This episode is the third 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. The first episode featured Vaishnavi Surendra, and the second featured Rohit Ticku. In this episode, Shruti speaks with Tanu Kumar about her research on housing subsidy programs in India and how these programs influence their beneficiaries’ political behavior. Kumar is a postdoctoral fellow at the Global Research Institute at William & Mary, specializing in urban politics and service delivery.
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 about their newly minted research on Indian political economy.
The third scholar in our young scholars series, Dr. Tanu Kumar is a postdoctoral fellow at William and Mary’s Global Research Institute. She is a political scientist studying urban politics and service delivery. I spoke with Tanu about her paper, “Home-Price Subsidies Increase Local-Level Political Participation in Urban India,” which has been conditionally accepted at the Journal of Politics. In her paper, Tanu studies the effects of a housing subsidy program in Mumbai through an original survey of winners and nonwinners of program lotteries, and its effect on local political participation.
We also discussed another one of Tanu’s papers, “The Human Capital Effects of Subsidized Government-Constructed Homes in Urban India,”on how this housing subsidy affects long-term investments in human capital.
For these papers and her other research visit https://tanukumar.com
Hi Tanu, welcome to the show.
TANU KUMAR: Hi, thank you so much for having me.
RAJAGOPALAN: I just want to start with the big, broad question. You find something really interesting in your study on subsidized housing transfers that are given in India. Normally, one imagines—at least what is reflected in the literature—that people who get these transfers engage less in democratic politics. Someone who got a subsidized house or subsidized toilet, or some other kind of transfer.
You find that people—when they get a subsidized house that they went through the lottery system—they engage more in politics. And it’s not a little bit more. It is substantially more than the people who didn’t get the transfer.
KUMAR: Just a broad overview of this paper is that the Indian government—and actually, governments everywhere—they invest a lot in making housing affordable and accessible to lower-income residents. So, I wanted to understand how these programs actually affect beneficiaries and shape their behavior and their decision-making.
Because these programs are such a large scale—maybe even 5 percent or more of the Indian population benefits from them—any effects on political behavior would have implications for the broader political landscape. What I find is really in line with what you just said—benefiting from a subsidized housing program in Mumbai makes people more politically active at the local level. They’re more likely to complain about local services, attend meetings about local public issues, and they also know more about local politics.
What’s particularly interesting is that they actually care more about local-level community issues like water, electricity, and sanitation. This is different from what we’ve seen in the past, where we find the people who benefit from different programs might participate less in politics. And the difference here is the outcomes that I focus on. I’m focusing more on, really, everyday politics, everyday making of complaints and stuff in cities to make services better as opposed to voting and turnout.
We might expect—given this literature on clientelism in India, where people argue that politicians are really trying to trade votes for cash or food or services—that these are things that might benefit the poor in particular. But making complaints about local services, no matter what income level I’m at, this is going to help me. It’s going to help everyone if the streets are better paved, if there’s less garbage in the roads.
Of course, if we’re interested in how winning subsidized housing affects behavior, it’s really difficult to establish causation. We can’t just survey beneficiaries of these programs and then compare them to nonbeneficiaries because the two groups are likely to differ in a number of ways, and it’s challenging to attribute any differences in behavior to just benefiting from the housing alone.
What’s helpful here is that, because a lot of these programs in India are oversubscribed, they’re implemented through a lottery system. So, among applicants, benefiting is random, and this was the case in Mumbai. You can think of it almost as a randomized control trial that you use to look at the effectiveness of a drug.
To learn about behavior, we have to survey both winners and nonwinners of lotteries. And again, this can create a lot of complications. I was serving these households three to five years after they won the lottery, so there’s a lot of worry that we wouldn’t be able to track them all down and survey them, and that we would lose the control group more than the treatment group because they didn’t own housing.
So this was a really big data collection process where I worked with a local NGO and people, neighbors, and community members over a period of nine months to really track them down, find them, and interview them.
RAJAGOPALAN: This is really interesting for a number of reasons. One, you actually talk to the people who are the beneficiaries and also the losers directly, and you’re able to estimate, in some sense, not the episodic nature of political surveys that we conduct, which is usually, “Did you vote? Who did you vote for? Do you identify with this party or that party?” But real political engagement.
