Why haven’t tariffs significantly damaged the economy?

The tariff increases during the second Trump administration have been historic in their magnitude and scope, dwarfing even those from Trump’s first term. Many feared that these tariffs would have dramatic impacts on the economy, but for the most part, the tariff impacts have been muted in the U.S. In a new BPEA study, Pablo Fajgelbaum and Amit Khandelwal document the effects of Trump’s tariffs, what goals they have (or have not) accomplished, and where the costs have fallen. On this episode of the Brookings Podcast on Economic Activity, the authors speak with Kari Heerman, Brookings’ director of trade and economic statecraft, about their findings and what to expect next.

• Listen to the Brookings Podcast on Economic Activity on Apple, Spotify, YouTube, or wherever you like to get podcasts.
• Learn about other Brookings podcasts from the Brookings Podcast Network.
• Sign up for the podcasts newsletter for occasional updates on featured episodes and new shows.
• Send feedback email to [email protected].

[music] 

EBERLY: I am Jan Eberly, the James R. and Helen D. Russell Professor of Finance at Northwestern University.  

STEINSSON: And I’m Jón Steinsson, Marek Professor of Public Policy and Economics at the University of California Berkeley.  

EBERLY: We are the co-editors of the Brookings Papers on Economic Activity, a semi-annual academic conference and journal that pairs rigorous research with real-time policy analysis to address the most urgent economic challenges of the day. 

STEINSSON: And this is the Brookings Podcast on Economic Activity, where we share conversations with leading economists on the research they do and how it will affect economic policy. 

EBERLY: While tariffs have been increasingly used by presidents over the last several administrations, President Donald Trump has taken their application to unprecedented levels, making tariffs the center of his economic and foreign policy regime. And yet, despite the predictions of many economic models, the negative impact of tariffs on the US economy seemed to have been muted. 

In a new paper, “Tariffs in 2025: Short-run impacts on the U.S. economy,” Pablo Fajgelbaum of the University of California Los Angeles and Amit Khandelwal of Yale University explore the breadth of Trump’s tariffs and their impacts thus far. Kari Heerman, Brookings’ director of trade and economic statecraft, interviews Pablo and Amit on today’s episode.  

STEINSSON: Amit and Pablo are really the perfect authors to assess the 2025 tariffs. They wrote one of the best papers on the tariffs Trump imposed in his first term, along with two co-authors. That paper found that the pass through of the tariffs into domestic prices went up was a hundred percent. 

In other words, the foreign countries that export to the United States didn’t absorb any of the tariffs. It was American firms and households in that case that paid the entire tariffs. This time around, the tariffs have been quite different. They are larger and applied to a broader set of goods and countries. So, it is interesting to see whether things have turned out differently when it comes to the impacts of the tariffs this time than in Trump’s first term. 

EBERLY: Now let’s hear from Pablo, Amit, and Kari. 

[2:27] 

HEERMAN: Thanks, Jan and Jon Welcome Pablo to the podcast.  

FAJGELBAUM: Thanks for having me.  

HEERMAN: And welcome Amit.  

KHANDELWAL: Hi, how are you? Good to be here.  

HEERMAN: Great, thanks for being here with me. and I’m really excited to talk about this very interesting paper, which is very timely as usual from the two of you. 

The headline of your paper is the result that the 2025 tariffs have only had small economic effects so far. What’s striking is that this comes after a very visible and historically large increase in tariffs. At least, so it says in the headlines, but in your paper, you show that the actual policy looks quite different. A large share of imports still came in duty free during this period. Applied rates were often lower than the statutory rates, and retaliation was fairly limited outside of China. 

Can you guys walk us through what actually changed in 2025 and why the tariff increases might have been smaller in practice than it appeared in the headlines? 

[3:30] 

KHANDELWAL: Sure. So, let’s take a bit of a step back. If you were sitting in 2024 in December of 2024, the average tariff rate in the US was about two and a half percent. And if you fast forward one year looking in December of 2025, the average tariff rate was just shy of about 10%. And so that magnitude of the change, about the 7.5% tariff change, is the biggest tariff change we’ve seen in, about a year. 

