Krishna and Rajagopalan examine trade agreements, tariffs, and economic growth

SHRUTI RAJAGOPALAN: Welcome to Ideas of India, where we examine the academic ideas that can propel India forward. My name is Shruti Rajagopalan, and I am a senior research fellow at the Mercatus Center at George Mason University.

Today my guest is Pravin Krishna is the Chung Ju Yung Distinguished Professor of International Economics and Business at Johns Hopkins University, at the School of Advanced International Studies (SAIS) and Department of Economics. We talked about history of preferential trade agreements, India’s approach to trade liberalization, whether such agreements are trade creating or diverting, and much more. 

For a full transcript of this conversation, including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts. 

Hi, Pravin. Welcome. It’s so nice to have you here.

PRAVIN KRISHNA: Real pleasure to be here. Thank you.

Essential Components of the Postwar Multilateral Trade System

RAJAGOPALAN: Before we get into the paper that you wrote for us for the 1991 Project, which is about preferential and free trade agreements that India has done, bilateral and regional agreements, I just want to get a sense of how the multilateral trade system emerged in the postwar period. Then if you can give us specifics on how India became part of that system and where we are today.

KRISHNA: Great. The multilateral trade system, the design that it has today, I think the origins of it are Bretton Woods. That sort of design in the 1940s was really shaped by the global experience with international trade, and international trade deals in the interwar period. You had the Great Depression, and there was a serious effort by countries around the world to try and do something using trade policy to direct domestic demand toward their own producers. What that meant in practice was some combination of tariffs as a protectionist measure and exchange rate policy.

The level of arbitrariness of the use of trade policy in that period, and the feeling that, if anything, it had worsened the economic downturn led people during the time to start thinking about ways in which once you got out of World War II, that you would have a more liberal trading order, a more rules-based system. One in which countries perhaps couldn’t deviate from agreements and their goals with respect to trade policy quite so easily. That’s where it all began.

As a practical matter what that meant was in terms of the actual rules of the world trade system, what we call the multilateral trade system, there are two essential components. One is the principle of nondiscrimination. Central principle what’s sometimes called the most favored nation [MFN] clause, really Article 1 of the General Agreement on Tariffs and Trade [GATT], essentially that was the precursor to the World Trade Organization. That says that an importing member country of the GATT cannot discriminate in its imports based on the country of origin of the export. So whoever’s exporting to—if it’s United States or importing from Japan or importing from France—you cannot have two different tariffs for the same exact good. That’s nondiscrimination.

Alongside that, there was this norm of reciprocity. Not quite a rule, more of a norm of reciprocity, which said that if I offer you some tariff concessions, then you’re expected to offer me reciprocal tariff concessions. In the face of it, these look somewhat separate from each other, different kind of ideas. What was important about the postwar system was that these two were exceptionally powerful in combination with each other.

Reciprocity meant that my tariff reductions, so on the one hand I offer some tariff reductions, they’re intended for you, but because of nondiscrimination these apply to everybody. Because of reciprocity, now everybody’s expected to reciprocate in my favor and so on and so forth. This created this kind of a spiraling out effect of liberalization where an initial round liberalization would apply to everyone, there would be reciprocal reductions by everybody that in turn would stimulate another round of reductions and so on and so forth. That was the idea, and that’s how it worked for most part at least for the first 30, 40, 50 years of the system.

That’s the multilateral system, the cornerstone of which is the MFN principle or the nondiscrimination idea, which we’ll come back and revisit in the context of free trade agreements and how they deviated from that.

India’s Role in the GATT and the Special Status of Developing Countries

With respect to India’s role, India is a founding member of the GATT. It’s one of the initial signatories of the GATT. For the most part, because the GATT had included in it Article 18, which gave a special status to developing countries, and that special status really meant that developing countries needn’t reciprocate. They could sit on the sidelines in effect, benefit from the liberalization undertaken by the other countries and not offer reciprocal concessions of their own.

RAJAGOPALAN: Or rather not benefit from the liberalization of the other countries and sit on the sidelines.

KRISHNA: Sit on the sidelines, if they could benefit, they could. It had its own problems, which we could talk about, if you like, in terms of not incentivizing developed countries to offer the kind of liberalization that would be of benefit to the development countries. Because they knew they would get nothing out of it. It was a relatively low-cost move for India to be part of it, but it was there.

The one thing which is interesting—and maybe we’ll come back to it again in the context of free trade areas and so forth—the one exception, and most people are not aware of this, the one exception that India sort of negotiated for itself in 1947. This will seem odd given the geopolitics of today, was India and Pakistan jointly negotiated that they be allowed to have better relations with each other than they would give other countries through the GATT. They wanted an exception from nondiscrimination, but one that would allow them to favor each other. Which again, given the world of today, that might seem counterintuitive and even a bit strange. That was in fact granted. If you look at the footnotes of the GATT, somewhere along the way there’s a line that says India–Pakistan would like special trade relations with each other. Because they were one country prior to this, this is going to be permitted. That’s the history of India’s engagement at least at the origin with the GATT.

India in the Global Trade System After 1991

RAJAGOPALAN: Now comes the 1991 moment, which you’ve also written about. There are two things that happened in 1991. One is, domestically there’s a whole set of licenses, controls, which are removed almost overnight. The second is there’s a move made toward systematically rejoining this global trade system. It’s not like they immediately join WTO on every single thing, but there’s an effort made to reduce tariffs across the board. Over a period of time, they slowly start liberalizing sector by sector, tariff line by tariff line.

How does that period work out post 1991 in India up to maybe, say, the pandemic, because that’s the next big break?

KRISHNA: You’re quite right. By the late 1980s, just a sentence of background, late 1970s, early 1980s India had turned itself into a near autarky with very high tariff rates. The top rate was probably the 300% range or whatever, it really didn’t matter.

RAJAGOPALAN: 356, I think, if I remember correctly.

KRISHNA: 350.

RAJAGOPALAN: It’s crazy.

KRISHNA: Extremely high, no imports coming in. Anybody that grew up in India during that period remembers that this country was almost completely shut off, and it was quite a rare and exotic thing to see a foreign good. The cost of that and the ways in which it was impacting the domestic economy had become very clear to everybody, including policymakers. By the early 1990s, for a variety of reasons, India decided to move away from that the old regime, the autarkic regime toward a more liberal—try and integrate itself with the rest of the world.

That actually proceeded in multiple steps. If you look at the average rates, or the top average rate across different goods, that might have been in the range of maybe 150% or something like that in the early 1990s. That went gradually, one reduction after another, after another to maybe by the mid-2000s, 2010 or so, to around 10%. There’s actually a very, very substantial liberalization.

The consequences of that were immediate to see. I think anybody that was in the country or reading about the country was interested. On the one hand, you saw greater imports, on the other hand, you saw greater variety of imports. Finally, and maybe most importantly, you saw a very significant increase in the quality of imports and goods being sold in the domestic market. The broad picture is it’s a textbook story in the sense that India was kind of an autarky. Liberalization led to a bunch of good things all that economists would’ve predicted and hoped for.

The Decline of the WTO and New Trade Dynamics

RAJAGOPALAN: When does the World Trade Organization start its decline? It seems like even before that decline really set in, there are now bilateral trade agreements, there are regional trade agreements. First of all, how does that whole thing start? It seems like it is places that are contiguous. Was it the European Union that kicked this off? But that seems to be the trend that dominated the last 20 years or maybe even more?

KRISHNA: That’s exactly right. One of the dominant trends in the world trade system is that the movement away from this multilateral architecture toward bilateral, or plurilateral trade agreements, as they’re called.

The origins of all of this were actually there in 1947 itself. This idea of whether there could be exemptions granted from nondiscrimination. If you formed a trade bloc with somebody, and you were a member country of the GATT, that really meant that you were giving some countries a preferential treatment, relative to what you were doing for somebody else. This was debated in the 1940s, and very actively. We’re talking about economists like Keynes, Cordell Hull, Harry Dexter White on the American side, very active debate.

There was a view expressed by the European members, in a sense anticipating the European Union already, that maybe a certain type of exemption should be granted from nondiscrimination and somehow made consistent with this nondiscriminatory architecture of the GATT. That was really done through Article 24 of the GATT which said, if you have a very close integration with each other, you really want to liberalize everything, nearly all trade with each other, and want to integrate in a very substantial way in the way that the European Union finally actually evolved, then that’s permitted.

You could form free trade agreements, you could form customs unions, where customs unions would involve some additional liberalization between the countries. You could have a common market. Customs unions would involve a common external tariff for everybody else or a common market, which also involves factor movements and so on. These, in a sense, were anticipated then and permitted.

The U.S. at that stage, meaning, again, the 1940s took a bit of a principled stand arguing against all this. Because it was already worried that there was going to be indiscriminate use of this window to dilute the main idea and derogate the main idea, which was the MFN idea, the nondiscrimination idea. But after some negotiation in which you saw very colorful language being used by Keynes, sometimes on one side of the debate, sometimes on the other side of the debate, finally was settled in favor of Article 24 included. Essentially, you’re being allowed to form trade blocs either in the form of free trade agreements or customs unions.

What’s important for the world today, one last institutional detail, which is very important, especially for India, is that both free trade areas and customs unions, which were sanctioned by Article 24, really required extreme levels of liberalization between the liberalizing parties. It really needed to liberalize almost everything. The legal language left a little room for what almost everything means, but for most part, you’re talking about substantial liberalizations.

