Amartya Lahiri on Fostering Economic Growth

Shruti Rajagopalan talks with Amartya Lahiri about monetary policy, regime uncertainty, inequality after liberalization and much more

In this episode, Shruti and Amartya Lahiri discuss the Indian economy in light of the country’s socialist past, distrust of markets, current monetary policy and continuing challenges. Lahiri is the Royal Bank Faculty Research Professor at the Vancouver School of Economics at the University of British Columbia. He has previously held positions at the University of California Los Angeles and the Federal Reserve Bank of New York. His current research interests include macro- and microeconomic developments in India over the past three decades.

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 today my guest is Amartya Lahiri, who is the Royal Bank Faculty Research Professor, Vancouver School of Economics at the University of British Columbia.

We spoke about the macroeconomic health of the Indian economy, the need to reform, demonetization, liberalization in India and its impact on poverty and inequality, Amartya’s intellectual influences and much more.

For a full transcript of this conversation, including helpful links of all the references, click the link in the show notes or visit

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

AMARTYA LAHIRI: Thank you for having me. It’s a pleasure, and it’s great meeting you for the first time. I’ve been following you for a long time. This has been an absolute pleasure to actually get to talk to you face-to-face.

Monetary and Fiscal Policy in Developing Countries

RAJAGOPALAN: You’ve argued—and this is in your paper with Urjit Patel—that large deficits, fiscal dominance by the state, especially when it’s combined with specific policies in India like the statutory liquidity ratio requirements, can render monetary policy quite ineffective.

Now, in the post-COVID world, given that India has undergone a severe economic contraction, and the prescription by most people— including yourself—is that India needs to announce a large fiscal spending package, how do we think about monetary policy, given that deficits are going to be large in the next decade or so? Does this mean that monetary policy is not going to be a very good tool in the policymaker’s tool kit?

LAHIRI: This issue has wider ramifications in India, but, in general, I think, in the developing world, there has been far more reticence in expanding fiscal support relative to the easy option, which is to lean on the central bank and say, “Do something, because it’s not coming out of my pocket directly.”

It just seems like an easy thing to do, and that’s why most developing country governments have tried to follow that strategy. They’ve relied a lot more on central banks to do the heavy lifting. The fiscal space is much more limited in developing countries, and that, in turn, reflects a fairly long history of living beyond their means. A lot of these countries have indeed had large fiscal deficits. India, in particular, for sure, has had big deficits for a long period of time.

As a result, the public debt in India is, I think, about 90% of GDP, so it’s pretty high, which limits how much you can potentially do. This whole thing that you raised earlier about this paper with Urjit Patel, where we were talking about various kinds of frictions that were introduced into the Indian monetary system were also ways in which the government was finding ways to acquire cheap access to credit. SLR, the statutory liquidity ratio, is exactly another one of those where—

RAJAGOPALAN: It’s the cheapest, actually.

LAHIRI: It’s the cheapest, saying, “Will you hold my debt, and I’ll give you some special breaks. Guess what? I’ll also make it compulsory for you to do this.” What that does is, it begins to commingle fiscal and monetary policy in a very, very organic way, whereas the approach, pretty much globally, has been to try and separate out those two arms. There’s a reason for wanting to separate the fiscal arm and the monetary arm.

If you think of a government as being an integrated government, that it’s collecting revenues and it’s spending on different spending goals, how does it matter whether we hand taxes to some notional subdivision of the government called the fiscal authority? Another source of revenue is that we are handing over to a monetary authority. Why not combine the two?

I think it has a lot to do with short-term versus long-term tradeoffs, and that’s one of the things about independence of the monetary authorities. The taxpayer is handing over part of his revenues directly to the fiscal guy, and part of his revenue he’s holding back and transferring to the government through this indirect route. The fiscal authority, I think everybody understands, is much more under the influence of direct political pressure of wanting to spend in order to maximize political gains beyond other standard economic arguments that you might have.

In order to minimize those things, you might want to put money into another agency of the government which is relatively siloed from these political pressures. And that’s the whole argument for independence because, without that, you might say, “Who cares?” You can have the central bank running as a separate entity within the finance ministry itself.

The flip side of it is, when you don’t have much fiscal space and the economy is basically on life support, if there’s ever a period or a time when you want spending to—and I’ll come to this issue of what my take is on why we need fiscal spending much more—but given the need right now to rely on a monetary authority to do the heavy lifting is potentially a problem, simply because the main instrument the monetary authority has is to make lending cheaper or more expensive.

Now, that depends on whether people want to borrow or not. People are choosing whether to borrow or not based on how far they think the return on this money is going to go, unless people say that it’s free—you don’t have to pay it back—which also has happened in the Indian banking system, where we have a mountain of nonperforming assets. You might think that, in technical terms, there’s a lot of adverse selection that has gone on, and moral hazard as well, that people have borrowed knowing they don’t have to pay back.

There is that aspect to it, but for the major part, when people borrow, they borrow because they’re going to put the money to work. They expect to earn a return from it, and part of that return is going to be used to repay the debt. If confidence is completely shaken, then nobody knows what the return is going to be on this money going forward. Making money cheaper is not going to induce people to borrow.

One of the things we’ve seen in India is that lending growth, credit growth has been correlating positively with the policy rate, the repo rate, which has been reduced. They’ve been cutting repo rates for three years actually now, and lending growth has been going down, so people aren’t borrowing even though rates are coming down.

What that also is showing is the mix of borrowers is worsening as well, which is why the spreads have stayed up. The guys who are willing to borrow are mostly the more risky guys.

Demand Contraction Due to the Pandemic

LAHIRI: The whole Indian story of growth since the 1990s has really been the structural transformation, which has been a key part of India, which is about people, employment and economic activity, moving from agriculture to non-agriculture. But within non-agriculture, it’s been moving a lot towards this service-sector-oriented stuff. That’s where a lot of the employment growth has happened.

The problem is, that is a segment that is basically serviced by a lot of lower-income urban workers, and the demand for it comes from high-end upper-middle-class urban consumers, and it relies a lot on face-to-face exchanges. Whether this is retail—all kinds of retail consumption—whether it’s going to malls, whether it’s shopping, whether it’s going to restaurants, whether it’s going to tourist destinations—all of that ecosystem is where a lot of this employment has grown, and that’s where it is centered.

At the same time, what has happened with this shock is, that demand for that stuff has just disappeared. That has completely collapsed. Now that spending is essentially discretionary spending of the upper-middle class and richer components of India.

Most of those are white-collar workers who can sit at home and do what they’re doing, like most of us are doing. They have been able to continue to have their income sources relatively less hammered than the other groups, and yet this discretionary avenue to spend has somehow disappeared. So what these guys have been doing is putting their money into stock markets, which is why they are—

RAJAGOPALAN: Which is right now booming.

LAHIRI: Booming. This is not just India. It’s booming everywhere, and a big part is exactly this. Saving rates of these groups has actually gone up, so they need spaces to spend.

The lower-middle class in the poorer sections of urban India, which are relying on this spending—that is their livelihood. That’s where they’re employed. These guys have nothing, and this is why they were heading back the moment everything got shut down. They have nothing to sustain themselves. They were heading back to their rural villages. This rural reverse migration that was happening around shutdown time is all related to this.

Until we have some sort of a fiscal spending which tries to take care of these guys, in particular, you’re going to see a very depressed economy. Two things have to happen. One is that the economy has to open up because without opening up, these face-to-face transactions are not going to happen. If they don’t happen, this segment of demand is not coming back. To do that, people have to feel safe enough to go and do the standard travel, hospitality, retail shopping, et cetera, that they used to do.

In order to do that, you need to have the sense that public health is under control. These workers—public health interventions are helping them in the sense that the vaccination drive is deep enough that most of these guys have been taken care of. All of this requires a lot of spending.

On top of that, just the pure social disruption of this massive chunk of urban India which is living essentially without any social insurance schemes to take care of them. This is where the government has to earn its keep. There are so many yojanas [schemes] we have, but we have none for the urban worker. We have to have some sort of an urban NREGA, without which I don’t know where these guys go.

Impact of Fiscal Dominance

RAJAGOPALAN: There’s a number of things I want to pick up on from here. The first is, when it comes to a government which has routinely had very, very large deficits—not just the current government, but historically, all governments—one of the big tools ends up being co-opting monetary policies such that rates are low enough for the government, and the government can, one, get cheap credit and, two, move a lot of its spending and debt off the books. That’s one part of what’s going on.

In the process of government control over the monetary policy framework and also the banks—because a lot of the big banks are nationalized—there’s also a cronyist element to this, which is the adverse selection that you talked about, which is the worst people in the business, who are least likely to pay things back and are very risky, end up getting big loans through politically connected means, and then, because of various incentives that nobody wants to get caught, they keep evergreening these loans.

The third part of that is that a lot of the private sector of credit operations are getting crowded out, so any kind of fixed-capital formation that takes place in an economy is now that much costlier when it’s private fixed-capital formation. If the government is having these huge deficits for its side of the capital formation, that’s great, but that doesn’t seem to be the case with the Indian government. It’s not really capital spending. It’s all very revenue-driven. They spend on fertilizer subsidies and things like that.

