositors. As has already been noted, this amounts to a negative real return on “micro� savings balances, after inflation and bank charges are taken into account, and creates a demand for lower-cost saving services such as are offered by MFIs and village banks in Fiji. This situation is likely to continue so long as overall economic growth remains sluggish and the banking system remains highly liquid. POVERTY IN FIJI The Fiji Poverty Report (UNDP 1997) focused attention on the rising incidence of poverty, estimated to affect a quarter of Fijian households, two-thirds of them in rural areas. This was seen as accompanied by a weakening of traditional social safety nets, and as due in part to the rapid growth of urban populations increasingly divorced from customary land ownership and the subsistence sector. While these changes were occurring, a growing discrepancy was apparent between numbers of job-seekers entering the Fijian labor market and the capacity of the formal economy to create employment opportunities. The Poverty Report heightened awareness of the need to stimulate self-employment and income generation in informal economic activities. It inspired the first National Conference on MF, mounted by the Ministry of Finance in 1998. This conference led in turn to the articulation of a National MF Action Plan in 1999. Meanwhile, a Women’s Plan of Action for the period 1999-2008 was prepared (Ministry of Women and Culture 1998) as the Government’s response 17 Reserve Bank of Fiji, Economic Growth and Employment in Fiji, Reserve Bank of Fiji Quarterly Review, December 2001, pp30-31. 18 Reserve Bank of Fiji Quarterly Review, December 2002. 29 to the UN World Conference for Women, in Beijing, 1995. This focused, inter alia, on mainstreaming ME development for women into government policy and programs. It proposed specific action to strengthen and create semi-formal financial institutions (or MFIs) providing financial services to low-income women. The first such MFI in Fiji, WOSED, had been operating with substantial support from the Ministry of Women since 1993. In 1999, the report of a Committee of Inquiry into Financial Services (Ministry of Finance 1999) drew attention to the limited availability and high cost of financial services for low-income and rural people. The coup of May 2000 dramatized the sense of marginalization felt by rural indigenous Fijians who perceived urban and communal bias in the distribution of the benefits of development. MF and ME development were among the policy instruments then selected by Government to advance the relative economic position of the indigenous people of Fiji, and for the alleviation of poverty more generally. Among the outcomes of the national conference on MF, mentioned above, was the NMFU. This was created in 1999 as an element in the National MF Action Plan and has continued as a pilot project of government. Initially, the NMFU was placed within the Ministry of Finance, but a change of government saw it relocated within the Ministry of Commerce, Business Development and Investment. NMFU has in its vision statement the goal of a sustainable micro finance sector/industry in Fiji supported by an integrated and enabling environment. However, for most of its short life the NMFU has devoted its resources primarily to funding and supervising the operations of two MFIs managed by Fijian NGOs and to conducting a program of lending to individual micro entrepreneurs on its own account. Its focus has thus proved to be more operational than systemic, and has been more consistent with a business development than a financial sector philosophy. ROLE OF THE RESERVE BANK OF FIJI In 2001, at the request of Government, RBF prepared two papers concerned with extending outreach of commercial services in rural areas. The first of these described the challenge facing government as one of creating “a rural credit delivery scheme through the banking system or through other alternative frameworks without perpetuating a sense of dependence on government support or a handout mentality." The emphasis in this first paper was on the extension of credit services, without explicit mention of savings services. In a second paper, the central bank examined options for cost-sharing among institutions which might contribute to extending rural financial services. This discussion assumed that commercial banks would be the service-delivering partners, and that the cost-sharing partner would be one (or a combination) of the national government, external donors and the central bank itself. In regard to the formal financial sector, in its Strategic Development Plan 2003-2005 (GoF 2002) the government indicated its support for the progressive liberalization of capital markets to mobilize domestic savings more efficiently and broaden the range of financial products and services available to investors. It expressed concern at the withdrawal of basic banking services from rural areas. The GoF signaled action to redress this imbalance in service 30 provision19 and indicated its intention to conduct a survey of rural demand for banking services in 2002 as a step in that direction. In 2002 the RBF returned to this theme with a new policy paper, which is under consideration by the GoF. It is based on the conviction that “methodologies are available for improving access of rural communities to basic banking services using available technologies� and that “there are ways of financing and pooling the costs of these methodologies but they require a strong lead by the Reserve Bank and a strong spirit of co-operation from the commercial banks.� In summary: To improve access of rural communities to basic banking services, the paper suggests extending PostFiji services to other banks and the related cost sharing arrangements. Other options include a banking industry initiative to establish a banking infrastructure development fund to finance development projects and designing products for the pooled savings of small rural savers who form self-help groups. The paper further suggests installing shared ATMs in strategic rural locations. One initiative that the banking industry can explore is to establish a banking infrastructure development fund. The fund will be used to finance projects that develop the banking infrastructure to develop and increase the client base for banks. One example, could be the use of shared ATMs in the rural areas. This is an area where the Reserve Bank could show its commitment to ensuring availability of banking services to rural areas. Reserve Bank can drive the concept by proposing it to banks, establishing the fund with a grant and then maintaining it in partnership with all commercial banks in Fiji. The fund could be legally established and governed by a Board comprising CEOs of the various banks, Permanent Secretary for Finance, and a representative of the Reserve Bank. Annual contributions to the fund by way of a levy on banks, for example, 0.01% of total assets 20, will maintain the fund. It is striking to see a central bank adopting MF principles wholeheartedly, and also to see the emphasis placed by the RBF on mobilizing savings as well as credit. Thus: Individuals who are not in formal employment form self-help groups. Members collectively pool their financial resources to help with each other’s savings and credit needs. Such groups range from rural credit unions, rural co-operatives and informal savings and credit associations. 19 Pages 88-89. 20 Total assets of the commercial banks were $F2.05 billion in November 2002, which would yield an annual levy of $F205,000. 31 To cater for small savers whose individual savings are too low to maintain a bank account, banks should be encouraged to design a product to meet the withdrawal and deposit needs of such people. The product should waive annual administrative fees. The Strategic Development Plan21 also stated the intention to create income-earning opportunities for the poor through MF measures, including:  increasing “micro-lending� to the poor;  establishing a coordinated monitoring mechanism for MF by 2003; and mounting a comprehensive review of all MF programs, by 2005. An affirmative action program outlined within the Plan also calls for women’s access to microcredit to be improved by 50% by 2004. More broadly, the Plan commented on the growth of a number of other MF-based institutions in Fiji, including cooperatives, credit union leagues and moneylenders. Noting that these serve a useful purpose in providing for the needs of lower and middle level income families, it stated the need for appropriate legislation and regulation to protect the interests of these institutions and their borrowers. These regulatory provisions are to be in place by 2004 and will involve the review of existing legislation covering credit unions, cooperatives and moneylenders. In terms of the broader financial system, the plan calls for a domestic credit rating bureau to be established during 2003. It is not stated whether this rating service would cover clients of MF institutions. As previously noted, the GoF commissioned a review of the Fiji Development Bank by the ADB in 2001, and received the report of this review in July 2002. The Strategic Development Plan commits government to evaluating and acting on the recommendations by 2004. The review pointed to some conflict among the goals of development, social responsibility, and commercial viability held for the bank. It concluded that ultimately it is for the GoF and the board of FDB to define the parameters within which the bank is to operate, with regard to target markets, loan sizes, products, and expected return on assets. It concluded that to address funding costs and compliance with commercial banking norms, the FDB should proceed to obtain a deposit-taking license, increase loan provisioning to the level recommended by the external auditors, and come under the supervisory umbrella of RBF. However, the discussion of FDB as a deposit-taking institution has never envisaged it as a provider of retail transaction services, but only as offering term deposits as a means of lowering its cost-of-funds. NATIONAL CENTRE FOR SMALL AND MICRO-ENTERPRISE DEVELOPMENT In 2002 the GoF passed a Small and Microenterprises Act, which defined ME as any enterprise with turnover or assets not more than $F30,000 and with not more than five employees. The legislation defines the two-fold objective of NCSMED, the National Centre for Small and Micro-Enterprise Development, as being to facilitate access to funds and to complement and supplement financing programs for small and microenterprises. Among its functions is to provide or source loans or grants to small and microenterprises. In carrying out this function, its Board may approve loans or grants to small and microenterprises. With the 21 Page 39. 32 approval of the Minister, the Board may borrow money from any source to carry out NCSMED’s functions. Finally, the Minister may prescribe procedures for applying for, issuance, and conditions of loans and grants. NCSMED will have divisions for Research and Development, Training, Treasury, and MF, as well as Corporate Services. The Treasury division will, inter alia, monitor and manage trends in SME lending activities and provide advisory services in finance to second tier institutions and SMEs. The MF division will have monitoring responsibilities, work to increase outreach, and assist MFIs to achieve financial sustainability, among other things. Financing functions appear to reside with Treasury, while the MF division will assume the role of capacity- building and monitoring. NCSMED will not work directly with SMEs in any of its operations, but will deal with so-called second tier organizations, defined as organizations that service SMEs directly. In regard to ME and the informal sector, the Strategic Development Plan22 declared GoF’s intention to implement user-friendly policies, laws and regulations that affect SMEs by 2003.23 These will be designed to remove, inter alia, restrictive town ordinances which impede the activities of informal and ME unnecessarily. With technical assistance from ILO, work had commenced in 1999 to prepare a policy framework for small and microenterprises in Fiji (ILO, nd). This had a number of outcomes, including the Small and Microenterprises Act (No. 1 of 2002) which provided for the establishment of NCSMED. This commenced operating at the start of 2003 and has important responsibilities under the Strategic Development Plan. The Strategic Development Plan calls for NCSMED to absorb the NMFU and for NCSMED then to play an important role in the provision of business support and micro finance to small and microenterprises in future. For this purpose it will also absorb the Small Business Advisory Unit now located within the Ministry of Commerce and Business Development, with the intention of creating a one stop shop for small and microenterprise development. However, it also possesses significant financing powers under the Act. The legislation enables it to facilitate access to funds and to complement and supplement financing programs for small and microenterprises. More specifically, it may provide or source loans or grants to small and microenterprises. Subject to ministerial guidelines, the Board of the National Centre may approve loans or grants to small and microenterprises. In parallel with the Strategic Development Plan 2003-2005, the GoF has also promulgated a long-range affirmative action plan. This is the 20-Year Development Plan (2001- 2020) for the Enhancement of Participation of Indigenous Fijians and Rotumans in the Socio- Economic Development of Fiji (subtitled 50/50 by Year 2020). This has elements relevant to financial sector development for the indigenous people and for their participation in ME. Measures proposed include establishing a compulsory savings scheme for indigenous Fijians, with balances held in the names of individuals to be available to them as equity for commercial undertakings, including the acquisition of capital goods or assets with income-generating 22 Page 49. 