The Quiet Trump-Harris Trade Agreement

Subscribe here: Apple Podcasts | Spotify | YouTube | Overcast | Pocket Casts

After former President Donald Trump’s surprise victory in 2016, his administration imposed several rounds of tariffs on China on everything from washing machines to steel. The move was described by the nonpartisan conservative organization the Tax Foundation as one of the “largest tax increases in decades.” And yet, protectionist economic thinking has since gained traction in both parties. In a rare instance of agreement, President Joe Biden retained most of his predecessor’s tariffs—and imposed even more earlier this year.

Treasury Secretary Janet Yellen described her own evolution on this topic succinctly: “People like me grew up with the view: If people send you cheap goods, you should send a thank-you note. That’s what standard economics basically says … I would never, ever again say, ‘Send a thank-you note.’” Essentially, Yellen used to think that if China wanted to flood the United States with cheap goods, why complain? Well, now she appears more concerned about the cost of all those cheap goods to the nation’s domestic manufacturing base.

On today’s episode of Good on Paper, I’m joined by the Cato Institute’s vice president of general economics, Scott Lincicome, to examine this popular narrative—one that he doesn’t put much stock in, largely because the high cost of tariffs are disproportionately borne by poorer people, but also because of the political dysfunction they sow:

“The economics of trade are counterintuitive,” Lincicome explains. “And so tariff policy is notoriously corrupt. And so there’s a lot of political dysfunction, along with just hiring all those lobbyists to get special tariffs or special exemptions. But also, it’s just a very politically perilous policy.”

The following is a transcript of the episode:

[Music]

Jerusalem Demsas: There was an interesting policy exchange about tariffs between former President Donald Trump and Vice President Kamala Harris during their debate last month.

Demsas: Then Trump hit back, pointing out that the Biden-Harris team had been all too happy to keep the tariffs going.

Demsas: This exchange flew by many people. There was a lot going on in that debate, and this happened in the first few minutes. But Trump is pointing out something interesting here—that while Harris is calling his tariffs a sales tax, she and Biden kept the majority of his tariffs when they came into office.

Looking back on 2019, Biden had similarly criticized Trump’s trade policy, arguing at the time that “any freshman econ student could tell you that the American people are paying his tariffs.”

While I think it’s important to highlight this similarity, it’s also important not to overstate it. Trump is now promising a 60 percent tariff on goods from China and a 20 percent tariff on everything else the U.S. imports. And in a speech last week, Trump said he’d “impose whatever tariffs are required—100 percent, 200 percent, 1,000 percent.” This is far greater than anything Biden or Harris have publicly considered.

[Music]

Demsas: Welcome to Good on Paper, a policy show that questions what we really know about popular narratives. I’m your host, Jerusalem Demsas, a staff writer here at The Atlantic. And today we’re talking tariffs, trade, protectionism, and more.

The standard economic narrative around tariffs is pretty negative. As my guest today has explained in a quip now famously memorialized on a novelty T-shirt: “Tariffs not only impose immense economic costs but also fail to achieve their primary policy aims and foster political dysfunction along the way.” It’s a busy shirt.

Scott Lincicome is the vice president of general economics at the Cato Institute and has written broadly, including here at The Atlantic, about why the parties shouldn’t be so quick to embrace tariffs.

[Music]

Demsas: Scott, welcome to the show.

Scott Lincicome: Well, thanks for having me.

Demsas: We’re going to talk about tariffs today, so I’m going to start with the simplest question: What is a tariff?

Lincicome: A tariff is a tax applied to an imported product, usually a good but, in theory, you could try to apply tariffs to services, as well.

Demsas: What kinds of things that people commonly buy have tariffs on them in the United States?

Lincicome: I think one of the most common examples we use is pickup trucks. In the 1960s, there was a dispute with the Europeans over chicken, of all things. That led to a tax on pickup trucks—

Demsas: Wait. Wait. Wait. Slow that down. How do we get from chickens to pickup trucks?

Lincicome: They were going after our chickens, so we put tariffs on their pickup trucks, and they stayed. Now we still have a 25 percent tax—tariff—on imported pickup trucks from everywhere except a few free-trade-agreement countries, like Mexico. So one of the reasons why we don’t have some of those cool little pickup trucks that you might see in Japan or whatever is because they’re subject to really ridiculous tariffs. And automakers abroad don’t want to have to deal with all the regulatory compliance and that kind of stuff and then pay another 25 percent tariff.

It’s actually a great example of the things that tariffs do beyond just raising prices. They limit availability and consumer choice, and they stick around forever. We have tariffs on the books on shoes and clothing and other things that go back to the Smoot-Hawley days. They’re really hard to remove once you get them into place.

Demsas: You just said something interesting. Who pays the tariff?

Lincicome: It’s a little complicated. Legally, the importer in the United States, in almost all cases, is paying the tariffs. If you are a U.S. company and you are importing stuff, you’re going to be paying the tariff, by law. There’s a little exception to that, but we don’t need to worry about that. The more complicated thing comes in who actually pays, because, in theory, a foreign exporter can lower his price to essentially absorb the tariff costs.

Let’s say you’re shipping widgets into the United States, and they’re $100. All of a sudden, a 25 percent tariff gets attached to it. You have, really, two basic choices: still sell at a hundred and have the importer pay $25 (25 percent of a hundred), or you lower your price to 80 and then have the importer pay $20 in tariffs. But to the importer, it’s all the same thing, right? It’s still $100. So the tariff hasn’t changed the calculus. In that sense, the foreign exporter is bearing the incidence of the tariff.

