Introduction

Tariffs have become an increasingly important part of United States foreign economic policy, particularly following the protectionist trade policies introduced during Donald Trump’s first administration and expanded substantially in the last few years. A tariff is a tax imposed by a government on imported goods, which increases the domestic price of foreign products relative to domestically produced alternative goods. Governments may use tariffs to protect domestic industries from foreign competition, encourage domestic production, battle unemployment, raise government revenue or pursue strategic and national-security objectives. However, tariffs also affect the allocation of resources within an economy and can increase the costs faced by consumers and businesses. As a result, whether tariffs improve economic welfare depends on whether the benefits generated for protected industries and their workers outweigh the costs imposed on the wider economy.

The debate is particularly relevant in the United States because the country has historically benefitted substantially from international trade and highly integrated global supply chains. The United States imports large quantities of both final consumer goods and raw materials used by domestic firms. Therefore, tariffs can affect the economy through several channels simultaneously. While protection from foreign competition may allow American firms producing tariffed goods to increase output, raise prices and potentially employ more workers, American businesses that rely on imported inputs may face higher production costs, leading to cost push inflation. These costs can ulteriorly be passed on to consumers through higher prices. In addition, trading partners may retaliate against US exports, reducing demand for American products and potentially harming employment in export-oriented industries.

The economic reasoning for tariffs can be illustrated through the concept of protectionism. In a market without trade restrictions, domestic consumers can purchase goods from both domestic and foreign producers, creating competition that tends to keep prices relatively low. The introduction of a tariff raises the price of imported goods, allowing domestic producers to become more competitive. Consequently, domestic production may increase as consumers substitute away from imports towards American-made products. This effect is supported by evidence from the United States International Trade Commission (USITC). Examining the Section 232 tariffs imposed on steel and aluminum beginning in 2018, the USITC found that steel imports covered by the tariffs declined by 24%, while US steel production increased by 1.9%. Similarly, aluminum imports declined by 31%, while domestic aluminum production increased by 3.6% (USITC, 2023). These findings suggest that tariffs can achieve one of their central policy objectives: increasing domestic production in protected industries.

The central hypothesis of this research is that the costs of US tariffs to American consumers and businesses outweigh the benefits they provide to protected domestic industries and their workers, resulting in a small negative net effect on overall US economic welfare. This hypothesis is based on evidence that tariffs substantially increase domestic import prices, raise input costs for businesses, provoke retaliatory trade measures and generate only slight increases in protected-sector production and employment.

1. Historical Background

Tariffs have played a shifting role in US economic policy for over two centuries. In the 19th and early 20th centuries, tariffs were primarily a revenue tool and a means of protecting infant industries, culminating in a series of increasingly protectionist measures. The most infamous episode is the Smoot-Hawley Tariff Act of 1930, which raised duties on thousands of imported goods and is widely blamed by economists for deepening the Great Depression by triggering retaliatory tariffs abroad and collapsing global trade volumes.

After World War II, US policy shifted decisively toward trade liberalisation, driven by the belief that open markets promote both prosperity and peace. This produced the General Agreement on Tariffs and Trade (GATT, 1947), later the World Trade Organisation (WTO) (1995) and regional agreements such as NAFTA (1994). Average US tariff rates fell from double digits to roughly 1–3% by the early 2000s.

This consensus began to fracture after China’s WTO accession in 2001, when the United States was exposed to a large, rapid surge of Chinese imports. Research on the resulting “China Shock” (discussed further below) documented significant, geographically concentrated job losses in manufacturing communities, feeding political momentum for renewed protectionism.

