Vaishnavi Surendra on Moneylenders and Formal Banking

This episode is the first in a series in which Shruti will speak with doctoral candidates and postdoctoral scholars about their research as they enter the job market and the world of academia. In this episode, Shruti speaks with Vaishnavi Surendra about her research on credit markets in rural India. Surendra is a postdoctoral scholar at the University of California, Berkeley.

This transcript has been slightly edited for clarity.

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, a podcast where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan, and for the next few weeks I will be speaking to young doctoral and postdoctoral candidates entering the academic job market and the policy world about their newly minted research on Indian political economy.

The first scholar in our young scholar series is Dr. Vaishnavi Surendra. Vaishnavi is a postdoctoral scholar at University of California, Berkley. She is a development economist working in the area of household finance, and her research is focused on studying credit markets in rural India. I spoke with Vaishnavi about her paper, titled “The Moneylender as Middleman: Formal Credit Supply and Informal Loans in Rural India.” Vaishnavi demonstrates that informal moneylenders borrow from the formal banking system and lend to households, acting as intermediaries to ease lending capital constraints in rural India. For this paper and her other research, visit her website, https://www.vaishnavisurendra.com. For a full transcript of this conversation, click the link in the show notes or visit mercatus.org/podcasts.

Hi, Vaishnavi. Welcome to the show.

VAISHNAVI SURENDRA: Hi, Shruti. Thank you for having me. It’s great to be here.

RAJAGOPALAN: The standard narratives on moneylenders in India, and also in large parts of South Asia and Africa, is that they charge very high rates. They typically exploit these bad harvest or drought years or some other bad situation in a particular geographical area. And if only governments in developing countries ensured that there was proper public-sector rural bank penetration, this problem would just go away. And your paper completely upends this standard narrative. Can you tell us what is happening with moneylenders and formal banking in rural India?

SURENDRA: Thanks for that question, Shruti, and thanks for laying out the context. I’d like to first start with a little bit of the historical context in India, which is where my paper is situated. Going back to the 1950s, a household in rural India could only ever get a loan from an informal source. And often that meant, as you said, an exorbitantly high-interest-rate loan from a village moneylender.

Starting from then, policymakers were concerned about these high-interest-rate loans potentially forcing households into debt traps. And the Reserve Bank of India, starting right then, in its attempts to expand access to credit, had an explicit goal of reducing the prevalence of moneylending. In fact, a report in the 1950s explicitly states that their goal was to put the moneylender in his place.

But fast-forwarding to today—while, of course, formal lending is much more prevalent across India, and the RBI does mandate that banks lend particularly for agriculture to rural households, to smallholder farmers and to more disadvantaged households, to their priority-sector lending mandates—a lot of households just aren’t able to access these loans, so moneylending continues to thrive. And so today, essentially, these two sectors coexist.

In my paper, I’m trying to understand what the interaction between these two essentially is. First, I find from moneylenders themselves that they rely on bank loans for part of that lending capital. So this, in itself, suggests that there’s some kind of vertical relationship with moneylenders on lending bank credit and acting as some sort of middleman in this credit market.

To understand the mechanics of this better, what I do is I exploit an exogenous increase in household credit demand. What I mean is that when household demand for credit goes up, and they borrow more from moneylenders, moneylenders themselves are borrowing more from banks. And this increase itself—for households, I mean—is much larger when a district has high bank credit supply then when it has low bank credit supply.

So effectively, what this suggests is that moneylenders are relying on bank loans to ease their lending capital constraints. Interestingly, what this means is that far from competing, these two sectors are essentially collaborating with each other.

RAJAGOPALAN: Normally, one imagines—whether it is these Hindi movies from the 1950s or all the narrative that you read—that these moneylenders are in places where they have this complete monopoly and they’re exploitative. But what you tell us is that it’s quite a competitive business. There are lots and lots of small moneylenders. You don’t have this one, typical, rich zamindar-type person in the village.

But what I found even more interesting is that it is also not exploitative in the sense that it doesn’t happen too much in bad years. Households actually borrow more in good years or non-drought years, and one would presume that is to smooth out consumption of some sort. Moneylenders, in turn, because of the kind of vertical relationship you mentioned, also borrow from banks more in good years to lend it out.