There was a separate history having to do with the British and various other colonial powers, so they had the system of imperial preferences. Various developed countries that are granted special trading rights and different forums to developing countries that was basically GATT illegal and increasingly started to become an issue in the minds of countries as to how does one take these activities, imperial preferences, in favor of colonies and so forth, former colonies, and how do you make it GATT legal.

By the time you got to the Tokyo Round in the 1970s, it was a very important issue, what was introduced was something called the enabling clause. The enabling clause essentially said that if the recipient of a preference is a developing country, then that’s okay. Developed countries could give preferences against their MFN tariffs to developing countries. There was a further discussion and developing countries said, “Well, why should developed countries be the only ones that have the right to do this? Why can’t developing countries give each other a preference as well?”

To accommodate that idea, developing countries were allowed to deviate from Article 1, deviate from nondiscrimination, by getting into preferential agreements with each other. But, importantly those preferential agreements were not subject to the discipline of Article 24, which is to say that they could be very arbitrary, they could be very limited. In principle, out of 10,000 or so plus goods that are listed in the list of goods that could be traded, you could have a preferential agreement in which you’ve liberalized trade for one single good. That was permitted. That’s the enabling clause.

All this was institutional detail. For most part, as you got into the 1980s, coming to the question of when did this trend toward preferential agreements really fully kick off, the really relevant agreement was really between the European countries, the European free trade agreement. There were other agreements, but these are agreements in name and really didn’t involve that much trade. Nobody really bothered about it. Varieties of agreements nobody cared about, for the most part.

Then the U.S.—there are different viewpoints on this—but for a variety of reasons, the U.S., maybe because it got a bit exhausted with the liberalization that had already been undertaken at the GATT. And maybe the GATT membership had grown to the point that it had made for the multilateral liberalization a bit unwieldy and difficult to achieve. It started to start thinking about abandoning its fairly principled position in favor of nondiscrimination and the multilateral architecture, started thinking about alternatives. The first step in that direction with the U.S.-Canada Free Trade Agreement, late 1980s. There was immediate discussion after that about the expansion of that to include Mexico into the North American Free Trade Agreement [NAFTA] which was finally done by the mid-1990s.

These two together, seeing that the Europeans had already gone in one direction with the European Union, the Americans going the direction of NAFTA, and with many other countries starting to express their interest in joining either Europe or the United States in these agreements, started to make everybody wonder if maybe they should be doing something similar.

You went from the late 1980s, or let’s say early 1980s, really a handful of agreements that were relevant, maybe just one, big one, to by the time you got through the 1990s, early 2000s, hundreds of agreements being negotiated. By the time you got to 2010, you’re talking about several hundred agreements in place. There is no country left in the world, member country of the GATT that is, that is not also a member of some type of preferential agreement.

Oddly, in a sense, the average number of preferential agreements that a country is part of today, member country of the GATT is part of, I believe is seven or greater than seven. That’s a very large number. It’s a very substantial dilution of the original idea of nondiscrimination. That’s the path that the world has taken.

The one last thing I’ll add there, which might then bring us to discussion about India, is that because the enabling clause permitted countries to get into “easy” preferential agreements—it’s a very easy window, especially for developing countries, just going to negotiate something basic or shallow and then announce it to the WTO, and then you have a new agreement. There’s a very large number of agreements that are currently notified that have been notified through the enabling clause as opposed to Article 24. As far as India is concerned, most of India’s agreements have been notified through the enabling clause.

RAJAGOPALAN: One thing I’ll add there is this is also in the post-Cold War ending and the Soviet bloc disintegrating. Until that point, the United States seems to have this very principled stand as the leader of the free trade bloc. Simultaneously that disappears.

Now, unfortunately, India is liberalizing exactly when it needs this the most. In some sense, countries like India, many countries in Africa, which have been this semi-autarkic, almost entirely autarkic, because of their own political economy problems and lobbying and protectionism within the country, need an external force that can be the disciplining mechanism, right?

KRISHNA: Right.

RAJAGOPALAN: India saw this in 2001 when the United States and other countries basically just took them to court, the dispute resolution system, and then another wave of liberalization followed.

Understanding the Small Percentages of Preferential Trade

Now, in the absence of this, what are these preferential trade agreements about? I noted this down because I wasn’t sure if I had gotten this right. You write that only 16% of global trade is preferential, and furthermore, less than 2% takes place in goods that receive tariff preference greater than 10%. It almost seems like lots of smoke and noise, and it doesn’t really do anything.

KRISHNA: That is true. Let me break that question into two or three pieces. One is—this is your final question—which is you’ve had hundreds of agreements. What does that actually amount to in terms of how much additional liberalization has actually take place? The reality of that, which is very striking, and it was first documented very ably by the World Trade Report, I think it was maybe in 2011, 2012, thereabouts, where they did all these calculations and noted this point that you’ve just made, which is that: It’s only a very small fraction of world trade that receives a preference anywhere, between any two countries, that receives a preference of greater than 10%, and that number is 2%.

At one level, it seems odd, maybe slightly bizarre that you’ve expend all this effort, political effort, negotiations over decades, and you’ve ended up in a situation where a preference is actually quite marginal. That’s still a bit of a question for me. Why is that the case? Why have countries done this? Some of it is political headline management. The governments maybe appear a bit more successful having announced one trade agreement or the other trade agreement, but that can’t be the entire explanation.

The other piece of this could be that, although we’re talking now about tariffs primarily or trade barriers, that there’s an investment component behind this cross-border flows of capital that somehow is more secure because there’s a trade agreement in place. There’s a certainty of outcomes that’s assured because of the fact of a trade agreement. Even if the tariffs were only liberalized by a little bit or maybe they were liberalized in the sectors that that particular investment flow cared about, then that was the reason for engaging in that type of trade agreement.

The reality of the matter, and this is certainly true for Indian trade agreements as well, is that while you have trade agreements announced and the amount of trade that’s actually liberalized, the volume of trade that’s liberalized, the number of tariff lines that are liberalized that are relevant—meaning that are actually active in terms of flow between those two countries that are engaging in that agreement—can be quite limited. This is the case with a large number of Indian trade agreements as well.

Indian Trade Liberalization and Alliances from 2010–2020

RAJAGOPALAN: Now coming back to India, between 2010 and say 2020, India did this famous U-turn on tariffs, which I believe is again, ebbing. I think we’re moving toward a more liberal direction. Now, while that was going on, India also signed a number of FTAs. This is free trade areas, custom unions, those sorts of things. Also simultaneously liberalized its internal market through things like the GST which were creating internal trade barriers.

Now focusing on the external, the paper that you wrote for us, basically your argument is, any benefits from these preferential trade agreements or free trade areas are extremely modest and marginal. This is mostly with the East Asian tigers, the ASEAN bloc, Japan. These are natural allies, the East Asian and South Asian allies. India is set to be like this important leader in the Global South. It starts negotiating these deals with everyone. Nothing much comes of it. The first question is what does that landscape look like? Second, why doesn’t anything come of it?

KRISHNA: Let me take that second piece first. The fact of the matter is that India’s liberalization in the context of bilateral trade agreements has historically been reasonably shallow, which is to say that the volume of trade actually liberalized is quite low. On top of it, many of these agreements were negotiated, I would say, late 2000s, early 2010s, thereabouts, that window, and most agreements have a phase-in period.

RAJAGOPALAN: That’s right.

KRISHNA: That phase-in period is 10 years, maybe 15 years. In the case of Indian agreements, by and large, at least my reading of them, sometimes the text of these can be quite hard to get through.

RAJAGOPALAN: On purpose.

KRISHNA: Hundreds of pages of each agreement and so on. My reading of it is that the actual liberalization was not frontloaded, was backloaded. Let’s say you look at the U.S.-Japan Trade Agreement, U.S.-Korea Trade Agreement, and so on, my recollection is that most of that liberalization was going to take place after 2017, 2018, although the agreements had been signed maybe 2008, 2009, thereabouts, if memory serves me right.

By the time you got to 2018, then you’re already approaching COVID and so on. I’m, actually, personally not even sure at this point how much of what was promised under the agreement, and which was supposed to be implemented a bit late, meaning toward the end of that 10-year phase-in period, how much of that has actually been undertaken. Perhaps it has, perhaps it has not. If it has now been undertaken, then maybe, we’re yet to see the full effects. Maybe wait a few more years and maybe what was really hoped for is now being achieved.

By and large, for Indian agreements as well as agreements around the world, you’re going to see this tendency toward a relatively shallow liberalization and a postponement of the actual liberalization. One shouldn’t be too terribly surprised that the trade patterns have not really changed very much. In India’s case, the growth of trade within the agreements for India looks very similar to the growth of trade for India outside the agreements.

At one level, this should be a comfort to those who argued that India’s trade agreements would be catastrophic for India because they haven’t done that. On the other hand, if anybody was hoping that India’s trade agreements would really catalyze India’s growth in some dramatic new ways, then that hasn’t happened either. I think both sides can relax a bit because not much has changed.

RAJAGOPALAN: Here, the policy question is, is the reason that they didn’t yield anything more than very modest effects because they didn’t have the teeth of the WTO and GATT, which was the dispute resolution mechanism? There was a disciplining mechanism in the large global multilateral framework, which is not necessarily included. Many of these trade agreements are just slightly above a gentleman’s handshake, right?