Then the last part of it is, the savers in the economy really get a terrible deal now because they’re essentially subsidizing the government.

Now, all these four things, put together, mean that there is very little cushion in the economy when things go wrong. Savers don’t have a huge incentive to save when the times are good, so when the times are bad, the cushions are very, very low—the incentive was to spend.

On the side of the governments, now it seems like my original question is almost moot because both fiscal policy and monetary policy have been rendered ineffective during the time of the pandemic. You have overspent so much in the past that you can’t announce a huge fiscal package, and given how much you overspent in the past and you intend to in the present, your monetary policy is going to be quite ineffective, given all the other institutions you’ve built around yourself. This seems very bleak.

LAHIRI: I like the way you slot it into a full-quadrant thing. I think that’s exactly right. The big issue here though is, is it just a chicken-and-egg kind of thing, or is there some way of breaking this vicious cycle? This is where I thought that the only hope right now is to somehow put in enough of a fiscal boost in three different areas. One is clearly a fiscal boost on just ramping up vaccine supplies and providing it for free instead of trying to nickel-and-dime the states, and so on.

The second part is just providing some sort of income support for the urban lower-middle classes. This is the group that actually has suffered the most. I’ve done some work on rural-urban disparities in India. One of the things you see is the gaps have actually shrunk a lot between rural and urban workers. The worst position to be in in India, actually, is being an urban lower-middle-class worker.

RAJAGOPALAN: Informal-sector workers, specifically.

LAHIRI: That has been the worst. There is virtually no job insurance. Food price spikes hit them the worst because their budgets are predominantly on food.

RAJAGOPALAN: And housing.

LAHIRI: And housing. They have no wealth to fall back on. These are the groups that have really been hammered in this last 15-20 years.

China dealt with it by just preventing people from moving or at least making it more expensive. We haven’t had that, so the price mechanism is completely screwed. It made it really bad for these informal urban workers.

That has to be a key component of spending because otherwise what’s going to happen is, these guys have nothing to fall back on. They need cash. You can give them a little bit of food, which is what the main intervention seems to have been, but they need cash to pay their rent. They need cash to buy things like milk.

There are these roles that the government plays. One is to create a legal and institutional framework within which the private sector operates. The other is to provide some social insurance, and the third is to intervene in public goods, which is public capital.

RAJAGOPALAN: Vaccinations and other such things.

LAHIRI: Exactly. At this point, it seems like two out of those three—the institutional structure they’ve inherited, so it’s not like anything’s been built, but the other two have just been rendered completely moribund.

In the lead-up to the previous national election, there seemed to have been a fair bit of welfare spending. There was some targeted spending over there, but I think a lot of that was funded on cheap oil prices and the excise. Now we are neither spending on welfare programs nor are we building public capital, which is like the worst of all worlds, as you rightly said.

The only way to spark demand right now is to actually create some cash transfers. Now, there is this bit of what’s going to spark demand for goods that are being sold by the urban informal sector through some part of the supply chain, which is being demanded by the higher-income white-collar guys.

A lot of that spending—this is why one of my suggestions was to cut the GST [goods and service tax] rate on consumer durables temporarily and make it seem like it’s a temporary cut. That might jump-start some spending in that sector. You think outside the box. Make it incentive-compatible for people to ramp up demand on things that the urban informal worker is somehow involved in, either in direct production or in the supply chain of it, which is going to spark this increase, but it requires some fiscal spending going on. It’s not going to come out of cutting repo rates anymore.

Returning to the Growth of the Early 1990s

RAJAGOPALAN: I want to go into the big picture for a minute. What is the way India should think about getting back on the old growth trajectory, the one that it enjoyed from, say, 1991 to almost 2014? That was a really nice run—high growth rates, a lot of confidence in the economy, though after 2010 it started faltering a little bit. But how do you get back to that part of the Indian economic story?

LAHIRI: The unsaid part of your question is the growth trajectory began slowing down, maybe 2010 to 2012-13. It slowed down, and then it actually did recover till about 2015-16. Since then, it’s been just a secular decline in growth rates. To answer your question, I think it’s useful for me to outline my view of what sparked the Indian pickup to start with, and then we can get into what needs to be done to get back on that path.

My take on what happened in India in the mid-1980s—I actually think Indian pickup and growth started in the mid-1980s, not necessarily the early ’90s. A lot of that was sequentially getting rid of frictions that prevented factors from moving to their highest returns, so that’s about labor. That’s about capital. We had set up an economy with a labyrinth of such restrictions that people and capital couldn’t really move to sectors or to areas where their particular own returns were the highest, and this is hugely problematic.

In a lot of modern economics, or the way of modern economic organization is about people making decisions under uncertainty. “I don’t know whether I’m truly supposed to be a baseball player. I don’t know whether I’m supposed to be an artist or I’m supposed to be an engineer.” We make decisions, some of which we are forced to make early. Others, we make decisions later on.

Same with capital—somebody has some investable funds, says, “Okay, I’m going to set up a textile firm, or I’m going to maybe set up a hotel or a restaurant.” Some work out, some don’t.

What has to happen at that point is that people need second chances. It’s not like I’m doomed by some decision I made at the age of 18 or at some point, and then I’m stuck with it. All healthy economies have churned, where people and resources are constantly re-optimizing and reorganizing themselves. That was the major problem with the Indian economy up until—the early ’90s is the very top of separation point, but as I said, in the mid-’80s, we had another round of that.

What it started doing was removing restrictions on the flow of factors and people. What ended up happening—if you just want to have a sort of caricature view of the Indian economy post-1990s—it was literally like we had lots of capital and labor scattered all across the economy, but mismatched.

A guy who was supposed to be an engineer was working as a banker. Same with capital. We just misallocated. Then the reforms allowed these guys to re-sort, so there was a huge amount of re-sorting that happened. Once you re-sort, you’re going to get capital to its optimal use.

You see a higher output, but it comes due to higher productivity, which was sparked by removing restrictions. That, of course, takes time. It’s not like we gradually did this and people started reorganizing themselves based on the new incentives that they saw. This gave us a 20-year run of gradually higher growth rates. Re-sorting—at some level, it’s like a one-shot game because you said, “Okay. If I could just do this all in one year, you would have seen the entire gains in one year, and that would have been the end of that.”

RAJAGOPALAN: The music stops at some point, and in our case, the music stopped around 2010 or so.

LAHIRI: Yes. At some level, one could see that as the end of pretty much what gains there were to be had from this reorganization. This is when we have been desperately crying out for a second generation of reforms. Now, when you think about what is needed—at some level, that doesn’t cost fiscal spending. It costs political capital because what is required is . . . The three big factors of production are capital, labor and land, and we have massive frictions in all three.

We’ve talked about the banking stuff a little bit, which is the major way in which capital gets invested in India—through banks—that’s the major intermediary, and we have massive frictions there in terms of the way we’ve set it up. Even in nonbank capital intermediation, there are all kinds of frictions about what can be done. The capital markets are limited in their ambit, and India’s stock market capitalization is, anyway, pretty low. We’re not really busting the doors on that.

Labor markets—everybody has been talking about it for a long time. This government responded, but the big part of the reforms is collecting 19 codes into 4. It’s not like getting rid of the codes.

RAJAGOPALAN: They streamlined it. They haven’t reformed it.

LAHIRI: It’s great packaging, but it’s not really any deep labor market reforms—some minor ones, about 300, 200. I’ll agree to that, but it’s nowhere close to what is needed.

Then the third one, which is the regulatory structure on the pending land—it’s a complete mess. The UPA government made a complete hash of it, got justly criticized for it, saying that these guys completely dropped the ball. But we’ve seen no movement on that either.

Those are the three main factors of production. If you have major frictions in the three main things that are needed to produce for any business or commercial entity, where is the growth going to come from? In some sense, it’s not a big surprise that we went from 3.5% growth. We, at some point, overshot, and then we were growing at 7% to 8%. But the fact that it’s going to settle—once all the transitional stuff is gone—at 5% sounds about right.

This is what we got as a long-run payoff to the changed regulatory environment but that’s where we’re going to go. At some point, having nothing to do, I was doing some baseline fiddling on if we keep growing at 5% as opposed to how long would it take. How long would it take for us to just catch up to Greece if Greece didn’t move at all? And that’s like 2060.

It’s ridiculous when you have these kinds of massive gaps. A 1 percentage point decrease in the growth rate has a huge effect on a massive chunk of humanity residing in a pretty small space of land. I think the human aspect of this is huge, of dropping the ball on the growth story which really needs these second-generation reforms.

Regime Uncertainty

RAJAGOPALAN: I agree that there was always an expiration date to the transitional part, and without a second and third stage of reforms, the party was going to end at some point. But I still feel like it might have ended a little bit prematurely because of some very poor decisions and a lot of regime uncertainty.

Now, this starts with some things going back more than a decade, like retroactive taxation on foreign entities, and this is all the UPA government’s doing. We have some massive regime uncertainty, of course, created under the Modi administration. Demonetization is a big one. The way GST was designed and executed, you don’t know what rates are going to apply to you. You don’t know if the system works. You don’t know how many filings you need to make.

