The pundit's dilemma

Trumpism will deindustrialize America. That's a hard pill for some thinkers to swallow.

“There are no American troops in Baghdad!” — Mohammed Saeed al-Sahhaf

Around the middle of 2022, it became apparent that the Biden administration was going to go all-in on industrial policy. The policies that Biden implemented — the CHIPS Act, and the green energy subsidies in the Inflation Reduction Act — had already been in the works for quite a while. But so had a bunch of other Democratic policy ideas that didn’t end up making it into law.¹ So it wasn’t until 2022 that we realized that industrial policy was going to be Biden’s Big Thing.

I was personally overjoyed about this trend. For years, I had worried that America’s eroding manufacturing base made it less capable of standing up to China and other authoritarian adversaries. I was also optimistic that a big push for green energy could accelerate decarbonization both at home and abroad. And for years, I had been deeply interested in the successful industrial policies of East Asian countries like Korea and Japan during their catch-up development phases. I was eager to see America try its hand at something similar, in order to revive its stagnant industrial base.

At the same time, I didn’t agree with everything the administration was trying to do. I realized that if the U.S. did experience a manufacturing revival, it would be by leveraging our strengths in automation and technology — in other words, even in the best-case scenario, the era of good plentiful blue-collar manufacturing jobs wasn’t going to come back. Some people in the Biden administration didn’t seem to understand that, and appeared to view industrial policy as a jobs program.

I also disagreed with some of the ways that the Biden administration tried to implement industrial policy. The “buy American” provisions were unnecessary and unhelpful (local demand will foster a network of domestic suppliers anyway). The “everything bagel” requirements for subsidies were too onerous, particularly the union labor requirements in states with very few union workers.

As one of the more prominent commentators writing about these policies, I had a choice to make. I had to decide whether to emphasize my criticisms of Biden’s approach, or my overall positivity about the idea. Then, as the years went by and we started to see some of the results of Biden’s policies, the choice became even more acute, because some of the policies succeeded and some of them failed.

The administration’s core industrial policies seemed to be bearing fruit; there was a massive factory construction boom, most of it financed with private money (the subsidies acting mainly as a nudge). The TSMC plant in Arizona got back on track after some initial hitches. But at the same time, Biden’s administration failed to build any EV chargers or rural broadband, and new transmission lines were stymied by regulatory hurdles. It seemed like the more the government was directly involved with a project, the less likely it was to ever happen; nudging the private sector worked, but direct government action turned into wasteful jobs programs that built nothing.

I had to make a decision: How vehemently and how frequently should I lambast the Biden administration’s failures, and how vehemently and how frequently should I praise its successes? If I leaned too hard toward praise, I might look like an apparatchik who just parroted administration talking points. And even worse, I might fail to perform one of my core functions as an independent commentator — criticizing policy to help steer it in the right direction. But on the other hand, if I leaned too much toward criticism, the people in the administration might decide that I was their political opponent, and stop listening to me entirely. And I might also throw the baby out with the bathwater by ignoring the real successes.

To be honest, I didn’t actually think very hard about any of these tradeoffs at the time. But perhaps subconsciously, I was making these sort of calculations. In any case, I think I ended up striking a pretty good balance between criticism and praise.

Conservative industrialists, however, are facing a much harder dilemma right now. Biden’s industrial policy was a mixed bag, with more successes than failures. But Trump’s tariff policy is a giant flaming disaster. The dollar is down, as investors flee American bonds, putting the country’s whole financial stability in danger. Forecasts for the real economy are getting more pessimistic by the day. Stocks are down yet again. Here’s a representative headline:

Any time you hear the words “worst since 1932” in connection with the American economy, you know things are not going well.

Given this disaster, commentators who support Trump have a hard choice to make. Do they denounce Trump’s policies, thus condemning themselves to certain excommunication from the MAGA movement, and dropping their influence to basically zero without altering Trump’s course in any perceptible way? Or do they scramble to find some way to put lipstick on the tariff pig?

The latter course of action will preserve their influence within MAGA-land, but will also make them look ridiculous to the rest of the country, who has no such ideological commitments to uphold. And in the process, the commentators will become an accomplice to this act of economic self-mutilation.

