What would a real anti-China trade strategy look like?

How we would do things if we were serious.

Trump’s current trade strategy will diminish American power and American technological capability, divide the U.S. from allies and partners, and give China an opening to become the world’s preeminent nation. I still think it’s unlikely that this is intentional; there’s an old adage that you should “never attribute to malice that which is adequately explained by stupidity.” The haphazard, last-minute, on-again-off-again way that Trump and his team have rolled out their tariff policy, and the fact that Congress has not chosen to use its power to revoke the President’s tariff authority, suggests that stupidity is the main factor in play here.

But in any case, there are obviously some people within the Trump administration and the MAGA movement who would like Trump to produce a trade strategy that helps to contain Chinese power. CEA Chair Stephen Miran has written that “China has chosen to double down on its mercantilist, export-led model to secure marginal income, much to the rest of the world’s consternation.” And Treasury Secretary Scott Bessent went even further, suggesting that containment of China should be the main goal of U.S. trade policy:

Scott Bessent emerged from this week’s market ructions as a perhaps-unexpected lead trade negotiator, offering a potential scenario for the coming months: US deals with longstanding partners that put pressure on China.

“They’ve been good military allies, not perfect economic allies,” the former hedge fund manager said Wednesday of some of these US friends. At the end of the day, the Trump administration can probably reach an agreement with them. “Then we can approach China as a group,” he said.

The nations Bessent said he’s looking to — Japan, South Korea, Vietnam and India — happen to be neighbors of China. They are countries with which the US could work to isolate China, something that’s been called a “grand encirclement” strategy.

This is actually a very realistic goal. Every day that Trump’s tariff chaos makes the U.S. look like a chaotic clown car makes it a less realistic goal, but as of right now, I still think that it would be possible for the U.S. to radically pivot its trade and industrial policies in order to create a coalition of nations that could economically balance, compete with, and even isolate China. And it’s not too hard to imagine what that strategy would look like.

But first, we should think about why we would want to economically pressure China, and what we might hope to accomplish. After all, in an ideal world, countries simply trade with each other and get rich, instead of fighting. And China has plenty of good stuff to offer the world — cool cars, cheap solar panels and batteries, and lots more. Why should we take an adversarial approach to trade with China?

The reason is geopolitics. Singing hymns to the gains from trade doesn’t change the fact that for whatever reason, the leaders of powerful countries sometimes want to dominate or even attack other nations. The world is an ungoverned place, and the balance of power is the only thing that keeps the peace.

Currently, China has become the world’s preeminent manufacturing nation. Its current leaders also think of the U.S. and many of its allies as either rivals or outright enemies. They appear determined to conquer Taiwan, carve off pieces of India, Japan, and the Philippines, and generally use Chinese power to dominate smaller countries. It makes sense to want to weaken China’s ability to do all this, while strengthening the other nations’ capacities to resist it.

The goals of trade policy with China should therefore probably include the following:

1. Preventing China from gaining an overwhelming military advantage over other nations

2. Reducing China’s ability to exert economic pressure on other nations

3. Reducing supply chain vulnerability in nations threatened by China, so that any future conflict with China wouldn’t crash those countries’ economies.

That doesn’t mean that prosperity and cool cars shouldn’t be goals of China trade policy, but merely that they should be augmented with these other geopolitical goals.

In any case, when I talk about economically “containing” China, that’s what I’m talking about. So here’s a list of things we would do if we were serious about that goal. Obviously this list is very, very far away from anything the Trump administration is doing or contemplating. But this is what I think it would take.

Zero trade barriers with any nations other than China

Manufacturers need scale to drive down costs and remain competitive. One reason China’s manufacturers are so formidable — and why American manufacturers were so formidable relative to their rivals 80 years ago — is that they have access to a huge domestic market. Chinese car companies like BYD can sell untold numbers of cars to their billion consumers; this allows those companies to scale up and drive down costs to levels no foreign competitor can match. BYD is currently building a single factory that’s bigger than the city of San Francisco.

