It is well-documented now that while Donald Trump may not be particularly ideological, he genuinely seems to believe the United States should deploy tariffs as a foreign-policy cudgel. In his first few days in office, the new U.S. president has already coerced Colombia into accepting deportation flights, while Mexico and Canada remain on tenterhooks to see if he will follow through on his proposal to levy 25 percent tariffs on their exports.

How exactly is Trump’s use of tariffs different from other American presidents? What are the costs and benefits of his proposals? And how might it change global trade? I put those and other questions to Brad Setser on the latest episode of FP Live. Setser is a senior fellow at the Council on Foreign Relations and previously served as a senior advisor to the U.S. Trade Representative.

Follow the free FP Live podcast for an audio version of our discussion. Subscribers can watch the full discussion on the video box atop this page. What follows is a condensed and edited transcript.

Ravi Agrawal: Let’s just start with a very basic definition. What is a tariff?

Brad Setser: I mean, a tariff is a tax on imported goods. It’s a tax the importer has to pay if they want the goods released from the port of entry. So it’s a pretty straightforward tax, actually.

RA: And tariffs have many uses, from protecting domestic industry to raising revenue to providing leverage in a negotiation. But none of this is new. What makes Trump’s use of tariffs so different from that of other American presidents?

BS: I’ll limit the answer to modern American presidents, those in the post-World War II era. Trump literally says that he loves tariffs. He prefers tariffs to other sources of leverage. He calls himself “the tariff man.”

Tariffs have been used for protection, but they haven’t been used very much for revenue until recently. So if Trump carries that out, that would be a significant change. They’ve always been a source of negotiating leverage. But Trump takes that to a new level because he likes the tool. Most presidents have viewed tariffs as a costly imposition on the U.S. economy or as something to negotiate down as part of a trade deal.

Trump is also very open to using tariffs as a source of leverage for goals other than trade. All of those make Trump’s use of tariffs a bit unique.

RA: Brad, you used the word costly there. Everything is a trade-off. Tariffs generate revenue, but how inflationary are they?

BS: I wouldn’t call it inflation. They are a one-off change to the cost of living. So in most instances, if you’re importing a good for a dollar and there’s a 10 percent tariff on it, the importer is going to pay $1.10. There’s some cases where the exporter may lower the price a bit, but empirically that effect has been very small. And then the importer will send that $0.10 to the government; they’ll either absorb that cost out of a lower profit margin or they’ll raise prices to the consumer. Now, many people perceive that as inflation, but I think the standard definition of inflation requires an ongoing process of price increases. This is a one-off change. If you’re passing this cost on to consumers, the consumer who is spending more on an imported avocado, for instance, will have less money to spend on a haircut. So you can see some reduction in demand for other goods throughout the economy.

So, if the tariffs are used as a revenue-raiser without offsets through other budget expansionary policies, like a tax cut, then there’s not more spending. I don’t actually think it’s a sustained change in the price level. That said, the way inflation is used in common parlance is much more about the price level, not the change in the price level. And so while the Fed may be willing to look through a one-off jump in prices, consumers certainly will perceive that as inflation.

RA: Let’s take a real-world example of this. So let’s say the United States does indeed impose 25 percent tariffs on Canada and on Mexico. In economic terms, what happens next?

BS: It’s actually a pretty bloody big shock, to be honest. So I doubt they’re going to do 25 percent across-the-board this weekend, just because of the scale of the shock.

But let’s start with the basics. We import some $400 billion of goods from Canada. The biggest component of that undoubtedly is oil, and a particular kind of oil that is heavier and tougher to refine. The refineries in the Midwest that take that specific grade of oil are optimized to use it. So it’s hard for Canada to find an alternative market, and it’s hard for us to find an equally efficient source of supply. We import roughly $500 billion of goods from Mexico, the dominant category of which is autos and auto parts. So it’s a big shock in the first instance to our auto sector. And then we also export a decent amount to Canada and Mexico. So it’s very much a two-way trade, unlike the more one-way trade with China.

As a share of GDP, imports from Mexico and Canada combined are about 3.2 percent of GDP. Exports are about 2.4 percent of GDP. So, the first-order effect is we’re imposing a 25 percent tax on 3.2 percent of GDP, which has a just-pay-it cost of about three-quarters of a point of U.S. GDP. There isn’t going to be any offsets either if you do it over the weekend. It’s just going to be a tax increase, and not a small one. Now, there could be adjustments that reflect the fact that this is potentially a temporary tariff. So maybe GM will run down its inventory in the U.S. so it doesn’t have to bring new cars and SUVs from its factories in Mexico and Canada into the U.S. And it will build up inventory in Mexico and Canada in anticipation of a deal. But if there is no deal, GM will eventually have to pay those tariffs. Same with auto parts. So then you’re going to have to decide whether to raise the price of the car that has Mexican or Canadian parts. More immediately, we import a lot of berries and vegetables from Mexico: avocados, out-of-season fresh produce. Those costs will almost certainly be fully passed on. So if you take this just-pay-it cost as an approximation of the overall impact of the economy, give or take, you get a meaningful shock to the U.S. economic momentum.