23 In Fiji, the acronym SME is used for ‘small and microenterprises’. This has to be distinguished from international practice where SME is used for ‘small and medium enterprise’. This report adopts Fijian usage. 33 potential. However, since the scheme would involve deductions from earnings at source, it is not clear how this could apply to the majority of indigenous Fijians outside formal wage employment. The 20-Year Development Plan also canvasses the range of grants and subsidized credit schemes available to Fijians through the Fiji Development Bank. With regard to microcredit as a means of financing the enterprises of the lowest income groups, it points out the anomaly occurring under current government policies (p.144): that those who receive MFI assistance, who tend to be the most disadvantaged in society, have … to bear the full cost of delivery to assure the long term sustainability of the programs… [while] those who receive FDB financing, who by definition are economically better off, actually pay a … lower interest rate. In fact, such instances are not confined to the programs of the FDB. Other examples of small grants for agriculture and business development are found in the programs of a number of ministries and compound the inconsistencies within the system. The 20-Year Plan also refers to the prevalence of informal money lending in rural areas, noting that commercial and official sources of credit are often not readily available in such communities. It states that measures will be taken to regulate such providers to ensure costs are not exorbitant. It may be a mistake to read too much into the comment about exorbitant interest rates, given the commitment of successive Governments since 1987 to interest rate liberalization (Ministry of Finance 1999). However, the suggestion does raise the question of whether any return to interest rate regulation could have the unintended consequence of inhibiting the development of sustainable MF operations. A more comprehensive review of financial sector legislation was promised in the Strategic Development Plan. It will include implementing recommendations of the Committee of Inquiry into Financial Services 1999 having to do with the commercial banking system, phased over 2003-2005, and revision of legislation for credit unions and consumer credit, and regulations for MF, all by 2004. CONCLUSION The above chronology of policy indicates how ME development and MF, with the latter considered primarily as the source of finance for ME, have been intertwined in GoF policy development in the period since the publication of the Fiji Poverty Report in 1997. It is clear that the GoF has committed to MF as a key policy tool in the achievement of its objectives for employment creation and poverty alleviation. Inconsistencies remain in the policy framework, among which the parallel promotion of expensive microcredit and cheap finance, the latter in the form of grants and credit subsidies, is one such.24 Another is the suggestion in some quarters of forms of compulsion, such as compulsory savings schemes as a remedy for the presumed lack of a savings culture among 24 For example, small ‘Self-Help Grants’ are available under a program administered by the Department of Regional Development with a budget of $F3 million for 2003. This compares with the NMFU’s 2003 budget allocation of $F1 million for grants of loan capital. 34 indigenous Fijians, and the suggestion of interest rate regulation for moneylenders as a form of protection against exploitation. Neither of these offers a remedy for financial services deficiencies affecting low-income people in Fiji, and could indeed be counter-productive. Comparatively little has been said of Government policy concerning the role and potential of the Fiji Development Bank as a MF provider. This is partly because of a question as to whether the role of a development bank as defined for the FDB at its establishment is consistent with a MF role, but primarily because the FDB itself appears to be in a state of flux. It has recently had a change of CEO while awaiting government’s response to the review conducted by ADB in 2002. Two significant strands of policy have emerged from the process of policy development in Fiji since 1997, outlined above. Both pose questions which cannot readily be answered at present, but which need to be addressed in the final section of this Review: One is the establishment of the National Center, NCSMED, which embodies the strategic, capacity-building and financing aspects of ME development. If it were permitted the full and literal exercise of its legislated financing powers it could become extremely influential in defining the framework for ME financing, and MF more generally, in Fiji. However, there is the potential for conflict of interest in having the functions of nurturing and financing ME under one roof, and managing this situation well will require transparent procedures and good judgment. It is essential that that NCSMED’s financing functions should be exercised responsibly and with awareness of their broader systemic ramifications. Perhaps the key to this is vigilant oversight of the Centre’s policies by its Board, as well as the commitment of management to fostering sustainable microfinance in Fiji. The other is the active role foreshadowed by the Reserve Bank in its policy proposals for increasing the outreach of financial services in rural Fiji. In the context of a highly profitable commercial banking industry which appears to have actively discouraged the participation of rural and low-income people in the banking system, it proposes cost-sharing arrangements for the extension of services based on a combination of electronic technology and self-help group methodology, with a strong emphasis on mobilizing the savings of poor and rural people. A comment was made above, about compulsion being a less than ideal method for achieving system-building objectives. The RBF would do better to achieve its objectives by exercising suasion, backed by political will, rather than compulsion of the banks to achieve its objectives. Similarly, the GoF would do better to raise the propensity to save of indigenous Fijians by assuring the availability of convenient savings facilities bearing positive real rates of interest than by any form of compulsory saving requirement targeted at one community within Fiji. 35 VI. CONCLUSIONS AND RECOMMENDATIONS ISSUE: CONFORMITY OF GOVERNMENT-SUPPORTED MICROFINANCE OPERATIONS WITH RECOGNIZED GOOD PRACTICE Conclusions Since 1999, GoF financial support for MF has been disbursed primarily via the NMFU. Systemic or sector-building objectives underlay the planning for NMFU. However, operational objectives were pushed to the fore by the decision to allocate substantial amounts of loan capital to NMFU and to set a short time-span for the initial program. This suggests a degree of urgency in the decision to launch MF quickly, and with an almost exclusive emphasis on microcredit. In consequence: The NMFU has not realized aspirations for it to play a systemic role in working towards the development of an integrated framework promoting a sustainable MF sector. Instead, its activities have reflected a narrow definition of the NMFU mandate and have contributed comparatively little to wider sector-building. On evidence of the unit’s performance, GoF’s plan for NMFU to be absorbed within the NCSMED appears appropriate. However, to avoid the potential for conflict of interest in having the functions of nurturing and financing ME under one roof appropriate safeguards may be necessary. The NMFU concentrated, at least initially, on encouraging the creation of Grameen Bank replications, acting as a wholesaler of funding from 1999. It added its own individual retail lending operations in 2001, but was slow to grasp the importance of savings or to consider the claims to assistance of existing financial institutions, notably the credit unions. While the Credit Union League quite early requested a grant to bolster its training infrastructure and to provide start-up capital for a central finance facility, or interlending mechanism, this did not come to fruition. More recently, however, there has been some broadening in the scope of NMFU operations. Thus: As a result of its experience of direct and indirect lending programs, NMFU has come to a greater appreciation of the importance of savings, and has attempted to find ways of cooperating with established institutions, including local authorities in rural areas (village and tikina banks) to extend the outreach of MF. It has not yet found means of assisting the credit union movement. In connection with its disbursement mechanisms for the programs it supported, the NMFU was found wanting. Thus: NMFU disbursed only around a quarter of approximately $F5.85 million in loanable funds made available to it by the GoF in the period 1999-2002. Disbursement in the last of these years rose to 80%, albeit on a reduced capital funds allocation. 36 Difficulties in disbursement were largely the result of limited absorptive capacity in the NGO sector where NMFU concentrated its efforts in the first three years. The way to break this constraint would have been to find more resources for capacity-building in the sector. However, For whatever reason, whether because of rigidities in the GoF budgetary system or an insufficient appreciation within the NMFU or the responsible Ministry of the need for capacity- building, no significant part of its substantial loan fund resources was diverted to capacity- building and the bulk of its funding was returned unspent at the end of each of the first three years. And further: In neglecting systemic issues, such as the creation of an enabling environment and capacity-building, the NMFU was unable to establish conditions in which its wholesale and retail MF initiatives might have flourished. Numbers of client institutions were limited and scale-up of their operations was modest. The experience of individual retail lending by NMFU supports the contention that such activities are inappropriate for a public sector agency which is not a regulated financial institution. In terms of collection rates for the funds disbursed and loaned,  NMFU claims an overall collection rate of some 85% on moneys disbursed by all the lending programs it supports. The cumulative repayment rates of the two MFIs to which it wholesales funds were stated to be 95% and 94%, respectively, at end-2002. For its own retail Individual Lending Program (IDL) the repayment rate was less than 62%. The last of these is clearly unsustainable as a lending operation, while the first two appeared marginal. A more detailed examination of the records of one of the MFIs indicated a cumulative repayment rate of 96% in January 2003, with portfolio at risk (PAR) of 31%. However, a judicious write-off of arrears dating from this MFI’s unfortunate early operations would have reduced PAR to around 18%. This would be a much better indication of the health of its current portfolio, while a less rigorous (though still acceptable) definition of PAR would have reduced the measure even further. Measures of on-time collection need to be calculated to confirm this conclusion. Among encouraging aspects of this MFI’s performance were its record of repeat borrowings from among its more than 1000 clients and its healthy savings mobilization. In summary:  There is evidence that MFIs in Fiji operating with group procedures can approach good practice repayment rates while continuing to expand outreach and achieve repeat lending. They can do this while providing group savings services which would be prohibitively expensive on an individual basis, and which are valued by clients. However, on the cost side, the performance of all modes of lending was relatively poor. NMFU calculated a lending cost of 44 cents in the dollar as at end-2002 for the its individual lending, although it is not clear to what extent overhead costs, including senior staff time, were included. By