Who’s the Culprit? Accessing Finance in Andhra Pradesh

The crackdown by the Andhra Pradesh state government on microfinance institutions was based on a number of assumptions about what’s happening with microfinance here in India.

[In case you haven’t been following what’s happened over the last month, check out a good overview of what’s happened by David Roodman here, and other analysis by N. Srinivasan, Vineet Rai, Beth Rhyne, and the Wall Street Journal. A sampling of articles in the Hyderabad press that helped precipitate the backlash is here:

Times of India September 29, 2010, Times of India October 18, 2010, Times of India October 26, 2010, Times of India October 28, 2010]

For those not familiar with Indian microfinance, an important distinction to make is that the controversy now is about one type of microfinance institution in India, which makes up about half of the microfinance market. These are what I’m calling “MFIs” here – institutions (both for-profit and not) that make loans to groups of five women in “joint-liability groups,” mostly following the original model of the Grameen Bank in Bangladesh. MFIs give loans directly to these groups of women, but are not allowed to accept deposits.

The other half of the microfinance market in India are “Self-Help Groups,” or SHGs. This model has been promoted by the Indian government, and involves larger groups of roughly 10-25 women, who save amongst themselves and distribute their pooled savings as credit amongst the group. Since the late 1980’s, these groups have been increasingly “linked” to commercial banks – meaning the bank gives a loan to the group, the group distributes credit to its members as needed (which is a lot more money than if they only lent out their pooled savings), and the group is responsible for repaying the bank loan, sometimes directly to the bank, sometimes through a separate “SHG promoting institution.”

The ordinance passed by the Andhra Pradesh state government was targeted at the first group of institutions, “MFIs,” and is at least partially (perhaps mostly) intended to stop them from competing with the second group of institutions, SHGs. It’s only the MFIs that are affected by the current crisis. Clients in Andhra Pradesh have essentially stopped repaying MFI loans, but SHG loans continue to be repaid for the time being. It’s worth noting that some of the loudest complaints about MFIs have come from the Andhra Pradesh state agency that oversees and promotes the SHG program. See here.

More than six months before these problems came up, the Centre for Micro Finance at IFMR Research, with funding from the Banker’s Institute for Rural Development at NABARD, conducted a household survey of 1,920 households in rural Andhra Pradesh to understand their access to and use of financial services. Led by Doug Johnson and Sushmita Meka, this was a representative survey of the state’s entire rural population, rich and poor, and collected detailed information on household savings and borrowing from SHGs, MFIs, banks, moneylenders, friends and family, and other sources. What we found is startling, and challenges many of the assumptions people have about microfinance in Andhra Pradesh.

Assumption 1: The poor in Andhra Pradesh are “over-indebted” because of microfinance institutions pushing loans on them that they don’t need.

Debt is indeed prevalent in rural Andhra Pradesh. An estimated 93% of all households have some sort of loan outstanding. But only an estimated 11% of rural Andhra households had a loan outstanding from an MFI. In contrast, 37% of rural households had a loan outstanding from commercial bank, 53% from an SHG, and a staggering 82% had a loan outstanding from an informal source, including friends, moneylenders, landlords, and others.

Overall indebtedness amounts are driven by informal loans and bank loans. For people with loans outstanding, median outstanding amounts are Rs. 35,000 ($778) for informal loans and Rs. 20,000 ($444) for banks, but only Rs. 8130 ($181) for MFIs and Rs. 4600 ($102) for SHGs.

Assumption 2: People frequently take loans from multiple MFIs, sometimes using one loan to pay off others.

“Multiple borrowing” is rampant. Eighty-four percent of households had at least two loans outstanding, and one household we spoke with had 19 loans. But the vast majority of these loans are informal. Of course, even if we ignore informal sources, multiple borrowing is still pretty common. But it’s not limited to people who lend from MFIs. Multiple borrowing from banks and SHGs is pretty common too.

• 17% of households with an SHG loan outstanding had multiple SHG loans outstanding, and 58% had at least one more loan from a formal source.
• 26% of households with a bank loan outstanding had multiple bank loans outstanding, and74% had at least one more loan from a formal source.
• 28% of households with an MFI loan outstanding had multiple MFI loans outstanding, and 82% had at least one more loan from a formal source.

We also find evidence that people who take multiple loans frequently take them out at the same time and for the same purpose, rather than staggering them as you would if you were using one to pay off the other. This bundling of several loans suggests that many people just find it difficult to get all of the credit they need from one place.

Of course, just asking people whether they use new loans to pay off old loans may be a simpler way to answer this. And of course they do. Twenty-five percent of MFI loans are used, at least in part, to pay off other debts. But so are 20% of SHG loans, and 15% of bank loans, though only 7% of informal loans.

Assumption 3: MFIs are specifically targeting existing Self-Help-Group members with loans.

Well…maybe. Seven percent of households have both an SHG loan and an MFI loan outstanding. And having an SHG loan makes a household 6% more likely to also get a loan from an MFI compared to households who don’t have SHG loans. Clearly there is some targeting of the same population, but not a lot.

These results call into question the common impression that rural households in Andhra Pradesh are saturated by private MFI lending. If anything, the survey shows that there may still be considerable room for expansion in the state given the high levels of borrowing from informal sources.

The point isn’t that Indian MFIs don’t need a greater degree of regulation – by their very success, Indian MFIs have achieved a level of importance that would warrant more careful and dedicated scrutiny by the appropriate regulatory authority (which is almost certainly not the state government). But ill-conceived measures based on knee-jerk reactions which don’t take into account the needs of microfinance customers will only end up hurting the poor in the end.