India Dispatch #2

November 18, 2010

I spent Tuesday in Delhi, attending the second half of the Microfinance India Summit, and flew down yesterday to Hyderabad, the capital of Andhra Pradesh, where the crisis is brewing.

Here, I will brood on two themes.

First, there is a political layer to the crisis that has mostly gone unreported in the Western press (including in the New York Times story). Amy Kazmin did get it in a Financial Times blog. (I had an interesting conversation with Amy on Monday about the difficulty she encounters and the delicacy she exercises in interviewing people who have just suffered misfortunes such as losing a loved one to suicide.) It is entirely possible that the political story is secondary. But at this point in my attempt to pierce the fog, I am not quite ready to conclude that.

In September 2009, the chief minister of Andhra Pradesh Y.S.R. Reddy—think prime minister of a state: more powerful than a U.S. governor because he sits atop the legislative and executive branches—died in a helicopter crash. In picking his successor, the top national power broker of the Congress party, Italian-born Sonia Gandhi, passed over his Reddy’s son. Jagan Reddy therefore has a beef with his usurper, Konijeti Rosaiah, and with the Gandhi family, including its scion Rahul Gandhi. As Vikram Akula, the founder of SKS, recounts in his new book, Rahul visited SKS in 2005, and was photographed next to Vikram. For those wanting to imply guilt by association with Vikram, there is also a picture of Sonia giving Vikram an award. Probably not by coincidence, Rahul just had photo ops with self-help groups in Uttar Pradesh, the politically favored alternative to MFIs (see below).

As far as I can tell, Jagan Reddy has kept a low profile in the press on microfinance. However, he owns part of the press. And according to Amy, “his newspapers and TV channels have whipped up public and political backlash against microfinance companies.” Meanwhile, openly pouncing on the issue since the beginning of October has been N. Chandrababu Naidu, who heads the opposition TDP party; is a former chief minister; and is the son-in-law of the party’s founder, who was also CM (sensing a pattern?). In the copy of India Today awaiting me in my Hyderabad hotel room, Naidu says that the microfinance institutions (MFIs) “are looting Andhra Pradesh.” “If the MFI agents harass you, tie them up, confine them to a room and call the TDP workers for support.” Consider: a leading politician in a state of 76 million people is calling for illegal violence against a small class of people; and, to my incomplete knowledge, no one of comparable stature has challenged him.

There’s some history here. When Naidu was CM, he oversaw the creation of a large program to start self-help groups (SHGs) and link them to banks. SHGs typically consist of 15 women, who first save for some months at a share bank account, then borrow from the bank even more than they have saved and divide the credit among themselves. SHGs are also platforms for other government services such as agricultural training, and put a strong cultural emphasis on solidarity, the idea that the members are there to help each other in times of need. The World Bank has pumped something like $1 billion into this program, initially called Velugu. In 2004, the Congress party, led by the helicopter-doomed Y.S.R. Reddy, swept Naidu out of power in part by promising to cut the interest rate on bank loans to SHGs from 12% to 3%. Congress kept its promise, in an unsavory way: SHGs still had to pay 12%, and would then be reimbursed for the 9% by government officials from a limited budget, which opened up all sorts of possibilities for petty tyranny, patronage, and corruption. In another sign of the SHGs’ political importance, Congress renamed Velugu after its original matriarch, Indira Gandhi.

Naidu is now trying to win back the support of women and “other backward castes” on the if-you-can’t-beat-them-join-them theory. Thus he says people should not repay MFIs until rates are cut to 3%.

But wait, there’s more. C.S. Reddy, the executive director of the SHG technical assistance group APMAS told me that Naidu promised in March 2009 to write off SHGs loans if he regained power. This, Reddy says, caused a sharp drop in repayments. Why repay loans that might be forgiven? In response, banks slowed disbursements of new loans to SHGs. MFIs dove into the breach with their millions of dollars in fresh equity. After the rumors of imminent forgiveness faded, SHGs had trouble rebuilding because MFIs had poached their best customers with much more convenient loans (no need for months of saving before you borrow). I’ve only heard this story from one (knowledgeable) source, so don’t take it as certain. But what is widely agreed is that those in government managing the SHG program were unhappy about widespread “poaching.” Several I talked to working on the SHG side believe that in switching credit providers, these women were acting against their own interests. The MFIs were undermining a decade of labor to build this politically important, flagship social program of the Andhra Pradesh government (and the World Bank). The poaching had to be stopped.