Then we have the empirical question. So the empirical question is: What actually happens? Well, what actually happens is, in the vast majority of cases, importers and consumers pay the tariffs. You only have a situation where foreign exporters pay tariffs when the market that the foreign exporter wants to sell into is just massive—really important—and the exporter says, You know what? I just want to maintain market share, so I’m going to lower my prices.

Typically, that’s not what happens. Typically, the consumer, the importer is going to pay the tariff. It might not be a hundred percent; the exporter might discount by a few bucks here and there. But, overall, as an empirical matter, typically consumers, importers pay. And that was certainly the case with the Trump-era tariffs on steel and aluminum and Chinese imports. Studies show that about 95 percent of the tariff incidence fell onto American companies and consumers.

Demsas: And so as any listener listening to this can tell, you don’t really like tariffs. Economists, in general, don’t really like tariffs. Why is that? Can you walk us through the standard economic story for why tariffs are bad?

Lincicome: I’ll start out with saying that economists are okay with tariffs in certain contexts—national security, for example. There’s a legitimate case that the United States—I’d say, a strong case the United States—shouldn’t be buying its tanks and planes and laser-guided missiles from China, that tariffs can serve a role there.

But economists don’t like tariffs for a few reasons. First is that they’re costly. A tariff is a tax. It’s a tax typically borne by consumers and importers. Those consumers and importers typically are poorer, so it’s a regressive tax, meaning: More burden is paid by poorer people. They spend a larger share of their incomes on, say, tariffed bananas or whatever.

But the second reason is that they are very inefficient taxes, meaning—so good tax policy is: You want a very broad base, and you want it to be very transparent, and you want to minimize gaming and other things that can poke holes and make the tax less distortionary.

A tariff doesn’t qualify for anything I just said, right? It’s applied on a narrow set of products. It’s very opaque. Unlike a sales tax, you don’t get a receipt on that pickup truck that says, Oh, you just paid an extra 25 percent for this, right? It’s subject to all sorts of gaming because tariffs will vary, typically, based on the type of products. You get what’s called tariff engineering, where you’ll classify—I’ll go back to cars. There’s a famous example: Ford vans were imported without seats to get a lower tariff, and then, literally at the docks, they installed the seats and then drove them off to the warehouses. So it’s a really distortionary and inefficient way to raise revenue or do anything else you want to do with them.

The other big thing, though, is that they’re pretty ineffective at boosting the companies that are getting protected and the workers that are getting protected. For example, I mentioned we have tariffs on shoes. Some of them are ridiculously high, more than 34—almost 40 percent. We have not saved any shoe jobs in the United States. We have almost no jobs in shoe manufacturing. You basically are just having consumers pay a tax for little to no good reason. And in case after case after case, what you see is: Most companies that are protected by tariffs either end up going away after the tariffs are lifted, or they’re seeking perpetual protection, right?

The other big thing is that tariffs, by insulating companies from competition, discourage them from innovating. If you have a guaranteed market, you’re probably not going to be hyper-focused on staying lean and mean and really focused on delivering the best value to your customer. You will get fat, lazy, and happy. You’ll spend a lot of money on lobbying to maintain the tariffs, less money on being productive.

For example, U.S. steel. So there’s probably no industry in the history of the United States that has received more protection than U.S. steel. It’s definitely on the Mount Rushmore of protectionist industries. And U.S. Steel is notoriously inefficient, in part because of that protection. It’s now trying to be bought out by Nippon Steel, a Japanese company. And the goal to—supporters of that deal, including U.S. Steel, by the way, say that Nippon Steel will help it innovate, provide it with better management practices, an influx of capital to upgrade its services.

So put that all together, and economists say, You get high cost, you don’t achieve your objectives, and this is pretty bad. And then you throw in—the historians have looked back at tariff history, especially in the 19th century but even most recently. And tariffs are really historically associated with corruption and cronyism. And that goes back to them being kind of a hidden tax. Also, they target foreigners, and that makes it easier to sell. The economics of trade are counterintuitive. And so tariff policy is notoriously corrupt. And so there’s a lot of political dysfunction, along with just hiring all those lobbyists to get special tariffs or special exemptions. But also, it’s just a very politically perilous policy, as well.

Demsas: You said a lot there. And I want to dig in on a few of these things, but I think as a broad overview, obviously, the idea is: You have to do a benefit-cost analysis of tariff policy. And you’ve obviously articulated a lot of reasons why there are high costs to tariffs, but, as you mentioned with national security, for instance, there are a lot of noneconomic things that policymakers are concerned with that they may want to use tariffs for. And so you think about the implications of what tariffs are trying to do, and often there’s this goal of, We want to spur some sort of industry in the United States. Often, it’s domestic manufacturing, right? You kind of asided to that with the shoe example.

But there’s a history of this, right? Actually, last week, we just had on the show Oliver Kim, who was talking to us about the East Asian development miracle. And one thing that a lot of East Asian countries are credited with doing is having protected native industries and ensuring that those industries were able to succeed on the world market. And there was a lot of protectionism that was involved in doing that, including tariffs.

And so what I guess I would ask you is, firstly, do you feel like that is a goal the U.S. government should have of trying to spur domestic manufacturing? Do you think that’s an important goal?

Lincicome: No. At least not via tariffs. I think there is a million things that the United States government could be doing to boost the manufacturing sector. I should note, of course, the United States is the world’s second-largest manufacturing nation in terms of output, in terms of productivity. So the stuff we make per worker—we’re absolutely crushing it. No. 1 in the world, basically, for large, industrialized nations, so it’s not like the United States is this weak, nothing-burger nation when it comes to manufacturing.

