Image Source: There is a widespread belief that the US labor market has been undergoing a period of unprecedented chance in the last decade or two. On one hand, David Deming, Christopher Ong, and Lawrence H. Summers case doubt on this historical claim in their essay, ” Technological Disruption in the US Labor Market”–that is, they argue that historical shifts in US occupations have been much larger during various periods of the late 19th and 20th century than the more recent shifts. However, they also point to some signs that although the big shift in US labor markets may not have happened yet, it could be on its way. The essay appears as a chapter in a collection published by the Aspen Economic Strategy Group, , edited by Melissa Kearney and Luke Pardue. You can download individual chapters or the book as a whole.For an historical perspective on shifts in US labor markets, consider the figure below. If you look back to the late 19th century, about 80% of all US jobs were either in farming or blue-collar production. But by 1960s, farming jobs were less than 10% of the US total and falling sharply. After about 1950 to 2000, the share of blue-collar production jobs falls from 40% of all jobs to 20%. On the other side, there are sharp rises in professional jobs, office and administrative jobs, and other “services” jobs.
In short, part of a dynamic US economy has always involved dramatic shifts in jobs. US jobs have been undergoing dramatic evolution for decades. At present, the share of office and administrative jobs has been falling since 1980, and jobs in retail sales are now showing a decline. But the shifts in US job categories in the last couple of decades certainly don’t stand out as extraordinary in the historical record.But is the US economy now in a situation, with the rise of new artificial intelligence technologies, that could lead to truly dramatic shifts in US job markets? Maybe! It’s hard to know at this stage how the new technologies might be used and how jobs will be affected: history teaches that, when it comes to jobs, new technologies often both giveth and taketh away. Deming, Ong, and Summers point to four ongoing shifts.
Trend 1: Job polarization has been replaced by general skill upgrading
[E]mployment growth was highly polarized in the 2000–2010 period, with large gains at the bottom and the top of the wage distribution and declines in the middle. Polarization continued from 2010 to 2016, although to a much lesser degree than in the decade before. However, the labor market has stopped polarizing since 2016. Between 2016 and 2022, low-skilled and middle-skilled jobs both declined by about 2 percentage points, while employment in high-skilled occupations increased by slightly more than 4 percentage points. Thus, employment growth since 2016 looks more like the kind associated with skill upgrading than the kind indicating polarization. …
Trend 2: Flat or declining employment in low-paid service occupations.
A key explanation for employment polarization in the first decade of the 2000s was the rapid growth of service sector jobs, which replaced middle-skilled (often unionized) production jobs and offered lower wages and fewer employment protections … [T]he growth of service jobs stalled in the early 2010s and was flat for most of the rest of the decade. Employment in food service and personal-care occupations fell rapidly in 2020 as a result of the COVID-19 pandemic and had recovered only partly by 2024. With the lone exception of health support occupations, service sector employment is now similar to what it was twenty years ago, early in the first decade of the 2000s. Service occupations have given back nearly all of the rapid job growth they experienced during that decade. …
Trend 3: Rapid employment growth in STEM occupations
STEM (science, technology, engineering, and math) jobs shrank as a share of USemployment between 2000 and 2012, while employment in non-STEM professional occupations grew rapidly (Deming 2017). Deming (2017) documents this surprising fact and argues that the demand for social skills is rising because social interaction is required for complex work and is not easily automated by current technologies. … [W]hile the growth in social skill–intensive management, business, education, and healthcare jobs has continued, STEM employment is now also increasing rapidly after having declined in the first decade of the 2000s. The share of all employment in STEM grew from 6.5 percent in 2010 to nearly 10 percent in 2024. About 60 percent of this growth is concentrated in computer occupations like software developers and programmers, although employment has also grown across a wide range of science and engineering occupations as well.
STEM employment growth has accelerated especially quickly in the last five years. Moreover, the rapid employment growth in business and management jobs is concentrated in occupations like science and engineering managers, management analysts, and other business operations specialists. Increased employment in STEM occupations is also matched by increased capital investment in AI-related technologies. …
Trend 4: Declining employment in retail sales
Retail sales jobs held steady at around 7.5 percent of US employment between 2003 and 2013. Between 2013 and 2023, the US economy added more than 19 million jobs. Yet retail sales declined by 850,000 jobs over the same period, causing their share of employment to drop from 7.5 to 5.7 percent, a reduction of 25 percent in just a decade. The decline in retail sales jobs began before the pandemic but has accelerated in the last few years. …Each of these four trends—the end of polarization, stalled growth of low-paid service jobs, rapidly increasing employment in STEM occupations, and employment declines in retail sales—suggests that the pace of labor market disruption has accelerated in recent years.Patterns of US economic activity and job growth were shocked by the pandemic, and it’s not clear (at least to me) how the changes will shake out. Will work-from-home, shop-from-home, and online medicine continue to rise, or will they fall back over time? Perhaps the new disruption of US labor markets is not so much across the broad job categories used in the historical analysis above, but instead will involve a shift in the skills demanded from workers in their current job categories.More By This Author:Argentina’s Economic Challenge In Context Some Patterns Of Long-Run Asset ReturnsDesigning An Industry Subsidy: Repayable Advances?
