The Case for Flat Tariffs

Amid all the critique of the Trump administration’s trade policy, it’s easy to forget that there is now broad agreement among the Washington cognoscenti that tariffs are at least sometimes an appropriate and effective tool of public policy. These days, prestige media is full of intellectuals and former officeholders writing essays defining the use cases for tariffs, which are always narrower than the president’s, but not so narrow as to exclude them from the policy menu entirely.¹

These intellectuals embrace what they call “strategic tariffs.” These are targeted tariffs on advanced manufactured products, such as electric vehicles and batteries, but only when exported from China. The approach is best summarized by Jake Sullivan, former national security advisor to President Biden, who explained that the Biden administration’s trade policy would be a “small yard and high fence.”² In a 2024 speech to the Brookings Institution, Sullivan explained further:

In multiple sectors that are important to our future, not just critical minerals, but solar cells, lithium-ion batteries, electric vehicles, we see a broad pattern emerging. The PRC is producing far more than domestic demand, dumping excess onto global markets at artificially low prices, driving manufacturers around the world out of business and creating a chokehold on supply chains.³

In response to this crisis, the Biden administration used Section 301 of the Trade Act of 1974 to tariff a list of advanced manufactured products from China, including all of the products named in Sullivan’s speech: a paradigmatic use of strategic tariffs.⁴ Sullivan contrasted strategic tariffs with the unsophisticated alternative:

Now, we know that indiscriminate, broad-based tariffs will harm workers, consumers, and businesses, both in the United States and our partners. The evidence on that is clear. That’s why we chose, instead, to target tariffs at unfair practices in strategic sectors where we and our allies are investing hundreds of billions of dollars to rebuild our manufacturing and our resilience.

President Trump ignored Sullivan’s warning and has chosen to be “indiscriminate.” Following the Supreme Court’s rejection of global tariffs under ieepa the administration is using an alternative statute, Section 122, to apply a temporary flat 10 percent tariff to virtually all imports from all countries, in addition to existing country- and sector-specific tariffs under Section 301 and Section 232. The original plan was to use Section 122 as a stopgap until the administration could impose more durable tariffs under Section 301. These would be designed to have broad effect and replace the ieepa tariffs that the Supreme Court rejected.

Now, however, that plan is threatened from all sides. The Court of International Trade issued an opinion holding that Trump’s use of Section 122 was legally invalid, although stopping short of issuing a universal injunction that would have directly blocked the flat tariff’s enforcement.⁵ Meanwhile, industry groups are doing their best to weaken the administration’s plan for broad global tariffs under Section 301. At hearings for the scope of the planned Section 301 tariffs, Politico reported that “prominent technology groups and trade associations” urged the administration to “reject broad-based tariffs and instead pursue narrowly targeted action focused on . . . China.”⁶

Even as predictions of spiraling inflation and strangled growth were falsified after more than a year of flat tariffs, the Trump approach still has the odor of the disreputable: the kind of thing you support if you don’t understand economics. Intellectuals invariably analogize it to a blunt force object, a hammer, or a wrecking ball. By contrast, the strategic approach is always a scalpel, which, in the words of economist and writer Noah Smith, “allows you to slice out exactly the types of imports you don’t want, while leaving the less strategically important stuff intact.”⁷ It sounds like the kind of precise operation well-suited for intellectuals, which is likely why they find it appealing. Indiscriminate hammering is for cavemen.

But the “strategic” approach ultimately has more to do with the aesthetics of strategy—that is, the performance of cognitively intense tasks like forecasting and targeting—than the object of strategy, which is to win the game. Most would agree with Sullivan that winning looks like “rebuilding our manufacturing and our resilience.” Yet while targeted tariffs on advanced manufactured products from China are essential, they are not enough on their own to accomplish this goal. For the game we are in, the crucial strategic move is to pair high tariffs on China with a global flat tariff. It is high time for the American policy elite to put down their scalpels, hit the gym, and start swinging blunt force objects.