comparison, lending costs of Aglow MFI were said to be 39 and 28 cents in the dollar for its Lautoka and Nadi branches, respectively. It is not known whether or how headquarters costs were allocated between these branches. Whatever the accuracy of these calculations, with MFIs there are issues concerning the costing of donated and voluntary inputs and for no institution has any allowance been made for the opportunity cost of funds from government grants. Thus: 37  The costs of providing MF services in Fiji using either the individual lending or group modes are high. In the case of Aglow, at the scale of operations thus far achieved by it, substantial operating losses are incurred despite an effective annual rate of interest of around 70%. The two MFIs are only sustained by annual capital grants which are depleted by their operating costs. NMFU is now to be absorbed within the NCSMED. However, on the basis of the evidence outlined above, and in relation to good practice in MF:  The continuing claim of NMFU to budgetary resources for loanable funds would be tenuous, were it to continue as a free-standing entity within the Department of Commerce. Once within the NCSMED, that claim would have to be judged anew in light of the mandate of the National Center and the approach it might take in discharging its duties. Recommendations: Conformity with Good Practice, and Sector Building NCSMED has a brief to fund and support “second tier� organizations in MF. This term should be interpreted liberally to include community-based organizations whose principal activity may, at least initially, be savings mobilization. Community-based organizational forms such as the village or tikina banks may prove to be the mechanism by which the most micro of enterprises are financed in Fiji. For NCSMED to support them would be consistent with the broader sector building motives of the central bank in championing such methods. If NMFU were to form the nucleus of the MF Division within NCSMED, it could apply the skills and institutional memory accumulated in its dealings with MFIs and community MF organizations to monitoring the activities of selected institutions and community groups, assisting them to increase their internal efficiency in the functions of lending, collection and savings mobilization, and assisting them to develop a range of financial services suited to the needs of their respective clienteles. Recalling the experience of the NMFU in failing to disburse most of the loanable funds made available to it, and the constraint to effective disbursement which followed from the lack of qualified MFIs, the NCSMED will need to allocate adequate resources to the MF and Training Divisions for the selection and preparation of MFIs and community MF groups and to the subsequent monitoring of their activities. Within NCSMED, the MF Division should work with the Training Division to develop training materials and courses of a generic nature, stressing the achievement of objective performance standards rather than the replication of particular models of MF, but including group methodology as appropriate. Should commercial banks express interest in collaboration with MFIs and community- based groups (for example, by means of “linkage� arrangements), NCSMED should be prepared to participate in pilot schemes and to analyze and fill the training needs associated with such schemes, both for staff of MFIs and of banks. Training activities of NCSMED should have an explicit sector building goal and not be restricted to second tier organizations directly funded by it. 38 NCSMED should also work with the credit unions and S&L cooperatives to develop training materials of value for them. The object should be to communicate insights from the experience of successful MF to those movements, notwithstanding their different traditions and methods of operation. Depending on the national policy framework that emerges from the deliberations of government, due for completion by 2005, deeper collaboration may be contemplated in future. Direct lending by the NMFU to individuals proved unsuccessful. The experience should serve to prevent any repetition of such activities by NCSMED or any other agency of government apart from a specialist financial institution. This is not an appropriate activity for a public sector agency outside the regulated financial sector. Improving the lending and collection performance of MFIs will be a primary activity of the MF Division. The improvement of management information systems, in conformity with the essentials of the CGAP appraisal format for MFIs, should be a priority as a means towards that end. This would assist in defining the objective performance standards referred to in recommendation 4, above. It would also assist NCSMED in its monitoring and evaluation of MFI performance and in its funding decisions. In the evaluation of MFI performance, and in its training activities, NCSMED should give particular attention to issues of governance and transparency. This is important for all MFIs and especially critical for the success of community-based initiatives in MF. In regard to funding decisions within NCSMED, the MF Division would appear, by virtue of its monitoring function, to have an important role in commenting on the capacity and suitability of institutions under consideration for funding. The Research and Development Division, by virtue of its involvement with pilot schemes, and the Training Division, by virtue of its preparation of candidate institutions, would both appear to have interests in the outcome of funding applications. To avoid any suggestion of conflict of interest, NCSMED will need to develop objective criteria and transparent procedures for its funding decisions. On the face of it, MF services provided by MFIs in Fiji appear likely to be expensive, in terms of cost per dollar lent. However, it is the function of NCSMED and its specialist divisions to prepare MFIs to achieve greater scale and more efficient operations in an effort to see how low costs can be pushed. Once the limits of best practice in Fiji have been determined, it is the prerogative of government to decide whether and how public resources should continue to be used to support MF in its various forms. At least initially, the National Centre might need access to external resources to equip it to meet the manifold demands likely to be placed upon it in fulfilling its mandate. 39 ISSUE: STRUCTURE AND MODALITIES OF GOVERNMENT AND DONOR SUPPORT Conclusions Support for MF comes from several agencies within the GoF, both as direct progenitors of programs and as supporters of activities initiated in the voluntary sector. The most notable source of government support at present is the NMFU, but the Ministry of Women has a longer (but not better) track record of direct assistance. While the RBF is currently exerting leadership in policy-formation, both the Ministry of Finance and the Reserve Bank of Fiji have been active in the field in recent years. Some very limited assistance has been received from the private banking sector, relating to community savings schemes. Government also receives assistance from external agencies to promote MF, both official and non-official. The role of multilateral agencies is notable, including UNDP and ILO and (prospectively) the ADB, but bilateral and private external sources of assistance are also engaged. NZODA is perhaps the most active of the bilaterals, but bilateral agencies in general are not significant in the sector. Private external sources of assistance are also active, in the NGO and credit union movements. Agencies with Potential to Take Lead Roles in the Sector Based on the discussion of donor and government activities in section II, above, and elsewhere, a logical division of labour between agencies can be suggested: In terms of funding support, the GoF has set up the NMFU as an apex institution for MFIs and it is to be incorporated in the NCSMED, which has been given credit functions in the legislation. Thought there are several options for the role of the NMFU within its new institutional home, it is a fait accompli that the NCSMED, under the umbrella of the Ministry of Commerce, Business Development and Investment, is to have a central role in relation to MF funding and capacity-building. UNDP has a track record of sustained interest in and technical assistance for MF and especially for MFIs originating within the voluntary or community sectors and with women as the target group. It has strong interest in developing rural savings facilities and also brings a regional perspective which permits the cross-fertilization of ideas throughout the South Pacific region. This is a region where the countries are very small and isolated, but also are also unique from other countries where MF has flourished and therefore regional cooperation is of particular value. ADB, through its Alternative Livelihoods Project, may choose to support the involvement of the cooperative and credit union movements in MF. This is proposed to occur through lines of credit and capacity-building at the level of both primary units and apex institutions. Given ADB’s concern for the establishment of a wholesale facility for MF by Government, it may be possible that ADB is interest in broader policy engagement. RBF is the lead institution for the GoF’s policy thrust to develop rural financial services. It is in a position to scrutinize the soundness of all policy proposals affecting the operations of financial institutions, in the interest of financial sector development. Given the potential significance of that issue for low-income Fijians, it should be regarded as a lead institution for 40 MF development. The RBF could also contribute to the effectiveness of the financial institution- building activities of the NCSMED, as suggested below. Finally, the discussion of capacity-building needs (Annex D, see also text below) suggests a possible role for the World Bank as the agency to coordinate assistance in this area. Recommendations Concerning the Structure and Modalities of Support While support for aspects of MF has been received from a number of government and external agencies, some scope exists for certain agencies to take the lead in particular aspects of assistance and to establish a division of labor. The following arrangements seem appropriate: It appears appropriate for NCSMED, which will incorporate the NMFU, to offer support through its various divisions to MFIs and other second tier MF institutions, for technical assistance and monitoring, training, research and development and piloting, and financing and financial advisory services. It will be a national reference point for operational and funding matters to do with MF and a point of contact for donors interested in the sector. It could also serve as the focal point for capacity-building within the sector. RBF, the central bank, is in a position to scrutinize the soundness of all policy proposals affecting the operations of financial institutions, whether or not they are formally regulated by it, in the interest of financial sector development. It is the lead institution for the GoF’s policy thrust to extend the outreach of rural banking services. Given the significance of that for low- income Fijians, RBF should be regarded as a lead institution for MF development. UNDP has a track record of sustained interest in and technical assistance for MF and especially for MFIs originating within the voluntary or community sectors and with women as the target group. It has strong interest in developing rural savings facilities and also brings a regional perspective which permits the cross-fertilization of ideas throughout the South Pacific region. This is a region where not only are the countries very small but they are also sufficiently different from others where MF has flourished for such regional cooperation to be of particular value. ADB, through its Alternative Livelihoods Project may choose to support the involvement of the cooperative and credit union movements in MF. This is proposed to occur through lines of credit and capacity-building at the level of both primary units and apex institutions and will complement activities conducted by other agencies. Its explicit support for the concept of a central financing facility established by the GoF suggests it may be interested in collaborating with the NCSMED. At the request of the Government, the World Bank could consider coordinating a program of assistance for capacity-building in the Microfinance sector and could work with the NCSMED, as the national focal point for capacity-building in the sector. 