KRISHNA: Right. I’m not sure that the dispute settlement played such a big role. One part of it is that many of India’s trade agreements, with neighboring countries, Afghanistan, Sri Lanka, et cetera, there’s some trade taking place. It’s not a huge amount of trade. Like we said, large bits of that trade are exempted in some different ways from the agreements in any case. The countries with whom we got into trade agreements, with some exceptions—you got Japan and Korea in there—but for most part, for the rest, these agreements were with countries we don’t trade a lot with in the first place. Maybe not much was expected to change in any event.

The liberalization margins were quite small. It’s not like you went all the way to zero, unlike what would’ve been expected through Article 24, because these were enabling clause agreements. You’re importing Sri Lankan tea and now you’ve given a tariff preference of 5% for Sri Lankan tea relative to somebody else. There’s going to be some effect, but maybe not at the level that you’re expecting. I’m not so sure it’s dispute settlement. I’m not even sure any particular disputes have arisen in the context of free trade agreements for India. I would say it’s more just the design of the agreements themselves.

Now, the Japanese agreement, the Korean agreement could be expected to yield more. Some of the more recent agreements that have been signed under the Modi government—including with Australia, with the UAE, with the Europeans and the agreements that are being contemplated—now you’re talking about the possible trade agreement with the European Union, possible trade agreement with the United States and so on, these at some level appear more ambitious. I’m fairly certain that if there was progress made with the European Union or England, so Great Britain, or with the U.S., then those agreements will demand more from India in terms of the liberalization that’s undertaken.

These are in the future, they’ve not happened yet. How India would react to actual negotiations for greater openness or greater integration of different forms is yet to be seen.

Viner: Trade Creation and Trade Diversion

RAJAGOPALAN: A lot of your academic work is about the political economy, lobbying, rent-seeking within a particular country, and how that impacts either trade liberalization more generally, or in particular preferential trade agreements. If you look at the history of thought on this, you have Jacob Viner, who’s talking about you can have these preferential trade agreements, which can end in one or two ways. They can be trade-creating, which means basically you start with these preferential trade agreements, and that’s the gateway to then greater liberalization. Because if you can reduce tariffs for Australia and not end up in some horrible place, then you can also do it for New Zealand. Then before you know it, you start doing it for European Union and the United States, and so on.

Or they could be trade-diverting, which is we’ve now signed this preferential trade agreement, and we’ve signaled that we’re willing to pick this low-hanging or almost irrelevant fruit. That’s enough of a signal plus effort. That actually strengthens lobbies such that we will not sign these big multilateral agreements, which are going to push us toward further liberalization.

First of all, is what I’m saying accurate in terms of the broad brush strokes? I know different people call it different things. Bhagwati and Panagariya call it stumbling blocks versus building blocks. You have trade creation versus trade diversion. Is this roughly the realm we are operating in when we think about political economy of trade?

KRISHNA: I think that’s about right, and you’ve summarized it very well. Let me just add one or two things, and this is out of a tribute to Jacob Viner, really, which is that—and I recommend his writings on this.

RAJAGOPALAN: Oh, they’re very readable.

KRISHNA: 1950 “The Customs Union Issue,” anticipating the European Union and so forth, brilliantly written. The one point that he made, along the lines of what you just said, which is, while these trade agreements are sometimes described as free trade agreements, and they might suggest therefore to whoever’s observing them as if this is a very substantial liberalization that’s taken place, in the context of trade diversion, they could actually be resulting in higher levels of tariffs, higher levels of protection. He notes this in his writing, saying, how is it that people who are normally opposed to free trade—business leaders and so on, business lobbies who are opposed to free trade—suddenly seem to get quite excited about trade agreements? It’s because they’re getting preferential access in somebody else’s market, which is a form of protected access. If anything, it’s protection.

Viner was really the first one to point out, at least in an analytically sharp and clear way, that what sounds like a free trade area, what’s being described as a free trade area, it could actually result in higher protection, not less protection, and hurt the country because of that fact. There’s more protection, by standard arguments in a national trade, more protection could lead you to a worsening.

That brings you to this question of if it’s hurting a country and there’s trade diversion taking place because there’s trade flows only taking place because of the preferential access that you’ve received in your partner country’s market, why would a country sign a deal like that, that hurts it? Then that brings us into the realm of political economy, and that brings us to appreciate the fact that, as we’ve just said, and as you pointed out, that business lobbies are interested in that access because it gives them additional protection. And if you have a situation where symmetrically the other country’s business lobbies want that protected access in your market, and you’re willing to provide that to them in exchange for the access that your lobbies are getting in their market, then that exchange could work.

RAJAGOPALAN: It’s sort of the narrow corridor of trade—

KRISHNA: Narrow corridor of trade.

RAJAGOPALAN: —to borrow a different phrase from a different development.

KRISHNA: That’s correct. It opens up this opportunity for vested interests, that in fact want more protection to sign what appears to be an agreement that’s giving less protection and seems liberal, whereas it’s the opposite of that. That’s one way of thinking about the political economy of preferential agreement.

RAJAGOPALAN: Actually, the way you phrase Jacob Viner is very nice because in your QJE paper, that’s the continuation point. In fact, you flip the corollary. You’re like, if people are not that excited about global trade, then they’re not going to be that excited about preferential trade that can eventually become the building block to global trade. Therefore, also your preferential trade agreements are going to be relatively toothless, if I understood the argument correctly, right?

KRISHNA: That’s right. The idea there in that QJE paper was that you could have this type of exchange of market access to vested interest that gives you this agreement, in a context in which although multilateral free trade is welfare improving, it is not actually chosen because it would hurt these lobbies rather than help them.

A different way of seeing this is to say if you already got into the preferential agreement—that was the order of the logic in that paper—let’s say you got into that preferential agreement and now you were given the opportunity, the country was given the opportunity of multilateral free trade. Why would they take it? They needn’t because now they’re enjoying preferential access in each other’s markets, which is all going to be diluted if you get back to multilateral liberalization.

In that sense, the paper is arguing that while some might imagine bilateral agreements as being some sort of a stepping stone toward multilateral free trade—economics’ utopia, multilateral free trade—that it need not be that easy path that countries might get stuck in these different areas because they don’t want to give up the preferential access that they already enjoy. By which I mean business interests don’t want to give those up.

RAJAGOPALAN: If I read it correctly, it’s almost harder if you have a preferential access versus if you’re starting from zero. Because now you’re giving up a larger pool of rents because you have access to two markets effectively, right?

KRISHNA: That’s correct.

RAJAGOPALAN: It’s almost as if it’s worse to have a preferential trade agreement, which protects rents in a very big way rather than have no trade agreement and start from scratch.

KRISHNA: That’s right. The paper was arguing really that it’s conceivable, it’s not always going to be the case, but you could have circumstances, in other words, theoretically where a multilateral agreement that would’ve gone through, that both sides could have agreed to, now is no longer feasible because of the fact that the business interests have gotten something higher through better rents, greater rents. Profits that they’re enjoying because of this preferred access, and in which they’ve shut out the rest of the world.

They would be reluctant to dilute that and get back to an agreement that they might have accepted in the first place. This was the other sort of, in a sense, criticism of the path that the world was taking at that time. Which is to say if you open this up as an option, meaning bilateral agreements as an option, you might miss out on multilateral liberalization that could still take place. That would no longer take place once you’ve gone down this bilateral path. It was an illustration of political economy reasons for why preferential agreements could become stumbling blocks toward multilateral liberalization.

More Optimistic View of Equilibrium

RAJAGOPALAN: Here I have a question. This is not a paper written by you, it’s by one of your co-authors. This is Devashish Mitra. This is his late ’90s AER paper where he’s also talking about this endogenous rent-seeking and lobbying. That’s a slightly more optimistic conclusion more generally for trade where he looks at, there’s one possibility where different interest groups may cancel each other out, and your equilibrium might be more free trade, lower levels of protectionism. As opposed to a system where there are one or two very powerful lobbies who just completely capture it.

My question is, is that compatible with what you’re saying? Why doesn’t that happen for preferential trade agreements, if that’s a possibility for more general multilateral trade agreements? Is it the narrowness of the PTAs, or is something else going on?

KRISHNA: Yes. First of all, that’s a brilliant paper by a good friend and co-author, Devashish Mitra.

RAJAGOPALAN: So elegantly written. I told him it’s one of the most elegant papers I have read.

KRISHNA: Yes, beautiful paper. He was working on it when he was a graduate student. We were graduate students together. He was my junior, behind me at Columbia, but I was well aware of his work at that time, and it’s a very nice paper. Now, that paper if you take it out of the context of preferential agreements altogether, just simply thinking about trade policy formation in a country. That paper is making an argument that you could consider a world in which somehow the barrier for lobby formation was very low.

Everybody is able to organize, everybody is able to form a lobby, everybody is able to represent themselves. You get this idea that was anticipated in the early political economy work of Gary Becker in the 1970s where Becker’s point was, listen, you can get the optimal outcome of welfare maximization when there are no lobbies. Nobody is pushing the government to do something special for them or when everybody is organized because they cancel each other out.

This complete organization of the system and the government, once again, is doing the same thing. In the middle, you don’t quite know, it really depends on what these lobbies are. Devashish’s paper was a bit about, at least one of the ways in which I’ve absorbed the messages of the paper, is under what circumstances can you get this endogenous organization of lobbies, if you will. How do people organize themselves in a way where you get to that utopia?