It’s taking three, four, five years to sort it out, which is a fairly long period of time. When you think of a reform, it’s a very short period of time, but in the life cycle of a business which is trying to earn a profit and stay afloat, it’s a fairly long period of time. Most small businesses don’t have two to three years’ worth of working capital to get through hard times.

Now, of course, these new forms of regime uncertainty—are we going to vaccinate, are we not going to vaccinate? Are we paying for it, are we not paying for it? I think we don’t take regime uncertainty as seriously in India as we should, probably because we’re just so used to the government being so fickle in its decision-making that maybe, as Indians, we’re just very accustomed to this kind of treatment. But it has this enormous impact on how people decide to earn and spend their money.

Of course, it’s career choices. Like you already mentioned, it’s the choice whether you borrow and invest in something. Am I going to expand my little store or not? Am I going to expand my cottage industry or not? Am I going to switch to a different fertilizer if the subsidy is going to end? It’s just never-ending, the downstream consequences of regime uncertainty. I know it’s everywhere, but what do you think is the genuine cost of this kind of governance system we have in India?

LAHIRI: I completely agree with you. Uncertainty is huge. Everything I said earlier was, if you knew what was going on, it was going to run out anyway. Actually, there’s a fair bit of work on this, on quantifying the role of uncertainty in economic activity and how big. It’s known as uncertainty shocks.

In the macro literature in the West, people have been trying to measure these things and assess the effect of it on overall economic growth and business cycle fluctuations. Can this generate slowdowns of its own? The answer is, you can get significant slowdowns just through uncertainty shocks, so it’s no surprise.

And the kinds of shocks that tend to hit Western countries, where this work is going on, is of a far lower order of magnitude than the ones that you were flagging for India, where we don’t even know if we’re going to have some very basic things operating the same way in a year’s time relative to now.

Where does that come from, I guess, is a deeper question. Why is policy making so much more uncertain in countries like ours relative to the West? Maybe there is some political economy to it. I haven’t thought about it much till you brought it up in this context, but I think we do tend to lurch a lot politically between this desire for reform and a desire for welfarist schemes because people view life as a constant call. As policymakers, there’s a constant call, a constant choice or tradeoff between inequality and growth.

The fact that growth, by itself, might reduce inequality doesn’t seem to be a big part of that, so what ends up happening is this constant swaying, whether it’s somebody saying “suit-boot ki sarkar” and then you go into this thing of “I’m going to ramp up on welfare spending as much as I can in order to establish my credentials.” Then comes this other thing, that suddenly somebody says, “Oh no, no, no, we’ve gone too far,” and so now we need to spark.

That is essentially the epitome of all kinds of uncertainty because as a potential capital owner who’s thinking about investing money, you have a time horizon of 10 to 15 to 20 years. That’s what you’re looking at. Any business potential which is going to have any longevity—anything below 10 years seems like a rather short horizon, but it’s almost impossible to plan.

The second part of that is related to this desperate search for revenues, which is again, related to the lack of fiscal space, so this retrospective taxation . . . I still actually don’t know what the initiative was for or what the thinking was behind demonetization. I’m just completely flummoxed by it. I have no rational explanation for it, other than that it was somebody’s whim, because it just doesn’t line up.

The retroactive taxation had at least this—I can see that there was a desperate thirst for revenue. “Where can I find some funds? I see something here that I can grab.” So it’s a very short-term view. Capital is way more mobile than labor is. Somehow both policymakers and trade unions traditionally tend to forget that.

I can see the reason for some of these things that you talked about, but demonetization, I don’t have any. But the deeper, overarching backdrop to this is this policy yo-yoing between these two goals of inequality, but never viewing them as being jointly determined in some sense. That, I think, is a more fundamental problem, and maybe this is where more research-based policy making as opposed to just policy making based on whims and fancies might be. Or just polling people—do you have policymakers or politicians with a longer horizon or not?

Dictators and Centralized Economies

LAHIRI: Sometimes I get it, that the political economy constraint implies that people will be making more short-term goals based on what the next election cycle is. That, I think, is always going to be true, but I refuse to say, therefore, that the solution is that we need to get away from a much more democratic structure into a world where you can have some benevolent dictator because the probability of getting a benevolent dictator is so low based on all—

RAJAGOPALAN: There’s only one.


RAJAGOPALAN: When I give that lecture, I always put Lee Kuan Yew’s picture up. We’ve only found one so far. We’re searching for the next. It’s not been very fruitful.

LAHIRI: Exactly. You can say maybe Deng was another one. That may be the other example where, somehow, China got blessed for that 20-year run based on one man, who had a big thing, but to put your eggs into a basket which has such low supply of good eggs just seems like a crazy model.

RAJAGOPALAN: Again, I want to unpack a couple of things. First, I think there’s a further issue in the political economy of decision-making. I think one is the socialist past. Economic decisions are highly centralized the way we make them in India. It’s not the kind of independent institutions and the back and forth between these independent institutions which gives a lot of stability to policy making. It’s a lot of economic policy by diktat, which make sense in our centrally planned economy.

A centrally planned economy is economic policy by diktat, but we never switched out of that in some sense.


RAJAGOPALAN: On demonetization, though I didn’t agree with it, I thought that the original goal was a one-time windfall, where they expected that 30% to 40% of the notes will not come back, and it’ll be a big gain for the government to help it write down its debt with the reserve bank.

LAHIRI: I genuinely think that probably what happened was, somebody got sold on a bad bill of goods, and that bill of goods would have had a few components to it. Each of those components individually might have seen, “Oh, this is probably a good thing.

The thing was announced on November 8th, I think. Over the succeeding one week, I had lots of exchanges with people involved in policy making at the time who wanted an assessment for what the numbers would be. The thinking there was basically what you were saying, that 50% was not going to come back of the demonetized bills, notes. That’s the overall revenue gain and so on. That thinking clearly had—

RAJAGOPALAN: A lot of buy-in.

LAHIRI: Yes, had a lot of buy-in precisely because they thought, “Okay, this is cheap money, I can just get it and we can—” Without thinking through all the disruption. Even if that was true, just because there are gains potentially—and I’m not saying these are not gains—that there are no costs.

RAJAGOPALAN: You and I are early critics of demonetization because we both felt like you can’t count benefits without counting the cost.

LAHIRI: It’s crazy. Even if I buy into that argument, I just don’t see the logic for it because you have to then ask what you’re giving up in the process. On top of that, there’s this notion that somehow people who are . . . This comes, I think, from people having consumed too many movies from the ’50s and ’60s, where the bad guys are sitting with cash. It’s only the very low-level businessman who’s carrying around big quantities of cash.

Cash component gets switched into noncash assets very quickly, and that is the big thing of somehow figuring out a way of converting them—money laundering or whatever you want to call it. If you have big money, you’re not sitting on cash except if you are a political party, which is the other.

RAJAGOPALAN: And they were exempt.

LAHIRI: They were exempt, but of course, they rely on donations from various . . . Business keeps aside cash, a lot of it around electoral cycles. Pretty much all big businesses will keep it aside because they know the grim reapers will be knocking at the door for funds, as you can see the situation with this ex-minister, Deshmukh. He would say, “Why single me out? Everybody is doing this.”

That’s the second bit of what this bill of goods might have had, which is the electoral consideration and what this is going to do potentially to electoral fortunes of one party versus the other. That might have had something to do with it because there were elections around the corner. The third one was, I think, just this notion that somehow this is going to, politically, look very good because I’m going after—

RAJAGOPALAN: The rich folks.

LAHIRI: Yes, and that clearly did succeed because, despite all of the pain and everything else, there was the sense that people seemingly thought we are all in this fight together. This is why, I think, this government is very good at political economy calculations because, even if it fails on all the other margins, you can build personal heft if you are viewed as someone who is willing to suffer costs for the greater good, and potentially saying, “Look, this is a guy who realized he took this big risk to try and clean things up.”

That narrative—you can spin it because I can imagine the spin would be “Look, I know that people would be possibly very angry with me for all this cost, but this had to be done.” That kind of political thinking is somewhat rare.

RAJAGOPALAN: Yes, and also the asceticism—the idea that he’s not personally profiting from it, unlike most politicians. His children and grandchildren and sons and daughters-in-law are not amassing massive sums of wealth. He’s relatively frugal personally, and that really builds into the typical caricature of a politician that all of us deal with who are just—

LAHIRI: Just not there. It’s very different in that sense, and I think this also built this thing. It does look like, all the time, all decisions we make—there are some potential gains and there are costs. When there is a move that is made which potentially has huge cost, or it ended up having huge costs, you can imagine a narrative gaining ground saying the guy didn’t care about the cost that potentially he might face, but he’s so set on trying to clean things up.

Long-Term Effects of Demonetization

RAJAGOPALAN: In your paper, you talk about how the demonetization policy didn’t quite realize the intended benefits, whether it’s cleaning up black money, which it didn’t because most of the holdings are not in cash, and anyway it’s not a stock but a flow, and so on. There was never much counterfeiting, to begin with, in the first place, so that was like a red herring.