KUMAR: Yeah, and one outcome that I was really surprised by was the knowledge outcome, where I asked people if they could provide me with the name of their local ward-level representative. When I was writing this question, I reflected that I didn’t know the name of my municipal ward-level representative, and control group knowledge is really quite low. I think about 19 percent of the control group could actually name their local ward-level representative.
But it was about 11 percentage points higher in the treatment group. This was pretty striking to me because it was actually a behavioral type of measure. It’s not just people reporting that they’re participating more, but we actually have evidence that they know more about local politics.
RAJAGOPALAN: What do you think is driving this? Is it because now people have succeeded once through winning the lottery for subsidized housing that it changes their perception of what is possible in terms of the interaction with the state? Is it that now the need for housing has been satisfied, they push their clientelist efforts towards getting other things?
Is it a locational thing? Now that the housing problem has been solved, they are geographically fixed, but they’re also fixed electorally. Now they know that they are constituents of a certain group of people, and maybe now they want to push more, given the geographical elements. Maybe some of these things wouldn’t transfer if it were a different kind of subsidy which wasn’t so geographically rooted. What do you think is driving this push for greater participation?
KUMAR: There could be many different things going on. It would probably vary across the whole population. But what I think is actually going on is two things. First of all, people have greater political capacity. They’re wealthier. They have more time.
I don’t really see more political participation across the board, but I actually see it targeted in a very specific way, like targeted around local, very community-level services. There is probably some element of having better expectations or changed expectations of what the government might provide, but it’s also action that’s very motivated by protecting the value of these homes, is what I argue.
RAJAGOPALAN: One of the things which is interesting is the way the Mumbai subsidy was designed. The beneficiaries of the subsidy can either use it themselves or they can rent it out and actually benefit as a larger income from winning the lottery. Do the people who rent it out and leave the housing location participate in the same way as those who don’t rent it out and remain?
KUMAR: Yeah, they do, actually. This gets back to your earlier question about whether this is something to do with location. There are two really interesting things that are going on. First of all, as you mentioned, not everyone has to move into this housing. About half of the people just rent it out. And I find that even people who rent out their housing report participating in local politics more in the places that they live. This builds upon the argument that they just have more political capacity, that they care more, and that they’re more basically habituated into participating in local politics.
But the other thing that I found that was really interesting is, not only are they participating more in politics where they live, but they also still care about the lottery apartments. A large percentage of these people who I call, basically, landlords—they’re traveling to their lottery apartments once a week. They live about an hour away on average, but they’re making this effort to travel there and participate in these local meetings and complaint-making where the lottery apartments are.
RAJAGOPALAN: And in a way, protect the investment, right?
RAJAGOPALAN: Typically, when it’s the case with randomized controlled trials or a lottery experiment, like what you find, the great thing about it is the design. You’re able to very correctly show us a snapshot of what’s happening.
But on the other hand, there’s the problem that these results are not easily generalizable to other contexts. What do you think are some of the broader learnings or some of the broader implications of what you find with the Mumbai lottery experiment?
KUMAR: The thing is, it’s really hard to shift political behavior. It’s really hard to change the way people engage in politics, to change the things they care about and the things they know about. And here we basically have an existing intervention that’s not devised artificially in a lab or an experiment, but a real policy that shifts behavior in a substantive and meaningful way. And this is a program that’s really common across all of India, and it’s implemented in a similar way in other cities across India. So we might expect the results to generalize to other cities.
We don’t yet know if the shift in behavior is a good or a bad thing. We could imagine that if beneficiaries care more about local services, that is great news for their community members, but we also might imagine that their caring more about street vendors or squatter settlements might hurt other members of society.
And thinking beyond housing, the research suggests that some of these other larger welfare policies in India might have similar effects. I’m thinking, in particular, about India’s National Rural Employment Guarantee Scheme. Here, too, is a program that makes people wealthier, and if they want to keep getting wealthier, they’ll have to get involved with local government and make sure that the program is implemented without corruption, that they’re working on projects that they care about.