And then if you adjust by the fact that the US is just a much more open economy now than it was, you know, a hundred years ago, roughly 10% of our GDP is, imports. The 10% increase or seven and a half percent increase in tariffs on about 10% of an imports to GDP ratio gives us a pretty sizable change. 

And so that’s as you said, this is a pretty historic event. And yet we ultimately concluded that the magnitude of the shock may be perhaps, smaller than what the headlines perceived. And that’s really for three broad reasons. The first is when we dug into the data, we found that the statutory tariff rates that the headlines were often picking up in the newspapers were larger than the tariffs that were ultimately applied. and a good reason for that is that the U.S. ended up exempting from tariffs, a good chunk of energy imports, a good chunk of computer related imports. 

And U.S. imports from Mexico and Canada still largely come in duty free. And so those three categories basically mean that, roughly 60% of the imports coming into the country in December of 2025 were still entering the country duty free. and so that’s one reason that the magnitude of the tariff changes were smaller than the headline. 

The second reason is that if you compare to the experience in the previous administration’s increases in tariffs, there we found that there was retaliations from our trade partners when we imposed tariffs on their imports, they retaliated in kind. And with the exception of China last year, our trading partners did not retaliate with increases in tariffs. 

And so that kind of blunted the magnitude of the effect on the U.S. economy, which is that the U.S. exports of goods were not hit with, sizable or, in kind increases in tariffs. And so those three kind of statistics, the applied rates are kind of lower than the statutory rates, the majority of imports are still entering the country duty free, and we have muted retaliations, ultimately leads, when we work through the numbers, to see an overall kind of muted response.  

HEERMAN: Thanks, Amit, yeah. So, you’re saying this is still a very big deal. 

This is a very substantial tariff increase. Bigger than what we’ve seen in generations, and yet the headlines may have overstated a little bit. 

[6:02] 

FAJGELBAUM: Amit just discussed the relationship between the type of news we read about how big the tariffs are and how big they actually were. 

Now the question is what effects of these tariffs we’re measuring? And I will qualify a bit, this notion that the effect is small. In the sense that we need to ask a small or large relative to what? So, the tariffs didn’t have sizable impacts on the economy. It is just that these impacts are not a massive share of GDP because the amount of economic activity being taxed is not a massive amount of GDP. 

So just to do a quick back of the envelope calculation. We just discussed that the tariff increased was about 7%, meaning the tariff rate went up by seven percentage points. On imports that about 11% of GDP. That means we are taxing about 0.8% of economic activity. 

So that’s the amount we are taxing, and we are finding net effects, meaning net creation of value or net destruction of value, basically how many dollars we are taking in or out of the US economy, that are plus minus 0.1%. So, if we do the math, we say the shock was about 0.8% of the US economy, and the net effect is like 0.1%, either way, positive or negative.  

It’s elasticity of one eighth. So that’s not zero, that’s not a very small effect. It’s just that the total amount of economic activity that’s being created or destroyed, it doesn’t feel so big because the shock wasn’t that large. 

So then let’s unpack where this total number is coming from. And like you say, so there are different components and broadly speaking, sort of the components are about by how much the prices that the buyers of imports, are paying, change versus by how much the tariff revenue collected by the government is changing versus how much the wages or the income of producers, are changing. 

So, this is a headline number we have, which is this pass through of 90%. This is something that’s going into prices, meaning for each dollar that we increase, in terms of tariffs, then 90% goes to higher prices paid by buyers of imports. 

Now, this is obviously consumer loss, but we need to keep in mind that the tariff itself is just an internal transfer to the U.S. government. 

So, the question is that transfer, how much value is actually creating or destroying, besides the loss of consumer surplus to the prices are not coming through the wages. And when we add these effects, then the net effects gives us this one to eight, elasticity that I mentioned before. 

Now there is something that now you say, well, but the total effect can be either positive meaning plus 0.1% of GDP or negative minus 0.1% of GDP and that depends on terms of trade effect. 

