I've been pretty oblique in recent posts about my evolving opinions of the Andhra Pradesh (AP) crisis. I've been trying to share my thought process with you. But that seems to have left me open to misinterpretation and criticism for poor construction. Perhaps I have taxed your patience. So let me be clear: in a week of talking to people in India about microfinance, I heard almost no one defend the behavior of the microfinance institutions (MFIs) in the villages and slums. Those actions appear indefensible. Loans were made too easily; interest rates were opaque, as they are in most of the world (though according to Chuck Waterfield, they are among the lowest in the world); collection practices were often aggressive; profits were disturbingly high. And that, I am now confident, is the core story. Whether the government response is ideal (when is it ever?), whether MFI leaders were cartoon tycoons acting on pure greed, the extent of the suicide link, the role of politics and the vested interested of the government-led self-help group (SHG) program---all those are more complex issues on which I think I am increasingly getting a grip.
The story behind the Ordinance
On Saturday, my last day there, I had an enlightening interview with B. Rajsekhar, the CEO of the Society for the Elimination of Rural Poverty. SERP, it turns out, was at the center of the Andhra Pradesh government's smackdown of the microcredit juggernaut in the form of that October 14 Ordinance. SERP was created a decade ago to implement the World Bank--financed Velugu program that provides finance and many other services to self-help groups in Andhra Pradesh. Through a hierarchy that mirrors the geographic divisions of the state (district, mandal, village) it bridges between the government and about 1 million SHGs with 10--15 members each. Formally it is non-governmental. But the state government funds it and the Chief Minister chairs its governing board.
Rajsekhar began with a soliloquy on intent. He pointed out that five years ago, before the MFIs had grown large, AP was already home to a huge number of self-help groups that help the poor save and borrow. If MFIs were truly committed to bringing financial services to those who lack, why didn't they go to states with fewer SHGs? Clearly, the MFIs are driven by greed. “Intent is not poverty alleviation. The intent is profit maximization on the MFI side." It was cheaper for MFIs to piggyback on SERP's years of hard labor organizing a million SHGs: the MFIs could just poach the SHG members, who were already screened for creditworthiness, organized into groups, and accustomed to credit. "It’s like SERP have cooked the food; it’s ready; MFIs can just come serve themselves and start eating."
He then told me the story of the Ordinance from his point of view. As early as the spring of 2010, local TV channels were broadcasting reports of microcredit-linked suicides. The coverage was sensational (whether sensationalized, I don't know): women spoke on camera of being pushed into prostitution. In response---and illustrating the key role of the media---SERP constituted task forces in each of AP's 23 districts to investigate such stories and file criminal charges where appropriate. The big MFIs resisted however, arguing that they were regulated under national, not state, law.
The media drumbeat intensified in July and August, focusing the minds of politicians and policymakers. One apparent reason was an unusually strong monsoon (think of the floods in Pakistan and, again, of global weirding). Under the impressive National Rural Employment Guarantee Act of 2005, the national government promises 100 days of paid work to every rural adult in the country. Most of the labor is unskilled and relates to public works, but it appears to have become a valued source of security for the poor. It pays weekly, which is convenient for people with loans requiring weekly payments. Sowmya Kidambi, a former activist now working for the program in AP, praised NREGA for getting cash into the countryside and preventing hunger this year despite higher global food prices. But it's hard to build earth works in a heavy monsoon. So some expected wages did not come. Another factor, which Rajsekhar did not mention but Sowmya did, is political unrest. A flyer into Hyderabad arrives at a huge and beautiful new airport, then rides into town on a divided highway through fallow pastures. On my way back to the airport, I traveled on the still-under-construction ring road. The flip side of all this investment catering to foreigners and the well-to-do, I was told, is the complete starvation of investment for the rest of the state, and that has stirred separatism. Activists want to split the state. Storefronts have been looted and buses set on fire. And political groups have called for bandh's---general strikes that shut down the region. Once more the poor get stomped on: wage laborers lose days of work. (See M. Rajshekhar's Economic Times article on both factors.)
Since intolerance for late payment is in the DNA of MFIs, pressure began to build on borrowers. Probably, in fact, the pressure had been building on many people for months and been held at bay with new borrowings. But that of course had to end.
Also in July, SKS went public. What was once cloaked was now spelled out with numerical precision in the papers: the investors and managers of SKS were making millions off the poor.
In August, the AP government formed the committee that would ultimately draft the Ordinance. It was led by Reddy Subramanyam, principal secretary of the state's rural development ministry, but staff at this non-governmental entity, SERP, did the bulk of the work. Rajsekhar described the committee as "racing against time." Despite the rush, they managed to study existing laws, consult their own legal department, and get input from legal luminaries.
Did they consult MFIs? "We didn't feel that we needed to discuss with them." When I asked why, he first cited lack of time. When I pointed out the inconsistency in this rationale, he referred to some bitter history. In fact, tensions have simmered between SERP and the MFIs for more than five years. In early 2006, they boiled over in the coastal Krishna district. An official seized the records and closed 57 offices of Share and Spandana, then the country's two largest MFIs. In a response to the attacks, and an apparent attempt to fend off regulation, the microcredit industry group Sa-Dhan announced a voluntary code of conduct for its members. Prabu Ghate recounts:
The RBI [Reserve Bank of India, the central bank] expressed its concern to the state government that the action it had taken could have wider repercussions by vitiating the MFI repayment culture in other parts of the state,... It set up a Co-ordination Forum to discuss issues of concern to stakeholders and resolve them as soon as possible. At a meeting of the forum held on April 20  it was claimed that the MFI movement was "eating into the SHG movement", MFI practices were "barbaric" and posed a serious law and order problem, and that even the lower interest rates suggested in the March 20 code of conduct of 21--24 per cent were usurious and illegal. Spandana and SHARE announced a reduction in their interest rates, including those on current loans outstanding, to 15 per cent on a declining balance basis. It was left to a respected MFI leader [must be Vijay Mahajan] to state that this rate was unacceptable to other Sa-Dhan members because it was not sustainable, and pointed out that the state government had no business to stipulate rates for [Non-Banking Financial Companies] regulated by RBI.
Having bared their fangs, the state government officials mostly retreated, effectively accepting the MFIs' promises to behave better and lower interest rates. (And, I am told, taking pay-offs from the MFIs.) So in Rajsekhar's view, the MFIs were given ample chance, five years, to behave themselves. And they blew it. Worse, some of them promised interest rates they could not deliver, making themselves liars. Meanwhile, the MFIs presumably resented the impossible demands for low interest rates and the state's fundamental misunderstanding of the cost-covering, commercial approach to microcredit. Thus by 2010, it seems that neither side respected the other. That's being polite: I suppose they held each other in contempt.
So in 2010 the government chose to ambush the MFIs with the Ordinance. The minister for rural development, of whom many spoke highly, briefed the Congress party's high command in Delhi, as well as the leadership in Hyderabad. A special state cabinet meeting was called on October 14, the first in history, Rajsekhar said, with one agenda item. The proposed Ordinance was approved in an hour. Since the legislature was not in session, the bill went next to the state's Governor (a sort of head of state, as it were) for interim approval on October 15, pending a vote in the legislature. The torpedo was launched.
I don't know that this little story has been aired before.
Critique of the Ordinance
The Ordinance required MFIs to immediately halt operations, to register, and to await processing of their registrations by an obviously hostile government before resuming. A court quickly granted MFIs the right to continue operating. However, through channels I do not understand, word went out to local officials to block all MFI employees from entering villages. The women I met in Yarvaguda, for example, were told two weeks ago to stop paying. This sub-legal directive is perhaps more important than the Ordinance, for it has frozen the MFIs in Andhra Pradesh and could bankrupt many of them.
The Ordinance has some good features: a requirement for clear interest rate disclosure; a "fast track" court system to resolve disputes; and a definition of coercion. Still, it "leaves a lot to be desired," according to N. Srinivasan, author of the 2010 microfinance State of the Sector report. I concur:
- Its preamble refers specifically to the need to "[protect] the interests of the SHGs" which "are being exploited by private Micro Finance Institutions." And the provisions only apply to MFI lending to SHGs and SHG members. SHGs, moreover, are defined as ones that have registered with SERP or its urban counterpart, the Mission for Elimination of Urban Poverty in Municipal Areas (MEUPMA) (hat tip C.S. Reddy of APMAS). But many SHGs predate these entities and have not registered with them. If women in unregistered SHGs are being driven to suicide by microcredit, that is apparently not a pressing concern for the government of Andhra Pradesh. If women belonging to no SHG at all are being driven to suicide by microcredit, that too is apparently not a pressing concern for the government of Andhra Pradesh. What then is the true intent? Occam's Razor says: protecting SERP's programs.
- The law includes several provisions that I expect would be unconstitutional in the U.S. and many other countries. For example, it makes it a crime to belong to more than one SHG. Imagine if the government of California jailed people for belonging to more than one book club. It makes it a crime for an MFI to lend to someone with two or more existing loans "irrespective of the source," which could include friends and family. Does the government not realize that 70% of households in rural AP already have three or more loans, most of those from informal sources? Meanwhile, MFIs must register with the governments of each district in which they operate, and those districts may revoke registrations at any time "after assigning sufficient reasons" even if they are only at the investigating stage. Imagine a California law empowering county governments to immediately shut down supermarkets for "sufficient reasons" without every defining that phrase. Imagine how, in India, this empowers local officials to demand bribes.
