Kunal Mangal on Chasing Government Jobs in India

Shruti Rajagopalan and Kunal Mangal discuss the competition for government jobs in India and what it reveals about the labor market

This episode is the sixth and final installment of a series in which Shruti speaks 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, the second episode featured Rohit Ticku, the third episode featured Tanu Kumar, the fourth episode featured Proma Ray Chaudhury, and the fifth episode featured Vaidehi Tandel. In this episode, Shruti talks with Kunal Mangal about his research on the pursuit of government jobs in India, why the competition for these jobs is so fierce and its broader social and economic consequences. Mangal is a PhD candidate in public policy at Harvard University. His research focuses on labor markets in developing countries, and he is currently studying India’s competitive exam system for government jobs.

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 this is the last episode in our special young scholars series where I speak with young doctoral and postdoctoral candidates entering the academic job market about their newly minted research on Indian political economy.

The sixth and last scholar in our young scholars series is Kunal Mangal. He is a PhD candidate in public policy at Harvard University specializing in labor markets in developing countries. I spoke with Kunal about his paper titled “Chasing Government Jobs: How Aggregate Labor Supply Responds to Public Sector Hiring Policy in India.” For the paper and his other research, and for a full transcript of this conversation, click on the link in the show notes or visit mercatus.org/podcasts.

Hi, Kunal. Thank you for coming on the show.

KUNAL MANGAL: Thanks for having me.

RAJAGOPALAN: I grew up in India, and when I was in college, and after that in law school, there were just so many people around me preparing for the civil services exam or the state services exams to get a government job. In fact, it was so commonplace that preparing for these exams is almost considered a full-time job in itself. It would be perfectly acceptable for someone to say for two or three years that they are preparing for exams.

It’s obviously not at all atypical of India. It’s an environment where government jobs are so prized, one, to those who are looking for jobs, but also to everyone else in the family, and so on. And you’ve studied this, so, my question to you is, in this kind of environment, what is the impact of these kinds of government services exams on those who are looking for jobs, but also overall on the labor market?

Government Services Exams and the Labor Market

MANGAL: As you said, one of these things that we see in India a lot is people studying full time for these exams just because they are so competitive. You have something like a million applicants for a thousand positions. If you really want to have a chance, most people feel like that’s necessary.

The way in which the availability of government jobs affects unemployment is a bit tricky, but it’s true that if you’re applying for a government job full time, then according to official statistics, you’d be counted as unemployed. If you look at unemployment statistics in India, one thing you’ll see is that almost all of it is concentrated among young people between the ages of 20 and 30, and almost all of that is concentrated among college graduates. And that is exactly the constituency that’s applying for government jobs.

Not only that. If you look at it by, let’s say, land holdings in rural areas, then that unemployment is—even though it’s concentrated among people with a medium amount of land holding, which is again the constituency that we’ve seen is most active in trying to get government jobs—it’s not the extremely rich. It’s not the extremely poor. It’s the kind of people who feel like they’re middle class, and government jobs is like the base of their middle classness.

Are they unemployed because they are applying? Or are they applying because they’re unemployed? Another way of thinking about that is that if the government weren’t recruiting so much, if this feature of the economy wasn’t there, if these institutions around competitive exams weren’t there, would we see as much unemployment? Or is it the case that people are unemployed and applying because the labor market is just very slack in general?

But it’s hard to say, if we change the system entirely, what will happen to unemployment? Or if we just stop hiring in the government, what will happen to unemployment? There’s a lot of people who think that if the government were to hire fewer people, and if so many people are unemployed because they’re trying to get one of these government jobs, then you will naturally expect that the unemployment rate will go down as people drop out, as those jobs become less available.

So, to try to get a sense of whether this actually happens, one thing I do in my paper is, I look at what happened when the government announced a partial hiring freeze for about five years in the state of Tamil Nadu. Governments change their hiring level all the time in India. What makes this episode pretty unique is that it was announced in a very public way, so, that provides everyone with the same information, and it allows people to react in a coordinated way.

