Key takeaways:

• The unemployment rates for young college graduates and young noncollege workers have risen slightly faster than the overall unemployment rate.
• But the rise in young college graduate unemployment in particular was mostly due to higher labor force participation: The employment-to-population ratio for young college graduates has held steady since 2024.
• Certain demographic groups, such as Black and Hispanic workers, face higher unemployment and lower hourly wages, even for young people with limited work experience.
• In the long run, the college degree is losing its edge: Unemployment for young college graduates has risen in historical terms, and the college wage premium has been flat or falling in recent years.

Over the last couple of years, the overall labor market has slowly weakened—with many arguing that the weakening is most pronounced for young college graduates (whom we define as young workers ages 22–27 with only a college degree).1 The evidence is actually pretty mixed—by some measures the young college graduate labor market is notably weaker, but their outcomes are largely no worse than those of noncollege young people or the labor market writ large.

In this first blog post of our series on young college graduates, we examine the labor market the college graduates of the Class of 2026 are entering. We look at unemployment rates and employment-to-population (EPOP) ratios for young workers with and without a college degree and examine wages for young college graduates by demographic characteristics. We also explore longer-term trends in unemployment driven by rising educational attainment, as well as changes in the college wage premium—the pay advantage college graduates earn over their high school graduate peers. In the next post, we will analyze trends in the industries and occupations young college graduates tend to work in, and take a closer look at the tech sector and any fingerprints of AI on labor market outcomes.

Unemployment on the rise for young college graduates—but mostly because of higher labor force participation

Over the last couple of years, the labor market has shown some signs of weakening, though some often reported measures are overstating it. For example, payroll employment growth has slowed significantly, but this is largely driven by much slower population growth over the past year and a half as net immigration has collapsed. The unemployment rate has slowly increased, though the share of the prime-age population—those 25 to 54 years old—with a job has remained high. Of most concern is the hires rate—the number of hires as a share of total employment—which has been steadily falling over the last three years. The hires rate is now at the same levels seen in 2013 and 2014, a period during the prolonged recovery from the Great Recession that saw unemployment rates 3.0 percentage points higher than they are today. Focusing just on unemployment rates, the softening of the overall labor market appears to be hitting young college graduates more acutely.

Figure A shows the overall unemployment rate, as well as the unemployment rate for young college graduates and young workers without a four-year college degree. Since 2023, the overall unemployment rate has risen from 3.6% to 4.3%, a slow and measured increase of 0.7 percentage points. The unemployment rate for young college graduates has increased from a low of 4.0% in July 2023 to its recent high of 5.3% in March 2026, a faster increase of 1.3 percentage points. Young workers without a college degree also experienced a rise in unemployment, though their rise began a little later than the other groups. Their unemployment rate has risen by 1.2 percentage points since March 2024, up from 5.9% to 7.1% by March 2026.

Figure A

Some have pointed to this disproportionate rise in young college graduates’ unemployment rates as evidence that AI is beginning to substitute for the white-collar jobs young graduates typically enter. But this conclusion is premature for several reasons. First, while the rise in the unemployment rate in the most recent period is faster for college graduates than for all workers, the same is true for other young workers without a college degree (see Figure A). This suggests that there isn’t anything particularly damaging to young college graduates happening today, such as AI specifically destroying their labor market prospects.

Further, the increase in the unemployment rate for young college graduates over the last two years appears to be driven by an increase in labor force participation rather than a declining probability of having a job. The EPOP for young college grads has held steady over the last two years as the unemployment rate rose. Nearly all (98%) of the increase in the unemployment rate between 2024 and 2026 for young college graduates was driven by the increase in the labor force—meaning more young workers are entering the labor market in search of opportunities as opposed to giving up and leaving the labor force or never entering it at all. This would actually fit a historic pattern in which labor force participation rates tend to respond with a surprisingly long lag to labor market developments. The historically strong labor markets of the early 2020s likely are still pushing up the labor force participation rates of young college graduates today.

When we look at EPOPs since 2019, shown in Figure B, we see that young workers, college and noncollege alike, fall in line with the overall trend. Not surprisingly, given the industries and occupations hit the hardest, young noncollege workers fared the worst in the pandemic recession, but now are faring similarly to their college-educated counterparts. Prime-age EPOPs have remained the most resilient through this business cycle.

