Not By Tariff Alone

Reorienting America’s economy requires a massive increase in industrial investment.

On March 3, President Donald Trump stood next to C.C. Wei, the CEO of the Taiwan Semiconductor Manufacturing Company (TSMC), and announced that the world’s largest chipmaker would be investing $100 billion to build five new factories in Arizona. This new capital would be in addition to TSMC’s existing $65 billion investment commitments. Secretary of Commerce Howard Lutnick explained:

Under the Biden administration, TSMC received a $6 billion grant, and that encouraged them to build $65 billion. So, America gave TSMC 10 percent of the money to build here. And now you’re seeing the power of Donald Trump’s presidency, because TSMC, the greatest manufacturer of chips in the world, is coming to America with $100 billion investment. And, of course, that is backed by the fact that they can come here because they can avoid paying tariffs.

On April 8, Trump got up in front of a Republican National Congressional Committee event and put it more bluntly: “TSMC, I gave them no money…all I did was say, if you don’t build your plant here, you’re going to pay a big tax…maybe 100 percent.” The implication was clear: tariffs can reshore industry, no need for extraneous and wasteful policy measures. 

The reality is more complicated. Tariffs can and do play an essential role in successful national industrial strategies, to the chagrin of both free trade absolutists and businesses unaccustomed to healthy constraints. But tariffs cannot do it by themselves. Many countries have won in key industries with the help of tariffs. None have accomplished the goal with tariffs alone.

There are at least two reasons for this. Tariffs tilt the scales in favor of domestic production, but they don’t address the full range of reasons a desired industrial investment might remain unattractive to investors, even despite tariff pressure. And relatedly, the domestic production spurred by tariffs will tend in the short-run to yield higher-cost products, which will be less competitive globally and less attractive to domestic customers—if this were not the case, the tariffs would not play a constructive role in the first place. Competition and experience can drive those costs down, but in critical industries where other countries are providing generous subsidies, and in concentrated ones where competition may impose less pressure, tariffs as the last word will not fully deliver. The swift investment that tariffs can prompt must build atop and then be buttressed by other policy measures. Semiconductors are a perfect case in point. 

Tariffs: Necessary but Not Sufficient

The administration’s tariff argument starts with a basic and entirely true premise: it matters whether things get made in America, and it especially matters whether particularly important things like semiconductors get made in America. The truth of this premise is a defining feature of the new economic thinking that has overtaken American politics since 2016, and especially post-pandemic. Trump is right. The United States has paid the price of deindustrialized weakness, is tired of it, and now wants to revitalize its capacity to make things. 

This dynamic is, of course, why the announcement of a $100 billion foreign investment commitment in new fabs on American soil is so politically valuable to the White House, just as TSMC’s previous commitments were to the prior administration. In this new area, politicians want to be seen as having successfully put the thumb on the scale for industrial production that the market wasn’t already pursuing and that wouldn’t have happened without their intervention. For Trump, that means pointing to the success of his tariff policy at prompting investment commitments.

But once policymakers acknowledge that they like tariffs because they care more about whether something gets made in America than about market efficiency—that is, once they admit that reindustrialization requires policy intervention where current profit incentives fail—they must also admit that other public investments make sense too. The logic of President Trump’s point is that producing chips in the United States has value beyond what private profit under current market conditions would imply—which, in the case of chips, has not been enough to support a domestic industry in recent years—and therefore his intervention was constructive.

But while tariffs will tilt the balance in favor of domestic supply chains, they will still only result in the  investment and production TSMC is able to turn a profit on, under less-than-ideal conditions. These will be fewer and less cutting-edge chips being produced in the United States, made at higher cost by relatively less productive workers. If the goal is instead to once again be a world leader in making cutting-edge chips that can supply a competitive domestic electronics industry—and perhaps even export these chips around the world in the long run—tariffs will help us on the path but not get us to the destination. 

