The problem with Trumpian mercantilism
"You can't just pour fertilizer into water and get a tree."
I get messages from people (sad people!) who still believe in free trade. I urge them to read Carlyle on the condition-of-England question.
Here is a modernized version. Suppose you had a drone which was a time machine: it could visit history, even alternate history, but only to fly over it and take video. When you watch the video of two versions of a city, you decide which one is doing better. Then, you check the actual statistics to see if you were right.
Imagine any independent observer looking at any minor American city or town in 1955 and 2025, with this “condition-of-America” drone. After comparing enough footage to be sure, you will find the idea that 2025 is doing better than 1955 outright preposterous. Now—view 2025, but if America only imported minerals we couldn’t mine ourselves, crops we couldn’t grow ourselves, etc, etc.
Do you really believe ours is the bestest 2025? Check the actual statistics to see if you are right. (Unfortunately, you can only do this with the real 1955.) Of course, I am doing a bit—the idea that you would believe a number over your own eyes is an insult. At least, I hope it’s an insult. “Check the actual statistics to see if you are right.”
This is just one of the ways in which Trump is right and the experts are wrong. But… the problem with Trumpian mercantilism is the problem with Trumpian everything.
Trump always has the right reflexes. But a reflex is not a plan. It is not his job, but the job of his administration, to translate reflexes into plans. While executing with great energy and enthusiasm, the administration has had a rocky start in this translation.
A reflex without a plan is not action, but the illusion of action. Action is defined by its purpose. The only purpose of a reflex is to do the right thing. The purpose of an action is a step toward an end state. A reflex has no end or intent. An action is proactive. A reflex is reactive. An action is strategic. A reflex is tactical. And so on.
Because of the influence of libertarianism and the beauty and power of spontaneous order, not to mention the essentially mythic characteristics of their historical faith (when thinking inside a myth, we suffer from just-world fallacy by definition), the conservative tends to operate in terms of reflexes. Nature, he assumes, will provide. Perhaps there is even some relic of the old Puritan doctrine of Providence.
Providence will get it done. Providence will provide. Providence will defund the bad, and fund the good. Providence will hire the staff. Providence will pay the lawyers. Providence will build the factories, get the permits, train the engineers, etc, etc, etc.
The purpose of the reflex is not to accomplish these things—but to show Providence that we are worthy of her favor. The fundamental hurdle of any project is spiritual. Inevitably, in the age of theater, the spiritual becomes the dramatic. The result is a kingdom of mere gestures—a cargo-cult empire, which cannot last long.
Closely related to providential thinking is the belief that America is a young country. Lol. America has every sign of old age that an aging empire has ever possessed. In the young, everything heals as if by providence. In the old, nothing, nothing fixes itself.
In case none of you zoomers is 50 yet, here is what happens. Things don’t just heal. While a human being is not a salamander, an old human is really not a salamander. Like: six months ago, I sprained my ankle. It still hurts, a little. (Also, did you know—aging is exponential?)
Sure: mercantilism in the 1970s would have saved America’s industries. Instead, free trade—championed by conservatives—killed them. That was then and this is now. Time does not run backwards. It is hard to heal. It is easy to kill. It is impossible to resurrect. It is difficult to rebuild—maybe even more difficult than building.
Imagine if the US after 1950 had never accepted manufactured imports—that it chose to import only raw materials, not manufactured goods. I am old enough to remember when an imported car was a novelty. What would the downside have been? Answer: the downside would have been purely hedonic. (Never forget that “GDP” is literally a metric of hedonism.)
But no, the post-hippie elite needed to puff around in their VW Beetles and park in city spaces with their little Honda Civics. The foreign goods were just better. French wine is better for the price than California wine—even in California. And for that purely hedonic victory—America looks like it was physically devastated by a war. Then you look more closely, and you see that there are no bullet marks. (Not saying that Soho, Palo Alto, etc, aren’t doing great.)
Go look at pictures of heavy industrial sites—factories, mines, etc—from the pre-1976 years when we ran a trade deficit. Or even the pre-NAFTA or pre-2008 years. Try to imagine which of these sites could be rehabilitated, and at what cost. No—as in any developing country, probably best to start with light industry.
In many ways, America is much more advanced than the China of 1980 or the South Korea of 1950. But it seems intuitively much more difficult to reindustrialize America, an old and rich country, than mid-century Asia, a young and poor country.
It is a matter of providential thinking to expect that such a thing will happen without planning. An new national production infrastructure needs to be designed and built. Because this is America, the new supply chain should not just be one big company. However, when you are doing anything fast, centralization is the most efficient—especially when chicken-egg conflicts abound.
For example: consider a plan to reshore American consumer goods. Everything you can buy in a Wal-Mart will be made in America. Cool! But—by when? By who? How? Where? With what equipment, what personnel, what raw materials?
In any modern supply chain, many links in the chain are useless without other links in the chain. If these links have to establish themselves as fully independent enterprises—imagine building the first car this way; who needs carburetors or mufflers, when there are no windshield wipers —it is near-impossible for any of them to get started. If our car companies died, the US would find ourselves in the automotive position of Cuba—able to import cars and able to repair them, but not able to make cars.