Figure B

Data from the Job Openings and Labor Turnover Survey provide useful insights into job openings, hires, quits, layoffs, and other separations, but they are not broken down by demographic, limiting our ability to analyze young workers. However, the depressed hires rate suggests that it is more difficult for new entrants to get a foothold in the labor market. The quits rate is down, signaling a reduction in the overall churn in the labor market as workers and employers sit tight through this period of economic uncertainty—likely related to chaotic policy decisions and implementation around tariffs, deportations, and the conflict with Iran. If the layoffs rate ticks up now, the unemployment rate is likely to spike quickly and could spell even more trouble for young people who tend to experience larger swings in unemployment with the business cycle.

Finally, there is no evidence that young college graduates are sheltering in school—i.e., going on to graduate school—to weather out the weakened labor market. In fact, enrollment rates among young college graduates have been falling slightly over the last couple of years, from 19.1% in 2024 and 18.8% in 2025 to 18.5% in 2026. Even though opportunities in the labor market are weaker, it’s perhaps not surprising that enrollment rates are on the decline. The Trump administration’s attacks on higher education have reduced available funding at colleges and the ending of student loan forgiveness and caps on borrowing make it increasingly difficult for students to make those educational investments.

Wages remain unequal across demographic groups

Real (inflation-adjusted) median wages of young college graduates rose slightly over the last year, up just 0.4% since 2025, consistent with the slowdown in wage growth for workers overall. Since 2019, young college graduate wages have grown 7.4% after adjusting for inflation.

Despite this positive wage growth, racial and gender wage gaps remain large even among young college graduates who are just starting their careers. Figure C shows that women are paid $4.18 less per hour than their male counterparts. At 85.9% of men’s pay, a young woman working full time with a college degree is paid $8,700 less over the year.

Young Asian American Pacific Islander (AAPI) college graduates are paid more than white, Hispanic, or Black workers. The demographic categories shown in Figure C are mutually exclusive: AAPI, white, and Black workers are non-Hispanic, while Hispanic workers can be of any race. Young white college graduates are paid $2.76 per hour less than their AAPI counterparts, while Black and Hispanic workers are paid $5.36 and $5.05 less, respectively. For a full-time worker, this translates into more than $10,000 in lower earnings over the year for Black and Hispanic workers. Not only are wages lower for these historically disadvantaged groups, but the unemployment rates of young Black college grads in particular are also higher. Therefore, their ability to secure employment at all—at any wage—is diminished.

Figure C Young college grads are competing against a wider labor force that is more educated 

As educational attainment has risen across the broader workforce, the advantage that young college graduates once enjoyed relative to the rest of the labor force in terms of lower unemployment and higher wages has steadily declined.

The young college graduate unemployment rate recently surpassed the overall unemployment rate, meaning a greater share of young college graduates are now out of work than workers writ large. The erosion of the unemployment advantage for young college grads, however, isn’t a sudden shift; as shown in Figure A, the trend has been building since 1979, when young college graduates had an unemployment rate of 4.0%, 1.9 percentage points below the national average. Over the following four decades, that advantage eroded. By February 2020, young college graduates had an unemployment rate of 3.8%, 0.2 percentage points above the overall rate, a slow but complete reversal of the historic edge. In recent years, the historic trend has continued—now the young college graduate unemployment rate is a full 1.0 percentage point above the overall. And the young college graduate unemployment rate is at historically high absolute levels today, currently sitting higher than it was during the worst of the 1990 and 2001 recessions.

The shift is not explained by young college graduates faring worse relative to their noncollege peers, as that gap has held relatively stable at around 2.0 percentage points. Instead, as the educational attainment of the overall workforce increased, young college graduates became less advantaged compared to the overall labor force. Further, as a greater share of young adults now attend college and are likely from a wider range of socioeconomic backgrounds, a college degree for somebody in their early 20s today is likely a less reliable marker of general economic privilege than it used to be.

Figure D displays educational attainment over time for young workers and all workers. From 1980 to 2026, the share of the workforce with a bachelor’s degree increased from 12.5% to 26.1%, more than doubling as a share of total employment. The overall level of college attainment for young adults rose from 18.0% in 1979 to 31.6% in 2026. If we include those with bachelor’s and/or an advanced degree in the overall workforce, the increase in educational attainment is even more stark, rising from 18.4% to 41.9%.

Figure D

The democratization of college degrees carries some clear upsides for productivity and the overall health of the U.S. economy. A more diverse set of people have accessed higher education and benefited from the advantages of being a degree holder in recent decades. This rise in college attainment has obviously not been costless, as many of these degrees could only be obtained by taking on large amounts of student debt, which may well provide some constraints on labor market opportunities and options for young adults.