Global Tariffs Address the Scale of Chinese Distortion

First, let us consider Sullivan’s statement of the problem. China is “dumping excess onto global markets at artificially low prices, driving manufacturers around the world out of business [emphasis mine].” In other words, Chinese overproduction is squeezing producers globally. Sullivan should follow his logic through to the conclusion. If it is worth excluding subsidized Chinese exports from the American market because they are artificially low-priced, why wouldn’t this same policy apply to exports from the rest of the world when they embody this same price distortion?

Chinese overproduction is a global phenomenon, not a targeted measure aimed at the United States. In product markets where prices are set on the margin, producers around the world must price against a Chinese supply glut, creating a race-to-the-bottom effect in which the subsidized Chinese price pulls down the global price. That price renders key sectors uninvestable for firms seeking market returns.

The response from the left-of-center policy elite has been to subsidize those key sectors while applying limited tariffs to China. But subsidies for firms that sell into a global market impacted by Chinese pricing are just money thrown down a black hole. Without price support, all the subsidies in the world can’t turn a production model into a self-sustaining business. And price support against China alone does not actually address the scale of the price distortion that China creates.

The critical minerals sector reflects this dynamic. The Biden administration applied strategic tariffs of 25 percent to critical mineral imports from China. Between 2022 and 2025, however, global spot prices for lithium declined by 80 percent and prices for cobalt, nickel, and graphite dropped by 40–50 percent due to China-driven oversupply.⁸ Despite clear strategic need, the International Energy Agency reported that global investment in critical mineral development and exploration was flatlining, as “low mineral prices are not providing the signal to invest.”⁹ It is world prices, not merely Chinese prices, that block U.S. producers from investing in a sector critical to American innovation and resilience.

The same effect can also be indirect. When China floods the world with subsidized exports, this deprives industrial firms in Europe and Asia of their own export markets, in addition to putting their domestic markets under pressure. Under normal conditions, these firms would simply exit the market, bringing supply back in line with demand. Instead, European and Asian governments subsidize their struggling industrial firms, further distorting global prices and creating new excess capacity outside of China.¹⁰ These firms also become newly desperate to sell into the United States, the one major market that is broadly insulated from the effect of Chinese exports. Without a global tariff, the United States would find itself on the receiving end of a diverted flow of exports that need somewhere to go after China pushed them out.

Essentially, Chinese industrial overcapacity is like a cancer that has already spread. Cutting out China alone isn’t enough, because the effects of its industrial policy have already been seeded into the rest of the world.

A logical solution would be to get the rest of the world to tariff China, cordoning off its distortionary price effect in a kind of anti-China customs union.¹¹ Over the longer term, this goal should be the animating principle of American trade policy. But this cannot be achieved in the short term, as the coordination required to effectively contain China from global markets is immense. Countries seeking to do this would not only need to tariff China, they would need to tariff every country that trades freely with China, or else Chinese exports would still be able to penetrate the union through third countries.

Tariffs on European and Asian allies are often presented as hampering U.S. diplomatic efforts to contain China. They are in fact an inducement to cooperation, helping us to more swiftly form an anti-China bloc.

When the U.S shuts itself from China, China diverts its exports away from the U.S. and towards the rest of the world, forcing everyone else to undergo painful “China Shock 2.0” deindustrialization just as we did in the 2000s. Trump’s tariffs caused U.S.-bound Chinese exports to drop by $130 billion in 2025, but China replaced this volume by cutting prices and increasing shipments to Europe by $60 billion and ASEAN countries by $100 billion.¹² The consequences of these diverted Chinese exports on Europe were detailed in a recent column by longtime Economist trade commentator Soumaya Keynes: massive job dislocation, a shriveled industrial base, and one-sided dependence on Chinese manufacturers.¹³

If the U.S. market were completely open to the rest of the world, European and Asian industrial firms might be able to survive Chinese export pressure by increasing their own exports into the United States, dodging the need to take action on China. As per usual, a global imbalance would be “solved” by the United States absorbing it.