RAJAGOPALAN: And it wasn’t just a small freeze, right? It was a very large drop in hiring.

MANGAL: It was a pretty large drop. Relative to the number of administrative posts that the government is hiring, the drop was about 85 percent. Hiring basically slowed to a trickle for about five years.

For people who had dedicated their lives to exam preparation, that meant that they didn’t expect those efforts were going to pay out in the short term at all. So, if you were to believe the story that, when the value of the exam preparation goes down, and people start to do something else, that the government job is what’s creating unemployment in the first place, then this will be a natural setting in which we would expect to see that happen.

And that’s not what I find at all. Instead, what I find is that unemployment increased. This was a bit counterintuitive to me, but when I started digging in the data more, one thing I saw is that there’s a set of things that are consistent with people actually spending more time on exam preparation. So, both employment rates drop and unemployment increases. But then you also see, among the few recruitments that were conducted during this hiring freeze, a massive increase in the application rate. Relative to the number of applications that the government attracted before, during the hiring-freeze period, the application rate increased by some 20 times.

RAJAGOPALAN: Basically, it’s not really a question from the government side of are they hiring too many people or too few people. It’s more a question, as you point out, that these jobs are so valuable.

They are so prized compared to other jobs available for these graduates in the market that instead of dropping out of the labor force when it’s hard to get one of these jobs, they actually double down. They work really hard to get these jobs, and they actually exacerbate the unemployment problem that is going on in the economy. Is that an accurate summation of the mechanics of what’s going on there?

MANGAL: I think that’s exactly right. You can imagine a situation in which somebody—a typical person who’s applying for one of your jobs—they see the outside option is something like returning to the family farm or working in their family business. If they stop studying and start working there, then the value of that is more or less fixed in their mind. It’s not really going anywhere.

But even if getting one of these government jobs is really hard, and it gets harder because of the drop in vacancies, the fact that studying pays these large dividends means that it’s worth gambling on really small probabilities. If you believe that, and if your efforts keep paying out as you keep on working, then it’s worth continuing.

Which is to say that you can imagine a world in which, after you study a hundred hours, no matter how much you study, you’re never going to get any better, then the logic of dropping out starts to make more sense. But if it’s the case that no matter how much you keep on studying, you’re better, and not only that, the people who study the most gain even larger advantage over the people who don’t, if you have this kind of rat race effect, then you can see how the increased pressure from decreased vacancies ends up just exacerbating that rat race effect.

Public-Sector vs. Private-Sector Jobs

RAJAGOPALAN: Absolutely. What I understand from your research is that the problem is somewhere else. It’s not that there are too many government jobs or too much government hiring, but in fact, that there is too large a gap between the value of a government job and the value of jobs outside of government, whether it’s in the family business, a family farm or the private sector.

Some of this has to do with the fact that a government job, of course, comes with a higher wage. A lot of it has to do with the fact that it is a lot more secure in its tenure because agriculture is almost entirely in the unorganized sector. Most businesses in India are in the unorganized sector; 80 percent of India’s labor works in the unorganized sector.

Now let me flip the problem. If government hiring is having this huge impact in this kind of an environment, do you think this problem will go away if there are certain reforms that are addressed towards changing the private-sector labor market, maybe labor law reforms, creating contract labor policies, anything that increases security of tenure and brings businesses into the organized sector in India?

MANGAL: I think you’re right in saying that this effect is going to be driven by the relative differences between the public and private sectors. Anything that makes people feel more comfortable about taking a private-sector job or feel like government jobs aren’t as great a steal as they are now, then that’s going to tip the balance. It’s going to make someone who feels, “Should I study for a third or fourth year?” They’ll feel, “Ah, no, it’s okay to drop out.”

At the same time though, some of what I’ve seen in my fieldwork is that preparing for a government job is almost a form of voting “none of the above” on the private sector, which is to say that there’s a large number of people who went to some college, and they went through the campus placement regimen, and they got a job.