41 POLICY FRAMEWORK FOR MICROFINANCE IN FIJI Conclusions Long-term trends in the macro economy are significant for the MF sector. Investment as a proportion of GDP has declined, slowing employment growth and increasing poverty. The decline in private investment has been particularly severe. These trends have influenced the lending environment for the banking system, with banks (focused primarily on corporate and medium-sized enterprises) perceiving a dearth of good lending opportunities. Liquidity in the system has been high and interest rates have trended down although lending margins have been maintained and bank profitability has remained high. One result of the situation: By the end of 2002, micro passbook savings accounts bore a zero nominal interest rate, offering negative real returns after inflation and bank charges. This situation reflected a lack of interest on the part of commercial banks in mobilizing deposits and created a demand for low- cost saving services such as those offered by MFIs and village banks in Fiji. This situation seemed likely to continue as long as economic growth is sluggish and the banking sector remains highly liquid. The Fiji Poverty Report focused attention on the rising incidence of poverty, estimated to affect a quarter of Fijian households, two-thirds of them in rural areas. It inspired the first National Conference on MF in 1998. In 1999, the report of a Committee of Inquiry into Financial Services drew attention to the limited availability and high cost of financial services for low-income and rural people. The NMFU was created in 1999 as an outcome of the national conference on MF. As a government entity it has subsequently conducted micro-lending programs, directly and via intermediary institutions at sustainable rates of interest. In addition to MF per se, other GoF programs involve direct disbursement of funds, whether by grant or soft loan, to potential candidates for MF services. These may have counterproductive outcomes for the sustainability of MF: Government grants and soft loans for ME activities can operate to undercut MF programs such as those supported by the NMFU by sending signals which conflict with the requirement to repay loans made at commercial rates of interest. In 2001, at the request of the GoF, the RBF turned its attention to the issue of commercial bank services in rural areas, focusing initially on credit and on means of cost-sharing among banks, government, the RBF itself and (possibly) donors, to facilitate outreach. GoF’s Strategic Development Plan 2003-2005 indicated support for the progressive liberalization of capital markets to mobilize domestic savings more efficiently and broaden the range of financial products and services available to investors and expressed concern at the withdrawal of basic banking services from rural areas. In 2002 the RBF returned to this theme, promulgating the view that methodologies are available for improving access of rural communities to basic banking services using available technologies and that there are ways of financing and pooling the costs of these methodologies. It is clear that such measures would require political will from the GoF, a strong lead by the Reserve Bank and co-operation from the commercial banks. At the time of writing the RBF’s proposals are still under consideration by government. In summary: 42 The RBF suggests extending PostFiji services to other banks in addition to CNB and related cost sharing arrangements. Options include a banking infrastructure development fund to finance development projects and designing products for the pooled savings of small rural savers who form self-help groups. The RBF further suggests installing shared ATMs in strategic rural locations. The RBF appears to contemplate a quite interventionist role. It would propose the infrastructure fund to banks, establish it with a grant and maintain it in partnership with them. The fund could be governed by a board comprising bank CEOs, the Permanent Secretary for Finance, and a representative of RBF. Annual contributions to the fund by way of a levy on banks, for example, 0.01% of total assets, would amount to just over $F200,000 in 2002. Espousing MF principles, RBF proposes banks should design products for the withdrawal and deposit needs of people organizing themselves into self-help groups. In this category it includes rural credit unions, rural co-operatives and informal savings and credit associations. It suggests these new products should bear no annual administrative fees. Such action would require innovation but successful examples can be found in the financial systems of other countries. The Strategic Development Plan 2003-2005 stated GoF’s intention to establish coordinating and monitoring mechanism for MF by 2003. It called for a domestic credit rating bureau to be established during 2003. The plan stated the need for legislation and regulation to protect the interests of a number of other MF-based institutions. These are to be in place by 2004 and will involve the review of existing legislation covering credit unions, cooperatives and moneylenders. Finally, the plan called for a comprehensive review of all MF programs, by 2005. In summary: The GoF has established the NCSMED which appears likely to play the coordinating and monitoring role for MF. Government will establish a credit rating bureau, although it is not clear whether this will encompass MFIs and other non-bank institutions serving lower-income people; and it will review outdated legislation governing the credit union movement. NCSMED was launched in 2003. The Strategic Development Plan calls for it to absorb the NMFU and to provide business support and MF to small and microenterprises. It also will absorb the Small Business Advisory Unit to create a one stop shop for small and microenterprise development. NCSMED possesses significant financing powers under the Act. The legislation enables it to facilitate access to funds and to complement and supplement financing programs for small and microenterprises. More specifically, it may ‘provide or source loans or grants to small and microenterprises’. Direct lending of this nature by the NCSMED would, however, be inappropriate. The concept of a one-stop-shop to accelerate the development of small and microenterprises has much to recommend it, especially in a relatively small economy such as Fiji. There is the potential for conflict of interest in having the functions of nurturing and financing ME under one roof, and managing this situation well will require transparent procedures and objective decision criteria. The MF functions of NCSMED will need to be conducted responsibly, with awareness of their broader financial sector ramifications, and its Board and senior management will need to be sensitive to these issues. Participation by the RBF 43 in the governance of NCSMED, by nominating a member to its Board, would help to assure the responsible discharge of those functions of the National Centre which have financial sector implications. There is some suggestion in the 20-Year Development Plan for indigenous Fijians of forms of compulsion, such as compulsory savings schemes, as a remedy for the presumed lack of a savings culture among them, and interest rate regulation for moneylenders as a form of protection against exploitation. Neither compulsory savings nor control of moneylender interest rates offers a remedy for financial services deficiencies affecting low-income people in Fiji. Such measures could even be counter-productive in terms of their unintended consequences for institutions attempting to provide MF services on a sustainable basis. Recommendations: What a Conducive Policy Framework for MF Might Look Like in Fiji and How Donors and Government Might Better Coordinate Their Support. Certain factors operate against the achievement of a conducive policy framework for sustainable MF in Fiji. These need to be addressed: The Macroeconomy and Interest Rate Structure. Better macroeconomic performance is needed to enable the removal of barriers to capital outflows, establish an improved outlook for private investment, and produce an increased demand for credit. These will in turn be needed to run down excess liquidity in the banking system. This should rekindle the interest of commercial banks in mobilizing deposits and make them more receptive to GoF efforts to expand the rural outreach of banking services, as well as permitting positive real interest rates to be paid again on small savings accounts. Grants and Subsidies as Contaminants. The comprehensive review of MF programs, promised by Government for completion in 2005, should address the issue of GoF programs involving grants or subsidized loans in amounts and for purposes which are might otherwise be the province of MF lending. Such programs are conducted by several Ministries with affirmative action objectives. The review should attempt to devise, and secure acceptance for, guidelines designed to assure that contamination of MF programs by grants and subsidies is avoided. Certain other regulatory measures need to be avoided, to prevent unintended consequences: Avoiding the Premature Imposition of Comprehensive Regulation. Leaving aside the case of the credit unions, there does not seem to be an urgent case for comprehensive regulation of MF. Drafting legislation to cover the circumstances of MFIs may create more problems than it is designed to solve, especially when a few key determinations or administrative decisions made on pragmatic bases may be enough to clarify the status and permissible activities of MFIs for the time being. The Strategic Development Plan 2003-2005 commits the GoF to establishing MF regulations by 2004, but this should be done with a light hand. The strongest argument against any attempt at comprehensive regulation is that its premature imposition on an embryonic industry may well stifle the creativity and experimentation necessary for workable Fijian models of MF to emerge. Avoiding Interest Rate Caps. Despite concern about high interest rates charged by moneylenders, proposals for capping interest rates for informal lending need to approached with 44 caution. These may have the unintended effect of inhibiting MF operations. A more effective means of protecting consumers from exorbitant interest rates is to develop a range of competitive micro- and consumer finance services accessible to all, which should bring interest rates down and give greater choice to consumers. CERTAIN POSITIVE INITIATIVES MAY BE TAKEN TO LOOSEN CONSTRAINTS ON MICROFINANCE: Fostering a Savings Culture. Indigenous Fijians are widely regarded as lacking a savings culture and some form of compulsory savings scheme for them has been proposed to address this deficiency. However a form of compulsion directed to one of Fiji’s ethnic communities may not be appropriate; instead, measures may be proposed to improve savings performance without compulsion. In parallel with the pickup of economic activity, GoF and RBF could mount a voluntary campaign for personal savings modeled along the lines of the ComeSave and WorkSave programs of UNDP. These employ community and workplace savings groups with some success. The village banks supported by NMFU may also provide a model. UNDP’s Pacific Sustainable Livelihoods Program would be a willing ally in this activity. A personal savings campaign would be consistent with the macroeconomic objective of returning the aggregate investment rate to the levels achieved prior to 1987 and would mesh with the initiatives in rural financial services currently proposed by the RBF. Extending the Outreach of Rural Financial Services. Rural financial service initiatives currently proposed by the central bank deserve support. These include requiring banks to finance a “banking infrastructure development fund� to finance rural service initiatives (to which the RBF would itself contribute), the extension and sharing of Post Fiji rural services for all banks and the extension of shared ATM facilities. Banks are also to be encouraged to design products catering for pooled savings. These measures may appear interventionist although they are within the RBF’s powers under its Act. However, the central bank resources to be expended are small, as is the banks’ proposed financial contribution. The measures appear justified as attempts to reduce financial repression resulting from a banking market where competition for deposits is weak and micro-lending is insignificant. Modernizing Credit Union Legislation. The current legislation does not encourage the CUs to act as efficient financial intermediaries, nor does it set appropriate prudential standards for capital reserves. It would be appropriate for the membership of individual CUs, acting democratically, to set their deposit and loan rates in line with current market conditions. The revised Act should make provision for CUs to have a peak body (as legislated for the cooperative movement) and for the registration and supervision functions to be located within a more appropriate ministry. Government is committed to revision of this legislation, the Credit Unions Act 1954, by 2004. The Credit Union could be an appropriate legal identity for rural self-help groups, such as those the RBF is proposing to encourage, to aim for as they mature. Indeed, consideration could be given to placing the credit union movement under the supervisory control of the central bank as an element