Brad Hershbein
banner
bradhershbein.bsky.social
Brad Hershbein
@bradhershbein.bsky.social

Senior economist and deputy director of research at the @upjohninstitute. My research focuses on the transition between education and career and how employers hire and compensate workers. Creator of New Hires Quality Index. #NHQI .. more

Education 39%
Economics 37%

Read the full release here: www.upjohn.org/sites/defaul..., and check out the NHQI here: www.upjohn.org/nhqi. 21/21 #econsky
www.upjohn.org

Taking a longer view, over the past 25 years, real wage growth of new hires has lagged considerably at the very bottom, even as inequality has closed between the lower-middle and the top. 20/20

This recent narrowing since the pandemic can be seen below, with the strongest cumulative gains in inflation-adjusted wages of new hires around the 30th percentile, with lower growth in the top half. 19/20

Consequently, while wage inequality among new hires had been narrowing sharply in recent years, this has stopped as of 2025, with inequality once again growing. 18/20

This means that the stagnation in mean wage growth, now in its 4th year, has been driven by different parts of the distribution since 2022. 17/20

Then, between 2023 and 2024, however, inflation-adjusted wages of new hires plunged 10% at the 90th percentile. Over the past 12 months, this reversed again, with wage growth fastest at the top. 16/20

These patterns have rapidly changed. As recently as the beginning of 2023, the top had seen the strongest cumulative growth, with the bottom lagging badly behind. 15/20

Since the COVID recovery began in the summer of 2020, cumulative growth has concentrated in the lower half of the distribution, with increases of about 14 percent at the 25th percentile and 11 percent at the 10th percentile. 14/20

Since 1999, the average inflation-adjusted wage of new hires has increased by 37.2% (1.22% annually), but the median has increased by somewhat more, while the 10th percentile has increased by slightly less. 13/20

These changes have been uneven, though. We can look at inflation-adjusted, self-reported hourly wages of new hires since 1999 at different parts of the wage distribution. 12/20

Netting out the 8.7 percent growth in the NHQI since 2005, composition-adjusted real wages of new hires have grown 18.0 percent, or about 0.83 percent per year. 11/20

Over the longer term, inflation-adjusted, self-reported hourly wages of new hires have grown 26.7 percent since 2005, with essentially all this growth occurring between 2015 and 2022. 10/20

In contrast, between July 2020 and July 2022, average real wage growth of new hires, controlling for changes in their occupations and demographics, rose 5.9 percent. 9/20

Roughly speaking, the difference between the two series implies that average real wage growth of new hires, controlling for changes in their occupations and demographics, fell 0.6 percent between 2022 and 2025. 8/20

Between July 2022 through July 2025, the actual inflation-adjusted wages of new hires rose 0.7 percent, while the NHQI wage index rose by 1.3 percent. 7/20

The spread between inflation-adjusted wage growth of new hires and that of incumbent workers (as captured by the Federal Reserve Bank of Atlanta’s Wage Growth Tracker) did narrow from last year: 1.8% vs +1.5%, but this is tempered by hiring rates being at record lows. 6/20

The recent pattern is a strong indicator of a deteriorating labor market, a weakening noted both in 2023 (www.upjohn.org/sites/defaul...) and 2024 (www.upjohn.org/sites/defaul...) 5/20
www.upjohn.org

This stagnation over the past 3.5 years follows a blistering 23 percent inflation-adjusted increase between January 2015 and January 2022. 4/20

The current (July 2025) average inflation-adjusted wage of new hires, $24.49, is up just $0.17 (0.7 percent) since July 2022, although it remains 9.3 percent above its level in February 2020. 3/20

Actual, inflation-adjusted wage growth of newly hired workers surged at an annualized rate of 3.5 percent between July 2020 and July 2022, then declined slightly over the next two years before recovering slightly in the 12 months to date. 2/20

This month’s @upjohninstitute.bsky.social NHQI, special for Labor Day, examines the *actual* reported wage growth of newly hired workers. (Typically, the NHQI is based on occupational wages, capturing “earnings power” but not individual self-reported wages.) www.upjohn.org/sites/defaul.... 1/20
www.upjohn.org

The labor market has weakened, for everyone, and young college grads aren't really suffering much more than all workers, as indicated by the employment rate chart above. That doesn't mean it's easy, just that the overwhelming focus on young BAs might be a little misplaced.

While Thompson makes a good argument that AI is changing how people (especially recent college grads) search for jobs, it is not AI that is making the hiring picture so gloomy for recent college grads. It's business uncertainty and a very unusual reference frame.

Notice a theme? Businesses don't like to hire when there's a lot of uncertainty about what's going to happen. Young people bear the brunt: research.upjohn.org/cgi/viewcont...
research.upjohn.org

These groups, coming out of school, are especially sensitive to hiring trends, which have a LOT of headwinds right now, from the simple reversion to the mean, to tariff uncertainty, interest rate uncertainty, geopolitical uncertainty, and AI uncertainty.

However, the labor market in 2021 and 2022 was unprecedently hot for young folks, regardless of education, and this sort of became the benchmark in many minds, especially for college students seeing their peers a few years ahead.

So what's really going on? The labor market has been weakening for everyone for the past two and a half years. Hiring in particular is way down, www.upjohn.org/sites/defaul..., even if layoffs remain relatively muted.
www.upjohn.org

The reason for the different patterns? Changing labor force participation rates. Young college grads are about as likely to be in the labor force as they were at the end of the Greta Recession. All working age folks, not so much, as they have gotten much older on average.

Now we see that the unemployment rate of young college grads has largely exceeded the unemployment rate of all working-age adults since the end of COVID. Unlike the *employment rate*, the gap today is much larger than in 2018, and this looks bad.

If we look at the *unemployment rate*--the fraction of people without a job and looking for one relative to people either with jobs or unemployed--we get this picture: