The biggest mystery about Mexico is understanding the numbers. They just don’t quite seem to add up. And that’s what was dogging me throughout the visit, including the two days spent in Mexico City talking to various actors in the sector.
I was lucky – it just so happened that ProDesarrollo was releasing its 2013-14 Sector Benchmarking, and I managed to get myself invited to the event. A great opportunity to network with many actors in the sector at once. I also got to see the presentation of the market figures. At the outset of the trip, I laid out several hypotheses. It seems to me that there are really only two that matter more than all the rest: 1) the number of unique clients and number of loans they hold, and 2) the profile of the MFI clients on which the market rests.
The question of market size continues to bother me. ProDesarrollo reports roughly 6.5 million active combined clients among its 82 affiliate MFIs. That’s by no means the whole market, but it is the bulk of it. It’s also not hugely different from the 2012 MIX Market figure of 6.0 million clients, so my earlier estimate of some 4 million unique clients is close enough. It’s also something that seemed reasonable by several microfinance practitioners I talked to. Of course, this figure is a guess. Without better data, it will remain that, at least for now.
It also remains the case that these clients are poor. How poor? Well, some months ago, I cited the IPA study of Compartamos clients, which pegged the average client’s household income per adult at 1,571 pesos per month ($120) and an average loan amount of 6,462 pesos ($500). Given a loan term of 16 weeks and an average APR of 110%, the weekly loan payment (principal and interest) would be about 480 pesos, or nearly 2,100 / month.
So we have a dilemma – the loan repayment is 130% of the client’s income. That’s a seemingly impossible ratio. Were all loans purely business loans, one could rationalize that the increase in business income would offset the repayment requirement (though such an increase wasn’t found in the IPA study). But it’s well known that many loans are taken for consumption-smoothing purposes. And if a family might manage a costly loan to tide over a difficult period (by relying on a spouse’s income, for example), consistent borrowing above one’s income is hardly a sustainable recipe. And that doesn’t even include the problem of multiple borrowing, which could magnify loan repayments by a factor of 2, 3 or more. To complicate matters further, the same study found essentially no change in the client’s financial situation.
It’s a dilemma I can’t resolve with the given data. The figures just don’t work. It may well be that the incomes captured by the IPA study are incomplete (and they do seem awfully low, even for poor families in Mexico). Or perhaps multiple borrowers tend to be wealthier. I’ve asked the study’s authors to share additional data they have collected during the study that may shed additional light (or disabuse me of a silly misunderstanding of what that data represents). I will share what I learn in due course.
For now, these core questions – who are Mexico’s microfinance clients, how many of them are there, and how much debt do they have? – remain unresolved. These questions shouldn’t be impossible to answer, but until then, Mexico will remain a mystery.
But therein lies the key to understanding the problem of overindebtedness in Mexico.