TSMC has already discovered that investing here comes with all the challenges endemic to making chips profitably in America. This is why market analysts have expressed skepticism that “TSMC’s $100 billion pledge will fully materialize,” according to Fortune. TSMC and the Taiwanese government have claimed the company’s $100 billion move was due to consumer demand and that pressure from the U.S. played no part. But the financial services firm Morningstar saw it differently: “We see the new fabs as a strong signal that TSMC’s investments are not commercially driven. The company has previously said construction and manufacturing are both more expensive in the U.S.” That is, tariffs can pressure TSMC to announce investment in the United States. But they do not remedy all the underlying factors that make it more difficult and expensive to produce here.

Contra the free trade absolutists, tariff pressure can achieve results by increasing the attractiveness of domestic production relative to imports. The mere threat of tariffs from Reagan is why the 1980s saw the explosion of a domestic Japanese auto sector—employing Americans and building cars with American parts—rather than the destruction of American automakers in the face of Japanese auto imports.

But the reason Reagan’s tariff threat worked in this context is because America still had a reasonably robust industrial base in the 1980s, to which the Japanese manufacturers could connect, and atop which they could quickly build out a supplier ecosystem. The same is true now. TSMC’s $100 billion commitment was prompted by President Trump’s tariff posture, but it was only plausible because the bipartisan Chips and Science Act induced and accelerated the first $65 billion of TSMC investment, which in turn has begun to spur wider workforce development and the reshoring of broader supply chains, upon which the next $100 billion can now build. And, if nothing else, TSMC almost certainly expects to benefit from the same CHIPS Act 25% investment tax credit that helped make its first $65 billion investment possible. Ensuring this commitment materializes will require ongoing dedication to making the full ecosystem competitive. Failure to do so will almost certainly result in higher prices for consumers in the long run, compared to what they had access to when chips were sourced from regions that reduced costs with subsidized construction, public investments in workforce development, saner permitting policies, and so on Could tariffs alone rebuild this ecosystem? Perhaps, theoretically. But it would take far longer and be far more expensive, which is not only a worse return on investment but much less politically feasible. This is true for chips; it is also true for critical industries and sectors generally. 

Timing is Everything

The administration needs to take these related questions of timelines and political dynamics very seriously, not just in the context of semiconductors, but across its plan for tariffs. The new China tariffs, which should be phased in rather than imposed overnight as announced, have been the most obvious case in point. Businesses can only relocate production and reorder supply chains so fast. There is no upside to inflicting immediate pain on businesses that could not feasibly avoid it, no matter how swiftly they try to adjust. But there are plenty of downsides. Imposing tariffs too steeply and quickly will negatively affect Americans as consumers, without the corresponding benefits that will come from rebuilding American focus on production, which will take longer to be fully felt by producers. It is for good reason that the administration is already contemplating a course correction toward phased-in China tariffs that discriminate between critical and non-critical goods, as congressional leaders and American Compass have recommended.

The result of failure to heed this reality will be economic pain leading to political consequences. Oren Cass put it well in the New York Times: 

There is a limit to the costs the American people and American allies will bear, especially without constant reminders of a coherent long-term vision. The president should want to minimize the short- and medium-term harm to businesses and supply chains that must survive disruption if they are to thrive in the long run…

This is why, in addition to implementing tariffs more wisely, the administration also:

needs to get serious about other policies necessary to support reindustrialization. If the United States is going to reduce its trade deficit quickly without painful cuts to domestic consumption, it’s going to have to increase production capacity just as quickly, either to expand exports to other markets or to substitute for imports at home. This requires industrial policy….

If the promised reindustrialization takes too long to materialize even as avoidable and needless short-term pain is not mitigated, domestic political backlash may kill the effort before it has a chance to succeed. Working-class voters are giving Trump more leeway on tariffs than his opponents care to admit. But they will not wait forever. Potential Republican challengers are already waiting in the wings to pounce should tariffs prove too unpopular. Democrats are contemplating an anti-tariff electoral strategy the potential effectiveness of which should not be discounted—especially if the Trump administration does not course correct. 

The other inconvenient truth is that while industrial timing affects the politics of tariffs, the politics of tariffs can also affect industrial timing. America’s trading partners have their own domestic dynamics that can extend the timelines of American reindustrialization efforts longer than they need to be. This underscores the urgency of a full-spectrum policy effort to speed those timelines up. 

