What Really Sustains Microfinance
Peter R. Crabb, Northwestern Nazarene University
The microfinance industry seeks to free itself from subsidies – both private and public. That is, microfinance institutions seek sustainability. This study reviews factors impacting the sustainability of these institutions and some sources of existing success. Both public reports and self-reported data from microfinance institutions are used here to show that subsidies have a significant negative impact and that overall economic conditions are significantly positive. Sound fiscal and monetary policy by the host country government should do more to serve the poor than continued subsidies to the microfinance industry.
Much praise has been awarded the microfinance industry since Muhammad Yunus received the Nobel Peace Prize for his work at the Grameen Bank of Bangladesh. According to the Nobel Prize committee, micro credit “must play a major part” in the effort to eliminate worldwide poverty. Much criticism has also been awarded the microfinance industry for its inability to free itself from the subsidies – both private and public. The Grameen Bank received grants and other support for most of its existence while the status of the poor remains roughly the same in Bangladesh. Grameen Bank began operations in 1974, but real per capita GDP in Bangladesh has grown at an annual rate of only 1.6% over the subsequent 20 years compared to 3.2% for neighboring India.
Yunus himself understands the need for the microfinance industry to free itself from outside support: “My experience is with Grameen Bank. We stopped taking any external money in 1995. Within our country, we borrowed in 1998 because we had a terrible flood. But we didn't go back to the donors. We went to the Central Bank. We issued bonds sold within the market to the banks. And we paid it back by 2002. After that, we didn't take out any loans. We cut off all donor links. We have plenty of money. We are expanding at high speed. With that experience, we say that anybody could have done that. But earlier, when donors wanted to give us money, we were always swayed and took the money. If donors hadn't given us the money, we would have discovered earlier that we have the strength”
According to its best spokesperson, the microfinance industry must rid itself of the dependency on subsidies and other assistance.
This study reviews the source of existing success, and possible reasons for the lack of greater success, on the part of the microfinance institutions. An extensive literature on microfinance now exists with studies of both the impact of microfinance institutions and their long-term viability. Many studies have looked at the Grameen Bank itself and the reasons for their ‘success’, including the group-lending methodology which mitigates risks. This study adds to the literature with the use of a broad cross-section of microfinance institutions to study sustainability with controls for both institutional and economy-wide factors.
Mahajan (2006) lists five fatal assumptions that underlie micro lending and limit the ability of such institutions to decrease poverty: encouraging self-employment over wage earnings, encouraging debt over savings, encouraging the use of credit over entrepreneurship, encouraging debt to low income households, and encouraging dependency through grants and subsidies. The first and last of these five problems are studied here. In the first argument, increased earnings in an economy contribute to the success of microfinance institutions as borrowers are better able to meet their obligations. Thus, the success of a microfinance institution is an indicator of improvement in the poverty conditions, not a factor in its improvement. In the second argument, successful microfinance institutions operate in countries with a high level of both private and government support to the industry. In these countries, their success is due to economies of scale achieved because of a high level of contributions, leading to a large number of borrowers and lower costs of service. In this manner, microfinance institutions serve simply as a vehicle to which public and private funds are distributed to the poor, but not a necessary contributor to the alleviation of poverty.
Both public reports and self-reported data from microfinance institutions are used here to argue that microfinance does not significantly improve the well-being of the poor in developing countries and remains operational only with governmental support. Additional arguments are made to support the contention that many microfinance institutions owe their success to improvements in the overall economic conditions offered all citizens and the ability of a countries citizens to operate freely. Sound fiscal and monetary policy does more to serve the poor than subsidizing an industry that encourages debt financing for low income households and diverts resources from the capital investment needed to employ more of the population. Conditions of economic freedom make it easier for clients of microfinance institutions to meet their debt obligations and thus create successful institutions.
The next section of this paper reviews previous work on the sustainability of microfinance and the subsidies awarded this industry. Section III describes the data used to answer the question of how economic conditions affect the success of microfinance institutions and how subsidies support their sustainability. Section IV reviews the results and section V concludes with suggestions for further work.
