Key Points

• Secular decline in the share of manufacturing jobs in the labor force largely reflects the shift of consumer expenditures from goods to services.
• High tariffs cannot restore manufacturing jobs to the 27 percent share of the labor force experienced 60 years ago.  
• Achieving Trump’s objective for eliminating the trade deficit in manufactures would require tariffs at least twice as high as those imposed through September 2025. 
• The annual cost to American consumers of shifting each job from service employment to manufacturing employment through high tariffs exceeds $200,000.

Introduction

Throughout his 2024 campaign for a second White House term, Donald J. Trump characterized himself as “tariff man” and declared “tariff” the most beautiful word in the dictionary. Trump displayed “tariff man” characteristics during his first White House term (2017–21), but the full exhibition emerged only after his inauguration for a second term in January 2025. 

Trump’s enthusiasm was demonstrated by executive orders (EOs) that imposed high tariffs on various countries and products, his choice of equally committed advisers and cabinet members, and his running commentary as he issued new orders. A crescendo was reached on April 2, 2025, which Trump labeled “Liberation Day,” when minimum baseline tariffs of 10 percent were imposed on all countries and higher tariffs were imposed on countries that ran bilateral merchandise trade surpluses with the United States. 

Trump announced five overlapping reasons to explain his tariff agenda: (1) to curb US trade deficits, (2) to boost the US manufacturing sector, (3) to retaliate against unfair foreign trade practices, (4) to create bargaining leverage against foreign countries, and (5) to raise revenue. This report evaluates only the manufacturing goal while recognizing that other goals also motivated Trump’s actions. 

In November 2021, Gary Clyde Hufbauer and Eujin Jung published an evaluation of US industrial policy episodes between 1970 and 2020.¹ During that time, tariffs were used several times to promote particular industries. These episodes largely failed to create industries that could compete in world markets, and they entailed quite high consumer costs per job created or saved. Trump’s agenda is a different order of magnitude from past industrial policy episodes: The president now seeks to boost the entire manufacturing sector through the application of tariffs. 

Accordingly, we begin this report with an overview of the US manufacturing sector, then turn to US trade in manufactures, and summarize US tariffs and foreign retaliatory duties. The final section roughly estimates the hypothetical tariff rate needed to eliminate the US trade deficit in manufactured goods, the number of jobs that would thereby shift from other sectors to manufacturing activity, and the economic cost of this endeavor. We conclude that the tariff agenda previewed by Trump’s declarations, though it is ambitious and led to a stock market collapse wiping out $10 trillion of equity value,² will not likely achieve self-sufficiency in the manufacturing sector.

US Manufacturing Output, Employment, and Wages 

To set the stage, we examine aspects of the US manufacturing sector between 2016, just before Trump’s first term, and 2024, just before his second term. During these years—a period interrupted by the COVID pandemic—Republicans and progressive Democrats complained ever more loudly that the US manufacturing sector was battered by imports, notably from China and Mexico. 

Figure 1 shows US manufacturing value added (“manufacturing output”) expressed in current dollars and as a percentage of gross domestic product (GDP). We focus on manufacturing value added rather than manufacturing sales because value-added figures are closely related to employment through wages paid and because sales figures include inputs from other sectors (mainly services but also agriculture and mining) and sales within the manufacturing sector. 

Expressed in current dollars, manufacturing output steadily rose between 2016 and 2024, but expressed as a percentage of GDP, manufacturing output fell by a little more than 1 percent over this period. Contrary to popular perception, manufacturing activity did not collapse in recent years. However, looking at a longer time frame, the share of manufacturing output in GDP receded compared with a period we characterize (following popular perception) as the “golden years”—namely, from 1960 to 1968³—a period when the US enjoyed a trade surplus in manufactures of around 1 percent of GDP⁴ and manufactured value added was around 25 percent of GDP.⁵ 

The main reason the share of manufacturing output in GDP fell is that as household income rises, a larger share of expenditure goes to services such as education, health, travel, and entertainment. This is a characteristic phenomenon in all advanced countries. A secondary reason, and the grievance of populist politicians, is that the US imports a large share of its manufactured consumption. 

Figure 2 shows US manufacturing employment expressed by number of workers and as a percentage of nonfarm employment between 2016 and 2024. Even during this relatively short period, lasting less than a decade, the share of workers engaged in manufacturing declined by about 0.6 percent. The declining share of manufacturing employment is a long-term trend, in the United States and other advanced countries. In the “golden years” of 1960 to 1968, on average some 27.5 percent of American workers held manufacturing jobs. A subtext of Trump’s tariff agenda is to recapture that period. 

Comparing the decline in manufacturing value added as a share of GDP between the 1960s and 2024 (from around 25 percent to 10 percent⁶) with the larger decline in manufacturing employment as a share of nonfarm employment (from around 27 percent to 8.3 percent,⁷ as shown in Figure 2, Panel B) leads to the conclusion that manufacturing productivity has grown somewhat faster than productivity in the economy at large. This is another persistent theme in the experience of the United States and other countries. Between 2016 and 2023, output per worker in manufacturing was 1.3 percentage points higher than in the economy as a whole, compared with 2.5 percent higher between 1987 and 2000 and 3.7 percent higher between 2000 to 2010.⁸ 

Another popular perception is that manufacturing jobs pay far better than other pursuits. Figure 3 shows wages and salaries per full-time-equivalent employee in recent years expressed in current dollars for manufacturing workers and all private employees. Wages and salaries in manufacturing during 2016–23 averaged about 9 percent higher than those of all private employees. In the “golden years” between 1960 and 1968, manufacturing paid about 13.6 percent higher than all private employment.⁹ Better manufacturing pay still prevails, but not so much as yesteryear and not so much as popular perception. 

It’s questionable whether protecting manufacturing output will sharply increase the gap between manufacturing pay and the pay of all private employees. US imports of manufactured goods are concentrated among products that require less advanced technology, such as clothing, furniture, toys, steel slabs, and household electronics. Making these products at home will not require the advanced skills required to assemble aircraft, construct heavy machinery and medical equipment, or refine chemicals. Accordingly, the gap between average manufacturing pay and average pay for all private employees may not sharply increase even with high tariffs on manufactured imports. 

US Manufactures Trade 

Imported manufactures are commonly blamed for the destruction of American industry and hard times in afflicted regions, especially Appalachia and the Midwest. These woes can largely be attributed to the decline in manufacturing jobs. Some of the job decline can be attributed to technology: the faster increase in manufacturing output per worker than the overall increase in national productivity. Much more of the job decline can be attributed to the fact that, as household incomes rose, Americans shifted an ever-larger share of their expenditures to services— health, education, travel, entertainment, and more. But the impact of manufactured imports on jobs plays a central role in Trump’s tariff agenda. 

Figure 4 shows manufactured imports in recent years expressed in dollars and as a percentage of US consumption of manufactures. Consumption is measured as manufacturing value added plus half of manufactured imports and minus half of manufactured exports. The one-half coefficient, estimated by Robert Z. Lawrence,¹⁰ reflects the commonly overlooked fact that manufacturing sales embody large inputs from other sectors of the economy, particularly services but also agriculture and mining. 

Between 2016 and 2024, imports were stable at around 41 percent of consumption, indicating that imports and consumption grew at around the same rate (Figure 4). Back in the “golden years” of the 1960s, imports amounted to roughly 3 percent of manufactures consumption.¹¹ The long-term rise in the imported share of consumption fuels the political drive to bring production back to the United States. 

In the telling of tariff advocates, the leading industry “victims” of manufactured imports are the five product categories depicted in Figure 5: computer and electronic products, transportation equipment, chemicals, machinery (except electrical), and electrical equipment, appliances, and components. Among these five categories, the share of computer and electronic products imports rose steadily, while the share of transportation equipment imports fluctuated. Otherwise, share changes were small. Tariff advocates especially vilify imports of autos and parts because of associated job losses. Unlike auto workers, chemical and machinery workers are not to the same extent symbols of national manufacturing strength. 

The leading national “villains,” as perceived by Trump, are the five countries shown in Figure 6 that sell the most manufactures to the US market: Mexico, China, Canada, Germany, and Japan. In US import share terms, China has steadily declined, while Mexico has risen somewhat. Import shares for the other three “villains” are roughly constant. 

China’s decline originated with Section 301 tariffs (alleging “unfair trade”) imposed by President Trump in his first term and essentially continued by President Joe Biden. The extreme tariffs Trump imposed in his second term presage a further sharp decline in imports from China, even if bilateral negotiations reduce the rate from 145 percent to medium double digits. To some extent, US imports from China have been replaced by imports from Mexico, but replacement imports are also arriving from Vietnam and other countries in Southeast Asia. 

Even politicians who castigate manufactured imports acknowledge (in a softer voice) the contribution of manufactured exports to US prosperity. Figure 7 shows the top five export product categories. Most of the leading export categories are the same as the leading import categories, with two exceptions: Petroleum and coal products are a leading export, while electrical equipment is a leading import, but not vice versa. 

“Comparative advantage,” first explained by David Ricardo in 1817, largely makes itself felt in the granular details of broad product categories. Inspection of 10-digit Harmonized Tariff Schedule (HTS) line items shows quite different specialties between imports and exports. For example, in the computer and electronic products category, Mexico leads with some $5.8 billion in US imports of machines used to receive, convert, transmit, or regenerate voice, images, or other types of data, while the US exported around $6 billion worth of central processing units to Mexico in 2024. In the same category, the US imported some $5 billion monitors from China in 2024, while it exported $5.4 billion of electronic integrated circuits, including processors and controllers, to China.¹² 

Figure 8 shows that four of the top five country export destinations are the same as the top country import sources. The only switch is between Japan (a top import source) and the United Kingdom (a top export source). 