Tariffs returned to the centre of US policy during 2018–2019, when the Trump administration-imposed tariffs of 10–25% on hundreds of billions of dollars of Chinese imports under Section 301 of the Trade Act of 1974, and on steel and aluminum globally under Section 232 of the Trade Expansion Act of 1962, citing unfair trade practices and national security concerns, respectively (Richmond Fed, 2025). This was followed by a much larger escalation in 2025, with a baseline 10% tariff on nearly all trading partners and steeper “reciprocal” rates on specific countries, pushing the average effective US tariff rate to roughly 18–27% — the highest level since the early-to-mid 20th century. Even after accounting for consumer substitution toward cheaper goods, the effective rate remained the highest since 1933 (Yale Budget Lab, 2025). Part of this 2025 tariff regime was later struck down: in February 2026, the Supreme Court ruled that the president could not impose tariffs under the International Emergency Economic Powers Act (IEEPA), though other tariff authorities under Sections 232 and 301 remained in effect (Tax Foundation, 2026).

2. Literature Review

The academic and policy literature on tariffs splits into two broad camps.

A. Studies emphasising costs to consumers and businesses.

Amiti, Redding and Weinstein (2019) found near-complete pass-through of the 2018 tariffs into domestic prices, with the full incidence falling on US consumers and importers rather than foreign exporters — a real income loss of approximately $1.4 billion per month by late 2018. Fajgelbaum, Goldberg, Kennedy and Khandelwal (2020) reached similar conclusions about the domestic burden of the 2018 tariffs in their study published in the Quarterly Journal of Economics. A Richmond Fed economic brief (2025) similarly found that the 2018–19 China tariffs significantly disrupted global supply chains, raised input costs for American businesses, raised consumer prices and contributed to a decline in manufacturing employment; firms largely shifted supply chains to other countries such as Mexico and Vietnam rather than reshoring production to the United States, so the expected employment boost was modest. The brief concludes that while these tariffs benefitted a narrow set of protected industries, they produced a net negative outcome for the broader economy.

More recent modelling of the 2025 tariffs reaches similar conclusions. The Yale Budget Lab (2025) found that the 2025 tariffs raised the average effective tariff rate by roughly 20 percentage points, increasing consumer prices by about 2.3% in the short run — equivalent to an average household loss of $3,800, with disproportionate effects on lower-income households and on categories such as apparel. The Penn Wharton Budget Model (2025) separately projected that the April 2025 tariffs would reduce long-run US GDP by about 6% and wages by 5%, with losses roughly twice as large as an equivalent corporate tax increase. Real-time price-tracking research by Cavallo, Llamas and Vazquez similarly found that tariffs produced both immediate and gradual retail price increases, with imported goods’ prices rising roughly twice as much as domestic goods’ prices relative to pre-tariff levels.

B. Studies and arguments emphasising benefits to workers and domestic industry.

The “China Shock” literature — Autor, Dorn and Hanson (2016, 2021), building on their earlier work (2012) — documents large, concentrated and long-lasting costs to specific American manufacturing communities from prior trade liberalisation. This body of research is frequently cited by tariff proponents as evidence that unmanaged trade exposure, rather than tariffs themselves, was the real source of harm to American workers. Advocates of an industrial-policy approach, such as the think tank American Compass, argue that tariffs are a legitimate tool to rebuild domestic manufacturing capacity, protect strategic industries from state-subsidised competition and reduce dependence on geopolitical rivals for critical goods. In a public debate with economist Jason Furman, American Compass economist Oren Cass argued that US tariff use has historical precedent under Lincoln, Theodore Roosevelt and Reagan — who used tariff threats against Japanese auto imports in the 1980s to induce Japanese manufacturers to build plants in the United States — and contended that China’s state-directed economy warrants a more confrontational trade posture than past disputes with market democracies. Proponents also point to tariff revenue as a fiscal benefit: federal customs duty collections rose from $79 billion in 2024 to $264 billion in 2025 after the new tariff rounds (Tax Foundation, 2026).