So what is really going on here? Are moneylenders—now that we know it’s not just exploiting a bad-year situation—are they easing credit requirements per se? Are they easing the problems of intermediaries? Is it that they are actually catering to a group of people who wouldn’t otherwise be able to get access to banks? What is the mechanism that is making this whole thing play out?

SURENDRA: As you said, the business is not necessarily today defined by one zamindar in a village. It’s true that there are many more moneylenders today than you might have expected, given these tropes in the movies or in the media. So that’s definitely one change—that there are many more moneylenders. And outside of my paper, there’s evidence in the economic anthropology literature and other work that suggests that there are just more people who are able to lend now.

Sometimes, the way people enter these businesses—this is based on a case study in a paper by Madhura Swaminathan [coauthored with VK Ramachandra], where they find that the moneylenders themselves got this lump-sum large loan from a bank, and that’s how they started their business. Or they got a large dowry, and that’s how they started their business. And so that enables more people to enter.

But in terms of what people borrow for, again, drawing a little bit [here and here] on economic anthropologists such as Isabelle Guérin, this is also an environment where households’ aspirations are increasing. People just want to borrow more for various things. They borrow for, say, a wedding celebration. They borrow to purchase assets. Some of this is to maintain their social status. It’s not all borrowing just to finance that agriculture. And it’s not always smoothing consumption in a bad year.

But having said that, that does happen. So I do see that in bad years, households borrow more, but they also borrow substantially more when it’s a good year, following a good monsoon, as you suggested. And that—going back to your question—is probably because not all households can access loans from banks. And even when they do, they probably cannot access as much credit as they would like.

For instance, if a household has taken a crop loan, it might be due in a year, but other loans may have terms of three years and so on. And so, if you need to borrow for something else, you probably also need to take this other loan.

RAJAGOPALAN: One really interesting thing that came out of your research is, clearly, the moneylenders charge a higher rate of interest than formal banking. That’s very, very clear from both your research and most of the literature here. So there must be some reason that regular folks are going to moneylenders when there is a formal banking system available in their district. And you showed that in places where there is more banking penetration, there are actually more clusters of moneylenders, and so on and so forth.

Now, the obvious question is, why are they doing that? One possibility, as you mentioned, is that they may not qualify for a bank loan, or the size of the loan that they need is not being given by banks, which leads me to question, what is it that banks are missing out? What is that subjective knowledge about that particular district or about this consumer base that they are missing out that the moneylender is able to see? Because that is really strange. That should not happen.

SURENDRA: Yeah, that’s true. This is something that’s established, that has been talked about in earlier literature as well. Part of this advantage that a moneylender has is some kind of localized knowledge, and part of the disadvantage for more formal-sector lenders, particularly banks, is lack of this knowledge, which makes it harder for them to enforce repayment. So that’s the standard way of doing that. And it seems part of that is responsible for this.

In my work, I also survey moneylenders themselves. Based on what they say, the way they operate is essentially relying on social capital. They only lend to people that they know. Sometimes, they do not lend to new clients, but for those that do take on new clients, they only take on new clients if there’s a personal reference from someone they know, so this lending is still within some kind of extended network, and that allows them to enforce repayment.

That’s not to say that moneylenders are these benevolent entities because a lot of times, this enforcement is not done in the best means. Some moneylenders themselves rely on the social capital to enforce repayment. Depending on the relationship with this borrower, if the borrower is unable to repay, they may be lax about it. For certain other borrowers, they may be okay with a later repayment by charging higher interest rates.

But it also extends as far as to say that if there’s default, they’re going to really apply some kind of pressure to make sure people repay. And that could extend from this personal contact who helped the borrower get a loan in the first place. That’s going to be their first point of contact. They’ll pressurize that person. That’s kind of pushing the social capital buttons they have at hand.

They might go to the village panchayat. And they even report they would go to a borrower’s house and just possess any durables. They may not have, necessarily, like land documents or housing documents with them. But even if they don’t, they would just go and possess some kind of durable. And one lender even went so far as to tell us that they might just send goons after borrowers.

RAJAGOPALAN: Yeah. They use violence, right?

SURENDRA: If they’re willing to tell us that in a survey, you never know what they might actually be capable of.