But a global tariff from the United States closes this outlet. Europe and Asia are forced to confront the consequences of absorbing Chinese overproduction alone. This creates a mutually reinforcing dynamic in which Europe and Asia become highly incentivized to tariff China while their need for access to the U.S. market, as a massive source of China-free demand, grows.¹⁴ We gain leverage to ask for cooperation on excluding China from the global trading system at the very moment that this enters the self-interest of trading partners. Indeed, Keynes, an erstwhile supporter of the WTO-based system of free trade, titled her column “Why Europe must embrace tariffs,” on China, not the United States. The winds are shifting.

The Trump trade team has seized the opportunity. Commentators on Trump’s trade policy expected retaliation and a spiraling trade war following Liberation Day. Instead, because of the dynamic described above, the administration was able to sign “Agreements on Reciprocal Trade” (“ARTs”) with 19 trade partners. These ARTs reduced tariffs on partners from the 30–50 percent range to 10–20 percent, in exchange for cooperation on a range of anti-China measures including collective tariffs, export controls, sanctions, investment screening, and critical minerals access, in addition to substantial commitments to invest in strategic U.S. sectors.¹⁵ These ARTs have not yet been operationalized, and the Supreme Court’s recent restriction of the president’s tariff power threatens their durability. Trade scholar Peter Harrell concludes, however, that the anti-China economic security provisions “could well outlast many of the other elements of Trump’s trade deal agenda.”¹⁶ The existence of these agreements demonstrates that U.S. global tariffs induce cooperation, rather than conflict, on the United States’s key strategic interests.

China, however, has an even more subversive method of distorting the global trading system than subsidized exports: it is increasingly externalizing production itself. The combination of massive government subsidization of production and chronically low domestic consumption has placed unprecedented margin pressure on Chinese manufacturers, who must go abroad in search of profits.¹⁷ Exports have exploded, but so has Chinese investment in offshore production, from aluminum smelters in Indonesia to electric vehicle plants in Eastern Europe to lithium mines in Australia.¹⁸ Goods produced in these countries reflect distorted Chinese production incentives and artificially cheap capital, but avoid tariffs applied to China-origin goods.

Take Spain, where the leftist prime minister has welcomed Chinese investment as a counterbalance to Trump. In March 2026, Chinese battery maker Hithium agreed to open a €400 million factory at a site in Navarra, which, incidentally, used to host a Bosch refrigerator plant before it shuttered in the face of cheap Chinese imports.¹⁹ The Hithium plant will further a trend in which annual Chinese investment in European manufacturing has tripled since 2022, to $12 billion. Under the existing rules of the global trading system, the batteries produced at this factory will be Spanish, and will benefit from whatever trade arrangement exists between the United States and the European Union. But, as the Financial Times reported, such factories “ultimately answer to Beijing.”²⁰ Their presence in global commerce is a result of Chinese industrial policy, not market forces, and distorts incentives for battery production just as surely as if the batteries had been produced and exported from China.

The usual policy response to this problem is to require rules of origin for favorable tariff treatment, in which exporters certify that their goods contain below a certain threshold of Chinese content or processing in order to avoid China-level tariffs. These rules are good so far as they go. But they are highly complex to administer and easy for importers to avoid through fraudulent certification.²¹ Further, rules of origin do not cover goods that are produced wholly outside of China, using non-Chinese inputs, but are nonetheless reflective of Chinese capital, control, and industrial policy.

This dynamic exposes a fundamental flaw with “strategic” tariffs. They consign us to a game of international whack-a-mole. When tariffs apply only to a specific geography, producers will look to change their geography, either on paper or in reality, instead of their trade practices. China can stand up manufacturing operations in virtually any country in the world. It can transship, circumvent, and evade any tariff order applied to it. It has, in the words of former U.S. Trade Representative Michael Froman, “near-unlimited capacity to mobilize capital and manipulate trade and investment policy in service of its long-term objectives.”²² If it needs to manufacture outside of China, it will manufacture outside of China. And that’s precisely what it’s doing.

