![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
This study explores the relationship between human rights and the success of microfinance institutions (MFIs). Microfinance emphasizes the empowerment of women, yet no study has examined whether the existing human rights environment, or the rights environment for women specifically, helps or hinders the effectiveness of this grassroots development approach. We test competing hypotheses, including the possibility that the rights environment affects MFI success, the possibility of an inverse relationship between levels of women's political or economic oppression and MFI success, and the expectation of no relationship. Our quantitative analysis of MFIs in the Opportunity International Network suggests that the overall human rights environment in which they operate has significant effects on repayment rates, while women's economic rights affect the operational self-sufficiency of MFIs. This has important implications for our understanding of the factors that make microfinance institutions viable, and for the degree of access that underserved communities have to credit.
Acknowledgements
The authors would like to thank Angela Bos, Amyaz Moledina, and James Warner for their comments and suggestions on previous drafts of this article and Sandeep Bhusal, Kyla McEntire, and Aneeb Sharif for their valuable research assistance on this project. This article does not represent the views of the World Economic Forum. The authors alone are solely responsible for its content.
Notes
1. Since traditional banks are not willing to give loans to poor people without collateral, the poor are not able to invest in ventures that could potentially get them out of poverty. Traditional banking systems have difficulty providing loans to the poor because of asymmetry of information between borrower and lender, including screening, monitoring, and enforcing credit contracts (Bhatt and Tang Citation1998: 119).
2. We owe some of these points to interviews by the second author with MFI officials from a country in Southeast Asia. Interviewees noted the need for MFI staff members to use bribery to keep local officials from impeding MFI work and detailed restrictions placed on movement within given regions by international donors seeking to monitor or aid MFIs in sensitive areas.
3. For an alternative set of critiques of the negative effects of social capital focused on how microcredit lending may actually reinforce gendered oppression, see Rankin (Citation2002).
4. However, some literature suggests that even though the woman would not be controlling the funds directly, this will indirectly improve her standing within the family and community by becoming an avenue of credit for a household (Wright 2000: 4).
5. When lenders charge high interest rates, it can negatively affect repayment rates by discouraging creditworthy borrowers or tempting borrowers to opt for riskier projects (Stiglitz and Weiss 1981).
6. A “Trust Group” is “10 to 30 entrepreneurs, usually women [who] pledge to guarantee each other's loans and support one another's businesses” (Opportunity International 2015: para. 1).
7. See Norell (2001) or Rosenberg (2006) for excellent discussions of the relative advantages of using Portfolio at Risk greater than 30 Days as a measure of arrears or loan repayment delinquency.
8. We owe this point to an interview by the second author with a cofounder of Conscience Community Development and Aid, a microfinance institution in Myanmar.
9. In the sample used in this study, there was little correlation between our measure of the MFI portfolio at risk greater than 30 days and operational self-sustainability (r = −.056).
10. Standard practice dictates that we recode all values of 0 as 1 before taking the natural log. However, doing so means that those values wind up greater than other values above 0 but below 1, since we have other observations in the sample with values of PAR > 30 greater than 0 but less than 1. Hence, our decision to recode zeroes as 0.01, which is very close to 0 but still less than the next lowest value of PAR > 30 in our data set that is 0.06.
11. This is only partially related to profitability, though both self-sustainability and profitability mean more capital available for loans or other financial services. We decided not to use an MFI's profit margin as an indicator of success in part because of critiques of it as a measure of MFI success (Cull, Demirgüc-Kunt, and Morduch Citation2007). Also, when used in place of operational self-sustainability, the dependent variable is almost completely determined by the lagged dependent variable, yielding little in terms of explanatory power to a study such as this one.
12. As Rosenberg notes, both Return on Assets and Return on Equity “are appropriate indicators for unsubsidized institutions. But donor interventions more typically deal with institutions that receive substantial subsidies, most often in the form of grants or loans at below-market interest rates. In such cases, the critical question is whether the institution will be able to maintain itself and grow when continuing subsidies are no longer available” (2006: 4). This suggests that self-sufficiency measures are more appropriate indicators of MFI success.
