For the last five years, I’ve given the same answer in response to questions about any one-month increase in the Black unemployment rate. Given the relatively small sample size used to calculate the number each month, we shouldn’t make too much of a single month’s increase but focus on longer-term patterns and see if the upward trend continues over the next few months. Well, as of August 2025, the Black unemployment rate has risen for three consecutive months and now stands at 7.5%. This post details three major conclusions I have drawn from this and supporting data:

1. There has been a clear deterioration in the labor market for Black workers this year: the unemployment rate is rising and employment is falling.
2. The decline in Black workers’ employment appears to be concentrated among Black women while Black men’s employment rates appear more stable.
3. Since January 2025, overall women’s employment has fallen most in professional and business services, manufacturing, and federal government—suggesting likely culprits for the decline in Black women’s employment.

An important signal that the rising Black unemployment rate may actually be more than a temporary blip in a notably volatile data series is that the share of employed Black adults between the ages of 25 and 54 is down compared to the last couple of years. After peaking at a historic annual high of 77.9% in 2024, the average so far this year is 76.6%. Until now, the rate had risen every year since 2021.

Another developing news story that has garnered increasing attention is the reported 300,000 Black women losing jobs and/or leaving the labor force in recent months. While a number that big certainly makes headlines, employment levels from the monthly household survey—especially those based on a small demographic slice of the monthly survey sample—should always be used with caution. While I’m not convinced that 300,000 is the most accurate accounting of the situation, Black women are uniquely experiencing a decline in employment that is not observed among other groups of women or Black men.

A clearer and more reliable indicator of how Black women are doing in the labor market is their employment-to-population ratio (EPOP). As Figure A shows, Black women’s employment has dropped sharply this year, but there has been a longer downward trend that started in early 2024. This stands in stark contrast to the trend for white women whose EPOP has changed little over the same time while Hispanic women have seen a slight increase.

Similarly, Figure B shows that the EPOP for Black men in the same age group has been much more stable over the last three years. Since the decline for Black women is not reflected in other group trends by gender or race alone, there appears to be something happening in the labor market that has been particularly damaging to Black women. Below, I explore possible explanations based on analysis of payroll employment data.

The 2020 pandemic recession showed how occupational segregation can contribute to a larger decline in employment among groups overrepresented in industries with the largest job losses. While the Bureau of Labor Statistics does not report industry-specific job losses by race and gender, they do produce a series on women’s payroll employment using data collected in the monthly establishment survey. And although the data from the household and establishment surveys are not directly comparable, checking for similar or related trends can be informative. With those caveats in mind, Figure C shows that women’s payroll employment has declined in eight industries between January and August of this year. The largest of those losses have occurred in professional and business services (–83,000), manufacturing (–41,000), and federal government (–33,000). Close to half of all workers in federal government and professional and business services are women, as are 29% of manufacturing workers (see Table 1).

Since federal job cuts have frequently been cited as a contributing factor for employment losses of Black women due to overrepresentation in that sector relative to their share of the total workforce, let’s start there. Table 1 shows that the year-over-year rate of decline in payroll employment of all women in federal government (–1.2%) was slower than the rate of decline in total employment (–2.4%) in that sector. This suggests that women aren’t losing jobs faster than men in the federal sector, but without additional information, we can’t rule out the idea that women’s federal losses are disproportionately falling on Black women.

The second-largest number of women’s job losses has been in manufacturing. As Table 1 shows, women are a smaller share of total employment in this industry compared with the federal government or professional and business services. However, unlike in the federal government, women have lost manufacturing jobs at a faster than average rate over the past year.

The largest number of women’s job losses has been in the professional and business services industry, which employs seven times more women than the federal government. As such, a higher rate of women’s job losses in this industry would be more likely to show up in declining EPOPs for Black women. Compared with the total year-over-year rate of job losses in the industry (–0.3%), women have lost jobs at a much faster pace (–1.3%).

As shown in Table 1, a closer look into the rate of employment decline in professional and business services shows that losses have been overwhelmingly concentrated in employment services—a decline of 3.2% between July 2024 and July 2025. Similarly, women’s losses within professional and business services were also skewed toward employment services.

Employment services accounts for roughly 15% of total employment within the professional and business services major industry. Women are nearly half (47.9%) of employment services workers and the most common occupation is human resource worker. This suggests the shedding of employment services jobs as a likely culprit behind large job losses among women in the industry. Again, additional information would be required to definitively conclude that those job losses are disproportionately falling on Black women. Well-documented patterns of occupational segregation often limit the representation of Black women in higher-level and higher-paying professional and management positions, resulting in overconcentration in service, administrative, and support occupations. Thus, it would not be a huge leap to assume that women’s job losses skewed toward employment services, over one of the subsectors beginning with “professional, scientific, and technical” or “management”, are probably having a negative impact on Black women’s employment.