Very clearly the costs of organization were very low. Now, you’re getting to the work of Mancur Olson and the collective action problems and various other things. All very important in the context of endogenous lobbying, that if the cost of organization into lobbies were super low, consumers are organized, producers are organized, intermediate users are organized. Everybody is organized, then you get free trade.

At a theoretical level, yes, that works in the context of a domestic economy, meaning absent any discussion of bilateral versus multilateral-type issues, just thinking about unilateral liberalization even. It would work and be relevant in the context of bilateral agreements. As a practical matter, the reality is that getting organized as a lobby is not that easy. Just directly from the fact that you don’t have free trade, already you could conclude that that kind of ideal equilibrium that was envisaged by Becker or elaborated on by Devashish is actually not taking place.

RAJAGOPALAN: Yes. I guess what I’m asking is not that the utopia doesn’t exist, but is there some additional problem in preferential trade agreements which prevents people from canceling each other out?

KRISHNA: I’m not sure that the context of preferential agreements creates any additional preference.

RAJAGOPALAN: It’s just whoever is the existing powerful lobby are the ones who are going to be driving you toward toothless preferential trade agreement or something that really protects rents. It’s one or the other.

KRISHNA: Protects rents. It could be the case—Devashish and I have worked actually building on his paper, the one that you mentioned in the AER—we have some additional work about kind of endogenous entry of lobbies. Let’s say you have a trade agreement of a certain sort, is the structure of lobbies going to stay the same? Probably not, right?

RAJAGOPALAN: Yes.

KRISHNA: Anticipating an agreement, or as a consequence of the agreement, it is conceivable that there are new entrants into this political space and political economy space. Not necessarily up to the point where they fully cancel each other out and bring you to a point of free trade. That’s conceivable, theoretically. In fact we’ve worked on that, again, theoretically. But as a practical matter, I don’t know.

RAJAGOPALAN: We haven’t seen that.

KRISHNA: In the context of preferential agreements, there may be some difference, in fact, in a given agreement of how those lobbies might get organized, anticipating of preferential trade agreement or how they might get organized differently, anticipating a multilateral agreement. I’m not going to say that the two situations in practice would be identical for any given country or given agreement that’s being imagined.

Conceptually, I don’t think there’s any real difference. Meaning, we needn’t see preferential agreements differently in terms of this particular idea of the incentives to get organized. What happens if everybody is organized? I think it’s probably, unless I’m missing something here, probably a similar context.

Foreign Lobbies in Domestic Markets

RAJAGOPALAN: You know here, it’s not all bad news because there’s the other paper you have, this 2006 REStat paper. I’m looking up my notes because I want to make sure I name your co-authors right. This is with Gawande and Robbins. There you actually show that foreign lobbies, which won entry into a particular domestic market, may actually have a canceling-out effect. It may not be one on one, literally, those who are importing and exporting cancel each other out. This is, again, the broad idea of typically, all these endogenous rent-seeking models, they come from Grossman and Helpman. That’s the foundational model.

The idea is, you’re going to have this government which wants to maximize certain things, and one of them is political contributions. Here it’s now a question of domestically enabled political contributions versus those that come from foreign firms. You wrote this paper when we were at the peak of the globalization and multinational, multicontinental firms. Is that logic a little bit different from what’s going on within a country? Why does the foreign firm work differently, aside from just plain different interests?

KRISHNA: Let me start by saying that this idea that a foreign lobby—let’s say you’re an exporter, let’s say you’re Toyota exporting to the U.S. market. Toyota’s interest would be the opposite of a U.S. car manufacturer. So, let’s say, Ford or Chrysler or whoever. They would like to have higher tariffs, and Toyota would like to have lower tariffs because they’re the ones exporting it to the country. Let’s say they’re operating in Japan, they’re exporting the car from Japan to the U.S.

Now, what’s interesting is the fact that Toyota would want lower tariffs, and those lower tariffs would be beneficial to the U.S. consumer, makes any potential effort by Toyota valuable. Although they’re not necessarily seeking to improve the welfare necessarily of the U.S. consumer, their self-interest in terms of wanting to lower tariffs, by coincidence, works to improve the welfare of the U.S. consumer.

This particular point, by the way, about foreign lobbies and their interests being coincident with that of the U.S. consumer was anticipated a long, long time ago by Cordell Hull, who had written about this, and who had said the idea of foreign lobbies operating the U.S. and so on, this is a complicated space, and there are multiple different things to think about. But at least in this narrow economic context, the fact of their coincidence of interest with the U.S. consumer, needs to be appreciated. That is what that paper was looking at.

It was a slightly cheeky take on what was going on with respect to trade liberalization and competing interests of domestic and foreign lobbies. What we did was to explore, using U.S. data on registered foreign agents who work on behalf of foreign interests, to see if those industries where there was such a representation or presence in the U.S. enjoyed lower tariffs. We did find that there were some effects in terms of trade barriers overall, but they were, in fact, lower barriers compared to those that didn’t have these lobbies operating.

RAJAGOPALAN: Do you expect 20 years later for this result to be a little bit different? I ask this in two specific contexts: One context is, as you said, this is also around the time, like the data that you’re looking at, at that time, the United States hadn’t completely abandoned this idea of, we need to move toward a global bloc that is liberalized and will trade with each other on somewhat of a level playing field. That idea has completely disappeared. One expectation is you would find the same, maybe more, foreign agents lobbying, but it may not result in a systematic reduction in tariffs. It’ll find its way in some other ways, some other subsidy, tech transfer, pattern, something else that’s going on. So that’s option one.

Option two is, we are now in a very different geopolitics from what we were in in 2004. Now a Chinese agent lobbying the United States, there’ll be this huge meltdown, and we’re not that far from Washington, D.C., where this meltdown will actually happen, or maybe on Twitter. The world will end if foreign agencies are now caught lobbying for what might be beneficial to the domestic consumer, but that’s absolutely not how it would be viewed.

KRISHNA: Yes, there’s multiple pieces to that. Let me say a couple things. One is that it was never the case that foreign agents could, say, make political contributions. There was a particular role that the foreign agents were Americans. They’re representing foreign interests, and they were free to go discuss these issues with Congress and so on so forth, but they could not make political contributions. What’s happened now is that this entire game of political contributions because of—

RAJAGOPALAN: Has changed.

KRISHNA: —the Supreme Court ruling under President Obama, during his tenure as president, the limits that used to exist before in terms of political contributions and how much could—those limits were shockingly small back in the day. A company could maybe make a contribution of $5,000, if I remember correctly, was the limit. All those limits have now been taken off. The politics, that space of who can lobby and how much money they could put behind it, as I understand, it has completely changed, on the one hand.

That already is a very dramatic change relative to the time that we wrote that paper, relative to 20 years ago. The other big piece is that geopolitics has changed very dramatically. There’s far greater sensitivity and far greater rate at which information gets transmitted and so on. There are still foreign agents, they’re all registered, everybody can go online and look up who these agents are.

RAJAGOPALAN: You could write the same paper again—

KRISHNA: You could write the same paper again.

RAJAGOPALAN: —but you’re not sure how much it will capture.

KRISHNA: There’s extraordinary sensitivity, I would think there will be more in the future, to any attempt by lobbies to actively lobby in areas that are somehow seen as challenging, important, from a U.S. and a national interest standpoint, strategic security interests and so forth. That would be very difficult. I think you’re completely right in that the general climate has changed.

I’m not sure what actually is going on with respect to scrutiny of these agents, whether they’re subject to greater scrutiny or less, and so on. I would imagine that type of operation, to take your example of Chinese agents representing Chinese businesses, lobbying on behalf—

RAJAGOPALAN: This may be for T-shirts being sold at Walmart. It’s not something insidious, necessarily.

KRISHNA: Necessarily. That probably, on the margin, I would expect has become harder, and maybe will become harder still.

RAJAGOPALAN: How would you view this for India? This paper was not written for India, and it’s near impossible to trace this for India. We don’t have this kind of transparency in political contributions, nor do we have this kind of registration of agents. Which way would your hunch go? When I just look at the foreign firms that have recently found success, they don’t have a domestic big counterpart making those same goods.

For instance, Apple has been the latest success, managed to create lots of jobs, it’s got part of this PLI scheme. They’re still struggling to get lower tariffs the way Vietnam and other countries offer them, but there’s some movement, at least the needle is moving somewhere. India didn’t have a giant Samsung, like the Chinese and the South Korea. It didn’t have this big behemoth that was already in that space which could have organized and become protectionist.

On the other hand, I feel like there’s maybe no room for something like auto manufacturing. That is completely sealed shut. 125% tariffs is never going to go away and hasn’t gone away in my lifetime situation. How does this play out in India? Is it that we’re going to see movement in places where India doesn’t have this big foothold or organized lobby?

KRISHNA: Yes. I think the Indian political economy is deeply rooted, it’s very well established. There are powerful players who, over decades, have known how to operate in New Delhi. If the question is, would it be easy for a foreign entrant to overcome the domestic politics and move policy in a direction that may favor the Indian consumer, along the lines of the argument in this paper, and favor them, this foreign supplier as well, my guess is that it would be challenging.

