The provocation was the “every president tries” part. If “trying” is defined as changing policy to consistently support employment growth in U.S. manufacturing, no president has tried in my lifetime to do this. Amazingly, doing nothing has indeed failed. Doing nothing was also the wrong choice.

The loss of manufacturing jobs

First, some data to define the problem. Furman focuses on the share of total employment that is in manufacturing. He notes that many structural non-policy forces (like technology and what people demand as countries get richer) put steady downward pressure on this in any growing country. There’s a lot of truth in that.

But the U.S. got much richer between 1965 and 2000—in fact it got richer at a far faster pace than it has since, so both technology and the different demands of a richer society should have been operating a lot less intensely since then. And yet the level of U.S. manufacturing employment was steady during that period, fluctuating roughly between 17.0 and 19.5 million depending on the state of the business cycle (see Figure A). After 35 years of stability, manufacturing jobs then cratered: 3 million manufacturing jobs were lost after the recession of 2001, and the 2003–2007 recovery saw essentially no gain at all in manufacturing jobs—the first manufacturing jobless recovery we’ve ever experienced. Then another 3 million jobs were lost during the Great Recession of 2008–09.

After falling from over 17 million to just over 11 million between 2000 and 2010, the sector has seen only very slow growth since. The new high point of manufacturing employment in the recent past was 12.9 million workers in early 2023.

Figure A

Manufacturing historically lost a disproportionate share of jobs during recessions, but what kept it from gaining jobs back quickly in the early 2000s and 2010s recoveries the way it usually had? One huge influence was the emergence of a large trade deficit in manufactured goods. In those decades, the deficit peaked at 4.4% of GDP in 2005 (see Figure B). After being forced into improvement by the Great Recession and the collapse of American spending on all goods and services (including imports), it has steadily moved back toward this peak and surpassed it in recent years.

Figure B

Policy measures can close the trade deficit and reshore manufacturing jobs

Tolerating this rise of the U.S. trade deficit was a policy choice. The deficit’s rise was driven by a dollar whose value is too high to allow balanced trade. A high dollar makes our exports expensive to foreign consumers and makes foreign imports cheap for U.S. residents. Hence, it leads directly to chronic trade deficits (see Figure C). Any serious effort at boosting manufacturing employment would require using policy levers to reduce the value of the U.S. dollar.

Figure C

What are these currency policy levers? First, policy would need to prevent other countries’ governments from actively managing the value of their currency to give their exports a competitive advantage against U.S.-produced goods. There are many ways to do this. Currency management is done through other countries’ governments (or their proxies) buying U.S. dollar-denominated assets (like Treasury bonds or mortgage-backed securities) to bid up the demand for dollars. There’s no particular reason the U.S. couldn’t undertake countervailing currency intervention and buy other countries’ assets whenever they bought ours in an effort to manage their currency’s value. Or we could tax foreign purchases of U.S. assets.

Second, we could raise taxes domestically to close fiscal deficits. In coming years unless we run into a recession (which the Iran conflict makes more likely), there is likely to be sustained upward pressure on interest rates stemming from the big increases in fiscal deficits locked in by the Republican mega tax and spending bill. Higher interest rates in the U.S. will attract foreign investors to U.S. assets, which will bid up the value of the U.S. dollar further and harm manufacturing.

Third, we could hasten the inevitable deflation of the AI-driven stock market bubble, which has attracted foreign investors looking to make high returns. All else equal, there would be less upward pressure on the U.S. dollar if foreign investors were not rushing in to buy dollars to purchase U.S. stocks.

Fourth, we could accelerate the transition to cleaner energy. The U.S. has swung from being a large net importer to a net exporter of oil and natural gas. This has greatly increased foreign demand for U.S. dollars simply to buy our energy supplies, which pushes up the value of the dollar and hurts U.S. manufacturing.

Finally, we could reform our corporate tax code to stop its bias toward offshoring both paper profits and real production. The swing toward a large trade deficit in the pharmaceuticals sector, for example, can be linked directly to the first Trump administration’s changes in the corporate tax code.

In short, taking currency seriously would mean going against some very powerful economic interests—finance, tech, pharmaceuticals, and fossil fuels—in the name of helping U.S. manufacturing. But it would be a good trade to make. And to be clear, dollar weakness that is caused not by intentional policy decisions but is simply an outcome of erratic policy decisions will not provide any sustained benefits to U.S. manufacturing. U.S. manufacturing needs a competitive value of the dollar and a healthy and stable domestic economy. Engineering dollar decline by sabotaging the stability of the domestic economy does not help.

How many jobs could be reshored if currency policy somehow closed the U.S. manufacturing trade deficit? Very roughly it would be close to 3 million. This would not change the long-run trend in the manufacturing share of employment, but it would boost manufacturing-based communities around the country.

Indifference to manufacturing was bad for economic dynamism

The long-run gains to rebuilding communities of manufacturing process knowledge in the U.S. could be large. U.S. losses and China’s growing dominance in manufacturing are in large part a story of deconstructing communities of process knowledge in the U.S. and building them in China. These communities are geographic clusters where firms and workers specialize in particular manufacturing sub-sectors. The agglomeration of knowledge and skills leads to steady innovation which further locks in the competitive advantage of the cluster and raises productivity growth.

Currency policy destroyed these clusters in the U.S. and provided ample space for them to grow in China. The large and constant pressure of an overvalued dollar in the U.S. imposes a heavy drag on the prospects of new manufacturing firms setting up shop and becoming a center for clusters like these. The currency policy of China surely acted as the reverse of this, clearing huge competitive space for new entrants and for further growth in communities of process knowledge.

Currency management was not China’s only industrial policy measure, but it is the one that allowed an across-the-board competitive advantage in all manufacturing industries. And it is the only industrial policy in the U.S. that would reclaim some of the across-the-board manufacturing disadvantage we’ve allowed to be imposed on our domestic industry. Targeted protection and subsidies for particular sub-industries in manufacturing have been important in crafting the exact patterns of trade, but it is currency policy that largely explains the manufacturing-wide trade deficit that the U.S. runs with China and other countries that manage their currency.

How big is this problem of losing expertise and process knowledge in manufacturing for the overall economy? Another sign of the indifference towards manufacturing shown by successive U.S. policymakers is that we don’t even really know—and this indifference and the ignorance it generates has grown over the past year of the Trump administration. The manufacturing sector used to be a source of productivity dynamism in the U.S. economy, but recent data indicate that as we hemorrhaged millions of jobs we also saw declining productivity growth in the sector. This productivity decline might not be entirely genuine—it might be a problem with statistical measurement. It would be nice to invest in our data-gathering infrastructure to shed more light on this issue, but instead the parts of the Bureau of Labor Statistics who have the expertise to do this have been gutted by the Trump administration and longer-run cuts. Another angle of taking manufacturing seriously would be supporting the public structures that provide needed inputs to know what’s even happening in the sector.

Doing nothing was a mistake

U.S. presidents have made the implicit judgement over the past 50 years that it’s a good trade for Americans to have a smaller domestic manufacturing sector in return for cheap imports of manufactured goods, even if that means we’re running chronic large trade deficits. It’s not so obvious to me that’s a good trade, and there’s one last angle that makes it even less obvious.

The foreign inflow of capital that is the mirror image of the trade deficit in manufactured goods is essentially investors abroad bidding against Americans who are looking to buy stocks and bonds and other assets to build their wealth. Bidding up the price of these assets means long-run returns will be lower. In short, this current system of trade imbalances lowers the returns to holding wealth for U.S. residents. One could argue that this is mostly a problem for wealthy U.S. households, who own the lion’s share of assets.

But there is also the issue of why the valuation of U.S. assets has grown in recent decades even aside from increased foreign demand. A huge part of this growth is a zero-sum transfer of income from labor earnings to corporate profits: Recent estimates have this transfer accounting for almost half of the entire nominal growth in the value of U.S. corporate equities in the last 40 years.

Absent foreign demand for U.S. assets, some of this loss to wages would have been counterbalanced for at least some subset of U.S. households by higher rates of return to their savings. To be clear, this zero-sum transfer from wages to wealth still would have been a negative development for the vast majority within the U.S. economy. But this transfer combined with the fact that most of the gains accrue to investors outside of the U.S. because of imbalances in trade and investment flows make it even more damaging. Essentially, U.S. households as workers feel all the pain of a campaign of wage suppression, but U.S. households as investors do not claim all of the benefits of this wage suppression.

Presidents have not tried to reverse manufacturing job loss

In the end, no president in my lifetime has made a serious and consistent effort to do what is necessary to make the U.S. dollar stay at values commensurate with balanced trade in manufacturing. Ronald Reagan famously negotiated the Plaza Accord, which pressured Germany and Japan (our two biggest trade-deficit partners at the time) to reflate their own economies and to stop currency intervention. But at the same time, Reagan ramped up military spending and made large tax cuts that put huge upward pressure on interest rates and led to huge trade deficits in the early 1980s. Bill Clinton oversaw smaller fiscal deficits but actively encouraged a “strong dollar policy” which saw the dollar hit some of its highest levels on record. This strong dollar policy and support for a punitive rescue package for countries slammed by the Asian financial crisis of the late 1990s led to another large increase in U.S. trade deficits. The Clinton administration’s support for permanent normalized trade relations (PNTR) with China and for China’s entry into the World Trade Organization (WTO) made it harder for subsequent administrations to apply pressure to China to abandon its significant currency management in the 2000s.

George W. Bush refused to address the Chinese currency management and undertook large tax cuts and increased military spending again, pushing up interest rates and leading to another round of large trade deficits. Barack Obama similarly failed to address currency management, even leaving it out of the Trans-Pacific Partnership (TPP) agreement he pushed hard in his final years in office. Donald Trump passed corporate tax changes that actively incentivized offshoring in his first term in office. His major trade policy change in the second term has been chaotic and fluctuating—though generally high and broad—tariffs across manufacturing. Manufacturing employment in 2025 averaged 157,000 lower than in 2024 even as the administration trumpeted these large tariff increases. That constitutes the worst non-recessionary year for manufacturing since 2004.

Furman is right that we have seen consistent presidential failure to support employment in manufacturing. And he’s right that most of these presidents made some rhetorical commitment to manufacturing that makes this failure jarring. But nothing serious was ever really tried, and that was a costly mistake.

It’s worth emphasizing that the upcoming Census data do not reflect any economic developments in 2025. Some policymakers will attempt to claim any good news from the data as validation of the current U.S. policy path, but this would be completely misleading given the radical policy shifts in 2025 under the Trump administration. In this piece, we argue:

• Data for 2024 will likely reflect continued labor market strength. Inflation decelerated rapidly in 2024, which should boost last year’s income growth. 
• Even the likely strong 2024 income and poverty data will still show an economy that has left many workers, families, and children in an economically precarious position. Racial disparities in income, for example, leave people of color much more vulnerable to economic insecurity and poverty.
• Trump administration policies—including chaotic and historically high tariffs, mass deportations, and attacks on the federal workforce—have already led to a softening labor market and more inflationary pressures in the economy. Given this, income and poverty measures are likely to worsen when these data are released next year for 2025.  
• In 2026 and beyond, cuts to food assistance and Medicaid that were part of the Republican-passed spending bill will increase food insecurity and the number of people without health insurance, particularly for families of color.
• The Census data are incredibly valuable and provide transparent and non-politicized data that allow Americans to make informed decisions about what policies are delivering economic security for working people. The Trump administration has begun attempting to politicize and erode trust in federal statistical agencies and to manipulate the reporting of anything that seems like bad news for the economy. This is deeply undemocratic.

The labor market mostly held strong in 2024 

Between 2021 and 2023, the labor market rebounded dramatically from the pandemic recession as large-scale policy interventions—like expanded unemployment insurance—helped families stay afloat and drove a recovery several times faster than the Great Recession. In 2024, the labor market remained relatively strong, growing by 2 million jobs over the year. The unemployment rate rose slightly but maintained a 4.0% average over the year. 

The prime-age employment-to-population ratio—the share of workers between the ages of 25 and 54 with a job—held steady at a high level of 80.7% in 2024. Prime-age Hispanic workers saw their employment rise, as prime-age Hispanic men increased their employment rates by 0.7 percentage points. Employment also rose slightly for prime-age Black and white women by 0.2 and 0.3 percentage points, respectively. At the same time, Black and white men experienced mild declines in their employment rates. 

Real (inflation-adjusted) wages continued to increase in 2024. Figure A shows that inflation fell sharply from 3.9% to 2.6% over the course of the year. At the same time, the strong labor market allowed workers to maintain a solid pace of nominal wage growth: nominal hourly wages and weekly earnings growth decelerated by much smaller amounts than price growth for goods and services. This combination of inflation falling faster than nominal wage growth is exactly the macroeconomic “soft landing” from the COVID-19 inflation shock that so many thought would be impossible to achieve. Together, this translated into a 1.4% increase in average real hourly wages over the year. Average real weekly earnings—perhaps a better signal for the annual income data out next week—rose by 0.9%. 

While these are meaningful averages, we also know that real wage growth was particularly strong for lower–wage workers and workers with lower levels of educational attainment. Along with steady employment, these advances bode well for improvements to income and poverty rates in next week’s report.

Figure A

Persistent economic disparities leave families of color disproportionately vulnerable

The upcoming Census release will continue to show persistent racial inequities that can only be corrected through years of sustained progress. In 2023, typical Black and Hispanic households were paid just 63 cents and 74 cents, respectively, for every dollar paid to the median non-Hispanic white household. These disparities are especially harmful to low-income families of color who live in a constant state of economic insecurity. In a new report, we find that Black and Hispanic families with children make up more than half (61.1%) of economically vulnerable families, defined as those with incomes below 200% of the federal poverty line. Even within the group of economically vulnerable families, Black and Hispanic workers are also more likely to have incomes below the poverty line (see Figure B below). Narrowing these racial disparities will demand stronger and more persistent income gains for these families in the years to come.  

Figure B

Policy changes are weakening the economy in 2025

There are several reasons to suspect that 2025 will be a worse year for incomes and poverty. For one, labor market indicators have weakened: unemployment inched up to 4.2% by July 2025, and job growth slowed to 85,000 per month compared with 168,000 in 2024. The last three months saw average job growth of just 35,000 per month. While layoffs have not yet surged, both employers and workers appear to be sitting tight in anticipation of a weaker economy going forward. Federal employment has fallen by 84,000 since January, which doesn’t include the many workers who will leave federal payrolls on September 30 at the end of the fiscal year. According to Trump official Scott Kupor, 2025 will end with 300,000 fewer federal workers.

The Trump administration’s damaging and chaotic tariff policy also threatens economic security, with nearly universal tariffs set at the highest level in a century or more. This is causing severe business uncertainty and already leading to higher prices for households because tariffs are taxes on both imported and domestically produced goods. Since lower-income families spend a higher share of their income on goods consumption, these tariffs will disproportionately harm their real incomes.

In addition, the administration’s mass deportation agenda will substantially harm the labor market. The damage will not just be felt by immigrant workers and their families—they will spill over and hurt U.S.-born workers as well. If the Trump administration successfully follows through on its goals of deporting 1 million people each year during their term, there will be 3.3 million fewer employed immigrants and 2.6 million fewer employed U.S.-born workers by 2029.

Health insurance coverage and access to food assistance will fall over the next several years

Health insurance, Supplemental Nutrition Assistance Program (SNAP) coverage, and Supplemental Poverty Measure (SPM) rates in 2024 will likely represent high-water marks over the next few years. That’s because the Republican spending bill passed in July cuts Medicaid spending by $793 billion and SNAP benefits by $186 billion over the next decade—all to pay for tax cuts for the richest Americans.

The share of the population without health insurance was 8.0% in 2023, or about 26.5 million people. This ranked near historic lows in the United States—driven by a strong labor market, enhanced Affordable Care Act (ACA) subsidies, and pandemic-era coverage protections (particularly in Medicaid). 2024 saw a slight rollback in some of the pandemic-era coverage protections, but the strong labor market and ACA subsidies likely kept uninsurance rates relatively low. However, we can expect uninsurance rates to climb in 2025 and beyond because the Republican spending bill both cut Medicaid and allowed the enhanced ACA subsides to lapse. This will lead to more than 14 million people losing health insurance coverage by 2035, increasing the number of uninsured people by more than 40%.1

The Republican budget will also lower incomes and increase food insecurity by cutting SNAP. Benefit reductions and more stringent eligibility requirements will reduce SNAP participation by an average of 2.4 million in the next decade. A weakening labor market will exacerbate this problem by making it more difficult to satisfy new SNAP work requirements as people work fewer hours due to shrinking job opportunities.

Black and Hispanic households will likely represent a disproportionate share of those losing health insurance coverage and access to SNAP benefits. Figure C below shows that people of color are more likely to rely on Medicaid and SNAP benefits. In 2023, SNAP lifted more than 3 million people out of poverty—over half of those were Black or Hispanic, and nearly 40% were children. Cuts to Medicaid, SNAP, and other government support programs mean that the 2024 rate of poverty as measured by the Supplemental Poverty Measure will also likely be the lowest for many years to come.

Figure C

Timely and accurate data are essential but under threat

Next week’s release will also mark the last year in which data from federal statistical agencies could be reliably assumed to be completely free of politicization or manipulation. Staffing cuts and politically motivated firings at government agencies threaten the credibility of future data releases. On August 1,^, Trump fired the Bureau of Labor Statistics Commissioner because he did not like the jobs numbers they released. High-quality public data inform how well the economy is delivering for the majority of working people—whether job opportunities exist, how families make ends meet, and whether families have access to vital services such as nutrition and health care. There is simply no substitute for the government data infrastructure, and pressure from the executive branch to alter data to fit political aims will damage a valuable public good that is critical for business decisions, policymaking, and planning by all stakeholders in the economy.

It is possible that the extreme competence and professionalism of federal workers who staff the statistical agencies will shield most of the data they release from manipulation or quality-erosion. But this will take near-heroic measures and is too much to ask of our civil service—they work hard enough collecting and analyzing this data in professional and non-politicized ways, they should not also have to be activists safeguarding its integrity.

Note

1. CBO projected that 32.4 out of 363.3 million people would be uninsured in 2034. Adding 10 million uninsured due to Medicaid cuts brings the uninsurance rate to 11.7%, compared with the actual 2023 uninsurance rate of 8.0%.

Summary

Tariffs do not automatically raise wages or create good jobs. While strong tariff policies can help preserve jobs in industries facing unfair competition, strong unions are a prerequisite for tariffs to translate into widespread job and wage gains. Without unions, corporate executives have no incentive to pass on the profits they’re gaining to their workers. Only unionized workers can secure a fair share of tariff-driven profits, while most nonunion workers will be left behind.

At the same time, the Trump administration is pursuing unprecedented union busting that will drive unionization rates even lower than they are today. Today only 16% of autoworkers and 18% of steelworkers are in unions compared with 62% of autoworkers and 50% of steelworkers in 1983. Without strong unions, tariffs simply funnel higher profits to corporate executives and shareholders. The potential benefits of tariffs will only be shared with workers if they are combined with improved labor rights and stronger labor unions.

How to fix it

Union leaders need to communicate this message to members and nonunion workers: Trump’s tariffs will benefit businesses, not workers. Tariffs must be combined with pro-labor reforms and support for unionization so that workers can have a voice in the workplace and in policymaking. 

Introduction

In 2024, Donald Trump campaigned on the benefits of tariffs for U.S. workers. He claimed that tariffs would boost wages and create good manufacturing jobs by protecting domestic industries from unfair foreign competition. On the face of it, it might seem like tariffs would automatically protect entire industries, increasing profits for employers and wages for workers. But whether employers share the benefits of tariff protection with workers depends on their bargaining power—something very few workers have without a union.

Without unions, tariffs will mostly just lead to higher corporate profits in protected industries. And with the Trump administration waging the worst union busting in recent American history (McNicholas et al. 2025), high tariffs will mean corporate executives and Wall Street shareholders will see the big payday, not workers. Tariffs alone do not increase wages or create good jobs, unless the industries being protected by tariffs have strong unions.

Since returning to office, President Donald Trump has aggressively increased U.S. tariffs on friend and foe alike. While some of these tariffs have been paused1 and others ruled unlawful by U.S. courts (Lynch and Zakrzewski 2025), Commerce Secretary Howard Lutnick has promised that the Trump administration’s overall approach to broad tariffs is “not going away” (Bacon 2025).

The case against tariffs

By now, there have been many critiques leveled against Trump’s tariffs. They will increase inflation and consumer prices (Peller 2025). They will decrease economic growth and trigger a recession (Burga 2025; Bianco 2025). They open the door to official corruption (Hersh and Bivens 2025). The chaos of tariff increases and reversals roiled the stock market and made it impossible for businesses to plan for the future (Rosen 2025). Tariffs alienate allies needed to strengthen U.S. supply chains (Mui 2025).

These academic arguments may all be correct, but they are unlikely to sway most working-class voters. Trump’s 2018 tariffs were attacked for all the same reasons, yet counties protected by those tariffs still swung toward Trump and GOP candidates in the next election (Autor et al. 2024). It is difficult to counter Trump’s populist promises that tariffs will increase wages and create new manufacturing jobs with elite concerns about the stock market and aggregate economic growth (White House 2025).

What’s tragic about this situation is that Trump’s new tariffs will likely not deliver for most workers, not even those in tariff-protected industries. One could understand working-class voters supporting a tariff policy that prioritized long-term investment and higher pay for workers at the expense of short-term returns for shareholders. But Trump’s tariffs threaten to do just the opposite—increasing profits for bosses while leaving workers far behind.

While it is true that higher tariffs may reduce new layoffs in U.S. manufacturing—a significant goal after decades of free trade caused de-industrialization and millions of job losses—maintaining the status quo is a far cry from Trump’s promise that tariffs will spur re-industrialization with higher wages and new jobs (Kimball and Scott 2014; Hersh and Scott 2021). Though tariffs can offer some job security to workers in trade-affected industries, tariffs will translate into job and wage gains only when companies work collaboratively with their employees.

To be clear, this report assumes that tariffs will boost profits in protected industries, but even that may be overly optimistic. The fact that automakers have posted billion-dollar losses is a reminder that tariffs alone are not a coherent industrial policy (Ewing 2025; Tucker 2025). Domestic industries need strong demand signals for their products. Instead, the Trump administration’s unpredictability has created deep uncertainty for businesses, discouraging investment and undermining long-term planning. Trump’s policy incoherence will likely yield a lose-lose scenario of lower profits and lower wages, while threatening job growth across the manufacturing sector. In other words, even the best-case scenario for Trump’s tariffs will fail to deliver higher wages or good jobs for the vast majority of U.S. workers who lack strong unions.

Tariffs don’t automatically benefit workers

To understand why, we need to revisit the conventional argument about tariffs and workers’ wages in protected industries. Tariffs shield U.S. producers from unfair trade practices and cheaper imports, enabling them to increase market share—and sometimes prices. But how do those higher prices lead to more jobs and higher wages? Here’s how the logic works in the theoretical (and unrealistic) models of an economic textbook.

Higher tariffs for a specific U.S. industry—for example, automobiles—increase the price of imported cars, which allows domestic car producers to raise prices without fear of losing market share. For industries facing chronic competition from unfair trade practices, tariffs help level the playing field by limiting unfairly valued competition. Absent that context, tariffs are a blunt instrument benefitting domestic producers without any accountability for worker pay. The tariff does not make it more expensive to produce cars in the United States but means that car makers can charge a higher price and earn higher profits for each car they produce. In turn, higher profits may induce increased investment and employment in making cars. Since these models assume full employment throughout the economy, no one is out looking for a job, and firms need to raise their wages to attract new workers.

But in the real world, there is little reason to expect that employers will pass on the higher profits they receive in the form of higher wages for their workers. There are two problems that get in the way of workers sharing in the benefits of high tariffs: unemployment and market power. 

Precarious employment breaks the connection between tariffs and wages

First, unemployment can break the connection between tariffs and wages. Although higher prices may lead firms to increase production, they often find a surplus of unemployed workers willing to work at the going wage. This is especially true when government policies like those in the Republican-led budget bill that Trump recently signed into law are sharply cutting social safety nets, pushing unemployed workers to accept jobs at even lower wages. This means that firms facing a slack labor market can turn higher tariffs into higher prices and profits without the need to increase workers’ wages. 

Part of the problem is that many U.S. counties with manufacturing-based economies still have higher unemployment rates from years of neoliberal globalization and cheap imports, even as the national unemployment rate remains relatively low (Autor, Dorn, and Hanson 2021). Given such unemployment in labor markets near manufacturing plants, we should not expect Trump’s new tariffs to increase workers’ wages. Unsurprisingly, this is exactly what recent studies find regarding Trump’s 2018 tariffs: They increased prices and profits, but nominal wages in the tradable sector of the U.S. economy increased by only 0.1%, or about $1.20 a week (Fajgelbaum et al. 2019).

Market power breaks the connection between tariffs and employment

Second, market power can break the connection between tariffs and employment. In competitive markets, higher prices lead firms to expand production and hire more workers. But when an industry is dominated by a small number of firms—as is the case in much of today’s U.S. economy—producers often behave like monopolists. Instead of increasing output to meet higher demand, they restrict production to keep prices—and profits—high, as we saw during the COVID-19 pandemic. Tariffs that raise prices in such markets can simply boost profits without leading to more jobs. Since the 1980s, the rise of corporate consolidation has given firms greater pricing power across the economy, including in key manufacturing sectors like autos and steel (De Loecker, Eeckhout, and Unger 2020).

Consider Stellantis, which chose to spend $3.3 billion on stock buybacks in 2024 after new car prices—but not manufacturing costs—spiked during the COVID-19 pandemic, rather than making investments that would increase production and employment (Stellantis 2024; BLS 2025). And in response to more recent tariff developments in April 2025, Stellantis went ahead with a $2.26 billion dividend payout to shareholders, rather than hiring workers or investing in expanded capacity (Lawrence 2025). For these firms, expanding production in response to higher tariffs would lower prices and erode profitability. Considering the uncertainty caused by Trump’s volatile tariff policy, it’s no surprise that most companies are delaying investments rather than hiring new workers (CBT News 2025). The outcome mirrors what happened in 2018: Tariffs raised prices and profits but failed to increase employment in newly protected industries (Autor et al. 2024)—though they likely helped preserve jobs in industries that might otherwise have lost further ground to unfair foreign competition.

In other words, when companies are given protection from competition that lets them charge more for each unit of output they produce, they do not automatically pass the benefits of this protection onto their workers in the form of higher compensation or expanded employment. So, how can workers make sure that higher tariffs lead to higher wages and more good jobs? One way is by joining a labor union, which increases workers’ bargaining power and helps them capture a share of their companies’ booming profits. Sadly, intentional policy decisions have made this option far too rare for most U.S. workers (McNicholas et al. 2019).

Unions boost profit sharing for workers

The UAW’s ‘Stand Up’ strikes support profit sharing for workers

Now consider the United Auto Workers (UAW), whose 2023 “Stand Up” strike won contracts that once again include profit sharing for workers (UAW 2023). The UAW’s current contracts with the Big-Three auto companies provide workers with profit-sharing bonuses of roughly $250 for every $250 million in company revenue (Martinez 2025). If Trump’s tariffs increase car prices and revenue for the Big Three, unionized autoworkers are guaranteed to see higher incomes. Even before the new auto tariffs, the UAW’s contract with GM led to a $14,500 profit-sharing bonus for union members in 2024. If the Big Three expand production, that would mean more jobs with union wages and benefits. The UAW’s new contracts also protect workers’ right to strike over future plant closures, giving the union some countervailing power against monopolistic companies placing profits over jobs (Bustamante 2023). And those gains for unionized workers also pressured some nonunion employers to raise wages too. But there is no inherent guarantee of good jobs if tariffs result in nonunion automakers increasing U.S. production at the expense of unionized facilities.

Or consider GM’s recent announcement that it would invest $4 billion to create new jobs in the U.S. auto industry. Although the Wall Street Journal (Otts 2025) reported this as a business decision based on Trump’s new tariffs, the reality is that the majority of these investments were negotiated with the UAW back in 2023 to end the “Stand Up” strike. The UAW plays a crucial role in making sure that profits for a company like GM translate into new investment and new jobs for workers. The UAW continues to push the Big Three to respond to tariff increases by reshoring production jobs to the U.S.—where these companies have unused capacity—rather than issuing stock buybacks and special dividends (UAW 2025).

This all means that auto tariffs are good news for UAW members, and it shouldn’t be a surprise that the UAW has offered tentative support for targeted tariff protection for the auto industry. But only 16% of the 1.4 million autoworkers in the U.S. are unionized, so the vast majority of autoworkers should not be surprised when Hyundai, Tesla, and other nonunion auto companies and auto parts suppliers refuse to share tariff-generated profits with their workers (Unionstats.com 2025). As Shawn Fain, president of the United Auto Workers, recently explained, “tariffs increase profits—but only unions increase wages” (Fain 2025).