There’s also this drift, like mission drift. It starts out with black money. It ends up with digitization—this whole moving towards digital currencies. Also, there’s a spike. Quickly everyone switches to Paytm for a short period of time, and things like that. Then it just comes back to trend very, very quickly. The benefits were clearly not realized.

At the time when you wrote the paper, you said it’s still too early to talk meaningfully about the kind of costs—so there is a short-term shock, but the long-term costs of demonetization are not yet clear. Do you have a better sense now? Or is it now completely clouded by other things that have happened since, such as GST and the pandemic and the lockdown? Will we ever really know the true costs of demonetization?

LAHIRI: It’s an excellent question. I don’t know. It’s going to be extremely hard. The GST thing was already confusing things a little bit because, even absent COVID, it was going to be hard to disentangle the two things because both of them had this massive, disproportionate effect on the informal economy, which is more cash-driven and, by nature of being informal, was not part of this whole tax structure. So both shocks have hit those sectors.

The reason I was concerned—from the perspective of demonetization—as to what it might do in terms of long-run shocks was, I could see two effects. It probably was going to disrupt the supply chains a lot in the short run because it made a lot of these informal suppliers essentially go out of business because if they couldn’t receive cash payments, they couldn’t pay salaries, and the guys who were relying on making cash payments to them were just not making them.

That, I thought, potentially could lead to a big dying out of these informal firms, which just didn’t have the sustaining capacity to be able to live through a shock like that. That, in turn, would spill down the chain to everybody who worked, and because they are also in this business of outsourcing a bunch of the contracts, they get to completely mom-and-pop enterprises operating out of their one-room tenements, and so on. I thought that was a big negative thing.

On the other hand, you could potentially see that it might lead to an expansion of some sort of vertical integration, where you would see some of the bigger firms integrating by expanding their upstream production lines as opposed to just relying on sourcing. What that will do to productivity is an open question, so there are lots of uncertainties.

In a very pure lab experiment, there are stories of either way it might . . . A greater upstream vertical integration could generate certain gains because you have some contracting gains. You’re getting rid of holdup problems, et cetera, et cetera. On the other hand, you potentially are losing certain gains from specialization that you might get. That’s how the world has organized, clearly, through much more specialization and linkages. It’s not clear in which direction this thing might go.

But then on top of that, you throw in GST, and now with COVID again, it’s completely decimated that entire informal sector. It has to be somebody very clever who will be able to figure something out through looking at some natural experiment, where at least one of these was very limited in its impact, and the other one did show up, and compare across geographic regions, something like that.

See, the demonetization—at least you can find some variation across regions and, simply because of penetration of cash, by where was the shock. You can do a little bit. The GST is much harder. COVID—maybe there is, but you can imagine drawing a Venn diagram. What is the overlay between the GST shock, the COVID shock and the demonetization shock? It’s likely going to be finding spaces where only one hit, and the other two didn’t hit.

RAJAGOPALAN: But even there, given how integrated the economy is, it’s really tough. Let’s say people in typically big cash businesses—like the farmers who would take gunny sacks full of cash to go and buy a new tractor—that’s your typical cash economy we’re talking about. Because of demonetization, now there’s this big imposition on cash, and they have to go deposit everything, and it takes a while to sort it out.

The tractor company, which is completely formal and has nothing to do with this side of the economy, in some sense—and it’s probably in an industrial part of the country somewhere, maybe the Sriperumbudur corridor or something, where the cash penetration is pretty high—is still going to feel a hit.

I can’t think of an easy geographical diff-in-diff when it comes to a problem like this, which is the economy being integrated across so many linkages. If you really take the economy seriously, not as a monolith but like a huge effort of coordination and cooperation, I think it becomes very difficult to separate this stuff.

I’m very distressed about this because I wanted some nail-in-the-coffin evidence to show that demonetization was a terrible idea, and any government in the future should tread carefully and never get into this. And now, it’s just been completely overshadowed by other blunders.

LAHIRI: It’s exactly like we are moving from one catastrophe to the next, which is dimming the memory of the previous bad shock. Even in one’s mind now, demonetization seems like, “Oh, whatever.” It was so far back.

Goods and Services Tax

LAHIRI: The same with GST. This GST thing was so massive in terms of its effects, and it was so poorly designed. That by itself would cut the legs off any economy that is relying 60% on the informal sector for sustenance. I have an example of a friend of mine. He was part of the Indian cricket management hierarchy. Because he’s getting paid—he’s a service consultant—whatever he does, he had to file 12 monthly returns, four quarterly returns and one annual return.

I have to do the same thing because I teach in India. I get paid maybe twice, and then imagine, I’m sitting over here, trying to figure out an ecosystem where it works relatively seamlessly, after taking six months to figure out that ecosystem. Imagine some person working, having this kind of outlet in Dadri or in even farther, more remote parts of India. The internet connections are so spotty, the system’s down half the time.

RAJAGOPALAN: The small firm in Dadri has two problems. One, it’s not easy to file, but until you file, you also can’t get your input credit back, which means double whammy. Now all your working capital is completely tied up with the state.

LAHIRI: Absolutely, absolutely. Then, on top of that, the average rate that was revised with the state was going to be in a revenue deficit, just by construction. Now you’ve got this odd situation where you were going to have an 18% average rate, when the previous system was averaging about 27%, 28%? I could’ve created three, four layers. But then you’ve got an average rate which is basically structured, centered at a revenue deficit point.

Now you’ve got, on top of that, the states have handed off all of their revenue independence. Now the center has come up with this. They have to fill up the coffers. They have these deficits, so the cesses they don’t have to share. It’s like a mad mess of frictions up and down. The states are not getting what they’re due.

RAJAGOPALAN: Yes, they were in constitutional default for three months, not paying the states. I think this, again, goes back to your previous point that there seems to be a fundamental misunderstanding about the nature of the economy—this whole growth versus inequality, or what does economic growth really do?

Now, at the current rate of growth in the economy, an 18% or a 15% rate is most likely going to be a revenue deficit. But the entire point of the GST was that if we can move to a single rate which is incentive-compatible, the size of the market is going to get bigger. There is going to be greater economic activity, and given three to four years after the adjustment takes place, actually, you’re not just going to be revenue-neutral. You’re going to increase revenues.

I recently wrote a paper; I think it’s going to come out in Social Philosophy and Policy. Originally, the GST tradeoff was, we can’t have an 18% rate because the person who buys the BMW and the person who buys the T-shirt or the Parle-G biscuit is going to have the same rates, and that’s completely unfair. You know that’s regressive.

We want to avoid regressive consumption taxes. We’re going to have this rate proliferation, so we ended up having five rates that now we have seven non-zero rates, twelve if you include cesses, and so on. But the part that the government forgot about is that the more rates you have, the more regressive it is on the production side, given a certain amount of compliance costs—the way it is, let’s say, for the small tenement shop in a slum in Dharavi.

Suddenly, these guys need to get chartered accountants. They need to have a firm. They need to get someone who is internet savvy. The cost of compliance is so high for the smaller firm that now it’s regressive on the production side, and that completely devastates the informal economy. Again, there seems to be a very big misunderstanding.

By the way, we are still at a point where Parle-G is at 18% GST tax, whereas Tanishq at below 1%. We haven’t even got the regressive inequality part of it right.

LAHIRI: Figured out. Yes, correct. Absolutely.

RAJAGOPALAN: We’ve messed it up on every single margin, but I think it still comes from this fundamental misunderstanding of exactly how the market works, how incentives work, and how certain types of costs which are imposed on certain economic activity could have much bigger downstream effects, than just having something which is much simpler, which may not seem that fair, but in the end, it’s going to increase the level for everyone.

Suspicion of Markets

LAHIRI: Yes. I think you’re absolutely right. This connects with this peeve of mine that I’ve had a long time, that this suspicion of markets in India is so endemic amongst policymakers. This is why I was alluding to people having watched too many bad movies from the ’50s and ’60s, where that was the basic narrative, that the capital—whether it was the shopkeeper or the capital owner or the industrialists—were all exploitative human beings who valued exploitation above all.

The fact that there can be a confluence of interests between capital owners and employers and employees, depending on how the joint output is going—that doesn’t seem to have figured. This, by the way, is true not just amongst policymakers or amongst politicians, but also amongst intelligentsia in India. Very left-leaning. Incredibly.

This whole cash versus in-kind transfers, whether you want to make it through the PDS system or whether you want to make it through Aadhaar or cash transfers, you have got an incredibly strong group. Of course, there will be examples you can find where those cash transfers were exploited in Jharkhand, or didn’t quite work as well—

RAJAGOPALAN: So were in-kind, frankly.

LAHIRI: Yes, it has to be, on the balance, which method is proving to be less leaky than the others.

RAJAGOPALAN: But it’s worse than that. It’s not like they check which side is more leaky. It is more that people are willing to live with an enormous amount of error when it comes to the state, and there should be no error at all when it comes to the market. Right?

LAHIRI: Yes, that’s right.