II. Review of Previous Work Microfinance Success
As The Economist concluded in the fall of 2006, microfinance has both virtues and limitations. The greatest limitation may be that these loans do little to eradicate poverty. Heart-warming case studies abound, but rigorous analyses are rare. The few studies that have been done suggest that small loans are beneficial, but not dramatically so. A further question is whether an approach emphasizing credit really can eradicate poverty: a ridiculously ambitious goal, though one that Mr. Yunus's evangelical view of the virtues of credit has perpetuated.
Brau and Woller (2004) review over 350 articles and studies on microfinance institutions (MFIs) and their impact on both economic growth and society. The review includes a section on sustainability of MFIs where they site many studies concluding that institutional sustainability is a necessary goal as subsidized loan funds generally are more fragile and less focused. Just as Yunus knows that these subsidies distort the incentives in the microfinance institution, others have further argued that subsidies undercut the efficiency and the scale of operations. Abrams and von Stauffenberg (2007) conclude that an increase in international support of microfinance by development institutions is “crowding-out” private investment. Development agencies are supporting the largest and most successful MFIs, increasing their scale, and discouraging support of these institutions that should be the primary market for private investors. Morduch (2005) argues that subsidies should be “time-limited and rule-bound”. Otherwise, an institution could be sustainable using standard measures but vulnerable to competition from new sources of credit. As the economic area in which the clients of the institution operate grows new lenders are likely enter. The data reported here shows that greater subsidies do lead to greater scale. However, we must also control for the benefits a microfinance institution may receive from changes in the overall economic conditions of the country in which they operate.
Khandker (1996) has a review of the many Grameen studies and further looks at its impact and sustainability. One important conclusion in that study was that higher economic growth is needed to support Grameen Bank, but more importantly, achieve the goal of moving the bank’s borrowers out of poverty. The author states that “the government thus has an important role to play in promoting and sustaining economic growth to reduce poverty on a sustainable basis”. Microfinance institutions alone are not what will sustain the microentrepreneur. Hudon and Traca (2006) find that a higher subsidy level in microfinance institutions is associated with lower sustainability. The authors use detailed data from a microfinance institution rating agency. The authors note that the sample is not representative in that these rated institutions are more likely to have achieved sustainability, having already achieved operational scale. Their study also did not control for overall economic activity. The study conducted here addresses these two issues by using both new and established institutions across many countries and controlling for economic activity in the country.
III. Available Data on Microfinance
Institutional level data for this study was provided by Opportunity International (OI). Each quarter OI collects the year-to-date income statement, balance sheet, and loan portfolio information for each of its member institutions. The OI institutions are located in developing countries throughout world – Eastern Europe, Africa, Asia, and Latin America. In total there are 36 different microfinance institutions studied over 16 quarters for both individual and group lending, providing more than 500 observations. These 36 institutions represent 32 different developing countries. (For more information on OI visit their website at www.opportunity.org.). The organization no longer provides to the public full financial statement reporting and therefore the sample period for this study ends in 2004.
This data set is better suited for the type of study presented here in that both well-off and struggling institutions are included. Previous empirical research described above is limited to those institutions that choose to report to the rating agency or other public sources of financial data on microfinance. Thus, the data used for this study should overcome any sample selection bias.
Country level economic data for this study was retrieved from Penn World Table Version 6.2, Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania, September 2006 (http://www.pwt.econ.upenn.edu). Data for countries was also taken from the Heritage Foundation’s Index of Economic Freedom (EF). The EF is a set of objective economic data in areas such as trade policy, fiscal burden of government, government intervention in the economy, monetary policy, capital flows and foreign investment, banking and finance, wages and prices, property rights, regulation, and informal market activity. Countries are measured in each area, and the ratings are averaged to create an overall level of “economic freedom”; low scores represent a low level of government involvement in the economy and high scores correspond to a high level of government involvement (For more information see http://www.heritage.org/research/features/index). A potential problem in all country level data used here is that it is reported only on an annual basis, whereas the institutional data is quarterly.