But that aside, there’s a couple caveats I think you need to include when you talk about Asian protectionism and industrial policy. First is: That came with a lot of free trade too. While, certainly, there was some protection for certain industries, there was also a lot of exposure to competition in export markets, in particular, but also in import markets. And, though, there was a lot of tariff liberalization for the things that manufacturers they were trying to support—that they needed. So it wasn’t this just blanket protectionist policy.

The second big thing, though, is that there is a bit of a correlation-versus-causation thing in a lot of East Asian industrial-policy narratives because they were doing a lot of other stuff at the exact same time. And there’s a great book by Arvind Panagariya, who actually looks at South Korea and Taiwan and others and says, Actually, these economies performed better when they weren’t being protectionist—when they weren’t engaging this heavy-handed industrial policy—than when they were. So we need to be a little bit cautious there.

But the third, and I think the most important one for the United States, is that the East Asian miracle applied to a radically different economy than the one in the United States in two big ways. One: Those were developing countries really trying to push infant industries, whereas most U.S. protection is—I mean, the U.S. is certainly not a developing economy. We’re a very developed economy. And most of our protection actually goes to lagging industries. It is not on the cutting edge, and one of the reasons—we have a lot of cutting-edge stuff. But typically, our protection goes to, again, shoes and steel and stuff like that—legacy industries.

But the other thing is that the United States has far-more-developed capital markets than Asian economies did—very open, very fluid. And that means we have much-more-efficient investment where there might be the potential for that success and that innovation. And so it’s less likely that government planners in the United States are going to be able to pick the right industries, pick the right companies, pick the right whatever, as opposed to capital markets and VCs and private equity and all that great stuff. And in general, though, it’s just a radically different environment than what existed in, say, South Korea in the 1970s.

Demsas: But then let’s take a look at the CHIPS and Science Act, for instance, right? That’s the 2022 law Joe Biden signed to bring semiconductor manufacturing to the United States. So during the pandemic, there’s a real concern about semiconductor chips, that we’re not going to be able to have as many. There’s obviously this big shortage. We’re really reliant on Taiwan, which is, of course, concerning because of its proximity to China and the threat that China poses to Taiwan’s freedom.

It’s clear that there is a need to produce, at least in—if not domestically, we need to “friendshore.” We need to make sure can get those supplies from ally countries that we’re less worried about having some kind of future political risk with, but also just domestically because there might be supply-chain problems in the future that are unprecedented, like a global pandemic that we had not been expecting.

And so the CHIPS Act is an industrial policy where there is a real push to get chips made here in the United States. We have factories opening up. I believe they are already producing chips. There’s an Arizona factory.

Lincicome: Yeah. TSMC is not quite up yet.

Demsas: Okay. Not up yet. But basically, we brought Taiwanese expertise to the United States, and they’re building here. We have American jobs that are being created here. And you may care about parts of that or not, but that seems like a policy where that’s on the cutting edge. It’s not confusing to make these chips, but it is a cutting-edge technology. It’s not a legacy industry. So how do you view the use of protection there?

Lincicome: Yeah. Two things: One is that it’s really important to start by noting that this CHIPS Act is subsidies and not tariffs. Now, Biden just imposed some tariffs on semiconductors from China but, in general, the CHIPS Act is just about throwing money at companies.

In general, if you’re going to ask an economist, What would you prefer: a domestic subsidy or a tariff? they’re going to say, A subsidy, nine times out of 10, right? That’s important because you’re at least—granted you’re subsidizing the production, but you’re at least—once the company gets up and running, going to be subjecting it to market forces and competition and its production and output and the rest. You’re not going to be artificially raising prices for downstream consumers and that kind of stuff. So a subsidy is definitely preferable to a tariff.

And in fact, we actually applied tariffs to semiconductors in the 1980s. We had a big industrial policy push in the ’80s related to chips, Japanese memory chips. We applied a bunch of different tariffs, any dumping duties. There was a trade agreement restricting Japanese semiconductors. And what happened? Well, it raised the price of semiconductors and pushed computer manufacturers offshore from the U.S. computer manufacturers. So tariffs, again—historically not very good at achieving your objectives.

But the other thing with the CHIPS Act is: It is starting to reveal some of the problems with industrial policy that we saw back then too. For example, back then, we actually picked—we, the government—picked the wrong type of semiconductor. The Department of Defense in the ’80s thought memory chips were going to be the big innovative thing of the future. So we targeted memory chips. Well, it turns out that the entire industry was actually moving towards logic chips, which are what we use today. And the government totally missed that, while imposing all of those costs.

Right now, we might have a bit of a similar situation because you mentioned TSMC—and TSMC is a global leader. Okay. Cool. But also, the biggest subsidy recipient was Intel. Intel is our national champion. Intel is struggling like crazy.

Intel is slated to receive as much as $45 billion in total subsidies because the CHIPS Act had grants, loans, and tax credits. So Intel is really in trouble.

So did we, once again, pick a loser, along with TSMC? So that’s, I think, a concern we have to deal with. And that’s a traditional issue with industrial policy. Now, why did Intel get all that money? Well, Intel is an American company. Intel has an army of lobbyists in Washington, was instrumental in getting the CHIPS Act passed. Intel decided to locate its facilities in Ohio, a politically important place. And thus, there are questions about whether the government should, again, be giving $45 billion taxpayer dollars to a struggling company like Intel.