So, what do we mean there? We mean that there is something that our analysis cannot really capture, and I will argue like most of the academic literature is sort of struggling with kind of pinning down the magnitude of this effect, which is perhaps, let’s say the wages, the overall level of wages in the U.S. or in China is changing with the tariffs. It’s an aggregate effect and everything we can measure here is relative. We are always comparing the prices of goods that are taxed versus not, taxed, or the wages in sectors that are protected versus not protected. But in general, we are unable to pin down an aggregate effect on the economy. 

And what we do in the paper in analysis is to make extreme assumptions. We either assume that there was absolutely no response on that margin, that that’s the case in which we get to the negative 0.1% of GDP or that the aggregate wages in the U.S. relative to China responded in the most optimistic way that will give you the plus 0.1% of GDP. 

[10:18] 

KHANDELWAL: So, I guess what we could try to unpack this overall effect that we find on the US economy. And I guess the way that we like to think about it is that there are three buckets that are changing. There’s, consumers who are purchasing the imports, then there’s gonna be domestic producers, which presumably are gonna be better off from getting protection. and then there’s gonna be some revenue that’s collected by the U.S. government. And so under the hood of this overall effect that we find, there’s a lot of action changing there, and that’s what Pablo was alluding to, where that, despite being, roughly a 10% increase in tariffs on roughly a 10% import to GDP ratio, there’s many things happening under the hood. 

So, let’s think about the first effect, which is imagine a tariff goes into effect, suppose, a Chinese company was sending, a hundred dollars good to you as the buyer. and it previously was sending it at a 0% tariff, and it now gets a 40% tariff. A key question then that we try to look for in the data is, does the tariff inclusive price go up to all the way to $140, or does the Chinese exporter reduce some of its price so that the tariff inclusive price, maybe it doesn’t go all the way to $140, but maybe goes to say $130. 

And as you said, we found that roughly 90% of the tariffs are passed into the U.S. economy, which means that U.S. buyers of imports are taking it on the chin. So, in our analysis we find that there’s a really big loss to U.S. consumers of imported goods. 

But there’s this flip side, and that’s, I think what’s often, not well understood, which is that there’s going to be some benefits to the protection, which is if you’re a domestic producer and you’re competing with, that Chinese exporter, the fact that you now have a tariff shield means that consumers would probably start to substitute and buy some of your goods. 

That’s going to lead to a gain to the producer side of the economy. And then of course, there’s also revenue that’s being collected. And so you have these three buckets. You have this negative loss to the consumers. You have producers who are gonna get a tariff shield that might make them better off. They might not be quite as better off because some of their imported components are also rising in expense. Their costs are going up, but you might expect them to be slightly better off. And then there’s revenues. So, you get this big negative and then you get these two positives and that’s what leads to this overall relatively small effect. 

HEERMAN: Let me just kind of summarize where we are because I think what you’re pointing out is that the headline might be that the overall effect is relatively small. But if you take that into the context of the amount of the economy that it’s affecting, it’s not small on the part of the economy that it is affecting. So, because traded goods are simply a very small part of our GDP, what we produce, what we consume in our economy now, what we spend our money on, the overall effect on our overall economy is relatively small. 

And where this effect comes from is three different places. One is, we are forcing the buyers of imported goods to pay higher prices for things. Two, on the other hand, offsetting that a little bit is we’re protecting some of the producers, so demand for what they produce goes up somewhat. And three, we’re getting some tariff revenue, which provides us something in our coffers to pay for the things that we need, that our government provides. So that’s kind of where we are. 

In your paper you do talk a fair amount about this thing called terms of trade effects, and that is something that I have found people trip over all the time. and it is kind of trade modeling magic, but if you could just walk us through what that means, I think it would be helpful. 

[13:59] 

FAJGELBAUM: Here what we mean, is that when the U.S. taxes, China, then there will be, relocation of demand away from China and into presumably U.S. made goods. this raises the level of demand in the U.S. and this in our models raises wages in the U.S. 

So, then this is really an effect that we have in our models, and it’s a very canonical effect. We are not using special or weird model in that way, this is really a canonical textbook way of thinking. It’s an aggregate wage effect. It’s a generally driven force that appears in our mouth. but it’s very hard, if not impossible to pin down empirically because it’s something that’s aggregated. 

 It changes from one period to the next, but we don’t have a way to identify by comparing different units. We will have to compare the wages of the U.S., how they change in this time period to the wages of say, Mexico. But of course, many other things are happening at the same time in the U.S. and Mexico, and that’s why this is really a model-based effect, this aggregate terms of trade. 

Then there are also, let’s call it the product level terms of trade effects, which are the ones we talk to and we do estimate when we talk about pass through. When we talk about the pass through, we are saying the price of a particular good that we are increasing the tariffs by 10% on then, will pass through, with a 90% rate. In this case, it will be 9% price increase on the consumers at the product level. 

But for the aggregate U.S., there is this aggregate terms of trade effect, and that’s the one that we make as a model assumptions, and then we make this range of assumptions either very optimistic or very pessimistic, so to speak, and that put us in this range of plus minus 0.1% of GDP  

HEERMAN: Great. So, you’re saying because the U.S. is really, really big source of demand for all things produced in the world. If we put a tax on things that are produced outside of the United States, it inevitably reduces demand for all the things outside of the United States, perhaps in a large enough way that their overall demand for the things that they produce falls enough that they can’t compensate. They have to drop the compensation for all the people that work in those economies, but you can’t really observe that. 

So you have to guess either it’s big or it’s non-existent, and what you do in your model is assume it’s non-existent and measure the outcomes, you get something a little negative, assume it’s big, measure the outcome, you get something a little positive, and that’s why you’re kind of on both sides of zero is because you have this factor that is important and possibly big because the United States is big.  

You said that there’s something we can’t really pin down. It’s hard to observe. That makes absolute sense. Is there any evidence that this is happening in the world? Is there something you could see where you could, try to assess whether the United States is big enough to make this happen, or not? What would you look for? Are you seeing anything? 

[16:48] 

KHANDELWAL: Yeah, so there’s a couple ways to test for it. In the paper, there’s one particular kind of test that we try to implement, which is a little technical, so I won’t go into those details. But there’s something that might be a bit more intuitive, which is when we talk about the terms of trade, what we’re thinking about is a measure that is the ratio of the prices that the U.S. receives for its exports relative to the prices that it pays for, its imports. 

And so, we talked a little bit about what’s happening to the prices of the imports, but then one could also look at data, for example, the Bureau of Labor Statistics regularly collects, the Export Price Index. It also collects information on producer price indexes. 

And we can check, for example, in the sectors in the United States that received more protection, do we see increases in producer prices? Do we see increases in export prices? 

And when we do this, just think about a simple kind of scatter plot where you’re trying to plot, the sector level tariff on those prices. We see kind of an upward sloping relationship, but it’s not particularly, sharp. And so that is why we are very cautious in the paper to conclude that this terms of trade effect, which determines the sign of this overall impact, whether it’s positive or negative, we’re very cautious to say whether, we’re in one extreme or the other because it’s not super, super conclusive. 

And it may be the case that as more time evolves, we have some ability to strengthen our credibility of that statement. But, as of 12 months of data, that’s kind of where we saw things land. 

HEERMAN: So, your model finds the small net effect that we’ve just kind of been through. You guys are really looking at fundamental elements of international trade and how you measure its impact on the economy. But your framework, your model is deliberately kind of stripped down. It doesn’t have all the whistles and bells that trade economists can integrate into models these days. 

 You’re abstracting from some things that trade economists have found are really, really important, in understanding how international trade affects the economy, dynamics, the effect of uncertainty, not knowing where policy is going on, what businesses do affects the impact of trade, and some longer run adjustments. 

So how should we think about that? Are those things important? Or unimportant? Is that why you left them out of the model? 

[19:09] 

FAJGELBAUM: So first I would say the way we did this exercise and we are building on previous work we have been doing, so we do wanna stay as close as possible to what we measure in the data, to what we can measure in the data and the way we construct our model. 