Young college graduates used to be more male, white, and likely to come from higher-income families—all characteristics rewarded (fairly or not) in labor markets. This growing diversity of college graduates may well mean that young grads are less likely to exit (or not enter) the labor force when job prospects are bad. In prior decades, it is possible that young college graduates were more likely to have resources to fall back on during periods of unemployment, and they clearly had less student loan debt. Now, with fewer fallback options and greater debt levels, the cost of being jobless may weigh more heavily on this group, leading people to continue actively searching for work instead of staying out of the labor force even when jobs are scarce, driving the unemployment rate higher.

The long-term rise in educational attainment may also have helped squeeze the wage advantage college graduates hold over those with just a high school degree. Some of the same reasons discussed above—the college-educated population becoming more economically diverse and more workers attaining advanced degrees—may also have eroded the measured earnings edge that once came with just a bachelor’s degree.

A useful way to measure this is the college wage premium. The college wage premium is the percentage boost in wages associated with holding a college degree, after controlling for demographic factors like race, gender, age, and geography. As Figure E shows, this premium peaked around 2015 and has declined slowly since. Today, the overall college wage premium stands at 55.2%, roughly where it was in the late 1990s. For younger workers ages 22–27, the premium is slightly lower, but follows the same pattern, also peaking around 2015 before flattening or trending downward.

Figure E

The labor market is weakening and young workers’ prospects seem worse-off than they did just a couple of years ago. But young college graduates are facing a weakened labor market only slightly worse than that experienced by other workers. Their unemployment rate has risen faster, though similarly to noncollege young workers, while their employment-to-population ratio has remained generally strong. A depressed hires rate may make it even harder for these young workers to get a foothold in the labor market. Much of these short-term trends of higher and rising unemployment are the continuation of a decades-long trend of worsening outcomes as the overall population increases their educational attainment.

1. Throughout this brief, we define young college graduates as people between the ages of 22 and 27 with only a college degree. Unlike similar analyses of young workers, We do not exclude young college graduates that are currently enrolled in school, but the results here are robust either way. Unless otherwise noted, data for 2026 represent a 12-month average from April 2025 through March 2026 for the most up to date and reliable estimates, which removes seasonality and increases sample sizes. Analysis for smaller demographic groups uses a 36-month average to improve reliability.

Key takeaways:

• Black unemployment rose and employment fell in Q1 2026, reflecting a deterioration in labor market conditions. In the first quarter of 2026, the Black unemployment rate (7.6%) was 1.2 percentage points higher than in the first three months of the second Trump administration.
• Black men’s employment-population (EPOP) ratio decreased by 1.7 percentage points (from 60.5% to 58.8%) since the first quarter of 2025, with noncollege graduates driving this decline.
• Black women’s EPOP ratio was the same in Q1 2026 as in Q1 2025 (56.4%), with gains among noncollege graduates offsetting losses among college graduates.

The rising Black unemployment rate and big employment losses among Black women made major news headlines in 2025. In a February 2026 analysis, I examined the nature of those losses, noting the large impact on Black women who were college graduates and public-sector workers. With so much of the Trump policy-induced 2025 labor market decline appearing to land first on Black workers who typically have relatively secure employment, the longer-term significance of those losses is of continuing interest. This post provides an update for the first quarter of 2026, examining changes in the overall Black unemployment rate and gender-specific employment trends for Black women and men relative to the first quarter of 2025. For consistency with the prior analysis, I apply the same mutually exclusive race and ethnicity categories used in EPI’s State of Working America Data Library and include all people age 16 or older when examining outcomes by gender. While these estimates differ slightly from those reported by the Bureau of Labor Statistics (BLS), they lead to similar conclusions.

In the first quarter of 2026, the Black unemployment rate (7.6%) was 1.2 percentage points higher than in the first three months of the second Trump administration. While a rise in the unemployment rate can sometimes be for “good” reasons—workers getting drawn into the labor force because of strengthening job opportunities—that was not the case here: the rise in the Black unemployment rate reflected a decline in employment. The Black employment-population ratio (EPOP)—the share of working-age people who are employed—declined 0.8 percentage points over the same period, from 58.3% in Q1 2025 to 57.5% in Q1 2026 (see Figure A).

Figure A

Looking more closely at changes in employment for Black women and men separately, Black women’s first quarter EPOP was the same in 2026 as in 2025 (56.4%), while employment among Black men was 1.7 percentage points lower (from 60.5% to 58.8%). BLS published estimates by race—limited to the sample of people age 20 or older and not exclusive of ethnicity—show a similar decline for Black men (-1.5 percentage points), but a 0.4 percentage point increase for Black women.

Figure B shows that among Black women, Q1 2026 employment was lower than Q1 2025 for college graduates but higher for noncollege graduates, resulting in essentially offsetting effects. The opposite was true among Black men, for whom the decline in employment was driven by lower employment among noncollege graduates and higher employment for college graduates.

Figure B

These first quarter 2026 estimates incorporate annual population adjustments applied to Current Population Survey (CPS) data each January to reflect updated population estimates from the U.S. Census Bureau. Since the previous year’s data are not adjusted, monthly data across the two years are not strictly comparable. This year, shifts in the demographic composition of the population also resulted in larger than usual discontinuities in labor force measures by race, ethnicity, and gender between December 2025 and January 2026—which is why this analysis is focused on a comparison between the first quarters of 2025 and 2026, when the population controls are the most up to date.

Based on this analysis, we can conclude that overall, labor market conditions for Black workers were not better in the first quarter of 2026 compared with the early months of the Trump administration. Black men’s employment is lower than what was reported in the first quarter of 2025, and while Black women’s employment is unchanged overall, employment among college-educated Black women is lower than first quarter 2025 estimates.  

May 1 is International Workers’ Day. Also known as “May Day,” its origins trace back to 1856 in Australia, where workers organized a day of stoppages and celebrations to demand an eight-hour workday. However, May 1 didn’t become a widespread international day for labor until after the infamous Haymarket Affair of 1886.

Workers in Chicago, including many immigrants, went on strike on May 1 to demand the eight-hour workday. At least four strikers were killed while picketing the McCormick Harvester factory, at that point the largest factory in the world. A large rally was held on May 4 to protest violence against peaceful picketers. As police moved to disperse the crowd, someone threw a bomb that killed seven officers. Police fired back indiscriminately, wounding and killing an undetermined number of workers.

What followed was a sweeping crackdown: police raids, the arrests of hundreds of men and women, and the indictment of eight people—five of whom were German immigrants. The partisan judge Joseph E. Gary conducted the trial where all 12 jurors acknowledged prejudice against the defendants. All defendants were convicted with no evidence and seven were sentenced to death; four were hanged, one died by suicide, and two had their sentences commuted. The trial is widely considered a miscarriage of justice.

In the aftermath, socialists and unionists worldwide began marking May 1st as a day of international worker solidarity. However, in 1894, U.S. President Grover Cleveland—looking to make peace with labor prior to the midterm elections after more than 30 workers were killed during the Pullman Strike—established Labor Day in early September. He did this explicitly to avoid associating it with May Day and the labor unrest it represented. In 1955, at the height of the Cold War, President Eisenhower proclaimed May 1 “Loyalty Day” instead of “May Day” in response to the holiday’s popularity in communist countries.

Labor unions today

Now 140 years after Haymarket, workers are still fighting for higher pay, better working conditions, and a voice on the job. In recent decades, policymakers have done little to stem the relentless tide of anti-union actions by employers, conservative governments, and a hostile Supreme Court. As workers’ rights have been eroded, the share of unionized workers fell from over 30% in the 1950s to just 11.2% in 2025. Fewer workers were involved in major strikes or work stoppages in 2025 (307,000) than during the Haymarket year of 1886 (610,000).

Nonetheless, there are clear signs of momentum in the labor movement. The post-pandemic period has brought a notable resurgence in labor’s popularity and organizing activity. Figure A shows that 68% of Americans now approve of labor unions, levels not seen since the 1960s. Unions are also more highly regarded among young people. Further, 43% of Americans want unions to have more influence in the country, a record high.

Figure A

Not only are unions more popular, but more workers have been trying to join a union. Figure B shows that the 2024–2025 period saw the highest number of newly unionized workers since at least 2000.

Figure B

Indeed, while the Trump administration has taken a decidedly hostile approach to unions and made labor organizing more difficult, union representation in the United States increased by 463,000 in 2025. More workers were represented by a union than at any point in the past 16 years, a sign that workers see unions as a means of resisting authoritarianism.

The time is ripe for policymakers to support workers’ struggles for dignity and respect. Key policies such as passing the Protecting the Right to Organize Act, ensuring workers can reach a first contract, expanding collective bargaining rights, and eliminating anti-union “right-to-work” laws can help workers organize their workplaces. Beyond improving the lives of their members, unions have spillover effects that benefit whole communities and democracy.

This May Day, workers and their unions across the country are holding thousands of events, encouraging participants to join an economic blackout and “demand a nation that puts workers over billionaires.” Just as workers around the world came together to demand fair hours and wages after the events of 1886, we can hope the workers of the future will find inspiration from May Day 2026.