RAJAGOPALAN: But it’s really interesting because even private banks, to recover car financing defaults and things like that in big metropolitan cities, they can’t formally rely on goons, but they do contract out and use goons to repossess cars, and so on and so forth. Given the legal system in India and how slow it is in contract enforcement, I’m not surprised that a lot of this happens in the shadows, even in the formal banking sector.

And of course, with moneylenders, there’s a fair bit of resorting to violence when there is a default. Now, in this particular case, you mentioned that they rely on networks. I assume these are typically caste networks and family networks because that is where you can put the most pressure and more stigma and so on.

My question is now from the moneylenders’ point of view. As you mentioned, they’re all quite small, relatively. They actually rely on bank capital to be able to lend out. They might be wealthy in a particular context, but they don’t have large amounts of funds at their disposal. In your paper, it’s also clear that while they cover marginal costs, a lot of them are not able to cover their average cost. That’s the other thing that is driving what’s going on with these moneylenders and keeping them relatively small.

So why don’t we see more moneylenders form, say, guilds or associations to have their own more certain supply of credit, given that there are so many of them? Is it a regulatory barrier? Is it informality? Because, say, medieval Scotland—you should actually see these guilds come out, these creditor guilds. And we just don’t see that in India.

SURENDRA: Yeah, that’s true. But I think part of this is also that moneylenders themselves are fairly heterogeneous. While it’s true that there are some lenders who do rely on bank capital, there are others who are quite wealthy and who just use profits from their business to make extended loans. Secondly, I think some of this heterogeneity also depends on where the lenders are. A rural moneylender is probably someone who has a smaller set of clients, and then they rely on these caste networks.

While I don’t talk necessarily about urban moneylending in my paper, some of the moneylenders I surveyed are in urban areas particularly. We also surveyed, say, around five lenders in Hyderabad, which is a fairly large city, and there, moneylenders are much larger. And there, perhaps, there’s some more implicit coordination like you’re suggesting. They don’t explicitly collude or coordinate, but they’re like, “We know what other lenders are charging.” And it’s similar across the board in these larger areas.

So I think there’s some heterogeneity. There may be some implicit common ground that they all reach. But in smaller areas, part of the reason that it’s a little more competitive is that there are these lenders outside in the urban areas. There’re also now more mobile lenders who just come to the village from outside and start lending, and so to maintain that competitive edge . . . It’s perhaps slightly different in the village context.

RAJAGOPALAN: How much are the various laws that were trying to abolish moneylenders and very high usury rates impacting this particular environment? I know you study a single state, which is Telangana, and a lot of these laws are at the state level. But say, in Andhra Pradesh, Telangana, what is the impact of something like that?

SURENDRA: You’re right. A lot of the money lending legislation is really old. In fact, the law in Telangana predates independence. It’s the same law that’s still in place in Telangana, and in fact, possibly in West Bengal and a few other states as well. These laws do regulate lenders and establish an interest-rate ceiling. Those interest-rate ceilings are never really observed.

These laws also require moneylenders to register themselves. Now, in Telangana, when we talked to moneylenders, they weren’t necessarily that forthcoming about all of their business practices. But looking at data on private moneylenders in a survey by the data on private money lenders in a survey by the National Sample Survey Organization, half of the lenders that they surveyed are just not registered. Some people do register. Half of them are just not registered.

So these laws, though they are in place, don’t seem to be very effective in any meaningful sense. They don’t successfully have moneylenders registered. That’s a far cry from regulating the actual interest rates they can charge.

RAJAGOPALAN: When it comes to moneylenders, I am reminded of this very interesting story I read in Dean Karlan and Jacob Appel’s book. This book is called More Than Good Intentions. As you know, they were part of the RCT revolution, and they talk about this flower seller Vijaya in the South of India. And they talk about how Vijaya wakes up every morning. She first goes to the moneylender; she borrows money. Then she goes and buys at the wholesale market. She sells the flowers. Every evening, the moneylender comes to collect the interest. It’s almost daily interest in some cases.

She has three or four loans outstanding at the same time. I don’t remember the exact details from the book, but one is for rent. One is to make sure she can keep her flower business going on a daily basis. There’s something for her children’s school fees, and so on.

They try to explain to her that she’s paying usually about 3 percent to 5 percent extra per month. And paying interest is becoming one of the largest expenses for the family. So they’re trying to teach her how, if she saved an additional dollar every week, then she would have to borrow $1 less each week, which means pay less interest. And in a certain number of weeks, she would be completely out of the clutches of the moneylender, and the entire capital outlay of $5 that she needs for her flower business—she can do it herself.