Demsas: You’re pointing out a glut of good reasons why it’s not the most optimally efficient policy. But it seems obvious to me, at least, that it’s important for us to make semiconductors here or at least friendshore them. Is there an alternative way to do this?

Lincicome: Yeah. Sure. Well, let me say one more thing about TSMC’s [fabrication facility], and then we’ll move onto your question. The other problem—and the thing I’m worried about—is that we’re actually not subsidizing bleeding-edge technology. TSMC’s fab that’ll be up and running next year is going to be very small, relative to its factories in Taiwan, and it’s not going to be producing the tippy-top-most innovative chip. It’s going to be producing four-nanometer chips instead of the industry two.

It’s also insanely costly. Apparently, it’s costing about 50 percent more to build. And then, of course, a lot of other chip companies that aren’t TSMC are getting money, too, and not just Intel. And they’re getting money to produce what we call legacy chips. So these are clunky commodity chips that really have no security or even, really, innovative nexus. So I think we should be concerned. I don’t know the answer yet. You know, it’s still early in the ballgame, but there are some warning signs.

Now, what could we do instead? A lot, because the big reason why companies weren’t producing a lot of chips here—although that’s a bit of a myth. About half of all chips consumed by American companies were still made in America before the CHIPS Act. But beyond that, we did lose some bleeding-edge capacity. Now, why did that happen? No. 1 reason is because of Intel. Intel was at the frontier and then totally botched it at 10 nanometers and has just become extremely behind the curve. So it’s just a corporate decision-making thing, nothing related to industrial policy.

But the other big reason is because it costs a darn fortune to build a semiconductor facility in the United States. Now, some of that is just because we’re the United States. Things are more expensive than in a developing country. But a lot of it is permitting issues and materials issues and immigration issues. The semiconductor industry is one of the most immigrant-dependent industries in the United States.

So tax issues, as well—we tax the construction of large structures at a much higher rate than we tax things like software and the rest. So you combine all these things, and there’s a free-market path to encouraging the onshoring of large manufacturing facilities, whether it’s semiconductors or anything else, and you could have tax reform and immigration reform and trade reform. Maybe we don’t put tariffs on construction materials and steel and everything else. So that’s a big part of it.

And to the extent even that didn’t do the job, then you could see a role for the government to provide a targeted subsidy for national-security-related chips, so things our Defense Department needs or the tippy-tippy-bleeding-edge stuff that we need for, like, government supercomputers and the rest. But we didn’t get that. You know, that’s maybe a $5 billion bill. And, instead, we got this $60 billion—and then plus another $200 billion in potential tax credit—slush fund that just goes to anything and everything. So I think that’s a problem. And that’s a problem with industrial policy. What starts out as maybe a decent idea on paper just morphs into kind of a political albatross.

Demsas: The only argument that I’ve seen that defends broad-based tariffs—because very few people will defend the, like, 60 percent tariff on goods from China, 20 percent on everything else the U.S. imports. I don’t think we even grow bananas. Even stuff we don’t actually make, no industries—coffee, stuff like that.

But the one argument I have heard is that, while you don’t see increases in domestic manufacturing from these smaller tariffs, if you were to do this really broad-based tariff, it would just force industries to invest in the American economy, because American demand is just both lucrative but also, it would just reshape how capital markets thought about where to invest in companies. It would reshape the kinds of entrepreneurship that would happen, because now we do have to figure out how to satiate this American demand that they’ve been priced out of buying these cheaper goods from abroad.

So setting aside the question about whether or not that would be good for the American consumer to have to now pay double or triple or whatever it is for these basic goods, why wouldn’t that work? Or what do you think would happen in a world where you actually saw these massive tariffs? You can go even higher. Like, you can say 60 percent tariffs on all goods outside the United States. What would actually happen here?

Lincicome: Yeah. So basically North Korea, right? And I joke, but the reality is that tariffs also come with a deadweight loss, an economic loss in terms of economic growth and the rest. Yes. The United States is a big, diverse country with a very productive workforce and a lot of smart people and wonderful capital markets. But if you started imposing giant tariff walls, you’d have a few problems—the biggest being slower economic growth.

By pushing workers into less productive industries, you would effectively be ensuring that the workforce, as a whole, is less productive. That means lower wages, less innovation outside of the sectors you’re targeting, right? You would push a lot of workers and resources into lower-value production. And let’s just leave aside the fact that you’d need giant greenhouses for bananas and stuff like that. We’ll leave that out.

Demsas: Or we just don’t have bananas. No bananas. Yeah.

Lincicome: Right. Heaven forbid. But I do think that’s the other thing that you would have to also consider. You would not just have lower economic growth and slower wages, but I think non-financially, it’d be a lower quality of life. And the grocery store is a wonderful example of that. I can remember back in the 1980s, the grocery store was not nearly as incredible as it is today. And a lot of that, today, is owed to open trade, globalization. And you would lose some of that. You would lose the variety and some of the things that make our lives richer. And I don’t just, by the way, mean bananas. And I don’t just mean food, although that’s a big one.

We have this big globalization series going on. And we talk about fashion and film, and you can go down and on and on down the list. And there’s a lot of aspects of trade and open markets that make our lives fuller and richer in ways that aren’t just about where we’re working or how much we’re making, right? And so that would mean a little less, if not a lot less, of that too. I mentioned at the beginning those cool European and Asian pickup trucks we don’t get. Well, we wouldn’t get those either. We would just have fewer varieties of those things, even if, let’s assume prices are a little bit higher. Sure. But we just also wouldn’t have the variety.