3. Comparative Case Studies

3.1 Section 232 of the Trade Expansion Act of 1962: Steel Tariff

One case of tariffs was the US 2018 steel tariffs. In 2018, Donald Trump’s administration put tariffs on steel and aluminum imports for crucial industries like the manufacturing and construction industry in America. He put a 25% tariff on steel imports and 10% tariff on aluminum imports. His administration stated his goals were to protect domestic steel and aluminum production and create more jobs.   While his goals were briefly accomplished as both production and jobs rose in these industries, in the long term, these tariffs also had negative effects. Higher import costs led to material shortages and soaring prices. Imports fell substantially for steel and aluminum in the American consumers and businesses for most of the cost for these tariffs. An estimated $125 billion was redirected from this shift. This also created problems for American companies that had already invested billions in manufacturing facilities in China and other cheaper production countries. Moreover, other countries retaliated with their own tariffs on steel and aluminum. While an estimated 35,400 jobs were created, there was also an estimated dead weight welfare loss of $232,000 per job saved due to changes in how companies invested, which is a significant amount of economic value lost from the American economy.

3.2 Section 201 of the Trade Act of 1974: The Washing Machine Tariffs

Another case of tariffs imposed by the Trump administration in 2018 was the washing machine tariffs. Trump’s administration imposed a “safeguard” on imported washing machines and dryers that ranged from 20-50% depending on how many machines were imported, leading to a surge of an estimated 12% in washing machines prices. This tariff had some positive benefits: for instance, many washing machine producers moved to America to produce their washing machines and an additional 1,800 jobs were created. However, even American-made washing machines became more expensive, and while jobs were created, American consumers ended up holding the weight of an additional $1.5 billion in total cost. Thus, while the tariffs did succeed in encouraging domestic production of washing machines, the price hike for the American consumer outweighed the benefits of this tariff.

3.3 China-US Trade War

A counterexample to the washing machine tariff is the 2018 China-US Trade War, initiated by the Trump administration. This trade war began with the steel tariff previously mentioned. As a result, China faced reduced access to the American economy and Chinese imports fell drastically. The Chinese did not necessarily lower their prices in response, which deemed their imports less profitable than American domestically produced goods. Overall, this created more jobs and gave the US economy more leverage in international economics.

Taken together, these case studies can help American economists predict what could happen in the future with the imposition of new tariffs. While products may become more expensive in America, it is difficult to tell whether future tariffs will overall benefit the American economy or hurt it.

4. Discussion

The results from the previous sections point to one main imbalance. Tariffs are defended to protect American workers and rebuild domestic industry, but the evidence shows that the benefits go to a small group of producers while the costs are spread across consumers, manufacturers who buy imported inputs and exporters. Studies find that: tariffs are passed through into domestic prices almost completely (Amiti, Redding & Weinstein, 2019; Fajgelbaum et al., 2020; USITC, 2023); manufacturing employment falls overall once input costs and retaliation are included (Flaaen & Pierce, 2024; Lake & Liu, 2025); the region’s tariffs that were meant to help show no measurable job gains (Autor et al., 2024); and that aggregate models predict lower output in the long run (Yale Budget Lab, 2026; Tax Foundation, 2026). Together, these findings undermine the claim that protection benefits American workers on net. The point is not that the gains are fake – they are real, and they matter a lot locally – but the problem is that they are smaller than the losses, and much easier to see.

4.1 DISTRIBUTIONAL ASYMMETRY AND THE POLITICAL ECONOMY OF PROTECTION

The clearest way to explain this is through the idea of concentrated benefits and dispersed costs (Olson, 1965). A tariff on steel helps a small group that is clustered in a few places and well organised politically. The cost falls on households as a few hundred dollars a year, which almost nobody can trace back to a specific policy. The Yale Budget Lab (2026) estimates that the 2026 tariffs raise the US price level by 0.5% to 1.1%, which works out at roughly $760 to $1,500 per household per year. That is a large total burden, but it is invisible to each individual household. This helps explain why tariffs remain popular even though economists agree strongly against them (Irwin, 2017).

