What Does the Data Tell Us? (Part 2)
In my last post, I examined data for Russia, Mexico and India, three large microfinance markets in different regions. We looked at the correlation between portfolio yield and loan size and discussed interest rates and profitability for MFIs with different focuses in terms of rural/urban populations and share of female clients. Comparing these markets, we observed a much broader range of yields in the Russian and Mexican market as compared to the Indian market.
How about operating expenses? It may be interesting to note that operating expenses are particularly high in Mexico, given that the cost of transportation, communication as well as salaries are higher than in many other countries and among the highest in Latin America. This is reflected in the median operating expense ratio reported by the MIX, which is 61.45% in Mexico, 16.21% in Russia and 11.12% in India. Looking at the median for only those MFIs with an ROA > 0 (sustainable), the median for the Mexican market is 48%, while it is 15% for Russia and 11% for India. From these numbers we can conclude that the median operating expense ratio for Mexican MFIs in considerably higher than for Indian MFIs, which could explain the higher market curve (yield) for Mexico. We might expect Mexican MFIs to be less profitable due to their higher operating costs, yet the analysis suggests pretty high ROEs. Could this be the result of an opaque pricing environment? What other factors could explain these findings? Will enhanced pricing transparency have an impact on this, as it promotes healthy consumer choice and competition among institutions?
How is the impact of consumer protection policies, particularly pricing transparency, reflected in these analyses? In Mexico, for example, regulated institutions are required to disclose the EIR of their loan products (called CAT in Mexico), however institutions can get a legal exemption fairly easily, while there is also evidence that clients do not understand the concept of the CAT and therefore don’t base their financial decisions on a comparison between the CATs of competing loan offers. Also, a large number of MFIs in Mexico are not regulated and don’t disclose effective interest rates. In India, pricing transparency efforts seem to be relatively stronger, and the most important networks and The Reserve Bank of India have established standards of consumer protection and pricing transparency for their member institutions. As we have seen, the range of yields made in the Indian market is much smaller than, for instance, in the Mexican market, despite even smaller average loan sizes.
How will the data collected by MFTransparency allow us to draw an informed conclusion? In this analysis we looked at real portfolio yield as a proxy for interest rates. As portfolio yield provides an average of the different prices charged on all of the lender’s products, it is not a useful indicator where MFIs offer a range of different products. MFTransparency’s country-level data allows us to compare the actual APRs and EIRs, so we don’t have to rely on total portfolio yield as a proxy. It allows us to compare the prices of all products offered in a given market, not just the average price of the range of products an MFI provides. The graphs presented in this post also don’t distinguish between different product uses, so we may be comparing apples with mangos. The interactive graphs on MFTransparency’s website allow us to analyze country-level data and distinguish between different purposes, such as business, consumer, education, housing and emergency loans, which have very different characteristics and therefore different prices. Our country surveys and the range of educational materials our team develops, will allow us to accurately interpret the data we collect in each country and take into account the unique characteristics of each microfinance sector worldwide as well as possible explanations for differences in loan prices, profitability and the shape of market curves.
What conclusions do you draw from this analysis?