Demsas: I agree with you on this, but then it also gets to a point where sometimes I’m talking to people, and I realize there’s a difference in value. Some people don’t care about this, or they think it’s less important. They think that if we could get more manufacturing jobs in the United States, then it’s okay for us not to have bananas. It’s okay for us not to have a great variety of trucks. Is that stuff important? And I wonder, doesn’t this fall then down to political value judgment about what kind of world looks best?

Lincicome: Yeah—yes and no, because I think if you started saying things like, Well, would you accept less medical innovation? Would you accept less scientific innovation outside of that? because resources are finite—so I think that if you gave people the fuller picture of the price of autarky, I think they would recoil. Particularly if you added things like, And also, your 401(k) is going to be smaller. Your houses are going to be smaller, there’s going to be less resiliency, not more.

You might remember the baby-formula crisis, right? Well, we made all baby formula in the United States, except—because of protectionism. We had walls— tariff wall, non-tariff wall—around the country. Ninety-eight percent of baby formula consumed here was made here. We had a one factory closure, and the entire supply chain collapsed for a year. So you would have actually a more brittle economy than a more resilient one. We would not, at the end of the day, enjoy the much lower living standards, overall, just because we had a few more manufacturing jobs that people don’t even want.

[Music]

Demsas: We’re going to take a quick break. More with Scott when we get back.

[Break]

Demsas: Something you mentioned earlier on I always think is interesting is: The connection between tariff loving and immigration hating I always find very bizarre. We’re at low unemployment right now, so if you’re trying to spur more people to work in domestic manufacturing, it means you’re moving people out of other industries to work towards manufacturing. And if you have the kind of broad-based tariffs that are being proposed by the Trump-Vance ticket—I mean, they’re proposing, like, 60 percent on goods from China and up to 20 percent on everything else from the U.S. imports. These are massive, massive tariffs.

That sort of thing means that you’re going to have the U.S. producing a ton more of the goods that Americans consume. And that would indicate that you would want more people coming here and working here. But at the same time, they’re opposed to immigration. So why do you think the anti-immigration and pro-tariff sentiments have gone hand in hand? They’re trying to deport millions of people too. I forgot about that.

Lincicome: Nationalist sentiment, right? Look—I don’t think there’s a lot of logic or coherence in most economic nationalist arguments. And I think this is just a great example of it, for the reason you said, right? This isn’t 2014 anymore. Native-born employment has flatlined. We are an economy that needs more workers if it wants to grow at the rates we have grown accustomed to in the past. And that means we’re going to need just warm bodies. Just in terms of warm bodies, we’re going to need more of them. And obviously, immigration is a great source. I mean, babies are great, too, but they take at least 18 years to become workers, right? So we can’t do that tomorrow—at least, not that I’m aware of. I don’t know what the science is doing—

Demsas: Latest technology? I think it’s still 18 years.

Lincicome: Right. So we’re going to have a while on that. So immigrants are the obvious source for, you know—if we’re going to be making toasters in America again, like J.D. Vance wants, we’re going to need workers to do that. And robots are great, but robots can’t fill the gap entirely, particularly, again, in the near term. So there’s a huge disconnect there.

And the other thing I’d note is that native-born Americans, in general, just don’t want to work in manufacturing. And this is something totally missed. We at Cato did a very expansive poll over the summer, asking people all sorts of questions. One of the questions we asked was a two-parter: One, Do you think the U.S. should have more manufacturing jobs? And it was, like, 80 percent yes. Yes. More people should work in manufacturing. Then we said, Do you want to work in manufacturing? And it was, like, 20 percent said yes. It’s almost the exact flip.

There was a great article in Bloomberg a couple of years ago about furniture manufacturing here in North Carolina, talking about how they can’t find workers. And this was pre-pandemic, so it’s certainly gotten worse since then. You look at—the textile-manufacturing jobs in South Carolina pay $11 an hour to start. These are not the glamorous jobs that a lot of our politicians think they are. So to the extent we want these jobs in the United States, I am ambivalent. I want the market to determine that. Big surprise. They’re just going to have to come to the reality that we’re going to need more workers to do that. And, again, immigration’s the source.

But there is another thing that I think the nationalists miss entirely, is that free trade actually can help reduce some of the immigration pressures in places like Central America, for example, because it’s going to boost the local economies and boost the stability of these places. Because a lot of immigration is that push-pull, right? People are living in terrible places. They’re like, I got to get the heck out of here. But also, the U.S. economy’s pulling them in. So to the extent that a trade agreement—and allowing companies to access the U.S. market to sell us shirts and stuff like that—can actually boost the local economies in places like, say, Guatemala, that’s going to actually reduce some of that push pressure on immigration, legal or otherwise.

And there’s a fantastic study that actually showed everything I just said, most recently, and it said that you could reduce illegal border crossings by several hundred thousand if you had truly free trade with Central America for textiles, for the reasons I just described. So is that a panacea for the border issues? No. But would it help? Yes. And it is completely lost on our anti-immigration, anti-trade folks, the idea that trading more with places would actually reduce some pressures for more immigrants. They just want no trade and no immigrants, which just doesn’t make a lot of sense.

Demsas: I want to get into some of the reasons for why tariffs haven’t been able to increase domestic manufacturing. There’s a really great study. Aaron Flaaen and Justin Pierce at the Fed—I hope, Aaron, I’m saying your last name correctly. I apologize if not. And they had this study where they looked at the Trump tariffs—the 2018, 2019 tariffs—and they find that the U.S. industries most exposed to tariff increases experienced reductions in employment.

