United States General Accounting Office GAO Testimony For Release FOWEIGW ASSISTANCE On Delivery 9:15 a.m. EDT September 26, AID's Microenterprise 1990 Assistance Program , Statement of Harold J. Johnson Director, Foreign Economic Assistance Issues National Security and International Affairs Division Before the Subcommittee on International Economic Policy and Trade Committee on Foreign Affairs, House of Representatives GAO/NSIAD/T-90-38 GAO Form 160 (12/87) Mr. Chairman and Members sf the Subcommittee: , We are pleased to be here today to provide a preliminary report on our on-going review of the Agency for International Development's (AID) microenterprisel program, The issues we have been asked to discuss today are (1) whether AID has adequately managed the microenterprise program-- including whether it has complied with congressional guidance on the size of loans disbursed, the economic status and gender of loan recipients, and the use of indigenous "grass roots" organizations to carry out the program; and (2) whether AID's March 1990 report to the Congress on the program is reasonably accurate and reliable. Upon completion of our work, we plan to issue a written report on these matters. BACKGROUND The Congress has had a longstanding interest in supporting microenterprise activities in developing countries. In fiscal year 1988, it earmarked $50 million to be used for credit and other mlcroenterprise assistance. This amount was raised to $75 million in fiscal years 1989 and 1990. The Conference Committee report accompanying the fiscal year 1988 earmarking urged AID to target 80 percent of the loans to people in the lowest 50 percent of the economic strata, with special emphasis on women-owned businesses, and those enterprises owned and operated by the poorest 20 percent of the population. The Committee also recommended that loan size IAID defines a microenterprise as a business having 10 or fewer employees. not exceed $300, unless there were indications that the loan size should be exceeded to accomplish the program objectives. This guidance was reemphasized by the Appropriations Committees in fiscal year 1989. While the congressional guidance is not mandatory, AID has incorporated the guidance in its policy determination. RESULTS IN BRIEF Prior to the fiscal year 1988 legislation, AID did not have an overall policy for structuring microenterprise activities and directing resources in support of them. In response to the legislation, the AID Administrator appointed an Advisory Committee to help develop policy guidance under which all microenterprise projects would be directed. The Advisory Committee incorporated the congressional guidance into an AID Policy Determination which was issued in October 1988 and transmitted to all of AID'S i missions. AID does not. have a system to track detailed information concerning its microenterprise activities. In preparing its March 1990 report, the AID missions we visited had supporting documents for amounts obligated for microenterprise projects, but most of the specific data concerning loan size and the recipient's gender and econorqic status was based on assumptions or estimates. As a result, the data AID reported to the Congress in March 1990 is of questionable reliability. We found some discrepancies in the data for eight of the 17 projects we examined. The lack of specific information should have been noted in AID's report. This was not 2 done, and consequently the impression is given that the data is more accurate or based on more factual information than is actually the case. i The fiscal year 1988 legislation clearly resulted in AID focusing greater management attention to its microenterprise assistance program. Recently AID took steps to further improve its management and oversight of these programs. In July this year, AID placed its microenterprise development projects within the Private Enterprise Bureau. According to AID officials, this was done to ensure more effective control over program implementation and resources. In an effort to increase its oversight and ability to respond to congressional interest, AID-contracted 'for a feasibility study to determine whether an information system can be developed to track AID's microenterprise programs and produce the types of information requested by the Congress. A draft of the study was recently submitted to AID. AID'S MANAGEMENTOF THE MICROENTERPRISE PROGRAMAND IMPLEMENTATION OF THE EARMARKS AID’S microenterprise program has evolved over the past 17 years, The 1973 "New Direction" legislation reoriented AID's overall mission towards addressing the needs of the "poor majority." AID began developing projects, including some microenterprise projects, in response to this new focus. It embarked on a long-term research and development program, which is the genesis of the present program, and it supported a number of centrally funded projects to obtain basic data on microenterprises and to determine 3 methods to best assist them. In 1985, AID began providing assistance to intermediary institutions to strengthen these institution's management of microenterprise programs. AID's Organizational Structure Before 1981, microenterprise development projects were not centrally coordinated. Project focus and objectives varied according to the AID office responsible for the project. For example, the Office of Science and Technology focused on (1) developing methods to improve technology and (2) applying research and evaluation methods to small enterprises. On the other hand, some projects within the Office of Agribusiness focused on agricultural