So, we just discussed like a really important aspect of welfare, which is these terms of trade, are there or not, and we don’t know, and we make this range of assumptions. So let me tell you, the moment we start to bring some of these things, these extra forces, then many of these are very, very hard to pin down. Of course, different economists have tackled different ways of doing it. But the question is, in this particular exercise we are looking at, yeah. 

And obviously not enough time has evolved for, for example, figuring out what was the effect of uncertainty in this case, and one will need to kind of come up with, some creative research designs to estimate that and then integrate into the model and so on. 

And this problem of, there are many things that we don’t really know what’s their kind of right quantitative magnitude, it’s magnified if we are talking about things such as, long run growth. Then we are relying on aspects of the economy that obviously are very important, but who’s like, empirical significance in response to a tariff in a dynamic way, it’s hard to pin down. 

So, we can bring those things and people have brought those things in different ways. But then this range will start to move. If it was minus 0.1% to plus 1%, then it will become bigger. And I don’t have a prior that it will become bigger, like way more positive or way more negative. We can start to speculate on a specific ones. 

For example, if we talk about growth, and dynamic effects. Yeah, there are papers that have convincingly shown that theoretically when you start to incorporate, for example, capital accumulation in a context where an economy is importing intermediate inputs, this will likely magnify, the losses because you are importing, something that is an input in production and this gets accumulated over time if this becomes more expensive then you will accumulate less and the economy will be poorer in the long run. 

That’s one force where, because of this particular dynamic force, if we were to incorporate it, and then you can say, oh, well probably the losses are bigger, but you can probably come with dynamic forces that can go perhaps the other way. You can argue that well, but also, you can get capital to allocate into the U.S. in ways that it wouldn’t have reallocated another way by taxing imports, so then you stimulated more foreign investment into your economy and this may actually magnify the terms of trade effects in your favor. 

And this is all very interesting, and I think that’s why, you know, I guess a lot of trade economists are kind of excited to study all these things. So, this is just to say I think we are providing a very good prior, and I will argue this is a prior that is as close to what we can measure as we can make it be close to what we can measure. 

And around that, yeah, there are many things that can be added and some of them we kind of go in which direction they will go and other may go, either way. So that’s say, yes, there are many forces that we don’t have, such as dynamics, uncertainty, all sort of frictions, and so on. 

And I wanna also say we obviously, we are not the only ones to put numbers, on the effects of the trade war. In the paper, we try to make as comprehensive as possible literature review of a lot of people that has done great job looking at, for example, like business cycle frequency effects, effects of having, trade deficits, as a source of welfare changing endogenously with that if some of these research is in theoretical stage and another is more quantitatively applies, so there are other people that have actually tried to put numbers on different aspects of what’s going on. 

And perhaps, for the best estimate, once you look at this range of papers and try to make that the range and different papers have looked at some particular things.  

HEERMAN: That’s really helpful. So your goal is to stay as close to the observed experience, the observed data as you can with this model, which arguably I think has a benefit to stay as close as possible to what people are feeling, with this increase in tariffs –that it feels not as big as it sounded like it was going to, and we’re not really sure where it’s going. I feel like this paper kind of comes out on the same plain as a lot of observers are coming out on. 

And you’re also saying that we just went through how you guys had to make these assumptions on how to get a range on terms of trade effects and what you were trying to avoid is having to do that eight times over and having to think about how do we parameterize and how do we think about, and how big are the dynamic effects and how do they feed in and how big is uncertainty and how does that feed in? 

Because those things are found in past research to be important and to have played out in a certain way in the past and we have a lot to learn from the future and you guys are professors of economics and international trade, so you have a lot to teach your students I assume as well. 

[24:03] 

FAJGELBAUM: But I also wanna emphasize like if there is interest in some forces that we don’t have in the paper, I’m sure that in the literature review that we do, people will find references to great recent papers that look at some of the things such as dynamics and so on. 