There is also a regressive side to this, which sits awkwardly with the pro-worker argument. Tariffs work like a tax on goods, and poorer households spend a larger share of their income on things like clothing, food and electronics, so the cost falls hardest on exactly the households that protectionist rhetoric says it is defending. Judging tariffs as a worker-protection policy therefore means weighing what workers gain as producers against what they lose as consumers. The Stolper-Samuelson model (Stolper & Samuelson, 1941) shows why this tension exists, but it does not settle it, because in reality workers cannot easily move between industries, and skills are often specific to one sector.

4.2 THE INPUT-COST CHANNEL

The most important finding in our analysis concerns intermediate goods. About half of what the US imports is not finished products but inputs and machinery. This means a tariff on one commodity is also a tax on every domestic firm that uses that commodity. Russ and Cox (2018) estimate that around 80 times more people work in industries that use steel than in producing it, so protecting steel raises costs for a far larger group than it helps. Flaaen and Pierce (2024) split the 2018-19 tariffs into three effects and find that a small gain from protection (+0.4%) was outweighed by higher input costs (−2.0%) and foreign retaliation (−1.1%), leaving manufacturing employment about 1.4% lower overall.

This is not a one-off result. Lake and Liu (2025) show that the 2002 Bush steel tariffs caused lasting job losses in areas that used a lot of steel, and that these losses continued long after the tariffs themselves ended 18 months later. The tariffs also failed to stop the long-term decline in steel employment. Two separate episodes, 16 years apart and very different in size, produced the same direction of result, suggesting a structural mechanism rather than bad luck. The 2025-26 data fits the same pattern: employment in primary metals rose, while total US manufacturing employment fell by nearly 70,000 (Cato Institute, 2026).

4.3 STRATEGIC AND SECURITY JUSTIFICATIONS

None of this means that no tariff can ever be justified. The strongest arguments for protection are the ones that do not depend on welfare gains at all. If a supply chain runs through a rival state or a single geographic chokepoint, then losing some efficiency can be a sensible insurance payment rather than a mistake. The Section 232 measures on semiconductors and pharmaceuticals, and the January 2026 US-Taiwan semiconductor agreement, make more sense on these grounds than as job policies. Tariffs can also work as bargaining leverage against state subsidies and overcapacity, which is the logic behind the Section 301 investigations opened in March 2026.

The key point is that these are different arguments, and they should be tested differently. A security argument should be judged by how likely a supply disruption is and how bad it would be, not by employment figures. Mixing the two allows a policy that fails one test to be rescued by appealing to the other.

4.4 POLICY UNCERTAINTY AS AN UNDER-COUNTED COST

One cost that standard models mostly leave out is instability itself. The Supreme Court ruled on 20th February 2026 that the IEEPA did not authorise tariffs, a Section 122 surcharge was introduced immediately afterwards, it expired on 24th July 2026 and over $150 billion in collected duties was refunded. The result is a situation where the landed cost of an import can change within days because of a court decision. Reshoring, however, requires firms to commit large amounts of capital for many years with no easy way to reverse the decision. This volatility works directly against the aim tariffs are supposed to serve. Protection becomes self-defeating: the more it relies on legal powers that can be challenged, the less able it is to produce the investment it is meant to encourage.

4.5 POLICY IMPLICATIONS

Three things follow. First, if the aim is distributional — keeping jobs in communities with few alternatives — then the real comparison is not tariffs against free trade, but tariffs against other tools. Wage subsidies, direct investment in industry and adjustment assistance can pursue the same goal without taxing firms downstream, and the evidence suggests they do so at lower cost. Second, how a tariff is designed matters more than how high it is. Narrow, temporary security-based measures combined with domestic investment are a different policy from broad surcharges on almost all imports, and they should not be judged as if they were the same thing. Third, where tariffs are kept for strategic reasons, that should be stated openly, because defending them with employment claims the evidence does not support ends up damaging the credibility of the security argument too.