credit. In 1981, AID placed responsibility for coordinating these microenterprise prolects within the Employment and Enterprise Development Division, Office of Rural and Institutional Development, Bureau of Science and Technology. This was done because of the division's capacity to conduct "...basic and applied research in employment and small enterprise issues, including policy, technology, factors of production and institutions...and to provide technical support on microenterprise activities to the regional bureaus, overseas missions and private voluntary organizations." In July 1990, AID placed its microenterprise development projects within the Private Enterprise Bureau. According to AID officials, this was done to ensure more effective control over program implementation and resources. AID's Response to the Microenterprise Legislation In February 1988, the AID Administrator appointed an Advisory Committee composed of individuals from nongovernmental organizations who had experience in implementing microenterprise credit projects in developing countries. The Advisory Committee was asked to develop guidelines for AID's microenterprise development program. This resulted in AID's first overall policy for structuring microenterprise activities and directing resources in support of them.2 In March 1988, the AID Administrator also sent a cable to each of AID's missions advising them of the legislative earmarks and the congressional guidance in implementing the microenterprise program. The missions were requested to report ongoing and planned programs in support of micro and small enterprises. The AID policy issued in 1988 stated that a microenterprise ‘I . . . should have no more than 10 employees and should have characteristics (assets, revenue, etc.) that fit well within the objectives of the program. Special emphasis should be placed on 2AID Policy Determination-17 (PD-l7), Microenterprise Development Program Guidelines, (Oct. 10, 1988). 5 small-sized and individually-owned productive activities.' The policy included guidance on (1) social and micro-economic policy reform, (2) use of institutional intermediaries in implementing programs, (3) the provision of technical assistance and training, and (4) terms to intermediate financial institutions. The Advisory Committee also incorporated into the policy the congressional guidance regarding loan size, program beneficiaries, and other matters; however, this guidance was not made mandatory. The objective of AID's microenterprise program, as stated in the policy, is "to help people with limited or no access to capital achieve a level and quality of business activity that will permit increased access to formal financing and technical services and expand productive employment and incomes." It broadly describes microenterprise activities as including "the whole spectrum of productive activities ranging from rural-based agribusinesses and handicraft production to urban-based trading, service, and manufacturing enterprises, many of which are labor intensive." Overall, AID's guidance gives mission officials the flexibility to adapt their program to specific country conditions. For example, -- Although the mission in Guatemala targets microenterprises in rural areas, the focus of its microenterprise development program is to promote non-agricultural activities. This was done in hopes of stemming the flow of migration to urban areas and reducing the population's dependency on agricultural 6 activities. The mission also targeted a large segment of the population that had been historically neglected. -- The Honduran microenterprise program emphasizes the creation of jobs in urban areas rather than subsistence self-employment initiatives. In Honduras, the majority of unemployed people are located in the urban areas, and, consequently, the vast majority of resources are directed towards businesses that have the potential to expand and generate employment In those areas. -- The mission in Senegal targets Its program resources to both agricultural and non-agricultural microenterprises in rural villages and small towns. However, the mission plans to expand its program to target microenterprise activities in Senegal's capital, Dakar. AID'S Approach to Assisting Microenterprises In adapting to specific country programs, AID's guidance provides for a variety of microenterprise activities, including training and technical assistance, institutional support, policy and regulatory reform, and credit. -- The Guatemalan government implemented a program, with AID funding, designed to (1) increase the productivity and income of microenterprises and (2) generate employment within the informal business sector. Its goal is to provide credit, technical assistance, and training to 20,000 microenterprises in 7 3 years. Three Guatemalan banks disburse funds held in trust for the private voluntary organizations (PVOs) that implement the program. The average loan size disbursed under this program is $753. -- The Honduran mission brought in a contractor to provide technical assistance and training to PVOs that implement microenterprise projects. The contractor provides assistance in the areas of marketing, credit systems, management information systems, training, overall program management, and adminrstration. The Honduran mission also supports direct training for small business owners and cooperative members through the Central American Peace Scholarship. Participants receive training in production processes, marketing, record keeping, inventory control, and new technologies. e i -- In Senegal, PVO staff were provided training in credit management, organizational development, financial accounting, and proposal writing. Literacy and numeracy training was also provided to village organizations. The missions in Guatemala, Honduras, and Senegal fund a variety of microenterprise credit programs that serve the needs of micro and small enterprises that do not have other sources of credit. These include, but are not limited to, small credit associations, such as village banks, cooperatives, credit unions, and solidarity groups. Group sizes varied according to the nature of the credit programs. 