And there’s qualitative evidence that MGNREGS is actually generating this type of local-level political participation, and my results support that evidence.
And finally, more broadly, in political science, we think of the main way of getting people involved in local politics, or politics in general, is to tax them. Once you tax them, they are going to demand more things. But my research actually suggests that one way of getting people to participate more in politics might be to give them things.
RAJAGOPALAN: Now, all else held equal, of course, it’s great to have more political participation and have people engaged, especially when it comes to provisioning of public goods, like you mentioned—roads and water and garbage being picked up and things like that. But on the other hand, the economist will typically ask the question, at what cost? And compared to what else? These are the standard questions we ask. Here I want to talk to you a little bit about some of the other literature on this kind of civic engagement.
One is, of course, Biju Rao’s work in his book, Oral Democracy [coauthored with Paromita Sanyal]. I had a chance to chat with him recently about this. He talks about how just the constitutional requirement of having the Gram Sabha, which was built into the 73rd amendment, dramatically increases this kind of civic engagement and deliberative democratic politics. You have people actually making demands of their representatives and the state.
On the other hand, in urban areas, you have someone like Adam Auerbach’s work. Again, I had a chance to chat with him recently on his book, Demanding Development, where he talks about how even in technically illegal or informal housing such as slums, you have this kind of massive demand coming bottom up from the people who are the residents of the slum to get things fixed. Now, those things, technically, are not a huge burden on the exchequer the way a huge housing subsidy program might be or a big labor subsidy program might be.
So now that you do know that this increases civic engagement, do you think this is the best way to do it? Or do you think they come at this huge cost to the exchequer and we need to think about other kinds of structural ways or designs to get people more engaged?
KUMAR: Yeah, that’s a good question. These programs are expensive in some way. They have a large opportunity cost to use this public land. They’re very politically appealing to people, so that’s one reason to do them. If you go through Mumbai or Delhi, you see these lottery apartments have the stamp of MHADA [Maharashtra Housing and Area Development Authority] or DDA [Delhi Development Authority] all over them. So it’s one way for the government to really signal to people that yes, we’re doing something, so they’re politically very attractive.
But I do think it’s really important in some way for citizens to have a stake in what their local government is doing. They have to care. It has to affect their well-being in the long term. One way for that to happen is what we see in the United States. When people buy housing on their own, when they buy stocks on their own, when they buy these assets and invest in things on their own, they’re just going to care about government more.
But if we really want to get lower-income people more involved in local government—people who might not be able to make these investments on their own—then I think it’s helpful to help them out.
RAJAGOPALAN: One of the trends we’ve seen in India—and this is most true for, let’s say, the Modi government starting with 2014—is there is a new trend of giving away private goods through the government program schemes like housing, LPG gas subsidy, and the Ujjwala scheme, as opposed to actually building public goods. But what you find is kind of interesting—housing tends to have this quasi-public and quasi-private characteristic.
It’s not exactly like giving an LPG, which is only used for private consumption. It does have certain benefits of maybe using public land better or gentrifying a particular area and having these kinds of externalities on public goods. But on the other hand, it is not exactly the same as using public money, public resources and infrastructure to deliver on water, to make sure there are sewage lines, garbage is picked up, there is no open defecation, and so on.
Which way do you think the externalities are going to fall? Are you going to get people who care more about these public goods because housing is a particularly weird kind of quasi subsidy—it’s public and private—or do you think it is going to be more rent-seeking behavior? “We got a house. Next, we’re going to ask for an LPG subsidy, and then next we’re going to ask for a college subsidy, and so on and so forth, and then we’re going to ask for a government job and reservations.”
Which way do you think this trend goes if we set off the clientelist behavior at lower-income levels with this kind of an intervention?
KUMAR: The thing about housing is that its value—unlike the value of a toilet or something—really is affected by whatever’s going on in the neighborhood. And actually, the people who I spoke to about their housing—they seemed to recognize that. When I asked them, what affects the value of your apartment, of your home, they all said the value of other apartments in the neighborhood and the value of other buildings in the neighborhood.