Tariffs provide a latent financial incentive, a kind of ambient corporate nutrition, to reshoring enterprises. That’s all. Even to harness that nutrition is a lot of cellular work—and its concentration, even with the initial Trump tariffs, is probably insufficient to nourish any significant growth of even primary industry, let alone a full supply chain. You can’t just pour fertilizer into water and get a tree.
Think of a supply chain as like a human body: full of complicated, wonderful organs, which turn into rotten offal if they lose their financial blood supply for five minutes. Many of us could use a body transplant. It takes 15+ years to grow one the normal way, the organs are only semi-compatible, and they yell when you try to cut their heads off.
Therefore, it’s better to print one out in a tank, from cell scaffolds laced with stem cells and growth hormones. We pump blood through these scaffolds till they mature into organs. Elapsed time: 26 weeks, or 20 with the express high-pressure package. Obviously, we don’t print a brain. But if we did, it would be ready to take the SAT.
In exchange for this insta-body, we accept the price of central planning. In the long run, firms are as small as possible, and markets will send the price signals that keep the body healthy. But to build the body—we need a big, complicated central printer.
Generally, the skills to make anything still exist in the US. Sometimes they need to be enormously expanded. Did you learn to code? Then AI put you out of work? Bummer. Maybe you can still be a machinist. (AI will never come for the machinists.)
What would a serious industrial reshoring look like? Picture it as a World War II style project. What would General Groves do? Dude would plan. He would make a central plan, like Stalin. He would build all the parts of a centralized supply chain, right down to the consumer products—making these parts logically independent organs that will become self-governing and self-financing companies once the whole body is running.
But at first—it would be somewhere between a vertically integrated manufacturer like Tesla, and a Japanese style keiretsu or zaibatsu, Korean chaebol, etc. These were created for the same purpose: rapid, systematic industrialization force-fed by finance. Maybe you want two or three competing ones, so they keep each other reasonably fit.
The financial energy for the creation of this system should come from three places: VC finance (which knows how to pick operators and stand up companies fast, and has tons of money), government debt (debt is fine when it finances tangible investment, not consumption), and (of course) American consumers.
Tariffs create the incentive for American consumers to buy American. If there is no way to buy American, or if the incentive is too low, the nutrient just hangs in the water and ferments. The tariff is just a consumption tax. Which is fine—but has to be understood as what it is.
Because taxes change prices, including capital prices, it is best to plan them carefully. If the plant is not ready for the nutrient, why put it in the water? The incentive should be crafted so that it appears when the plant will be ready, in just the concentrations it will need. Trump’s tariffs seem too high in the near term, too low in the long term.
Ideally, the US has a plan to end manufacturing imports entirely within the next four years—by ramping up tariffs to the point where they are simply unaffordable. Given the security that it will be impossible to import toasters, except some kind of insane Topf & Sons German luxury toaster, in four years, even venture capitalists will see a legitimate profit opportunity in investing in toaster factories.
It might be best to centralize the new American heavy industry into specific zones, rather than trying to reintroduce industry into even old brownfield industrial areas. Regulation can be relaxed in special industrial areas—endorrheic basins are a good place for new industrial chemical plants. Pollution debt is another kind of debt—it can be easier to get to a mature clean industry by passing through an early polluting phase, then imposing regulations and cleaning up the pollution. Most light factories, though, can go anywhere.
Most of all: industrial self-sufficiency is only one of the purposes of such a project, and not the main one. The main purpose is to match labor demand to labor supply—to give every American not just a job, but a good job matched to his or her skill set. Ideally, no one should even have to move—but moving is not the end of the world.
A major piece of the design for reshoring is actually an understanding of the human component—the labor supply that needs to be re-employed. Who are they? Where are they? What are they used to doing? What are they generally good at? What can they be taught, and what can’t they be taught?
One way to define this question is to imagine if the government gave every adult American a button to resign from their life. No, not suicide—not Canadian style MAID. You could just say: this isn’t working out for me.
And the government would say: okay. You’ll go here. You’ll live there. You’ll do that. And at the end of this process, you’ll have a new life with a new career that works for you and/or your family, more or less permanently. Maybe where you are. Maybe in Boise. If it has to be Boise, is that so bad?
Personally, given modern technology—especially the AI robotics stuff coming down the pike—I am not super convinced that this can be done with just factories. The age of Dickensian factory labor, with its human robots, is over. The “dark factory” arrives.
Economically, a robot or even a machine is just another type of foreigner—without any of the human externalities foreigners can create. They use less healthcare, but more electricity. A guest worker is half a foreigner. The robot is a Gastarbeiter from the machine world, whose support contract is his remittance.
The ultimate tariff is simply the requirement that goods be made by hand, by humans. In this future America, when we buy clothes, we have to buy them on Etsy—and Etsy’s rules are enforced by the FBI. Every time they capture a floating brick of Temu shirts, dropped by some fleeing speedboat, they display it on a card table for the local news.
“These Chinese junk clothes were trying to take American housewives’ businesses away. Now, Sergeant Hernandez will set them on fire. Look how well they burn—that’s pure polyester. Just the fabric, without a historical-clothing tag, would get you five years for artificial fiber, aggravated. Imagine if we still used that stuff! Why, the shepherd could not afford his bread and cheese, the cotton-picker his energy drink… everywhere, the economic fabric would collapse. It’d be like the late 20th century.”
If you like this kind of stuff, read my previous trade post, or just: