The Impact of Institutional Variety in Latin American Microfinance on Its Borrowers


Commercialization has been a growing trend for microfinance institutions (MFIs) hoping to improve their sustainability (Cull et al. 2009; Hoque et al. 2011; Armendáriz and Morduch 2010). The mission of microfinance is to provide poor households with access to financial services so they can eventually improve their overall standards of living. Academic literature suggests that it is challenging for commercialized MFIs to find a balance between achieving this mission and maximizing their profits (Armendáriz and Morduch 2010; Hoque et al. 2011; Cull et al. 2009; Hermes et al. 2011). In order to understand the potential benefits of microfinance on impoverished communities the entire microfinance system should be taken into consideration. Do borrowers benefit more from variety or homogeneity in microfinance systems? My hypothesis is that heterogeneous Latin American microfinance systems will have more services available to borrowers and higher repayment rates than homogenous systems. However, I was unable to complete my analysis due to challenges I experienced while collecting data. I will discuss the insufficiency of microfinance social data and suggest solutions to these limitations.

Table of Contents: 


    Microfinance provides poor entrepreneurs with access to credit. The borrowers use these microloans to create or expand their businesses, with the goal of achieving financial self-sufficiency at a reasonable income level. In many developing countries their formal, traditional, financial sectors do not give loans to low-income families because impoverished households lack the collateral necessary to ensure some form of repayment if they default on their loans (Armendáriz and Morduch 2010). This is why microfinance has flourished in developing nations where households with collateral are rare. The purpose of microfinance is to create stable microenterprises that economically and socially benefit the borrowers, their families, and their communities. 

    A traditional microfinance organization is a not-for-profit organization. Recently, microfinance institutions (MFIs) have adopted for-profit banking models to increase their sustainability and outreach, referred to as “commercialization” (Armendáriz and Morduch 2010).  While commercialized microfinance institutions have been studied to understand their profitability, sustainability, and impact, I wish to look at the functioning microfinance system and consider the overall benefit to all the borrowers in a microfinance system when they have a wider variety of choices from commercialized to nonprofit MFIs. I will ask the following questions: with more options and services available to borrowers in microfinance systems, comprised of both commercialized and non-profit MFIs, will microloans have a greater impact in Latin American countries? Also, is institutional variety beneficial to state-level communities? I hypothesize that Latin American nations with institutional variety in their microfinance systems will have more services available to borrowers and higher repayment rates than in nations where MFIs are homogenous and predominantly nonprofits.

    In this paper I will review the relevant literature regarding commercialization, impact and outreach, data collection, and the future of the industry. Then I will introduce my methodology—including the indicators chosen to measure benefit and the proposed method of regression analysis. Due to limitations during the data collection process I was unable to complete the analysis; therefore, I will examine these limitations in microfinance data. Finally, there will be a discussion about the importance of improved data and the potential policy implications of greater monitoring and evaluation.

    Literature Review

    Microfinance, being heavily studied by academics, is a much debated industry. There has been substantial debate over its effectiveness to empower those excluded from the formal financial market. In 1976, microfinance first emerged from Muhammad Yunus’s creation of Grameen Bank in Bangladesh (Campbell 2010). Since that time, it has been a rapidly expanding industry now spanning every region of the world. In this paper, I hope to explore how the latest microfinance trend, commercialization, impacts the system. I hypothesize that it creates institutional variety, potentially giving more options to its borrowers. To understand if there is a relationship between variety, outreach, and the effectiveness of microfinance systems, I reviewed literature regarding four subject areas: commercialization, outreach and impact, data collection, and the future of microfinance. It is important to consider the potential solutions to the problems raised by researchers in order to understand varying perspectives and visions for microfinance.


    In the literature there is general consensus about the positive characteristics of commercialized MFIs. Commercial banks have the ability to reach more borrowers, because they have more funding and can reach a higher level of sustainability through profit creation (Armendáriz and Morduch 2010; Campbell 2010; Cull, et al. 2009). However, commercialized MFIs are more likely to experience “mission drift,” (the scaling-up of MFIs increasingly providing loans to wealthier borrowers) because they focus on profit maximization, leverage and sustainability (Armendáriz and Morduch 2010; Cull et al. 2009; Hoque et al. 2011; Hermes et al. 2011).

