One of the major aims of the current US tariff policy has been to reshore manufacturing back to the United States.1 Over the years, US manufacturing has lost ground to services in its contribution to the overall US economy. Real gross value added (GVA) in manufacturing has grown at just 1.5% on average per year since 2000—lower than the corresponding 2.1% average annual growth in real gross domestic product.2 Hence, manufacturing’s 9.4% share in the economy (in the second quarter of 2025) is much less than the 15.1% it held 25 years ago.3

The United States is not alone: Across key advanced economies, manufacturing’s share in the economy has declined, albeit unevenly (figure 1).4 The sector, however, still has a sizable presence in industrial powerhouses like Germany and Japan.5

For the United States itself, tariffs may, however, not be the silver bullet needed to revive manufacturing. To drive growth, a sector may need not only trade protection and investments but also a skilled workforce with the right experience. As of September 2025, there were 12.7 million people on US manufacturing payrolls—well below the 17.2 million in 2000.6 Even if a mix of policy intervention and private sector participation to train potential workers were to succeed, any increase in the workforce will likely be in high value–added, capital-intensive sectors. Low value–added, labor-intensive manufacturing may remain in emerging and developing economies due to their surplus labor and wage competitiveness relative to the United States and other advanced economies.7

High value–added manufacturing has fared better

Within US manufacturing, major durable goods sectors have consistently outpaced others in growth. Since 2000, real GVA in durables manufacturing has risen by about 2.1% on average per year—nearly three times the pace seen in nondurable goods during this period. Out of the eight broad categories of nondurable goods, more than half have seen output decline during this period, with the sharpest contractions recorded in textile mills and products of such mills (a 2.9% decline on average per year) and in apparel, leather, and allied products (a 2.2% decline on average per year).

In contrast, durables have generally fared better, although real output has contracted for 5 out of 11 broad sectors (figure 2). The biggest increase has been in computer and electronic products: real GVA for this sector is up 6.9% on average per year since 2000. Other sectors within durables to witness strong growth are primary metals; motor vehicles, bodies and trailers, and parts; and other transportation equipment (figure 2). 

Figure 2 also illustrates how high value–added and capital-intensive sectors have consistently performed better than low value–added ones such as textiles, apparel, paper products, and furniture. Primarily, this is due to two reasons: First, advanced economies like the United States have focused on high-skill, high value–added manufacturing and services.8 This trend is likely to intensify, especially given the sharp increase in investment in technology in recent years, including AI-related investment, with more commitments on the way.9 Second, global value chains have reshaped the way goods and services are produced and exchanged across the world. According to the Organization for Economic Cooperation and Development, nearly 70% of global trade now takes place through these global value chains.10 In manufacturing, most of these supply chains are structured in a way that low value–added manufacturing has shifted to emerging and developing economies with surplus labor and lower wages.11

Employment patterns have shifted in tune with the changing nature of US manufacturing

Shifting trends in manufacturing output have led to changes in employment in key labor-intensive sectors within manufacturing. Between 2000 and 2025, the fastest decline in payrolls was in apparel (a 6.8% decline on average per year), followed by textile mills (5.8% decline on average per year). That’s not surprising, given that output in these sectors has declined sharply over the years (figure 2).

Payrolls, however, have declined over the years in key durables sectors with strong GVA growth (figure 3). Payrolls in computer and electronic products manufacturing, for example, have declined by 2.3% on average per year between 2000 and 2025 despite real GVA growth during that period. The trend is similar for sectors like transportation equipment and primary metals.

What explains this dichotomy? First, this is due to high value–added manufacturing that the United States focuses on, primarily using its high-skilled labor. These sectors are usually tech-heavy and capital-intensive.12 And with strong productivity gains over the years, these sectors haven’t produced the employment gains13 that low value–added manufacturing has created in labor-surplus emerging and developing economies.14 Labor productivity in US manufacturing, as measured by real output per hour of all people employed in the sector, has risen by 42.2% since 2000, with almost all gains taking place in the period up to 2010.15

Second, manufacturing global value chains also include a host of services such as research, design, computer coding, and marketing, thereby enabling the spread of these activities across different firms.16 For example, services such as customer support and marketing may not be handled by the firm that is manufacturing and designing computers and electronic products or automobiles.

Occupational composition of the US manufacturing workforce has also changed

Changes in the manufacturing workforce are also evident from occupational employment and wage statistics from the Bureau of Labor Statistics.17 Production occupations,18 which include assemblers and fabricators, machinists, tool and die makers, and metal and plastic machine workers, continue to account for the largest share of manufacturing employment, but their role is receding. In 2024, these jobs made up 48.7% of total manufacturing employment—lower than their 51.9% share in 2003. This decline stems from automation, rising productivity due to new technology, and higher outsourcing.19

During the same period, occupations such as health care support, office and administrative support, and building and grounds cleaning and maintenance also witnessed declining employment, thereby reducing their shares of total manufacturing employment (figure 4). This is likely due to the outsourcing of certain services to other firms,20 while technological progress (like word-processing or accounting software) has reduced the need for office and administrative support personnel in the wider economy.21

The share of low-wage occupations22 in the manufacturing workforce has declined—but remains high—relative to high-wage occupations (figure 4). Within high-wage occupations, the shares of business and financial operations, computer and mathematical science, and management have risen. In the overall economy, computer and mathematical science occupations constitute more than 80% of the tech workforce, which in turn has been growing steadily over the years.23 A similar trend is visible within the manufacturing workforce as well: As high-tech manufacturing has gained ground, the share of computer and mathematical science occupations has increased.