4.6 LIMITATIONS OF THE PRESENT ANALYSIS

Two limitations should be noted. Much of our assessment depends on modelled estimates from the Yale Budget Lab, the Penn Wharton Budget Model and the Tax Foundation, and these results are sensitive to assumptions about monetary policy and how easily consumers substitute between goods. In addition, most of the causal evidence comes from the 2018-19 trade war, which was smaller and more targeted than the 2025-26 regime. Whether those estimates carry over to a near-universal tariff structure is still an open question that only later data can answer.

Overall, these findings indicate that the benefits tariffs give American workers and domestic industries do not outweigh the costs they impose on American consumers and businesses, at least when measured in conventional welfare terms. That conclusion does depend on treating aggregate efficiency as the deciding criterion. 

In conclusion, the evidence reviewed here indicates that the economic case for broad-based tariffs is considerably weaker than the political appeal of protectionism would suggest. While narrowly applicable measures may enhance the fortunes of protected industries, the benefits of protection must be weighed against their impact on consumers, downstream producers and export-competitors at home and abroad. The empirical results from the 2018-19 trade war suggest that the winners tended to be those industries directly protected by tariffs, whereas the costs were diffusely borne by the manufacturing sector as a whole; similar conclusions can be drawn from more recent estimates of the macroeconomic impact of protectionism. Meanwhile, the assumption that tariffs are always an inefficient policy instrument ignores the possibility that strategic trade policies may serve purposes beyond traditional economic efficiency.

The key insight to be drawn from the body of research reviewed here is that the economic case for protection through tariffs depends critically on the objective function to be optimised. If the goal is to raise wages and reduce unemployment in communities vulnerable to import competition, then policies such as direct subsidies or investment in key industries may substitute for protectionist measures without the accompanying distortion to the price mechanism. If the goal is national security, then the analysis should explicitly consider the implications of tariffs for the reliability and resilience of supply chains. The failure to distinguish between these objectives leads to the kind of conflation that dooms protectionist policies to be ineffective as either economic or security instruments.

The essential point, then, is that the benefits of protection through tariffs are rarely sufficient to outweigh the costs in terms of national economic welfare. This does not necessarily preclude the use of tariffs as a tool of trade policy, but it does suggest that when such tariffs are imposed, they are best justified as targeted and temporally limited measures. The justification for protection should be based on a clear-eyed assessment of the costs and benefits in relation to specific objectives, whether these relate to labour markets, industrial structure or national security. The analysis reviewed here indicates that the latter considerations can rarely justify protectionist measures that extend beyond a narrow set of industries and input sectors.

Conclusion

This paper asked whether the benefits tariffs deliver to protect American industries and their workers outweigh the costs they impose on the wider economy. The evidence assembled here suggests they do not, at least when judged by conventional welfare criteria.

Four examples were examined: the 2002 Bush steel tariffs, the 2018 Section 232 metals tariffs, the 2018 washing machine safeguard and the near-universal tariff regime of 2025–26. They differ in scale, in legal basis and in the industries they covered, yet they produce the same result. Employment rises in the protected sector, and falls by more across the manufacturing economy as a whole. Flaaen and Pierce (2024) show why: a protective gain of 0.4% is outweighed by a 2.0% loss through higher input costs and a further 1.1% through retaliation. Because roughly half of US imports are intermediate goods, a tariff on a commodity is simultaneously a tax on every domestic firm that uses it, and Russ and Cox (2018) estimate that around 80 times more Americans work in steel-using industries than in steel production. That the same sign appears across the examples spanning twenty-four years points to a structural mechanism rather than to circumstance.

The political durability of tariffs is explained not by their economic performance but by their distributional shape. Following Olson (1965), the gains are concentrated, geographically visible and countable, while the costs arrive as a few hundred dollars per household per year that almost no one traces back to a specific policy. The burden is also regressive, falling hardest on the lower-income households that protectionist rhetoric claims to defend. The gains are real and locally significant. They are simply smaller than the losses, and far easier to see.