in a more comprehensive approach to MF provision. Establishing MFI Access to a Credit Rating Bureau. MFIs should be permitted to file client information with the proposed credit reference bureau. They should have access to the bureau’s data base on reasonable terms, both for their own benefit and for the benefit of commercial lenders. The GoF has announced its intention to have a credit rating bureau operating in 2003. If MFIs were permitted to participate, they could offer clients the incentive of 45 establishing a credit rating. This would provide motivation for better repayment performance as well as reducing the asymmetry of information facing commercial lenders when dealing with small borrowers. Conclusion: An Architecture for the Microfinance System in Fiji A comprehensive review of financial sector legislation was promised in the Strategic Development Plan. It will include implementing recommendations of the Committee of Inquiry into Financial Services 1999 having to do with the commercial banking system, revision of legislation for credit unions and consumer credit, and regulations for MF. This Review has already expressed the belief that the MF sector will fare better under a light regulatory hand. GoF commissioned an ADB review of the Fiji Development Bank in 2002, and has announced its intention to act on the recommendations of that review. ADB itself has foreshadowed assistance to the FDB to provide larger MF loans. This could have value in providing finance for graduating MF clients who have moved to the limits of financing available from more conventional MF institutions. Whether or not the FDB becomes a deposit-taking institution, this seems unlikely to have any relevance for MF, since it would be limited to term deposits. The framework within which MF will operate in Fiji will be based on NCSMED as the primary source of funding and technical assistance for MFIs. The NCSMED will have as its overriding concern the promotion of small and microenterprises, and financial service provision will be subordinate to that objective. This will not matter so long as the integrity of its financial sector activities is preserved. Some degree of oversight by the RBF would contribute to that outcome. If NCSMED takes a sufficiently broad and long-term approach to its responsibilities it will see merit in progressing the village-based savings and credit initiatives of the NMFU to assist the most micro element of its constituency. This would place NCSMED in a directly complementary role in regard to the financial service outreach initiatives of the Reserve Bank of Fiji. These initiatives are likely to become an important element in the framework for MF, using that term in the broader and less conventional sense which has been employed in this Review. An active and creative role for the central bank in incorporating MF within the broader financial sector is highly desirable. The initiatives foreshadowed by the RBF will depend on the cooperation of the commercial banks, which will in any case have to deal with GoF’s implementation of recommendations from the Committee of Inquiry. These may also have the effect of requiring banks to give more attention to social obligations, and if so there should be scope for some collaboration between the banks and MFIs and community-based organizations. New legislation for the credit union movement may presage the revival of the Credit Union League and its interlending mechanism, though here the attitude of the NCSMED to funding and otherwise supporting the credit unions (a move contemplated by the NMFU but not pursued) will be important, as will the nature of any future ADB support under its new project. The Credit Union legislation may prove hospitable to community savings and loans groups which seek a more defined legal identity and structure of support as they mature. If so, this formalization of self-help groups could prove complementary to the rural initiatives of RBF, in strengthening the rural financial infrastructure. 46 The division of labor suggested above in which, besides NCSMED and RBF, UNDP, the ADB and the World Bank play major roles, provides a firm basis for support. It would be preferable if RBF were to play the coordinating role, even though in a nominal sense that role might be thought to reside within the Treasury Division of NCSMED. RBF’s responsibility for financial sector building suggests it should remain closely involved, and training recommendations, capacity-building are designed to support the central bank’s continuing interest and expertise in the field. 47 Annex A Basic Assessment Criteria for the NMFU: Key Indicators (ADAPTED FROM THE CGAP APPRAISAL FORMAT FOR MICROFINANCE INSTITIONS)25 Indicators cover the following aspects of each MFI examined: 1. Clientele and geographic location 2. Institutional Capacity 3. Financial Performance 4. Products Offered 1. Clientele 1.1 Location of branches i) Where has the MFI located branches: what are the reasons for location? ii) Please classify your branches by location, and list them by name under the following headings :  Major urban (Suva)  Suburban Suva  Other major towns  Minor towns  Rural iii) Define the target clientele of each branch and estimate the number of eligible families that could benefit from Microfinance in the area of each branch. iv) Do your define your target group in relation to some assumed “poverty� level of family income?If so, what level of income do you use for this purpose? v) How do you characterize your target clients in relation to the broader community? Do you use indicators other than income to judge poverty? vi) What other agencies or financial institutions provide financial services to the same target group, and how do their operations affect your ability to reach that target group? 2. Financial Performance Are you able to account for income and expenditure in the following format? 2.1 Income and Expenditure Statement 26 25 Available at 48 2.1.1 Operating Income Interest and fee income from loans Sale of passbooks Service fees (savings) Income from investments Total Operating Income 2.1.2 Operating Expenses Interest and fee expense Loan loss provision expense Admin. Expense (personnel) Other admin. Expenses (up to 10 categories 27 ) Total Operating Expense 2.1.3 Net Operating profit (loss) 2.1.4 Non Operational income In-kind donations for financial services 2.1.5 Non Operational expenses 2.1.6 Total consolidated profit (loss) Are you able to prepare a balance sheet in the following form? 2.2 Balance Sheet 2.2.1 Assets Cash Loan portfolio 28 (Loan loss reserve)29 Net fixed assets 30 Total assets 26 Prepare on a cash basis, not accruals. 27 Include small items of capital equipment here 28 Include loans past due, but not written off 29 Provision for estimated future loan losses (a negative item). Method of calculating provision should be explained fully in a footnote. CGAP Suggests the following:-  1-30 days late: 10 percent of full remaining unpaid balance  31-190 days late: 25 percent of full remaining unpaid balance  91-180 days late: 50 percent of full remaining unpaid balance  More than 180 days late: 100 percent of full remaining unpaid balance. 30 Adjusted for depreciation 49 2.2.2 Liabilities Savings accounts (compulsory) Savings accounts (voluntary) Other short term liabilities 31 Long-term liabilities Total liabilities 2.2.3 Equity Equity of shareholders/members Donated equity, prior years (cumulative) Donated equity, current year Retained earnings (prior years) Retained earning (current year) Total equity Total equity and liabilities 2.3 Adjustments for Inflation and Subsidies32 First example: Year 1 of Operations (from 1 July) $F $F $F Say, TOTAL OPERATING INCOME (2.1.1) 20,000 2.3.1 Adjustment for Inflation Initial equity 33 (from 2.2.3) 100,000 less Net fixed assets (from 2.2.1) 9,000 Equity subject to erosion by inflation 91,000 less Inflation factor (say, 4%) 3,640 3,640 Inflated-adjusted value of equity 87,360 31 Accounts payable 32 Certain problems of averaging are ignored here, for simplicity. For further information consult the CGAP handbook. 33 Assumes initial grant is made at start of financial year (i.e., 1 July) 50 2.3.2 Adjustment for real cost of funds Initial grant (source of equity) 100,000 Commercial interest cost (say, 10% per annum) 10,000 Total Operating expenses (2.1.2) 22,500 2.3.3 Adjusted operating expenses 36,140 36,140 2.3.4 Adjusted Operating profit (loss) (16,140) ====== Second example34 : Year 2 of Operations (from 1 July) $F $F $F Say, Total Operating income (2.1.1) 44,000 2.3.1 Adjustment for inflation Initial equity35 (from 2.2.3) 187,360 36 less Net fixed assets (from 2.2.1) 13,000 Equity subject to erosion by inflation 174,360 less Inflation factor (say, 3.5%) 4,980 4,980 Inflation – adjusted value of equity at end Year 2 169,380 2.3.2 Adjustment for real cost of funds Cumulated grants (from 2.3.3) 34 Again, certain problems of averaging are ignored. 35 Assumes second grant of $100,000 is donated at start of financial year. 36 Cumulated purchases of fixed assets, after depreciation. 51 Commercial interest cost 200,000 (say, 9.5% p.a.) 19,000 Total Operating expenses (2.1.2) 44,000 Adjusted operating expenses 2.3.3 67,980 67,980 Adjusted profit (loss) 2.3.4 (13,980) 2.4 Measures of Profitability Are you able to calculate the following measures? 2.4.1 Return on assets Operating profit (2.1.3) divided by average total assets37 2.4.2 Adjusted return on assets Adjusted operating profit (2.3.4) divided by average total equity38 (from 2.2.3) 2.4.3 Adjusted return on equity Adjusted operating profit (2.3.4) divided by average total equity (2.2.3) 2.4.4 Operational self-sufficiency (excluding cost of funds Operating income (from 2.1.1) divided by the sum of loan loss, personnel and other Admin. (from 2.1.2) 2.4.5 Operational self-sufficiency Operating income (from 2.1.1) divided by total operating expenses (from 2.1.2) 2.4.6 Financial self-sufficiency Operating Income (from 2.1.1) divided by adjusted operating expenses (from 2.3.3) 37 Average total assets (end of Year 2) = Total Assets (end Year 1) + Total Assets (end Year 2) 2 38 Average equity (end of Year 2) = Total equity (end Year 1) + Total equity (end Year 2) 2 52 2.5 Measures of Efficiency 2.5.1 Administrative efficiency Personnel expense (from 2.1.2) plus other Admin. (from 2.1.2) plus in- kind donations (from 2.1.4) divided by average net portfolio39 (from 2.2.1) minus loan loss reserve (from 2.2.1). 2.5.2 Operational efficiency Operating expenses (from 2.1.2) plus in-kind donation (2.1.4) divided by average net portfolio 2.5.3 Administrative cost per active loan Personnel expense (from 2.1.2) plus other Admin. (from 2.1.2) plus in- kind donation (from 2.1.4) divided by average number of active loans for the year. 2.5.4 Personnel costs as percentage of total Admin, costs Personnel expense (from 2.1.2) divided by personnel expense (from 2.1.2) plus other Admin. (from 2.1.2) plus in-kind donations (from 2.1.4) 2.5.5 Number of line staff as % of total staff, at year-end 2.5.6 Number of active loans per staff member, at year-end 2.5.7 Number of active loans per loan officer, at year-end 2.5.8 Outstanding portfolio (from 2.2.1) per loan officer, at year-end 2.5.9 Number of loans per branch office, at year-end. Table A.1 Portfolio data Loan product:_________________ 2.6.1 Total principal balance outstanding, end of period 2.6.2 Number of active loans (clients), end of period 39 Average net portfolio (at end Year 2) = loan portfolio (end Year 1) + loan portfolio (end Year 2) 2 53 2.6.3 Average principal balance per client (2.6.1 divided by 2.6.2) 2.6.4 Average principal balance outstanding over the period 2.6.5 Loan losses written off over the period 2.6.6 Increase in loan loss reserve over the period 2.6.7 Loan loss rate 2.6.8 Total outstanding balance associated with loans that are : On time (and never refinanced) On time (but have been refinanced) Late (at least 1 payment)  1 – 30 days  31 – 60 days  61 – 90 days  91 – 180 days  181 - 360 days  1 year or more 2.6.9 Portfolio at-risk delinquency rate (more than 30 days late) Loan product 2:__________________   All loan products   3. Institutional Capacity 54 3.1 Legal Structure  Legal identity  Any legislation to which the MFI is subject 3.2 History  Date of foundation  Date commenced MF  Significant events influencing progress of MF activities Table A.2 DONOR SUPPORT FOR MICROFINANCE Source Of Date Amount Terms Status Support ($F) for example: “Fully Government of Fiji Grant ? disbursed and non-renewable� Or loan ? or NZODA (if loan, give “Partly terms of disbursed and UNDP repayment non-renewable� or ILO “Renewable annually� (state period if fixed Foreign Foundation, term applies). etc. 3.3 Your Board of Directors  Membership and affiliation of members  Process of selection or appointment  Frequency of meetings  Extent to which records of meetings are kept  Are records (Minutes) available for inspection? 3.4 Alliances 55  Significant partnerships supporting MF activities (please give details) 3.5 Human Resource Management i) Personnel policies  does the MFI have explicit, written, personnel policies ?  are these set our in a manual ?  is the manual available for inspection ? ii) How does the MFI recruit staff ? iii) What staff training procedures exist ?  internal  external iv) Describe a typical loan officer  gender  age  education  experience  socio-economic background  salary and other compensation v) Is compensation based on performance to any degree? If so, how ? vi) Describe the degree of dependence of your organization on foreign or external Consultants Table A.3 STAFFING SUMMARY FY2001/2002 FY-2002/2003 (to 31/1/03) Total staff (at end of period) New staff hired during period Staff who left during period Number of loan officers (end of period) Number of admin. Staff (end of period) Average total salary* (Loan Officers) Total salary* (Senior Loan Officers) 56 Average loan size Staff training costs (as % of annual budget * Salary stated should include pay, plus value of other benefits (FNPF, etc.). 3.7 Organizational Structure i) Please provide your current Organization Chart. ii) Branches  does the structure include branches ?  how is the performance of branch managers assessed ?  is each branch treated as a separate cost or profit centre ?  are head office costs allocated to the branches to calculate their profit or loss?  are management decisions, including loan approvals, delegated to branch level ?  how standardized are operational procedures across branches?  is there an operational manual to assure a standardized approach? 3.8 Management Information System (MIS) i) What hardware and software does your MIS system have? ii) What reports does the system produce, and how frequently ? iii) How soon after each period is each report available ? iv) To whom is each report made available ? v) What use is made of reports; what follow-up action is taken ? vi) Is there a manual of procedures for the MIS system ? 3.9 Internal Audit i) Describe your system of internal controls ii) How secure is your cash management? iii) Is there a manual for the internal control system ? External Audit i) Is your accounting system audited externally? ii) By whom is it audited? iii) How do auditors report, and to whom? iv) Are you subject to any scrutiny, including audit, by government? 4. Products Offered 57 4.1 Loans Describe the institution’s loan product (or products), including any requirement for compulsory savings, and indicate whether the loan is individual or delivered through a solidarity group. 4.2 Loan delivery methodology 4.2.1 Preparation phase: for each category of loan, describe very briefly the steps taken to prepare each client for borrowing. 4.2.2 Again, for each category of loan, explain how long it takes between the institution’s first contact with a potential client and the disbursement of the client’s initial loan. 4.2.3 Explain the terms on which a typical loan in each category is given. Describe the essential matters such as the amount and timing of all payments by the client, including the interest, commission or other fees, and complete details of any compulsory deposits associated with the loan. Give sufficient detail to allow the next step, below, to be completed: the calculation of the theoretical interest yield for the institution and the cost of the loan to the client. 4.2.4 Calculate the theoretical interest yield for the institution, and the cost to the client for each category of loan. (See Annex for method to be used). 4.2.5 Give details of the initial loan size which is permitted, and state the rules for how loan sizes are permitted to increase with the second and subsequent loans. 4.2.6 Give details of how your loans are distributed across the range of loan sizes from smallest to largest. For example:  For each category of loan in your program, and for each of the past three years, make up a table to show how your loans are distributed across the size range. Show only the principle sum; do not include interest even if this is your customary method of recording the amount owing in your accounts. Choose the intervals (say, $0-99, or $0-199) to suit your particular program’s circumstances. Loan ‘size’ could mean the size of each loan when it is given, or if you have the data you could make up the table on the basis of the outstanding balance of each account. Whichever of these two definitions of ‘size’ you choose, you should make clear in the heading of the table which one you are using.  In addition, if you have the information, please give details of the median loan size (initial loan or balance outstanding) for each of the last three years. 58 4.2.7 Calculate the Dropout rate among the clients of your program, using the following formula: DR = 1- ( FL divided by LPO).  DR is the dropout rate for a period (say, one year, or even the whole period since you commenced lending).  FL is the number of ‘ followup’ loans. These are all new loans issued during the period, other than loans given to new first-time clients.  LPO is the number of loans paid off during the period. Table A.4 - Size distribution of loans, financial year 2001-02 $0-200 $201-400 $401-600 $600+ Number % Number % Number % Number % 4.2.8 Voluntary savings (Note that we are not here talking about compulsory (or ‘contractual’ ) savings, which have already been covered above.) Provide the following instrument for each voluntary deposit product:  The number of current depositors  The distribution of accounts by size (preparing a table similar to table A.4, above)  The average passbook account balance of each client who saves voluntarily  Costs paid by each client for this service, and  Your estimate of the administrative cost of providing this service: a) as a proportion of fees collected for providing the service, and b) as a proportion of your total administration costs. 59 Annex B Microfinance Provider Mechanisms: Aglow - a Grameen Bank Replication Aglow is a faith-based NGO affiliated with Aglow International, an organization headquartered in the US. Among its initial evangelizing activities were visits to women in urban and peri-urban settlements and to market vendors visiting towns, including Nadi, Lautoka and Ba in Western Viti Levu, the main island of Fiji. This early experience made Aglow members aware of the economic circumstances of the people in its target groups. When the GoF called for expressions of interest from community organizations to conduct MF in 1999, Aglow applied and was accepted for accreditation and funding under the NMFU pilot program. Aglow was given a package of operational procedures and reporting requirements by NMFU. This was based on fairly standard Grameen Bank methods of group organization, which Aglow has now modified somewhat in the light of experience. Essentially, however, the Aglow MF operation remains a Grameen replication exercise, operating from three sub-units located in Lautoka (where it is headquartered), Nadi and Ba. The client base is predominantly, but by no means exclusively, female and indigenous Fijian. No breakdown by ethnicity or gender was available at the writing of this report. Aglow set up its MF unit as a separate entity, Aglow MF Unit (AMFU), and received its initial grant early in 2000. After group selection and training AMFU commenced financial operations, by ill-chance, less than a month before the May 2000 coup. The staff of Aglow MF had limited experience at that time. They commenced operations with lending, rather than (as has now become standard procedure) an introductory period of saving by all clients. Amid the unstable political environment, the MF Unit experienced substantial collection problems, ending the year with arrears of $F13,000. The borrowers concerned are pioneer delinquents in the accounts (and institutional memory) of AMFU. Their arrears have not been written off and are shown separately in the periodic management reports of NMFU, while continuing efforts are made to recover the amounts owing. In February 2003, the pioneer arrears had been reduced to $F9,000. Total arrears of the program by then totaled around $F18,000, after rather less than three years of operation and a cumulative loan disbursement of some $F360,000 (table B.1). Performance Summary Data collected are from discussion with AMFU headquarters staff, and review of NMFU’s periodic operational and financial reports. In January 2003 AMFU had slightly more than 1000 clients. It had opened more than 1650 loan accounts since its inauguration, more than 1200 of which had been paid off in full. First loans totaled 1,012, with 427 second cycle loans and 169 third cycle loans over the period of less than three years of AMFU’s operations. A single client had entered her seventh loan cycle in that period. Among contracts current in January, some 270 were either up-to-date or prepaid, while 111 were in arrears. The contract periods of a further 57 borrowers had expired, with balances remaining unpaid. In addition, 90 pioneer 60 delinquents remained, hangovers from the early and unhappy experience of AMFU. As mentioned above, their debts still formed half of NMFU’s arrears in January 2003. In February 2003, savings balances held in the names of Aglow clients totaled $104,000 (table B.3). Gross savings (i.e. before withdrawals) totaled $142,000 to the end of 2002. AMFU holds member savings in a single master account with a commercial bank, keeping individual records for each saver and charging a one-off fee of three cents in the dollar on deposits to cover its transaction costs. Around $F75,000 was held as the voluntary savings of borrowers, with contractual or group savings totaling some $F18,600. Another 320 prospective clients were placing savings with Aglow at that time, in the amount of more than $F10,000. Aglow’s savings mobilization performance is the more striking in that Aglow pays no interest on savings balances. It succeeds in retaining former borrowers as savers and has a significant proportion of members who regard the savings facility as the primary advantage of membership. It is quite opportunistic in regard to savings mobilization, visiting market-places and collecting directly from vendor-members in situ. Market activity reaches a peak after paydays, usually on Thursday, with high turnover in the period to the end of the weekend. At this time, of course, the banks are closed. Market-vendors appreciate being able to put aside money quickly after its receipt. Management Information System Aglow has a computerized management information system (MIS) which permits tracking savings, lending and repayment performance of each of its three sub-units, as well as aggregate performance. Reports are prepared weekly, in line with the repayment schedule for the single loan product, a 24-week loan with weekly repayments. In February 2003, at the time of Mission’s visit, the weekly reports for two of the three sub-units were available for the week of February 14. Data entry for the third was in process. This suggests Aglow has the capacity for timely reporting. In the aggregate, AMFU reports the following loan portfolio measures (table B.1). Table B.1: Aglow Loan Portfolio Measures Measure Value ($F) as at 1/24/2003 Total amount of loans released as of 358,776 period Total amount of loans outstanding as 59,021 of period Total amount of loans received as of 299,755 period Total amount of loans due as of period 311,084 Total amount of loans paid in advance 5,018 61 Measure Value ($F) as at 1/24/2003 Total amount of loans past due 3,658 Total amount of loans overdue 3,622 Total amount of pioneer delinquents 9,077 Total portfolio at risk as of period 18,087 Total amount past due for 11,339 repayment40. From these aggregates it calculates the following indicators of performance. Numbers shown are from the weekly report as at January 24, 2003. Table B.2: Aglow Loan Portfolio Performance Indicators Indicator Level at 1/24/2003 % of repayment overall (c/d) 96% % of advance payment (e/c) 2% % of past due (f/a) 1% % of overdue (g/a) 1% % of pioneer delinquency rate (h/a) 3% % of portfolio at risk (i/b) 31% % of past due for repayment (j/a) 4% The system has records for each client, which include the cycle number of the current loan in the borrower’s loan history (1st, 2nd, 3rd, etc.). It can run out reports classifying all loans by cycle number, thus providing useful information on repeat borrowing. For each client in arrears, it records the number of the payment (e.g. No.18 of 24), at which arrears commenced. It records data for the repayment performance of the personal client groups of each of the Field Credit Officers (FCOs), permitting comparisons of individual staff performance. 40 This appears to correspond to the sum of the three categories of arrears (past due, overdue, delinquent) minus the advance payments, although why prepayments should be deducted is not clear. 62 Savings data are similarly recorded at the subunit level and in the aggregate, and for the clients of each responsible FCO. Savings data are available for prospective borrowers (clients who have not yet advanced to the point of borrowing, or are not currently doing so) and for the personal and group (that is, contractual) savings of borrowers. The MIS operative was able to produce printouts for combinations of the above variables rapidly upon request. The data for savings are presented in table B.3. Table B.3: Savings Mobilized by AMFU Actual and Prospective Clients, February 2003 Personal savings $F75,495 Group savings $F18,598 Candidates’ savings $F10,336 Total savings $F104,38241 For individual borrowers, it is possible to track borrowing and repayment in the weekly statement. The initial amount of each loan is recorded as the amount of principal and interest due under the contract (125% of the principal sum, repayable in equal installments over 24 weeks and amounting to an annual effective rate of interest of 71.4%). Principal and interest were not recorded separately in any of the printouts requested, but this would not be a difficult modification, given that interest and principal form fixed proportions of each payment. In its cash flow statements, Aglow presents monthly data distinguishing between interest and principal received. However, the figure for loans released appears to include the interest due on loans, which is misleading as an indication of the volume of credit disbursed. The indicators of portfolio performance also appear to include interest due in both the numerator and denominator of each calculation. This does not affect the accuracy of calculation for the indicators concerned, but does blur the analysis of the aggregate loans made. Reporting Arrears For each borrower, the weekly statement records the amount paid to date and the loan balance. The first time a full or partial payment is missed, the deficit is recorded as past due and the whole of the unpaid balance of the contract is recorded as portfolio at risk (PAR). The calculation provides an accurate measure of the proportion of loans outstanding which is due from borrowers in arrears as at the reporting date. However, this measure of PAR may be considered draconian by normal standards. Many MFIs do not recognize arrears for PAR until after some longer interval, depending on factors such as the term of loans and the required frequency of repayments. The proportion given for PAR in the report of January 24, 2003 is 31% 41 The individual totals given in the savings report for February 2003 actually sum to $F104,429. As mentioned elsewhere in the text of this review, minor arithmetical errors are relatively common in the AMFU periodic reports. 63 (some $F18,000 at risk on loans outstanding of around $F59,000). However, the indicator is unstable, since only a week earlier it had been recorded as 41% (PAR of $F22,450 on portfolio of $F54,500). This instability is likely due to the immediate recognition of risk practiced by AMFU and reflects short-term errors in collection performance and loan-giving. Using this sudden death definition of PAR may be a practical necessity for Aglow, because of difficulties it appears to experience with recording the time-profile, or ageing of arrears. The system makes provision for the aging of arrears (1-30 days, and so on, up to 180 days) for individual clients but these records appear erratic, on the basis of material reviewed. The records record each client’s payment history and status, but are not easy to interpret in regard to the aging of arrears. Moreover, Aglow does not appear to present an aggregated report showing the time-profile of its arrears. In any case, given its short loan term (24 weeks) and weekly repayments, it would be more appropriate for Aglow to record arrears in the intervals 1-7 days, 8-14 days, 15- 28 days, and so on. It might also be better for it to recognize an account as being in arrears, for the calculation of PAR, after 7 days, rather than immediately. Whether the difficulties of interpretation of the individual records are due to data entry errors or conceptual problems is unclear. Minor arithmetical errors are relatively common in the weekly statements. However, it should be possible to iron out these difficulties with some training and technical assistance (to clarify the appropriate definitions, calculate the necessary algorithms and train staff in their entry and analysis). Aglow currently possesses staff fully capable of benefiting from such technical assistance. The payment indicator, percentage of repayment overall, is shown above (table B.2) at 96%. This is a cumulative rate, relating total repayments received to total repayments having fallen due as at the reporting date, over the whole history of Aglow. On the face of it, this appears to be an accurate measure. From an analytical point of view the measure has some value as an indicator of long-term performance, although it can be misleading on its own, especially given the relatively short-term loans granted by AMFU. For analysis of current performance it would be more useful for Aglow to relate arrears in the current period to the amounts falling due in the period, using a measure of collection rate. For example, the on-time collection rate is a useful indicator of current repayment performance. It measures amounts paid on-time during a period as a proportion of all amounts falling due for the first time during the period in question.42 With regard to the analysis of arrears, Aglow reports a measure of PAR, as discussed above. This is a standard measure of portfolio health, indicating the outcome of a scenario in which all borrowers currently in arrears (in Aglow’s case, even those who have missed a single payment) fail completely to discharge their remaining loan obligations. For the analysis of arrears, Aglow also records separate categories of past due and overdue payments. The former are defined as arrears existing prior to the end of the loan contract term, while the latter are payments owing on contracts which have run past their 42 The measurement and analysis of the arrears of MFIs is dealt with in Measuring Microcredit Delinquency, CGAP Occasional paper No. 3. June 1999, which is available on the CGAP website at www.cgap.org. 64 completion date. For Aglow’s purposes, the past due measure is, as mentioned above, less useful than measures of current collection performance, such as the on-time collection rate. The overdue category is not a useful category for analytical purposes. It appears to be an expedient made necessary by Aglow’s failure to “bite the bullet� and write off its long-term arrears. As mentioned, Aglow continues to press the pioneer group of delinquents for repayment, no doubt believing that this is necessary for its credibility as a lending agency in Western Viti Levu. However, Aglow may need to accept the need to write off the pioneer arrears in order to have its balance sheet and arrears records more accurately reflect current performance. The decision to write off bad loans is quite distinct from the process of collection; the latter may continue as long as the Board of Aglow deems collection worthwhile. In the meantime, while the delinquents are quarantined from the measures of past due and overdue accounts reported by Aglow, they continue to inflate the measure of PAR, obscuring the analysis of current performance. Thus, if the delinquent accounts totaling $F9,000 had been written off in a previous accounting period, the proportion of PAR to outstanding portfolio as at January 24, 2003 would have been around 18% (approximately $F9,000 at risk on outstandings of $F50,000) instead of the 31% reported. Financial Statements Financial reports reviewed were limited to records of cash flow and expenses for calendar 2002. AMFU reports cash flow on a monthly basis. Savings collected are not passed through the operating cash flow statement, but are recorded separately. The statement of inflows distinguishes between receipts of principal and interest. The record of expenses is also produced on a monthly basis and distinguishes current and capital outlays. At year-end it notes variances against the annual expenditure budget. Neither NMFU’s balance sheet, nor its income and expenditure statement, was reviewed; however, casual inspection of the cash flow and expenditure records suggests that NMFU is far from operational sustainability at its present scale of operations. Thus, personnel outlays for 2002 amounted to some $F84,000, with total operating expenses in the order of $F112,000, as against interest income of less than $F33,000. Fees on savings may have added another F$4,000 or so to income during the period. It should be noted that in addition to relying on government grants for its initial loanable funds and continuing shortfall of operating income, AMFU also benefits to some extent from in-kind contributions, including some voluntary labor and rent-free premises in one of its sub-units. 65 Annex C Microfinance Provider Mechanisms: The Individual Lending program of the NMFU The decision by NMFU to commence a direct lending program gave it the character of a quasi-MFI. This had implications for its focus and effectiveness as an institution. The aggregate dimensions of the IDL are noted in table 3.2. NMFU’s individual lending program was financed by disbursements totaling $F0.199M from NMFU’s capital fund. Including recycled capital, the IDL program had loaned $F0.232M to more than 280 individual loans at a mean loan size of just over $F800 by end-2002. There had been 238 first-cycle loans, 45 second-cycle loans, and one client had entered a third cycle contract. NMFU claimed a collection rate of around 62% as at that time and recorded 187 accounts as delinquent. Accumulated savings of borrowers amounted to approximately $F34,000. The program was in full operation from April 2001 for only about a year, after which new lending was suspended, pending the results of an external review. Although lending has not resumed, NMFU continues to collect from a small number of clients who were granted second or third cycle loans, as well as attempting to collect arrears. The last of the current loans was discharged April 2003. According to the external review, the IDL was launched hastily and without adequate preparation. NMFU management felt pressure of slow disbursement of loan capital placed at its disposal by government. This no doubt contributed to the urgency of the decision to initiate individual loans in larger size ranges than had been sanctioned for the NGO programs, without benefit of the borrower preparation which should mark their group lending approach. While preparations for the program had been set within NMFU, in the form of a market study and the drafting of an interim manual of procedures, the external review found that lending commenced without benefit of either of these tools. The external review noted that the decision to implement the IDL directly from the NMFU office was done without assuring compliance with government accounting and money- handling requirements. Not all staff had the appropriate delegation to conduct cash operations and NMFU’s activities as a quasi-MFI were constrained in a variety of other ways by civil service procedures. These limited the mobility of the field operating staff and developed the wrong work culture. Among other criticisms was that no time was taken to develop the necessary internal control mechanisms to assure proper screening and selection of loan applicants, with the result that many borrowers were ill-prepared for the challenge of handing their loans. Further, clients were scattered in widely-spread locations, increasing costs and impeding client monitoring and the provision of assistance for those in difficulty. Writing after the suspension of lending, the external review found that, in the setting of the NMFU, the IDL had failed to show potential to evolve beyond its pilot phase. The great majority of loans were extended for periods of 24, 36 and 52 weeks, repayable at 4- week intervals. Interest on the loans was calculated on a flat rate basis of 15% for 24 weeks, 20% for 36 weeks, and 25 % for 52 weeks. These are equivalent to effective annual interest rates of 37.9%, 33.1% and 28.1%, respectively. Two longer contracts, for 78 and 108 weeks, 66 respectively, were observed in the records and each bore a 25% flat rate interest charge. These are equivalent to effective annual rates of 17.5% and 12.7%, respectively. Loans were for the most part below $F1,000, although occasional loans of $F2,000 and up to $F3,000 were observed. Larger loans were generally contracted for longer terms. In terms of portfolio quality, IDL reported a collection rate of 61.74% to end-2002 (table 3.2, above). This is the proportion of all payments of principal and interest received by end-2002 to the total of principal and interest due by end-2002. Hence the rate shown is cumulative rather than current, although given the short lifespan of the IDL program the discrepancy is not significant. The distinction between current collection rates and cumulative repayment rates is explored in some detail in Annex B, in connection with the portfolio quality of Aglow