I think, as you were saying, it would be more challenging the more powerful the domestic interests are. Certain sectors you’re going to see this, whether you think about synthetics, fabrics that are used to make apparel, you think about auto and so on so forth. There are certainly areas where it’s very obvious that the Indian political economy itself is quite strong and equilibrium is strong, and to shake that or move that would be quite challenging.

Where there are new areas, again, Apple cellphones and so forth, or maybe areas in which the domestic players are not necessarily so deeply entrenched or so strong, could there be some room there? Possibly, yes. For me, the greater part, in terms of Indian liberalization—of course, there’s a political economy of it, and then there’s a separate issue of political will. Maybe even broader than that is this idea of what is it that we actually want to do? Do we want to go in a more liberal direction? Do we believe in greater global integration? Do we believe that unilateral liberalization that we could be undertaking will help us? Or is it somehow being seen as some great concession that’s being done under the threat of the WTO or somebody else, or some external player?

RAJAGOPALAN: Some foreign influence.

KRISHNA: The foreign influence. These are the bigger questions. On the margin, I think over the last many years, is certainly a greater interest, it seems to me, at least, compared to decades past in bringing in foreign capital, bringing in foreign manufacturers and so forth. Undoubtedly, there may be some things done to accommodate that. And given that India needs all of this investment, maybe it’s a wise thing to do. But there’s a broader, bigger question of how we see our future, how we see our own growth and what the role is of international trade policy within it.

Are we going to be pursuing self-reliance? Is that just a slogan? Are we actually interested in self-reliance of a certain type in certain goods that have strategic interest, and everything else is we want to have market competition? These are the questions that we need to be thinking carefully about, I think, and that are going to determine our future.

Just pick a number

RAJAGOPALAN: When it comes to India and tariffs, maybe I’m being too simplistic, I’m just like, pick a number, keep it low, keep it uniform. India’s economy now is too deep. This is not the late ’80s where we can tinker and tweak and we know the total number of goods being produced. Now it’s, we don’t know this input-output matrix anymore. Any lever we press and start increasing to protect one kind of industry, it’s going to affect something else, and it’s going to impact imports for that industry, which is going to make that industry’s export competitiveness worse. One, do you think I’m being too simplistic? Is the intuition, right? If yes, what’s the number you would pick? Then let’s just do that and be done with this. Let’s put you out of work.

KRISHNA: No. Listen, I mean, on the one hand, I think the argument that you’re making for a uniform tariff, it’s an excellent argument at a theoretical level. It has the benefit of simplicity. It has the benefit that once you set that and it’s locked in, then maybe it removes all the political lobbying from it because nothing’s going to change the tariff. It has the additional benefit in a narrower input-output sense for the economy that it doesn’t disincentivize certain types of producers because you do have the problem today, and the government is well aware of it, the Indian government, of tariff inversion, where the intermediate inputs are being taxed at a higher rate than our final goods. That disincentivizes value addition within the economy. With uniform tariff, that’s gone as well. Lots of reasons for why that would be great.

RAJAGOPALAN: I would add one more, which is, for foreign investment, one of the most important things is certainty. Forget tariff inversion. One part is tariff inversion is a really big thing hanging over their heads. But just to be able to estimate what your costs look like for imports and your manufacturing process over the next eight, 10 years, that’s what determines if big foreign capital investment flows in. And it hasn’t. I think that’s another big piece of this puzzle. That’s actually the main reason, other than lobbying, that I would vote for that because foreign investment is down.

KRISHNA: Without a doubt. The predictability of tariffs, both for Indian suppliers who rely on intermediates, for foreign exporters, for any foreign investment that’s taking place in India that can rely on intermediate inputs, everybody. That would be great. If you asked me to pick a number, I would say maybe a tariff revenue-neutral number. We have some reliance in India still on tariff revenues or trade customs duties to finance the budget.

Maybe that number is in the two to two and a half trillion rupees range. At some point, I’d looked into this, and maybe the number of a uniform tariff that would not hurt the revenues that come from trade taxes might be in the 7% range. Seven works.

RAJAGOPALAN: Magic number. Let’s do it.

KRISHNA: Five, seven, 10, whatever these numbers are, but let’s say it’s 7%. That would work. As a practical matter, there seems to be very little appetite for that, just as there’s been very little appetite, even in the context of just the Indian domestic economy itself. If you look at GST, you could make a similar kind of an argument. Why don’t we have a uniform tax across goods and so on and so forth? You don’t. You have multiple tiers and the tiers are getting more complicated, and there’s always pushes and pulls by different players.

RAJAGOPALAN: Yes. We have seven nonzero rate and 53 cesses. I’m aghast at how we have managed this.

KRISHNA: For income tax, for that matter, I’m sure you’ve heard these arguments, and I have, and we’ve also had meetings where somebody could make an argument for a flat tax, 20% and just keep it simple, no exemption.

RAJAGOPALAN: That’s a little bit different because here, there’s an intermediate goods problem with both GST and with tariffs, which is not exactly there for income tax. Even if you want to have progressive rates or a slightly more complicated thing for income tax, that may be less distortionary overall in the economy, I think, relative to what GST and tariffs do, and also the number of income taxpayers we have is so tiny that it’s almost irrelevant. In principle, I totally agree with you.

KRISHNA: Right. The point of me comparing those two or bringing them both into the argument was just a simplicity point, right?

RAJAGOPALAN: Absolutely.

KRISHNA: Just a very simple tax, no exceptions, no exemptions for anybody. An economist’s delight. As a practical matter, I’m not sure.

RAJAGOPALAN: Is that because the geopolitics don’t allow for it, or is it because bureaucrats don’t want to let go of power? Which rent-seeking model is driving the move away from simplicity?

KRISHNA: I’m not so sure about geopolitics because even when our geopolitics were simpler, we had complicated tax structures.

RAJAGOPALAN: That’s true.

KRISHNA: Complicated tariff structures, on the one hand. Bureaucracy, how alien this idea might be, just how strange it would appear, to suddenly say everybody can go home because we’ve just decided on one tax, so there’s nothing for anybody to do. Whether there’s bureaucratic rents that are being earned by people who can decide on these things and adjudicate these issues, maybe.

Equally, interest groups who want something better for themselves, more protection, something else, less protection on their intermediate inputs, and so on. It’s a combination of all of those. The other reality is that you’re dealing with tax systems that are already in place and already quite complex. It’s a modification of that if this is all being designed from scratch, moment of birth or whatever.

RAJAGOPALAN: We’ve inherited the whole amount of—

KRISHNA: Complications and moving taxes and reducing the benefits of somebody would be challenging, I would think.

The Impact of Trade Liberalization

RAJAGOPALAN: There’s one aspect of trade liberalization, which you’ve written about. This is actually a very nice paper that you and Devashish did long, long time ago. I think you need to redo this JDE paper where you’re looking at the impact of trade liberalization, and this is early on. It’s just a few years, a handful of years after trade liberalization.

What you find is that it increases productivity and economic growth. What is the mechanism through which this is happening? Is it just increased competition? Is it something else? Is it learning by doing because of increased competition? Is it tech transfers? What is the mechanism? That is, it’s crystal clear that trade liberalization increases productivity and economic growth, but why?

KRISHNA: Right. Actually, funny little anecdote here, which is that it’s interesting you asked that question because this paper with Devashish, which came out in the Journal of Development Economics, we started working on it shortly after the initial trade reforms. We were very excited about the impact that it would have and it became one of the chapters of my dissertation.

For my Ph.D. defense actually, Jagdish Bhagwati, who was one of my advisers at the defense, asked me exactly this question. He said, “Listen, as much as I’m thrilled by the finding that productivity is improved, I have to ask”— because the time period, because my Ph.D. defense would have been in 1995. It’s just been a few years, really. “Why would you expect there to be this type of productivity improvement? What do you think is happening?” The truth of the matter is that—

RAJAGOPALAN: He probably asked it differently than I did.

KRISHNA: Very close, actually. Very close.

He did ask the question and my answer then to him—I guess it’s the same today—is that by that point, it would not have been reasonable to have expected there to have been technological change. This new technology, which should happen later, shortly after, so as you got mid-1990s to the 2000s and so on, all of these things happened. Firms, joint ventures, Indian domestic affairs, joint ventures, various companies abroad, whether you’re looking at the automobile sector, motorcycles, whatever, there’s a whole range of products where technology was actually brought in, improved and so on. This was too early for that.

By that stage, the only thing we could really have expected was that these were the competitive effects. Standard X-efficiency argument for a productivity gain, you know that the liberalization has happened. You know that more is expected. The only way you’re going to stay in business is if you somehow cleaned up your act. Our guess—because we couldn’t get any further with the data that we had in terms of why is it taking place. At that time, we really didn’t have much of an opportunity to survey firms to find out what was going on either. This was all statistical.

You have data on outputs, you have data on inputs, you have to figure out productivity. My guess would be to the extent that there were these productivity increases that we found that these were probably driven by competition X-efficiency type of arguments. Maybe there were a few companies that brought in technology very rapidly, immediately after ’91, quickly enough that we were able to see these increases in productivity.

RAJAGOPALAN: That would’ve mostly been IT.

KRISHNA: It would have been mostly IT, maybe a bit in manufacturing, but I don’t know enough, frankly.

RAJAGOPALAN: Not across the board.

KRISHNA: Certainly not across the board.

RAJAGOPALAN: Tariffs are a tax on the domestic consumer and the foreign producer. The License Permit Raj is a tax on the domestic consumer and the domestic producer. Is there a way to parse out the productivity gain that came from dismantling the licensing system from the tariff system?