And of course the lucrative IPO of SKS microfinance in July galled many. It generated $11 million for Vikram (plus ~$50 million more in stock he can’t sell yet) and $117 million for Vinod Khosla.

Work on the crisis-spawning ordinance to regulate microfinance may have begun months ago. I hope to talk to a drafter on Saturday. But there seems little doubt that the current chief minister felt threatened by the attacks from the opposition (and perhaps by those from within his party) and that this is why the ordinance arrived with such suddenness and severity on October 14. (Government-issued ordinances can hold force of law for six months, pending parliamentary approval.)

What is less clear is whether these political and governmental machinations are the real cause of the crisis. Were they the firebomb or merely the spark that lit the tinder? Nachiket Mor, a member of CGD’s Task Force on Access to Financial Services has emphatically argued along with his deputy Bindu Ananth that the government’s capriciousness is the primary cause.

I am prepared to take this view seriously because I am mindful of how easy it is for a society, with coordination from the top, to turn against a minority in the worst way. There are many examples. And sometimes the minorities have been moneylenders—the Jews in Europe, the Chettiars in Burma. Lenders are easy to vilify. As Nachiket and Bindu emphasize, the situation in AP is unconscionably repressive. Village leaders are taking their cues from the center, forbidding MFI employees from entering and probably intimidating women who want to repay.

Ironically, another theory about the politics is that the real moneylenders are supporting the backlash. Think about it: who has the most to gain from the demise of MFIs? A lot of those village leaders probably are moneylenders, or are related to them.

Many have been tempted to see the Andhra crisis as a simple story of greed and comeuppance, bubble and bust. As you can see, a lot of the comeuppance is unsavory.

Nevertheless, it is possible the simple story is basically right. Which brings me to the second theme of this post: was it a bubble?

Today I met two groups of very thoughtful people at two respectable organizations with long experience in microfinance in Andhra Pradesh: BASIX, stands out among the big MFIs in doing more than pure microcredit; and APMAS, the aforementioned SHG advisory group. The BASIX people said there was no bubble. And they should know. The APMAS people said there was obviously a bubble. And they should know.

A couple nights ago at dinner I talked with Greg Chen, Daniel Rozas, Karuna Krishnaswamy, and others about this puzzle. I uncovered a few distinctions. First, there are asset bubbles and credit bubbles. Asset bubbles are straightforward: people buy something because its price has gone up—gold, tech stocks, etc. Credit bubbles are different: new loans make old ones look good. Of course the mother of all credit bubbles, in the U.S., was linked to houses, and so had the character of an asset bubble. Whereas microcredit bubbles are not generally linked to assets. Maybe there is a theory paper in there.

Second, and more practically, there is ability to pay and willingness to pay. Perhaps in Andhra Pradesh, and in other countries with crises I have called bubbles, people could repay most of their loans. Yet perhaps as the loan balances rose, they reached a tipping point beyond which it was in their interest not to repay. Access to future loans, secured by staying in good standing, became less valuable than the huge savings from defaulting. Now, this tipping point has no fixed location. A metaphorical guy with a megaphone can exhort everyone to default at once and shift the tipping point into lower-balance territory. The lender can’t single you out for punishment if everyone is doing it. Thus the tipping point is a function of political and social developments. What we can say, as Daniel pointed out, is that the higher the average loan balances, the more vulnerable the industry is to a sudden reversal. In this view, even in theory there is no clearcut line between bubbles and not-bubbles. Interestingly, the definition of bubbles of John Kenneth Galbraith (Kennedy’s ambassador to India) accommodates this fuzziness: “All crises have involved debt that, in one fashion or another, has become dangerously out of scale in relation to the underlying means of payment.”

What we really need to make some progress on this question is data on borrowings. Some numbers are available to shed a bit of light, but I will stop here for now.

If I were a betting man with a promise of perfect knowledge someday hence, I’d wager that the MFIs did overshoot and that the politics are secondary to the crash. That said, the politics helps explain the severity of crisis. On Monday, Sam Daley-Harris described it as a near-death experience. The U.S. financial crisis rightly did not destroy the mortgage as a product, nor the NASDAQ bubble the publicly traded share. But this one threatens to decimate an industry.

Maybe after talking to villagers tomorrow and examining that data, I’ll have firmer opinions.