HEERMAN: In your paper, you do a really nice job of laying out an evaluation of whether the tariffs have achieved some of the multiple stated objectives. And you’ve given some really nice, clear benchmarks against which to measure whether there has been success, and that is a huge contribution. But ultimately, you say in the end, the dice are still rolling. We need more time and more data to fully assess the impacts on the economy, and I fully agree with that. 

Now the three of us have known each other for some time, and part of the reason is that over the, last several years, the two of you have done very relevant and very insightful work on trade policy and its effects on the economy, particularly starting in and around 2016. And I have called you from several jobs, from the office of the U.S. Trade Representative, from the Council of Economic Advisors, and from the State Department, as I’m putting together, advice for policymakers, and I have said, Amit, Pablo, help me explain this. What are you finding? Do you have anything new on de minimus? Do you have anything coming out on the longer-term effects from the 2018 tariffs? And you have always responded, and I thank you for that. 

Now suppose my next job is to take the current tariffs as they are today as a starting point. And I get to decide what to do next, and I call the two of you for input. And I can’t wait to see where the dice land, and I have to make a decision mid adjustment Now, I’m gonna present my idea and certainly I’m brilliant, so they’ll do whatever I say, but they’re still gonna ask me some follow up questions. 

If all I can say is there’s some small aggregate effects of what we’ve done so far, they might conclude this doesn’t even matter very much. We’ve talked a lot about how that small effect isn’t really small for the targeted, part of the economy. and you’ve talked about how there’s a number of transfers under the surface. But how might the existence of some of those transfers within that small part of the economy affect my decision about whether to leave the current structure intact or to make some changes? That’s question number one, and then I’ll have a follow up after that. 

[26:24] 

KHANDELWAL: Oh yeah, that’s a great question. What I would say is that tariffs are one instrument that a policymaker has to change incentives within an economy, and it can be a pretty blunt instrument. 

So, in the paper we kind of worked out what we believe are six or seven different rationales that policymakers have had for, justifying the increases in the tariffs. And so, if you start to read through these rationales, for example, you want to increase manufacturing employment, but at the same time, push some of the incidents of those tariffs onto the foreign producers. 

Those rationales can be in conflict. If you want to increase manufacturing employment, you really do need to make imported goods expensive. Otherwise, domestic consumers are not gonna switch towards the domestically produced goods. And so one of the things you kind of immediately realize is that if you want to use tariffs to achieve a particular goal, I would say it’s important to specify that goal and to be very clear that it’s going to come at the expense of something else. 

And we have found, for example, in looking at the structure of the tariffs, that some of the stated goals do not appear to be achieved through how the US has imposed its tariffs. For example, we went into this project expecting tariffs on geopolitical allies to be lower than geopolitical rivals, and we don’t find evidence of that. We went into this project expecting tariffs to be higher on, goods that were particularly important for national security or advanced technologies, and we don’t find evidence of that. If anything, actually we have higher tariffs on apparel and footwear than we do on advanced technology products. 

And so I think one of the lessons that comes out as we went through this is that, you know, having many possible rationales and expecting terrorists to hit all of them, is not gonna be achieved, and if you do limit yourself to one particular rationale, it’s going to come at the expense of others and I think if you’re very clear to the public that, you know, we want to increase manufacturing employment, we’re gonna do this, we’re gonna expect some efficiency losses and all the other problems that tariffs may create in order to achieve this objective, but that there’s going to be a, cost that comes with that potential benefit. 

I think, that’s one way that I would kind of go about explaining what tariffs can and cannot achieve.  

HEERMAN: That’s very helpful. You have said a couple of things that are really, really important and really insightful, which is that these are inevitably gonna come at the expense of other things. 

Now, suppose my next bosses are really focused on a couple of things and they’re clear about it and they have very clear objectives. And two of those things are, what they really, like about tariffs are the revenue they get, and they also really, really, really want to diminish reliance on foreign suppliers. You’ve just said that for critical goods, the tariffs aren’t even there. So, consider that part of the tariff structure that we may or may not wanna change. 