The United States expends significant resources combating such behavior. These enforcement efforts have succeeded at reducing imports from China below levels not seen since 2001, when China joined the World Trade Organization.²³ This is a significant win for the cause of decoupling, but it is not a complete strategy. Without a global tariff that creates a baseline level of geography-agnostic protection for American producers, we become wholly dependent on a costly investigative process to discover Chinese content and capital behind every shipping container that arrives on our shores.

A global tariff is thus not merely better “targeted” to the actual, worldwide price effects of Chinese overproduction. It cuts out a meaningful part of the benefit that China would gain by evading even those tariffs directly applied to it, all while creating an inducement for tariffed countries to cooperate on excluding China and reducing bureaucratic complexity. This makes it the policy most likely to actually alter fundamental trade flows rather than incentivize gamesmanship.

Indeed, macroeconomist Michael Pettis has argued that only global tariffs, not bilateral tariffs, are capable of structurally shifting American consumption toward American production:

The American economy more or less automatically absorbs excess production from trade partners who have implemented beggar-my-neighbor policies. It is the global consumer of last resort. The purpose of tariffs for the United States should be to cancel this role, so that American producers would no longer have to adjust their production according to the needs of foreign producers. For that reason, such tariffs should be simple, transparent, and widely applied (perhaps excluding trade partners that commit to balancing trade domestically). The aim would not be to protect specific manufacturing sectors or national champions but to counter the United States’ pro-consumption antiproduction orientation.²⁴

Taking these factors together, a coherent strategy comes into view. The United States can initially combine high tariffs on China with a global flat tariff designed to reshape fundamental trade flows. Later, the United States could lower the global tariff for countries that commit to (1) mirroring high U.S. tariffs on China; (2) mirroring high U.S. tariffs on countries that do not commit to item (1); and (3) balanced trade within the anti-China bloc, such that the United States can increase the global tariff in proportion to its overall trade deficit with members of the bloc. Former U.S. Trade Representative Robert Lighthizer recently proposed such a system.²⁵

Tariffs on China should always remain significantly higher than tariffs on the rest of the world, to ensure that U.S. importers have an incentive to decouple away from Chinese suppliers and to express our special concern, rooted in national security, with commercial dependence on Chinese production. But the kind of China-specific tariff championed by wonks and experts cannot be truly effective without its disreputable, global counterpart.

Flat Tariffs Are Good Industrial Policy

In addition to being global, the tariff should be flat: that is, applicable to virtually all kinds of imported products, at the same rate of duty. This concept scandalizes the policy elite. They can recognize that the United States should care about manufacturing advanced products like semiconductors and EV batteries. But why drive up input costs for manufacturers who import components if we’re trying to grow the manufacturing sector? Why drive up the cost of imported T-shirts, which we shouldn’t care about making domestically, or bananas, which we can’t? Haven’t the Trump people heard about deadweight loss? Far better, as Noah Smith put it, to “slice out exactly the types of imports you don’t want, while leaving the less strategically important stuff intact.”

Return again to Sullivan’s definition of winning: “rebuilding our manufacturing and our resilience.” After so much lost ground, it is a fantasy to imagine that we can parachute in at the top of the manufacturing value chain and capture durable market share only in the highest-value, end-use products. This “scalpel” approach would not achieve the United States’ strategic aims and does not reflect the reality of how manufacturing sectors grow and innovate.