Three qualifications limit how far this conclusion can be pushed. First, much of the aggregate evidence rests on modelled estimates whose results are sensitive to assumptions about monetary policy and consumer substitution. Second, most causal evidence comes from the 2018-19 trade war, which was smaller and more targeted than the current regime; whether those estimates transfer to a near-universal tariff structure remains open. Third, the argument is framed in efficiency terms, and security-based justifications do not depend on welfare gains at all. Where a supply chain runs through a rival state or a single chokepoint, accepting some inefficiency may be a reasonable insurance payment.

That distinction carries the paper’s main policy implication. Economic and security arguments for protection should be tested separately because conflating them allows a policy that fails one test to be rescued by appealing to the other. Where the objective is distributional, tariffs should be compared not against free trade but against wage subsidies, industrial investment and adjustment assistance, which pursue the same end without taxing downstream producers. Where the objective is security, the measure should be narrow, durable in law and defended on its own terms. Future work will need the post-2026 data to establish whether the near-universal regime follows the pattern its four predecessors established, or breaks it.

American Compass. (n.d.). Home page. American Compass [online]. <https://americancompass.org>

Amiti, M., Redding, S. J., & Weinstein, D. (2019). The impact of the 2018 trade war on U.S. prices and welfare. National Bureau of Economic Research, Working Paper No. 25672.

Autor, D., Dorn, D., & Hanson, G. (2012). The China syndrome: Local labor market effects of import competition in the United States. National Bureau of Economic Research, Working Paper No. 18054. 

Autor, D., Dorn, D., & Hanson, G. (2016). The China shock: Learning from labor-market adjustment to large changes in trade. Annual Review of Economics, 8, pp. 205–240.

Autor, D., Dorn, D., & Hanson, G. (2021). On the persistence of the China shock. Brookings Papers on Economic Activity, Working Paper No. 29401.

Boller, L., Carmody, K., Huntley, J., Reichling, F., & Smetters, K. (2025). The economic effects of President Trump’s tariffs. Penn Wharton Budget Model, University of Pennsylvania [online]. <https://budgetmodel.wharton.upenn.edu/p/2025-04-10-the-economic-effects-of-president-trumps-tariffs/>

Cavallo, A., Llamas, F., & Vazquez, J. (2026) Tracking the short-run price impact of U.S. tariffs [pdf]. <https://www.pricinglab.org/files/TrackingTariffs_Cavallo_Llamas_Vazquez.pdf>

Fajgelbaum, P. D., Goldberg, P. K., Kennedy, P. J., & Khandelwal, A. K. (2020). The return to protectionism. Quarterly Journal of Economics, 135(1), pp. 1–55.

PBS NewsHour. (2025). Economists offer differing views on Trump’s tariffs and trade war. PBS NewsHour [online]. <https://www.pbs.org/newshour/show/economists-offer-differing-views-on-trumps-tariffs-and-trade-war>

Richmond Fed. (2025) Tariffs: Estimating the economic impact of the 2025 measures and proposals, Economic Brief No. 25-12 (Federal Reserve Bank of Richmond).

Tax Foundation (2026) Tracking the impact of the Trump tariffs & trade war. Tax Foundation [online]. <https://taxfoundation.org/research/federal-tax/trump-tariffs-trade-war/>

Tax Foundation. (2026). Tariffs increased retail prices of imports by 7 percentage points prior to Supreme Court ruling. Tax Foundation [online]. <https://taxfoundation.org/blog/trump-tariffs-raise-prices-consumers/>

Yale Budget Lab (2025) Where we stand: The fiscal, economic, and distributional effects of all U.S. tariffs enacted in 2025 through April 2. Yale Budget Lab [online]. <https://budgetlab.yale.edu/research/where-we-stand-fiscal-economic-and-distributional-effects-all-us-tariffs-enacted-2025-through-april>

AI Disclosure: This paper was written and researched by the authors. AI tools (Claude, Anthropic) were used to assist with researching supporting sources, drafting sections of the literature review and formatting citations. All content was reviewed, verified and edited by the authors, who take full responsibility for the accuracy and originality of the final work.