Roodman Reviews Role Reversal Revisited

The microfinance rating agency MicroRate recently released an update of its Role Reversal report, which received wide attention five years ago. Like the original, Role Reversal Revisited makes a strong case that public agencies such as the World Bank's International Finance Corporation are investing heavily in microfinance institutions that are also capable of attracting finance from private investors. Thus public investors are "crowding out" private investors, in contravention of their own policies. The report is succinct and punchy and raises tough, billion-dollar questions. It is well worth reading.When the first Role Reversal report appeared, CGAP hosted an online public discussion about it. I was a newbie in microfinance, so my contribution to that forum was tentative in tone. I felt that R² presented compelling evidence of crowding out---but did not provide as-compelling evidence for why I should care. Roughly speaking, I feel the same about R³.But there is a difference this time around. In addition to being more confident, my view has evolved. I've learned a lot more and written a book. In the last chapter, I confront the problem of too much capital going into microcredit. I conclude that while "the rule that public investors should exit when private investors enter is blunt and has a fuzzy rationale,...curtailing the flow of public money into microcredit portfolios would probably make the world a better place."So I will explain why I don't quite find the report's arguments compelling yet broadly share its conclusions. One virtue of the report is its clarity. It is succinct and full of evidence and leaves no doubt about its point of view. To explain its starting point, I can do no better than quote it:

It starts from the premise that as the microfinance industry matures, its growth can and should be financed by private resources. The role of development finance institutions is to pave the way for those resources: DFIs [public development finance institutions] should only go where private lenders don’t yet dare to tread and act as catalysts for private funding but they should not compete with it.

The idea that public investors should step in only where private ones fear to tread is the basis for the sharp public-private distinction that runs through the report. Prima facie, the principle is reasonable. But it is also abstract. And the report never justifies it: truly, it is a premise. Since the lives of millions of people are affected by the actions debated here, this unjustified premise deserves scrutiny. Reading the report, a bunch of rationales for the premise came to mind, most of which I do not find convincing:

1. It is wrong for public agencies to interfere in private markets. It's not the American way, you might say. I don't presume that this is the view of the authors, Damian von Stauffenberg and Daniel Rozas, but the idea is in the genes of the premier DFI, the International Finance Corporation, and I would assume other DFIs as well. It lies behind the prohibition in the IFC's Articles of Agreement (III.3.i) on investing where private capital is available, a prohibition that MicroRate invokes. So dominant was the U.S. government in the World Bank in the 1950s that the debate over the creation of the IFC was primarily an American one. Staff of the young Bank saw the need for a way to invest in the private sector. But the U.S. Treasury and Federal Reserve opposed the proposal with intensity. In the end, they lost, but their opposition formed the crucible for the IFC:

   Several reasons accounted for the negativism of the school of thought represented by this group. Foremost among them was the strong conviction that participation of an intergovernmental institution in equity ownership of private enterprises was basically wrong. Accordingly, it was pointed out that such participation involved a kind of partnership between government and private enterprise that essentially ran counter to the principles of our economic system and therefore placed these principles in direct jeopardy. (Bronislaw Matecki, 1956)

   But the poor cannot eat market morality. And we should not weep for the wealthy social investors whose returns are depressed by competition from public investors.
2. According to the letter or spirit of their rules, DFIs are supposed to exit when private investors enter. Certainly institutions should obey their own rules. But I have trouble getting exercised about this, and clearly the people at MicroRate do not wake up each day inspired by the mission of making large public institutions follow their rules. Again, the impacts on poor people and an industry that aims to serve them are what matter most.
3. Especially if private money is plentiful in microfinance, public money can do more good elsewhere. This gets to the other idea encoded in the genes of the IFC: that the success of a DFI should be measured by how much private investment it catalyzes. If it funds projects so uneconomical as to repel private capital, or dumps in so much public capital that there is no room for private, then it is failing to leverage its limited resources. I think that's a strong argument. But it applies to private social investors too, who I suspect generate the majority of private capital going into the microfinance investment funds whom MicroRate would defend. If DFIs would do better by investing elsewhere, maybe the social investors would too. They could both make more room for commercial capital.
4. The microfinance industry will "inevitably outgrow the ability of even the largest DFIs to satisfy its funding needs"; and it is better for the inevitable to happen sooner (page 15). An odd argument on the face of it.
5. It is the fullest realization of the movement for microfinance institutions to untether from donors and connect in rich ways with other private actors. This is the perspective of my chapter 8. It views economic development as an exploratory process of increasing complexity, rather like the development of a child or the evolution of an organism.

It is the last rationale I buy most; I suspect it is closest to the MicroRate philosophy too. But consider: will the microfinance industry realize its full potential to enrich the economic fabric and serve the poor via a shift from channeling public, overseas, philanthropic capital into microcredit to channeling private, overseas, philanthropic capital into microcredit? I think not---and with more confidence than when I made the point five years ago. Much greater enrichment (not to mention stability) of the financial fabric comes when microfinance institutions move beyond credit, to taking savings locally, and becoming full intermediaries. The report suggests as much:

Compartamos has held a banking license since 2006, though as of year-end 2010, it had not raised any funds from retail deposits. It is thus worth asking---could large DFI loans to mature MFIs like Compartamos provide them with a disincentive from raising deposits? Could they thus be hindering the deepening of the financial offerings to the poor that the DFIs claim to be supporting?

But if this is the long-term goal for the industry---and I believe the report is concerned with what is best for the industry in the long-term---how different, really, are public and private capital? Cannot both create disincentives for deposit-taking? Cannot both be overstimulated by naive expectations about the efficacy of microcredit? Cannot both, out of the best of intentions, create bubbles?Now, while I don't think the report absorbs this argument to the extent it should---that would require abandoning the sharp public-private split---it does recognize the similarities of public and private finance in passing, in an interesting way. After pointing out, as I have, that both public and private investors inflated microcredit bubbles in such places as Bosnia and Morocco, it asks whether private investors are less the bull in the china shop, less apt to dump too much finance into a microcredit market:

Would private Funds have been more cautious than their government-owned counterparts? Since private lenders played a minor role in Morocco, one can only speculate. The example of Bosnia raises doubts that they would have. But it is a fact that markets punish private lenders much more harshly for mistakes: a disappointing return affects a Fund’s ability to attract more money, whereas the governments that own DFIs would hardly notice loan losses.”

In other words: Private investors are flawed too. But on balance, we believe they are less flawed.Not the most compelling case for the superiority of private investors---but one I take seriously all the same. Here's how I put it in my book:

...conversations with industry insiders have persuaded me that the public–private distinction has teeth. The nub of the matter is that most private investors are small specialists. Bad calls can permanently damage their reputations. Such mortal risk focuses the corporate mind. As private companies with staffs numbering in the dozens rather than thousands, they are much less rule-bound, more agile. In contrast, major public investors such as the International Finance Corporation (IFC) are muscle-bound generalists. After the recent crises, the private investors seemed quicker to recognize and respond to troubles than the public investors; they even gently discouraged prospective investors in their funds.

One of those industry insiders, I suppose I can now reveal, was Damian von Stauffenberg, lead author of R³.But the principal problem here is overcrowding, not crowding out. Shortly before Damian released R², he warned at a CGD-sponsored event that the industry would soon lose its innocence: somewhere, probably within a year, a microfinance institution would fail. He was off on the timing, but right in spirit. Five years on, bubbles have popped and MFIs have gone bust. It is more obvious than when the original report appeared that the greatest threat to the greatest strength of the microfinance movement---its capacity to enrich the local institutional fabric and extend its reach to the poor---is an abundance of investment, public and private.