The most pressing example of short-term risk is China doing exactly what attentive leaders have long warned it could do: use its supply chain dominance to choke off the flow of critical minerals, used in everything from high tech weapons systems to MRI machines, into the United States. The intent is to strike at the heart of the Trump administration’s industrial goals, including the revitalization of America’s defense industrial base, which will be seriously impeded without these key inputs, as they are currently and overwhelmingly sourced from or flowing through China. In the meantime, America’s economic security will remain vulnerable. 

The options are to capitulate to China or to launch as aggressive an industrial policy campaign as possible to establish alternative supply chains. The administration has taken some needed steps on critical minerals. It must do more, like invoke Title III of the Defense Production Act to offer price guarantees to the domestic market. 

But the problem is not limited to our enemies. Lost in the celebration around TMSC’s March announcement was the threat that Taiwan’s domestic politics could pose to the promised investment’s future. Major foreign investments of this kind require the Taiwanese government’s approval, and it is not at all clear that future approvals will be easy to secure. In Taiwan, Wei’s announcement was met with extreme criticism from the opposition party. Former President Ma Ying-jeou castigated the ruling Democratic Progressive Party (DPP) for “selling TSMC” as a “protection fee” paid to Trump, and thus precipitating “a major national security crisis.” The current president had to stage a joint event with CC Wei to reassure the public that TSMC would expand at home as well. 

It is not hard to see the results of a future Taiwanese election imperiling the investment Trump is currently boasting about. For Taiwan and others ruffled by the administration’s tariff posture, Trump’s bet on the centrality of the American market is probably correct: economic necessity will eventually compel nations to comply, or at least to genuflect. But the domestic politics of America’s trading partners can nevertheless drag out full cooperation and lead these countries to offer it on less advantageous terms. When playing the tariff card runs into this barrier, it will be helpful to have other cards in our hand. 

The Full Toolkit

Making the whole American industrial ecosystem more competitive, and thereby speeding up results, mitigating pain, and better insulating the effort from political vagaries both foreign and domestic, will require the U.S. government to embrace the full suite of policy tools available. The president should embrace a clearer, whole-of-government reindustrialization strategy, which would better serve both his policy and political purposes. This would need to include measures the administration likes, those it might like but hasn’t yet pursued, and those it doesn’t like but should.

The administration has been strong out of the gate regarding regulatory hurdles to industrial investment. The newly announced Investment Accelerator Office, designed to ease major foreign investments in the United States by helping investors navigate the regulatory maze (and, where possible, actively reducing regulatory burdens), is a very positive move. Broad environmental permitting reform from Congress must follow; the President or his Cabinet-level designate should demand it loudly. The Biden administration failed to solve this problem; the Trump administration could. 

Massive and aggressive workforce development investment is also required. The lack of adequate skilled workers is routinely cited as a serious impediment by both American and foreign companies interested in expanding their industrial efforts, including flagship companies like TSMC. The administration seems keen to either take over or dismantle elite universities on ideological grounds. Its energy would be better spent articulating a constructive vision for American workforce training, funded by redirecting taxpayer dollars away from a college-for-all system that has indeed failed and toward dignified non-college pathways. Smart ideas are available that would make good on the Trump-Vance campaign’s promise to prioritize working Americans. The American Workforce Act, introduced in 2024 by Senator Tom Cotton, then-Senator JD Vance, and Congressman Max Miller, would establish a per-trainee grant for high school graduates to offset the cost of on-the-job training. Employers know what skills are needed; let them work with community colleges, unions, trade groups, and other organizations to develop training better targeted to the economy’s real-time needs. There is no reason to entrust the vast majority of taxpayer dollars to higher ed when it is failing to produce the workforce America requires. We can support the employers who will do the hiring, and the workers they will hire, instead. 

A comprehensive industrial plan would also address the structural deficits in the American financial sector. The United States not only has an offshoring problem, it also has a domestic financialization problem—an obsession with making money through financial engineering and speculation rather than investment, and with distributing profits rather than reinvesting them. Industrialization requires investors willing to deploy their capital for that purpose. But the American corporate sector, as American Compass has noted, is instead characterized by:

[T]he increasing tendency of firms to maximize short-term shareholder returns and disgorge profits back to financial markets rather than pursuing investment at all. In the 1970s, publicly traded companies returned roughly 50% of their profits to shareholders each year. From 2008–17 that rate averaged 100%, and in four of those years exceeded 100%.