13. We use this measure in part because some of these rights are relevant to the typical clients of MFIs, and in part because we want measures of rights from within the same data set (CIRI) so as to be able to better compare rights across type. We are sensitive to the possibility that other measures of gender equality, which are either directly or indirectly related to women's economic rights, should also be considered. An excellent alternative measure is the UN Development Programme's (UNDP) Gendered Inequality Index (GII):
The Gender Inequality Index (GII) reflects women's disadvantage in three dimensions—reproductive health, empowerment and the labour market—for as many countries as data of reasonable quality allow. The index shows the loss in human development due to inequality between female and male achievements in these dimensions. It ranges from 0, which indicates that women and men fare equally, to 1, which indicates that women fare as poorly as possible in all measured dimensions. The health dimension is measured by two indicators: maternal mortality ratio and the adolescent fertility rate. The empowerment dimension is also measured by two indicators: the share of parliamentary seats held by each sex and by secondary and higher education attainment levels. The labour dimension is measured by women's participation in the work force. The Gender Inequality Index is designed to reveal the extent to which national achievements in these aspects of human development are eroded by gender inequality, and to provide empirical foundations for policy analysis and advocacy efforts. (UNDP Citation2013: para. 1)
Unfortunately, the GII is only available for select years (2000, 2005, and 2010) in the time frame of interest. Using this measure as a substitute for women's economic rights would dramatically reduce our sample size, so we chose not to employ it. However, to ensure that Women's Economic Rights still captures the essence of these issues, we checked the correlations between it and the GII. In 2000 (r = –.62), 2005 (r = –.68), and 2010 (r = –.71), these indicators were highly and negatively correlated, and all correlations were statistically significant (p < .01). That is, greater economic rights for women are strongly associated with less gendered economic inequality. As such, we are confident that we have accounted both directly for those factors included in the CIRI Women's Economic Rights and indirectly for the economic and social rights accounted for by the GII.
14. We had hoped to include Women's Social Rights as well, but the CIRI data set does not have consistent yearly data across space and time for that indicator. This would dramatically reduce (by more than half) the number of observations in the study. Moreover, while Women's Economic and Political Rights are (perhaps surprisingly) not highly correlated within this study's sample (r = .120), Women's Social Rights are strongly correlated with Women's Economic Rights (r = .481). Therefore, we omitted Women's Social Rights from this study.
15. None of the four human-rights-related indicators are strongly correlated with one another. All correlations between and among the human rights indicators have correlation coefficients of r < .200. However, to account for the possibility that including all of these indicators in one model simultaneously might affect the results, we reran our models for all three dependent variables removing one or both of the nongendered rights variables. The substantive results barely changed. Therefore, we include in this article only those models that include all four human rights variables.
16. Rosenberg argues that “[t]his indicator is more useful than the cumulative number of loans made or of clients served during a period. Among other distortions, cumulative numbers make an MFI offering short-term loans look better than one providing longer-term loans. The recommended measure counts active clients rather than ‘members’ in order to reflect actual service delivery: members may be inactive for long periods of time, especially in financial cooperatives” (2006: 2).
17. We also considered including the Average Loan Balance per Borrower, indicating the typical size of the loans given by the MFI, as a control for the depth of outreach to the poorest of the poor, and by extension the poverty level of the MFI clientele (Rosenberg 2006: 2). Rosenberg explains that the Average Loan Balance per Borrower is a rough proxy for poverty of the MFI clientele “because better-off clients tend to be uninterested in smaller loans” (2006: 2). However, he cautions that “[l]ow loan sizes do not guarantee a poor clientele. Likewise, growth in average loan size does not necessarily mean that a MFI is suffering ‘mission drift.’ As an MFI matures and growth slows, a lower percentage of its clients are first-time borrowers, and average loan sizes will rise even if there has been no shift in the market it is serving” (2006: 2). This methodological concern, plus the high degree to which Average Loan Balance per Borrower is correlated with some of our other variables, convinced us to omit this variable from our models.
18. As a robustness check, we also ran an alternative specification of this model, which included the lagged version of Portfolio at Risk > 30 Days to account for the possibility that problems with loan repayment might affect Operational Self-Sufficiency of MFIs. That variable turned out to not be significant, nor did the rest of the model see substantive changes in effects or significance, so we did not include these results here due to space considerations. They are available on the authors' website (http://discover.wooster.edu/mkrain/research/) along with replication data for all of the models described here.
Additional information
Notes on contributors
Elisabeth c. Bremer
Elisabeth C. Bremer is a Consultant to international public and private sectors, working most recently with the Russian Direct Investment Fund and the World Economic Forum. She graduated with a BA in International Relations from the College of Wooster in 2008, where she wrote her Independent Study Thesis, “A Gendered Analysis of Microcredit Lending: The Effects of Women's Rights on Grassroots Development.”
Matthew Krain
Matthew Krain is Professor of Political Science at the College of Wooster. His research examines the causes and consequences of repression, human rights abuses, and large-scale political violence and the role of international actors in causing or preventing conflict and violence.