And they also find that counties more exposed to rising tariffs show increases in unemployment—more people are unemployed in counties that are more exposed to rising tariffs—and, of course, declines in labor-force participation. So people are just exiting the labor force entirely there. Why is that happening? Because why is even this narrow case of tariffs—they’re big tariffs, but they’re nothing like they’re being proposed now—why did that not improve domestic manufacturing?

Lincicome: Right. For the moment, let’s just leave aside that the vast majority of us work in services. And if you work in services, you’re basically hurt by tariffs, regardless of anything.

Demsas: Okay. This is one of my hobbyhorses, that whenever everyone talks about the working class, we pretend like everyone’s a manufacturer, but really everyone’s in the service industry, and it’s like, No one cares about those people. McDonald’s? Don’t care about them. It’s just bizarre.

Lincicome: It’s crazy. Even for male-dominated professions—because we’re all worried about men not working and stuff—there are four times as many male-dominated, blue-collar jobs in services than there are in manufacturing. And we never talk about any of it, like you said. Whether it’s construction or security or repair, like automotive repair, you name it, there’s tons of jobs. Nobody talks about them. But anyway, we’re going to ignore all of those folks.

Demsas: Just like our political leaders.

Lincicome: Right. We’re going to ignore them. Sorry. Sorry, folks. We can get back to them later.

Manufacturing—there’s three big reasons why tariffs actually harm American manufacturing. The first is that American manufacturing today is very much global. About half of everything we import into the United States is actually stuff used by American manufacturers to make other stuff—things like steel or machinery and semiconductors. The huge example of that is: The most advanced semiconductor-production technology comes from the Netherlands.

We import that equipment to support semiconductor production in the United States, right?

Demsas: These are intermediate goods.

Lincicome: Yeah—intermediate. Oh, look at you! Nice. Yes. Exactly.

Demsas: (Laughs.)

Lincicome: When people use trade wonky terms, I’m always impressed. That’s great.

So all these intermediate goods—you raise the price of those goods, which tariffs do, and you raise costs for manufacturers. That means those manufacturers spend less on employment and investment and the rest. You’re just raising their costs. It’s like a corporate tax but only for manufacturers that consume imports, which, again, is most of them.

The second big channel is the export side, and that is through retaliation. Foreign governments typically don’t just sit there after a tariff is imposed on products they’re exporting and say, Oh, you got us. We’re toast. No. They retaliate. And they retaliate because they have their own domestic political considerations. They have strategic considerations about preventing even more tariffs. So that harms American manufacturers that export—American manufacturers that are already hurt because they’re facing higher import costs. So those companies are getting hit two ways: higher input costs and retaliation.

The third channel is currency, and I won’t get into the weeds, but tariffs tend to increase the value of the domestic currency. So the dollar gets stronger. As the dollar gets stronger, there’s a good thing: That means that imports get a little cheaper. So it’ll offset some of that tariff pain. The bad thing is that it makes exports more expensive, and anybody who’s gone abroad and has a really strong dollar knows you can buy a ton abroad. That’s actually an import. You’re getting cheap imports. But if the dollar gets really weak and you go abroad, it’s the opposite. So just kind of think of it—it’s kind of those mechanisms, right?

So those three channels, effectively, eliminate any benefit that manufacturers might get from tariff protection. And thus, like you said, the literature tends to show that countries with higher tariffs don’t have wonderful trade surpluses or burgeoning manufacturing industries. And in the United States, the empirical research from the Trump era shows much the same thing.

Demsas: You’ve talked about the narrow cases in which tariffs make sense to you, which I think, largely, is around national security. But I think once you accept that logic, then it just becomes a political question about what things people value, right?

There is this sense that people really care about protecting the manufacturing legacy of specific areas in the United States. And this is, I think, a legacy of 2016, when a lot of people were surprised by the victory of Donald Trump to the presidency. There was a lot of indexing on the fact that he won Wisconsin, Michigan, and Pennsylvania and seeing that this narrative—that he really spoke to the white working class who had been disaffected by free trade.

And this, of course, is right when the “China Shock” paper is becoming really central to the discourse. And so there’s a level here where I wonder if there’s a political-narrative thing that’s going on here, too, where, regardless of all the stuff that we’re talking about, if people want to win national elections, is this just necessary?

Lincicome: No. I’m a firm believer that a lot of what’s going on with our protectionist moment right now is political. The conventional wisdom in Washington today is that, to win national elections, you need to win a handful of gettable votes—so Obama–Trump voters, basically, people who flipped—in a handful of important places, mainly in the industrial Midwest. And to win those votes, you need to offer lots of protectionism and industrial policy too—manufacturing-centric policy.

And I think that is the reality—the conventional wisdom is. I think that is the case. I don’t necessarily agree with it, but I’m not a political consultant, so I won’t dare to question it. And there was a good paper recently by Autor, Dorn, and Hanson, another person—the “China Shock” authors—that said that Republicans did gain a little bit in places, thanks to the tariffs. Even though those places didn’t actually benefit economically, the tariffs were a political winner for Republicans, thanks to the idea that they were being protected. They weren’t actually being protected. The economy was actually a little worse. But they thought they were, and they were rewarding politicians for that.

So I think that is the case. And it’s unfortunate because, first, I am not entirely convinced that tariffs and protectionism were what tipped the 2016 election.