They found that environment so abusive and so unfulfilling, and they felt so stuck and undervalued and worthless, that they felt like the only way of making something of themselves and providing some sense of upward mobility for their family is to try to get a government job.

Some people ask me whether people who are applying just have these misplaced beliefs about what it’s like to work in the public or private sector. Maybe they’re just buying into certain narratives, but there’s at least a fair number of people I’ve met who tried it out. They went through the process to get a private-sector job, and they left really disillusioned. To that extent, I think something like contractualization, for example, probably would make people double down on government jobs even more.

On the other hand, I don’t think this phenomenon is entirely due to just the relative values. Going back to the logic I was describing before about how, if you were to double down, then that doubling down has to matter, meaning the extra effort you’re putting in should lead to extra return in terms of the likelihood of getting a government job.

If that’s not the case, then you can imagine this actually working out a little bit differently. For example, you can imagine that part of the reason people are studying full time and you have this intense rat race effects is because the way the system is set up is that you only benefit . . . It’s all about being in, let’s say, the top thousand or top several thousand out of a million applicants.

Basically what you want to do is somehow deconvexify the returns to effort here, which is to say that these exams provide value in sorting between different applicants, but it is hard to imagine that there are really sharp differences between a person in the fifth spot and in the sixth spot, or even the thousandth and thousand-tenth spot.

Suppose, rather than saying that “we are going to only select, let’s say, the top thousand in a particular exam,” what we could say is, “We need people with certain skills. If you get more than 160 on this exam, then you demonstrate that you have these skills.” And you can imagine randomizing or drawing chits among the people who pass that score threshold. What that would end up doing is, it would mean that this kind of perpetual rat race would stop to matter once you get beyond a certain score threshold.

RAJAGOPALAN: Basically, you’re recommending something like, say, a standardized test where the capital investment is not based on how much you cram or how many years you prepare, but it’s a more sensible metric of your abilities. Then, of course, that’s not going to sort the same way that is required for a thousand people being hired out of a million applications. So, at that level you use some other policy, right?

You could use a reservation policy, or you could use randomization. You could use a combination of the two, and so on and so forth. But that, according to you, will not have this crazy impact on unemployment and doubling down and those sorts of effects that are persisting in the economies.

MANGAL: Right. Whether you choose to get a government job depends on the value of preparing. And that value of preparing has two components: both the value of the job itself and your sense of your likelihood of getting that job, and how that depends on your effort.

That means the policy can adjust on both margins. You can either do things that change the relative value of a government or a private-sector job. Or you can do things that change how effort translates into your chance of getting selected. Right now, I think we haven’t really thought that much about that margin of how effort turns into success. And it seems like, under the current system, a lot of that effort is wasted.


MANGAL: Or there is a lot of effort put into margins that don’t seem to be really productive in terms of thinking about what makes a good public-sector employee. Getting a couple of extra questions, getting a 99 percent instead of a 98 percent probably doesn’t make you that much more effective, but that could be where all the effort is going if that’s the main margin that determines whether you get selected or not.

Reserving Jobs on the Basis of Historical Disadvantage

RAJAGOPALAN: And you also see this in the demands in the slightly different area in the market, which is for reservation of jobs and reservation policy. This effect that you just explained—that’s exactly what is driving even erstwhile, say, upper caste, landowning caste to now say they want reservations in jobs because agriculture becomes so unproductive, and it is just so darn hard for anyone other than the very elite in some sense, or the very, very well prepared, to get one of these government jobs.

So, you see a push, which is kind of proving your theory from a completely different direction, which is caste reservation.

MANGAL: You’re right. My question is, proving the theory in what way? Can you elaborate a little bit more?