A policy to tariff only strategic end-use products, while continuing to import cheap inputs to manufacture those products, would consign the United States to a commoditized final-assembly role, shutting us out of most of the productivity and innovation gains that advanced manufacturing can deliver. That’s not “resilience,” and it certainly isn’t a manufacturing renaissance. A recent study used firm-level data to find that firms that offshored intermediate inputs to China and Mexico between 2002 and 2011 ended up spending 15 percent less on R&D.²⁶ Pointing to a long line of engineering case studies, the authors observed that “offshoring to low-wage countries can take so much cost out of older-generation technologies that it becomes unattractive to develop next-generation technologies.” That’s the track we were on.

A supporter of tariffs on strategic end-use products should support a broad tariff that covers the supply chain for that product. A 10 percent tax on imported EVs, serving as an inducement to purchase domestically made EVs instead of imported EVs, does not add much to resilience or manufacturing growth if the domestic process is simply assembling imported components. The same 10 percent inducement to domestic supply should apply to the battery, the motor, the electrical components, the chemicals, and the steel and aluminum that make up the vehicle. These steps are where the value is added, not at the final stage.

The truth is that there is no easily severable strategic interest between the final car and the components necessary to make the car. A shortage of steel bolts will grind an EV plant to a halt just as surely as a shortage of lithium-ion batteries. We don’t need to implement autarky in steel bolt production, but on the margin, we should express a preference that we increase our domestic output of even basic inputs if we are serious about resilience.

A flat global tariff at the low end, say, 10 percent, will not reshore production of basic inputs by itself. But it will, by definition, drive reshoring of products where American producers’ margin of price competitiveness is within 10 percent, which are precisely the producers most likely to turn support into sustainable gains. In combination with other industrial policy measures, including stacking with strategic tariffs through Section 301 and Section 232, phase-in periods to encourage supplier shifts, and preferential financing and amortization schedules for machine tools, a relatively small flat tariff can be a durable platform for reshoring up and down the supply chain – including, perhaps, a share of the steel bolt market.²⁷

The Europeans are apparently alive to the danger of becoming a mere assembler of imported inputs. As China has located more factories in Europe, the EU has attempted to ensure that inputs for such factories are locally sourced, with EU industry commissioner Stéphane Séjourné warning that “Europe must be a complete industrial base and not a mere assembly platform.”²⁸ Unfortunately for Europe, China is opening those factories “to foster foreign dependence on China’s advanced manufacturing, which Beijing sees as a source of leverage in an era of geopolitical shocks.”²⁹ China knows better than anyone the importance of local content requirements for inputs into domestic production, having perfected the strategy. They are unlikely to heed the commissioner’s warning.

Much of the concern with tariffing inputs focuses on losing competitiveness with our trading partners on global markets. As former Treasury Secretary Larry Summers argues:

Lower priced inputs mean more competitive exports. Take, for example, the steel tariffs that my friends here favor. A hundred times as many people work in industries that use steel as work in the steel industry. When we protect steel, we reduce our competitiveness.³⁰

But the opportunity to restore American manufacturing competitiveness is not in global markets. As Jake Sullivan pointed out, those are being flooded by an unprecedented glut of subsidized Chinese supply. No market economy is beating BYD on price, because BYD’s prices do not reflect market economics.

Rather, the opportunity is in the massive U.S. domestic market, of which the $1 trillion trade deficit reflects the annual amount of demand currently going to imports that could be gained for domestic supply.³¹ The relevant question is not “would a flat tariff make American exports less competitive?” Unless we are planning to replace the American way of life with China-style industrial subsidies and overproduction, the export-based growth opportunity has been lost. The key question instead is “would a flat tariff make domestic finished goods more attractive relative to imported finished goods in the American market?” The answer is yes.