There’s a lot of other stuff bubbling under the surface. But the other big thing is: You actually look at the effect of import competition on these places pre-Trump, and it’s not nearly as devastating as the narrative makes it sound. Whether it’s the China Shock or NAFTA or anything else, these things undoubtedly had a small but significant negative effect on certain places, but it was small. There’s a lot of bigger things going on in terms of manufacturing job loss, in terms of communities surviving or dying.

There’s a great study a few years ago from Brookings that found that, like, 80 percent of old industrial cities in the United States had transitioned to successful economies—places like Pittsburgh. So not every place ended up being like Youngstown, Ohio, right? Yet there’s this narrative that it was all trade, and every place got crushed. And that’s just not the reality, you know?

And the other thing we ignore entirely is interstate competition. A lot of the jobs in the Rust Belt manufacturing—they’re still in the United States. They’re just not in the Rust Belt anymore. They’re in the Sun Belt. We don’t talk about that at all, either. It is all trade, trade, trade. And I think that’s really unfortunate.

At the end of the day, what does that do? It means that the real solutions—and there are a lot of policies that could be pursued to help people adjust, to give them better training and education, to help them move if they need to move by lowering housing prices (you know all about that)—we don’t do any of those. Or, at least, we don’t focus on those. Instead, it’s like, Ah. We’ll just slap a tariff on a toaster, and suddenly Youngstown will be thriving again. And that’s just not reality, not just in the literature. It just doesn’t make any sense. But that’s politics for you, right?

Demsas: Yeah. Yeah. I also think that one of the things that I wanted to get your—because you’ve thought about this for years as someone who’s working in trade. The political dynamics of tariffs, I think, are really important to understand. I think, broadly, my question for you is: Why are tariffs so popular if they’re so harmful? What is going on that, if you’re right, it’s creating all these problems, from baby-formula shortages, which is extremely politically costly, other kinds of shortages during the pandemic—very, very costly. If it’s leading to lower growth—all this stuff—what’s happening? Why doesn’t the political party just win 300 electoral votes by campaigning against tariffs?

Lincicome: Right. Because they are extremely politically attractive to voters.

There’s a guy named Bryan Caplan who wrote a wonderful book several years ago called The Myth of the Rational Voter. He’s a George Mason economist, libertarian guy. But this is more political-science oriented. He ticks through a bunch of biases we all have. And bias is kind of a bad connotation, but I don’t mean it that way. I just mean things that we innately feel.

And tariffs check all of the boxes: an anti-foreign bias, a make-work bias. We like things that produce jobs, right? We have a status-quo bias. Like, we want to protect things that we see that are right in front of us. We are less inclined to want the unseen or the things we don’t know. We can, in fact, fear them. You can go down the list, and tariffs check all of those boxes. So that’s the first thing. Voters innately think, Oh, that’s great. You’re going to protect jobs with that tariff. Wonderful.

But beyond that, the economics of tariffs are hard. It is counterintuitive that a tariff might actually reduce manufacturing output, right? It is counterintuitive, I think, that a trade deficit isn’t necessarily a bad thing. It sounds terrible, right? And it’s counterintuitive that if you cut imports, you actually cut exports too. So there’s all these little things in trade economics that make it a hard sell.

And then, finally: It’s opaque. I mentioned before, when you go to the gas station, you see how prices change. So even some voters that are somewhat connected to the news can be like, Oh, wow. There’s this new conflict with Iran and Israel, and gas prices are going to go up. I get that connection. You don’t really get that with tariffs.

Demsas: So you need a tariff ticker in grocery stores to show—

Lincicome: Yes. I’ve actually long said we need—just the gas station ticker, you need that as well. I think that if you got a receipt from the grocery store and a lot of the line items was the tariff amount, I think that probably would change a few minds. And then, finally, the other thing is that tariffs are oftentimes a corporate tax, and corporate taxes can be hidden. They can either be absorbed by companies or passed on to consumers, again, in invisible ways. And that makes it hard too. So it’s a very, very tough sell.

Now, I’ll note: We’ve known everything I just said for decades, if not centuries. And politicians came up with a fix. It’s called a trade agreement. Trade agreements are not, contrary to popular belief, primarily economic or even about foreign policy. They’re primarily political. There are ways for governments to tie their own hands when it comes to tariff policy. They’re like, I can’t be trusted with this. We went through Smoot-Hawley and all these other bad tariff episodes. We can’t be trusted with guiding tariff policy. So we’re going to delegate it all to the president, which, by the way, that was not the best idea, given Trump. But beyond that, we’re entering into agreements that essentially say that if we go back on our promises, well, what happens? Then the countries we’re trading with can retaliate, there can be litigation and the rest. And that can act as a check.

The other big thing is: We’re going to offset import-competing industries. We’re going to offset their political power with exporters, and trade agreements are going to do that too. Because that’s the other political attractiveness, right? Concentrated benefits and diffuse costs. The benefits of protectionism are very narrow, like the steel industry. Costs of protection are diffused. We all bear those little costs—again, an invisible cost.

So how do you offset that? Well, a trade agreement does that, too, because you have exporters that are like, Oh, but I want access to that market. And I don’t just mean Boeing. I mean financial services and other companies. And so that was the political solution. Now, trade agreements have problems, but they were reasonably successful for 80 years in liberalizing trade, integrating economies, and checking the protectionist impulses of our political class. It was only in the last decade that Donald Trump hacked the whole machine. And we’re basically dealing with the aftermath.