The US manufacturing workforce may grow but gains may be limited

As commitments to set up more production units in the United States increase—aided by new trade agreements like the one with the European Union, Japan, and South Korea24— their success will depend not only on achieving cost effectiveness due to tariffs25 but also on the availability of a skilled workforce. In August 2025, there were approximately 409,000 unfilled positions in manufacturing.26 By 2033, the industry may need 3.8 million new workers, with nearly 1.9 million of those roles at risk of going unfilled if preexisting workforce challenges remain unaddressed.27 In a survey of over 500 manufacturers in early 2025 by the Reshoring Initiative, 30% of original equipment manufacturers (OEMs) responded that they would reshore production to the United States if the “workforce had higher skills and were in abundant supply.”28 The same survey shows that a skilled workforce is a bigger focus for these OEMs than policy (like tariffs and taxes), regulations, and currency movements.29

One way to attract more talent is higher wages. Indeed, wage and compensation growth in manufacturing has been roughly keeping pace with that in services-producing industries since the pandemic (figure 5).30 And manufacturers have also been focusing on improving the workforce experience to attract and retain employees.31 Nevertheless, getting additional skilled workers ready to run new production facilities will not only require interested workers but also adequate training, potentially incentivized by public policy.

It may also require skilled foreign workers to run and oversee manufacturing facilities initially while training US workers for the future.32 And since population growth has slowed in the United States over the years, a larger workforce may also need more immigrant workers to fill gaps in low- to medium-skilled roles. In fact, analysis by the Bureau of Labor Statistics suggests that, in 2024, foreign-born workers were more likely to be employed in production, transportation, and material-moving occupations compared to native-born workers.33 Recent restrictions on immigration,34 however, will likely present some challenges.

Finally, the reshoring of labor-intensive manufacturing back to the United States is expected to cost money and time, even with high tariffs in place. It may just lead to an inefficient allocation of resources, thereby raising costs for producers and consumers.35 As tariff rates stabilize, manufacturers may still prefer to keep low-value operations in emerging and developing economies while passing on part of their tariff costs to consumers.36

by

Akrur Barua

India

Endnotes

  1. The White House, “Regulating imports with a reciprocal tariff to rectify trade practices that contribute to large and persistent annual United States goods trade deficits,” April 2, 2025.

  2. US Bureau of Economic Analysis, National Income and Product Accounts, sourced using Haver Analytics in December 2025. All data on gross value added (real and nominal) are taken from this source. Unless stated otherwise, all data is sourced using Haver Analytics.

  3. Ibid.

  4. Eurostat and the Organization for Economic Cooperation and Development (OECD), National Accounts data on gross value added, sourced using Haver Analytics in December 2025.

  5. Ibid.

  6. US Bureau of Labor Statistics, Establishment Survey, sourced using Haver Analytics in December 2025. All data on payrolls (employment) in this article is taken from this source.

  7. Faezeh Raei and Anna Ignatenko, “Global value chains: What are the benefits and why do countries participate?” International Monetary Fund (IMF), Jan. 18, 2019; David Dollar, “Value chains transform manufacturing—and distort the globalization debate,” IMF, June 2019.

  8. World Bank Group, “Global value chain development report 2019: Technological innovation, supply chain trade, and workers in a globalized world (English),” accessed December 2025.

  9. US Bureau of Economic Analysis, National Income and Product Accounts, sourced using Haver Analytics in December 2025; Reuters, “From OpenAI to Google, firms channel billions into AI infrastructure as demand booms,” Nov. 18, 2025; Steve Shepley, Kate Hardin, John Morehouse, and Kruttika Dwivedi, “2026 Manufacturing Industry Outlook,” Deloitte Insights, Nov. 13, 2025. 

  10. OECD, “Global value and supply chains,” accessed December 2025; OECD, Trade in value-added, sourced using Haver Analytics in December 2025.

  11. Raei and Ignatenko, “Global value chains”; Dollar, “Value chains transform manufacturing.”

  12. John Coykendall, Kate Hardin, Adam Routh, John Morehouse, and Matt Sloane, “Accelerating the resurgence of American manufacturing,” Deloitte Insights, March 26, 2025.

  13. Gene Kindberg-Hanlon, “The technology-employment trade-off: Industry, automation, and income effects,” World Bank Blogs, Feb. 10, 2021.