Because if it’s competition, then foreign competition is really what’s at play here, which is totally understandable because India limited the number of players in every sector because competition was considered wasteful. If it is liberalizing the domestic markets such that now Bajaj can have improvements and efficiency and it can make more scooters and people can get to work faster and their productivity at their firm improves—is it easy or impossible to distinguish this? If it is the same time period, that’s fine.

KRISHNA: That is the problem I think because the sets of things that were taking place were all happening simultaneously. You had the License Permit Raj and quantitative constraints being taken down domestically, on the international margin, about the same time that you had tariff changes.

RAJAGOPALAN: The tariffs come down.

KRISHNA: Tax structures being changed within the country as well. It would be very hard, but maybe not beyond the scope of my imagination at least, that some of our younger colleagues in the field who are laser focused on this idea of identification and maybe there are particular sectors. There’s a question that may be worthwhile getting back to. Were there sectors that were not liberalized despite the License Raj going to dismantling of that regime, but which experienced some tariff reform and others which had both and maybe yet others that had nothing?

Maybe there’s some comparisons there that are still possible to do. But broadly—and once again, you’ve raised this question earlier and we’ve talked about—the intermediate goods: Once you recognize that, then it becomes even more complicated. Because even if your sector in terms of the final goods was not affected by policy change, certainly the intermediates that we’re using would’ve been affected or most likely they would’ve been affected. That could be a little challenging to do in terms of separating these out.

RAJAGOPALAN: I think the core of your argument still doesn’t change in the sense that it is competition. Is it domestically driven competition or foreign competition that’s driving the actual outcome? I’m not saying it’s irrelevant but I’m saying it doesn’t matter for the immediate mechanism at hand here. Because both de-licensing and reducing tariffs have this impact of unleashing competition because now you could compete on the margin of price and quantity which you couldn’t do before, and therefore quality.

KRISHNA: That’s right. I will say two things. One is that the paper with Devashish in the JDE looks simultaneously at both things, right? It’s looking at markups. A monopoly’s markups, so to speak, price over marginal cost. And it’s looking at productivity.

RAJAGOPALAN: You see the monopoly markups going down.

KRISHNA: Yes. For me, that’s in a sense the stronger result of the paper. And that you would expect to take place relatively quickly because as soon as you take down import barriers—

RAJAGOPALAN: Price competition is the most immediate visible thing.

KRISHNA:  to happen. We had very consistent drops in markups across the board alongside this productivity gain. Competition being a central factor that is a 100% for sure.

The second quick thing I’d say is that I do remember—and this is more in the nature of an anecdote—was shortly after liberalization maybe around ’94 probably, ’95 actually, there was a conference at Harvard that took place. And I think Ashutosh Varshney might have organized this. A very big conference. Lots of people came in, including business folks from India and so on. Montek Singh was there and Mr. Rahul Bajaj was there.

I remember Mr. Bajaj standing up and saying, “I should just mention this that this whole time because I was constrained by the License Raj in terms of how much I could produce, how many scooters I could produce, I didn’t have to have an advertising budget because the shortages were so many years now. Now I have to worry about all these things. These are good worries for me because now we can expand production, compete and so on and so forth.” Along the lines of what you’re saying, right? It just created a different dynamic, a rather different competitive culture within the Indian private sector.

RAJAGOPALAN: Yes. The Bajaj scooter anecdote is one of my favorites because they had these quantity controls and if you needed to expand, you needed an expansion permit that was given both from the ministry of industry, and then you needed of course foreign exchange for the things that you need to import. I think they got denied that once or multiple times. By the mid to late ’70s there was almost a seven-to-10-year wait for a Bajaj scooter, which meant that the secondhand Bajaj scooter price was far above marked up than a firsthand Bajaj scooter price. Because the secondhand, you could get immediately. The firsthand, you have to wait for four or five years.

KRISHNA: That’s right.

RAJAGOPALAN: Then of course the joke is it became a major dowry demand because that’s the one time you can really squeeze the parents of the bride. If a daughter was born, they’d say book your Bajaj scooter now. By the time she’s a marriageable age, you’ll get your allotment.

KRISHNA: Exactly. That was exactly what I was thinking of, which was the dowry stories of those days.

RAJAGOPALAN: Godrej almirah and Bajaj scooters and Khaitan table fans—the whole thing is bananas.

KRISHNA: The whole thing was bananas. There were all these brokers, intermediaries who if you really needed that scooter—

RAJAGOPALAN: They could get it done.

KRISHNA: —if this guy was not going to marry your daughter because of the scooter, then they could get it to you for a price.

Labor Elasticity in Relation to Trade Openness

RAJAGOPALAN: I have a couple of questions about some of your work on trade liberalization but now I want to talk about labor elasticity. I think it’s very important in what’s happening today in the United States and across the world. You, along with Devashish and Sajid Chinoy, have tested whether increase in trade openness and liberalization changes liberal elasticities in the home country. You looked at it for Turkey. And then Devashish and Ramaswamy and Rana Hasan and those guys looked at it for India.

Basically, the relationship is quite clear. In fact, in Devashish and Hasan and Ramaswamy paper they even find it at the subnational level. Is that why there has been this move away from trade openness? I want to link it to what’s happening in the U.S. If you look at the last few elections, it is really the labor elasticity and shock problem which has driven the move away from trade openness.

This is the famous swing states and the Rust Belt states which it may feel like it’s a Trump phenomenon but it’s been going on for almost 15, 20 years. That’s the group which is so vocally against trade liberalization. Now we’re going to look inwards, we’re going to increase tariffs, we’re going to bring back industries. We are going to start with industrial targeting. The whole thing. One, did you get the sense when you were writing the paper on labor elasticities? Do you see that connection now? Is that what’s driving the move against trade liberalization? Did we just underestimate it all along?

KRISHNA: I would say that at the time that we wrote the paper and what motivated us to write that paper was certainly the awareness. It was already being discussed right here in Washington, D.C. that what is happening over time or what had happened over time was that American workers—it was mostly being discussed in the context of the U.S.—that American workers had lost their power in the labor market. That would happen if labor demand elasticities were higher.

If you plugged in some higher labor demand elasticity into some standard bargaining model workers and firms and so forth, then greater elasticities may mean less power for the worker. That’s what motivated this. At that time the questions were, is it because capital can threaten to move away? Is it because when you have a greater international trade you have greater elasticity of product demand? Now we’re getting very nerdy here, in a sense, but greater elasticity of product demand is going to translate into greater elasticity of labor demand and so on and so forth.

RAJAGOPALAN: Which is just basically responsiveness of either the particular quantity of the product or the labor and the wages to the changes in price or wage.

KRISHNA: Broadly speaking, responsiveness to global opportunity.

RAJAGOPALAN: Exactly.

KRISHNA: There are other products that you could buy. There are other places where you could produce. This is all going to increase the elasticity.

RAJAGOPALAN: Therefore, volatility.

KRISHNA: That’s right. Result finally in lesser power to workers. Is that what’s happened in the U.S.? Maybe, to some extent. Is technological change what’s happened in the U.S.? Maybe, also, to some extent. This is a longstanding debate in the field, and you’re aware of it. How much of it was technology? How much of it was declining power of workers? How much of it was China entering the WTO and so forth?

All of these things have happened simultaneously. Without a doubt, the U.S. was already deindustrializing for a very long period of time. One should be aware of that. Sometimes people think of this as purely a 1990s phenomenon, which it’s not. Maybe things were accelerated a little bit more in terms of the deindustrialization process. For me, even that is a bit of a question mark. But I’m persuaded by various colleagues in the field, multiple papers that show that this deindustrialization process, getting out of manufacturing increased a bit.

This labor demand elasticity issue is an important one. Nonetheless, I don’t know that if one looked at, say, India, with what’s going on in India with respect to shifts and out of agriculture and other sectors, that that story’s the most relevant one at the moment. For the U.S., I certainly see its relevance. There’s a sense certainly by looking at incomes and union workers and what’s happened to wages and unionization even over time, that that story has salience.

And it was certainly motivated by the kinds of things that we have eventually seen. This was already being discussed, like I said, by various people in Washington at the time. The argument was already being made. That the kinds of things that are happening, it was actually largely started, at least for me, I was mostly exposed to this argument in the context of the NAFTA debates, pre-1995.

RAJAGOPALAN: This is the Clinton era—

KRISHNA: Clinton era.

RAJAGOPALAN: —when it starts. That’s when the unions also start moving away from the Dems slowly.

KRISHNA: That’s right. That’s when various economists, think tanks in Washington were raising these questions about what’s going to happen to workers. Is their power going to be decreasing somehow, or is going to decrease somehow? I think one way or the other, just the stagnation of real wages for low-skilled workers in the U.S. tells us that something has happened. Whether all of that can be loaded just onto international trade, whether it’s outsourcing, whether it’s technology, whether it’s WTO, China, et cetera, these are active research questions.

RAJAGOPALAN: My intuition, and I have no way of even figuring out how one would test this empirically, is it’s the degree of specialization. In the last 40, 50 years, that has just taken off in a way that because things are so hyperspecialized, in some sense, it increases the overall importance of human capital within the production process. On the other hand, it’s going to reduce the power of the generic wage earner, or the generic laborer or the factory worker. Of course, that’s hastened by trade liberalization, or technology or so on. I think when we dig through this, at the bottom, we will find it’s just specialization.