The United Steelworkers win improvements in profit sharing for U.S. Steel and ArcelorMittal workers

Another example is the steel industry in which the United Steelworkers had to fight U.S. Steel and ArcelorMittal for a share of the profits generated by Trump’s 2018 tariffs. Back then, tariffs increased U.S. steel prices by 30%, and companies like U.S. Steel announced $2 billion in profits (Keller 2018). But when it came time to negotiate a contract with its workers, U.S. Steel offered a small wage increase that would be wiped out by deep cuts to workers’ health care benefits (Lindstrom 2018). It was only after USW members overwhelmingly voted to authorize a strike that the companies agreed to new contracts that increased wages, maintained health care benefits, and improved the companies’ profit-sharing arrangement (USW 2018).

The profit-sharing provision in the current USW contract with U.S. Steel requires the company to share roughly 6% of earnings with steelworkers every quarter (Dolph-Smith 2024). That means that Trump’s new 50% steel tariffs are good news for USW members, but the union only represents 18% of the 295,000 workers in the U.S. steel industry (Unionstats.com 2025). When union density is high, these gains help raise wages for nonunion workers as well. But when unions are against the ropes, as is the case for a vast majority of workers, Trump’s tariffs will likely mean higher prices and corporate profits, but not higher pay.

Workers’ wages haven’t kept pace with increases in their productivity

The inability of most workers to share in their company’s profits is not unique to trade policy. It is just another example of workers’ wages lagging behind increases in worker productivity when they lack strong unions. From 1948 through 1979, when labor unions represented between 20% and 30% of all U.S. workers, wages grew along with increased productivity (Romero and Whittaker 2023). But since 1980, precipitous declines in union density—now at 6% in the private sector—have left wages lagging far behind productivity gains (Durbin 2024). From 1979 to 2019, worker productivity grew by roughly 60%, while workers’ total compensation grew by less than 16% (Mishel 2021).

Nor is the inability of many workers to share in their company’s profits specific to the United States. I examined the relationship between wage growth and productivity growth in 28 manufacturing industries, across 117 countries, from 1986 to 2002. The analysis relies on labor rights data that measure the degree to which a country respects their workers’ rights to act collectively (Dean 2015a). The results demonstrate that when labor rights are weakly protected, the conventional textbook model discussed above systematically exaggerates how much workers gain from tariffs that benefit their industry. Wages rise along with productivity growth when labor rights are well protected, but wages don’t go up at all if labor rights are regularly violated (Dean 2016; Bivens, Hersh, and Weller 2005).

There was a time when labor unions like the UAW and USW were powerful enough to secure higher wages and profit sharing for the majority of U.S. auto and steel workers. In 1983, the UAW represented 62% of all U.S. autoworkers, and the USW represented 50% of all U.S. steel workers. With unionization now down to 16% and 18% of these industries respectively, unions have less power to translate gains to the wider economy.

Trump combines high tariffs with union busting

The already narrow benefits of Trump’s new tariffs shrink even further when we look holistically at the administration’s trade and labor policies together. Since workers need strong unions to share in the benefits of high tariffs, it is crucial to understand that the Trump administration is systematically weakening American unions. Trump has quickly taken a hatchet to workers’ rights, illegally firing government employees, illegally terminating the collective bargaining agreement with Transportation Security Administration workers, and illegally revoking the collective bargaining rights for roughly 1 million federal workers (McNicholas et al. 2025; Berger and Leibenluft 2025; AFGE 2025; EPI 2025). Labor historian Joseph McCartin (Glass 2025) characterized the Trump administration’s activity as “by far the largest single action of union-busting in American history.”

Trump kneecaps the National Labor Relations Board

Perhaps most troubling for unions like the UAW and USW, which mostly represent workers in the private sector, are Trump’s efforts to paralyze the National Labor Relations Board (NLRB), the federal agency that adjudicates disputes about union elections and unfair labor practices (Wiessner 2025). In January 2025, Trump illegally fired Gwynne Wilcox, a member of the National Labor Relations Board, before the end of her term (Kaye and Davis O’Brien 2025). This unprecedented action left the board without a quorum, effectively halting its operations (Cohen 2025). Now, when employers break the law by firing union organizers or intimidating workers to vote against joining a union, workers will literally be denied their day in court. 

Even with a functioning NLRB, such illegal union busting was extremely common; in union elections in workplaces with more than 60 employees, employers were charged with violating federal labor law 54% of the time (McNicholas et al. 2019). Union-busting employers will only be emboldened in their illegal activity now that the NLRB lacks the quorum needed to make final decisions and levy penalties for unfair labor practices.

Weak enforcement of labor law has a chilling effect on unions

Trump’s effective termination of federal labor law enforcement is the exact opposite of what U.S. workers need. In fact, many scholars believe that the already weak enforcement of labor law before Trump’s new term was a main cause of union decline. At the heart of this debate is the growing gap between increasingly high public support for unions and declining union density. Recent surveys find that 71% of Americans have a favorable view of unions, and roughly half of workers say that they would vote to join a union if they had the opportunity (AFL-CIO 2023; Poydock et al. 2025). According to law professor Kate Andrias (2016), the problem is that American labor law is failing to protect workers’ right to join a union: “weak enforcement mechanisms, slight penalties, and lengthy delays—all of which are routinely exploited by employers resisting unionization—fail to protect workers’ ability to organize and bargain collectively with their employers.” Americans want unions, but Trump is making sure they can’t have them.

Trump’s anti-union reforms, especially his paralyzing of the NLRB, pose an existential threat to the country’s remaining powerful unions like the UAW. The growing presence of nonunion auto plants run by multinational companies in the American South has gradually eroded the UAW’s bargaining power with the Big-Three auto companies in the Midwest. These companies point to lower wages and benefits in Southern plants to justify offering lower wages and benefits (Butzel Long 2013). As Bob King, ex-president of the UAW, explained more than 10 years ago, “if we don’t organize these transnationals, I don’t think there’s a long-term future for the UAW” (Krisher 2023). Since the early 1980s, union density in the U.S. auto industry has dropped by 74%. With Trump’s anti-union reforms, high auto tariffs will only mean higher profits for increasingly nonunion auto companies.

How U.S. economic policies compare with those in other countries

Trump’s combination of high tariffs and weak labor rights is moving the U.S. toward a development strategy that will funnel the benefits of re-industrialization to capital instead of labor. If we zoom out and compare Trump’s economic policies with those in other countries, there are striking similarities with Italy in the 1930s, and Brazil and South Korea in the 1960s and 1970s. These are all countries in which authoritarian governments used high tariffs to protect manufacturing industries alongside labor repression to break unions and ensure that wages lagged behind productivity (Zamagni 1993; Seidman 1994; Deyo 1989). 

Of course, there are other examples that the U.S. could follow. There are countries whose governments have combined tariff protection with pro-union policies that made sure workers shared in the benefits of industrialization. In the 1930s, Norway used a combination of high tariffs and pro-labor reforms to build an industrial economy that led to shared economic growth (Grytten 2002). In the middle of the 20th century, Argentina and India both used high tariffs and pro-union policies to spur rapid industrial growth as well as stronger unions, higher wages, and lower income inequality (James 1988; Kohli 2012). Argentina’s and India’s overall development strategies faltered in the late 20th century, but for decades, they produced gains that were broadly shared with workers.

Proposals to expand investment in sustainable new energy industries similarly envision an industrial policy that combines tariffs and subsidies for green industries like wind turbines and electrical vehicles with pro-labor reforms that will make it easier for workers in these industries to join a union and secure a share of the benefits (Tucker et al. 2024). Something similar is playing out now in the European Union, where the trade protection and strong unions of the Green Deal Industrial Plan jointly support good jobs in green industries (European Commission 2023).

History shows that worker support for tariffs isn’t set in stone

When discussing how tariffs will make America great again, Trump often praises President McKinley’s high tariffs in the 1890s (Cabral 2025). Ironically, there is no better example to illustrate that workers will only share in the benefits of high tariffs if they have a strong union. Even better, it shows that the spell can be broken— that workers will understand that high tariffs and weak unions promise nothing more than wealth extraction.

The crucial historical precedent is the infamous Homestead steel strike, which culminated in the election of Grover Cleveland, a Democrat, in 1892 (Dean 2015b). The Republican Party of the late 19th century promised that high tariffs would deliver higher wages for U.S. workers. When a Republican Congress passed the protectionist McKinley tariff of 1890, steel prices and profits started to soar.

Just months before the 1892 general election, the Amalgamated Association of Iron and Steel Workers (the precursor of today’s United Steelworkers) was locked in contract negotiations with the Carnegie Steel Company (the precursor of today’s U.S. Steel). The union argued that “with the metal tariff as it is… there is no reason why the labor of the mills should not have part of the plum of profits that the ownership has been enjoying” (National Labor Tribune 1890). In response, the company locked out the workers and proclaimed that “hereafter, the Homestead steel works will be operated as a non-union mill” (Brody 1960).

The ensuing conflict ultimately ended with four striking workers killed, the town of Homestead occupied by the Pennsylvania National Guard, and the union decimated (Burgoyne 1893).

Although the steelworkers lost the strike, the conflict became a major focal point in the 1892 presidential election. Workers throughout the country began to question whether the GOP’s high tariffs actually delivered benefits to U.S. workers.

While the steelworkers had loyally voted for Republicans for decades, their new views on tariffs led many of them to throw their support to the Democratic Party (Arizona Republican 1892). The industrial states of Illinois, Indiana, New York, and Wisconsin all voted for Grover Cleveland, the victorious Democratic candidate for president.

What does all this mean for today’s politics? 

Trump’s 2018 tariffs, which increased average U.S. tariffs from 2% to 4%, helped him win voters in 2020 (Tradingeconomics.com 2025; Autor at al. 2024). Will Trump’s more aggressive 2025 tariffs, which are set to increase average tariffs to 16% or more, do the same (The Budget Lab 2025)? Given the recent memory of high inflation, many Democrats hope that linking Trump’s new tariffs to rising consumer prices will help win back working-class voters (DNC 2025).

There is some evidence that this time might be different, with overall public support for tariffs decreasing dramatically since Trump began increasing tariffs at the start of his second term. In early June, a survey found that 64% of independent voters opposed Trump’s tariff plans (Keene 2025).

And yet, Trump’s tariffs still resonate with many Americans. According to a recent Gallup poll, half of Americans expect tariffs to create more manufacturing jobs, and 69% are willing to accept at least some economic disruption to realize such gains (Brenan 2025). Or consider a recent focus group with Pennsylvania voters who voted for Biden in 2020 but Trump in 2024 (Talev 2025). Most of these swing voters expressed continued support for Trump and his tariffs because “things have been so screwed up for so long and he’s finally doing something about it.”

When the next elections arrive, will voters in swing states like Pennsylvania and Michigan reward Trump for following through on one of his main campaign promises, or will they reward Democrats for attacking tariffs in ways that sound a lot like defending free trade (Stiglitz 2024)?

How to construct a progressive message on tariffs

Thankfully, there is an alternative. Progressives can explain that Trump’s economic policies are meant to benefit businesses and not workers. As long as working-class voters believe that they benefit from tariffs, they will only be further alienated by politicians who critique tariffs for roiling the stock market and creating business uncertainty. Progressives will defeat Trumpism by embracing a progressive version of economic populism that combines tariffs and pro-labor reform to deliver shared prosperity (Reston 2025).

Union leaders need to communicate this message to rank-and-file members, as well as nonunion workers. Trump’s tariffs will only benefit the small minority of workers that have the bargaining power to win improvements in the workplace. And with Trump’s anti-labor policies simultaneously crushing unions, those benefits will be enjoyed by fewer and fewer workers. In the short term, workers who want to share in the potential benefits of high tariffs need to join a union. In the long term, workers who want to share in the prosperity of re-industrialization need to fight for policies that strengthen workers’ unions as well as their industries.

About the author

Adam Dean is an associate professor in the Department of Political Science at George Washington University. He is the author of two books on international trade and labor politics: From Conflict to Coalition (2016) and Opening Up by Cracking Down (2022), both of which were published by Cambridge University Press.

Notes

1. Proclamation No. 14266, 90 Fed. Reg. 15625 (April 9, 2025).

References

American Federation of Government Employees (AFGE). 2025. “AFGE Sues Trump Administration for Illegally Terminating TSA Officers’ Union Contract” (web page). March 17, 2025.

American Federation of Labor and Congress of Industrial Organizations (AFL-CIO). 2023. “AFL-CIO’s Shuler: State of the Unions Is Strong with Record Public Support, Unprecedented Activism and Organizing this Labor Day” (press release). August 29, 2023.

Andrias, Kate. 2016. “The New Labor Law.” Yale Law Journal 126, no. 1: 2–100.

Arizona Republican. 1892. “In Doubt!” November 9, 1892.

Autor, David, Anne Beck, David Dorn, and Gordon H. Hanson. 2024. “Help for the Heartland? The Employment and Electoral Effects of the Trump Tariffs in the United States.” National Bureau of Economic Research Working Paper no. 32082, January 2024.

Autor, David, David Dorn, and Gordon H. Hanson. 2021. “On the Persistence of the China Shock.” National Bureau of Economic Research Working Paper no. 29401, October 2021.

Bacon, Auzinea. 2025. “Commerce Secretary Says ‘Tariffs Are Not Going Away’ as Trump’s Trade Plans Face Legal Battles.” CNN, June 1, 2025.

Berger, Sam, and Jacob Leibenluft. 2025. Trump Administration’s Mass Layoffs of Federal Workers Are Illegal. Center on Budget and Policy Priorities, May 2, 2025.

Bianco, Ali. 2025. “Musk Goes Scorched Earth: Trump Will Cause Recession, Implies He Should Be Impeached.” Politico, June 5, 2025.

Bivens, L. Josh, Adam S. Hersh, and Christian E. Weller. 2005. “How Well Are We Measuring Workers’ Rights?” Review of Radical Political Economics 37, no. 3 (September): 302–310.

Brenan, Megan. 2025. Most Americans Skeptical About Benefits of Tariffs. Gallup, April 28, 2025.

Brody, David. 1960. Steelworkers in America: The Nonunion Era. Cambridge, M.A.: Harvard Univ. Press.

Bureau of Labor Statistics (BLS). 2025. “Producer Price Index by Industry: Automobile, Light Truck and Utility Vehicle Manufacturing [PCU336110336110].” Retrieved from FRED, Federal Reserve Bank of St. Louis, August 20, 2025.

Burga, Solcyré. 2025. “Trump’s Tariffs Hit Harder Than Expected as OECD Slashes U.S. Growth Forecast.” Time, June 3, 2025.