Demsas: Yeah. This diffuse-benefits, concentrated-cost thing—I think it’s just so key. Also, because even after the political constituency has died, it’s kind of hard—in general, once a law gets passed, it’s really durable. Repealing that law, on not just tariff policy but all policies—it kind of just lives on its own. It develops a constituency, whether it’s in the government or outside the government, that wants its continuation. And there’s also very few people who are going to make their political hobbyhorse to do good-governance reforms.

But I write about housing policy a lot. And it’s funny—everyone is talking about housing policy now. Everyone’s talking about how to reduce the cost of housing, make it easier to do construction, all this sort of thing. I’ll have people who are in the Democratic Party or in the administration saying things like, Jerusalem, we need to lower costs. We need innovative ways for the federal government to do this. It’s really hard. It’s all at state and local level.

And I’ll often just say, Hey. Did you guys know there are, like, massive tariffs on Canadian lumber, on Canadian softwood lumber? And they doubled those tariffs in August. And there’s none of this thinking about the diffuse costs to the American people. Like, Congress isn’t working on fixing that. It’s just a level at which I believe that they all care about lowering the cost of housing. I think that’s not a fake thing that they’re talking about here. But we don’t even think about tariff policy when we’re thinking about broad economic costs to the public. We only think about them narrowly in the question of, How does it hurt or benefit this specific industry? and not, What is the harm to the rest of the public?

Lincicome: For sure. And every time you bring up potentially lifting the tariffs that are in place, what happens? Well, big lumber comes to your congressional office or big dairy when the—the dairy industry in the United States, highly protected. When the baby-formula thing was going down, they were vigorously opposing a long-term elimination of the tariffs on baby formula. Now, think about that for a second: baby formula. And these guys are out there, big dairy is out there fighting it. And it worked. Congress has not eliminated those tariffs, even though it’s the most sympathetic consumer possible, right?

Demsas: And it was broadly unpopular. It’s very unpopular, what happened with the baby formula.

Lincicome: Exactly. And every time you scratch a tariff, there’s a crony underneath, and they’re going to fight like heck to keep their windfall profits.

And they’re paying attention. They’re editing Wikipedia pages to make the protectionism sound better. They are the ones laser focused on keeping the protection in place, while the rest of us are like, Well, five cents for some food that’s subject to a tariff, a few dollars here and there extra for a refrigerator or washing machine. Oh, well.

But that stuff adds up, of course. Studies show that if you eliminated all of the protectionism that’s remaining in the U.S. economy—and we’re a pretty open economy—you would save consumers hundreds of dollars a year, if not more. And yet, because it is 10 cents here and 10 cents there, it just doesn’t resonate. And the other side is extremely motivated.

Demsas: Well, thank you so much, Scott. I have one last question for you. And it’s: What’s an idea that you had that seemed good at the time but turned out to only be good on paper?

Lincicome: Yeah. I struggled with this question.

Demsas: Because you’ve always been right? Yeah. (Laughs.)

Lincicome: Well, no, no, no. Because I wanted to find a good one. Self-checkout is my answer.

Demsas: Oh, yeah?

Lincicome: Yeah. I am a huge fan of self-checkout. And being me, I’m also a big fan of just efficiency, right? Waiting in line is terrible. I wrote a whole column about why you should never wait in line, because of the opportunity cost of doing so.

So self-checkout—in theory, self-checkout is this amazing life hack. And I still love it, but I’m realizing that—let’s face it—and companies are realizing that self-checkout is not nearly the labor-saving, time-saving miracle that we think. And that’s because humans, alas, are still human. And for every guy like me who literally treats it like I’m trying to beat my best time ever at Costco when I’m going through the self-check—my daughter’s, like, handing me stuff. I mean, we’re literally gamifying it. It’s so great.

Demsas: This is how I feel in the airport security line. I get so angry.

Lincicome: For every person like me, who’s trying to get out of there as soon as possible and trying to break his own personal record, there are, like, 74 other people who are utterly confused by the technology, in no rush, wanting to maybe chat with the person behind the counter, wanting to pay by a check, confused by their coupons, or trying to steal. That’s the other big thing. And so, unfortunately, it has turned out that self-check is not the miracle technology that I was hoping. So it looked good on paper but less so in reality.

Demsas: There’s a Safeway near my house. I moved recently, so I was checking out the nearby grocery stores. And the self-checkout is, like, I don’t know, an armed state. It’s so insane. You can’t exit the checkout without scanning your receipt. And I usually just throw my receipt away immediately, so I had to go get the receipt out of the trash. It wasn’t even functioning. Someone had to come and let me out and then look at all my stuff and make sure I wasn’t stealing. It was just this level of just—it genuinely would have taken me so much less time to wait in this line. But every time, I still go to the self-check. I don’t know why I’m doing it to myself.

Lincicome: Of course. No. And I have a dream of opening up my own supermarket where we actually time people, and there’s, like, posted records of all this. But no. Alas, it still runs into problems.

Demsas: Well, Scott, thank you so much for coming on the show.

Lincicome: My pleasure. Thanks for having me. Great talk.

[Music]

Demsas: Good on Paper is produced by Jinae West. It was edited by Dave Shaw, fact-checked by Ena Alvarado, and engineered by Erica Huang. Our theme music is composed by Rob Smierciak. Claudine Ebeid is the executive producer of Atlantic audio, and Andrea Valdez is our managing editor.

And hey, if you like what you’re hearing, please leave us a rating and review on Apple Podcasts.

I’m Jerusalem Demsas, and we’ll see you next week.

Lincicome: I’ve really worked this out on Twitter a few times. You’d put a bar right at the checkout area, so people could watch, and stadium seating around it. It’d be great. Scott Mart!