Consider how a flat tariff would impact TSMC’s new fabs in Arizona—which, as the Wall Street Journal reported, have captured investment originally destined for fabs in other countries in order to avoid U.S. tariffs on imported chips.³² The Arizona fab will need to pay 10 percent on its imported lithography machines and silicon, while a Taiwan fab will not. But in tariff terms, the Arizona chips will still be preferable to the Taiwan chips in the American market. Oren Cass explains why:

The tariff will tilt an American buyer’s choice toward the Arizona-made chip, because a chip from Taiwan will have the 10 percent tariff added to its entire value, whereas a chip from Arizona will have the tariff embedded only in the foreign elements of its supply chain. The tariff also, therefore, creates an incentive for TSMC Arizona to find domestic suppliers, and for foreign suppliers to set up shop on U.S. soil.³³

The 10 percent tariff on inputs will necessarily amount to less than 10 percent of the final price because chips do not sell at marginal cost. Further, the cost of the tariff on capital equipment will be amortized by the Arizona fab, an option not available for the tariff on the Taiwan fab’s chips. Assuming all else is equal, a flat tariff will mean that the Arizona chips will be cheaper than the Taiwan chips.

The availability of the domestic-led growth strategy is a unique privilege. We have it because we own the largest consumer market in the world. Other countries hoping to build their manufacturing sectors must seek growth from the global market, in competition with Chinese exports, because their domestic markets simply can’t provide enough demand to build scale. Germany, for example, doesn’t have enough domestic demand by itself to grow a world-class auto industry, so it depends on exports. Over 75 percent of German auto production is exported, compared to 15–20 percent for the American industry. But firms that depend on export markets lie directly in the path of the Chinese overproduction monster, and German manufacturers are now hemorrhaging market share.³⁴ As owners of the largest market in the world, we are far more protected from this threat. So long as we ensure, through active antitrust policy, that our domestic market remains highly competitive, we can provide enough demand to grow our manufacturing sector without having to undertake a martyrdom operation to compete with Chinese exports on cost.

Flat Tariffs Are Pro-Innovation

Nonetheless, it is highly likely that China has built enough of a lead in the current generation of advanced manufactured products––such as EVs, lithium-ion batteries, and solar equipment––that it will be impossible for the United States to come from behind and surpass China as the market leader. Unfortunately, these are the precise industries that Sullivan identified as most important to protect. While the United States can and should defend industries it still possesses, the heart of our industrial policy should be to build our productive capacities to capture the next generation of manufacturing, using advanced materials, new processes, and AI-enabled production systems.

This kind of innovation will not come from islands of “strategic” production based on what planners believe is important. Rather, it will emerge unplanned, out of an “industrial commons” of producers sharing knowledge across wide domains of expertise from every point in the value chain, both upstream and downstream.³⁵ That includes expertise in what appear to be non-strategic manufacturing processes and non-strategic products.

In February 2026, the hedge fund Palliser Capital published a presentation outlining its thesis that the Japanese toilet manufacturer Toto occupies a critical node in the semiconductor supply chain.³⁶ Specifically, Palliser argued that Toto’s “Advanced Ceramics” unit, which had come to represent the majority of the firm’s profits, “is a critical enabler of next‑generation semiconductor manufacturing, supplying highly specialized electrostatic chucks used in cryogenic dielectric etching tools for advanced 3D NAND memory production.” The stock is up 55 percent year-over-year.

Toto is far from the only legacy manufacturer that started with an unremarkable business and later found itself commercializing a critical innovation. In fact, the pattern is extremely common. Over the past century, numerous advanced industries first emerged out of processes for making products that no expert would have described as strategic.

When Western countries offshored consumer electronics assembly to Asia, expertise in battery technology moved with it and the lithium-ion battery industry took off in Japan and South Korea even though the technology was first invented in the United States. The same relationship existed between the camera industry and semiconductor lithography. The American auto industry grew out of the machine shops and supplier base propagated by the bicycle industry, which had adopted key processes such as electric resistance welding and precision tooling while popularizing the use of tubular steel frames, ball bearings, differential gearing, and other innovations.³⁷ Bill Knudsen, the legendary General Motors executive and central figure in Arthur Herman’s now canonical book Freedom’s Forge, began at a bicycle company called Keim and was first exposed to automobiles when Keim won contracts to make brake drums and steel axle housings for Oldsmobile and Ford.³⁸ (At the time Knudsen won those contracts, the tariff on bicycles, automobiles, and auto parts was 45 percent. But it’s probably nothing.)