MFI. Other NMFU data permit calculation of on-time collection rates for the IDL for 2001 and 2002, at 71.4 and 57.5%, respectively. This indicates a significant deterioration in portfolio quality during the second year of the program. It will be recalled that new lending was suspended after the first quarter of 2002, which no doubt sent a negative signal to borrowers with outstanding balances at that time. In terms of its sustainability as a lending operation, it is necessary to consider IDL’s collection performance, which impacts directly on the income side of operations, and the costs incurred by it in implementing the program. As seen above, collection performance was very poor. On the cost side, IDL operated out of a Government office with high overheads by MFI standards, particularly salaries of the staff dedicated to the program. This would be particularly so if all appropriate costs, including senior management time, had been allocated rigorously to the IDL. It is not apparent the extent to which this has been done. Other costs, including transport, were also of an order not typical of MFI operations. This experience supports the view that public sector entities should not be part of the financial sector and cannot become sustainable long-term providers of financial services. In the event, NMFU calculated a lending cost of 44 cents in the dollar as at end-2002 for the IDL. Total costs amounted to more than $F102,000 in that year as against interest due (although not necessarily collected, as has been shown above) of slightly less than $F40,000. By comparison, NMFU calculated the lending costs of Aglow MFI at 39 and 28 cents in the dollar for its Lautoka and Nadi branches, respectively. The comparison is more striking when one considers that Aglow, unlike the NMFU, is exclusively concerned with direct lending so that the allocation of costs between activities is not an issue. Moreover, Aglow’s collection performance, discussed below, is much superior to that of IDL. 67 Annex D Capacity-building Options for the Microfinance Sector The institutions active in the sector at present are few, and the number of groups involved is small. While the emphasis should be on developing in-country capacity as far as possible for training, monitoring and supervision, this cannot be achieved without some investment in international training and exposure for key personnel. Fiji is comparatively remote from the CGAP-initiated training “hubs� located in Southeast and South Asia, where the training courses are typically short (3 to 5 days) so that costs are likely to be high per unit of training. However, there may be occasions when short courses are offered back-to-back and if so these opportunities should be grasped and perhaps combined with visits to relevant institutions in the country of study. The Boulder, Colorado, summer school in MF is three weeks in duration and is clearly a feasible option. It would be desirable to designate a focal point for capacity-building in the sector, and for this the NCSMED appears well-placed. At least initially, the National Centre would benefit from external resources to equip it to discharge this responsibility and the World Bank could be an appropriate external agency to support NCSMED in this function, perhaps in the context of a broader program of support for Private Sector Development. Capacity-building options for providers, regulators and auditors: RBF rather than the NCSMED, is regarded as the putative regulator for the purposes of this Review. Other key institutions are the two operating MFIs, the CUL and credit unions, the NMFU and the NCSMED itself. The NMFU is likely to be absorbed into NCSMED after the end of its pilot phase, scheduled to finish at end-2003. These suggestions for capacity-building assume that the new MF Division of the NCSMED will be composed essentially of the staff of the present NMFU. However, if resources are available before the incorporation of the NMFU, consideration should be given to extending training and capacity-building opportunities to staff of the NMFU as it is now constituted. A discussion of various training options follows. Suggestions are made for training opportunities for MF providers, but not for specialist auditors. At this stage of the industry’s development, given the small numbers involved, it seems more sensible to build capacity within the NMFU and/or NCSMED to scrutinize the accounts of MFIs, and the same applies to the Credit Union League in the case of CUs, or to the CCSLA in the case of cane farmers’ cooperative S & Ls. Reserve Bank of Fiji Specific legislation for MF would be premature in Fiji at this early stage of the sector’s development. Nonetheless, it is important that the Reserve Bank of Fiji, Financial Institutions Department, should have at least one senior officer with a sound conceptual understanding of MF. For this purpose it is recommended that a senior staff member of the staff of the Financial Institutions Department should attend the annual MF Training Program in Boulder, Colorado, where there is a stream of courses especially designed for regulators. In addition, the Chair of the NCSMED, Former RBF Governor Mr. Savenaca Siwatibau, expressed interest in the linkage 68 mechanism for commercial relationships between banks and MFIs. For this reason, it is recommended that an RBF staff member be given exposure to the operations of the National Bank for Agriculture and Rural Development and associated commercial banks in India, and should also visit the Reserve Bank of India for briefing on this approach to MF. Table D.1: Capacity-Building Options for the MF Sector AREAS OF National Village Credit Union Credit Reserve NEED/ MF Unit NCSMED MFIs Banks League Unions Bank Fiji INSTITUTIONS Policy & Snr staff of regulation Financial Institutions Dept. to Boulder T/A and Director to Director MF, to monitoring for Boulder Boulder MFIs Apex operations, Director, Treasury, donor relations to Boulder MFI Managment CEOs or senior staff issues, incl to CGAP, Manila accounting & training hub and/or financial analysis, CGAP South Asia, MIS, and/or in-country training by an delinquency appropriate management, institution from product design. Australia or N.Zealand MIS for MFIs FNTC, training for MFI staff Accntng, fin-ancial Staff of MF Field staff analysis, division, to be management of conducted by arrears, MIS, Philippine NGO or governance an Aust. or NZ institution MFI governance Boards, CEOs, conducted by Director MF Division Management, incl. FNTC training in CUFA inter-lending. management, periodic MIS. accounting & consultancy, Training of computer skills for CUL trainers, management, Governance issues. interlending, training of Lessons from MF. trainers. 69 AREAS OF National Village Credit Union Credit Reserve NEED/ MF Unit NCSMED MFIs Banks League Unions Bank Fiji INSTITUTIONS Accounting, FNTC computer CUL,after arrears mngt. skills training of MIS, governance. trainers Group methods, Officials, arrears mngt, field simple acctng. staff, conducte d by staff of MF Division Linkage banking Snr staff Financial Instns Dept. to visit Reserve Bank India and NABARD Note: all training in-country unless specifically stated. It would be beneficial for the Reserve Bank to take the opportunity to observe the operations of the ADB project MF and Employment (loan No. PNG 32472-01) for which the PNG central bank (the Bank of Papua New Guinea) is the implementing agency. Yet another resource available to the RBF is the Pacific Financial Technical Assistance Center, PFTAC, located in Suva. The International Finance Corporation and the IMF have some interest in MF and the PFTAC may be in a position to facilitate some useful contacts for the Reserve Bank. The PNG MF and Employment Project had as one of its early goals the establishment of a MF Competence Center. The problems of MF in PNG, and the cultural milieu, are closer to those of Fiji than is the case in any Asian country. It would be a useful early initiative for the RBF to explore the opportunities for Fijian staff of MFIs and other appropriate institutions to attend training at this center. This possibility is not canvassed in Table D.1, but it should be considered seriously. Exchanges between NCSMED and the MF Competence Center, as well as the wholesale MF financing facility to be established within the Bank of PNG could also prove valuable. Within NCSMED there is a Treasury Division and there is to be a MF Division (presumably, as mentioned above, drawn from the present NMFU). The former will be tasked to monitor and manage trends in SME lending activities and provide advisory services in finance to second tier organizations and SMEs. This officer will also manage relations with donors. The MF Division is to have functions including monitoring, increasing the outreach of MF, and enable MFIs to achieve financial sustainability. It seems a division of labor will be established that sees Treasury concerned with flows of funding and relationships with wholesalers, while MF Division will provide technical assistance to retailers and perhaps have responsibility for sectoral initiatives of a technical nature. On this basis it is recommended that the Directors of the Treasury and MF Divisions of the National Center (or the Director of the NMFU if its 70 annexation by NCSMED is delayed) should attend the Boulder summer school, with appropriate choice of elective courses in each case. Staff of MF Division (or the NMFU in the event of undue delay) should have the opportunity for in-country training in MFI accounting and financial analysis, the management of arrears, management information systems and governance. There are a number of options for the source of this training. Philippine NGOs have good training capacities (for example the group of NGOs associated with TSPI in Manila) and Philippine staff of UNDP have proved successful in Fiji. Australia and New Zealand have NGOs and research institutions familiar with the Pacific and with appropriate MF experience, including CARE Australia, IDSS (the consulting arm of Oxfam Community Aid Abroad) and the Foundation for Development Cooperation. The numbers of trainees could be augmented by staff from the MFIs. As a longer-term goal, the NCSMED should be looking to develop the in-country capacity to conduct progressively more of this training using domestic resources. One domestic training resource which should be tapped for the MF sector is the Fiji National Training Council. FNTC conducts computer training courses and MFIs should have the option of upgrading their computer skills with this institution. For MF providers, specifically MFIs, CEOs and perhaps some specialist staff should be sent to various CGAP regional training opportunities. Suitable courses are offered in both locations in MFI management issues, including accounting and financial analysis, information systems, delinquency management, and new product development, among others. There would also be opportunities for visits to “best practice� MFIs in either location. In-country options could also be considered, with the use of trainers from the Philippines, Australia or New Zealand, as discussed above. As mentioned in the previous paragraph, middle-level staff of the MFIs should be given opportunities to attend in-country training with staff of the MF Division. An important additional capacity-building initiative should be conducted personally by the Director of the MF Division, in order to lend the necessary prestige to the exercise. This would be training in issues of governance for MFIs, and would involve Boards and CEOs of these institutions. FNTC facilities could also be made available to staff of the credit unions as part of a drive to see all CUs move over a reasonable period to full computerization of records. Staff of the Fiji Credit Union League could also benefit from a wider range of FNTC course offerings in management, accounting and computer skills. However, the most comprehensive assistance that might be offered to the CUL would be the upgrading of its existing relationship with the Credit Union Foundation of Australia, by funding a continuing consultancy involving periodic visits and a range of activities to strengthen CUL management and governance, to conduct training of trainers and revive the interlending mechanism (for which purpose an initial capital injection would be necessary). For the embryonic village banks there would be value in training conducted by staff of the MF Division. This could deal with issues of governance, management of groups and simple accounting methods. There would be scope for cooperation with the rural financial services initiatives being considered by the RBF. The target group would be local level officials associated with the village and tikina banks and their field staff, or collectors. Finally, so far as its resources and other commitments permit, the UNDP resident staff in Suva with interest and expertise in MF should be involved in the capacity-building program. This valuable relationship should be maintained. 71