    When MFIs change their priorities towards commercialization they change their borrower profile as well, and begin catering to a wealthier clientele, rather than loaning to poor clients and women—borrowers for whom microfinance was intended. However, there are contrasting opinions on how commercialization is affecting microfinance. Some of the suggested impacts of commercialization include: implementing regulations that improve savings, internal controls, and liabilities, serving different markets than NGOs (Armendáriz and Morduch 2010), lowering default rates (though this is disputed) (Campbell 2010), reducing dependency on subsidies, attracting more profit-maximizing investors (Cull et al. 2009; Assefa et al. 2013) and creating a stronger emphasis on efficiency (Hermes et al. 2011).

    There is disagreement on the effects of commercialization in microfinance. Two of these studies compare commercial banks to NGOs (Cull et al. 2009; Hoque et al. 2011). Others look at commercial banks independently from other MFIs (Armendáriz and Morduch 2010; Hermes et al. 2011; Montgomery and Weiss 2011). However, none of these studies look at commercial MFIs working in conjunction with non-profit MFIs. A study of the overall system can give us a realistic perspective on how commercial banks function within their fields.

    Measuring Impact and Outreach

    The impact of microloans on borrowers and society can be difficult to measure because of the ambiguity involved in determining their direct influence. Outreach is a widely used indicator of impact because it defines who is receiving the loans by measuring the average loan balance, number of active borrowers, or number of women borrowers. If one assumes that an organization has a lower average loan balance, it can be inferred that poorer populations are being served (Hermes et al. 2011; Hoque et al. 2011; Assefa et al. 2013). However, not all researchers believe that loan balances are a proper measurement of outreach (Cull et al. 2009).

    There are many external variables that influence changes in people’s welfare, and this causes an inaccurate measurement of the impact of microloans (Meyer 2007; Montgomery and Weiss 2011). The design of a study, data collection process, and the analysis all greatly affect the outcome of research. This has been a main reason few conclusions have been accepted by academics (Meyers 2007). An important way to measure impact is by statistics, such as default, repayment, and dropout rates, which imply that borrowers are struggling or becoming indebted by their access to credit (Hoque et al. 2011; Armendáriz and Morduch 2010; Meyer 2007; Campbell 2010). The main conclusion that can be drawn from the literature is that it is important to choose clear measurements when determining the full effect of microloans on its borrowers. Because impact and outreach are vague terms that can be defined in a variety of ways, academics must be specific and justify their choice of indicators.

    Data Collection

    In Jonathan Buachet and Jonathan Morduch’s paper “Selective Knowledge: Reporting Biases in Microfinance Data,” they examine how voluntary reporting to the MixMarket and Microcredit Summit Campaign has biases and inconsistent data. The Microfinance Information Exchange, Inc. (MIX) has an Internet platform, MixMarket, which contains a collection of financial and social data from microfinance institutions (Bauchet and Morduch 2010). The Microcredit Summit Campaign (MSC) collects socially driven data from microfinance institutions (Bauchet and Morduch 2010). When comparing the two sets of data, the article states “the MixMarket dataset provides richer detail” than the Microcredit Summit Campaign (Bauchet and Morduch 2010). The data are inconsistent, because the MFIs are voluntarily reporting information to the MixMarket and/or the MSC, and this causes self-selection bias (Bauchet and Morduch 2010). There is a difference between the MFIs reporting any data and those not reporting at all. Some MFIs only report to one of the outlets, and even if a MFI chooses to report information they may report only some indicators or only during certain years (Bauchet and Morduch 2010). This makes microfinance data unreliable and incomplete.

    To further examine the self-selection bias Bauchet and Morduch, using data from both outlets, studied the relationship between operational self-sufficiency and the percentage of women borrowers (Bauchet and Morduch 2010). While the MixMarket data indicated that a MFI could not be both sustainable and serve women borrowers, the MSC data suggested the contrary (2010). This finding shows just how strong the self-selection bias is in microfinance data and how it has a significant impact on conclusions drawn about the industry.

    Data reporting even varies greatly by region; which could potentially complicate any attempts at comparative analyses using regional data (Bauchet and Morduch 2010). Some of the differences are: South Asian microfinance institutions are more likely to report poverty indicators, MFIs that are more mission-focused are less likely to report to the MixMarket, Latin American MFIs are more “professional” when they report their data, and some MFIs are selective of what data they report (Bauchet and Morduch 2010).  Microfinance institutions that voluntarily report to outlets cause irregularities between the MixMarket and MSC data, but also create inconsistent data annually, regionally, and even internally, since some MFIs do not report every indicator.