If Canada and Mexico retaliate, it’s over 2 percent of U.S. GDP in exports to their economies, that’s a complicated choice. We export some things that are pretty important for their ability to make goods. It’d be hard for Canadian factories to operate without U.S. parts. On the other hand, if we’re tariffing parts coming in, they may feel obligated to tariff our parts going out. And so then we would typically see some adjustment down in our exports as well. Certainly in the first instance, you would see a fall in our exports because again, if it’s temporary, folks would try to find ways to not import things across the border and instead, use up inventory in anticipation of a deal. But over time, you’re going to end up paying. And then you’re going to have to have a negotiation between the Canadian oil producers and the American refiners to decide how much of a discount to the current price of oil will those producers accept to sell into the U.S., knowing that the refiners are going to have to pay 25 percent? So you’re really looking at a shock to the U.S. economy of a percentage point of U.S. GDP if you do 25 percent and it stays.

We’re very integrated with Mexico and Canada. It’s much more difficult to put these kinds of tariffs on without feeling it than it was with the tariffs on China, which also were rolled out in phases. So that’s why I think it’s unlikely you’re going to see the full 25 percent across-the-board. It’s too costly to ourselves.

RA: You touched on retaliation. It strikes me that different countries will have different responses. We saw recently how Colombia gave in and agreed to accept deportation flights after Trump threatened massive tariffs. How much does the size and strength of a country matter when it comes to retaliations or tariff negotiations?

BS: Well, if large blocs stick together, they tend to have the capacity to impose bigger retaliation in response to our actions. So in that sense, there’s strength in being a bigger economy that can retaliate more effectively. One asterisk: China’s economy’s kind of unbalanced, so China had fewer retaliation options than you might expect for an economy of its size.

Your retaliation options are also a function of your willingness to hurt yourselves. Are you willing to tariff parts knowing that will limit your ability to produce for the global market, or not? That said, for very small economies, they’re very conscious of the fact that if they are not willing to fight on occasion, they run the risk of being bullied. So sometimes countries will stand up even if it is not in their straightforward interest. I think in Colombia’s case, they’ve been able to argue that they didn’t completely fold because they addressed concerns about the treatment of Colombians on the flights by sending their own aircraft to pick up their citizens. There usually is scope for some kind of negotiated outcome. But countries don’t like to be picked on, and bigger blocs have a capacity to punch back.

RA: It strikes me that Trump, perhaps more than other recent presidents, is extremely sensitive to market reactions. If you are being threatened with tariffs, are there any stock market levers you can pull to try and play the negotiating game Trump is asking you to play?

BS: If I want to be a true geo-financial strategist, I would say China, through its reserve managers and the China Investment [Corporation], actually holds a decent portfolio of liquid U.S. equities. So if it really wanted to play the stock market game, China could start selling some of its own equity portfolio. But in general, the countries who are on the receiving side of the threats can impact the stock market primarily through how they choose to retaliate or threaten to retaliate. But much of the reaction is going to be a function of the nature of the tariffs and how those tariffs impact U.S. firms relying on imports.

RA: The more the United States uses tariffs as a foreign-policy cudgel, what are the chances that other countries look for other routes to grow their trade without the United States? Ruchir Sharma pointed out in the Financial Times recently that over the last eight years, some 80 percent of countries have seen trade rise as a share of their national GDP. But the United States is a big exception here. It’s growing faster than its peers, but with no boost from trade. Is the United States waning as a trading power?

BS: If you just look at the export side of our economy, we’ve been waning as a trading power for a long time. That’s not solvable just through the Trump trade policies. It takes changes in tax and changes in our exchange rates. But the amount we export as a country is roughly at the same level, in proportion to our economy, as it was in the 1980s. So we haven’t been an exporting power for a long time. And exports as a share of U.S. GDP have basically been trending down for the past 10 years, not because of Trump’s trade policies in his term, but primarily because the dollar’s been pretty strong and because we have a tax policy that penalizes companies that export from the U.S. So they can achieve much lower tax rates by putting their intellectual property outside of the U.S. We encourage companies not to produce in the U.S. and instead to produce abroad and just to book offshore profits. So there’s a set of policies that have made the U.S. a very weak exporting power. I think we have to be honest with ourselves. That is a real problem. It’s not because of unfair trade practices abroad.