8 Cooperative membership included up to 250 members, village bank membership ranged from 10 to 45 members, and solidarity group membership ranged from.2 to 6 members. These programs typically used the "group loan methodology," which employs group pressure as an incentive to repay loans. Loans disbursed through these organizations ranged from about $45 to almost $1,200 per individual. The Role of Intermediaries ~11 three countries we visited used intermediary organizations in the design and implementation of their microenterprise projects. These groups were PVOs and nongovernmental organizations (NGOs) based in the United States or abroad. Most of these organizations implemented their microenterprise programs either through or in conjunction with locally based community organizations, often called grass roots organizations. However, the extent to which grass roots organizations were used varied among the countries, -- The majority of PVOs in Guatemala worked directly with the beneficiaries of the program. -- The mission in Honduras funded two umbrella PVOs, which supported nearly 25 community-based organizations that implemented a variety of microenterprise projects. -- In Senegal, the PVOs and NGOs worked closely with cooperatives and local community organizations in funding eight local 9 community organizations and one PVO that provide credit and literacy and numeracy training directly to 57 village organizations. Difficulty in Tarqeting the Poorest of the Poor None of the three countries we visited targeted their microenterprise projects to the poorest of the poor. However, some PVO officials with whom we talked said that these people were benefiting from their projects. Other PVO officials said that they did not have the financial and management capacity to target these groups, especially if the poor reside in remote, inaccessible areas. According to these officials, this group might be better served by social welfare programs that address problems of health care, nutrition, or literacy. A problem in targeting a particular economic strata, such as the lowest 20 percent, is identifying who they are since there is no agreed upon criteria. The general consensus among AID officials was that standard criteria to identify the poorest 20 percent of the population would be difficult and costly to develop. However, officials at the mission in Guatemala and at AID/Washington stated that they are conducting studies to develop poverty indicators that may be helpful in creating the necessary criteria for identifying the poorest of the poor. According to these officials, it would be difficult to measure relative poverty levels based on income alone because it is of questionable reliability. These officials said that other factors must be considered, such as the differences 10 between urban and rural areas. The urban poor lack access to land and other natural resources upon which the rural poor can subsist. However, the rural poor usually live in remote areas and may not have access to basic government services such as health care, education, and nutrition programs. AID'S PREPARATION OF ITS MARCH REPORT ON MICROENTERPRISE LOAN PROGRAMS In late31988 and 1989, several members of the Congress requested that AID provide specific data to document that it was complying with the legislative earmarks for the microenterprise program and the guidance on loan size and recipients' gender and economic status as specified in the accompanying Committee reports. In replying to the requests, AID stated that it had complied with the law; however, it stated that the specific data requested was not readily available and would be difficult to compile. In February 1990, shortly after we began our review, AID sent a survey questionnaire to its missions requesting detailed information-- including the types of funding, loan sizes, and beneficiary gender and economic status-- on their microenterprise projects. The questionnaire was in response to a congressional request for information concerning AID's compliance with the earmarking legislation and congressional guidance, AID issued its report in March 1990.3 3Report to the Congress, AID Microenterprise Development Program, (Mar. 30, 1990). 11 The report presents statistics from 47 missions reporting microenterprise projects for 3 fiscal years--1988, 1989, and 1990. We focused on AID's loan data for fiscal year 1989, which was the most recently completed fiscal year. We examined the methodology and supporting documentation used by three of AID's missions, Guatemala, Honduras, and Senegal. We selected these to examine in detail because, according to AID's report, they accounted for about 27 percent of the $30.3 million for microenterprise loans in fiscal year 1989 --$3.2 million in Guatemala, $1.6 million in Honduras, and $3.3 million in Senegal.4 Observations on Mission Loan Data None of the three missions we visited had a system to reliably and accurately obtain the loan data requested by AID/Washington's questionnaire. We found supporting documents at the missions for amounts obligated for microenterprise projects, but most of the other data on loan size and beneficiary characteristics had to be obtained from PVOs or NGOs implementing the projects. However, these groups also did not routinely track this data, nor were they required to do so by AID. Also, in some cases, they were not able to provide data by the U.S. government fiscal year because their accounting systems were on a different fiscal year basis. Therefore, much of the data reported by the overseas missions was 4The top eight countries reporting microenterprise loan projects were El Salvador with $5.5 million, Senegal with $3.3 million, Guatemala with $3.2 million, Indonesia with $2.4, Madagascar with $2.1 million, Jamaica with $2.0, the Philippines with $1,8, and Honduras with $1.6 million. 12 WI th a t 33 percent o f th e 3 ,9 1 5 loans m a d e u n d e r o n e o f th e a b o v e projects w e r e m a d e to w o m e n ; h o w e v e r , th e s u p p o r tin g d o c u m e n ta tio n a t th e m ission showed 25 percent o f th e loans w e r e m a d e to w o m e n . W e could n o t d e te r m i n e th e reasons for th e differences b e c a u s e th e R A ID o fficial who had compiled th e d a ta for th e q u e s tio n n a i r e was no longer with th e m ission a n d did n o t leave a n y inform a tio n on how th e d a ta h a d b e e n c o m p i l e d . t Honduras In H o n d u r a s w e fo u n d th r e e discrepancies in th e d a ta reporte d for o n e large m icroenterprise project. T h e m ission reporte d th a t -- $ 1 .2 m illion in loans h a d b e e n fDnded in local currency under th e project, b u t s u p p o r tin g records s h o w e d th a t $ 1 .6 m illion had b e e n fu n d e d ; -- $ 6 .6 m illion in m icroenterprise loans h a d b e e n m a d e , but m ission records s h o w e d th a t th e a m o u n t w a s $ 7 .2 m illion: and -- 5,891 m icroenterprise loans h a d b e e n m a d e o n th e project, w h e r e a s th e m ission records w e reviewed s h o w e d th a t 6 ,5 7 2 loans had been made. 14 We were told by mission officials that, after the reply to questionnaire was prepared, the mission hired a contractor to work with the PVOs and NGOs to improve their information systems. The documentation we reviewed was based on the improved information developed with the assistance of the mission's contractor. We were also told by mission officials that the loan data reported to AID/Washington included funding from sources other than AID. The PVOs and NGOs do not keep track of microenterprise loan information by funding source. The mission made note of this in its reply to the questionnaire, and an AID/Washington official told us that the Honduran information was adjusted to account for this problem. Without more detailed information readily available, we believe the adjustments were appropriate but may have contributed to the apparent errors in the reported data. The Honduran mission also did not report the number of loan recipients and amounts to the poorest 50 percent or 20 percent of the population. Mission officials stated that they did not have criteria to determine whether the loan recipients were in these categories. Senegal In Senegal, we found a discrepancy in the mission's calculation of the average loan size for one microenterprise project. The project also highlights how the missions had to rely on estimates and assumptions in replying to AID/Washington's questionnaire. Under 15 EFFORTS TO ESTABLISH AN INFORMATION SYSTEM TO TRACK LOAN DATA In view of congressional interest for better oversight of Its microenterprise assistance program, AID awarded a contract to the consulting firm that compiled the March 1990 report to conduct a feasibility study on whether an information system could be designed to track AID's microenterprise programs and produce the type of information requested by the Congress. A draft of the study was recently submitted to AID. We reviewed the draft study and noted that the system proposed by AID's contractor would have the ability to systematically track and report the funding sources and beneficiary characteristics of AID's microenterprise loan programs. The study proposes to use data and reporting structures that should already be in place at the project level. However, as noted in the study, a number of matters need to be resolved before the system can be implemented. They include -- How can AID-provided funds be separately identified when PVOs and NGOs have multiple funding sources and funds are commingled. -- How should fluctuating or multiple exchange rates be treated. -- What criteria or parameters should be used or developed to establish income percentiles and poverty levels. -- How shoula non-financial microenterprise activities, and their contributions, be accounted for. 17 R ): In summary, we believe AID has improved its management of the R 3 microenterprise assistance program, and is attempting to implement r the congressionally recommended program guidance. We also believe R accurate and 1 that while AID desired to provide the Congress with an I reliable report on the microenterprise program in March 1990, full disclosure of the limitations AID encountered in accumulating the data would have improved the report's credibility. This concludes my prepared remarks. We would be pleased to respond to any questions you or members of your Subcommittee may have. R
Foreign Assistance: AID's Microenterprise Assistance Program
Published by the Government Accountability Office on 1990-09-26.
Below is a raw (and likely hideous) rendition of the original report. (PDF)