    The Future of Microfinance

    Many have suggested ways for microfinance to improve. Some suggest that following a specific model, such as the Grameen Bank model, will be the best way for the field to proceed (Hoque et al. 2011). Most of the authors agree that technology will facilitate improvements in providing loans to their borrowers (Armendáriz and Morduch 2010; Cull et al. 2009; Campbell 2010; Hermes et al. 2011). However, there have been conflicting opinions about the role of commercial banks in microfinance. There are strong opinions that commercial banks do not belong in microfinance, and the focus should instead be on savings to increase investments (Hoque et al. 2011). However, most people see commercial banks improving microfinance regulation (Assefa et al. 2013), moving toward “cross-subsidization” between wealthy and poor borrowers (Armendáriz and Morduch 2010), incentivizing the innovation of new products (Cull et al. 2009), downscaling the large banks working in commercial financial markets (Armendáriz and Morduch 2010), or working with non-profit MFIs (Campbell 2010).

    All the studies agree that microfinance must embrace innovation and new technologies to improve both outreach and the welfare of borrowers and communities. The literature shows that microfinance is in a period of transition and that change is necessary for the improvement of the field. The most recent transformation within microfinance has been the movement toward commercialization. It has been researched through case studies, in comparison with non-profit microfinance organizations, and as an institution. However, there has been a lack of research that investigates commercialized MFIs working in cooperation with other types of MFIs, such as non-profit institutions. To understand the impact of commercial banks in microfinance they should be studied as a part of a whole system; rather than as an isolated entity operating separately from all other institutions. I would like to understand their relationship with other institutions within Latin America in order to understand how they contribute to the entire microfinance system.


    To examine the impact of institutional variety within microfinance systems I will divide twenty-one Latin American and Caribbean countries into two nearly equal categories based on the diversity of the firms working in those nations and the size of the microfinance system. The two categories will be homogenous microfinance industries and heterogeneous microfinance industries. I chose to only study a single region, Latin America and the Caribbean, rather than microfinance on a global scale in order to minimize other potential factors contributing to variations in my results. I specifically chose Latin America and the Caribbean, because microfinance is well established in this region. Also, it has many MFIs that operate at a range of levels from local to international, but there are not any individual organizations that dominate the market, as seen with the Grameen Bank and BRAC in Bangladesh and Southern Asia. I would collect data at the country level because there is a lack of studying commercialization at this level and it allows me to see how commercial MFIs interact with their system.

    The indicators I will use to measure impact will be default rates, number of borrowers, variability in income levels of the borrowers, average loan balance per borrowers, retention rates of borrowers, increase incomes for borrowers, number of jobs created by financed microenterprises, and the percentage of borrowers attending financial education classes. These indicators will measure both the breadth and depth of the impact of the microloans. Default rates will measure how well borrowers are managing their loans and how successfully they are able to repay those loans. The number of borrowers highlights the outreach of the microfinance system. Variability of income levels of borrowers suggests the different social groups active in microfinance. Average loan balance per borrower gives an idea of how many borrowers are below the poverty line. The retention rates measure how many borrowers return for a subsequent loan, which would suggest that they found their microloans beneficial. Increased incomes of borrowers is a direct indicator of financial improvements. The number of jobs created by financed microenterprises measures the community benefit. Finally, the percentage of borrowers attending financial education classes measures complete production inputs. Borrowers are better equipped to spend their money effectively if they are financially literate.

    I propose to collect my data from MIX Market, which contains relevant financial and social data about performances of the MFIs. I chose to use the MixMarket database rather than the Microcredit Summit Campaign because the MixMarket has more comprehensive data, as discussed by Bauchet and Morduch. I will run a regression analysis using ordinary least squares estimators. I will then compare the impact of homogenous microfinance industries to the diverse microfinance industries.

    Problems with Data Collection

    While collecting data for the analysis I encountered problems finding sufficient data. The MixMarket has a range of financial and social data reported by microfinance institutions. Yet, the financial data on the MixMarket is stronger and more ample than the social data. Default rates, average loan balance per borrowers, the number of women borrowers, and the retention rates of borrowers were the only indicators adequately available.

    However, some of the indicators that I was going to use to measure the impact of institutional homogenous and heterogeneous microfinance systems on its borrowers were absolutely or significantly unavailable. The first unavailable indicator was the variability of borrower’s income. MixMarket collects state level data on “clients below the first poverty line” as well as “clients below the second poverty line.” However the “clients below the second poverty line” data was insufficient and had many missing data sets. With only “clients below the first poverty line” there is not any indication of the number of borrowers at the other income levels. The second unavailable social indicator was regarding financial education. Financial literacy is essential for borrowers to spend their microloans effectively. Thirdly, there was very limited data on how microloans were spent. The MixMarket had some data available on how many microenterprises financed, but not for all of the Latin American countries. Also, the number of microenterprises financed does not give any insight into what constitutes as “financed” and exactly how the microloans benefited these businesses. The fourth unavailable indicator was how many jobs were created as a result of the microloans. The MixMarket had insufficient data since some of the MFIs did not collect certain data about the social impact of microloans.