RAJAGOPALAN: Well, technically, if the really big problem was backwardness or rather erstwhile backward caste who didn't have education and things like that, was the big barrier to entry in terms of cracking these services exams, and the original goal was, there needs to be reservation for those groups, right? Now, even groups that did not have that kind of historical disadvantage, who are landowning, whose parents are technically privileged, who did have access to schooling, who can crack these exams, are now not able to crack them because it’s literally a million-to-one or a 100,000-to-one ratio of getting in.

Now there are newer and newer groups who try and play disadvantage on different margins, so that they can get into the reserved category, so that it becomes slightly easier, on the margin, to get one of these jobs.

MANGAL: Right. I see what you’re saying.

RAJAGOPALAN: And Tamil Nadu already has some of the highest reservations in government.

MANGAL: If you look at the people who are applying for administrative posts, there is some sort of reservation category that applies to almost everybody who applies.

RAJAGOPALAN: So, in a sense people who don’t have the reservation benefit don’t even try because they know they literally have no chance of getting in.

MANGAL: Tamil Nadu is a little bit different compared to north India because the nonreserved groups are just a very small population of the total.


MANGAL: So much so that in the official categorization, there are all these different reserve groups. In Tamil Nadu, the rest are called “others.”

RAJAGOPALAN: Or even Andhra Pradesh and Telangana, where there is a huge push for reservation, but they’re not yet at the same levels as Tamil Nadu. And you do see that in Maharashtra, now, you have the Marathas who are asking for reservations. In Gujarat you have the Patidars. In Andhra Pradesh, you have the Kapus. They’re all land-owning caste.

MANGAL: So, the irony of a lot of these pushes is that there is an analysis that tends to show that you actually don’t get that many more seats when you’re reserved, just because you end up occupying a lot of the general seats to begin with. For example, the Modi government announced this economically weaker section reservation.

RAJAGOPALAN: The 103rd amendment. It covers 85 percent of the population.

MANGAL: It covers almost everybody. Almost everyone who would have gotten the job anyways is covered by it. So, there is one sense in which it seems like some of their protesting is either symbolic, or people don’t quite realize how little it would change their chances of getting a job.

I think part of what’s going on there is that this is one thing that people who are studying for jobs now tell me, that their parents don’t realize how much harder it’s gotten.

RAJAGOPALAN: Absolutely. The demographic is out of whack compared to when their parents were looking at the job market.

MANGAL: Exactly. When their parents were applying, college completion rates were probably half of what you have now, and bureaucratization rates haven’t kept up. It’s not the case that we’ve expanded our bureaucracy to meet the growing size of our population. What that means is that, as you were saying, this has just gotten that much more competitive.

But it doesn’t mean that there’s been a concomitant change. People have still held onto this idea of a permanent job as the archetype of proper employment, to the point that it varies across state, but if you go to Bihar, somebody has a job that doesn’t have permanent tenure, they call themselves berozagaar. They call themselves “unemployed.”


MANGAL: And I ask them, “You’ve been employed for the past 30 years. You’ve been doing work. You’ve been earning a wage and taking care of your family.” But they’re like, “Yeah, that’s not a real job.” Aaj hai, kal nahi.  You have it today. You don’t have it tomorrow.

RAJAGOPALAN: So, it’s clearly the uncertainty in the unorganized part of the economy which drives so much of this government job as an aspiration sort of thing.

Does Studying for the Government Services Exams Lead to Better Outcomes?

Now, I want to go in a slightly different direction. One would imagine that if there are so many young people who are investing so much of their time studying for exams, it must be making them smarter or building up their human capital in some way that has some long-term benefit. Even if they don’t end up getting the government job, maybe that actually leads to higher wages and earnings for themselves throughout their lifetime when they do get another job in the private sector or in the family business or something like that.

Do you think that plays out in India? Or do you think these exams are just so ridiculous because you are cramming details of the Salt Act in colonial India, or something else, that it doesn’t quite build human capital in a way that’s mobile and portable?