RAJAGOPALAN: The market better be a zero-error bet. Otherwise, we’re going to clamp down on it instantly.

LAHIRI: Yes. Market is mala fide unless proven otherwise. It’s the basic way in which people think, and that shows up in public policy all the time, in various things that people do in terms of the regulatory structure, in terms of taxation schemes. It just shows up, whether it’s COVID or policies of purchasing vaccines. It’s the same thing. Markets are just not your friend.

Now, it might well be true in certain cases with externalities . . . There are many, many examples where unfettered market is not what one is talking about. But starting with this prior that the market is your enemy is a very defeatist thing, where most Indians are essentially involved—extract their lifestyle through the market. It’s a strange thing of cutting off your own legs in an economy.

For some reason, this narrative has gained so much ground. This is, once again, when you talk to businessmen in India—small businessmen, not the bigger guys. They’re all so incredibly suspicious of, whether it’s public-sector banks versus private-sector banks. They are, “No, I’d rather put my money in the public-sector bank. I don’t trust the private guys,” even though they themselves are working—it’s like a dissonant space, almost like people not trusting themselves.

RAJAGOPALAN: Is it because during socialism and the mixed economy, the only way to operate in the market was by being very corrupt, and giving bribes and so on, and breaking a hundred rules, that now, somewhere at the back of your mind, that is the version that we have?

Because if you start from an enlightenment set of ideas, you think the market’s wonderful. If you grew up in the United States, you don’t have this horrible experience in the marketplace. You don’t have shortages. You don’t have clever capitalists trying to bootleg things stolen from ration shops and trying to capitalize on people’s misery. But under socialism, everything gets corrupted. Even the market gets distorted, and really, the worst get on top. Do you think it’s that kind of a hangover? It’s so bizarre where this comes from.

LAHIRI: It could be. I think it has something to do with the only guys who could succeed in the market economy were those who were willing to . . . Bribing at some level was all about public sector. You’re trying to bribe some person who is involved in the public-sector space is what drove most of that, including me trying to get a telephone line. You would have to bribe the linemen to get what you were owed to start with. A gas connection—again, you would have to pay to just hold your line in the queue.

What does that have to do with the market? It’s just got to do with the fact that the rules, the regulatory structure . . . Now, whether people have conflated that with the market being corrupt is a meta way of thinking about markets, saying that the whole corruption world was the market trying to find the right price, and therefore, I think the market is corrupt. We were sold on this notion that somehow, big industry in particular was all in the business of exploiting. The only ones who seem to have escaped these images are the Tatas, but other than that—

RAJAGOPALAN: That’s because they predate the Republic, I think.

LAHIRI: Yes, maybe. Maybe they didn’t grow during this period. But this is an endemic view up and down the system, that big industry is completely exploitative, and that is their main goal, to exploit.

Even though the fact that the biggest escape out of poverty has happened since 1991, in the history of modern India, and the biggest regression into poverty has happened through the growth slowdown post-COVID, should make clear that, without growth, you’re not going to have any escape from any of this. Growth is not going to come out of the state leading you to growth.

That’s the other part of what is sad about India—the way it operates is, it seems that everybody’s looking to the state to solve problems. Again, it’s 70 years of being fed this kind of line, this “I will take care of you.” The state is this mai baap. This is what we’re going to do, and the entire attention is, what is the government going to do this year? What is the government going to do on this? The focus is on the government solving our problems.

RAJAGOPALAN: And not getting out of the way instead, which is, again, the enlightenment way of thinking.

LAHIRI: Correct. It’s been 70 years of proof that the government is not the solution. The government is the problem, so what we have to do is not more government but get the government out of the way, as you rightly put it, and that doesn’t cost any money for the government.

RAJAGOPALAN: It costs a lot of political capital that’s hard to come by.

LAHIRI: Yes, costs political capital, but it doesn’t cost direct, so this fiscal stuff that you worry about—reforms, et cetera—regulatory reform is actually cheap financially. Fiscally it’s cheap; politically, maybe expensive. Again, it also shows up in the attraction of individuals, whether we’re always looking for the Great Leader, the Purushottam who is going to lead us to salvation. Looking for the next Moses—it’s not just India. It’s in a lot of places.

At some level, I think maybe one should be more self-effacing and admit that this notion of the democratic structure as being the way to organize life is a very modern, new experiment in the grander scheme of mankind, in the history of mankind, or any herd-based groups, which, typically, have always been led by one.

That seems to be the natural order, whether it be out in the world of always looking at one leader, and you can’t have a choice by aggregation of views. It’s a relatively new thing. That’s why you see, at the first sign of problems, people are regressing to looking for this—

RAJAGOPALAN: Very centralized solution.

LAHIRI: Very centralized, yes.

Liberalization and Inequality

RAJAGOPALAN: I want to talk to you a little bit about liberalization and the many things that followed. The first part is inequality. Now, it’s fairly well established that liberalization between, say, the late ’80s and for the two, two and a half decades that followed, India might have lifted anywhere between 250 million to 300 million people out of poverty, depending on where you start and when you end. That’s an extraordinary achievement in a fairly short period of time.

Now, the idea that liberalization has increased the income levels generally of everyone seems to have a lot of buy-in. The typical accusation against it is, it’s increased inequality. First of all, where are we on increase in income levels across all the different groups in India? And where are we on increases in inequality?

LAHIRI: There are two aspects to inequality. One can do it as, you can divide up a country or any society into constituent groups with some identified or well-defined markers. In India, it could be caste, it could be geography, it could be gender. You could do it on multiple margins. What you find is, when you look at inequality based on inequality across groups, or between groups, if you will, those things seem to have shrunk.

Whether it’s rural-urban inequality, or whether it’s caste-based inequalities, at least at the broader level, as scheduled castes and tribes versus non-scheduled castes and tribes, those gaps seem to have shrunk. It’s also showing up in faster paces of intergenerational mobility from amongst some of the relatively economically backward castes or economically backward regions catching up. There is a fair bit of evidence that that has happened over the last 30 years. We’ve seen big gender gaps have actually been decreasing, as well, quite rapidly.

The other aspect of inequality is within-group inequality. Do you see people within rural areas having inequality? What has happened to that? Within urban areas, within castes, within gender. My sense is, there is evidence that within-group inequality has widened, but that by itself shouldn’t be that surprising that you’re going to find within groups some winners and losers. That’s exactly how merit-based systems should be working.

We’re not all identically endowed and everything. If that were the case, then, of course, differences in outcomes would be completely random and would tend to revert themselves. But inasmuch as a market system, as a merit-based system, is rewarding people for their ability, you should see within-group inequality widening as you move more and more towards market-based outcomes.

You’re not trying to manage outcomes. You’re outsourcing that to the market. But you should also see across-group, unless you believe that groups, by themselves, are differently endowed. Then you will find that, as long as groups are not differently endowed, on average, any gaps are evidence of some preexisting distortion of the system.

What we have seen in India is those gaps decreasing, but within-group disparities potentially widening. It’s important any time there’s inequality discussions to keep these two different aspects of it in mind. Somehow, it often gets conflated, the two, but I think it’s important to distinguish between those two things.

Causes of Upward Mobility for Marginalized Groups

RAJAGOPALAN: I think another very uplifting thing in the empirical evidence, as you point out, is that historically, the poorest groups, who were also the most oppressed—this is the Dalits and the Adivasis—two things seem to be happening. One is their overall socio-economic level has increased, and the second is intergenerational mobility within these groups and families have actually shown upward mobility.

The children of Dalits are more likely to deviate from the caste-ordained occupation and more likely to take on a more lucrative job. Now, of course, it’s not that they’re all suddenly rich, but they are more likely to have this upward social mobility. This is work that you’ve done with Sourabh Paul and Viktoria Hnatkovska. Now, how much of this is because of affirmative action programs and this huge push towards constitutional equality and reservations? And how much of it is because of markets? Is there a way to disentangle that?

LAHIRI: People tried different things. I actually just finished a paper on this topic, trying to understand this, decouple these two effects—growth and affirmative action—on outcomes. Of course, the way we do this is by looking at it through the lens of a model, and do the identification. The idea in the paper is, let’s write down a model, a structure. The structure will generate gaps between groups. We try to rig the model such that it gets the gaps between groups exactly right for 1983.

Then we hold those gaps—the rigging—constant. We just hold the calibration parameters constant. Then we hit the model with just growth, the growth that we actually saw between 1983 and 2012. That’s the period we have. And then we ask, what does that growth do to intergroup inequality? What is the predicted inequality in 2012? And how well do we match that with the actual gaps we saw in the data in 2012? What we find is that growth alone can explain upwards of 80% of the overall shrinking in the wage gap.

How important is affirmative action? Clearly, in our structure, affirmative action has two parts to it. One is the education component, and then there is the job reservations in the public sector. Now, public-sector employment is a very small fraction of Indian employment. If you just add everything up, including the states, it’s a very, very small part.

The Indian workforce is close to about 650-700 million people. Public-sector employment is about close to, I would say, 20 [million]. It’s not more than that. Once you add up everything, even across states, the federal level is much lower. The big part of Indian employment is actually outside the public-sector space.