KRISHNA: Maybe a different way of putting it—maybe it’s exactly what you’re saying—but in a world that’s fully globalized, where things could be produced anywhere, and when some other country has the same technology that you do, it’s impossible to sustain or be very challenging to sustain a wage for a given worker that’s much higher than the wage for a similar worker elsewhere.

The minute you ended up with technology flows on the one hand, China entering the WTO and getting all the technology that was previously here, or having the same technology as was here, how can there be a big wage difference between an American worker and China? That’s not going to happen. At the same time, automation, if you’re not super terribly skilled and you’re not doing something complex that a machine would find difficult to replicate, and your job can be easily automated away, then how can you sustain high wages for long periods of time? On top of it, you have these outsourcing issues and international trade issues, which as we’ve just discussed, lead to these labor demand elasticities rising and so forth. Overall, it’s a multidimensional pressure, all incentivizing, in some sense, higher human capital, or supporting higher human capital workers relative to those that have less.

Predicting the Near Future Impact of U.S. Trade Tariffs

RAJAGOPALAN: I know you’re going to have more work to do with the Trump 2.0 threat of tariffs, so that’s good for you. In general, what are you expecting, and is the overall narrative of increasing tariffs across the board for anyone who doesn’t get in line—is this even feasible? Is this all just crazy? Because the old universe is, these things take lots of negotiation. People like you, bureaucrats, diplomats are involved. They produce these 200-word agreements, which no one can read. Now everyone’s saying, “No, Trump is going to tweet this stuff.” I’m not entirely sure how trade policy will be made in the future. Is this feasible?

KRISHNA: It’s feasible. What is infeasible, just to step back, and let’s take the man at his word. What are the things that he’s actually said? The things that he’s actually said in the context of the campaign. Of course, this might change, but what he said in the context of the campaign is multiple things: “One, I’m going to have a tariff on everybody that’s in the 10% to 20% range. Two, I’m going to only give you the tariff that you give me on a good-by-good basis.” This is completely getting away from GATT Article 1, Article whatever—

RAJAGOPALAN: Oh, boy.

KRISHNA: Over. “Three, I’m going to have another 60% tariff on China. Four, I’m going to replace the income taxes with trade taxes.” That last bit is the one piece that is completely infeasible because the—

RAJAGOPALAN: Yes, the numbers don’t add up.

KRISHNA: The numbers don’t add up. Given the volume of trade that the U.S. has, maybe I think Maurice Obstfeld and Kimberly Clausing at, the Peterson Institute had a nice calculation that the tariff revenue maximizing rate is probably around 50%. It’s going to only give you a fraction of what you get by income taxes. The idea of replacing income tax with that.

RAJAGOPALAN: It’s cuckoo.

KRISHNA: That’s cuckoo. But can you see some movement in that direction? Certainly, you can. You might not get all the way to replacement, and even that would be very dramatic. If you started to think about replacing 30% of income tax with trade taxes—

RAJAGOPALAN: Wow.

KRISHNA: —that would be very large. Even a country like India that’s far more reliant on trade taxes than the U.S., maybe gets 3%, 4%, something like that of overall government revenues from trade taxes. The U.S. is much lower, maybe 1% to 2%. That would be a very dramatic shift, and I don’t expect it to be fully realized for sure, because it can’t.

RAJAGOPALAN: Also the downstream consequences, there will be the supply-side inflation as they call it, which will just be bonkers.

KRISHNA: All kinds of things.

RAJAGOPALAN: Everything from your toaster to the mug to this mic, everything is going to be more expensive. Nobody will like it.

KRISHNA: For sure. With taxes that are high enough to offset income tax revenues. Now, whether you can do a 10% to 20%, and whether you can put tariffs on China that are 60%, you can break that down, I think at least into two or three pieces. One is, as president, given that Congress in the U.S. actually has the authority over tariff making—it’s not the president, necessarily—how can Trump do what he’s threatening to do? Will he be opposed by Congress? Maybe even that’s not an issue now because of the Republican majorities in the House and the Senate and so forth. Let’s say, is there any constitutional break?

The reality is that, Congress over the years has delegated authorities to the president. There are two particular windows that Trump has used in Trump, the 1.0 previous administration. One of these is Section 232, which essentially says that in the interest of U.S. national security, putting it in simple terms, in the interest of U.S. national security, the U.S. president can do whatever they choose to do, with respect to tariffs. That’s one. The other one is Section 301, which says, “If there’s unfair trading practices taking place somewhere by some other country, then I can address them with whatever trade measures I see fit.”

Again, speaking somewhat casually, both of those are in the president’s hands. In principle, you can wake up tomorrow morning as the U.S. president, and do what it is that you’ve half-jokingly indicated, tweet out a new trade policy, tweet out something else. They’re all internal mechanisms within the U.S. Commerce Department which says in Section 301, has to go through Commerce Department. There are potential legal challenges and so forth. Can they be done? Yes. They can be done.

RAJAGOPALAN: Oh, boy.

KRISHNA: I should say that I was one of those in 2016—when already in a similar timeframe when Trump had won the election, had not quite come into office yet. Even shortly after he came to office, things were being heard around Washington, and various representatives of the Trump administration were already indicating what type of trade policy they would pursue.

At that point, I felt I was one of those mistakenly felt that this was just campaign bluster, and would not be realized, not realizing the extent of which they did actually pursue it. They pursued greater tariffs against allies. They pursued greater tariffs against—

RAJAGOPALAN: Against China.

KRISHNA: —“adversaries.” And the adversarial relationship with China, it went well beyond that. It started with the steel and the aluminum tariffs. India was one of the targets of that. One of the external constraints that normally would’ve been in place, which you mentioned a couple times earlier, is the dispute settlement mechanism of the WTO. Normally countries would’ve been able to go to the WTO and say this is a fault of the United States, and we challenge this.

Successive U.S. administrations, Trump administration, actually the Obama administration prior to them, as well and the Biden administration that followed Trump, none of them have done much to allow the dispute settlement mechanism to function. Actually, I’ve misstated this. The dispute settlement mechanism needs judges. Judges need to be appointed.

RAJAGOPALAN: The U.S. is not appointing them.

KRISHNA: —U.S. is not appointing them. This started before Trump. It continued under Trump. It reached a dramatic point under Trump because the number of judges fell to zero, essentially.

RAJAGOPALAN: Biden didn’t improve on it at all.

KRISHNA: Biden didn’t improve on it. There seems to be some bipartisan consensus on certain issues with respect to international trade and with respect to China in the U.S.

RAJAGOPALAN: Yes. Because Biden’s continued the Trump tariffs from 1.0.

KRISHNA: Absolutely, through industrial policy of certain forms. If anything, has intensified this particular margin of the competition with China. The internal constraints, at least legally, they aren’t there. The international constraints, this dispute settlement mechanism is not there. Even if dispute settlement mechanism was revived in some manner, which will take some time, it’s not clear that it acts quickly enough to push back on any of these things.

A lot of this is now left to his instincts and his feeling for whether this is the right thing to do. It’s conceivable from what we’ve all observed of Trump, that this is just some negotiating tactic and that he’s willing to completely deviate from WTO rules of nondiscrimination, but go country by country and say, “You got a 20% tariff, you got a 30% tariff, what can you do for me?” That could be. What can you do for the U.S.?

In some minority of cases, this might even be good. There are still countries that have very high tariffs. I could imagine some hypothetical situation where you’re saying, “Okay, listen, why do you have a 50% tariff on me? I’m going to put 50% on you unless you reduce your tariffs.” Some other country unilaterally reduces the tariffs and that could be good.

RAJAGOPALAN: Will that only be for the U.S. or will it go back to that GATT world where they reduced it for the U.S. and now it’s reduced for everyone? Is this Trump’s way of getting the world back to WTO, but it’s so secret and we don’t know it?

KRISHNA: I don’t think it’s his way of getting back to WTO because that would mean to really insist on nondiscrimination and equal treatment for all and so on, that would completely constrain the way that he—

RAJAGOPALAN: That’s true.

KRISHNA: —would operate, could operate, maybe wants to operate. I think individual using the might of the U.S.

RAJAGOPALAN: Then it’s going to be the U.S. with bilateral agreements with each trading partner. Oh, my God.

KRISHNA: Right. Complete deviation from this whole architecture that the U.S. put in place.

RAJAGOPALAN: I hope you have lots of graduate students. You have so much work cut out for you.

KRISHNA: In reaction to the absurdities of the interwar period in which this is one by the way.

RAJAGOPALAN: This is. Yes.

KRISHNA: There was a conditional MFN—these are details—but conditional MFN in trade agreements between 1918 and the 1930s, which was extremely chaotic, and it was on a reciprocal basis. The kind of thing that Trump is in. Extremely chaotic because you really have to work out tariffs for thousands, tens of thousands of goods for 160 countries, so extremely complicated.

RAJAGOPALAN: Lots of PTAs.

KRISHNA: Lots of PTAs. We’ll see where this actually goes.

How the New Administration’s Plans Might Impact India

RAJAGOPALAN: What should India do if this Trump 2.0 version is going to pan out, right? I still think the 7% uniform is the way to go because it checks all the boxes, but it also opens India up to China. Is this going to be outside of China and maybe a couple of countries? We just reduce tariffs unilaterally for everyone with whom we’re not in a border dispute or another conflict?