The close match between export destinations and import sources illustrates the power of economic gravity: Geographic proximity and national economic size largely determine the magnitude of two-way trade flows. Accordingly, US neighbors Mexico and Canada are the top US trading partners. China is third, owing to the economic size of the US and China, despite their separation by the Pacific Ocean. 

The political characterization of imports as bad and exports as good leads Trump to focus on bilateral trade deficits. Figure 9 portrays the evolution of US manufactures trade deficits in the top five product categories, while Figure 10 portrays manufactures trade deficits with the top five country partners. 

Among product categories, Trump focuses on transportation equipment, chiefly autos and parts, for high tariffs. Even though they are not major import categories, Trump also singles out steel and aluminum as products requiring national security protection under the provisions of Section 232 of the Trade Expansion Act of 1962. In his second term, Trump also added semiconductors, part of the computer and electronics import category, to his list of “national security” products needing protection. 

Partner countries responsible for large US bilateral deficits are the leading targets for Trump’s misnamed “reciprocal” tariffs announced on April 2, 2025. As Figure 10 shows, that makes China the top target. Mexico is a distant second but will probably ascend as Mexico replaces Chinese exports. Germany, South Korea, and Vietnam are also on the list, much to the chagrin of the three US allies. 

Trump’s Tariff War 

Starting in the middle of his first term (2017–21) and sharply escalating early in his second term (2025–29), Trump imposed high tariffs on select countries and products characterized by US trade deficits. Table 1 summarizes effective US tariff rates for each exporting country as announced in Trump’s April 2 “Liberation Day” EO, along with the US bilateral trade deficit in 2024. For China, the rate includes tariffs imposed during Trump’s first term (extended by Biden) and tariffs imposed in Trump’s second term.

In addition to country-specific tariffs, Trump imposed global tariffs or announced investigations on specific products, largely manufactured goods that he considers especially vital. As of May 2025, the designated products are steel, aluminum, copper, autos, semiconductors, lumber, and pharmaceuticals. Investigations were initiated for lumber and pharmaceuticals in March and April, but proposed tariffs are not yet announced. Product-specific tariffs will apparently not be stacked on the baseline universal tariff of 10 percent or country-specific tariffs.¹³

Based on 2024 trade patterns, Table 2 estimates actual tariff rates on imports from China and the rest of the world in January 2025 (at the end of Biden’s term in office), scheduled tariff rates following Trump’s April 2 “Liberation Day” EO, and hypothetical tariff rates to achieve zero trade deficits. Following the April 2 EO, the US imposed universal baseline 10 percent tariffs on imports from all countries— with the important exception of US-Mexico-Canada Agreement (USMCA)–compliant imports from Mexico and Canada—and country-specific tariffs (up to 74 percent) on countries that ran bilateral trade surpluses with the US in 2024.¹⁴ 

As Table 2 shows, following the EO, scheduled tariffs on all manufactured product categories imported from China exceeded 50 percent; that figure was raised to 145 percent by a new EO on April 9, 2025. However, on May 12, 2025, Trump paused many of the Chinese tariffs for 90 days. Scheduled tariffs on manufactured product categories imported from the rest of the world were around 10 percent. 

According to one estimate, Trump’s April 2 EO escalated the scheduled average US tariff rate on imports from all countries to 22.5 percent,¹⁵ actually higher than the Smoot–Hawley average tariff because the 1930 law continued duty-free imports of products not extensively grown or mined in the United States (e.g., cocoa, bananas, lead, and zinc). 

After Trump’s inauguration on January 20, 2025, and the succession of EOs imposing new tariffs, the stock market swooned, and consumer confidence plunged. Blockbuster “Liberation Day” tariffs sparked a global sell-off in equity markets. These repercussions prompted Trump to declare a 90-day “pause” on April 9, suspending all April 2 tariffs except tariffs on imports from China and the universal baseline 10 percent tariff. At the same time, Trump increased the tariff on all imports from China to 125 percent. This action was taken in response to China’s retaliatory 84 percent tariff on all imports from the US after the April 2 EO. The new 125 percent tariff comes on top of the April 2 EO’s 20 percent tariff on China, bringing the total rate to 145 percent. China then raised its tariff on US goods to 125 percent. De minimis imports from China and Hong Kong were also terminated effective on May 2.¹⁶ 

On May 12, the US and China published a joint statement agreeing to roll back tariffs and started another 90-day pause, reducing the general average to 30 percent (20 percent before April 2 and 10 percent amended from April 2).¹⁷ The two pauses left in place product-specific global 25 percent tariffs on automobiles, steel, and aluminum; 25 percent tariffs against Canadian and Mexican imports of non– USMCA compliant imports; and 10 percent tariffs on imports of fossil energy and potash from Canada. During the two 90-day pauses, the Trump administration sought further concessions from foreign partners, especially those running bilateral surpluses with the United States. 