But higher education does matter once you start thinking of the economy as a layered structure where people can work with almost no training in agriculture, with very minimal training. Then to access some employment in baseline manufacturing, you need to get greater skills. You at least need to be able to follow instructions, maybe read some written instructions.

Go a bit more up the technical ladder, you need some polytechnic-based competence or some technical education. Then to move up into the white-collar service sector, you need more, and so on. That’s the kind of structure we have in the model.

Education reservations do matter for transiting out from untrained agriculture towards more trained vocations. That’s where these reservations might be important. Of course, we keep the reservations constant because reservations, in an overall sense, haven’t really changed. Based on what we had, we’ve continued having since the early ’50s. We’ve had that in place. The only thing we’ve done is, we’ve expanded OBC reservations, which really helps the others, not the SC/STs, but in spite of that, we see this.

What we find is that, if you remove reservations in education, you would see an impact on the wage gaps. The wage gaps would actually widen, would worsen, but it’s more like a static effect. In terms of the dynamic effect of growth, you would have seen a convergence, a pretty sharp convergence between groups, with or without reservations. All that the reservations do is, in some sense, move the whole wage gap, the level effect. Affirmative action is almost like a level effect, as we have it in the model. We don’t know whether that’s the right model, but at least from the lens of that model, it’s a level effect.

The growth part just keeps going down at the same rate, except both the starting and the end points come down if you have reservations. If you don’t have reservations, both the starting and the end points go up, but then the dynamic effect is often growth. That’s the sense in which we think that growth basically lifted all boats in the Indian context. That’s the dynamic effect.

That said, one might think, in a more long-run structure, you could have this propagation coming from affirmative action, which can have dynamic effects because it might change the initial distribution of endowments because, inasmuch as we are kids of people who went to college, the human capital we are getting—just throw in the nurturing component—starts transmitting itself or propagating itself.

As time goes on, as more and more people from SC/STs and those backgrounds are getting educated, they’re going to be passing on more of that inbuilt knowledge to their kids. It generates a better initial distribution, which is a mechanism we don’t have in the model.

RAJAGOPALAN: But it’s important?

LAHIRI: It’s an important mechanism, but I see that as a more long-run thing, not in a 20-year period. You’re unlikely to see big changes coming from that margin.

RAJAGOPALAN: On job reservation, do you think there’s a pull effect? I know it’s a fraction of the jobs in the economy, but there is such a rat race to get one of those jobs. It’s such a prized thing to get a government job—reserved or otherwise—that people make massive investments in human capital to get that job, and it’s not like the human capital dies if they don’t get the job.

LAHIRI: I agree with you. One should always be a bit skeptical of using just baseline proportions to make arguments because sometimes people also make these arguments that trade doesn’t really matter for the U.S. because just 1% of overall U.S. economy is trade, but it does really matter. In that sense, it’s true that it doesn’t matter because pricing is always at the margin. As long as there is this open fringe, it’s going to create that marginal guy who’s going to start pulling all prices.

I totally hear you there, that it certainly, probably has an effect on people, but the only thing I would say on that is that that mechanism has been there since the mid-’50s.

RAJAGOPALAN: It shouldn’t suddenly spike after ’83, ’85.

LAHIRI: Exactly. What we have seen is this dramatic change after the 1990s. The thing that really changed was growth. This is where you see this effect all across India, across groups, exactly this thing that you’re mentioning, parents working incredibly hard to invest in their kids through sending them to whatever ramshackle structure that is promising English medium education for their kids in a one-room thing where some guy comes in.

They think that that’s their ticket out, that there’s some urban employment for their kids, which might be facilitated by them knowing a little bit of English, whether that’s being a driver or just some low-level office employee in one of these big peri-urban centers and so on. That has been the big driver, though, the thinking that those are the jobs that are opening up—not so much the public sector. At least that’s my take on it because the public-sector employment bit has not changed.

Lack of Social Mobility for Muslims

RAJAGOPALAN: What about, say, a group like Muslims? There are some papers on this, like Paul Novosad and Sam Asher have written. Their conclusions are very much in line with yours on the upward social mobility in Dalit and Adivasi groups. But they find, actually, that benefit is entirely offset by the lack of mobility among Muslims. What do you think is driving some of that?

Now, one possibility is if the growth is really being driven by affirmative action, then, of course, the Muslims are traditionally excluded from it because it’s caste-based and not religion-based. But if it is mostly market-driven, then what is going on with this particular socioeconomic group?

Is it one of those Timur Kuran long-run divergence explanations, where institutions of lending, financial intermediation, how they hold assets, how they invest capital as a structure among Islamic groups—is not letting them participate in markets as effectively? Or is it regime uncertainty because of discrimination? Is it access to education? There must be something going on that one particular group doesn’t see the same kind of upward social mobility.

LAHIRI: My take is, it’s less to do with affirmative action, and it has more to do with these other factors that you mentioned. First up, Muslims were doing much better than SC/STs. Up until the early ’80s, they were doing much better. One way to focus on this is to say, “Okay, I’m going to start looking at all religious minorities.” Because if it’s affirmative action-based, then what you should see is SC/STs doing better than Christians, doing better than Buddhists, doing better than Jains, doing better than Muslims.

There’s no reason to single out Muslims as the only minority that, somehow, this affirmative action has messed with, whereas it doesn’t seem to have done. Absent that kind of study, I’m not completely convinced that affirmative action is a story that is potentially explaining what’s happened to Muslims. There are these numbers issues, but exactly the argument that you made earlier is, they are not directly persuaded by numbers. However, I think the regime uncertainty might have a lot to do with it.

More than that, I worry about the fact that there exists a parallel institutionalized education delivery mechanism for certain groups. The Muslims, in particular, have one. What that has done, whether that curricula has been evolving to be consistent with modern market needs or not—I’m not sure what these madrasas are teaching. I think this requires a much more in-depth analysis of precisely where do kids from madrasas go? Are they looking at employment in the modern economy? Or are they looking at employment in very, very segmented labor markets? Why is that?

Are there specific kinds of markets that they look at? That suggests more some sort of social-religious issue that is driving choices being made by that group, as opposed to a pure issue of discrimination with the market. It may be that there’s a push effect, that there is some market-based discrimination which is generating this. It could be a perception-based issue. But I think it’s a much more long-running issue than the last five years or something.

RAJAGOPALAN: Even if you don’t have discrimination in the market, just social discrimination can sometimes nudge you in certain directions. For instance, if there is the sense of discrimination or some kind of unfair treatment of Muslims, they are more likely, for reasons of job security, to take the traditional employment that their parents might have followed or their kin groups might have followed because that just gives you more certainty than engaging in manufacturing or some big industrial shop floor or something like that.

Even if firms are not directly discriminating against Muslims—if the same people outside of hiring decisions are, then it can create this weird effect. I don’t know if there’s something like that going on because it seems to have done that when it comes to housing choices. There’s been greater and greater ghettoization of Muslim communities across all incomes in urban areas, in a way that wasn’t true even 40, 50 years ago. This is a very new urban thing, so I’m wondering if there’s a similar effect going on in employment.

LAHIRI: Maybe an interesting thing to do at some point is the trajectory of Pakistan post-1980, post-Zia, which is when this institutionalized education . . . There was a massive focus on religious, seminary-based education which started picking up a lot of Pakistani children from rural bases. How much does that mirror what we’ve been seeing in India, to try to understand? Because that’s a case where there is no minority-based persecution or anything. Just to do with a different view of what—

RAJAGOPALAN: Human capital building.

LAHIRI: Yes, building is supposed to be. I think that may be the way to look and see if you see different trajectories or something similar happened there as well because that, I think, is an important question. I’m personally aware of this urban housing discrimination that leads to ghettoization.

By the way, it’s not just religion-based. My father, many years back, was stationed in Ahmedabad. At that time, I used to visit him there, and he was living in a government accommodation because he was working for one of the central government services. I had some cousins there as well.

One of the key things that would come out is this housing discrimination based on food habits. If you were a non-vegetarian— that’s the first question that a homeowner would ask if you went to rent, do you eat or not? If not, you go to the Muslim part, would be the standard thing.

This kind of ghettoization has been there at various levels. It’s certainly much more endemic with Muslims, has become much more endemic with them over the last 20-odd years. I didn’t remember hearing so much of it up until the mid-’90s. After that, one was hearing much more.

RAJAGOPALAN: Post-Babri. That seems to be the break.

LAHIRI: Yes, exactly. Post-Babri is the break. This is this thing that suddenly society normalized expressing thoughts which were previously viewed as being non-kosher. Now it’s normalized. You’re empowered to say whatever you want. I think that’s partly what has happened.

Rural Versus Urban Mobility

RAJAGOPALAN: One kind of convergence we just talked about is caste groups and intergenerational mobility. The other question I had is rural-urban mobility. You already hinted on this earlier when you were talking about the nature of migrants and things like that. It seems like that is shrinking. . . .

LAHIRI: The gaps are shrinking.