Illustrating the point in present context, manufacturing entrepreneur Greg Koenig posted on X to explain why “a toaster factory and a drone factory are basically the same thing”:

A factory is a collection of processes that manipulate raw material through a series of steps using capital equipment. Modern factories are highly flexible. A drone and a toaster share a huge amount of commonality – injection molding, stamping, machining, PCB assembly, wire forming, precision assembly, packaging, QC, etc. etc. You learn to run the wire forming process for the heater on a toaster? You learn 90% of what you need to know to wind the stator for a drone. You learn how the CNC mill works to clean up the die castings for the toaster? You know 95% of what you need to know to mill the drone core from bar stock. The injection molding machine that makes the slide lever for the toaster? Swap out the mold in it and now it makes the blades for the drone. The list goes on and on.³⁹

The lesson is that America can’t be reindustrialized by selecting only industries that appear “strategic” for support. By the time an industry is clearly identifiable as strategic, it likely is too late for it to regain a lost position as the global leader. Rather, the task is to rebuild an industrial commons that seeds innovation by developing manufacturing expertise broadly. But America won’t be able to grow this industrial commons if it freely trades with China outside of a few “strategic” areas. China makes all the “strategic” drones because it was making all the non-strategic toasters. And because it makes all the drones, it is in pole position to make whatever comes next. It turns out that toasters are more strategic than anyone in the Biden administration thought. Maybe we should make some.

A flat tariff provides broad protection for American manufacturing, protection that doesn’t require a planner to pre-select any product as strategic and instead boosts industrial production as an entire practice. Supporters of strategic-only tariffs often frame their position as more compatible with market principles. But compared to the flat tariff, their approach shares more DNA with socialist planning. The flat tariff does not pick winners and losers. It does not attempt to figure out which products should receive support and which should not. It merely expresses a generalized preference for domestic production over imports. After that, it lets the market decide where America’s core competencies lie.

To the extent it intervenes in prices, it does so to correct a global distortion created by China’s state-led strategy to flood the world with subsidized exports. The China-origin distortion it corrects affects prices on virtually all products that China exports, from drones to toasters, not merely “strategic” products. In this context, a flat tariff is not itself a non-market distortion; it makes prices more like what would prevail under normal market conditions.

In the realm of vibes, critics (and Chinese bots) deride the flat tariff with memes of miserable Americans working in sweatshops sewing T-shirts.⁴⁰ Surely it would be more efficient to leave labor-intensive manufacturing to the Vietnamese? Not necessarily. The availability of cheap labor abroad holds back innovation in America that would otherwise increase productivity and reduce labor intensity.⁴¹ It’s easier for firms to underpay armies of laborers than invest in productivity-enhancing technology. When an American stapler company moved from Long Island City to Nogales, Mexico following nafta, here’s what a journalist saw:

I found the stapler-assembly and testing sections, rows of four men and women facing four others across long tables, putting the staplers together by hand . . . . Labor was so cheap in Nogales that stapler assembly had been deautomated. There was no sign of the Bodine assembly machines I’d seen in Long Island City.⁴²

Cheap imports de-skill American firms just as surely as illegal immigrants in the American labor market. The only difference is that the products are imported instead of the people.

Taxing imports thus becomes an inducement to innovation. The United States still does $10 billion of cut-and-sew apparel manufacturing, and American companies like Framework Automation and Anatar are busy developing automated production processes specific to apparel. It is entirely possible that a highly automated American apparel industry powered by advanced technology could compete with Vietnamese sweatshops. Across the board, as automation and AI reshape production, the advantage conferred by low labor costs will shrink, and making T-shirts may enter America’s competence while looking nothing like the process to make a T-shirt in Vietnam. Meanwhile, the process knowledge from innovating and operating in an automated apparel industry spills out into the rest of the industrial commons, giving more people the skills and experience to come up with the next idea in advanced manufacturing.