Oren Cass, the chief economist at American Compass and a prominent advocate of industrial policy from the right, has chosen the second path. He is continuing to fiercely defend the idea that tariffs are a way to reindustrialize the American economy. In a post on April 4th, shortly after Trump’s “Liberation Day” tariffs, Cass defended the tariffs and made some minor suggestions for how Trump might tweak the policy:

As the economic carnage from the tariffs became more apparent, Cass continued to double down. In a debate with Jason Furman and others a week later, Cass argued that trade had devastated American manufacturing in the 2000s. A few days after that, he argued that many Americans will enjoy working in factories:

[I]f fully 25% of poll respondents say they’d prefer a factory job to their current job, that suggests massive potential unmet by the current labor market…One in four?! That would be 40 million of 160 million employed Americans!…When we talk about reindustrializing America, we are talking about adding millions of manufacturing jobs, not tens of millions—several percentage points of the labor force, not a quarter or more…[P]lease, let’s stop with the “nobody wants to work in a factory,” and even more so the “I don’t see you working in factory.”

And a few days after that, Cass argued that Trump’s tariff strategy is primed to succeed, because meeting American domestic demand is more important than exporting:

Our $1 trillion trade deficit presents an enormous opportunity. Most economic analysis of the American position in the global economy seems to assume that export markets represent the key opportunity for U.S. industry and thus success depends upon winning in those markets. From that perspective, for instance, tariffs on the intermediate goods that manufacturers might use in their own production is a disaster. How can a U.S. producer hope to compete with a German producer if the U.S. producer has to pay a tariff on components from China and the German producer does not?

This assumption makes good sense in the typical developing-country situation where the domestic market is small and trade imbalances are likely a minor factor…But the top priority and major opportunity for the United States is not higher exports, it is recovering the capacity to meet domestic demand with domestic supply…Indeed, the U.S. will be building leading-edge semiconductor fabs to meet domestic demand for a long time before it needs to consider export markets at all.

If you’re going to try and mount an intellectual defense of Trump’s tariff policies — instead of just screaming out memes, pointing fingers, and trying to distract people by talking about immigration instead — this is basically how you have to do it. Nothing good is coming of Trump’s tariffs right now, so if you’re going to defend them, you basically have to argue that they represent short term pain for long term gain. In this case, the long-term “gain” is economic self-sufficiency and a bonanza of factory jobs.

Formally speaking, you can’t prove that this is wrong, except by waiting a bunch of years and then observing that reindustrialization didn’t happen. There are no hard and fast economic laws of the Universe; we have a lot of theories, but economies are complex beasts, and the past is an imperfect guide to the future. I can’t completely rule out the possibility that Trump’s tariffs will cause a vast crop of steel factories and shoe factories and semiconductor factories to spring up from the American topsoil like mushrooms after the rain.

And yet when we look at what’s actually happening to American manufacturing in real time, it doesn’t look anything at all like the beginning of the reindustrialization that Cass imagines. Here’s a quick roundup of news from just the last few days:

• Volvo is cutting 800 jobs at three U.S. factories, citing tariff uncertainty.

• The Philadelphia Manufacturing Survey is plunging, signaling extreme pessimism among U.S. manufacturers.

• Howmet Aerospace, a major aircraft parts manufacturer based in Pittsburgh, has declared that it may halt production due to Trump’s tariffs.

• The April NY Manufacturing Survey is recording some of the most negative conditions that it has ever recorded, with new orders and shipments falling off a cliff:

Source: Heather Long

• The Philadelphia Fed’s survey of manufacturers is also showing a massive dropoff in new orders.

• Falling orders are causing Volvo to lay off hundreds of American workers at two factories.

• Plenty of evidence shows that American manufacturers are pulling back on their capital spending plans because of tariffs:

  Manufacturing pessimism has spiked dramatically in April as industrial producers deal with changing tariff plans and try to assess how global trade policy will impact their costs and operations in the coming months…Several surveys released this week show huge swings in business confidence between January, when most manufacturers had a positive outlook for near-term conditions…and April, when sentiment changed for the much, much worse…Some of the most dramatic declines came from the Equipment Leasing Finance Foundation (ELFF), an organization that represents lenders that help manufacturers obtain new capital equipment for factories. In March, more than half of manufacturers surveyed by that group expected capital spending to increase or stay about the same in the next four months. By April, more than 61% said they expect spending to fall.