- The law imposes other burdens on MFIs without equivalent demands on SHGs. MFIs, for example, must obtain permission from the district government to lend to SHG members---but not vice versa.
- It contains other strange provisions. It makes it illegal for MFIs to accept collateral even though that would soften the pressure for full repayment. It limits interest to 100% of principal---which seems odd where rates tend to be in the 20s, 30s, and 40s. But as Srinivasan points out this rule could, as it were, make my mortgage illegal since I will pay more interest than principal on it over 30 years.
As Srinivasan says, "The objectives of the ordinance are laudable." But the execution reveals bias, and bias of a particular kind. The law does not merely view SHGs as better. It views MFIs as malevolent. It does not outlaw MFIs but, one wonders, perhaps only because it cannot. The presumption of guilt on the part of MFIs is clear in its draconian requirements.
My bottom line, for now
While the government responded to a real problem, there are real problems in the response. It was hurried because of the media drumbeat (Rajsekhar referred to the TV news several times) and the associated political drumbeat. I believe the government acted in no small part for the best of reasons. Still, India is home to a million small tragedies a day. This is a country where low-level officials routinely steal food coupons meant for the indigent. Solicitousness for the poor does not suffice to explain why microcredit literally came to dominate the government agenda. SERP's list of alleged suicide cases (symbolizing a link I find plausible in principle) were verified, Rajsekhar said, by “third parties”: district revenue agents and certain local non-governmental groups. But causality in suicide cases is delicate, and I don’t know enough about these third parties to trust their judgment in such a politicized context.
The main problem in the response, as Vijay Mahajan put it to me, is that the government is an unfair referee. It’s a player and a referee. While SERP is not technically part of the government, it is as a matter of political economy an extension of it. SERP felt its interests directly threatened by MFIs.
Clearly self-regulation for microfinance failed miserably in Andhra Pradesh. That calls for the government to step in. But this is this is not how regulation should be done. Regulations should be written and enforced by disinterested parties and published in draft, with a public comment period. I would tentatively suggest (not knowing enough about India to be sure) that this will contribute to the foreign perception of AP as no longer such a good place to do business: the government can slam the private sector at any time. The Ordinance comes off as assuming that MFIs are devils---companies that act out of pure greed rather than a mix of that with the pursuit of growth and genuine commitment to the poor---and assuming that SHGs and district officials to whom MFIs must now pay obeisance are angels. It’s parochial.
So, as is my wont as a child of divorce, I see some symmetry in the tragedy: the government people and the MFI people are both imperfect, acting out of a mix of motives. (Let's not forget that greed--the interest in making new opportunities for bribes--could be playing a role in the political economy on the government side too.) Neither side respects the other. One could say that the MFIs deserve what they got, maybe even needed it in order to force them to act more responsibly. They blew their chance at self-regulation while exuding disrespect for the government. So they got smacked.
But what’s important is not what’s fair to MFIs but what helps the poor most. The Ordinance may be better than nothing because it froze a situation that was spinning out of control for many; it may well have saved lives. But it is far from ideal. Despite Rajsekhar's assertions to the contrary, it is not realistic to expect the SHG system, for all the good it does, to meet all the financial needs of the poor. The private sector can help fill that gap. My hope is that this brutal game will ultimately lead the industry to a better equilibrium than before. But mostly it seems that the government wants to get rid of the MFIs, and is pursuing that goal quite efficiently. I believe the SERP is about as committed to serving the poor as is Vikram Akula, founder of SKS; but that does not mean that SERP, any more than he, is acting perfectly in the interests of the poor.
These events should also be cause for introspection at the World Bank, which has financed both sides, but especially the government and SERP (with $1 billion or so). The SERP-administered SHG program may well be doing much good. But World Bank money has also beefed up a political economy hostile to private sector solutions.
Still, the true bottom line is this: credit, the poor, and business-like insistence on regular repayment are a dangerous combination. Pushed too hard, credit can easily become a buzz saw. Change any one those three elements, and it is safer: savings instead of credit (cf. Gates Foundation), the well-off instead of the poor, the flexible and somewhat subsidized communality of SHGs instead of the hard-nosed efficiency of MFIs. If microcredit is to safely serve the poor, it must soften its edges. There are many ways to do that. But probably all are harder when growth is rapid. Fast growth in credit to the poor is therefore dangerous, and often unworthy of the label "development."