Another key factor that makes Chinese manufacturers so powerful is domestic supply chains. Practically everything that goes into a Chinese EV, particularly the battery, the metal, and the chips, is produced in-country. That makes it very quick and easy for Chinese manufacturers to source everything they need, instead of having to struggle to import it from overseas.

It’s inherently very hard for American manufacturers can match those two advantages. The U.S. is much smaller than China — our consumption is larger in dollar terms, but we have far fewer people, and so our companies can’t ship as many units domestically. Chinese people buy about double the number of cars every year that Americans do.

Of course this problem is even more acute for America’s allies, like Japan and Korea. Smaller countries compensate by finding highly specialized niches to be competitive in. But this leaves their supply chains and defense-industrial bases at a disadvantage; China, because it’s so huge, can more easily create a fully self-sufficient manufacturing ecosystem (which it has, in fact, spent the last two decades trying to do).

The only possible solution way for China’s rivals to match it in size is to gang up. And in this case, what “gang up” means is to form a free trade zone amongst each other, with zero trade barriers between them.

If the U.S. had zero trade barriers with Europe, Japan, Korea, India, and the countries of Southeast Asia, those countries wouldn’t become exactly like one huge “domestic” market. There would still be language barriers, geographic distance, exchange rate fluctuations, and national regulatory differences that end up accidentally restricting trade. But it would go a long way toward allowing American manufacturers — and European, Japanese, Korean, Indian, and Southeast Asian manufacturers — to attain the sort of economies of scale and supply-chain networks that China enjoys within its borders.

Basically, to balance China, you’d need to start thinking of “Non-China” as a single vast economic entity.

If this sounds familiar, well, it should. Two trade treaties, the TPP with Asia and the TTIP with Europe, would have gone a long way toward creating this sort of common market among non-Chinese manufacturing nations. Both were killed by Donald Trump.

But in any case, if you want to economically balance China and reduce economic dependence on China, this is the first thing you’d do.

Tariffs on Chinese intermediate goods, and data collection on supply chains

The next thing you’d need to deal with is supply chain vulnerabilities among non-Chinese nations. The ideal would be to make sure that Non-China has the ability to make everything it needs to make, so that A) Non-China can be self-sufficient in case of a major war, and B) China can’t dominate the nations of Non-China by exerting pressure on key supply chain vulnerabilities (like it’s doing right now with rare earths).

One thing you need here is targeted protectionism. The idea is to prevent China from being able to put Non-China manufacturers out of business with a sudden flood of subsidized exports. For example, suppose China decided to destroy the American, Japanese, Korean, and Taiwanese chip industries by unleashing a massive flood of subsidized computer chips. The only way to prevent this strategy from working is protectionism.

So you need the ability to put up targeted trade barriers very quickly, in sectors that China is making a bid to conquer. Note that this is very different from Trump’s tariff policy — it’s far more targeted in terms of industries, it’s only on China, and it has nothing to do with trade deficits or other macro imbalances. It’s more like the tariffs Biden put on some Chinese products.

But there’s a problem here, which is that standard tariffs don’t hit intermediate goods. If China makes a phone, takes it apart, then ships the pieces to Vietnam, where Vietnamese workers snap it back together and sell it to America, our tariffs think that this phone is “made in Vietnam”. If laptops made in Mexico and sold in America contain Chinese chips, those chips aren’t subject to the tariff rate on Chinese goods — they’re only subject to the tariff rate on Mexican goods. Stephen Miran recognizes this fact in his 2024 note.¹

The solution to this is to apply tariffs not based on the country where something was finally assembled, but to the countries where the value was added. Doing this would allow us to put tariffs on Chinese intermediate goods like computer chips and batteries, in addition to final goods like phones and cars.

Of course, applying tariffs in this way would require much better data collection. We’d need to figure out where the components in each imported good originated. This would require, among other things, a small army of bureaucrats.

Industrial policy for strategic industries

In order to give Non-China a self-sufficient, robust manufacturing ecosystem, we’d need to do a lot more than just stop China from poking new holes in that ecosystem. We’d have to fix the existing holes as well. For example, China already makes most of the world’s batteries and processes most of the world’s rare earths. Those are vulnerabilities that need to be dealt with.