The final column of Table 2, following the US trade representative (USTR) methodology, makes two strong assumptions to calculate hypothetical tariff rates needed to eliminate US trade deficits for each product category. The first assumption is that foreign countries would not retaliate against US exports. This leads to the unrealistic premise that US exports would not be affected by the trade war. But China has already retaliated with a reciprocal 125 percent tariff. The second assumption is that a 10 percent tariff reduces the volume of imports by 10 percent. This is the assumption made, in a roundabout way, by the USTR explanatory report for calculating “reciprocal tariffs” on countries running bilateral trade surpluses with the United States.¹⁸ 

Hence, if the trade deficit for a product category amounts to just 10 percent of imports, a 10 percent tariff will supposedly eliminate the deficit. Under these strong assumptions, the hypothetical tariff on electrical equipment, for example, is calculated at almost 60 percent to eliminate the US trade deficit, while the hypothetical tariff on chemicals is calculated at 31 percent to achieve the same goal. 

Table 2 shows that scheduled US tariff rates after the April 2 EO were high enough to choke off almost all US manufactured deficits with China, based on USTR assumptions. Trump’s subsequent imposition of another 125 percent China duty in his April 9 EO would finish the task if not subsequently reduced. However, scheduled tariffs on manufactured imports shipped from the rest of the world, in the range of 10 percent, are well short of the hypothetical levels needed to end those manufactured deficits (reported in Table 3). 

Early in May 2025, the news media speculated that the US and China would dial back their extreme tariffs. This was subsequently done, reducing the general tariff on US imports from China to 30 percent. 

Since July 2025, more negotiations are in progress. The latest “reciprocal” rates have been announced, including for top US trading partners Japan and Korea,¹⁹ lowering their rates to 15 percent. While Canada and Mexico are exempt from the 10 percent universal tariffs, both countries, along with China, are still subject to the International Emergency Economic Powers Act tariffs aimed at addressing illicit drugs (China: 20 percent; Canada: 35 percent; Mexico: 25 percent).²⁰ 

Foreign Retaliation 

Of course, US trade partners are not quietly accepting Trump’s tariffs. They are retaliating with their own tariffs on US merchandise exports and with other measures. Table 3 summarizes retaliation actions announced as of April 9, 2025. 

On that fateful day, President Trump declared a 90-day pause in all tariffs announced on April 2, 2025, except the baseline universal 10 percent tariff. The pause is discussed below. But because retaliation reduces US exports, the goal of eliminating US trade deficits became more elusive. Moreover, as Abba P. Lerner explained in 1936, a tariff on imports acts as a tax on exports.²¹ As well, analysis by the International Monetary Fund indicates that US tariffs may appreciate the US dollar exchange rate by about half of the average tariff, encouraging imports and discouraging exports.²² By May 2025, however, the dollar had actually depreciated against other currencies owing to chaotic tariff announcements. 

Nevertheless, the three reasons cited may imply the need for higher tariffs than the hypothetical rates calculated in Table 2 to achieve zero trade balances. Moreover, since the macroeconomic excess of US expenditure compared with US production drives the aggregate trade deficit,²³ tariffs that curb imports of selected products, or imports from targeted countries, will almost certainly increase trade deficits for other goods, services, and countries not subject to the initial tariffs. 

US Imports from Foreign Affiliates 

Production abroad by US-based multinational corporations attracts special scorn from critics of US trade deficits. “Runaway plants” are often blamed for increasing US imports from low-wage countries like China and Mexico. Critics suggest that tariffs will persuade wayward multinational corporations to reshore production in the United States. 

Without endorsing this view, Figure 11 shows total US imports during 2016–22 shipped by all foreign affiliates to their US parent firms, about $375 billion in 2022. Imports of affiliate manufactures generally increased during this period, accounting for around 90 percent of total imports (including agricultural and mineral products) by parent firms from their foreign affiliates. 

Table 4 shows imports of manufactures shipped by majority-owned foreign affiliates (not all foreign affiliates) to their US parent firms in 2022. Notably, Canada, Mexico, and the EU together accounted for almost 67 percent of such imports to US parent firms, while China’s share was quite low at under 4 percent. 

The Cost of Slashing the US Trade Deficit in Manufactures 

Based on presidential statements and actions, it seems fair to conclude that the roughly $250 billion of US imports from majority-owned foreign affiliates and the roughly $300 billion of US manufactured trade deficits with China are prime targets for Trump’s deficit reduction campaign. But it’s an open question whether the campaign—if resumed after the 90-day tariff pause—can achieve anything close to eliminating the US global deficit in manufactures trade. 

Table 5 shows the hypothetical tariff on all manufactured imports from all countries needed to eliminate the trade deficit in manufactured goods. That calculation draws on six strong assumptions: 

• No foreign retaliation. 

• No “Lerner effect” on US manufactured exports. 

• No change in the dollar exchange rate. 

• No change in the US macroeconomic balance between expenditure and production. 

• The tariff “pass-through” coefficient is 0.5, and import demand elasticity is 2.0. 

• Hence, a 10 percent tariff on all manufactured imports reduces import volume by 10 percent. 

The USTR uses the last two assumptions to calculate “reciprocal” tariffs on countries that run a bilateral trade surplus with the US.²⁴ 

Table 5 indicates that, under these six assumptions, a 42.5 percent tariff would be needed to eliminate the manufactures trade deficit. This compares with a trade weighted average of the April 2 EO product category tariffs, reported in Table 2, of 17.8 percent.²⁵ Likewise, before the April 9 pause and the 145 percent China tariff, the Budget Lab at Yale calculated that Trump’s scheduled tariffs averaged 22.5 percent, about half the deficit-elimination rate.²⁶ 

If the US trade deficit in manufactured goods disappears through the replacement of imports by domestic production, part of the labor force would shift from other activities to manufacturing. Table 6 calculates the labor equivalent of the manufacturing trade balance during 2016–23 as explained in the table note. Between 2016 and 2023, the manufacturing trade deficit of roughly 4 percent of GDP annually translated into about 1.7 percent of full-time-equivalent employees. In 2023, 1.7 percent of full-time-equivalent employees equated to about 2.45 million workers.²⁷ 

Table 7 offers a conservative calculation for the cost of eliminating the US trade deficit in manufactures output through high tariffs. The calculation starts with the requisite tariff of 42.5 percent, conservatively estimated based on the assumptions underlying Table 5. Following the USTR report,²⁸ the calculation then assumes that foreign suppliers of US manufactured imports absorb half the tariff, while US purchasers (households and business) pay the other half. This assumes a tariff pass-through coefficient of just 0.5, much smaller than that found in econometric estimates.²⁹ Accordingly, the “landed price”—that is, the tariff-inclusive price of imports paid by US purchasers—increases by 21.3 percent. This is a crucial assumption, since doubling the pass-through coefficient to 1.0 (as estimated by independent scholars) doubles the total cost of a given tariff to US households and business firms and doubles the cost per job-year.

To eliminate the trade deficit, $1,211 billion worth of manufactured imports are hypothetically foreclosed by the tariff. The surviving $1,638 billion are now 21.3 percent more expensive to US purchasers— meaning an additional cost of $348 billion (21.3 percent times $1,638 billion). 

The next step is to calculate the higher cost of domestic production that replaces the $1,211 billion in imports hypothetically foreclosed by the 42.5 percent tariff. We assume that this production is forthcoming at a price increase of 21.3 percent, the same as the increase in the landed cost of imports. Hence, the higher cost of the replacement domestic production is $257 billion (21.3 percent of $1,211 billion). 

Together, the higher landed cost of surviving imports plus the higher replacement cost of domestic production comes to $606 billion. Because this calculation assumes a small pass-through coefficient ( just 0.5) and a constant level of US manufactured exports, the calculated costs are more modest than alternative calculations might suggest. Nevertheless, the cost per manufacturing job-year created is quite high for the rest of the economy. Replacement production of $1,211 billion of manufactured imports would entail about $606 billion of new manufacturing value added (half the value of imports). The annual value added per full-time-equivalent worker is about $225,085.³⁰ Hence, $606 billion of new manufacturing value would create 2.69 million jobs. 

The total cost to US purchasers is the sum of the higher cost of surviving imports ($348.2 billion) plus the higher cost of replacement domestic production ($257.4 billion), or $605.6 billion annually. Each job created entails an annual cost of about $225,000 for an indefinite period ($606 billion cost divided by 2.69 million). Unless the tariff rate drops, the cost per job-year created will continue year after year. The cost is paid through higher prices of manufactured goods purchased by US households and business firms. 

The foregoing calculation assumes a pass-through coefficient of 0.5, following the USTR report,³¹ meaning that a 42.5 percent tariff increases the landed cost of imports by only 21.3 percent. Independent scholars have estimated pass-through coefficients closer to 1.0. At that pass-through coefficient, the cost per job-year doubles to around $550,000 for an indefinite period, and the total cost to the economy doubles to around $1,211 billion annually. Hence, the calculation in Table 7 is quite conservative. 

Trump’s tariffs, while extreme by US historical experience, are well short of levels that would eliminate the trade deficit in manufactures. Accordingly, Table 8 offers cost calculations for the “reciprocal” tariffs announced in the April 2 EO and augmented with respect to China by the April 9 EO (raising the tariff on imports from China to 145 percent). For this calculation, we ignore the 90-day pause in reciprocal tariffs announced in the April 9 EO and the temporary exemption of a large array of electronic imports from China and elsewhere, notified by Customs and Border Protection on April 11, 2025.³² If the pause and exemption are extended for the duration of Trump’s White House term, the Table 8 calculations overstate the cost of Trump’s tariff agenda. 