    The MixMarket has Social Performance Indicators developed by the Social Performance Task Force (SPTF) in the interest of transparency. They use eleven indicator categories to measure microfinance social performance of MFIs (MixMarket 2010).  The eleven indicators are: the institution’s social mission, proper social performance management training by MFIs’ board of directors, range of products and services, dedication to “Smart Campaign Client Protection Principles” by the MFIs and clients, cost of services transparency, poverty levels of clients, lending methodology, number of enterprises financed, and retention rate of clients (2010). The purpose of these indicators is to measure social performance by all actors involved in microfinance, from the MIF board of directors to the borrowers. However, there are questions about whether these indicators adequately represent social performance as it relates to impact.


    Currently, microfinance data is very limited because it is predominantly voluntary data from microfinance organizations. MFIs make decisions on how to collect data, what data to collect, which databases to report to, and what years to report; therefore, the data is not strong. While the Social Performance Indicators from the MixMarket provide important, measurable data, there are some improvements that would increase the potential to understand the impact of MFIs. Not only should the data outlets, MixMarket and Microcredit Summit Campaign, include more social indicators, they should also diversify how they collect data and improve the uniformity in how organizations collect it. The following social indicators should be included: variability of borrowers’ income, the percentage of borrowers that attended financial education courses, how the money was spent by its borrowers and within its microenterprises, how many jobs were created due to the microloans, and the change in income after the borrowing, if any. Also, there should be indicators measuring how the money was spent and how it benefits the families of the borrowers. All of these indicators give greater insight into how the microloans are being spent, as well as and what impact they are having on the families and the communities they are serving.  

    Also, there should be changes to how microfinance data is collected. First, the data is provided by the MFIs on a voluntary basis and then the data is reviewed for cohesion and consistency by the outlet to which the data is submitted (Bauchet and Morduch 2010). However, because this data is voluntary, some MFIs do not collect data for all the indicators published by MixMarket. Missing data inhibits researchers from studying and comparing certain topics if that missing indicator needs to be included in their analysis. An example of the limitations of the current data is the variability in the number of clients surveyed for poverty measurements. According to state level data of Latin American and Caribbean countries on the MixMarket, the “Number of Clients Surveyed for Poverty Measurement” by each country ranged from 0 to 56,000 people for the year 2010. This is too wide of a range to compare national data on poverty, because there is not enough accuracy and continuity in the number of people surveyed to consider the data sufficient.

    Furthermore, data should be collected at all levels rather than only relying on MFIs, because MFIs’ data will likely be heavily focused on institutional level data. Other organizations, such as universities, governments, and external nongovernmental organizations, should become more actively involved in microfinance data collection to create more abundant and expansive knowledge and datasets. Because collecting data is costly, there cannot be a dependence on only MFIs to improve microfinance data. There should be more borrower-level surveys and interviews to get a better measurement of impact. More borrower-level interviews could also expand the accountability of borrowers, ensuring they spend their microloans effectively and properly. However, more borrower interviews and surveys could have some unintended consequences, especially in conservative communities where too much interaction would discourage potential borrowers.

    After forty years, the impact of microfinance is still widely debated. With improvements in technology, loan techniques, and business models, collecting data could get easier and more extensive. Also, as more MFIs become commercialized, there will be an increase in regulation of the industry, because they function as profit-seeking institutions that are accountable to their stakeholders. According to Armendáriz and Morduch’s chapter on commercialization and regulation in The Economics of Microfinance: Second Edition, commercialization calls for improved internal controls by creating more formalized and less flexible business practices than those adopted by non-profits (Armendáriz and Morduch 2010). More regulations could create optimal circumstances for MFIs to collect better social data.

    However, while improved social data is necessary to understand the true impact of microloans on its borrowers and their communities, collecting data is difficult for several reasons. First, collecting data is timely and costly, and microfinance institutions have limited resources to dedicate to it. Second, there are few organizations doing the data collection; therefore, it is hard to get complete data from every region. Third, when interviewing or surveying borrowers the trustworthiness of the response could potentially skew social data. Finally, proving that the benefit or impact on borrowers is directly related to microloans can be difficult to prove, because there are many external factors influencing a person’s social and economic success. These all contribute to the current state of microfinance data and could hinder the improvement of data collection in the future.