MANGAL: This is one of those theses that I found really plausible when I was first looking at this, thinking about this topic just because studying for these exams isn’t quite that different from many other sorts of school preparation that people do in their 10th or 12th standard or for college.

But what I ended up finding is that the same groups of people who spend more time preparing for the exams are also the ones who end up earning less about a decade after the hiring freeze that they set. So, the evidence—even though candidates often tell, and coaching center owners especially often tell me that education never goes to waste, that the skills that people are learning are going to serve them throughout their life—at least, in pure material, labor-market econ terms, I don’t find that to be the case.

RAJAGOPALAN: Do you think that’s happening because they are studying so hard for these exams—sometimes for two, three, five years at a time—that they are unemployed and not building the relevant skills in the job market, like they would have learned certain skills had they had a job? Are the lower wages linked to the fact that they were unemployed for so many years, studying for these exams?

Or is it something else? Is it that these candidates get disillusioned? As you said, they don’t like the idea of private-sector jobs, so they may not perform as well there. What is the mechanism that leads to lower earnings, even when they have spent so much time working on their human capital?

MANGAL: The size of the effects that I see—they are too large for them to be driven just by time out of the labor market per se. The returns to experience in India are something on the order of 3 percent for college graduates, at least from what we see from observational data. The data that I’m using don’t allow me to measure this super precisely, but I’m seeing changes in unemployment in average earnings on the order of about 4 percent.

So, the only way to really square that out is to think about some of these other channels that you were describing, like feeling kind of disillusioned or dejected, or not really occupying the same place in the household as you would otherwise.

And that’s at least what some of the ethnographic work to date has suggested, that part of what ends up happening is that people end up delaying marriage. They end up delaying family formation. What I see is that it’s not really the case that a decade out, the households don’t look worse off, but they’re contributing a smaller share to that household.

For those of us who have spent time in India, I think most of us have come across cases where there is somebody in the family who has fallen behind and who does odd jobs here and there or works here and there, but isn’t a full contributing member as much.

RAJAGOPALAN: Yes, your classic Hindu united, a Hindu undivided family underemployment that you see everywhere from agriculture to family business, across the board, right?

MANGAL: Exactly. The uncle who everyone is pressuring to do more, and he’s talking about how he’s still maintaining those dreams or is bitter about where he’s at. The data don’t really allow me to say that, but at least that would be my guess.

RAJAGOPALAN: But if that’s true, then what you’re describing is not just a labor market problem. We’re creating a huge social problem of people who are well educated, come from good families, college graduates who have this huge aspiration. For whatever reason, despite working incredibly hard, it didn’t quite pan out.

Then, it has this outsized effect on the rest of their life, on their contribution to their family, on who they marry, how many children they have, their mobility—both geographical mobility and upward social mobility. It has this huge social impact, much larger than what the numbers would suggest in unemployment.

MANGAL: I think that’s right. There are a number of candidate advocacy groups, people who advocate for people applying for government jobs, and there are different groups in different states. One thing that they’ve started to talk about is mental health issues in this community and the need to take care of that. I don’t think that these social issues are unique to this community. But it’s dangerous if a lot of people start to believe that their only hope of contributing to society is to make it past this cut-off. Otherwise, their contribution is nothing.

Current and Future Research

RAJAGOPALAN: Absolutely. What are some of the other projects you’re working on?

MANGAL: I’m pushing on two different fronts. One, I’m trying to learn more about this really opaque institution. One project is trying to look at, for example, gender gaps that we see in selection outcomes. One thing that puzzles me is that gender gaps in education have closed a lot in the past 50 years, and women tend to perform just as well as men in the secondary schooling.

Yet, we see the need for reservations for women across the board for government jobs in most states. And it seems to be the case that without those reservations, you wouldn’t actually have that many women. That’s just puzzling. Why would it be the case that a selection mechanism that depends on doing well on exams doesn’t work well in a setting where women often tend to outperform men in school?

