Microfinance as Subprime

Having done a fair amount of work in microfinance and closely related areas (development finance involving business clients with larger-than-microfinance loans) in the developing world, I am overall a big fan.  As many people are.  The question that has long loomed, however, is whether it can or should scale upwards to become a full-fledged part of the global capital markets, or whether it should remain a highly subsidized development activity for very poor people or, most plausibly, some of both.  I wrote about this problem in an article in 2002 – asking whether sufficient attention had been given in the conceptualization of microfinance to the question of whether it was supposed to serve as:

  • a genuinely economic connection between very poor people and the capital markets, or instead
  • a kind of “faux-market” in which the tools of the market were deployed as a form of artificially sustained discipline over the efficient use of resource, but nonetheless massively subsidized and, in that sense, never genuinely part of the global capital markets but instead always some sort of philanthropy.

I, like everyone else I have known in this field, have wanted to see some of the first, some of the second and, most crucially, some kind of “venture philanthropy” merger of the two that would somehow combine:

  • the discipline of genuine capital markets to induce efficient use of capital to promote geniune economic growth;
  • access to much larger pools of capital than are available to government aidagencies or NGOs, through the commercial capital markets;
  • subsidies or guarantees to facilitate the entry of for-profit entities into the sector, in order to help them gain experience with loan-making, monitoring, default, and other costs of microfinance, and to overcome the problem of microfinance’s problematic diseconomies of scale compared to other commercial lending;
  • the many social benefits of microfinance for both very poor people and not-so-poor but still poor people as separate groups; and above all,
  • scalability.

So, back in the 1990s, I proposed internally to the Open Society Institute structures of credit guarantee facilities that would allow a consortium of philanthropic and government aid agencies to offer part-guarantees to banking institutions seeking to enter the sector, with the aim of doing all the above good things.  At the time – and in most situations in which I’ve inquired about this since, with the very particular exception of India – the response from the microfinance organizations was, well, that’s nice – but as a matter of fact, at this point we don’t suffer from a general lack of capital.  We can get capital at a zero or negative capital cost in the form of interest free loans from governments or straight out donations.  We don’t need to tap the capital markets, even in a subsidized form at this point, thanks very much.  Maybe someday; not now.

The reasons why this is so are important.  The microfinance providers with whom I was speaking were generally in the business of microfinance for poor people in which the transaction costs were clearly extraordinarily high for the size of the loan and possible rate of return, if one took into account all the monitoring and active involvement with the borrowers, etc., etc.  And that was leaving aside completely the transaction costs of the foreign donors and any other upstead costs; it was just the narrow cost of a local NGO engaging in microfinance loans.  Everyone likes to tout – or anyway did like to tout – the fantastic repayment rates of these microloans as evidence of client creditworthiness .  But within the sector, practitioners have always been very clear that this is on account of large investments at the front end of monitoring and reliance upon the heavy hand of social stigma and joint and several liability (as a substitute for material collateral) of other members of a “lending circle” as disciplinary mechanisms to ensure repayment.

This is nothing new; microfinance practitioners, although sometimes evangelical in their zeal for it as a development tool, have a pretty decently practical streak, and recognize that this is a subsidized – heavily subsidized – activity when it comes to most clients.  It is another instance of the problem that much of development, as William Easterly tirelessly points out and Jeffrey Sachs seems gradually to be acknowledging, is not a scalable activity.  It takes place at the capillaries, and the blockages are not the mass flows of capital – it is what happens in the “last mile.”  Talking with a finance academic who has decided to start teaching in this area – he remarked somewhat ruefully, I can’t get my students interested in this because the whole point of finance is scalability.  But there are many extraordinarily bright and experienced finance experts, people who perhaps made some money on Wall Street and decided to do something more personally satisfying in the last fifteen years, who have been bringing an immense amount of sophistication to the problem of applying finance to development.  Parts of it have worked, and parts of it are showing the problems, which is a somewhat understated way of stating the current banking for the poor crisis in India.

The grail of transforming at least part of the sector into something that is genuinely economically sustaining, in the sense of covering its costs, and being able to scale up to the point that tapping the commercial markets for capital, has never gone away.  The attraction is greatest in India – second would perhaps be South Africa – places with many very poor people, many pretty poor people, many poor people, but globally connected, globally sophisticated, utterly first world banking sectors.