Burgoyne, Arthur Gordon. 1893. Homestead. A Complete History of the Struggle of July, 1892, Between the Carnegie Steel Company, Limited, and the Amalgamated Association of Iron and Steel Workers. Pittsburgh, P.A.: Rawsthorne Engraving and Printing Co.

Bustamante, Alí R. 2023. “The UAW Strike Was a Success for Workers and Economic Democracy” (blog post). Roosevelt Institute website, November 20, 2023.

Butzel Long. 2013. “The UAW’s Organizing Campaign at Volkswagen’s Chattanooga, Tennessee Plant” (Client Alert: Labor and Employment). November 11, 2013.

Cabral, A. R. 2025. “Meet Trump’s Hero, William McKinley—the OG Tariff Guy.” U.S. News and World Report, April 24, 2025.

CBT News. 2025. “U.S. Tariffs Cost Automakers 18 Months of Planning, S&P Global Says: Automakers Are Delaying Investment and Production Decisions Due to the Uncertainty.” June 6, 2025.

Cohen, Marshall. 2025. “Court Lets Trump Fire Labor and Worker Protection Board Members While They Fight to Keep Their Jobs.” CNN, March 28, 2025.

Dean, Adam. 2015a. “Power over Profits: The Political Economy of Workers and Wages.” Politics & Society 43, no. 3: 333–360.

Dean, Adam. 2015b. “The Gilded Wage: Profit-Sharing Institutions and the Political Economy of Trade.” International Studies Quarterly 59, no. 2: 316–329.

Dean, Adam. 2016. From Conflict to Coalition: Profit-Sharing Institutions and the Political Economy of Trade. New York: Cambridge Univ. Press.

De Loecker, Jan, Jan Eeckhout, and Gabriel Unger. 2020. “The Rise of Market Power and the Macroeconomic Implications.” Quarterly Journal of Economics 135, no. 2: 561–644.

Democratic National Committee (DNC). 2025. “Trump’s Economy: Rising Prices, Shrinking Industry, and Distressed Consumers” (website). June 18, 2025.

Deyo, Frederic. 1989. Beneath the Miracle: Labor Subordination in the New Asian Industrialism. Berkeley, C.A.: Univ. of California Press.

Dolph-Smith, Sarah. 2024. “2024 1st Quarter Profit Sharing” (blog post). USW Local 6787 website, April 25, 2024.

Durbin, Dee-Ann. 2024. “US Unions Flexed Their Muscles Last Year, but Membership Rates Fall to All-Time Low.” Associated Press, January 23, 2024.

Economic Policy Institute (EPI). 2025. “Executive Order on ‘Exclusions from Federal Labor-Management Relations Programs.’” Federal Policy Watch Blog (Economic Policy Institute), August 11, 2025.

European Commission. 2023. “The Green Deal Industrial Plan: Putting Europe’s Net-Zero Industry in the Lead” (press release), January 31, 2023.

Ewing, Jack. 2025. “G.M. Profit Shrinks on Billion-Dollar Tariff Hit.” New York Times, July 22, 2025.

Fain, Shawn. 2025. “Labor Voices: Tariffs Can Be Effective If Implemented Correctly.” Detroit News, April 10, 2025.

Fajgelbaum, Pablo D., Pinelopi K. Goldberg, Patrick J. Kennedy, and Amit K. Khandelwal. 2019. “The Return to Protectionism.” Quarterly Journal of Economics 135, no. 1: 1–55.

Glass, Aurelia. 2025. The Trump Administration Ended Collective Bargaining for 1 Million Federal Workers. Center for American Progress, May 22, 2025.

Grytten, Ola Honningdal. 2002. “Norwegian Policy Response to the International Economic Disintegration During the Inter-War Years.” Norwegian School of Economics and Business Administration Discussion Paper no. 23/02, December 12, 2002.

Hersh, Adam S., and L. Josh Bivens. 2025. Tariffs—Everything You Need to Know but Were Afraid to Ask (FAQ). Economic Policy Institute, February 2025.

Hersh, Adam S., and Robert Scott. 2021. Why Global Steel Surpluses Warrant U.S. Section 232 Import Measures. Economic Policy Institute, March 2021.

James, Daniel. 1988. Resistance and Integration: Peronism and the Argentine Working Class, 1946–1976. New York: Cambridge Univ. Press.

Kaye, Danielle, and Rebecca Davis O’Brien. 2025. “Trump Firings at Labor Board Paralyze the Agency.” New York Times, January 28, 2025.

Keene, Camille. 2025. “Americans Continue to Disapprove of Trump and His Handling of the Economy.” Navigator website, June 18, 2025.

Keller, Megan. 2018. “Steel Workers Union Leaders Demand Pay Increase as Profits Rise After Tariffs.” The Hill, September 11, 2018.

Kimball, Will, and Robert Scott. 2014. China Trade, Outsourcing and Jobs: Growing U.S. Trade Deficit with China Cost 3.2 Million Jobs Between 2001 and 2013, with Job Losses in Every State. Economic Policy Institute, December 2014.

Kohli, Atul. 2012. Poverty amid Plenty in the New India. New York: Cambridge Univ. Press.

Krisher, Tom. 2023. “UAW’s Clash with Big 3 Automakers Shows Off a More Confrontational Union as Strike Deadline Looms.” The Hill, September 4, 2023.

Lawrence, Eric D. 2025. “Stellantis Shareholders OK $2.26B Dividend, but Stock Buybacks on Hold.” Detroit Free Press, April 15, 2025.

Lindstrom, Natasha. 2018. “Steelworkers Scoff At U.S. Steel’s Latest Proposal, Move Closer Toward Strike.” TribLive.com, September 12, 2018.

Lynch, David, and Cat Zakrzewski. 2025. “Trump’s ‘Liberation Day’ Tariffs Halted by Court of International Trade.” Washington Post, May 28, 2025.

Martinez, Michael. 2025. “Stellantis UAW Bonuses Lowest Since 2014 While GM Payouts Average Record $14,500.” Automotive News, February 26, 2025.

McNicholas, Celine, Margaret Poydock, Julia Wolfe, Ben Zipperer, Gordon Lafer, and Lola Loustaunau. 2019. Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns. Economic Policy Institute, December 2019.

McNicholas, Celine, Samantha Sanders, Josh Bivens, Margaret Poydock, and Daniel Costa. 2025. 100 Ways Trump Has Hurt Workers in His First 100 Days. Economic Policy Institute, April 2025.

Mishel, Lawrence. 2021. “Growing Inequalities, Reflecting Growing Employer Power, Have Generated a Productivity–Pay Gap since 1979: Productivity Has Grown 3.5 Times as Much as Pay for a Typical Worker.” Working Economics Blog (Economic Policy Institute), September 2, 2021.

Mui, Christine. 2025. “What Trump’s Trade War Means for ‘Friendshoring.’” Politico, February 5, 2025.

National Labor Tribune. 1890. “Business of Pittsburgh Mills.” June 7, 1890.

Otts, Christopher. 2025. “GM Plans $4 Billion Investment to Boost U.S. Manufacturing.” Wall Street Journal, June 10, 2025.

Peller, Lauren. 2025. “CBO Says Trump’s Tariffs Will Shrink the Economy and Add to Inflation but Reduce the Deficit.” ABC News, June 4, 2025.

Poydock, Margaret, Celine McNicholas, Jennifer Sherer, and Heidi Shierholz. 2025. 16 Million Workers Were Unionized in 2024: Millions More Want to Join Unions but Couldn’t. Economic Policy Institute, January 2025.

Reston, Maeve. 2025. “Sanders Says Harris Fell Short with Working Class. He Has a Plan to Fix That.” Washington Post, May 31, 2025.

Romero, Paul D., and Julie M. Whittaker. 2023. A Brief Examination of Union Membership Data. Congressional Research Service R47596, June 16, 2023.

Rosen, Phil. 2025. “Tariffs Have Spooked Investors Out of the U.S. Stock Market.” Inc., May 30, 2025.

Seidman, Gay W. 1994. Manufacturing Militance: Workers’ Movements in Brazil and South Africa, 1970–1985. Berkeley, C.A.: Univ. of California Press.

Stellantis. 2024. “2024 Share Buyback Program” (website). Retrieved August 20, 2025.

Stiglitz, Joseph. 2024. “The Message to Democrats Is Clear: You Must Dump Neoliberal Economics.” Guardian, November 28, 2024.

Talev, Margaret. 2025. “Focus Groups: Pennsylvania Swing Voters Voice Tariff Confusion.” Axios, April 11, 2025.

The Budget Lab at Yale (The Budget Lab). 2025. “State of U.S. Tariffs: May 12, 2025” [Excel files], May 12, 2025.

Tradingeconomics.com. 2025. “United States—Customs and Other Import Duties (% of Tax Revenue).” Sourced from the World Bank. Retrieved August 20, 2025.

Tucker, Todd N. 2025. Trump’s Tariff Tantrum: How Sweeping Tariffs Came to Be and Why It Matters. Roosevelt Institute, April 3, 2025.

Tucker, Todd, Kyunghoon Kim, Saule T. Omarova, Jonas Algers, Andrea Furnaro, César F. Rosado Marzán, and Lenore Palladino. 2024. Industrial Policy 2025: Bringing the State Back In (Again). Roosevelt Institute, February 11, 2024.

Unionstats.com. 2025. “Union Membership, Coverage, and Earnings from the CPS.” Retrieved August 20, 2025.

United Auto Workers (UAW). 2023. “Supplemental Agreement Covering Profit Sharing Plan.” October 30, 2023.

United Auto Workers (UAW). 2025. “New UAW Analysis Shows U.S. Can Create Tens of Thousands of Auto Jobs Using Existing Capacity.” April 30, 2025.

United Steelworkers (USW). 2018. “USW Members Vote to Ratify 4-Year Contract with U.S. Steel.” November 13, 2018.

Wiessner, Daniel. 2025. “Trump Paralyzes US Labor Board by Firing Democratic Member.” Reuters, January 28, 2025.

White House. 2025. President Donald J. Trump Increases Section 232 Tariffs on Steel and Aluminum (fact sheet). June 3, 2025.

Zamagni, Vera. 1993. The Economic History of Italy: 1860–1990. New York: Oxford Univ. Press.

This is a fair-enough characterization of what the blog post said about the steel and aluminum tariffs it was analyzing, but it misleads in applying this conclusion to the tariffs being proposed yesterday (and in previous weeks) by the current Trump administration. Specifically, the steel and aluminum tariffs examined in that EPI blog post were narrowly targeted and applied to roughly $50 billion of imports, less than 2% of the imports that yesterday’s announced tariffs would apply to. In terms of assessing the effect of tariffs, this large difference in scale matters crucially.

It is hard to overstate what a radical change yesterday’s tariff announcements are relative to either the steel and aluminum tariffs we examined in that blog post or even to the full suite of tariffs passed in the first Trump administration. The first Trump administration’s tariffs were far narrower and only undertaken after a lot more study and consultation. For example, seven months into his first term, President Trump ordered the United States Trade Representative to investigate Chinese trade practices. The report was released in March 2018 and found a number of instances of China engaging in unfair practices. In that same month (over a year into the term), the Trump administration announced tariffs on up to $60 billion worth of imports from China.

By the end of the first Trump administration, further tariffs had been implemented. With these additional tariffs, roughly $380 billion worth of goods were subject to these taxes, which had been phased in over four full years.

Now, Trump is imposing tariffs on essentially all goods imports—more than $3 trillion—after just over two months into his term. Tariffs can be a legitimate and useful tool in industrial policy for well-defined strategic goals, but broad-based tariffs that significantly raise the average effective tariff rate in the United States are unwise. Further, the second Trump administration’s rationale, parameters, and timeline for tariffs have been ever-shifting. As the original post cited by the administration argues, tariffs should not be a goal unto themselves, but a strategic tool to pair with other efforts to restore American competitiveness in narrowly targeted industrial sectors.

So far, the “stock market veto” has turned out to be nothing more than wishful thinking. In his second term, President Trump has continued to loudly proclaim his support for the anti-growth policies of broad and high tariffs and mass deportations. He has also added a new anti-growth twist of arbitrary and illegal firings of federal employees and cancellations of federal contracts. Finally, he has enthusiastically backed a U.S. House budget resolution that would slash disposable incomes for the bottom half of U.S. households. Besides being substantively unwise, these policy efforts have been undertaken with maximum chaos.

And the stock market has indeed rebelled. The S&P 500, for example, is down 8% in the last month.

This raises the question: Was there ever anything useful in the “stock market veto” view of the world? After all, there is no general correlation between stock market movements and what’s good for broadly shared growth, so it seemed odd to think a stock market veto would somehow come to the aid of the broader U.S. economy.

In what follows, we’ll present the good and the bad of stock market ups and downs and what they might mean for the trajectory of economic policy. In the end, it turns out that today the stock market and the economy are mirroring each other: Stock market weakness is reflecting broader economic weakness. In short, if there was ever going to be a “stock market veto” of broadly anti-growth policies, it is past time for it to kick in.

What does the stock market measure?

Most of the time when people are talking about the stock market, they are referring to an index that measures changes in the prices of stocks of a number of publicly traded companies. Examples are the S&P 500 index (which includes 500 firms) and the Dow Jones Industrial Average (which includes about 30 firms).

In theory, the stock price of any given firm reflects investors’ expectations about the returns they will see from holding its stock into the future. These returns include the annual dividends paid out from these firms to shareholders and the capital gain that would be available if shares were sold in the future (with the capital gain being the difference in price between when it is sold and when it was bought).

The stock market does not measure broader economic health

Because stock market prices generate a new data point every day (or even every hour), they tend to be discussed far more than broader measures of economic health. However, many know that the stock market has little to do with the economic health of U.S. households. For one, the stock market doesn’t tell us anything about jobs and wages. There are times when broad-based economic strength pushes up jobs, wages, and the stock market, and other times when broad-based weakness causes these all to fall together. But more often what is happening to stock prices gives us no insight into the wider economy. And as we will talk more about below, there are times when stock market strength can be a pure zero-sum transfer away from the wages of typical workers—reflecting stronger profits earned solely from successful wage suppression strategies.

Because most households depend overwhelmingly on wages from work as their primary source of income and not returns from wealth-holding, the stock market tells us nothing about these households’ economic situations. The wealthiest top 10% of households own over 85% of all corporate stock, and the top 1% alone own roughly 50%. Roughly half of all U.S. households have essentially zero invested in the stock market, even when including indirect investments they might have, like holdings in 401(k)s.

What causes stock prices to rise or fall?