On the importing side, we are the 400-pound gorilla. And I disagree with Ruchir Sharma’s overall assessment in some important ways. Our import growth is still the main factor that drives global trade growth. When U.S. imports shot up during the pandemic, that drove global trade. Recently, China has been exporting like crazy, and the United States is the market that’s seen rapid growth in its imports. It’s a recovery in consumer imports. So I don’t think it is true to say the U.S. no longer matters to global trade. We matter enormously to global trade.

The fact that we have been willing to import without much exporting and that we run a very large aggregate trade deficit is the reason why many countries around the world can run trade surpluses. The U.K., India, Turkey, all have deficits, too, but the biggest is in the U.S.

There’s a tendency by Ruchir Sharma and others to think that it is easier to get around the U.S. than it is. There is no other country in the global economy that runs a $1.2 trillion goods trade deficit. Without that, a whole bunch of global trade would come to an end.

That said, if we become a coercive trading power, if we use our leverage to, for instance, try to force Canada to become the 51st state or to completely transform its trading patterns, Canada will reevaluate whether it should have its economy fully integrated with the U.S. economy. It’s normal to have more intense trade with your neighbors, particularly when you share a continent. But if your neighbor is a bully, is coercive, is insisting on unfair terms, then Canada may decide it will be better off reorienting its economy to be more integrated with Europe, even though it will be costly.

RA: That hypothetical threat on Canada’s part was put forward by its former Canadian finance minister, Chrystia Freeland, recently.

If you interpret some of this as bending the rules—or interpreting the rules—in ways that favor you, what does that mean for the global trading system? What does that do to the rules-based order of, say, the WTO [World Trade Organization], if rules don’t matter as much?

BS: I think there are two systems. There’s the WTO system, and there’s a broader sense of informal rules and norms. One of those informal rules and norms of the past 70 years is that you don’t use tariff threats to try to expand your territory. That’s not written into the WTO, but that’s been an informal norm. It’s implicitly written into the WTO, because you’re only supposed to use tariff threats for violations of the WTO.

The big issue is, are you going to use tariffs to not only incite a change in the WTO system of rules, but a change in the broader sense of global order. One where big countries have not sought territorial expansion through the threat of tariffs or through armed force. That would be an enormous change. And I think it’d be a very destructive change. The United States hasn’t been following WTO rules for a while. WTO rules basically stipulate that you don’t treat China, Canada, Japan, and Europe differently. You can treat Canada a bit better because you have a free-trade agreement with Canada—USMCA—but, in theory, you’re going to treat all members of the WTO equally. As a matter of practice, the United States imposes much bigger tariffs on China because of the 301, which was done outside of the WTO process, than it imposes on Europe or on Japan. So we are in no way fully compliant with the WTO principle of nondiscrimination right now. The bottom line is we want to be tougher on China than on other WTO members. And that’s a fundamental problem, in my view, with the current set of rules. Those rules are basically out of date. And I would go further and say China doesn’t really follow the WTO rules either. The rules were not designed for an economy that delivers subsidies through off-budget government investment funds and then has further government investment funds run by local governments. And then, the state banks offer very low-cost financing, because they are state banks and they have to follow the dictates of [Chinese President] Xi [Jinping] and the goals of the party. That puts China outside of the rules while it pretends to be inside of the rules.

So, we’re going to be outside of the WTO rules in a lot of areas. The Biden administration was outside of a narrow read of the WTO rules in the Inflation Reduction Act. And its unequal provision of subsidies for EVs and other products to all WTO members. The Trump administration was outside of the WTO only in their arguments, and there are exceptions in its steel tariffs and in its China tariffs. But other countries have argued the United States used it too broadly in a way that wasn’t consistent with the WTO norms.

But if you’re not following the WTO rules, people are going to ask, “Well, what rules are you following?” If you’re, instead, expanding your foreign-policy goals to include things like convincing a country to sell part of its territory to you or convincing a country to give up some of its territory, like the [Panama] Canal, that is an enormous expansion of the use of tariffs. And it goes beyond not following the rules to inventing enormous new rules.

RA: As we think about how this new Trump administration is going, about the implementation of tariffs or the threat of tariffs, why do you think they’ve begun with Colombia and Mexico and Canada, but not China, the one that was talked about so much more on the campaign trail and for which there is more history?