India, particularly, because the size of the internal market – in this as in many other things – but also an underdeveloped and underserved one, with takeoff underway in so many other sectors, has reasons to be attracted to this model.  Economic takeoff is going to require banking models that can reach to poor people in a commercial way; it’s not precisely microfinance, and not microfinance in the “faux-market” development sense I suggested earlier.  It is the search for a genuinely commercial product of banking for the poor that provides capital and banking services – but which manages to cover costs and return a profit.  NGO development programs cannot possibly serve the needs of all those people at that level; their specialization is with a different population of very poor people.

For all these reasons and more, I have been supportive of the efforts to try and commercialize banking for poor people, in India and elsewhere.  I’ve supported rich philanthropists putting money into these businesses in an effort to try and meld the doing good and doing well.  However, the melt-down underway in India of the current model certainly gives me pause, and the belief that a fundamental rethink of the intersection of doing good and doing well is in order.  The New York Times and the Economist each have good stories this week on the crisis in India for a company that went public in India as a microfinance lender.

It’s an economic, political, and social mess in India.  Yet, although it will indeed set back the commercialization of banking for the poor for quite a while, I am persuaded that it is not a bad thing to have to sit down and re-consider the premises of venture philanthropy and combined social-profit motivations.  I say this as someone who if, for example, the Open Society Institute or some philanthropist had invited to get involved in advising things, would have leaped in – I am confident I would have led down exactly this path.  I plead no special powers to have seen ahead.

However, with the benefit of hindsight, a couple of things are becoming clear.  The banking for the poor model has important similarities to the US subprime crisis.  Particular regions of India were deluged with capital that came cheap, in part because of the subsidies for it both implicit and explicit. Lending standards were relaxed, in part because the lenders were seeking to overcome the enormous hurdle of diseconomies of scale in tiny loans – the monitoring and loan-making costs for tiny loans.  But let’s add one important difference.  So far, microcredit – crucially and more exactly, within the analytic terminology of this post, “banking for poor people” – has not yet been securitized directly, nor has it had credit derivatives built on top of it.  As someone who has been occasionally involved trying to dream up upstream financing structures that could do securitizations and derivatives in this sector, I just would like to say that I’m glad that up to this point, the sector has not yet been leveraged up in those ways.  I’m not opposed in principle to the idea that “prudent” leverage could draw more capital efficiently into the sector; I just don’t think we have any way at this point of figuring out short of meltdown what that level is.  This should, of course, sound familiar.

The problems in India are problems of an excess of capital; poor incentives among the lenders (volume not quality, for example); and a failure to be realistic about the rates of return actually achievable by poor people even when they have availability to capital; etc., etc.  But they are still mostly at the level of the limits of poor borrowers; or, again, more precisely, what happens when poor borrowers meet global capital, in the form of expected returns that can’t really be expected (i.e., opportunity cost for global capital).  That is half of it – if you’re going to attract real global capital, you have to somehow manage to pay global capital rates and that requires economic activities that produce at least that net rate of return.

But, crucially, the other half that drives this sector is apparently opposite, but actually helps crucially swell the bubble, ratchets it up, because it is the nip that draws the cat of capital out of some better return elsewhere and into this particular place, so producing a bubble.  That other half consists of rich people, rich philanthropists, for whom the amounts are simply too tiny to worry about, not really.  It helps lead the herd of capital to indisciplined lending – not the only thing, of course, but an important component, and important component in the lack of clarity that surrounds rational choice in the sector.

But it also arises from some of the most celebrated new lending models that take advantage of the “retailization” of every encounter globally via the internet; one can exchange illegal child porn, or play chess, or make microloans all the way around the world.  The model, growing in popularity, for direct person to person lending, individual rich-worlder to individual poor-worlder, is great in one way – but let’s ask ourselves, is it such a good idea to have one-to-one lending on this basis?  Should we maybe ask ourselves why in the developed world, we use intermediaries and banks.  Sure, one answer is that a huge amount of informal lending takes place through friends and family, not intermediaries, and in a sense this model replicates that.  But, well, it doesn’t, because as we all know by now, the internet creates internet friends and family, not actual friends and family.  The social virtues, as Adam Smith would have described them, are not precisely the same across continents and over the internet as they are with people with whom one has actual, not virtual, social intercourse.