There are dozens of reasons why an individual firm’s stock price might rise or fall. A drug company might announce the discovery of a new blockbuster drug. An oil company’s stock might rise if the global price of oil begins rising. For the stock market as a whole, it is usually macroeconomic trends that cause increases or decreases in stock indexes. But just because it is macroeconomic trends that drive overall stock indexes, this does not mean they always move in the same direction—sometimes favorable macroeconomic trends can push up stock prices, but sometimes they can actually push stock prices down.

Overall pace of growth

One macroeconomic influence that can affect stock prices broadly is expectations about the pace of economic growth in the future. If, for example, investors believe that economic growth will accelerate in coming years, this should (all else equal) lead them to believe that profits will accelerate, and this should lead to some combination of faster dividend growth or higher capital gains in the future. Expectations about the pace of future economic growth can be driven either by expectations of long-run productivity growth (how fast the economy can grow on average over long periods of time holding inputs fixed) or by expectations about the business cycle (i.e., whether the economy is about to enter or emerge from a recession).

Redistribution from wages to profits

A redistribution of income from wages to profits is another macroeconomic influence that affects stock prices. For any given rate of expected growth in corporate sector income, the value of owning stock rises if profit’s share of this income rises. A number of things influence this distribution of corporate income between profits and wages. A rise in monopoly power in product markets, for example, will boost profits and therefore raise the share of corporate income claimed by shareholders. Similarly, a decline in labor’s bargaining power in labor markets will also cause such a redistribution by boosting profits. For example, some estimates indicate that nearly half of the rise in real corporate sector wealth between 1989 and 2017 can be attributed to a zero-sum transfer from wages to profits.

Reduction in corporate income taxes

U.S. corporations pay income taxes on their profits. Reductions in this corporate income tax rate (either through legislation or tax avoidance strategies) will hence make any $1 in pre-tax income more valuable to investors, and this will broadly bid up the price of stock.

Interest rate movements

We’ve established that the price of a given stock represents investors’ expectations of the returns to owning this stock in the future. But whenever future gains must be assigned an economic value to compare with current values, one must use an appropriate discount rate, which measures how much more valuable $1 is today than $1 is a year from now. Even ignoring inflation, there are reasons for this discounting. For example, people are impatient and $1 in consumption today is valued more highly than $1 next year. Further, every $1 in consumption foregone today can be invested and yield more than $1 next year.

The measure of how much money invested today can earn over the next year risk-free is usually proxied by something like the rate of return to U.S. Treasury bonds, which are assumed to never default. As discount rates fall, the value of wealth today rises. This means that falling interest rates push up the current value (and price) of the stock market broadly. This effect is sometimes underrated in how powerful it can be. Say that you think stock prices generally reflect what investors believe returns will be over the next 10 years. In this case, a fall in interest rates from 7% to 3%—essentially the fall we have seen over the past 30 years in the U.S. economy—will boost the value of stock prices by almost 20%.

The Trump administration’s policy agenda is weakening the economy—including the stock market

The real value of the S&P 500 rose a stunning 27% in the last year of the Biden administration. Further, the share of corporate sector income claimed by profits instead of wages remains extraordinarily high in historical terms. Despite this, many felt after the election that stock markets would continue rising during the Trump administration because they inherited a fundamentally strong macroeconomy that would continue to generate profit (and wage) growth. Further, the inflationary spike of the immediate post-COVID recovery was clearly over, and there seemed to be some room for the Federal Reserve to reduce interest rates.

There was zero concern among investors that the Trump administration would raise taxes on corporations, and even some hopes that they would be cut further. Finally, the first Trump administration was a never-ending assault on institutions and policies that provide bulwarks to typical workers’ leverage and bargaining power in labor markets. These assaults had been successful in keeping real wage growth lower than one would have expected based on historical experiences with unemployment rates as low as what prevailed in the late 2010s. In short, as long as the second Trump administration did not do something extremely stupid to upend the strong inertial growth they were inheriting, their tax policies and their policy aim of redistributing money from wages to corporate profits were expected to bolster stock prices.

And yet stock prices have been falling in the past month as the Trump administration’s policy agenda comes into clearer view. The agenda is anti-growth and inflationary—a rare and bad combination. Slowing overall growth will slow profit (and wage) growth in coming years, which will depress stock prices. The administration’s illegal cutbacks to federal employment and contracts—and the spending cuts in the House budget resolution (if passed)—will drag sharply on economy-wide demand. The chaotic tariff policy and threatened mass deportations will constitute a large and inflationary supply-side shock. This will both slow growth and provide a real obstacle to the Federal Reserve continuing to cut interest rates.

The current Trump administration has predictably launched a number of attacks on workers’ bargaining power, and these might prove effective in boosting profits at the expense of wages. But the profit share of income in the corporate sector is already starting from an extremely elevated level, making further increases likely harder to attain. Further, the effect on stock prices of any such redistribution will likely be overwhelmed by the broader downward pressures noted above.

If a recession caused by the Trump policy agenda (as well as by any consumption cutbacks spurred by the falling stock market itself) is sharp enough, the Federal Reserve will cut interest rates. Given what we noted about the powerful effect of interest rate cuts on stock prices, this could halt the fall in stock prices and even provide some slight rebound. But by then the damage to the real economy— households’ jobs and wages—will have been done.

In short, while the stock market isn’t the economy, the stock market declines we have seen in recent weeks are genuinely worrying. They are a symptom of much larger dysfunctional macroeconomic policy that will likely soon start showing up in higher unemployment and slower wage growth for the vast majority. If ever there was a time for the “stock market veto” to activate, it is right now.

Instead, the administration seems determined to squander and wreck the strong economy. Each of the individual policies they are pursuing—illegal layoffs of federal workers, mass deportations, constant threats and retractions of broad-based tariffs, and Medicaid spending cuts—would be bad for the economy. But each policy is also being pursued with maximum levels of chaos and incoordination, creating unprecedented levels of economic uncertainty. This uncertainty is itself a serious economic threat.

Below, we sketch out the macroeconomic dangers posed by each of the administration’s big policy initiatives so far, and end with an assessment of where this leaves the U.S. economy. The best outcome that could result from continuing these policies would be avoiding a recession but still sharply reducing growth and creating an U.S. economy that is significantly poorer than it would have otherwise been. The most likely outcome, however, is a recession in the coming year.

DOGE and the illegal attack on the federal workforce and payments

Among all the Trump administration blunders, the so-called Department of Government Efficiency’s (DOGE) illegal layoffs and contract cuts will show up first in macroeconomic data (they might already be showing up in unemployment insurance claims). The size of these cuts to date are not large enough to cause a recession by themselves, but if DOGE continues to pursue further cuts and they begin to add up to significant shares of the federal workforce, they could certainly spark a recession.

Further, the medium- and long-term damage of DOGE’s efforts are profound. DOGE is arbitrarily hacking away at key state capacity that is incredibly valuable and would be hard to reassemble quickly in the future. Their cuts also open up a number of new possible channels to trigger an economic crisis. Any one of these potential crises has a low probability of coming to pass, but creating multiple new possibilities for an economic wreck is extraordinarily reckless.

For example, cuts might erode our ability to detect and manage new pandemics. Or they might hamstring key public services that facilitate private-sector activity (such as air travel safety or scientific research). Given DOGE’s multiple demonstrated instances of misunderstanding basic facts about the government systems they are meddling with, it is far from inconceivable that DOGE could inadvertently cause key federal payments (say, interest payments to U.S. Treasury holders) to be missed or delayed. All in all, the unanticipated illegal DOGE attack on the public sector is the clearest current danger to the economy, which is saying a lot given the other bad policy efforts underway.

Further, the DOGE effort also creates huge uncertainty about how long the legal challenges will take to play out, whether Republican-appointed judges will be willing to cross the Trump administration, how long and in what form restitution will take, and even whether the administration will further the constitutional crisis by ignoring court decisions.

The Republican budget marching through the reconciliation process

The House budget resolution calls for cutting spending programs like Medicaid to help pay for a tax cut primarily for the rich, which would transfer income away from the bottom half of U.S. families and toward the very top. This is a terrible outcome by itself, one that would increase economic misery and inequality.

But the budget resolution if passed would also noticeably drag on overall demand growth in the economy. When disposable incomes fall for families living paycheck to paycheck, their spending falls in lockstep. Conversely, when rich families get a tax cut, they tend to save most of it. This decline in demand alone would force the Federal Reserve to cut interest rates to keep unemployment from rising. In turn, this would give the Fed much less ammunition to push back on any other recessionary force that arises in coming months.

This reconciliation process is also creating great uncertainty. It seems fairly sure that Republicans will extend the so-called Tax Cuts and Jobs Act (TCJA)—it is a universal Republican priority to keep taxes low for the rich. But it is still unclear whether this will be financed through cuts to spending like Medicaid (which would destroy demand) or just adding to the deficit (which would expand demand). Added on top of these layers of uncertainty are the possibility of a government shutdown and fights over the debt ceiling in the coming months.

Mass deportations

Should the Trump administration’s stated desires for mass deportations come to pass, this would constitute a profound shock to both demand and supply in the macroeconomy. In sectors where immigrants are a much larger share of workers than consumers—like food production, construction, and child care—prices will rise sharply as labor supply craters. In sectors where immigrants are more likely to be consumers than workers—retail, professional services, and durable manufacturing—demand will dry up.

Research has shown clearly that as immigrant workers are forced out of economies by enforcement measures, these economies suffer sharp employment declines of both immigrant and U.S.-born workers.

The final scope of the mass deportation efforts is also highly uncertain in macroeconomic terms (in human terms, we know this will cause great suffering). The administration might decide they are fine with highly performative acts of cruelty but not directly target millions of families. Or they might actually try to deport the millions they claimed they would target during the presidential campaign. Forecasting overall labor demand and supply given this uncertainty is extremely hard.

Unstrategic and chaotic tariff policy

Tariffs can be a useful policy tool. But the Trump administration’s approach to trade policy has been completely bizarre, with the strategic goal they claim to be targeting changing by the day (or even hour). And instead of using tariffs surgically to meet discrete goals, the Trump administration tariffs have been sweeping and across the board—often falling on goods (like many agricultural products) that the U.S. has no ability at all to produce domestically.

The initial rollout of widespread and high tariffs with no strategic goal was bad enough—these would have caused lots of pain to consumers while doing little to help workers in industries (like manufacturing) who have been genuinely damaged by past decades’ trade policy. But the administration has threatened—and even imposed—such tariffs only to yank them away again and again.

Radical uncertainty for its own sake

Each of these policy measures would be bad for the economy and typical families on their own. The fact that each comes with huge uncertainties creates even further damage.

This approach of pursuing destructive policy measures in the most chaotic and unpredictable way possible is tailor-made to freeze business investment in plants and equipment. What business will want to make plans about the future when key questions of demand, supply, the potential workforce, and the costs of inputs are entirely up in the air?

Swings in business investment are the most volatile parts of gross domestic product (GDP) in normal times, often leading the economy into recession. A policy approach that has led to the highest levels of economic uncertainty in history—higher than even the first months of the COVID-19 pandemic—is not a pro-investment agenda.

It is again worth noting the incredibly favorable investment environment that was handed off to the Trump administration. Business investment during the Biden administration grew faster than from 2017–2019, and significantly faster than it had averaged between 1979 and 2019. Further, the share of profits in corporate sector income hit historic highs in the pandemic recovery, offering huge incentives for businesses to keep investing in plants and equipment.

A policy agenda that manages to swamp this favorable investment momentum and send it into reverse would be a disaster, yet that seems to be where the radical uncertainty created by the Trump administration is taking the U.S. economy.

Uncertainty handcuffs the Federal Reserve

Finally, this uncertainty greatly complicates the Fed’s job. Inflation is currently running ahead of the Fed’s desired target, yet they have admirably tested the waters of lower interest rates because they have forecasted continued progress on lowering inflation even with slightly lower interest rates. But their forecasts are now far more uncertain. Multiple policy whipsaws that undermine both demand and supply growth will leave them very unsure about when and how much they should move interest rates to counter these policy shocks. Given that Fed policy is the first line of defense against recessions, policy uncertainty that makes it harder for them to recognize when the balance of demand and supply is getting out of line will see them acting less quickly than needed to recessionary shocks.

The U.S. will unambiguously be poorer at the end of Trump’s term

Absent a radical reversal of the current policy agenda, the U.S. will be a poorer country at the end of Trump’s term than it should have been. The only open question is how rapidly this de-growth will happen, whether more quickly through a sharp recession or more slowly as the supply destruction outpaces demand destruction.

My personal view is that avoiding a recession over the next year will require huge luck. The luck will either take the form of the administration quickly backtracking on many of its worst ideas, or in the fact that supply destruction will happen quickly enough that the demand destruction will not lead to spiking unemployment. But this second scenario is still very bad for U.S. economic growth, and either scenario will cause economic pain that was easily avoidable.

Far fewer federal employees than expected voluntarily accepted the offer of deferred resignation by the Trump administration’s deadline, meaning the administration will need to turn to more drastic measures to hit their target of shedding at least 10% of the federal workforce. Public reporting this week indicates that the administration has instructed at least some agencies to fire employees in their probationary period—meaning they have generally worked for the federal government for less than a year and thus lack the full job protections and right to appeal wrongful firings that most federal employees have. The American Federation of Government Employees (AFGE) announced their intention to continue to pursue every available legal challenge to these mass firings.

The Trump administration continued to illegally fire leadership of independent agencies—those agencies outside of the President’s cabinet that were specifically established to be insulated from the White House and political influence. These include firing the Chair of the Federal Labor Relations Authority and a member of the Merit Systems Protection Board. While the work of these agencies is often under the radar, they have an important role in defending the legal rights of federal employees and ensuring smooth labor relations between the government and those who make it run. The Trump administration also effectively stopped work at the Consumer Financial Protection Bureau this week, an independent agency responsible for protecting Americans from fraudulent financial products and services.

As the Trump administration farms out most of its work to kneecap agencies to Elon Musk’s DOGE team, it’s important to note that Musk has been under investigation for potential illegal activities at his companies by many of the agencies now in his crosshairs. The Occupational Safety and Health Administration (OSHA) investigated a Tesla facility after the death of a worker last year. And among independent agencies, the National Labor Relations Board (NLRB) ruled that SpaceX illegally fired workers for taking collective workplace action. Musk has also faced scrutiny from the Securities and Exchange Commission (SEC) into whether he misled investors or committed securities fraud. Without transparent disclosure of his financial conflicts—and with political appointees under his direction accessing sensitive agency documents and privileged government data—Musk may well get personal financial benefits out of DOGE’s operations.