Possibly Related Posts

• India Dispatch
• India’s Microcredit Monsoon

Tweet  

1. Nachiket Mor Says:
   November 18th, 2010 at 9:30 pm

   In order to answer the question of whether there is “over” lending or not, in my view, Andhra Pradesh is not the best place now — the waters are far too muddied. Look at any other state, district or village where microfinance has been operational and no diktats, ultimatums or fatwas have been issued by those in power. Collections continue at 99% plus week after week and the average amounts of loans continue to stay stable or fall even in some cases. In general, in the JLG version of microfinance in India, the outstanding amounts stay within the savings envelope of what the borrowers can save annually from existing income streams and therefore the leverage levels stay very-very low unlike mortagages where the debt-equity ratios sometimes go upto as much as 20 or even higher.

   There is also the report that Centre for Microfinance at IFMR has recently brought out that you refer to in an earlier post of yours and not this one (http://bit.ly/AccessAndhra) which points out that the reality is pretty complex and that the MFIs are only a very small part of that puzzle. The Economist Magazine, referring to that very same study comes to a conclusion (http://econ.st/901XRx) that is very similar to the one that Bindu and I reach in our IFMR Blog Post (www.bit.ly/AndhraCrisis).

   So if you were a betting man and did place a bet, my advice would be to not take on too much leverage while placing the bet.

2. Nikhilesh Sinha Says:
   November 19th, 2010 at 8:52 am

   I have been following this blog with interest and look forward the book in its final form. I’d like to make a couple of comments. Firstly muddied as it is AP remains an excellent petri dish, given the high level of penetration of both SHG model and the private MFI forms of microfinance. Discerning how much of the crisis has been perpetuated by political interests is a challenge, but it is certainly a persuasive argument. As far as the role of moneylenders is concerned, I’m not convinced. I worked with APMAS for a while doing a study on chronic indebtedness in rural AP, and I found that micro-credit had by no means supplanted the money-lender. Urgent requirements for sums in excess of Rs 10,000, for medical emergencies, or to to fund weddings and funerals for example continue to keep the moneylender in business

   As for whether there is a bubble, the fact that numerous studies and my own conversations with villagers indicate multiple simultaneous loans are the norm rather than the exception, would suggest that there is a case for charging MFIs and SHG-linked bank schemes with indiscriminate lending. Perhaps group liability is effective in ensuring repayment (even in cases that would constitute involuntary default), but there is not enough credible data on loan utilization. So we don’t know how much ‘productive investment’ is actually being carried out. Surely this is vital information. If loans are being utilized largely for consumption, then the crisis is less of a surprise.

   Moving forward, is additional screening necessary, and how would that impact interest rates?

3. Phil Says:
   November 19th, 2010 at 3:04 pm

   David, did you really compare the situation of the MFIs in Andhra Pradesh with that of the Jews in Europe?? I should think any concrete explanation of why the analogy is wholly inappropriate must be unnecessary.

   I would also urge you to check your “media frenzy” argument with regards to Indian broadsheets you have unjustly lumped with some AP media. The motivation of Sakshi (which sits in Banjara Hills very close to both SKS and BASIX and the AP government) in reporting the MFI suicides is a wholly different matter from the stories The Times, The Hindu, etc., have run on the topic.

   When you say … “What is less clear is whether these political and governmental machinations are the real cause of the crisis. Were they the firebomb or merely the spark that lit the tinder?” … I wonder, why even ask the question, if from there on your story is the simplest of conspiracy theories where patrimonial government arbitrarily strikes down the hero MFIs in order to serve its own selfish interests by appeasing its key constituents. The beauty of this simple story is mainly that it appeals to the rich-country prejudice that all poor-country politics must be inherently and hopelessly corrupt, and all politicians must just be goons in suits. But perhaps your post just has an unclear structure, because you end on a different note.

   You end with a discussion of ability to pay and willingness to pay. The capacity of the poor to repay is the asset underlying microfinance, so when that asset becomes overvalued (for instance through easy credit) and then devalued (for instance through overindebtedness), one can see how the market would come unhinged. But I think not ability and willingness, but rather *appropriateness* should be considered – i.e. has the point been reached where it would be financially inappropriate, or in their own financial interest, for microborrowers not to repay? (Of course consideration of that point assumes the MFIs are primarily on a poverty alleviation mission.) That point is clearly an economic one, but can be politically shifted.

   If the point has been passed where repayment will make a borrower poorer than if they never had taken the loan, then the AP government intervention would only have been the trigger for something that should have happened either way. In that sense, I wouldn’t waste my energy on searching for sinister motivations, but rather check whether the economics was at all sound.