And then there is the problem that what is little money to George Soros or to me or you is really big money to someone in the poor world.  They need the money, but its efficient use requires that we take it seriously and that they take it seriously.  We don’t take it seriously.  How could we?  And yet the consequences of us not taking it seriously are an unsustainable bubble, asset inflation in already poor zones, many other bad things.  We just log off and go back to our real lives, but the effects can be – are – very real.

This is what makes The Onion so hilariously right, as it nearly always is – the ludicrousness of anyone in the first world pretending that this “lending” is anything other than “donating”:

Representatives from One World Finance, a U.S.-based microcredit provider, confirmed Monday that they had initiated foreclosure proceedings on a goat in southern India following a borrower’s repeated failure to make her $2.20 monthly loan payments. “I tried to work with Ms. [Subha] Thangam on this, but once she fell a full $6.10 behind, I had to repossess the goat,” said loan officer Michael Conrad, who stated that he was just doing his job and that it was “not [his] fault” if certain subsistence farmers were living beyond their means.

Let me be utterly clear that this is not an argument for not deploying the money; anything but.  It is needed in many places.  But the combination of easy money from the rich world that, if liberated from market discipline on the local end, can create vast problems, is one that has to be re-thought at this point, in the actual practice and alignment of incentives in this sector.  We are not willing to take the goat; but if we wish to avoid bubbles and the very real damage they cause, as well as what we somewhat too anodynely call “efficiently allocate capital,” we need to ensure that our capital goes to some entity that is willing to contemplate something close to that.  Otherwise it is not operating on the margin that matters in that society, and the results are almost certainly a bubble.  I can’t do that; you can’t do that; we need not to interact in a touchy feely way with our borrower-donee, but instead hand our money over to an intermediary that has the right incentives to find that margin.

One can pile up important similarities and differences, in other words, between India’s microfinance bursting bubble and the subprime crisis.  But let me focus on one that is perhaps less noticed.  I notice it as a similarity because it’s something that (as someone who works out in the gym in Fannie Mae’s basement in Washington DC), I have heard a lot over the past dozen years: a tendency to play a self-deceptive bait and switch between doing good and doing well.  I.e., the many conversations with Fannie Mae senior staff who, when things were going well, thought (what they thought of as) their mixed social-profit model must be great, and as things weren’t going so well, took comfort in the idea that they were doing good and this was merely a cost of doing “good” business.  Something like that seems to have been present here – which hardly surprises me because I confess to having been tempted to it many times, working in or advising organizations with similarly mixed motives.

The invitation to self-deception is high, in other words.  Bertolt Brecht wrote a play – famous in its day, and one of his writings that deserves to  live on – The Good Person of Szechuan, in which a young woman of tender and generous heart inherits a tiny shop, but discovers that she cannot keep it afloat because she cannot say no to all the need around her.  So she goes on a journey and then her cousin comes to run the shop – ruthlessly and with an iron hand to make it profitable again.  And so it goes several times round.  Brecht thought of this as a condemnation of capitalism; it is perhaps rather more instructive of the virtues of not mixing motives.  I remain as committed to microfinance, as a development tool for the very poor in “faux markets,” on the one hand; and to banking for poor people as a genuinely commercial activity, on the other, subsidized in various ways.  But I do think it is time for some deep reconceptualization of the latter, particularly, and its model of capital and its social uses.

ps.  There is a vast literature, much of it excellent, on the theory and practice of microfinance.  But if you’re interested in further academic reading on this particular topic, the “upstream” funding issues, let me recommend two recent law articles.  The first is by Kevin Davis and my American University colleague Anna Gelpern.  The second is by my co-author on financial regulation, Duke University’s Steven L. Schwarcz.  My own essay on this topic, as mentioned above, is here.  And, okay, let’s acknowledge, as usual, The Onion got there first.  (HT: Insta commenter.)