The labor-intensive Vietnamese T-shirt will likely remain a good bargain, and Americans are free to import them even with a flat tariff. But on the margin, fewer will, and that helps speed the future of American innovation into being.

Deadweight Loss?

What about bananas? There are certain tropical products that America can never make, and conceivably a flat tariff could exempt such products. But it probably shouldn’t. Flat tariffs represent a preference for consumption that employs Americans and benefits American firms rather than foreign ones. A 10 percent flat tariff might make bananas cost a nickel more. Some people will pay the nickel, but some will substitute away from bananas, which aren’t grown on American farms, to apples, which are. Is our social preference for one kind of consumption versus the other worth the nickel? Probably.

Economists call the substitution of the apple for the banana “deadweight loss,” since it wasn’t the consumer’s first choice. The invocation of deadweight loss is always meant to be a conversation-ender. But all interventions in a voluntary market technically create deadweight loss; whether this is a good idea depends on the social effect of the intervention, as Democrats well know. And if the nickel was enough to make the consumer choose the apple, the banana must not have been adding very much welfare in the first place.

Meanwhile, the nickel isn’t thrown away. It enters the Treasury as revenue, where it can pay down the debt, fund spending, or pay for tax cuts that leave firms and consumers no worse off on net than pre-tariff. Before the Supreme Court overturned Trump’s ieepa tariffs, they had raised $166 billion in revenue in less than a year and were projected to raise nearly a trillion over ten years.⁴³ They offset roughly half of the fiscal cost of the tax cuts in the One Big Beautiful Bill, and when they were overturned, U.S. treasury yields increased, signaling bond market displeasure.⁴⁴ Trump’s discovery of a path to raising significant revenue without either crashing the economy or fighting through Congressional gridlock was a generational achievement. A flat tariff, far from simple deadweight loss, may be our only realistic path to fiscal health. Signaling that the tariff may not really be flat, first by exempting sympathetic products like bananas, may open a Pandora’s box of additional exemptions that would cripple the tariff’s effectiveness.

A flat tariff’s effectiveness can be maximized in other ways. It should be phased in over time, to give firms time to adjust and scale up domestic alternatives. It should be combined with industrial policy measures such as preferred financing for industrial equipment, so that firms don’t have to pay more for the machines that enable them to make machines. And most critical of all, it should be passed in a manner that creates certainty, ideally through Congressional legislation. Firms won’t spend money changing their behavior if they think the policy itself will change overnight.

The year of Trump’s tariffs saw improvement in manufacturing performance, including growing industrial output, demand for capital equipment, and purchasing manager survey ratings, all with slowing inflation and accelerating growth.⁴⁵ Performance was especially strong in the first quarter of 2026, with manufacturing labor productivity increasing by 3.6 percent and output by 3.3 percent alongside a 6.1 percent increase in hourly compensation.⁴⁶ Even manufacturing employment, which continued to decline, declined at a slower rate after Liberation Day than before. Although investment has held steady, it hasn’t yet surged. To the extent that the tariffs have so far failed to create transformational volumes of new investment in American manufacturing, lack of certainty is the likely culprit.

The expert consensus is right that tariffs should be strategic. But it is wrong about what strategy requires. A tariff regime designed around planners identifying products for protection will always arrive late, play whack-a-mole with circumvention, ignore key parts of the value chain, and target the current generation of technology instead of the next one.

The more strategic move is simpler, dumber even: tariff everything. Create a durable preference for production in the United States, across the whole industrial base, and let entrepreneurs, engineers, and manufacturers discover where the next breakthroughs will come. The scalpel has its place, but the blunt force object is the tool for this job. For those who really want to “rebuild our manufacturing and our resilience,” it’s time to start hammering.

This article is an American Affairs online exclusive, published June 2, 2026.

---

Notes