IV. The Impact of Subsidies and Wages on Microfinance Institutions
Table 1 reports descriptive statistics for all variables used in the study. There exists substantial variation across these 36 institutions for both the size of the operations (Value of Outstanding Portfolio) and the extent to which they rely on subsidies (Subsidy Income). The economic growth rates and level of consumption activity also vary substantial. The average score of 3.3 for Economic Freedom suggest that these institutions operate in countries considered “Mostly Unfree” by the index methodology. Table 2 shows the Pearson correlation coefficients for each variable used in the study. All institutional variables and the Score variable show the substantial statistical relationship with the dependent variable of interest, Financial Sustainability. The Cost per Unit Lent and the Return on Portfolio show, not surprisingly, a strong negative correlation.
A. Consumption and Microfinance Success
As indicated in Table 2, both measures of economic activity in a country have a positive impact, albeit small, on the financial sustainability of the microfinance institution. Figure 1 depicts the relationship in this sample between the financial sustainability of the microfinance institution and the extent to which consumption makes up real gross domestic product in the host country. This positive correlation is not statistically significant which is likely due to the limited annual data. Quarterly variation in economic activity may improve the relationship, but this result suggests, nonetheless, that economic activity should be controlled for when measuring the impact of any factor on the ability of the institution to sustain its operations.
B. Subsidy and Sustainability
The correlations in Table 2 also show a negative relationship between the financial sustainability of an institution and the level of subsidies received each quarter. This relationship is depicted in Figure 2. As the level of subsidy income rises, the respective institution’s financial sustainability falls. Many have argued that subsidies help microfinance institutions reach the needed operational size. This is true for the sample presented here. However, as discussed in the previous literature, these institutions may actually be doing less good as they receive more assistance. Figure 3 shows the relationship between size and subsidy income. Institutions with more subsidy income, have a higher level loans outstanding, that is greater scale. This result may reflect the crowding out effect described above - an increasing amount of grants and other subsidies are being directed to microfinance institutions that have already achieved a level of operations needed for their own continued success.
C. Multiple Regression Analysis of Both Questions.
Tables 3, 4, and 5 report multiple regression analysis designed to address both questions above. In Table 3, the dependent variable, Financial Sustainability, is impacted by each of the institutional control variables and the model is overall significant. The level of subsidies in the institution decreases the sustainability, although the marginal effect is small as indicated by the coefficient. Overall GDP growth is not a statistically significant factor. This may be due to the limited annual data described above or could be explained but the likelihood that a microfinance institution’s poor clients are not impacted by the many factors affecting an economy’s total output. With the assumption that these poor clients service local consumption needs, rather than producing for capital investment or export, the same model is estimated using consumption share of real GDP as a proxy for the ability of microfinance clients to sell their products. Table 4 reports this result and again the model is overall significant. Financial Sustainability is positively and significantly impacted by the variation across this sample in the consumption share of GDP.
Table 5 shows the results of the model using the index score of economic freedom for the host country. Previous results hold and the model remains significant. The economic freedom of a country plays a significant role in the financial sustainability of the institution. The negative sign on the coefficient means that as economic freedom is lost in a country, the sustainability of a microfinance institution declines. Thus, if a government wants to contribute to microlending a long-term and effective poverty alleviation tool they should simply increase the amount of economic freedom in their country.
V. Conclusions and Suggestions for Further Work
The results of this sample - a broad cross-section of institutions with controls for both institutional and economy-wide factors affecting sustainability- show that microfinance institutions are not directly benefiting from continued subsidies and that overall economic freedom is possibly more important. It is also likely that increased earnings in an economy contribute more to the success of microfinance institutions than additional grant assistance. The successful microfinance institutions in this sample are operating in countries with a high level of both private and government support to the industry. However, their success is likely due to economies of scale achieved because of a high level of contributions, leading to a large number of borrowers and lower costs of service. This does not show that those served by the institutions are actually breaking the cycle of poverty. They may be just as well or better served by improvements in overall market conditions.
In summary, sound fiscal and monetary policy should do more to serve the poor than subsidizing an industry that encourages debt financing for low income households and diverts resources from the capital investment needed to employ more of the population. The work shown here and throughout this literature would benefit from further econometric tests. The wide variation in institutional size suggests possible heterogeneity in the data. There also remains a question of the causal relationship. Are microfinance institutions contributing to overall economic growth, or does economic growth contribute to the success of microfinance institutions as was found here. More theoretical work in this area should help define the model more specifically. The popularity of microfinance as embodied in the Nobel Prize to Mr. Yunus will continue to encourage research in this area.