The way to do that is industrial policy — we need to start making things that we currently don’t make (or that we make very little of). Maybe given the right long-term incentives, those industries would reappear in Non-China on their own, but giving them a helping hand fixes the problem much more quickly.

And sometimes, industrial policy can help create robustness within Non-China as well. For example, if Taiwan gets invaded or bombed by China or struck by a massive earthquake, the world’s chip supply could be seriously damaged, because most of the factories of TSMC — the world’s dominant chipmaker — are in Taiwan. Thus, it makes sense to pressure or cajole TSMC into moving some of its factories to safer locations — the U.S., Japan, and elsewhere.

This was the cornerstone of Biden’s approach to industrial policy, with the CHIPS Act for chips and the Inflation Reduction Act for batteries and renewable energy tech. But this was just an exploratory phase — just two sectors out of many. Other industrial policies should be added for other sectors — drones, electric motors, machine tools, robots, telecom, and of course rare earths and mineral processing. They don’t have to be as big and splashy and expensive as the CHIPS Act and IRA, but they should be in the mix.

Of course, it’s not known whether Biden’s approach to industrial policy — which is similar to China’s, though smaller in scale — is the best one. In an interesting post, Balaji Srinivasan suggests an alternative strategy based on government-organized industry consortia like SEMATECH in the 1990s. This is similar to how Japan did many of its industrial policies during its boom years.

In any case, industrial policy should make a comeback if the U.S. and the broader Non-China world wants to compete with China.

Smart pro-investment policies here at home

There’s one more big reason China is such a manufacturing superpower — it has structured its government policies around building lots of factories. That pro-investment policy has introduced macroeconomic distortions, but it has also allowed Chinese manufacturers to iterate quickly, to expand the ecosystem of suppliers, to scale up, and generally to do all the other things that make manufacturing work.

I’m not suggesting that the U.S. allow wholesale pollution of its rivers or kick millions of people off of their land in order to build factories to compete with China. But over the past half century, the U.S., even more than other rich countries, has thrown up a vast thicket of procedural barriers that block the building of new factories. Simply eliminating many of these barriers would go a long way toward making American manufacturing competitive again.

To its credit, the Trump administration has actually been making some moves in this direction. For example, Trump has issued executive orders eliminating a bunch of rules regarding the implementation of NEPA, one of the biggest procedural barriers to development in the U.S. Experts on the harms of NEPA are optimistic that this change could mean a significant weakening of NIMBYs’ ability to block factories, housing, and other development projects.

And although the U.S. shouldn’t aim to invest as much of its GDP as China does, increasing the amount from its current low level should also be a priority. Two policies, suggested by JD Vance and widely believed to be effective, are 100% bonus depreciation and full expensing of R&D spending. The Trump administration is also experimenting with government loans for manufacturers, under the Office of Strategic Capital. That’s a good idea, though of course it’ll be subject to some amount of waste and corruption.

Much more can be done. Private banks could be encouraged to make loans to manufacturers looking to scale up. Export promotion, and promotion of greenfield FDI in manufacturing, are also promising ideas.

In any case, this is all aspirational on my part. The Trump administration is totally focused on its unhelpful and damaging tariff policy. What’s more, zero tariffs on non-China countries, expansions of state capacity, and expanding on the legacy of Biden’s industrial policies definitely don’t seem like the sort of things this administration would be interested in.

But if you did want to turn the global economy into a fortress against Chinese power, this is basically how you’d do it.

---

1. Miran does make a pretty substantial mistake, though. He wrote:

“Freeman, Baldwin and Theodorakoplous (2023) find that, while just over 60% of manufacturing intermediates imported into the U.S. came directly from China, incorporating the value-added of manufacturing intermediates that originated in China but were imported from other trade partners brought that number above 90%.

These numbers are badly off. You can see Freeman, Baldwin, and Theodorakoplous’ estimates in Figure 2.3 in their paper:

This is the “look-through” exposure — in other words, an estimate of the total value-added of manufacturing intermediaries that originated in China. China is ultimately responsible for 3.5% of all U.S. intermediate goods, which is about 20% of the value of imported inputs. Not 90%. Miran is just way, way off base with his numbers here.