So, my boss says, I don’t wanna tank the economy. I don’t want to have, bad outcomes for households in the U.S. economy. I really need the revenue. I really want to move away from relying on adversaries for some critical inputs. 

How does your paper help me explain what this comes at the expense of, how to make some of the trade-offs that we might have, and how do I articulate efficiency losses and some of the other negative things that come with maintaining tariffs where they are or adjusting them. 

[29:59] 

FAJGELBAUM: About the tariff revenue, this is a pretty textbook answer, but there is the basic laffer curve logic, and there is only so far you can go with tariff revenue replacing for other sources of taxes and the logic there. for whether you want them or not, is a pretty standard public finance logic. An optimal taxation structure should try to somehow minimize the distortions. So then on the margin, it may be that the tariff is less distorted than other sources of tax revenue. that’s, I guess, up for scrutiny and, and recent papers that we cite have actually made points along those lines. 

But again, like the Laffer Curve logic, which just to remind, people means that you increase the tax on some activity, that activity will go down. In this case, you can keep increasing the tariffs, but at some point, the imports will fall and there will be a maximum amount of revenue will be able to achieve. all the regional calculations people have done say that, you know, you cannot, go super, super far in funding the federal, budget with just tariffs, or, largely with tariffs. 

So, that’s the first point. It can be less distorted than other taxes on the margin. That’s a valid discussion, but at the same time, there is a ceiling to how far you can go. 

[31:12] 

KHANDELWAL: Just to caveat that point is even after we saw that this increase in tariffs by last year, it was about 5% of government receipts. And so, again, because imports to GDP is about 10%, there’s kind of an upper bound on how much we can fund government expenditures from, tariffs, and irrespective of whether you like tariffs or not, you need other sources of revenue. 

And then there’s this additional element, as Pablo mentioned about whether the tariffs are in fact the most efficient way to raise that revenue. 

[31:38] 

FAJGELBAUM: And then of course there is a question about production, that can be a goal that the government has. So first and empirically, again, I don’t think we have like very strong evidence from the 2018 that they raised production significantly in the protected areas or sectors. And again, we mentioned research that has looked into that empirically. We have not, then another kind of textbook sort of response is that we typically think of if the goal is to target production, usually we think of direct incentives to producers as being less distortive than like tariff that distort both consumption and product. 

So in that way, coming back to the discussion about what are the best tools to achieve a goal, if the goal is to, incentivize the sector, it’s not always you wanna do it with tariffs as opposed to, with tools that do not directly target trade, but rather target the incentives to hire, for example. 

[32:25] 

KHANDELWAL: And I guess I would further add that we know from a lot of experiences of the use of tariffs around the world, which is that once the tariffs go into effect the initial kind of protection that you’re providing to a particular sector, might start to dissipate as that sector becomes, used to the protection that it receives. And it’s therefore not necessary for them to become as competitive as possible. 

And so often what we see, in discussions of kind of the best way to use tariffs to stimulate activity would be kind of a protection, but, perhaps dissipating pretty quickly as the sector kind of gets off that infant stage and is able to compete without the necessary protection. 

[33:03] 

FAJGELBAUM: And of course, if a cost of tariffs that is, less visible or less important for other tools, is that tariffs are zero sum. Zero sum, I mean between the country impose the tariff and the other countries if there are positive terms of trade for a country. 

And if you’re going to grow industrial production or production in some sectors, it will be through these increasing hiring in these sectors. If you’re thinking of tariffs and it’ll be through terms of trade improvements. If you’re thinking tariffs as the mechanism to do it, this comes at the expense of the rest of the world. 

So, there is the extra cost that, you know, you’re hurting trade partners by achieving your domestic goal. The U.S. does, being a large economy affects trade partners, even like a domestic subsidy to a sector will affect trade partners in a way but the tariff does it very directly. 

[33:51] 

KHANDELWAL: And we had mentioned that we saw limited retaliations last year. In our model if our foreign trade partners had retaliated the magnitude of the losses would be, through about three times larger. And so, again, going back to this point that Pablo just raised, maybe they haven’t retaliated now, but the longer the tariffs are in effect, it may give them, space, as time falls for them to retaliate and increase, tariffs on U.S. exports. 