BS: I will confess that I am not close enough to the people inside the Trump administration to have any clue why they chose this particular order. Presumably it’s a function of the president wanting to start putting pressure on countries that he thinks need to do more on migration.

With China, the other factor is clearly there’s a bit of a desire on the part of both Xi and Trump to have a fancy summit meeting and see if some sort of deal can be done. It is also the case that tariffs in China, beyond the current tariffs, will generate a certain set of tensions. About half of the trade that existed when Trump took office for the first time in 2017 has already been tariffed to 25 percent. Some of the remaining trade is tariffed to 7.5 percent, which doesn’t have a huge impact, and some are categories which Trump didn’t want to tariff when he was in charge last time. The most important example is iPhones. Trump decided during the China trade war to exclude iPhones because [Apple CEO] Tim Cook told him, correctly, that Apple makes its phones in China and Samsung makes its phones in Vietnam. So, putting tariffs on China is putting tariffs on Apple, while giving Samsung an advantage. And Trump didn’t want to do that.

So, you’ve already hit the low-hanging fruit in tariffing China. You’re going to have to pick some sectors where trade with China was less disadvantageous last time. China clearly has a capacity to retaliate. China does have the ability to buy a lot more U.S. stuff and do things that led to the so-called phase one deal from just before the pandemic.

Trump’s administration has to decide whether the end goal is decoupling or is your end goal balanced trade, where China buys a lot more, probably somewhat arbitrarily using the power of the state to divert Chinese demand toward the United States and away from others and therefore create what appears to be a more balanced trading relationship? I don’t think it is a complete surprise that it will take some time to figure out what the approach to China is, strategically and economically.

RA: One more question on China. It’s often cited that more countries count China as their biggest trading partner than they do the United States. And as we think about potential future tariffs on China, where does that leave U.S.-China power dynamics globally? Because China has leverage not just with the United States, but also with all these other countries, some of which are American allies. China could hurt other countries through other retaliations, but American tariffs elsewhere could also push precarious American allies into Chinese arms.

So, how are you thinking about this new era we seem to be entering of “tariff wars”? And how the two biggest economies in the world are wielding them or could wield them to essentially carve up the world?

BS: I don’t accept a lot of the standard foreign-policy discussion that says that China can just step in and substitute for the U.S. on trade. Why? China has a very unbalanced pattern of trade. Yes, it’s many countries’ biggest trading partner. That’s because a lot of countries import a ton from China. And if you’re not making commodities and you’re not Taiwan selling high-end chips, you export next to nothing to China. China’s imports of manufactured goods as a share of China’s GDP have been falling since 2005. And if you take out imports of parts, China’s manufacturing imports are less than 4 percent of its GDP. It is squeezing manufactured goods out of its economy. And Xi’s pretty explicit about that. So, trading with China is a heavy lift if you’re not just sending goods to China to be sold to the rest of the world. If you’re really relying on Chinese demand, that’s a tough one.

The United States, by contrast, has been a big net source of demand, and manufacturing demand in particular. So I don’t think you can easily swing from trading with the United States, with our manufacturing deficit, to trading with China and its manufacturing surplus. China is a horrible trading partner for most of the rest of the world. It’s been a horrible trading partner for Europe for the past few years. Europe’s deficit with China has soared. It’s been a horrible trading partner with Korea. It squeezed Hyundai out of the Chinese auto market. And if you look at Chinese exports last year in volume terms, exports were up close to 15 percent. Imports were only growing at 2-3 percent. That’s an enormous discrepancy.

The problem for the rest of the world is that if you trade with the U.S., you have to deal with Trump. And if you trade with China, it’s not clear that you’re going to actually be able to sell anything other than commodities to China.

I think the alternative is forming tighter relationships with countries that are not China, like European or Latin American countries. Now, certainly some of its neighbors in Southeast Asia are more willing to form closer trading partnerships with China. But I note that Southeast Asia, as a whole, runs a trade surplus. And China, as a whole, runs a very big trade surplus. So, even if they grow more tightly with each other, unless they get rid of that combined surplus, they’re still going to have to trade a lot with the rest of the world. Europe runs a trade surplus. The Saudis are going to soon swing to a trade deficit because they are building these new cities in the desert, so maybe you can sell some of this surplus to Saudi Arabia. But, Saudi Arabia is not that big. So you really do come back to a world where you need a big economy to keep the current pattern of trade. There really isn’t an obvious substitute for the United States and its deficit, and, frankly, for India and its deficit. So there are limits to how much people can go around the United States. That said, if you don’t take China as the center, if you take Europe as the center, there’s a little bit of a surplus. So I’m actually bearish on the ability of China to replace the United States.