KRISHNA: Yes. I think the we’re entering a more challenging, chaotic world, that’s for sure.

RAJAGOPALAN: No, but that’s my question, are we? As a size of the amount of trade taking place, India’s largest exports are to the U.S., and its largest imports are from China. After the U.S., you have of course, European Union and the Middle Eastern bloc because of oil. Is it really that chaotic? It just seems like this could actually end up in a reasonable direction in a non-WTO world?

KRISHNA: Being that the U.S. is the largest importer of India’s exports, high U.S. tariffs could create problems for India’s exporters.

RAJAGOPALAN: On the other hand, I thought the idea was that if India reduces, the U.S. will reduce.

KRISHNA: Could, but do we have the appetite for that? Indian importing interests and exporting interests are not the same parties. These are not the same individuals saying, “Okay, this is a deal that can work for me, so let’s do it.” These are vastly different groups of people, on the one hand.

The trade relationship with China was always going to be challenging because in many ways they used to be this kind of low-skill abundant country. We are a low-skill abundant country. We’d like to do the things that they do and what we don’t necessarily want is to be—at least this is a feeling in the policy world, New Delhi—we don’t want to be drowning in Chinese imports in those sectors where we have potential for growing ourselves. All complicated by the events of 2020. The geopolitics on that margin has made things more challenging.

RAJAGOPALAN: And exiting RCEP. RCEP would’ve been the perfect way to integrate that.

KRISHNA: They did it partially because of these concerns of not—of course, 2020 had not happened by then yet, but there was a fear of this. Now what options does that leave for India?

On the one hand, there’s been all this talk. For the last 25 years there’s been the talk about some U.S.-India free trade agreement. I think that’d be difficult, not impossible, but difficult. RCEP is difficult because of China’s primacy within it. I’m not sure to what extent, there’s some recent news about the geopolitics having cleared up. I’m not sure how much that alters the strategic calculations around RCEP there, but my guess would be that it still would be quite difficult. There is the TPP, or what’s called the CPTPP, which is the TPP without the U.S. in it, after Trump in fact exited from the TPP. That’s a pretty large group of countries. Pacific Rim countries, many countries that India would consider to be allies: Japan, Korea, Australia, I believe, and so on.

RAJAGOPALAN: Australia, New Zealand.

KRISHNA: If India is seeking some level of improved market access and it wants to get into this game of global value chains, then that seems to me to be maybe among the more promising possibilities, if we pursue that. As far as the chaos, I think it will be chaotic. Partially because it’s not just what might come out of Washington against India directly. It’s what might come out of Washington with respect to everybody else, how the world reacts to this. Are all of these movements going to be favorable to India or not?

That being said, it’s also very clear from the experience of the last eight years that that China Plus One opportunity that people were talking about with the—

RAJAGOPALAN: Hasn’t panned out.

KRISHNA: —U.S.-China trade war has not quite panned out in the way that we hoped for.

RAJAGOPALAN: It’s panned out for Vietnam, it’s not panned out for India. It’s not panned out for lots of others.

KRISHNA: Lots of countries. That being said, one of the big changes that I’m very excited about, that matters for international trade, is all of the infrastructure investment that’s taken place. India is a more productive place.

RAJAGOPALAN: Absolutely.

KRISHNA: At some basic level in terms of public goods provision by the government and so on.

RAJAGOPALAN: The ports between GST, the new airports, all of it. It’s definitely an improvement.

KRISHNA: Improvement. I’m hoping that the PLI, although that it has its own critics, that it at least has enough of a demonstration effect, the kinds of investment that is coming.

RAJAGOPALAN: Like Apple, so now we know we can do it.

KRISHNA: You can do it. One can hope that there’s more value addition taking place, even in the context of Apple within India and so forth. These are all good things. But beyond that, one can hope that Apple’s presence and success and expansion within India has demonstrated that India can be a factory for the world and that that has its own positive effects. It’s conceivable that whatever comes out of Washington in the next couple of years actually allows India to finally take that China Plus One—

RAJAGOPALAN: Leap.

KRISHNA: —leap, but who knows. These are difficult questions at the moment.

RAJAGOPALAN: My hunch is it’ll be a combination of reducing tariffs and joining CPTPP, things like that. Like integrating in a meaningful way and more industrial policy like PLI.

KRISHNA: Correct.

RAJAGOPALAN: Because that’s the only real way to please the protectionist folks and also please the allies and also protect against China. That’s, I think, one.

The second, I’m a little bit more worried about the case-by-case nature of it. Even in industrial policy and providing PLIs and subsidies, it’s going to be where are you coming from. If it’s Tesla, they’ll get the red carpet. Whereas if it’s BYD, even though BYD wanted to partner with an Indian firm, and I think they gave lots of money—the Indian counterpart gave lots of money through the electoral bonds and things like that—it just got rejected. I think to gate keep against China, they’ll just try a case-by-case method, which goes against everything that I actually believe in. We may already be in this world of third and fourth bests. I think that’s where I see this potentially going, even in a really good scenario where they reduce tariffs.

Future Trade Relations Between India and China

KRISHNA: The China question is an important one, difficult one for India. There are obviously strategic concerns. There are economic concerns, geopolitical concerns. Everybody is aware of it. For most part in New Delhi, everyone’s aware, people have talked about it, talked about it publicly. The economic survey this year, Dr. Anantha Nageswaran has a very nice discussion of it. Arvind Panagariya, professor Panagariya has written extensively on this as well.

Is it possible to work out some kind of trade relationship or even bilateral investment relationship with China that, wherein, we somehow exclude all of these sensitive sectors? Define sensitive strategic in a narrow sense rather than think of everything as being strategic. Is it possible to do that? I think it’s possible. Will we end up with an ideal outcome? I’m not so sure. Industrial policy has its own pressures. They’re subject to similar sorts of lobbying and so forth. It’s not inconceivable that there’s some growth there, while we’re being mindful of our own strategic interest.

The one thing I will say is the following, which is, look, all the market access in the world is not going to help you if you’re not competitive. All the competitiveness in the world is not going to help you if you don’t have market access. The current levels of market access, the multilateral system as it is today, with all of its problems and deficiencies and so forth, is a wide enough open space for India. Given where we are. Given what our footprint in international trade is, which is actually very small. That at one level you could say, listen, none of these things are worries for us. We don’t need to think in terms of greater market access.

We need to think strategically about just how can we improve our presence of global value chains? How can we improve exports? How can we improve competitiveness, first? If we improve our competitiveness, in other words, the level of market access that we enjoy today is sufficient for us to follow nicely in the footsteps of a country like China, achieve a meaningful structural transformation with workers coming out of low-pay agricultural jobs into manufacturing services and so forth.

That’s all possible. Additional market access is icing on the cake. I would say that there are certain areas like global value chains where that type of frictionless flow is crucial. It’s not just about market access, but it’s just—

RAJAGOPALAN: It’s integrating.

KRISHNA: —integrating in a business environment that sees you as being a friction-free point through which to enter and exit. For those reasons, you might seek different types of market access in this type of integration with TPP and so forth.

Look at China. China achieved everything that it did between 1990 and 2015, let’s say, or 2020, without a single substantial free trade agreement. It didn’t have free trade agreements with the U.S. It didn’t have free trade agreements with anybody else.

RAJAGOPALAN: It had WTO.

KRISHNA: So do we.

RAJAGOPALAN: No. It had WTO in the heyday of WTO.

KRISHNA: So do we. Nothing has changed in that sense.

RAJAGOPALAN: That’s true, actually.

KRISHNA: The global market is what it is. The demand for textiles is what it is.

RAJAGOPALAN: That’s true.

KRISHNA: Of course, things are taking place in the background: technological changes, automation, et cetera. Our market share is 2%.

RAJAGOPALAN: It has to go up.

KRISHNA: From two to five, these are not constraints. If you are 30 and you’re saying, I want to go from 30 to 40—

RAJAGOPALAN: Yes. China has the constraint, actually. We don’t.

KRISHNA: We don’t have the constraint.

RAJAGOPALAN: China has to have a different structural transformation to turn into South Korea or something, which at their scale is very difficult. We have actually the lesser challenge, which is we know that throughout the world, they had already outsourced or imported from China. Now they can do it from another country, which they’re doing from Vietnam and Bangladesh, why not us, right?

KRISHNA: We have less. In case I wasn’t clear, absent any further trade deals, and improvement in our competitiveness will be good for us on the trade margin. If on top of it through trade agreements, free trade agreements, TPP, whatever it is, we’re able to gain greater market access then so much the better.

RAJAGOPALAN: When you say competitiveness, that really does come from lower protectionism, just to be clear?

KRISHNA: Lower protectionism. Also the things we talked about already. Better roads, better ports.

RAJAGOPALAN: All of it.

KRISHNA: A more predictable business environment.

RAJAGOPALAN: Better taxes.

KRISHNA: All those things that go into the infamous doing business with indicators. Everything that’s in there at some level. Regardless of the value of the particular ranking system.

RAJAGOPALAN: That actually really matters. Because in India we know different states have different competitiveness. They all have the same trade treaty. Tamil Nadu doesn’t get any special preference over Bihar when it comes to international trade. Tamil Nadu has got its affairs in order, which means it’s just far more competitive than other states. Yes, that makes sense.

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

KRISHNA: A real pleasure. Thank you so much.