Table 8 calculations start by calculating the impact of 145 percent tariffs on imports from China. We assume this extreme tariff all but eliminates US manufactured imports from China, some $425 billion in 2024. However, we optimistically assume that the same volume of imports arrives from other countries, such as Malaysia, Mexico, South Korea, and Vietnam. We assume that China-replacement imports, along with customary imports from the rest of the world, all pay the average tariff of 22.5 percent, following the April 2 EO, as estimated by the Yale report.³³ 

Total US manufactured imports in 2024 were $2,849 billion. Based on the USTR report assumptions,³⁴ the average 22.5 percent tariff will reduce the volume of imports by 22.5 percent, or $641 billion. Surviving manufactured imports will amount to $2,208 billion. The landed cost of surviving imports (again by USTR report assumptions) will increase by 11.3 percent, or $248 billion (11.3 percent times $2,208 billion).

We assume that replacement production for the foreclosed $641 billion imports can be achieved at the same increase in landed cost, or just 11.3 percent. That entails a cost of $72 billion (11.3 percent of $641 billion). 

Together, the higher landed cost of surviving imports plus the higher replacement cost of domestic production comes to $321 billion annually. Replacement production of $641 billion manufactured imports would entail about $321 billion of new manufacturing value added (half the value of manufactured imports). In turn, the additional manufacturing value added of $321 billion entails roughly 1.42 million jobs.³⁵ Each job created entails an annual cost of $225,000 for an indefinite period ($321 billion divided by 1.42 million). The cost is paid through higher prices of manufactures purchased by US households and business firms. The cost per job created in the manufacturing sector is the same as in the trade deficit–elimination scenario (Table 7), but since the reduction of imports is not so drastic, the annual total cost to US purchasers of manufactured goods is not so great—$321 billion each year instead of $605 billion. 

However, this calculation critically depends on the assumed 0.5 pass-through coefficient. If the coefficient is actually close to 1.0, as estimated by independent scholars, the total cost would be $642 billion annually, and the cost per job-year would be $550,000. 

Conclusion 

Four points summarize our findings. 

First, there is no going back to the “golden years” of the 1960s when manufacturing jobs accounted for 27 percent of the labor force. Even extreme tariffs that eliminated the US trade deficit in manufactures would increase manufacturing jobs to only about 10 percent of the labor force. 

Second, the tariffs declared by President Trump on April 2 and 9, 2025, even if resumed in full force after the two 90-day pauses and “temporary” electronic exemptions, will at most increase manufacturing jobs to a little over 9 percent of the labor force. 

Third, the cost to American purchasers of manufactured goods per manufacturing job created by tariff protection will be high, at least $225,000 annually per job-year for an indefinite period. More realistic calculations, with a tariff pass-through coefficient of 1.0, indicate a cost of $550,000 annually per job-year. 

Fourth, because of the high cost of tariff protection relative to jobs created, two alternative approaches would make far more sense: (1) provide federal financing for training programs to create the high-skilled workforce that manufacturing firms demand and (2) provide direct federal assistance to depressed communities.³⁶ 

Notes

Gary Clyde Hufbauer and Eujin Jung, Scoring 50 Years of US Industrial Policy, 1970–2020, Peterson Institute for International
Economics, November 2021, https://www.piie.com/sites/default/files/documents/piieb21-5.pdf.

Nuño Rodrigo, “The Tariff Bomb Wipes Out $10 Trillion in Stock Market Value, Half of the EU’s Entire GDP,” El Pais, April 8,
2025, https://english.elpais.com/economy-and-business/2025-04-08/the-tariff-bomb-wipes-out-10-trillion-in-stock-market-valuehalf-of-the-eus-entire-gdp.html.

Jack L. Hervey, “Changing US Trade Patterns,” Economic Perspectives 14, no. 2 (1990): 2–12, https://www.chicagofed.org/publications/economic-perspectives/1990/03marapr1990-part1-hervey.

Brian Reinbold and Yi Wen, “Understanding the Roots of the US Trade Deficit,” Regional Economist, October 9, 2018, https://www.stlouisfed.org/publications/regional-economist/third-quarter-2018/understanding-roots-trade-deficit.

US Department of Commerce, Bureau of Economic Analysis, “Value Added by Industry as a Percentage of Gross Domestic Product,” November 16, 2021, https://apps.bea.gov/iTable/?reqid=147&step=2.

US Department of Commerce, Bureau of Economic Analysis, “Value Added by Industry as a Percentage of Gross Domestic Product.”

Robert Z. Lawrence, “Recent Manufacturing Employment Growth: The Exception That Proves the Rule,” Working Paper No. 24151 (National Bureau of Economic Research, December 2017), https://www.nber.org/system/files/working_papers/w24151/w24151.pdf.

Robert Z. Lawrence, Is the United States Undergoing a Manufacturing Renaissance That Will Boost the Middle Class?, Peterson Institute for International Economics, October 2024, https://www.piie.com/sites/default/files/2024-10/pb24-12.pdf.

HathiTrust, “The National Income and Product Accounts of the United States, 1929–74: Statistical Tables,” December 22, 2023, https://babel.hathitrust.org/cgi/pt?id=umn.319510030494767&seq=227.

Robert Z. Lawrence, Behind the Curve: Can Manufacturing Still Provide Inclusive Growth? (Peterson Institute for International Economics, 2024), chap. 4.

Manufactured trade data for the 1960s are not available on the US International Trade Convention DataWeb. The calculations are rough estimates, using manufactured exports and imports as a percentage of merchandise exports and imports from the World Bank for 1962–69. Data for 1960 and 1961 are unavailable.

Exports and imports value are under the following HTS-10 categories: 8517620090, 8528520000, 8542310045, and 8542310075.

13. The White House, “Executive Order: Addressing Certain Tariffs on Imported Articles,” April 29, 2025, https://www.whitehouse.gov/presidential-actions/2025/04/addressing-certain-tariffs-on-imported-articles/.

Chad P. Brown, “Trump’s Trade War Timeline 2.0: An Up-to-Date Guide,” Peterson Institute for International Economics, June 16, 2025, https://www.piie.com/blogs/realtime-economics/2025/trumps-trade-war-timeline-20-date-guide.

Yale University, Budget Lab, “Where We Stand: The Fiscal, Economic, and Distributional Effects of All U.S. Tariffs Enacted in 2025 Through April 2,” April 2, 2025, https://budgetlab.yale.edu/research/where-we-stand-fiscal-economic-and-distributional-effects-all-us-tariffs-enacted-2025-through-april.

The White House, “Executive Order: Amendment to Reciprocal Tariffs and Updated Duties as Applied to Low-Value Imports from the People’s Republic of China,” April 8, 2025, https://www.whitehouse.gov/presidential-actions/2025/04/amendment-to-recipricol-tariffs-and-updated-duties-as-applied-to-low-value-imports-from-the-peoples-republic-of-china/.

The White House, “Joint Statement on U.S.-China Economic and Trade Meeting in Geneva,” May 12, 2025, https://www.whitehouse.gov/briefings-statements/2025/05/joint-statement-on-u-s-china-economic-and-trade-meeting-in-geneva/.

Executive Office of the President, Office of the United States Trade Representative, “Presidential Tariff Actions,” https://ustr.gov/trade-topics/presidential-tariff-actions. The USTR report assumes an import demand elasticity of 2.0 and a pass-through of tariffs to domestic buyers of 0.5. Hence, the US demand elasticity is 1.0 (2.0 times 0.5), meaning a 10 percent tariff reduces the USquantity demanded by 10 percent. Based on this assumption, the USTR report calculates the tariff needed to eliminate a bilateral trade deficit as the amount of the deficit divided by US imports from the partner country. The USTR report assumes that the amount of the bilateral trade deficit reflects the partner country’s aggregate unfair trade practices against US exports.

White House, “Implementing the United States–Japan Agreement,” September 4, 2025, https://www.whitehouse.gov/presidential-actions/2025/09/implementing-the-united-states-japan-agreement/; and Donald J. Trump (@realDonaldTrump), “I am pleased to announce that the United States of America has agreed to a Full and Complete Trade Deal with the Republic of Korea. The Deal is that South Korea will give to the United States $350 Billion Dollars for Investments owned and controlled by the United States, and selected by myself, as President. Additionally, South Korea will purchase $100 Billion Dollars of LNG, or other Energy products and, further, South Korea has agreed to invest a large sum of money for their Investment purposes. This sum will be announced within the next two weeks when the President of South Korea, Lee Jae Myung, comes to the White House for a Bilateral Meeting. I would also like to congratulate the new President on his Electoral Success. It is also agreed that South Korea will be completely OPEN TO TRADE with the United States, and that they will accept American product including Cars and Trucks, Agriculture, etc. We have agreed to a Tariff for South Korea of 15%. America will not be charged a Tariff. I would like to thank the Trade Representatives who came forward today. It was an Honor to meet them, and talk about the Great Success of their Country!,” Truth Social, July 30, 2025, https://truthsocial.com/@realDonaldTrump/posts/114944494894008041.

White House, “Fact Sheet: President Donald J. Trump Modifies the Scope of Reciprocal Tariffs and Establishes Procedures for Implementing Trade Deals,” September 5, 2025, https://www.whitehouse.gov/fact-sheets/2025/09/fact-sheet-president-donald-j-trump-modifies-the-scope-of-reciprocal-tariffs-and-establishes-procedures-for-implementing-trade-deals/.

Abba P. Lerner, “The Symmetry Between Import and Export Taxes,” Economica 3, no. 11 (1936): 306–13, https://doi.org/10.2307/2549223.

Lukas Boer and Malte Rieth, “The Macroeconomic Consequences of Import Tariffs and Trade Policy Uncertainty,” International Monetary Fund, January 19, 2024, https://www.elibrary.imf.org/view/journals/001/2024/013/article-A001-en.xml.