To the extent that American companies are currently investing in overseas production, tariffs may push that capital toward production at home instead. But it does not follow that the financial sector will abandon its broad bias toward capital-light, shorter-term investments, or its even greater bias toward disgorging profits to shareholders rather than pursuing actual investment in the first place. Addressing this should appeal to the administration, which has a demonstrated appetite to discipline sectors it believes to be acting contrary to the American interest, including Wall Street. It could start by banning stock buybacks—corporate America’s preferred method of dumping cash rather than productively investing it. 

Looking forward, the administration should establish a domestic development bank or industrial finance corporation, of the kind JD Vance was working on with Senator Chris Coons before he left the Senate. A years-long legislative battle to support every critical industry is not a viable way to execute a sound national industrial strategy. Neither is it an optimal use of political capital. A development bank that can leverage public capital to deliver private investment across the full range of critical industries, on an ongoing and flexible basis, can fill the financing gaps that the current financial sector does not. Imagine if, rather than each industry having to beg Congress for years, American industry broadly could turn to a financing source that could quickly make investments to stand up new chip fabs or critical mineral processing facilities. This kind of standing capacity is an essential tool and is exactly the kind of thing President Trump’s proposed sovereign wealth fund could be used to accomplish. 

Finally, the administration should rethink its posture toward the CHIPS Act, which despite deficits in its implementation has undeniably worked. The CHIPS Act’s construction tax credits, loans, and direct grants have spiked manufacturing construction through the roof and set the United States on the path to reestablishing a domestic chip industry. U.S fab capacity is expected to triple, and the global share of advanced logic capacity is expected to rise from 0% to 28%, by 2032, largely as a result of this commitment. Investment in manufacturing construction almost doubled between 2019 and the middle of 2023, due largely to the CHIPS Act; construction in electronics manufacturing increased around ten-fold in the same period. It is encouraging that President Trump’s criticisms of the measure have been met with strong resistance from congressional Republicans, who are committed to its effective implementation. 

An industrial strategy overly dependent on subsidies has downsides, certainly. Overuse of subsidies can lead to gluts and dumping, for example; tariffs do not. But abandoning their proper use—swift, short-term correction for the commercial disincentives to invest in a critical industry, before other tools have had a chance to fix the relevant structural problems—is as foolish as anti-tariff hysteria. Subsidies are, of course, exactly why an industry like semiconductors left the United States to begin with. Weak trade protections allowed it to leave, but foreign subsidies were the draw and are the primary reason why producing chips overseas has been so much cheaper.  

Admitting all this hardly makes one soft on trade. Most successful national industrial strategies incorporate judicious use of tariffs, but tariffs are never the only element of that strategy. This is as true of our own history as of other more recent success stories, like the rapid post-war industrialization of Japan, South Korea, and Taiwan. American supporters of smart tariff policy are rightly fond of citing Alexander Hamilton and the other pro-tariff fathers of the American System that eventually birthed American industrial might. “Give us a protective tariff and we will have the greatest nation on earth,” Abraham Lincoln famously said. Lincoln also understood that massive federal infrastructure spending was required for his vision of American economic greatness to work. So did Hamilton. 

Ha-Joon Chang, a heterodox economist who studies how nations industrialize, is about as pro-tariff as they come. In his book Kicking Away the Ladder: Development Strategy in Historical Perspective, which argues that all rich nations tend to get rich behind high tariff barriers, he offers a caution from history. Evoking the scriptural message that man does not live by bread alone, he warns that reindustrialization happens “not by tariff alone”: 

Important as tariff protection may have been in the development of most NDCs [now-developed countries], it was – I repeat – by no means the only, nor even necessarily the most important, policy tool used by these countries in promoting infant industries…Tariffs were not, and are not, the only policy tool available to a state intent on developing new industries or upgrading old ones. 

The United States has long needed a smart, coherent industrial strategy that addresses the full range of impediments confronting successful reindustrialization. A full court press on tariffs does not mitigate the need for this. It makes it even more urgent.