Baron, J., and J. Greene. 1996. “Determinants of Insensitivity to Quantity in Valuation of Public Goods: Contribution, Warm Glow, Budget Constraints, Availability, and Prominence.” Journal of Experimental Psychology: Applied 2(2):107–25.Google Scholar
Bester, H. 1992. “Bertrand Equilibrium in a Differentiated Duopoly.” International Economic Review 33(2):443–8.Google Scholar
Brock, W., and S. Durlauf. 2001. “Interactions-Based Models.” In Handbook of econometrics, edited by J. Heckman and E. Leamer, Vol. 5. Amsterdam: North Holland.Google Scholar
Dupas, P., S. Green, A. Keats, and J. Robinson. Forthcoming. Challenges in Banking the Rural Poor: Evidence from Kenya’s Western Province. NBER Africa Project Conference Volume.Google Scholar
Hollo, D. 2010. “Estimating Price Elasticities on the Hungarian Consumer Lending and Deposit Markets: Demand Effects and Their Possible Consequences.” Focus on European Economic Integration 2010(1):73–89.Google Scholar
Hopkins, R. and Scott, C.D. 1999. The Economics of Non-Governmental Organisations. Available at SSRN: http://ssrn.com/abstract=1126984
Jaumandreu, J., and J. Lorences. 2002. “Modelling Price Competition under Product Differentiation and Many Firms (An Application to the Spanish Loans Market).” European Economic Review 46:93–115.CrossrefGoogle Scholar
Kapoor, M., J. Morduch, and S. Ravi. 2007. “From Microfinance to m-Finance.” Innovations, Technology, Governance, Globalization 2(1–2):82:90.Google Scholar
Karlan, D., and J. Zinman. 2010. “Expanding Credit Access: Using Randomized Supply Decisions to Estimate the Impacts.” Review of Financial Studies Oxford University Press for Society for Financial Studies. 23(1):433–64.Web of ScienceCrossrefGoogle Scholar
Manski, C. 1995. Identification Problems in the Social Sciences. Cambridge: Harvard University Press.Google Scholar
Morduch, J. 1999. “The Role of Subsidies in Microfinance: Evidence from the Grameen Bank.” Journal of Development Economics 60:1999.Google Scholar
Navajas, S., J. Conning, and C. Gonzales-Vega. 2001. “Lending Technologies, Competition and Consolidation in the Market for Microfinance in Bolivia.” Journal of International Development 15(6):747–770.Google Scholar
Wooldridge, J.M. 2002. Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: Massachusetts Institute of Technology.Google Scholar
About the article
Published Online: 2013-08-28
Microfinance Institutions (MFIs) offer $100–$5,000 loans to customers with little-to-no collateral.
Cull, Demirguc-Kunt, and Morduch (2009) provide further discussion of the composition of firms in the microfinance industry.
Kapoor, Morduch, and Ravi (2007) find many examples where microfinance borrowers can pay installments on loans via phone drastically reducing the transaction costs associated with micro-lending.
In general, little has been done on the welfare implications from microloans except for Karlan and Zinman (2010). Using South African survey data they find that when a provider relaxed borrowing restrictions, the marginal borrower was actual relatively more productive and MFI profits as well as consumer welfare increased.
Consumer surplus may go up due to more customers (shifting out of demand) or existing customers facing lower prices (movements along demand curve).
The model is similar to Dick (2008) which looks at competition in the U.S. deposit market.
McIntosh, deJanvry, and Sadoulet (2005) find in Uganda that increased competition results in multiple loan-taking by borrowers. This further substantiates the need for a continuous choice framework.
Data discussion follows in Section 3 of the paper. Data used to compile Table 1 was gathered by the Microfinance Information Exchange (MIX) and is available from MIX market website: http://www.mixmarket.org/.
Any reference to product or firm is synonymous as we presume that each MFI produces a single product. The product differentiation exists not within the firm but between firms. This structure maps into the data very well.
The warm glow effect suggests a consumer preference for firms which appear to behave altruistically. For instance, consumers are willing to pay a price premium to shop at Whole Foods in part because of the clean and green image they offer (Baron and Greene 1996).
The set of institution types ultimately used in the analysis includes: Non-governmental organizations (NGOs), traditional banks, non-banking financial institutions, credit unions, and rural banks.
In microfinance almost all collateral is non-traditional. It varies from household assets, like sewing machines, to forced savings or guarantees from peers.
In addition to maximizing borrower welfare, even NGOs likely maximize profit to some degree given donor-pressures to eventually become self-sustaining. Cull, Demiguc-Kunt, and Morduch (2009) note that over the past two decades, all MFIs have been encouraged to achieve financial self-sustainability by earning ample “profits”.
MFI i and the group of competitors in can take on any characteristic and are in competition with all other MFIs in m.
To estimate the model, attributes and prices in are measured as the average of all other MFIs (exclusive of MFI i) within the country m where MFI i is located.
Note that theoretically appears as a constant. We estimate the empirical model allowing for multiple specifications around this assumption. Note also that .
It follows from prior literature that the cross-price elasticity of demand for non-collateralized consumer lending in developing countries is approximately 0.25 (Hollo 2010). This elasticity is used to construct an estimate for for each firm at each point in time. Methods used to construct are derived in the appendix. We use this estimate, since no other estimate for the cross-price slope exists in the literature. We note later in the paper how other parameters of demand are simple monotonic transformations of this estimate.
Data available from MIX market website: http://www.mixmarket.org/
The structural model outlines results based on marginal cost. Ideally the data would include this variable as well. Since it does not, we qualify results with the caveat that assumes equal average variable cost and marginal cost.
A breakdown of these statistics for each country is not included in this article. That being said, the country-level breakdowns strongly support the trends in Table 5. Market shares subdivided by institutional types show very little evolution over time. These results are available upon request.
Region and year fixed effects are not reported but available upon request.
Tests of the null hypothesis that also cannot be rejected.
As noted earlier we do not include the exact location of each MFI within a country as an additional attribute due to data limitations. This omission may be introducing some bias in our estimates.
Models that include region–attribute interactions are estimated with little impact on the main results of the paper. (i.e., The overall impact of these minor differences has little effect on the end results that relate to consumers’ surplus.) These results are available upon request from the authors.
See appendix for derivation
Results are similar if we estimate CS using the parameters from a pooled OLS regression of column 2 in Table 6. We do not estimate consumers’ surplus using regressions that control for lending methodology, since the sample of countries that includes these controls is significantly smaller.
This includes NGOs, banks, credit unions, non-bank financial institutions, rural banks, or other financial institutions.