The administration retreats on trade policy

Much hay was made about the administration’s teased announcement of a reciprocal tariffs policy this week. However, the White House’s presidential memorandum is ultimately all talk and no action—it simply orders more studies and policy recommendations from agencies to be added to an April report, and does not actually impose any new tariffs. It’s notable that when it comes to trade, the White House has either rolled back or soft-walked most of their policy actions. The chaotic announcement in early February of large, broad-based tariffs on Canada, China, and Mexico faced such blowback from allied nations, the business community, and consumers that the supposed tariffs on Canada and Mexico were almost immediately placed on pause pending further negotiations. President Trump also walked back the suspension of a tax exemption on cheaper imports from China. On trade policy, the administration appears to be more interested in creating chaos and the illusion of action.  

As a reminder, you can find a more comprehensive catalogue of administration policies relevant to working people and the economy at Federal Policy Watch, our online tool documenting actions by the Trump administration, Congress, federal agencies, and the courts. You can subscribe to daily Federal Policy Watch updates here.

Updated March 28, 2025 to include the impact of other countries’ value-added taxes (VATs) on U.S. exports. Jump to the update →

This document will be periodically updated as new questions arise. 

During his presidential campaign, President Trump pledged to impose universal tariffs of 10–60% on all U.S. imports—a whopping $4.2 trillion in goods and services purchased from abroad in 2024. This was always a real possibility.

The International Economic Emergency Powers Act gives the president broad authority to do so. In early February the Trump administration seemed to be making good on the threat to enact extremely high and broad-based tariffs. They announced tariffs of 25% on all goods from Mexico and all goods (except energy goods) from Canada, as well as tariffs of 10% on all goods from China, though ultimately punting on action against our neighbors for one month. These three countries combined account for over 40% of goods imports to the United States. Tariffs this high and applied to such a broad scope of U.S. imports would have constituted a highly significant change in economic policy. Almost immediately, the Mexican and Canadian tariffs were suspended for a month. Yet, all this highlights that historically large and broad-based tariffs remain a very possible policy outcome in the coming years.

This FAQ provides information on the likely effects of these tariffs and, crucially, the effects that will not occur due to these tariffs. First, let’s define it. 

What is a tariff? 

A tariff is a tax levied on imports to the United States. By raising the cost of foreign-produced goods or services relative to U.S.-produced ones, a tariff redistributes some of the benefits of trading from U.S. consumers and foreign producers to U.S. producers of import-competing goods, allowing domestic businesses to also raise prices. In this way, a tariff acts as a tax on consumption as well as allowing import-competing businesses to raise prices without losing market share to foreign producers. Until recently, tariffs on U.S. imports averaged 2.2% in accordance with numerous international agreements negotiated by the U.S. government–though many items entered tax free.

Can tariffs ever be effectively used to target smart policymaking goals?

In brief: Yes. Tariffs can do a number of useful things. Three broad uses include: 

• providing effective protection for domestic production in specific economic sectors 
• shielding U.S. workers from unfair forms of competition from specific trading partners (like those with abusive labor rights regimes)
• complementing a country’s strong domestic climate policy when trading partners’ policies are not as strong

Because tariffs are most effective when they focus on well-defined and narrowly tailored goals, they work best as part of a larger strategy.

In detail: The most direct benefit of tariffs is protection for domestic sectors in the U.S. economy that warrant strategic support. For example, some sectors are harmed when our trading partners take actions to support their own domestic exporters or undercut labor and clean air and water standards, or are critical for economic or national security. As a recent example, U.S. steel and aluminum producers have faced chronic global oversupply that has largely been caused by subsidies (direct and indirect) that trading partner governments have given their own domestic producers—who are among the world’s worst polluters.

As part of a strategic suite of complementary industrial policies, tariffs can help sustain and support the development of key industries and maintain them during periods when trading partners are engaged in market-distorting subsidization of their exports. Tariffs can help correct these pervasive distortions and improve economic efficiency, allowing firms to thrive in the face of these distortions.

Other reasons for wanting to target more domestic production from specific sectors include national security concerns, the underinvestment of private actors in the resilience of key nodes of supply chains, and combating monopolization of key inputs by another country—a lesson learned painfully during the COVID-19 pandemic when everyone was scrambling to source personal protective equipment (PPE), respirators, and critical medicines unavailable domestically at the necessary scale. Tariffs can help internalize the social costs of fragile global production chains otherwise created by profit-maximizing corporations. In short, tariffs are a valid, and often useful, industrial policy tool that can provide narrow and targeted protection for key sectors. 

Tariffs can also be used to shield U.S. workers from low-road practices (like labor abuses) among trading partners. For example, if tariffs were higher for countries that routinely failed to protect workers’ fundamental rights (or if tariffs were lowered when a country made a genuine commitment to protect these rights), the benefits of pursuing international competitiveness through wage suppression would be reduced.

Similarly, tariffs could also be useful in complementing high-road competition in environmental standards, for example by embedding the costs of greenhouse gas emissions (GHGs) from manufacturing and transportation in low-standard countries. This would incentivize clean air while also making sure U.S. workers in trade-exposed, energy-intensive industries do not bear the extra burden of adjusting to new climate policies. An approach that explicitly used tariffs to internalize the social costs of labor and environmental exploitation in low-standard countries would help correct these problems and provide transparent incentives for countries to pursue pro-worker and pro-environment policies.

These uses are not trivial: Tariffs are absolutely a key tool of smart industrial and trade policy. But on their own, tariffs cannot and should not be the centerpiece of a national economic strategy. Doing so would represent a gross overuse of a tool for a task it’s not suited for and would cause damage to the wider economy.

Can high and broad-based tariffs fix the U.S. trade deficit or rebuild manufacturing employment?

In brief: No, mostly because high and broad-based tariffs will also reduce exports along with imports, and this will leave the balance of trade mostly unchanged. Exports fall when tariffs are introduced for a number of reasons. The first is that many U.S. exports use imports as intermediate inputs to final goods produced in the United States. Making these inputs more expensive with tariffs will boost the price of these U.S. exports and make them less competitive in global markets. Second, trading partners are highly likely to retaliate to U.S. tariffs with tariffs of their own, making exports more expensive in international markets—which we’ve seen on “Made in America” goods from Boeing airplanes to Kentucky bourbon. And finally, tariffs will put upward pressure on the value of the U.S. dollar in global markets, which will make our exports more expensive and will increase the attractiveness of imports to U.S. customers—primary causes of U.S. trade deficits and manufacturing job losses.

In detail: It is true that unbalanced trade has suppressed employment in manufacturing in the United States for decades. We consistently import far more manufactured goods than we export (and the difference is nowhere near made up in the services trade). This trade deficit drives a wedge between domestic consumption of manufactured goods and domestic production. Closing this trade deficit would, hence, substantially boost job opportunities in the manufacturing sector in the U.S. However, large and broad-based tariffs on all manufactured imports will not do much to close this deficit for several reasons.

First, many U.S. exports are produced using a large share of imported inputs. The slicing of global value chains in recent decades means that parts of a final good are often sourced from several different countries. Tariffs would, hence, make these inputs more expensive, and this would, in turn, push up the price of U.S. exports using these inputs, weakening the competitiveness of U.S. exports in global markets.

Second, and most obviously, tariffs are rarely unidirectional. When we impose tariffs, our trading partners are likely to retaliate with reciprocal tariffs on U.S. goods, pricing U.S. exporters out of international markets. This is not speculative—it absolutely was the result of the tariffs imposed during the first Trump administration.

American farmers and ranchers incurred the most widespread damage from this retaliation following the 2018 tariffs. The damage was so great that the Trump administration authorized $61 billion in emergency relief payments to cushion farmers and ranchers from the blow of this retaliation, an amount roughly equivalent to all of the tariff revenue collected from U.S. businesses. Big manufacturers like Boeing also lost access to international markets. Prior to 2018, China accounted for 25% of Boeing’s sales, but after the tariffs, China stopped ordering Boeing aircraft and created an opening for China’s homegrown COMAC C919—a direct competitor to Boeing’s 737 series planes. Not only will U.S. exporters lose markets abroad, but the lost exports will increase the supply of their goods to U.S. markets, putting downward pressure on the price of goods they sell domestically, reducing corporate profits.

Third, large and broad-based tariffs would put upward pressure on the value of the U.S. dollar, making U.S. exports more expensive to foreign buyers and imports cheaper and more attractive to U.S. businesses and consumers. This often happens as trading partners intentionally push down the value of their own currency against the dollar through exchange rate management policies to offset the competitive ground lost in U.S. markets to the new tariffs.

But it will also happen essentially mechanically. Countries use the dollars they earn from importing to the United States to purchase exports from the U.S. If tariffs reduce the dollars countries earn from importing to the U.S., this will either lead them to reduce what they purchase as U.S. exports, or they will need to purchase dollars on international capital markets in order to maintain their level of U.S. export purchases. This increased demand for dollars in these capital markets will push up demand for dollars, and the exchange rate will increase.

Again, this is not speculative—it is exactly what happened in 2018 in response to Trump’s first-term tariffs on Chinese technology goods and broader steel and aluminum imports, when China depreciated its currency by roughly 10% against the dollar. Against an international basket of currencies, the dollar rose by about 7.5%. Already since the November 2024 election, the value of China’s currency has fallen 1.1% against the dollar.

As a result of these influences, the U.S. trade deficit—how much we export minus how much we import—saw no improvement through the first Trump administration even as tariffs were increased. The tariffs did work to change the composition of the trade deficit as Chinese exporters sought to circumvent U.S. tariffs on Chinese goods by rerouting trade and expanding investment in third countries—in particular, taking advantage of the U.S.-Mexico-Canada trade agreement negotiated by President Trump to use Mexico as a platform to export to U.S. markets. Since the 2018 tariffs took effect, imports from Mexico have increased 63%, and the U.S. trade deficit with Mexico increased by 159%.

Reducing damaging trade deficits cannot be achieved solely through trade policy—except in the extreme case where trade policy measures are so severe that they essentially shut down all international trade, which would cause radical disruption to the U.S. economy. Instead, more balanced trade will only result from macroeconomic policies that are consistent with lower trade deficits—including exchange rate management to realign an overvalued U.S. dollar and a reasonable mix of fiscal and monetary policies.

Are tariffs the same as an industrial policy for the United States?

In brief: No, tariffs are only one tool in the industrial policy toolkit, and they need supporting policies in strategic endeavors to effectively boost domestic sectors.

In detail: Tariffs, on their own, are an incomplete industrial policy strategy, even for the narrow goal of supporting a strategic domestic sector. New research confirms the efficacy and pervasiveness of industrial policies when applied strategically. While these policies can take a variety of forms, all successful industrial policies do three main things, mostly aimed at addressing key market failures:

• promote positive economic spillovers that provide economic benefits beyond the targeted industry—such as by creating innovation that benefits other industries or by supplying complementary goods or services that make other investments viable—and limit or abate negative economic spillovers that impose costs on other industries, consumers, or the public. For example, industrial policy to intentionally spread out the production of key inputs like semiconductors so that bottlenecks specific to a single country don’t choke global supply chains in the future creates the positive externality of resilience—left to their own private profit-making devices, individual companies will not have the incentive to make these investments.
• provide complementary and industry-specific public inputs. Key examples include infrastructure, research and development, and workforce development investments that complement and crowd-in private investment.
• provide coordination of disparate actors where complementary and collective actions are needed for an industry’s success, but market mechanisms are incapable of playing this role. One example of this is publicly provided monitoring of potential supply-chain stresses to keep private actors informed of how they can plan deliveries and marshal inputs to solve blockages before they happen. Industrial policy can also provide market-creating guarantees to crowd investment into cutting-edge technologies that individual investors might consider too risky to support, such as the COVID-19 vaccine development. 

Tariffs can be part of this formulation when there is a compelling public interest to support a particular industry. But they are insufficient on their own to ensure that industries critical to U.S. economic and national security—from primary metals, to critical medicines and health equipment, to semiconductors and other advanced technologies—can overcome market failures and unfair competition.

Tariffs change the price signals in markets from which investors decide to shift resources between different sectors. But price signals alone are often not sufficient to ensure key market failures are overcome. But there are many market failures besides getting prices wrong that domestic capital owners and workers need to overcome in order to be willing to invest in producing in tradeable goods sectors.

At the technology frontier, by definition, no one knows the likelihood of success in achieving technological advances or what the market potential is for such innovation—although technological progress is highly desirable, the potential risks and rewards cannot be accurately priced. Another example is where complementary investments are needed to make an individual investment financially viable, such as the need to upgrade electrical grids and build charging infrastructure for investments in electric vehicle manufacturing to be viable. Further, because it’s easy to change tariffs on short notice, capital owners and workers are unlikely to see tariffs alone as a sufficient signal that they should make costly, long-term investments in the production of tradeable goods.

Who ‘pays for’ tariffs imposed on U.S. imports?

In brief: American households will bear most of the burden of higher tariffs. This will mostly come through higher prices for imported goods and, crucially, higher prices for domestic goods that compete with imports.

In detail: Tariffs are a tax on imported foreign goods and services. The legal incidence of these taxes falls on the U.S. company doing the importing. If a company imports $100 worth of goods and tariffs are 20%, the company must pay a tax of $20 to the federal government. However, one of the useful insights of economics is that the legal incidence of a tax and the economic incidence are different. Taxes set off a cascade of adjustments that can spread or concentrate their ultimate economic burden. In the case of tariffs, these adjustments essentially lead to U.S. households paying higher prices. Importers who pay the tax initially will typically raise prices to pass this additional cost along to consumers, known as “price pass-through.” The precise degree of pass-through will differ by good and sector: It is driven largely by factors such as the degree of a company’s market power and consumer sensitivity to price changes. But substantial research convincingly demonstrates that it is U.S. households who ultimately pay for tariffs.

It is important to note that if tariffs do not raise prices in the U.S. market, they will not shift consumer preferences to domestic goods from foreign goods, thereby failing to provide any useful protection to domestic industries. And if they are not providing effective protection to domestic producers, it is hard to see the point of imposing them. Tariffs provide effective protection to domestic producers by raising U.S. prices of foreign goods and services relative to similar domestically produced goods and services. This enables companies producing import-competing goods in the United States to raise prices, too, without fear of losing market share to lower-priced foreign competition. These higher prices enable domestic firms to maintain or expand production at a viable scale.

Because the tariff on a competing foreign good does not change a U.S. company’s production costs, in the short-run, the higher prices U.S. consumers pay for import-competing goods go directly into higher profits for the company. In the longer-run, and with complementary supporting policies, some of those profits might be redirected to investment and wages as the import-competing sector looks to expand its output.

