Micro-lending in the Philippines : the case for a pilot project

NUMBER 135 r~~~~~~~~~M X * ED Precis X Operations Evaluation Department January 1997 Micro lending in the Philippines: The Case for a Pilot Project New approaches to microlending Background been tried only among high-income should be tested through pilot projects professions in western Europe, and before being introduced on a large Cottage firms in the Philippines not in a developing country. The scale, according to an audit by OED.* have almost no access to bank credit Philippines project also differed Between 1990 and 1993, the World because most of their owners do not from others in that it did not plan Bank financed a project aimed at mak- have real estate titles they can use as to use an existing network of rural ing fuinds widely available to cottage collateral besides their homes, which commercial banks, relying instead enterprises and developing a viable are already mortgaged. What collat- on individual banks and lending guarantee system for lendinig to eral these family-run businesses do institutions. inicroenterprises in the Philippines. have usually consists of small ma- Designed at a time when very fewv chines that can be easily moved and The $15 million project was ap- Bank-supported microlending projects are difficult to value or repossess. As proved in May 1989. It was de- existed anywhere else, the operation a result, banks are generally unwilling signed specifically for the informal did not include a pilot project and to provide credit to microenterprises. sector, which in the Philippines is tried to break new ground by findinlg very important to the economy. The substitutes for collateral througlh mu- The World Bank in 1991 sup- MGA concept appeared to fit well tual guarantee associations, which ported a project aimed at boosting with the country's social environ- grouped potential borrowers and pro- the microenterprise sector in the ment, which values community ac- vided guarantees for loans made to Philippines by developing a system tivities very highly. The project, their members. of loan guarantees for retail lending however, ran into a series of ob- to cottage industries, and by provid- stacles, the most serious of which Although microlendinig is much ing funds countrywide to a signifi- turned out to be the lack of previous needed in the Plhilippines, where the cant number of women entrepreneurs, experience with MGAs. The MGA informal sector is very important to who make up the majority of micro- concept for lending to cottage in- the economy, the project met a series enterprise owners. It also aimed at dustries in the Philippines was en- of difficulties. Commnercial banks were increasing both capacity and interest tirely the Bank's idea, and ran unlinterested in lendinig to the guaran- of the financial system in providing counter to an analysis by the tee associations, and the procedures for loans to cottage enterprises, espe- country's National Economic establishinig the associations were cially in the countryside. Development Agency. complicated and expensive. The project was complex and could not The microlending operation in 'Perfornt aldLit report. rely on prior experience in guarantee the Philippines attempted to create Pili11ipines. Cottage Enterprise association schemes. It also did not a new substitute for collateral by de- Finance Project, by Nicolas mnke extensive use of an existing net- veloping mutual guarantee associa- Mathieu Report No. 15834, July zvork of rural banks, and intenitionially tions (MGAs). The associations 24, 1996. Available to Bank ex- opted to work through accredited fi- would group potential borrowers ecutive directors and staff from nanicial institutions. The project zvas with something in common, such as the Internal Documents Unit and unsuccessful, but it did provide vali- professional or family relationships, from regional information service able lessons for the rapidly growing or geographical proximity, and issue .c 'nhrs. Pr&cis written bv. Stefiavio practice of microlendinlg in developing guarantees to banks on their mem- counltries. bers' behalf. The MGA concept had Petrucci.  Outcome MGA allowed members to become should be taken carefully into con- eligible for a credit of up to 90,000 sideration in the planning stage. The project had a poor outcome. pesos ($3,600). But commercial Few financial institutions were will- banks considered this figure too If MGAs are to be adopted suc- ing to lend to MGAs and few MGAs high for new and unfamiliar bor- cessfully in developing countries, were successfully established. rowers. (Even with a maximum they need to be carefully tailored to contribution of 30,000 pesos, only specific conditions during project From 1990 to October 1993, 47 percent of an individual loan preparation and testing. The lessons when the project was closed one was covered by MGA guarantees.) learned from the experience in the year ahead of schedule, only 39 Furthermore, many MGAs were in Philippines highlight some of these MGAs had been legally established, remote areas and relatively far from issues: compared to an initial target of 60. the banks they were trying to bor- Today, 21 MGAs still legally exist, row from. * Project design should remain but only 12 are operating. Ten of simple and allow easy access to these are considered institutionally The MGAs themselves had their credit, particularly where there strong and likely to remain active in own problems. The administrative is poor information and limited the long term, but many have suf- process of creating MGAs was cum- banking technology. Regulations fered since their very start from a bersome and more expensive than governing MGAs should not be lack of trained administrators. Dur- had been anticipated. For example, too expensive or burdensome. ing the project, the targeted size of MGAs had to register with the secu- each MGA was reduced from 60 to rities and exchange commission * To make their clients more 40 members. Today, each MGA has (SEC); separate agreements were attractive to commercial banks, an average of 35 members. needed for matching loan funds, microfinancing projects based on guarantee agreements, registration MGAs may need to set up a small Commercial banks did not with SEC, memorandums of under- matching fund facility so they can consider MGAs attractive clients: standing, registration of individuals, guarantee a higher percentage of difficulties in obtaining adequate in- preparation of financial statements, each loan. formation from borrowers resulted and notarization of documents. As in high processing costs, while a result, some MGAs were set up * Lenders and borrowers should onlending arrangements made the artificially just to meet deadlines, be located in the same area, to re- interest margins too low. Further- with reduced sizes and members duce travel times and allow long- more, commercial banks in general often having little in common with term personal contacts to develop. seem uninterested in lending to one another. small-scale enterprises, which are * Promotional efforts should able to borrow from a number of The project also had problems clearly explain the commercial as- subsidized schemes. with information dissemination. pects of the plan to borrowers, and In spite of the staff's promotional more intensive dialogue with the In the end, most banks agreed to efforts, many potential borrowers government and implementing lend only to those enterprises they believed the loans included some agencies is needed during project felt had growth potential, and on form of subsidy and were under preparation. condition that the proposed interme- the impression that repayment con- diation margin be attractive and that ditions would be flexible. * MGAs may require temporary borrowers make deposits. In 1993, subsidies for seed capital and to loans approved to MGA members Lessons help cover initial expenses in the totaled only 72.8 million pesos first years of operation. (about $2.5 million) with individual The main lesson of the project is loans averaging between $2,700 that an untested approach for lend- * MGAs need to select participants and $2,900. ing to microenterprises should not carefully: borrowers with good busi- be adopted on a national scale with- ness development potential; rural Another problem was that bank- out first trying a pilot operation, al- banks and credit unions who know ers did not know their prospective lowing the borrower and the Bank their customers; and well-trained MGA clients, who in many cases to test the feasibility of the scheme. MGA leaders, preferably with a lived far away. An individual contri- Moreover, the views of the borrower higher education and a successful bution of 10,000 pesos ($400) to an on what can and cannot work business. OED Precis is produced by the Operations Evaluation Department of the World Bank to help disseminate recent evaluation findings to development professionals within and outside the World Bank. The views here are those of the Operations Evaluation staff and should not be attributed to the World Bank or its affiliated organizations. This and other OED publications can be found on the Internet, at http: / / www.worldbank.org/html/oed. Please address comments and inquiries to the managing editor, Rachel Weaving, tel: 1-202/473-1719, fax: 1-202/522-3200, e-mail: rweaving@worldbank.org Januiiary 1997