Her response staggered me. She said, “I can’t save anything in this house. Whatever I bring home, my husband drinks it.” It’s really interesting that Vijaya goes to the moneylender not so much because she’s not aware of other possibilities or even that she doesn’t know how to save herself. It’s a commitment device. She’s optimizing not just on credit; she’s optimizing also to make sure that she minimizes the husband’s alcoholism.

How much of this kind of behavior do you see happening with moneylenders in rural areas? How much of it is them resorting to a commitment device, making sure that the moneylender’s only going to give them a certain amount of money, and he will give it to them to buy durables, which the bank may not give them, or for a wedding, which a bank may not give them. But it’s still a very severe commitment device for any family, given that they’ll send goons after you or stigmatize you in the village.

SURENDRA: That’s a great point, too. One thing I do observe is that a lot of the borrowing in rural areas from these informal lenders is driven by households that don’t necessarily have savings. Now, that’s the extent of what I can say using my data. That does suggest that a lack of meaningful savings driving borrowing could be because they need a commitment device. It could be because of, say, a present bias behavior. It could be a variety of things, and what I find is consistent with all of that. So that’s very possible that that’s what drives this borrowing for some set of households.

RAJAGOPALAN: But it’s really odd because you also say that the recovery rates are very high. So it’s not like these people have a saving problem or a defaulting problem. If they’re able to pay back moneylenders with such high rates of interest for whatever it is they’re buying, one would imagine that they’re sophisticated enough to save it themselves.

So what gets in the way of household savings? Is it lack of access to formal credit? Do they not trust the banks? Are they worried that if they save the money, it would be spent on something else?

SURENDRA: I honestly don’t know. A lot of it is just speculation. The lack of a commitment device or just present biased preferences—those are just conjectures based on correlations that I see, that it’s people who don’t have savings who are borrowing. But I don’t think I can say anything much more meaningful.

RAJAGOPALAN: Now, you worked on one of the big important questions in the area of rural credit, right? What are some of the other major questions in this area, according to you?

SURENDRA: I think that relates to what you just mentioned as well. In my view, one thing that we need to better understand is what constitutes meaningful financial inclusion. For instance, in India right now, there’s this huge emphasis on access to finances being having a bank account, especially through this Jan Dhan Yojana.

Looking at data, say, from 2017, from the Global Findex database, you can see that this is fairly successful. Eighty percent of individuals they surveyed have a bank account, but only half of them use it at all. Half of them report not saving, not withdrawing, not making any deposits. So, while people have an indicator of doing so, they don’t really do so.

The second dimension to that is just credit. While these 80 percent of people have access to banks, clearly—they have a bank account—very few of them are using it, and far fewer are even borrowing from banks. I think understanding what really constitutes meaningful financial inclusion relates to what hinders saving. Why are households behaving like this? All of that is part of the puzzle of what it really means.

RAJAGOPALAN: That’s a great observation on your part. We seem to have two things running in parallel. We want to have this really formal banking system. We want to have interventions through this formal banking system that are intended to help poor rural beneficiaries and so on. But the poor rural beneficiaries are really not part of that pipeline in any meaningful way. They finally get that credit through an additional layer of intermediaries, which you point out as the local moneylender, in that same banking pipeline that the government has created.

It is useful to understand, what is the local context in which people actually save and borrow? I don’t think the Indian government or the policymakers have a very good handle on that, and your paper really highlights that element of how little we know as people in policy on what’s really going on with this behavior. So that’s really great.

What are the other projects you’re working on?

SURENDRA: In the paper we’ve just talked about, I look at this interaction between formal and informal credit and find that there’s this vertical relationship. In some sense, it’s intentionally entrenching moneylenders, since they’re successfully able to arbitrage across these sectors and enrich themselves, and that could, in some sense, contribute to increasing inequality.

But in other work with my coauthors and collaborators, Vivian Hoffmann, Biju Rao and Upamanyu Datta, we see that in a different context, the interaction between informal moneylending and a more formal credit option is quite different. In that work, we look at a self-help group-lending program in Bihar. This is the program called Ajeevika under the National Rural Livelihoods Mission.