    This paper was designed to study the impact of institutional variety in Latin American microfinance systems on its borrowers. To do this, I planned on separating twenty-one Latin American and Caribbean nations into two categories based on the variability of institutions in their systems: homogenous or heterogeneous. Then, using eight social and financial indicators, I planned to run a regression analysis to measure any difference in impact between the two groups to see if institutional variety affects its borrowers more one way or another.

    Unfortunately, the MixMarket currently lacks the data to properly analyze impact on borrowers. A significant number of indicators were unavailable, such as borrowers participating in financial literacy classes, increase in income as a result to borrowing, and whether jobs were created. Another indicator, variability of borrower’s income, had limited data that did not give enough information to adequately fulfill these indicators. Overall, because microfinance institutions collect and report the majority of microfinance data, MixMarket, the most expansive microfinance database, has strong financial data and but only limited social data,.

    The mission of microfinance is to provide access to credit for impoverished entrepreneurs, so they can improve their businesses, generate sustainable sources of income and strengthen their local economies. To measure whether microfinance systems are achieving their mission social data is imperative. Improved data will give us a better understanding of the social impact of MIFs, reinforce accountability of MFIs and how the microloans are spent, and allow for more accurate analysis of microfinance.

    The following recommendations would allow for improved data collection. First, there must be greater variability in the levels of data collection. Data collection cannot just at the institutional level. For example, there should be greater emphasis on microfinance data at the borrower and state level. Second, other organizations must be involved in data collection. Currently, only MFIs collect and report data. This is very limiting, because they decide which data to collect and how to collect it. More organizations, including governments, universities, and research institutions, should be collecting data on microfinance to contribute to its robustness. Finally, more surveys and interviews of borrowers should be conducted to better understand how they are using their microloans and how those loans are impacting their lives.

    Improved social data about microfinance could have potential policy implications, because there would be more information available. A greater understanding of which loan practices are beneficial to borrowers would lead to better practices being adopted by MFIs and ineffective practices being omitted. Improved measurements of impact would create better accountability in microfinance. This would cause regulation in microfinance without regulation being implemented. Finally, national governments could give more support to MFIs if they understood the extent of microfinance’s impact. Microfinance has the potential to make a difference in impoverished communities and lift people out of poverty. Social data could help to stimulate growth and efficiency of microfinance organizations.


    • Assefa, Esubalew, Niels Hermes, and Aljar Meesters. 2013. "Competition and the performance of microfinance institutions." Applied Financial Economics 23 (9): 767-782.
    • Armendáriz, Beatriz, and Jonathan Morduch. 2010. “Commercialization and Regulation.” The Economics of Microfinance: Second Edition, 239-264. Cambridge: Massachusetts Institute of Technology.
    • Bauchet, Jonathan, and Jonathan Morduch. 2010. “Selective Knowledge: Reporting Biases in Microfinance Data.” Perspectives on Global Development and Technology: 240-269. doi: 10.1163/156914910X499705.
    • Campbell, Gregor. 2010. "Microfinancing the Developing World: How Small Loans Empower Local Economies and Catalyse Neoliberalism's Endgame." Third World Quarterly 31, no. 7:1081-1090. World History Collection, EBSCOhost (accessed February 3, 2013).
    • Cull, Robert, Asli Demirgüç-Kunt, and Jonathan Morduch. 2009. "Microfinance Meets the Market." The Journal of Economic Perspectives 23 (1): 167-192.
    • Hermes, Niels, Robert Lensink and Aljar Meesters. 2011. "Outreach and Efficiency of Microfinance Institutions." World Development 39 (6): 938-948. doi:10.1016/j.worlddev.2009.10.018.
    • Hoque, Monzurul, Muhammad Chishty, and Rashid Halloway. 2011. "Commercialization and Changes in Capital Structure in Microfinance Institutions." Managerial Finance 37 (5): 414-425. doi:
    • Meyer, Richard L. 2007. “Measuring the Impact of Microfinance.” In What’s Wrong With Microfinance?, edited by Thomas Dichter and Malcolm Harper, 225-238. Warwickshire: Practical Action Publishing.
    • Microfinance Information Exchange, Inc. 2010. “Social Performance Indicators.” Accessed July 18, 2013.
    • Montgomery, Heather and John Weiss. 2011. "Can Commercially-oriented Microfinance Help Meet the Millennium Development Goals? Evidence from Pakistan." World Development 39 (1): 87-109. doi:10.1016/j.worlddev.2010.09.001