What’s more, for goods that the United States does not or cannot produce domestically at adequate levels to meet demand, tariffs raise prices for U.S. consumers and businesses without giving a boost to domestic industries. This includes a wide range of agricultural goods (e.g., coffee, avocados, and bananas) and commodities and minerals that are relatively scarce in U.S. territory. Tariffs on imports that do not compete with “Made in America” goods simply raise prices for U.S. consumers without spurring domestic production of these goods. They represent a pure cost to U.S. consumers without any countervailing benefit.

Should policymakers try to make tariffs a significant revenue source for government spending?

In brief: No. Tariffs are essentially a tax on consumption and are, hence, more regressive than most current federal revenue sources. This means that with tariffs, people with lower incomes will pay a larger share of their earnings in taxes than high-income people. For the significant amount of revenue we need to raise in the coming years, we should build on the existing progressive revenue sources we have (income and estate taxes) and institute new progressive taxes.

In detail: President Trump has suggested that tax revenues from new tariffs could replace the federal income tax. This would require tariffs to reach historically high and broad levels and would constitute a large, regressive shift in who finances the federal government. In this scenario, the tax burden would shift from higher-income households to low- and moderate-income households. Large, across-the-board tariffs of this magnitude would also have many negative economic side effects relative to income taxes.

In 2024, the federal government will collect an estimated $2.5 trillion in individual income tax revenues. To raise this much revenue, the base of any tariff would have to be extremely broad—effectively universal, falling on all imports. If one starts with the implausible assumption that a universal tariff would not change U.S. demand for imports, the tariff rate would need to be 78% to replace the individual income tax. More realistically, if tariffs deter Americans from importing goods, then these taxes on imports would need to be significantly higher to replace income tax revenues. In fact, most estimates of how sensitive U.S. purchases of imports are to their prices indicate that tariffs literally could not replace even half of the income tax.

Further, even if tariffs could somehow replace income tax revenue one for one, this would be an extremely damaging shift for most U.S. families who would end up paying a greater share of their incomes in taxes. Revenues from progressive income taxes are preferable to those from tariffs for a few reasons.

First, tariffs are a regressive tax, meaning people with lower incomes will pay a larger share of their earnings in taxes than high-income people. Tariffs are essentially a consumption tax, and consumption as a share of income tends to fall as incomes rise.

Second, progressive taxation achieves more than merely raising revenues to fund essential public services supplied by our government from those who are most able to pay—it also creates incentives that shape economic behavior in socially productive directions. The runaway growth of incomes for top earners are driven by rent-seeking practices—income gained from the exploitation of power that is unrelated to an individual’s contribution to overall economic growth. Progressive taxation disincentivizes such rent-seeking among those with power and instead allows incomes to be more broadly dispersed. This disincentive effect of the federal income tax has been greatly eroded since the 1980s as top marginal rates have fallen, but relative to a scenario with no income tax, it is significant and worth preserving.

Finally, tariffs lead to efficiency losses as the potential benefits of international specialization are lost. At tariff levels that have persisted for the past 70 years, these efficiency losses are quite small (and often very exaggerated by economic commentary). But at tariff levels needed to replace the federal income tax, these efficiency losses would be high. Progressive income taxes also have some distortionary effects that might reduce economic output, but they also have countervailing influences that might boost economic output and welfare. For example, by raising revenue from the rich and financing federal spending targeted toward low- and middle-income families, this progressive redistribution supports growth in economywide demand. Again, richer households save higher shares of their income, so, redistribution away from them boosts spending. Substituting tariffs for progressive taxation forgoes these economic benefits beyond tax revenues.

Are tariffs easier and more transparent to collect than other forms of taxes?

In brief: No, tariffs involve multiple compliance costs, and across-the-board tariffs will offer many more chances for corrupt dealing than exist under current taxes.

In detail: In 2018, tariffs imposed under Sec. 301 and Sec. 232 authorities by the first Trump administration included a process whereby importers could petition for tariff exclusion. Essentially this provided a huge new tax loophole for politically connected large companies to exploit. In total, the Trump administration granted more than 100,000 exclusions from the tariffs. Audits by the U.S. Government Accountability Office and the Department of Commerce’s Inspector General found the exclusion petition processes were plagued by a lack of transparency, followed capricious and inconsistent internal procedures, and issued contradictory and seemingly arbitrary decisions.

New empirical research indeed confirms that tariff exclusions have been used systematically to reward political contributions, as well as to punish political opponents, rather than to accommodate economic need. In one instance in 2019, Apple lobbied President Trump to secure exemptions for iPhone imports from China and pledged to repatriate some Mac computer manufacturing from China to the United States, though they never delivered on this promise. The scope of companies that will be incentivized to apply for exclusions (and potentially offer improper favors in exchange for them) and the financial benefit of avoiding tariff charges will grow enormously if the Trump tariffs that were promised during the campaign actually come to pass.

Will the impact of tariffs on consumers vary a lot across U.S. states?

In brief: Not a lot—the effect of tariffs on consumers’ purchasing power will mostly depend on the share of consumer spending on categories of goods and services facing new tariffs. This doesn’t differ widely by state—consumers in California, for example, aren’t necessarily more likely to spend vastly more money on durable goods or food (products facing high new tariffs) than consumers in Montana. Essentially, residents of all states are in the same boat when it comes to these price increases.

In detail: Tariffs are a tax on imports that, in effect, redistribute “surplus” from U.S. consumers and foreign producers to subsidize domestic producers of import-competing goods. When used strategically to support specific sectors, tariffs can be an effective tool of industrial policy—but consumers will face lower purchasing power as a result.

The effects of these price increases from tariffs will be mostly uniform across states. Below, we provide a rough estimate of the potential impact on U.S. consumers’ purchasing power in each state from President Trump’s new tariffs on Canada, Mexico, and China implemented on March 4, 2025. These tariffs apply a 25% tax on goods imported from Canada and Mexico, except for Canadian energy products that will be taxed at 10%, and raise tariffs on goods imported from China from 10% to 20%. On March 6, President Trump suspended these tariffs on many Mexican and Canadian goods for one month, but given that the administration has now twice announced and backed down from these tariff threats, we thought calculating their potential impact if fully implemented would be a useful addition to the policy debate. The final estimates are shown in the map below.

To estimate the impact on consumers, we analyzed personal consumption expenditures in each state, decomposed into durable goods, non-durable goods, and services consumption from the Bureau of Economic Analysis, using the most recent available data from 2023. We applied research from Federal Reserve Bank of San Francisco economists that identifies the import content of consumer spending in each of these consumption categories and then estimate the tariff-impacted component of imported consumption.

The main variation in how one state’s consumers will face higher or lower costs from import tariffs hence depends on whether their consumption skews more or less toward those spending categories that are intensive in imports (i.e., goods). It is likely the case that states with high housing costs spend less (as a share) on goods and will hence face lower tariff costs, simply driven by the fact that housing is not import-intensive, and if it’s expensive it is by definition claiming a bigger share of personal spending in those states.

Finally, to present a dollar value of tariff costs per capita, we adjusted the prices of 2023 consumption to January 2025 prices—the most recent measurement available from the Personal Consumption Expenditures price index.

There are some estimates of import destination by state, but in the end, we did not use this information to allocate tariff effects. In this import destination data, Canada, Mexico, and China account for 43% of all U.S. imports, though there is considerable variation across states—from a mere 13% of total imports in Hawaii to 94% of all imports in Montana. However, many imports are intermediate inputs used as components in the production of more finished goods—just because something is imported to a state does not mean it will be consumed in that state. For example, 75% of Michigan’s imports are from these three countries, but much of this represents parts that will be assembled into cars and trucks sold all across the country. It’s not Michigan consumers alone who would face any particularly large cost increase from tariffs.

Finally, it’s important to note that this analysis only allocates the effect of new U.S. tariffs across states. The potential effect of retaliation from trading partners might be much more focused on particular states. These three countries are the top export destination for 45 states, but retaliation might not stop at restricting U.S. exports. For example, a possible response by Canada could be to raise electricity prices to the U.S. states supplied by Canadian generation. This is a small subset of states that would face much greater costs than others. Other responses could include reduced tourism, as Canadian Prime Minister Trudeau suggested, which would also be more concentrated in particular states.

Are other countries’ value-added taxes (VATs) an unfair barrier to U.S. exports?

In brief: No. Value-added taxes as they are administered in trading partners of the United States are neutral with respect to trade. They are not a barrier to U.S. exports and hence should not be penalized under “reciprocal” trade protection.

In detail: A value-added tax is a consumption tax on goods and services levied at each stage of the supply chain. Unlike sales taxes in many U.S. states and localities which tax the consumption of a good or service at its final user, a VAT levies taxes incrementally on each stage of production. President Trump’s February 13, 2025, Memorandum on Reciprocal Trade and Tariffs falsely claims that value-added taxes are “unfair, discriminatory, or extraterritorial taxes imposed by our trading partners.” President Trump has suggested treating VATs as a kind of tariff to be met with reciprocal tariffs from the United States. This is a mistake commonly made by people who don’t really understand how these taxes or trade work.

Economists and tax practitioners have long recognized that in theory VATs do not inherently affect international trade flows, as summarized by Nobel Prize-winner Paul Krugman and chief economic advisor to President Reagan Martin Feldstein. In practice, when VATs fall more heavily on tradeable goods (like manufactured goods) than on non-tradables (like housing and health care), they will reduce the size of a country’s tradeable goods sector, leading to reductions in both imports and exports. Further, if a VAT (which is a consumption tax) substitutes one for one for an income tax, this can increase national savings and hence provide a small boost to net exports. But these real-world twists on VATs still in no way constitute an unfair barrier to U.S. exports.

According to the International Monetary Fund, more than 160 countries use a VAT; the United States is somewhat of an outlier in relying on a single-stage retail sales tax to raise revenues from consumption. To illustrate how a VAT taxes each stage of production, imagine that a farmer sells wheat to a grain mill, which makes it into flour, which is in turn purchased by a bakery to produce bread for final sale to consumers. Under a VAT:

1. The grain mill pays a tax on the value of the wheat.
2. The bakery pays a tax on the value of the flour minus the value of the wheat (which has already been taxed).
3. Consumers pay a tax on the value of the bread minus the values of the wheat and the flour (which have both already been taxed).

The value taxed for consumers of bread under a VAT in the third step already embody the taxes paid for the values of intermediate steps—1) wheat and 2) flour—in the supply chain. In contrast to a VAT, a retail sales tax like in U.S. states levies a tax on the full retail price of bread because, as intermediate inputs to production, the wheat and flour are exempted from tax.

It is true that VATs are levied on goods and services imported to the VAT country and rebated on products that the VAT country exports. But this treatment is exactly what keeps VATs neutral with respect to trade flows. In effect, this works like existing U.S. sales taxes: U.S. exporters do not pay a U.S. sales tax on goods that are sold outside the U.S., but U.S. consumers do pay a sales tax on imported goods. Similarly, when a good is sold across state lines into a state with no sales tax, businesses pay no sales tax in the “exporting” state. And consumers in a sales tax state pay taxes for “imported” goods from another state. Levying a VAT on imports and rebating VATs on exports merely serves to give all products the equivalent tax treatment.  

In other words, with a VAT, there simply is no penalty for imports nor a subsidy for exports. Hence, other countries’ VATS do not constitute trade barriers that demand reciprocal protection from the United States.

Dear Commissioners:

Thank you for the opportunity to comment on the upcoming 2025 Report on the USMCA Automotive Rules of Origin. Although these rules are only really beginning to come into effect, it is already clear that the incremental tightening—along with the labor Rapid Response Mechanism—are insufficient to overcome the fatal flaws in the NAFTA model that allowed multinational producers to threaten and actually relocate work to lower-wage and more readily exploitable places like Mexico. i Thirty years after the start of NAFTA, this corporate strategy is still suppressing wages for motor vehicle and parts workers and their communities in all 3 countries. USMCA has done little to change this equation and the updated Rules of Origin (ROOs) seem unlikely to change the course dwindling job quality and quantity in the U.S. motor vehicles and parts industries.

ROOs are too leaky to ensure high quality jobs

Even before the 2023 “rolling up” decision, USMCA’s content rules were too porous and the labor value standard too low to achieve the competitive North American automotive industry with high quality jobs that we all want. The rules allow non-originating content to be magically transformed into originating content and creep into North American supply chains with duty-free market access, even when that content is produced with illegal subsidization and exploited workers in third countries. Leakage from USMCA’s ROOs undercuts all North American workers by pitting them against non-USMCA producers who have not made the same commitments to worker, environmental, and consumer safety standards, and have not extended reciprocal market access to similar North American producers.

The porous rules have a number of implications: First, the more complicated an intermediate auto part is – i.e., the more underlying, lower-tier components are required to make an intermediate part – the more foreign content can masquerade as “Made in North America.” Even under USMCA RVC calculations, substantial non-originating content will enter at 0% duty. Thus, non-originating content qualifying for USMCA ROOs can expand at an exponential rate.

Second, the core parts list omits critical new technology goods for electric vehicle related components and autonomous vehicle related components. Fortunately, USMCA Article 3.10 provides a mechanism for the parties to expand the list of core components covered by North American content rules, though this is not a foregone conclusion and, in the meantime, content leakages deter the establishment of North American competition to illegally-subsidized and market-dominant Chinese technologies. The faster the industry evolves, the more vehicle content will be uncovered by USMCA ROOs.

Third, for most places in the United States $16/hour is at or near a poverty wage, on par with fast food and other low-skilled service wages. But because President Trump did not negotiate an inflation adjustment, already the Labor Value Content wage has declined 7% in real terms since he signed USMCA into law. And this wage floor—which will continue eroding into complete irrelevance before Mexican wages converge on U.S. and Canadian wages—covers less than half of a vehicle’s content, meaning it is not a meaningfully binding wage standard.

Finally, not only did USMCA leave these gaping holes for non-originating content, but the agreement largely relies on these same corporate entities to self-certify their compliance with the ROO regime. This creates a clear incentive and opportunity for manufacturers to cheat compliance. The less government monitors and enforces these rules, the more incentive manufactures have to evade USMCA ROOs.

In order to improve upon these shortcomings, USTR in conjunction with the Departments of Labor, Treasury, and Commerce should maintain a real-time database of facilities certified and in compliance with all USMCA ROOs for parts content, labor value, and primary metals content. Congress should establish the capacity to investigate whether duty-free treatment for a vehicle was provided due to errors or omissions in the producer’s certifications or preference claims, and to claw back duties with penalties when found in violation. What’s more, U.S. agencies should be required affirmatively to notify the Secretary of the Treasury of any malfeasance or whistleblower violation in response to reporting fraud, in the ROO calculations, so that duty-free treatment could be revoked.