We find that when villages get access to that program, this new, more formal credit option is able to crowd out moneylending. And that essentially hinges on the fact that Ajeevika targets particularly women from more disadvantaged households. So they’re essentially targeting women from Dalit and Adivasi households, make sure that they are included. And when these women are able to access credit, it has a more meaningful and a different impact on the informal moneylending market.

RAJAGOPALAN: What kind of impact does it have?

SURENDRA: What we see there is that with this program, households are able to access relatively cheaper credit. This is credit at, say, 2 percent a month, as opposed to 5 percent a month from a moneylender, so households are able to switch away from moneylending loans to these self-help group loans.

And overall, at the village level, the moneylender interest rates are significantly lower after that. That’s the crowding-out effect, and we find some evidence that there are also fewer moneylenders operating in that village. So this has a very different interaction than what we see with money lent by banks.

RAJAGOPALAN: But it’s still very consistent with your story, which is that some kind of intermediary is clearly required in Indian rural areas between the formal credit system and the individual.

SURENDRA: That’s true.

RAJAGOPALAN: So, either that vacuum is getting filled through caste networks and those moneylenders, or it is getting filled through NGOs, micro-credit programs, or Ajeevika, as you mentioned, which target particular groups. But they all are basically competing in the same space of intermediaries, and what they’re really competing on, in some sense, is local knowledge. That seems to be the missing magic of however the government-run, public-sector, rural banking system is working. There’s something that they are missing that intermediaries are required to fill the vacuum.

SURENDRA: Yeah, that's absolutely right.

RAJAGOPALAN: What are some of the big future projects, if you are in a position to share that with us?

SURENDRA: Another part of my research is looking at women’s employment in India. This is part of this larger observation by policymakers and researchers alike that female labor force participation in India has been declining since the mid-2000s. And it’s quite in contrast to India’s South Asian neighbors. India’s women’s labor force participation is now much lower than Pakistan and Bangladesh as well. And so there’s a lot that’s going on behind this. Understanding that is something I’m interested in as well.

RAJAGOPALAN: Yeah, that’s a great question. One of the things I learned, which surprised me about the female labor force participation rate, is that in India, that’s an inferior good in the sense that, as incomes increase, women are less likely to engage in the labor market, unless you’re talking about the top 1 percent or 2 percent of the elites who have a particular kind of exposure and information and education and so on.

But it’s really the poor women, who absolutely need to work, who work. And as they get richer, the preference within families is the women must raise the kids. And if there isn’t a need, they must not work. Is that an accurate picture of what’s going on in India?

SURENDRA: That’s definitely true. There are papers that do document that, as household incomes go up, women are dropping out of the labor force. And related to that is something that I observed in this context with Ajeevika, though not necessarily incomes. What I see is that when households have access to more credit, the impact on women’s labor force participation on the labor supply is quite different, depending on who these women are.

There’s some impact on women from more privileged households. That’s from more privileged caste groups where they are increasing, to some extent, participation in, say, self-employment.

But I see this large decrease in labor supply by both women and men from more disadvantaged households. It’s women and men from the Dalit and Adivasi households that are reducing labor supply when they can now access more credit. So that suggests that some of that labor supply is also a way for them to smooth income. And if they have a credit option, maybe that’s not something they need to do anymore. And that also suggests that maybe this is an inferior good, as you have mentioned.

RAJAGOPALAN: I think that’s a great line of research.

Now, Vaishnavi, to ask you the most important question during the COVID pandemic—what have you been binge-watching?

SURENDRA: [laughs] Well, that has really varied over the many months. It’s varied from binge-watching, say, The West Wing to Star Trek, and recently binged-watched The Expanse. Right now, looking for something new to watch, but it’s been a mix of political dramas and sci-fi.

RAJAGOPALAN: That sounds wonderful. Thank you so much for doing this. I look forward to reading all your research in the future, and good luck.

SURENDRA: Thank you so much, Shruti. It’s been a pleasure talking to you, and I appreciate the opportunity to talk about my work.

Thanks for listening to Ideas of India. If you enjoy this podcast, please help us grow by sharing with like-minded friends. You can connect with Shruti on Twitter @srajagopalan. Our next episode features the research of Rohit Ticku, a postdoctoral fellow at the Institute for the Study of Religion, Economics, and Society at Chapman University, about his research on religion, culture and identity from an economic point of view.