As we’ve been discussing, a lot of success depends on making a lot of attempts. If you’re in a setting in which women are expected to get married earlier, and if their careers are considered pretty secondary, then what might end up happening is that there is less tolerance for failure in the family for women. They are less willing to continue to subsidize their studying if they fail out a couple of times. So, if men make more attempts than women, then you might end up in the situation where women—they study just as well. They’re just as capable, but you still need reservations in order to bring parity.

And I think what we learn from there is not just something about what is the role of female reservations, but also there are a lot of other institutions in which there is a certain gatekeeping role before you enter a profession. And there’s general consternation about why we see underrepresentation of women there. I think we learn something general about the way in which social expectations and institutions end up shaping women’s ability to get represented there.

MANGAL: There is a second thing that I’m pushing on, which is, part of what made this project possible and part of what are these big barriers to learning more about the labor market, especially in India, is that there is just not that much data available.

In the US, we have the CPS [Current Population Survey], and it’s collected every month. We get this nice monthly time series on almost everything that’s going on. In India, we usually collect data every five years, and now it’s gotten to the point where the government just waffles about whether or not to release it.

Yet at the same time, state governments, especially, are constantly experimenting with policy changes. Rajasthan has substantially changed contracting laws. There’s pushes to expand maternity leave. There’s all sorts of dramatic changes happening in the labor market, but we really can’t get a handle on what’s going on if we don’t have a way of comparing before and after.

So, what I’m doing with some coauthors is trying to get a sense of whether online vacancy data through these major—

RAJAGOPALAN: Jobs platforms.

MANGAL: —jobs platforms could be used to get a sense of what’s happening in the broader labor market. There has already been extensive work on this in settings like the US. There’s a firm called Burning Glass here based out of Boston that compiles this data for the whole country. And economists have used that really profitably.

But in a setting like India where informality is really large, it’s still a very open question about what these data exactly tell us. So, part of what we’re trying to do is combine online vacancy data with measures of what’s going on at the ground, and try to play them off each other a little bit and figure out what can we learn about the aggregate labor market from these kinds of statistics when we compile them together in a meaningful way.

As an empirical person, what I think about is, what are things that smart people have already talked about in theory? And yet, the conversation is stuck because we don’t really have a sense of where that conversation could go next.

RAJAGOPALAN: I think you’re asking all the right questions and looking at some of the big vacuums, which also have big policy implications.

Now I’m going to ask you our last question, which is also our most important question, which is, what have you been binge-watching during the COVID pandemic?

MANGAL: Couple of different things. I did recently discover the Avatar series. I’ve been really impressed by how it goes between both being a kid’s cartoon and talking about very, very serious things in a way that doesn’t dumb them down, things like responsibility, arrogance, feelings of betrayal or honor. I haven’t gone into even one season yet, but I’ve picked that up so far and I thought that it was really . . .

The other thing I’ve binge-watched was The Great British Baking Show.

RAJAGOPALAN: Oh, I obsessively watch The Great British Bake Off.

MANGAL: Yeah. But I’ve stopped watching it because I started to get pretty fat, because what happens is the baking on the show would inspire baking in real life. And then, the baking in real life would inspire . . .


RAJAGOPALAN: I cannot tell you how much I feel that right now because I was getting into a first-world loop, where I started buying more bananas than I needed. So, the bananas would go ripe, and then I would have to bake banana bread just so that I can deal with it.

Thank you for doing this, Kunal. This was such a pleasure.

MANGAL: Thank you, Shruti. You were very generous.

RAJAGOPALAN: Thanks for listening to Ideas of India. If you enjoy this podcast, please help us grow by sharing it with like-minded friends. You can connect with me on Twitter @srajagopalan. Thank you for listening to the six excellent researchers in our special young scholar series. We will be back to our regular podcast in two weeks on January 28, joined by Arvind Panagariya on free trade.

About Ideas of India

Host Shruti Rajagopalan examines the academic ideas that can propel India forward. Subscribe in your favorite podcast app