RAJAGOPALAN: Part of it is because the urban incomes in the informal sector are not as high as we had hoped, and second, that the rural areas are transforming into non-agrarian parts, but they’re still rural or semi-urban or peri-urban—whatever they’re calling it these days. They’re moving out of agriculture, which we know has been declining in productivity, but they are not yet moving to urban areas, and a lot of rural areas are actually doing quite well in non-agrarian activities. Is that a good way of thinking about this convergence?

LAHIRI: Yes, it is. One of the ways in which we traditionally think about rural-urban is that rural is agriculture, urban is all non-agriculture, and that, I think, is an oversimplification. It’s not just in India. I think this is true of patterns of structural transformation that have happened even in the West, that rural areas tend to reinvent themselves as they go along.

As structural transformation happens with agriculture beginning to shrink, people have to reoptimize. They need to find ways of accessing jobs in the expanding non-agricultural sectors, which are mostly nonrural, which are urban. As a result, some people are able to. The traditional models of this, like the Harris-Todaro kinds of models, were written based on this difference between urban and rural.

In response, some people clearly move to urban areas and try to find jobs in these expanding non-agricultural jobs in urban areas, but rural areas begin to reinvent themselves as well. Clearly, once you start looking at areas in the West—look at the U.S. or look at Europe—traditional villages. In the U.S., we don’t talk about places being villages, but in Europe, people talk about villages. There’ll be a guy who is working in the IT sector who will tell you that he lives in a village on the outskirts of Berlin.

Village culture is very alive, but the village is not an agriculture village. It didn’t happen overnight. It happened gradually along the process of structural transformation. India is no exception. The only thing is—and I think this might have happened in the West as well—the way, typically, when structural transformation happens, there are requirements for greater connectivity across regions, across cities. The transportation network expands very sharply.

You have highway systems being laid down, possibly, increasing train connections being laid down. A lot of the newer activities that rural areas come up with are centered around the commerce abutting these newer highways, et cetera. That’s where you see this big churn happening, people quitting agriculture and going into these roadside restaurants, roadside small mechanic shops, tire outlets, and so on. There’s a tremendous amount of entrepreneurship that exists in India, as opposed to this notion that somehow we’re just a job-seeking group, but that’s not what it is.

Ninety percent of Indian service-sector employment is actually Own Account Enterprises, which is mom-and-pop shops, where it’s one guy with his wife and maybe a kid, and they run the whole thing. That’s entrepreneurship in India, and that’s how rural areas in India have been reinventing themselves. A big component of the rural pickup is actually expansion on those non-agricultural things.

Urbanization and Manufacturing Are Necessary for Growth

RAJAGOPALAN: Does this mean that India can somehow get out and have this great structural transformation without urbanizing and without manufacturing, which have traditionally been the ways by which countries get out of that poverty trap and go into the middle-to-high income? Is this a way out? Or will this also hit some ceiling very quickly when these mom-and-pop stores can’t scale, and when everything has been reorganized, and the music stops?

LAHIRI: No, I don’t think that there is any escape from manufacturing because the flip side of having 90% of service-sector employment in Own Account Enterprises is that these are, by themselves, not high growth, high productivity. You can do just so much selling bananas and peanuts, and it’s not going to be a big growth sector.

Our view of services as being this driver of growth is really a three-city narrative, based on Bangalore, Bombay, Delhi and maybe some Hyderabad, and that’s the majority. It’s very high value-added but very low employment relative to the scale of the economy. If you’re putting out a million people a month, you’re not going to have—

RAJAGOPALAN: They can’t absorb them.

LAHIRI: It’s not going to be, and it’s futile for some person who’s educated until grade 10 or grade 8 to expect employment at Infosys. He has to be aspiring for a job like a machine operator. That’s what his skill set is going to allow him to do, the best that it can allow him to do. There’s a mismatch if you somehow say that we’re going to be expanding, the service sector is going to spark our growth. It’s not going to spark our growth. It can tide you over in the interim while you’re waiting for this. Our only hope is low-tech, large-scale manufacturing.

RAJAGOPALAN: And a majority who are young. Half of India is below 25. It’s frightening.

LAHIRI: We are looking at the difference between demographic dividend and demographic curse—a very thin line.

RAJAGOPALAN: We’re on the brink.

LAHIRI: Yes, we’re on the line, and it could turn into a South Africa very quickly of gangs of unemployed people walking around cities with huge issues of security and all the associated manifestations of that. We are very close to that if you are unable to find large-scale employment.

RAJAGOPALAN: You see this happening already in parts. Snigdha Poonam, in her book “Dreamers,” has written about this, how people are aspiring to be gau rakshaks because it just gives them a certain status in society in a way that they can’t get because they have no jobs, and they have no future. It’s terrifying. That is not a job, and it’s not a good substitute for a job.

LAHIRI: Absolutely, I agree. I think of the rise of vigilantism, and people just looking for some anchor or some place to validate their self-worth, and if the market doesn’t provide it, these are the alternative ways in which you find it. I can completely see that. It’s very scary, and I don’t think we are appreciating it enough.

In some sense, I think this spending so much of societal resources on welfare schemes of the types that we did—this was like a one-time bonanza of money that we had because of this drop in oil prices. I’m not sure we spent it in the optimal way. I think it was good for the people who benefited from it. Some of these programs are great. Swachh Bharat Abhiyan, I think, was outstanding in terms of, at least, its conceptualization and what it was aiming to do. The benefits are massive. Some others were not quite as well thought out.

RAJAGOPALAN: All your loan forgiveness and fertilizer subsidy and free water and free electricity—this should have ended 20 years ago. There’s no way India can afford it. It’s 2% of the GDP almost—all these agricultural subsidies put together in a very unproductive sector. The expiration date has come and gone, and we’re still hanging onto this stuff.

LAHIRI: I know, and the sad part is, politics remains like competitive socialism. That’s what Indian politics is. Every electoral cycle is just this. Each party is trying to outdo the other in terms of what amazing thing they’re going to deliver. It’s a vicious circle because once one guy does it, everybody expects it. And so, both on the receiving end and on the giving end, it’s just a bad ecosystem to be in.

The short answer is, I don’t think there’s any escaping expansion of low-tech, large-scale manufacturing. That is the panacea for a country like ours. If we were a small country, that would have been something else.

RAJAGOPALAN: Look at Bangladesh. Just look at Bangladesh. That is the model in terms of very similar people, language, culture, South Asia. It was just completely devastated by a civil war, and now it seems to be on a growth trajectory that can easily outdo India in the coming decades.

Why Economics?

RAJAGOPALAN: I wanted to ask you some questions about yourself. One is, how did you become an economist? Or, if you are named Amartya, it’s just predestined?


LAHIRI: Actually, funnily enough, I was named after him.

RAJAGOPALAN: Oh, that’s lovely.

LAHIRI: It’s not that common a name, at least in my generation. It’s become more common in the generation after mine. In Northern Bengal, I think, in the ’40s, ’50s, ’60s, it has that feel of what we hear about in France, of intellectuals being fettered, being almost public personas, and so on. Amartya Sen was, I think, very much like that.

He was very renowned in terms of all his achievements and youngest professor ever at Jadavpur University. People would know in Bengal at that time—that generation of Bengalis would know the names of those who finished in the top 10 in high school, who finished in the top ten—they would be tracking. Amartya Sen was like a rock star, just like Lévy’s a rock star in France amongst philosophers. That’s how he was. I think that’s where the name came from.

RAJAGOPALAN: Were your parents economists? Or they wanted you to become an economist?

LAHIRI: No, my father was a mathematician. He was a few years after Sen in Presidency College, but another of my uncles was actually a cohort of his in college, of Sen. My father was a college professor for a couple of years, and then, while deciding whether he wanted to go abroad for a Ph.D., he’d taken the Civil Services Exam as well, and so he switched, and that’s it. He quit his academic life about two years after he started it.

My mother was a historian initially, a history major, but then switched to languages. After she had me, she was staying at home for a while. She’d finished her M.Phil. but she’d just started a Ph.D. in international relations, which she stopped. Then, when she came back after I had grown a bit, she found she didn’t want to do that anymore, so she switched to languages. She retired as a professor of Italian language and literature in Delhi University some years ago.

I was named after him, but my choice of economics was a bit more practical. It had nothing to do with great love for the subject because I had barely spent much time studying. My first choice of career was cricket. I was actually a very active cricketer growing up. I was playing for my state at the age limits. I played up to the Delhi Under-22 level.

When I was choosing colleges, the big thing was that I needed to do something which would not force me to attend too many classes and yet survive. That took out the sciences because the lab. That’s what my cousin, my father’s elder brother’s son—he’s an economist as well—Ashok Lahiri. He’s the one who was of great influence. He said, “Take up economics,” and that’s how I got into economics.

I was playing for my state, I played for Delhi Under-15, Delhi Under-19, Delhi Under-22. I was playing active club cricket and college cricket. I was in St. Stephen’s College. I was playing for them. That was a full-time thing for me—that was my first goal. I actually played as a professional in Bengal in the Calcutta League for a season as well.