• Ford is halting sales of some American-made cars to China. (Note: This will increase America’s trade deficit with China).

• GM is laying off American factory workers too.

• News of various other factory layoffs is proliferating. Cleveland Cliffs, the steel company, is laying off 1200 workers.

It’s not hard to understand why American manufacturing is getting hit hard by tariffs. As I’ve said in many posts, and as knowledgeable folks are screaming from the rooftops, broad tariffs of the type Trump has imposed raise the price of imported components and makes U.S. manufacturers less competitive. Even manufacturers who initially scoffed at this effect are quickly learning what a big problem it is:

And on top of the effect of the tariffs themselves, uncertainty about future tariffs — which has spiked to record levels under Trump — hurts manufacturing even more, because manufacturers don’t know how to plan their future supply chains.

This is not a difficult principle to understand. And yet somehow, Oren Cass appears not to grasp it. In his post that I quoted above, Cass argues that the “imported components” problem only applies to exports, not to manufacturing for the domestic market. That is obviously wrong. For an American auto factory in Kentucky, the cost of components matters just as much whether you’re selling your cars to customers in Dubai or Dallas. Without the ability to source cheap car parts from Mexico and Canada, American manufacturers will see their costs for domestic manufacturing rise.

This is why you can’t just measure the U.S. trade deficit and assume that this amount will be added to U.S. manufacturing if we wall off our economy (as Cass assumes in his post). The amount of domestic demand for manufactured products is not fixed. More expensive components will mean more expensive manufactured products, which will cause Americans to consume less. Americans will become poorer, and they’ll also substitute their consumption away from goods and toward services (whose price will go up by less).

Cass’ hand-waving dismissal of exports also ignores another crucial factor: scale effects. Export markets help American manufacturers scale up their production, which lowers their costs — and thus helps them sell even more to American consumers, capturing more of the U.S. market as well. Commentators like Sam Hammond, who argue that the U.S. should focus on export promotion instead of import substitution, are correct for this reason.

(On top of all that, it’s not clear how much tariffs even reduce trade deficits. As Arnaud Costinot and Ivan Werning show in a new theory paper, it’s possible that even very high tariffs might leave trade deficits mostly unchanged, while simply making an economy poorer.)

All this is theoretical, of course. Cass and other tariff defenders can simply continue to assume that tariffs will reduce trade deficits by a lot, and that this reduction will translate into a whole bunch of new U.S. factories. Perhaps they assume that the unfolding carnage in the U.S. manufacturing industry is simply a temporary disruption, and that after a short period of suffering, American manufacturers will get back on their feet, invest in domestic production capacity, and eventually do better than ever.

Except if that were true, we’d probably see two things. First, we’d see manufacturing stocks doing well, because far-sighted investors would know if American manufacturers would eventually benefit from tariffs. Instead, the stocks of American manufacturers have been crashing. Also, if tariffs provided a protective shield under which American manufacturers could flourish, we’d probably see a bunch of them rushing to invest and build new capacity. But we don’t; capex in the industry is plummeting.

In other words, neither investors nor manufacturers themselves believes in the long-term reindustrialized future that Oren simply assumes will come about. Instead, it appears that tariffs are dramatically accelerating America’s deindustrialization.

And if that’s true, Cass’ other arguments all go up in smoke. Who cares if people like working factory jobs when there are fewer factory jobs to work in? And so on. Every defense of the tariffs is based on this assumption of reindustrialization; if, as looks very likely, that turns out to be a fantasy, the whole edifice collapses.

Oren Cass and other tariff-defending pundits have thus hitched their wagons to a stagecoach that is driving straight off a cliff. They have managed to remain inside the favored circle of the MAGA movement, but this required them to make a terrible sacrifice — they lost their freedom to look out the window and see the calamity unfolding.

I am glad that during the Biden administration, I didn’t have to make any such choice.

---

1. Some of these — like the big subsidies for care industries in the Build Back Better bill — got killed because of lack of political support from the center. Others, like wealth taxes, co-determination, and sectoral bargaining, just never really caught fire.