HEERMAN: Sure. But I’m just looking at what I’ve seen. I’m looking at the news; I’m looking at your paper. It’s stripped down. It really jives with my experience and you’re telling me I’ve got a couple of things I wanna do. They’re really serious things. They’re really important, and you’re saying no, there’s limits on what tariffs can do. 

Maybe they can help with some revenue, but there’s a cap and maybe they can help, you know, adjust a little bit what the incentives look like for where things are produced, but maybe not that much. It would be better if we used subsidies, but, you know, then we’d have to get that money from the Congress and, oh, that’s a difficult thing. 

So, what I’m hearing from you is, eh, they’re not that good for it. And so far, what we’re seeing is, I don’t know; it doesn’t seem to have a huge effect on the economy. The economy, the part of the economy at effects, is small. Tell me what it is that this is coming at the expense of. What are we giving up? What are the transfers looking like in your model that you talk about? How does the distribution change? 

[35:18] 

FAJGELBAUM: The big redistribution of, of course, is away from buyers of goods into producers and the government. So that’s, the main sort of margin of gross, gains and losses that add up to this sort of smallish, net effect, in the sense that we describe it in the beginning. 

So, it’s going to create winners and losers. So that’s going to be the main thing I will keep in mind. 

[35:40] 

KHANDELWAL: There is also this kind of opportunity cost of time. So, the tariffs may do something to tilt incentives. It may not be the best thing that one could do, but then that means you’re giving up this opportunity. So if the structure of the tariffs are not being used exactly to achieve national security objectives, then that might be coming at the expense of policymakers, kind of bandwidth to think about, you know, they might be thinking that the tariffs can achieve this goal, and we may be looking back, in three years and realizing that, that it may not be there. 

And so, I think, a discussion about what the right instrument is to achieve a particular objective is quite important in part because it helps set the debate. And, as you mentioned, we may not have the fiscal space for subsidies, but if it is clearly a priority, that’s a debate that Congress and the American public should know about and should see. 

HEERMAN: You’ve each made an excellent point, Pablo, pointing out that there is a distributional impact that you guys find in your paper where these tariffs are regressive essentially. They’re hurting consumers and transferring wealth, to protected producers and to the government. Generally speaking, my boss might wanna know about that. 

Second, the point made by Amit is excellent as well. If we think we’re solving a problem that we’re not ultimately solving, and it is a national security problem, that’s something we really wanna have a sharp understanding of. So, that’s great. I have just one last question, and this is a gotcha question. 

I hope you guys are ready. Your paper starts with a quote from Seinfeld. How long have you known about this quote and how excited were you to put it in the paper? 

[37:19] 

KHANDELWAL: Oh, well, I’ll take credit for that. It is my dream to have combined my two great interests in life: economics and Seinfeld. 

And I was able to achieve that in this paper. 

HEERMAN: Fantastic. Well, it’s been great being with you on the podcast, and I hope to see you in person soon. 

FAJGELBAUM: Thank you.  

EBERLY: Thanks so much, Kari.  

[music] 

STEINSSON: Once again, I’m Jón Steinsson  

EBERLY: And I’m Jan Eberly.  

STEINSSON: And this has been the Brookings Podcast on Economic Activity. Thanks to our guests for this great conversation and be sure to subscribe to get notifications about new releases of this podcast.  

EBERLY: The Brookings Podcast on Economic Activity is produced by the Brookings Podcast Network. Learn more about this and our other podcasts at Brookings dot edu slash podcasts. Send feedback to podcasts at Brookings dot edu and find out more about the Brookings Papers on Economic Activity online at Brookings dot edu slash B-P-E-A.  

STEINSSON: Thanks to the team that makes this podcast possible. Fred Dews, supervising producer Chris Miller, co-producer, Gaston Reboredo, co-producer and audio engineer. Show Art was designed by Katie Meris. And promotional support comes from our colleagues in Brookings Communications.