That’s a very specific thing about housing because we’re thinking of it as a longer-term investment that’s rooted in a specific place. And that might not be the case with other types of transfers.
One big and important question that’s still open in any study of politics is what shapes political behavior. And we have some ideas about what might shape partisanship or turn off behavior. But there’s this new literature out there by people like Gabrielle Kruks-Wisner and Jennifer Bussell that highlights that there are other forms of political behavior, like this complaint-making, like this participation in local politics and through public services. These are demands for better services and transparency, and we’re just learning about them.
I think that in addition to just voting behavior, this type of behavior actually affects the types of services that we see in cities. Understanding what affects service delivery in cities—it’s going to become more and more important as time goes on because these cities are growing. As more and more people arrive and live in cities, all these people will need housing, water, and electricity, and all of these other things. We need to better understand what leads to the allocation of these services in ways that are generally efficient and also beneficial for different groups of the population.
RAJAGOPALAN: Why is any of this research important? Why should people care about this?
KUMAR: There are three reasons that I think this is important. First of all, when we’re thinking of housing alone, this is a hugely important policy. The government spends so much money on these policies, not only in India, but in the United States as well. Affordable housing is a huge government initiative here, almost everyone who buys a home benefits from some type of mortgage subsidy. And we see these programs in Kenya, Ethiopia, Brazil, literally everywhere. So we need to understand how they shape behavior.
And then second of all, the fact that other welfare programs, like MGNREGS or maybe even cash transfers—the fact that they might affect behavior in a similar way is also an important thing to study, especially as people begin to pay more attention to things like universal basic income or these other larger, newer types of welfare policies. We need to know how they might shape behavior and affect politics down the road as well.
And then finally, this is related slightly to this paper, but another piece of work that connects to this paper, where I look not only at participation in politics, but households’ investment decisions.
RAJAGOPALAN: This is your paper on human capital investment, right?
KUMAR: Yeah, exactly. I find that benefiting from this policy leads households to invest much more in their children’s education. And getting poor households to really have the bandwidth time and energy to invest in these longer-term types of things is really difficult to do.
There’s been some research on how things like cash transfers might lead to these types of changes in behavior. But we don’t know very much about what existing policies that are already politically attractive, that we don’t have to persuade governments to do—we don’t know very much about what those existing policies are doing to affect this really important type of behavior.
RAJAGOPALAN: I want to ask you a little bit about the human capital investment. I found that absolutely fascinating. We already know that in India, there’s this huge premium on schooling and education. We know that even very poor households sometimes spend up to 50 percent of their income to send their kids to budget private schools, just to get a better education than whatever is lacking in the public school system.
We know this is a really important personal household consumption issue for the poor as well as a political issue for the poor. And what you find is quite remarkable—that an intervention or a welfare transfer in one area actually changes investment decisions, which are much more durable, in a completely different area. Can you tell me what you think is the mechanism? What is going on there that affects human capital and schooling choices for children in such a particular way?
KUMAR: Just to summarize, what I find is that beneficiaries of this program are sending their children to school more, or their children are attending school more. It’s both what their parents do, whether or not they succeed in school. And in particular, I find that beneficiaries’ children are much more likely to make it past 10th grade, make it past these big scary board exams, make it past 12th grade, and actually complete college.
And what I think is going on here is that, again, I know that when I win this lottery apartment, I know that my overall household wealth, my permanent income has increased by a lot. Even though these programs are interesting—I can’t resell within the time period of the study that I did. I can’t resell for 10 years.
So even though I don’t get access to the subsidy right away, I know that I have it in the future. And this peace of mind, this mental accounting allows me to invest more in the future. Speaking to beneficiaries in my fieldwork—they generally said that they felt less stressed, that they could think more about the future, and that they felt that their children’s futures were more secure.
RAJAGOPALAN: It’s taking away a certain kind of financial uncertainty or instability.
RAJAGOPALAN: So my question is, do you think it’s specific to giving housing? Or do you think a large wealth transfer would have done the trick? If you remember, Prime Minister Modi famously promised an additional 15 lakhs in every household as one of his electoral promises. Do you think that would have done the trick about the same way as housing?