Maurice Obstfeld, “Trump’s Tariffs Are Designed for Maximum Damage—to America,” Peterson Institute for International Economics, April 4, 2025, https://www.piie.com/blogs/realtime-economics/2025/trumps-tariffs-are-designed-maximum-damageamerica.

Executive Office of the President, Office of the United States Trade Representative, “Presidential Tariff Actions.” The USTR assumes that the pass-through coefficient for tariffs into higher domestic prices is 0.5 (much lower than other scholars have estimated) and that the domestic elasticity of demand for imports is 2.0. The combination of these two parameters results in the assumption that a tariff of 10 percent reduces the volume of imports by 10 percent.

The corresponding tariff rates for China and the rest of the world for each of the 21 NAICS-3 categories (including the five categories listed in Table 2) are weighted by their import shares in total manufactured imports, amounting to $2,849.3 billion in 2024. This weighted tariff rate is a rough estimate, calculated as the sum of the weighted tariffs for China and the rest of the world across all categories. Import values are based on trade flows under NAICS-3, chapters 31–33.

Yale University, Budget Lab, “Where We Stand.”

There were 144.1 million full-time-equivalent employees in the US in 2023. US Department of Commerce, Bureau of Economic Analysis, “Table 6.5D. Full-Time Equivalent Employees by Industry,” September 27, 2024, https://apps.bea.gov/iTable/?reqid=19&step=2&isuri=1&categories=survey#eyJhcHBpZCI6MTksInN0ZXBzIjpbMSwyLDNdLCJkYXRhIjpbWyJjYXRlZ29yaWVzIiwiU3VydmV5Il0sWyJOSVBBX1RhYmxlX0xpc3QiLCIxOTciXV19.

Executive Office of the President, Office of the United States Trade Representative, “Presidential Tariff Actions.”

Econometric estimates of pass-through coefficients for tariffs imposed in the first Trump term are close to 1.0. See, for example, Pablo D. Fajgelbaum et al., “The Return to Protectionism,” Quarterly Journal of Economics 135, no. 1 (2020): 1–55, https://doi.org/10.1093/qje/qjz036; Mary Amiti et al., “Who’s Paying for the US Tariffs? A Longer-Term Perspective,” American Economic Association Papers and Proceedings 110 (2020): 541–46, https://www.princeton.edu/~reddings/pubpapers/ARW-May-2020.pdf; Alberto Cavallo et al., “Tariff Passthrough at the Border and at the Store: Evidence from US Trade Policy,” Working Paper No. 26396 (National Bureau of Economic Research, October 2019), https://www.nber.org/system/files/working_papers/w26396/w26396.pdf;and Mary E. Lovely, “Key Economic Strategies for Leveling the U.S.-China Playing Field: Trade, Investment, and Technology,” testimony before the US-China Economic and Security Review Commission, May 23, 2024, https://www.uscc.gov/sites/default/files/2024-05/Mary_Lovely_Testimony.pdf.

In 2023, the manufacturing industry generated $2,840 billion value added, with 12,614,000 full-time-equivalent employees. US Department of Commerce, Bureau of Economic Analysis, GDP by Industry, https://apps.bea.gov/iTable/?reqid=150&step=2; and US Department of Commerce, Bureau of Economic Analysis, “Table 6.5D. Full-Time Equivalent Employees by Industry.” Based on these figures, value added per full-time-equivalent employee is estimated to be $225,085 per worker ($2,840 billion divided by 12,614,000). Following Lawrence, only half of the $1,211 billion reduction in the trade deficit is attributed to the manufacturing sector ($605 billion), which in turn entails 2.69 million jobs ($605 billion divided by $225,085). Lawrence, “Recent Manufacturing Employment Growth: The Exception That Proves the Rule.”

Executive Office of the President, Office of the United States Trade Representative, “Presidential Tariff Actions.”

US Customs and Border Protection, Cargo Systems Messaging Service, “Updated Guidance—Reciprocal Tariff Exclusion for Specified Products,” April 5, 2025, https://content.govdelivery.com/bulletins/gd/USDHSCBP-3db9e55?wgt_ref=USDHSCBP_WIDGET_2.

Yale University, Budget Lab, “Where We Stand.”

Executive Office of the President, Office of the United States Trade Representative, “Presidential Tariff Actions.”

Following the calculation explained in note 27.

Similar conclusions have been expressed in articles published by The Wall Street Journal and Financial Times. See Justin Lahart,
“How the U.S. Lost Its Place as the World’s Manufacturing Powerhouse,” The Wall Street Journal, April 13, 2025, https://www.wsj.com/economy/us-manufacturing-decline-service-economy-ee97a1e2; and Tej Parikh, “Nostalgia for Manufacturing Will Make the US Poorer,” Financial Times, April 13, 2025, https://www.ft.com/content/845917ed-41a5-449f-946f-70263adbaeb7.