But then I decided, as I was finishing my undergrad, that I should find a career with—talk about uncertainty shocks—with less uncertainty about what’s going to happen. I decided to stop playing cricket, and that’s when I joined Delhi School of Economics because, at that point, it was the logical place to go because I’d just done an undergraduate in economics.

The love really came in while in graduate school. When I went to Delhi School, it was really learning economics from scratch because I just hadn’t paid much attention. It’s one thing getting through exams. It’s quite another to really know the subject. It was literally being thrown into a cauldron, trying to also cover all the stuff that I had not paid any attention to over the previous three years.

Those two years, I had great teachers. I had some great influencers, actually, who somehow gave me the sense that I could hack it as an economist. Professor Badal Mukherji was a professor in the Delhi School of Economics. For some reason—I think because he liked cricket or liked sports—he somehow took a personal liking to me, and he was a great influence on me. Then another professor, Kaushik Basu, was great. He ended up writing recommendation letters for me as well.

The two of them were big influences in the sense that, somehow, I got the sense that they thought that I could hack it, which was a big thing because without that, I would not. Then there was the push factor. You talked about the pull and push factor. I was doing my M.A. in 1988 to 1990, and at that time, this was pre-reforms. The thing was, either you then went to one of these IIMs and tried to aim for a private-sector job, or you took the Civil Services Exam.

My mother actually was quite keen on that, that I take the Civil Services. My father was in the Civil Services. I just didn’t want to do that. This seemed like the ticket out that, apply and go outside, and then we’ll figure something out. It’ll be my ticket, my escape from these two professions that I was staring at, and so that’s why I applied for a Ph.D. program.

But then, when I came to graduate school, one of the great things that I thought was a major difference between the education model in India versus the West was this de-emphasizing just rote memory and emphasizing deeper learning about concepts and what exactly is driving things. Make yourself a person who is sitting in this environment you’re studying and ask. That way of thinking was very different. It took the stress out of exams. That was the big thing. Once that stress was gone, and I started reading for the sake of reading—that’s when I think I fell in love with the subject.

RAJAGOPALAN: Who was at [the University of] Maryland in those years?

LAHIRI: Well, I had two advisers who were both incredibly intellectual and one person who was not on my advisory committee but, again, took a lot of personal interest in me. One of my two advisers was Allan Drazen. He used to do much more macroeconomics at the time, but has since become a political economy person. He’s got a big book and so on. He was making that transition at the time from macro to political economy. And then Guillermo Calvo. They were my joint supervisors, the two of them, who took a lot of interest in what I was doing, gave me incredible access with a lot of warmth.

Then there was another gentleman called John Haltiwanger, who’s in Maryland, who’s a macroeconomist. These were the three people. You start doing things often because you want to mimic people. Both Calvo and Drazen—something about the way they would present facts or present ideas was a tremendous influence on me that I found almost cool, if one can do something like this and pull it off. They seemed to know a bunch of things. They seemed to be able to present things with remarkable clarity and always insightful.

And role models—you always need role models somewhere along, good role models to follow. Sometimes one is just not very lucky in finding them. In fact, when I started my graduate program I wanted to do theory in micro because that’s what all Indians do.

RAJAGOPALAN: Yes, that’s what Indians do. You’re one of the few in the ’90s who went in that direction. You might be the only.

LAHIRI: It helped that Maryland at that time didn’t have any micro theory people who were—

RAJAGOPALAN: Schelling hadn’t come to Maryland by then?

LAHIRI: No, I was actually a TA for Schelling. I actually did spend two years with him very closely. It was a remarkable experience. Even when I didn’t need to, I would sit in on the classes because I always found them incredibly amazing. What I always remember about Schelling is, he would walk into these classes with no notes, no nothing. Actually, all three of these people I mentioned had this unbelievable thing.

Schelling, of course, would talk. He would introduce a topic like incarceration as a public policy. “Let’s have a discussion about it.” And then it would flow. That’s the whole class—50 minutes, or 1 hour 15 minutes. He would just engage in debates with people. The first person that he would call upon to say something, and he would take the opposite argument, and then the thing would just keep flowing.

The whole test would be . . . the power of the mind of the man was whether each interpersonal exchange—could somebody last three questions, or could somebody last four questions before an inconsistency would be pointed out between what the person started off. Just the power of the mind.

I decided to subject myself to this kind of examination with him when I was coming up with Ph.D. thesis topics. I twice, actually, went up to him saying, “Could you give some time? This is what I’m thinking of. Tell me.” The first time I went to him with something—of course, this is not an area he directly used to work in—I think I lasted four questions, and then there was an inconsistency with some line of thought I had. The second time I lasted about the full hour. I thought, “Okay this one probably has more legs to it.”

Just an incredible mind, and it was a great privilege. But also, a very warm human being on top of that, which is the other memory.

RAJAGOPALAN: But not a theorist’s theorist, right?

LAHIRI: No, not at all. I actually never saw him write any equation, no nothing. He was just pure logic, and the clarity and the acuteness of the logic was just awe inspiring in terms of how clearly he could think and see. It was like a chess player who can see six, seven steps ahead as to where he was going. Phenomenal. I forgot about Schelling; I’m glad you mentioned him. He was a great influence.

Writing Process

RAJAGOPALAN: What is your writing process like?

LAHIRI: It depends on what I’m doing. There is this popular writing stuff. I’ve tried to keep away from writing syndicated things, which implies trying to produce something every two weeks.

At some point, I was in India for a couple of years, in 2017 to 2019, and I found myself writing because I was much more in the Indian political space. Then these guys would start reaching out, saying, “Why don’t you write another piece?” And I found that I didn’t enjoy that because it felt like I was forced to write.

RAJAGOPALAN: It’s a job with a deadline.

LAHIRI: Yes, and often you’re forced to write something on things that you may not feel. It feels more contrived because you try to come up with something to write on, as opposed to “I really want to write on this.”

That kind of writing is things that fester through my head. I can be walking or biking or whatever. When I sit down to write on something that has been playing on my mind, that can be just a two-hour process, and I write something out. That end product would probably take two hours to write. There’s a lot of this festering in my mind for the previous week while I keep thinking about it. The actual writing can be very quick in the end, but that’s because of this festering done in the head.

Academic writing is a bit different because that has become a much more long, drawn-out process now because there’s a data component to it. This notion of having an idea and then completing a paper involves so many different steps.

Economics, professional academic writing, is now becoming more and more factory flow output. It’s because referees are expecting things, and the editors respond to the referees. You’ve got to respond to both of them. What it has meant is ideas are just a small component of it. People do much better at this process than I do in terms of publication. I’m maybe not the best model for this.

But you need to have an idea. Then you need to have some preliminary evidence for the idea, that the idea is motivated by some facts. You can call these stylized facts, or you can call them some motivating facts that say “Okay here is something intriguing that I want to look at.” Then you come up with an implementation of the idea in terms of some rough model which will give you some testable hypothesis, and so on and so forth.

Then you go to the next stage, which is where you actually take the model and take it to the data. That’s a structural implementation of the model. That could involve just calibrating a model and running some exogenous data shocks through the model, or you could be estimating the model in an econometric way. Either of them. That is confronting the model directly with the data.

The model has to then pass the data test, and then you do some counterfactuals on the model to talk about some policy implication. Once you’ve done that, then there’s going to be some key driver or mechanism that you’ve embedded in the model. Then you need some independent evidence to back up that key driver. Can you find that? That’s five steps along the way.

Even a quick one, like the COVID paper, which we thought was really quick—it took about three or four months to write. And then, by the time the whole process ended, it was still a year. That was because this journal was looking for COVID papers, so the end process got a bit quicker. Otherwise, you’re looking at three years. By the time some paper gets published, you’re so sick of it, and it has become remote in your own mind, and that’s if you’re lucky.

I have a grouse with the profession in the sense that it has gotten so besotted with dotting the i’s and crossing the t’s that it has lost, sometimes, sight of the woods for the trees.

RAJAGOPALAN: And the magic of the idea.

LAHIRI: Yes, this is a balancing act. Clearly you want some rigor, but you shouldn’t be throwing the baby out with the bathwater and end up in this quest to chase down. I think that is something that we need to worry about it.

RAJAGOPALAN: Yes, we’ve got to make room for the Schellings. We seem to have completely flushed them out.

I have one last question, which I think is the most important question during the pandemic. What are you binge-watching?

LAHIRI: [laughs] I binged-watched Money Heist at some point, which I hadn’t seen before. That was fun. One Indian show, Aarya, I got into at some point, and I ran through that one. Then I went through an old series, Monk. Then my wife suddenly got into The Crown, so I started watching The Crown. I do remember watching something else. I don’t know if you’ve seen a series called Ambassadors.

RAJAGOPALAN: Thank you so much for doing this, Amartya. This was such a pleasure.

LAHIRI: Thanks again. I enjoyed this tremendously. It was great chatting with you, and I look forward to some informal chats with you at some point.

RAJAGOPALAN: Absolutely.

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 me on Twitter @srajagopalan. In the next episode of Ideas of India, I speak with Salil Tripathi about what India can learn from Bangladesh.

About Ideas of India

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