Or once again, there was something about this kind of intervention – housing - which leaves people very financially vulnerable and also physically vulnerable in a big city, and solving that problem opens up capacity in virtually every other area of life, especially for long-term decisions.
KUMAR: Yeah, that’s interesting. I think a feature of housing is that it is somewhat illiquid. It’s different from getting a 15-lakh cash grant because I can’t spend it right away. I know that I have it, but I’m not able to spend it right away. So it’s not really clear to me what somebody, especially a poor, very constrained family might do if they got that kind of money. I would say it’s different because it’s illiquid.
The other reason that I think it’s different is that it’s much easier to give away housing for local governments. You have this public land that’s often earmarked for social purposes. I can’t use it for anything else.
Also, because of the way this program works, the actual construction costs are covered by the beneficiary because I do have to pay something upfront for this house, or I do have a mortgage. But the actual wealth transfer is arriving because the market value of these homes is much larger than the construction costs. But for local government, there are no real outlays in the same way that giving families a huge amount of money would be.
RAJAGOPALAN: This is the study that you conduct in erstwhile mill land, which is right in the middle of Mumbai, and it’s very, very highly valued and, in this particular instance, was socially earmarked.
A slightly different sort of direction—aside from the stuff that you were working on, what do you think are some of the big and most important questions in your field of research?
KUMAR: I think we still don’t know a lot about just how services and how public goods are delivered. There’s a lot of research out there about how MLAs might direct more resources to places that voted for them, or people might allocate resources to areas that have more higher caste members or something.
But what I’m really interested in or what we know really little bit about is how things like pressure, like demands from citizens affect this type of service delivery, but then also how lower-level bureaucrats make decisions and responses in response to these demands.
There’s a little bit of new research out on this. Aditya Dasgupta and Devesh Kapur wrote a paper on how, when bureaucrats don’t have a lot of funding, they face a lot of cognitive constraints, and it’s difficult for them to make these decisions. I think this individual-level political behavior answer and understanding what service allocation looks like as a result is something that we need a lot more research on.
RAJAGOPALAN: How did you get interested in this area of inquiry?
KUMAR: The real thing that I’m most interested in is what shapes behavior, and I’ve always been interested in how people interact with their government. I’ve noticed that in my family there are a lot of people who really care about what’s going on with local government or even national government. And then there’re some people who do not care at all. So I’ve been really interested in what shapes that type of behavior and what gets people excited or motivated to participate.
While I’m a political scientist, I do like to think of my work as really interdisciplinary. There’s a lot of intersection with development economics. I’m also getting more and more interested in psychology, especially as we’re learning more about behavioral economics and the psychology of poverty.
I think the other thing that sets my work apart is that I’m really interested in new and original data collection. Almost every project that I participate in does have some aspect of new data, something that we haven’t seen before. And I’d like to keep doing that as I go forward.
RAJAGOPALAN: And the kind of data collection you’re talking about is ethnographic data, right? You actually go and you talk to people, and you survey them or conduct informal interviews.
KUMAR: Two answers to that question. I’ve definitely surveyed data. I’m also becoming more interested in the type of administrative data that we can scrape from websites and stuff. But I do believe that it’s important to always spend some time with the people that you’re studying to really understand what’s going on, so fieldwork and ethnographic data collection will always be a part of that.
RAJAGOPALAN: Excellent. And the most important question during the pandemic is, what are you binge-watching to keep yourself sane?
KUMAR: I’ve actually begun watching a lot of new French shows. There’s one called French Village [Un Village Français], which is about a French village in World War II, and this has just opened the door of French TV to me, which is actually really fun.
RAJAGOPALAN: Well, that’s great. Are you fluent enough to just understand? Or do you watch it with subtitles?
KUMAR: No, I watch it with subtitles. Now, I think I speak French because I’ve been watching so many of these shows.
RAJAGOPALAN: [laughs] Thank you so much for doing this. This was such a pleasure.
KUMAR: Yeah, thank you so much.
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 Proma Ray Chaudhury, who is a PhD candidate at the School of Law and Government in Dublin City University. Her research is on female populist leaders in contemporary India.