  14. Dollar, “Value chains transform manufacturing.”

  15. US Bureau of Labor Statistics, Productivity and Costs, sourced using Haver Analytics in December 2025.

  16. Sébastien Miroudot and Charles Cadestin, “Services in global value chains: From inputs to value-creating,” OECD Trade Policy Papers, no. 197 (2017); OECD, Trade in value-added, sourced using Haver Analytics in December 2025.

  17. US Bureau of Labor Statistics, “Occupational employment and wage statistics,” accessed through Haver Analytics in December 2025. The OEWS categorizes occupations into 22 broad occupations and sub-occupations within each. In this article, the authors look at only the 22 broad occupations. For more on this list of broad occupations, refer to figure 3 in the article. The authors analyzed each sector in manufacturing to gather the data for that sector and aggregated the data across all sectors to arrive at the total number for occupational employment in manufacturing. Since most data for sectors within manufacturing is available from 2003 to 2024, the analysis in this section is based on that time horizon.

  18. Production occupations include a long list of sub-occupations. For more details, refer to: US Bureau of Labor Statistics, “51-0000 production occupations (major group),” accessed Dec. 16, 2025.

  19. Michael J. Handel, “Growth trends for selected occupations considered at risk from automation,” Monthly Labor Review, US Bureau of Labor Statistics, July 2022.

  20. IMF, “Challenges with measuring production abroad (global production),” Oct. 28, 2013. This meeting tried to address issues related to calculating services and production outsourcing in manufacturing, and the issue has been discussed in subsequent meetings as well. For decisions from the latest meeting, refer to: International Monetary Fund, “IMF Committee on Balance of Payments Statistics: 2024 annual report,” March 4, 2025.

  21. Patricia Buckley and Akrur Barua, “Income inequality takes a step back as compensation rises for a changing workforce,” Deloitte Insights, Nov. 22, 2023.

  22. The authors have categorized the 22 broad occupations in the economy into high-wage, medium-wage, and low-wage occupations based on median wages in these occupations and how these wages have trended over time. The same categorization is applied to the manufacturing workforce in this analysis. For a view into how we have analyzed these occupations through a wage lens, refer to the link here for a previous article on the subject.

  23. Akrur Barua, “The tech workforce is expanding—and changing—as different sectors battle for talent,” Deloitte Insights, Dec. 16, 2021.

  24. The White House, “Joint Statement on a United States-European Union framework on an agreement on reciprocal, fair, and balanced trade,” Aug. 21, 2025; The White House, “Fact sheet: President Donald J. Trump drives forward billions in investments from Japan,” Oct. 28, 2025; The White House, “Joint fact sheet on President Donald J. Trump’s meeting with President Lee Jae Myung,” Nov. 13, 2025.

  25. Jim Kilpatrick, Kristine Dozier, Kate Hardin, and John Morehouse, “Enhancing supply chain resilience in a new era of policy,” Deloitte Insights, April 1, 2025; refer to the section titled “Identifying a tipping point for reshoring production to the United States” for a sample test case of cost-based reshoring decisions. 

  26. The Manufacturing Institute, “Manufacturers need as many as 3.8 million new employees by 2033,” accessed Dec. 15, 2025; John Coykendall, Kate Hardin, John Morehouse, Victor Reyes, and Gardner Carrick, “Taking charge: Manufacturers support growth with active workforce strategies, Deloitte and The Manufacturing Institute, April 2, 2024.

  27. Ibid.

  28. Reshoring Initiative, “2025 reshoring survey report,” accessed December 2025. 

  29. Ibid. 

  30. US Bureau of Labor Statistics, Employment Cost Index, sourced using Haver Analytics in December 2025.

  31. Coykendall, Hardin, Morehouse, Reyes, and Carrick, “Taking charge.”

  32. Reuters, “Trump says foreign companies need to train Americans after Hyundai raid,” Sept. 8, 2025; Trevor Hunnicutt, Ben Blanchard, and Yimou Lee, “Exclusive: Trump team wants Taiwan to train US chip plant workers, sources say,” Reuters, Nov. 27, 2025.

  33. US Bureau of Labor Statistics, “Foreign-born workers: Labor force characteristics (2024),” press release, May 20, 2025.

  34. Immigration Policy Tracking Project, “Tracking Trump administration immigration policies,” accessed December 2025.

  35. Akrur Barua and Michael Wolf, “Tariffs will impact the economy... and so will uncertainty,” Deloitte Insights, April 11, 2025; Michael Wolf, “United States Economic Forecast: Q3 2025,” Deloitte Insights, Sept. 30, 2025.

  36. Ibid.

Acknowledgments

The authors would like to thank Ira Kalish, chief global and US economist, and John Morehouse, US research leader for Industrial Products Manufacturing, for their reviews and suggestions.

Editorial (including production and copyediting): Arpan SahaSayanika Bordoloi, and Anu Augustine

Design: Harry Wedel and Govindh Raj

Audience development: Atira Anderson

Knowledge services: Bichapogu Rishitha

Cover image by: Harry Wedel

Copyright