The impact of loose ROOs

Rather than working to retain production in the United States with incentives for high-quality work, the trade data already show a widening U.S. bilateral automotive trade deficit with Mexico since USMCA’s inception.

Figure 1 shows that U.S. deficit in vehicles traded with Mexico is climbing under USMCA. For light duty vehicles—passenger cars and light pickup trucks—the U.S. deficit improved moderately after 2019 with the onset of the pandemic, automotive supply-chain shortages, and shifts in demand toward higher-value vehicles—more of which are produced in the United States. As a result, the light duty vehicle deficit was cut nearly in half by 2022. In 2023, with a strengthened U.S. economic recovery, even into the headwinds of U.S. monetary policy tightening, the light duty vehicle deficit began climbing once again. For medium- and heavy-duty vehicles, Figure 1 shows that prior to USMCA the U.S. maintained a small-but-steady trade deficit with Mexico. But that all changed with USMCA: the trade deficit jumped to more than 1 million trucks by 2023 from just 123,000 trucks in 2020.

In U.S. dollar terms, the bilateral U.S. trade deficit with Mexico in light-duty vehicles reached a pre-pandemic business cycle peak of nearly $56 billion in 2019 (Figure 2). As the U.S. economy has recovered, the light-duty trade deficit is climbing once again, to $40 billion 2023. The deficit in medium- and heavy-duty trucks, which averaged less than $9 billion annually in the pre-pandemic business cycle expansion jumped to $38 billion in 2022 and $43 billion in 2023 (Figure 3).

A similar pattern can be seen in U.S.-Mexico bilateral trade in motor vehicle parts. Parts imports from Mexico showed a steadily increasing trend over the pre-pandemic business cycle, from a $7 billion deficit in 2009 to a $27 billion deficit by 2019 (Figure 4). These data cover trade in tariff codes for motor vehicle parts specified by the ITA. ii However, since the onset of USMCA, U.S. vehicle producers have gone on a shopping spree for parts from Mexico. In just three years, U.S. imports shot up to $78 billion annually from $51 billion and the parts deficit ballooned to $42 billion from $26 billion.

Widening trade deficits in vehicles and parts with Mexico reflect motor vehicle producers’ preference to sight more of the North American industry in low-wage manufacturing centers, despite tighter USMCA ROOs. But they also reflect a realignment of global automotive supply chains following 2018 U.S. Secs. 301 and 232 tariffs and a slate of antidumping and countervailing duty determinations as production and trade was rapidly reorganized to third countries. A key example of this phenomenon can be seen in the case of tire imports to the U.S. (Figure 5). In anticipation of and following ITC determinations of injury from Certain Passenger Vehicle and Light Truck Tires from China in 2015, Chinese tire manufacturers quickly expanded offshore in countries like Thailand and Vietnam, among others, substituting for Chinese production for U.S. markets.iii As a result, tire imports from these countries surged and led to subsequent ITC import injury investigations from these countries.iv

Increasing Chinese penetration of North American motor vehicle supply chains

USMCA is not equipped to handle the coming influx of automotive content originating with Chinese-owned and Chinese-affiliated firms—often the beneficiaries of robust, illegal government subsidization and regulatory forbearance. In the aftermath of U.S. tariffs and trade enforcement measures, Chinese-owned and Chinese-affiliated firms have been repositioning their global footprint to launder the origins of their production chains through countries with more favorable tariff treatment in U.S. and North American markets. Following trends in other industries including steel and aluminum products, production chains oriented around Chinese-owned and -affiliated firms are rapidly penetrating North American automotive supply chains with growing manufacturing platforms in Mexico and a range of third countries with surging auto parts exports to the U.S.

China’s ballooning outward direct investment position since 2018 indicates a rapidly expanding overseas footprint for production organized around Chinese value chains. Since 2018, Chinese outward FDI positions increased by 126% in Mexico, 40% in Thailand, 246% in India, 119% in Vietnam, 97% in Malaysia, 73% in Indonesia, and 12% in the Philippines—in total a $42 billion expansion.v At the same time, exports of Chinese-made manufacturing machinery and equipment to these emerging automotive production hubs has surged since 2017: 134% to Mexico, 92% to Thailand, 79% to India, 149% to Vietnam, 159% to Malaysia, 119% to the Philippines, and exponentially to Indonesia.

For Mexico, depending on the kind of machinery, this has meant an 8-fold to 41-fold increase of manufacturing equipment imports from China (Figure 6). Among this wave of manufacturing investment are significant projects to establish platforms in Mexico for Chinese motor vehicle and parts producers aiming to penetrate the USMCA market.vi Mexican imports of Chinese core parts—chassis fitted with engines and bodies and cabs—have increased 132-times and 670-times, respectively, over the previous business cycle.

These examples illustrate the tip of the iceberg for the potential of state-supported Chinese content to penetrate North American markets. While imports of complete vehicles from China is still limited, it won’t be long until Americans are importing Chinese vehicles made in Mexico. Subsidization rates, for example seen in BYD’s Dolphin sold at widely ranging prices in different markets, are substantial enough to dwarf the gap between USMCA duty rates and MFN rates.vii

Benchmarking ROOs impact on corporate operations

Finally, I want to offer the commission some context on ROOs for motor vehicle manufacturing operations. We should not be surprised when companies blame regulatory compliance with USMCA, among other factors, for their efforts to shift more North American manufacturing to low-wage production centers. But we also should recognize these claims ring hollow. Motor vehicle manufacturers earn more than enough to offer workers dignified pay while still enjoying healthy profits and complying with regulatory requirements. Despite last year’s UAW strike and record new contract, Big 3 automakers still made $33bn in profits and spent $14bn on share buybacks and dividend payments. This year, the Big 3 have already spent more than $7bn on share buybacks.

The reality of more moderate cost pressures on manufacturers can be seen in data covering the recent U.S. inflationary episode. Producer prices for light duty vehicles peaked at 5% annualized inflation and for heavy duty vehicles at 3.2% in 2023 (Table 1). Producer prices for imported semiconductors, which faced critical scarcity in the pandemic, reached an inflationary peak of 5.6% annually. Meanwhile, though, consumer prices for new vehicles reached 13.2% annualized inflation. These data points indicate that it was corporate profit-push rather than wage- or cost-pushes driving inflation. In other words, the behavior should be attributed to corporate greed rather than economic necessity.

Thank you.

Adam S. Hersh, Ph.D.
Senior Economist, Economic Policy Institute

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Table 1

Notes

i. In 2017, EPI economists outlined 6 points for NAFTA reforms: Robert E. Scott, Josh Bivens, and Samantha Sanders, Renegotiating NAFTA: What Should the Priorities Be, Economic Policy Institute Policy, December 7, 2017.

ii. International Trade Administration (ITA). 2024. Harmonized Tariff System Codes and Schedule B Codes for Trade of Motor Vehicles; Harmonized Tariff System codes, Schedule B codes, and North American Industry Classification Schedule codes for automotive parts. Accessed September 9, 2024.

iii. See for example, “Thailand-made Double Coin tires arrive in U.S.” Fleet Owner. March 30, 2018; “Linglong opens Thailand plant.” Rubber News. March 6, 2014; “Sentury Tire in Thailand: establishing a world class tire brand.” China Daily. January 21, 2016.

iv. Passenger Vehicle and Light Truck Tires from Korea, Taiwan, Thailand, and Vietnam; Invs 701-TA-647 and 731-TA-1517-1520 (Preliminary)

v. See Table 1, https://www.epi.org/publication/us-mexico-canada-agreement/

vi. “BYD weighs 3 states for electric vehicle plant.” Mexico News Daily. August 26, 2024; Machuca, Elizabeth (2024). “China’s BYD Plans Expansion Into Mexico, Rules Out U.S.” Wards Auto. May 6, 2024; Enright, Merritt (2024). “How China became the leading car supplier to Mexico and what it means for the U.S.” August 23, 2024; Mahoney, Noi (2024). “China-based automaker to invest $3B in Mexico EV plant.” April 3, 2023; “MG Motor to build manufacturing plant, R&D center in Mexico.” Reuters. August 8, 2024.

vii. Hersh, Adam. 2024. “EPI comments to the Office of the United States Trade Representative on the US-Mexico-Canada Agreement with respect to automotive goods.” January 17, 2024.

For now, we’ll set aside the question of whether supporting the administration’s concrete actions but disliking their rhetoric merits this level of blistering criticism. Instead, we’ll point out two things. First, Summers’s description of the aims of these industrial policy bills (and hence the “doctrine” of Bidenomics) is obviously inaccurate. Second, to the degree that the administration really has made an intellectual break with the past (including past Democratic administrations), it’s a welcome and necessary break. This is especially true regarding Summers’s claim that the pre-Trump approach to trade policy is a model that should be restored.

Summers describes the Biden administration doctrine as “manufacturing-centric economic nationalism.” Almost every bit of that seems overblown. The main doctrinal provocation that Summers is responding to is an April speech by National Security Advisor Jake Sullivan. In this 4,800-word speech, Sullivan uses the word “manufacturing” four times and “nationalism” or “nationalist” zero times. Summers implicitly claims that the Biden administration has made restoring historic levels of employment in U.S. manufacturing a goal of their industrial policy efforts (he’s not alone in pushing this claim—it was repeated by esteemed economics writer Martin Wolf in a number of venues recently). But that’s just not true—the Biden administration has made no such promise or goal. The trio of bills passed in 2021 and 2022 are actually narrowly tailored, matching up the tools of industrial policy tightly to the economic challenges they’re targeting (the renewable energy transition and supply-chain resiliency, with particular focus on semi-conductor supply chains). In short, the doctrine seems simply to be using public policy to address market failures that won’t be fixed by private markets.

We should be clear that the U.S. could certainly benefit from an intentional “manufacturing policy” that tries to address some of the broader-based challenges to domestic manufacturing production. Key among these challenges are exchange rate misalignments (including those driven by the rapid rise of the U.S. as an oil exporter) and poor macroeconomic policies in trading partners. But, the three industrial policy bills do not constitute an overall manufacturing strategy and they’re not meant to—they have other hugely important virtues.

Among his short list of substantive criticisms, Summers objects to attaching strong labor standards to public subsidies. In his view, even mentioning co-benefits of clean energy investments like job creation or job-quality betrays unseriousness in addressing the core targets of these industrial policy interventions. This obviously does not follow. Nobody looks back on the U.S. defense buildup in the late 1930s and early 1940s and thinks that it got derailed by extraneous labor standards. Instead, this buildup is universally seen as wildly successful and as one of the keys to winning World War II. And yet, the labor standards attached to defense contracts in that time were hugely ambitious and led to the creation of the U.S. middle class. Labor peace agreements as a condition for defense contracts led to large jumps in unionization, and non-discrimination mandates contributed to a large reduction in Black–white wage differentials economywide. Summers argues that any industrial policy effort should be done with as light a touch as possible to ensure that it does not disturb wider economic outcomes. But if these wider economic outcomes are going poorly, why not implement industrial policy in a way to nudge them in better directions?

Summers’s largest criticism is that the Biden administration does not concede that the pre-Trump trade policy regime was a series of successes that we should seek to return to. Summers is right that there is a real intellectual shift here, but it’s a welcome one. Summers argues that the laser focus of trade (and, incidentally, anti-trust) policy should be on costs, full stop. The Biden administration’s welcome pivot in trade policy to also centering workers and their wages is summed up well by United States Trade Representative Katherine Tai:

“Our new approach to trade recognizes people as more than just consumers, but also producers—the workers, wage-earners, providers, and community members that comprise a vibrant middle class. Our focus has shifted from liberalization and the pursuit of efficiency and low costs—at any cost—to raising standards, building resiliency, driving sustainability, and fostering more inclusive prosperity at home and abroad.”

Recognizing that households are comprised of both consumers and workers is a cornerstone to making more progressive policy across a number of areas. For example, the same forces that lead to international trade flows reducing import prices are also those that lead to depressed wages for non-college labor in the United States. Celebrating only the cost declines while ignoring the wage suppression is intellectual malpractice and leads to policies that damage far too many.

Summers tries to support his argument for pre-Trump trade policy by claiming that removing the Trump-era tariffs would have measurably reduced inflation over the past two years and that this, in turn, would have boosted real wage growth. Honestly, I find this claim wildly overstated. The Trump tariffs were introduced in 2018, and there was no measurable inflationary effect. The idea that removing them would have led to large inflationary reductions makes little sense—and the magnitudes of the tariffs just don’t add up to large potential price declines. Finally, and this is too inside baseball for most, but tariff removal changes relative prices, not the absolute price level. A key claim of inflation hawks in recent years is exactly that focusing on relative prices is irrelevant to the larger debate over inflation. I personally think that’s wrong and that relative prices do matter a lot, but it’s a clear (and pretty opportunistic) contradiction for somebody like Summers to make. This confusion between what might change relative prices and what might meaningfully reduce inflation also applies to his implicit claim that anti-trust policy that focused only on consumer prices would reduce inflation, and that somehow the broader focus of the Biden administration’s anti-trust policy was making inflation worse.

Does all of this mean we think the Biden administration has hit every note perfect, in both deed and word? Of course not. Given the extreme political constraints they’re facing, it would be shocking if they did. For one, we’d certainly like to hear more from the administration about what they see as the unfinished business of “Bidenomics.” The public investments are welcome and extremely well-targeted, but huge problems remain in the economy—like high levels of inequality—that aren’t well-targeted by industrial policy and need other policy levers to fix. The administration has been admirable in talking a lot about some of these policy levers—like targeting full employment—but could talk more about labor standards like fundamental labor law reform and the need for movement on the federal minimum wage. And, contra Summers, there is plenty of room to expand even the labor standards directly associated with the investments made by the trio of industrial policy bills.

For another, this section of Sullivan’s speech didn’t crystallize what was wrong with pre-Trump trade policy in a way that felt right to me:

“The main international economic project of the 1990s was reducing tariffs… Simply put: In today’s world, trade policy needs to be about more than tariff reduction, and trade policy needs to be fully integrated into our economic strategy, at home and abroad.”

The main problem with pre-Trump trade policy was not that it focused too much on tariff reduction. Ninety percent or more of the text of trade agreements from NAFTA onwards dealt with non-tariff issues and sought extremely deep integration of signatory countries. The problem was that the agreements were captured by corporate interests and were exercises in selective protectionism that exposed some actors (mostly manufacturing workers) to fierce global competition but protected others (mostly corporate exporters and monopolists). The “depth” of integration is not the problem—it’s on whose behalf this integration works. Much of what the Biden administration has done on these issues seems to grasp this more fundamental critique, and I hope it shapes their approach going forward.