Report

The COVID-19 pandemic had a profoundly negative impact on employment across the U.S. workforce. However, the impact was less for workers in STEM occupations than it was for those in non-STEM occupations (figure 5-1). Unemployment rates for STEM workers increased 1.5 percentage points between 2019 and 2021 (from 2.1% to 3.6%), whereas unemployment for non-STEM workers increased 2.8 percentage points (from 3.9% to 6.7%) during this period.

Unemployment rates reflect both employer and employee sides of employment. Unemployment rates are increased by job loss (employer side effect) and are decreased by workers leaving the workforce, either voluntarily or due to layoffs or job loss and not seeking reemployment. The pandemic increased the demand for some occupations and reduced the demand for others.

Unemployment rates for both men and women in STEM and non-STEM occupations increased over this 3-year period. Within each occupation group, the amount of increase for men and women was comparable. However, for both years (2019 and 2021) and both occupation groups (STEM and non-STEM), the unemployment rates for men were higher than those for women.

Within the racial and ethnic groups, increases in STEM unemployment rates between 2019 and 2021 were statistically significant for Whites, Hispanics, and Blacks but not for Asians. However, when comparing among groups, the increase was greater for Blacks (3.4 percentage points from 3.2% to 6.6%) and Hispanics (2.9 percentage points from 2.8% to 5.7%) than it was for Whites (1.0 percentage point from 1.9% to 2.9%). For non-STEM occupations, the unemployment rates in 2021 were higher than the rates in 2019 for White, Asian, Hispanic, and Black workers. The increases over the 3-year period for Asians (4.6 percentage points from 2.9% to 7.6%), Hispanics (3.9 percentage points from 4.3% to 8.2%), and Blacks (3.8 percentage points from 6.7% to 10.5%) were higher than those for Whites (2.0 percentage points 3.2% to 5.2%). Both in 2019 and 2021 and for all racial and ethnic groups, those working in STEM occupations had lower unemployment rates than those in non-STEM occupations.

Overall, the unemployment rates of workers with or without one or more disabilities were lower if they worked in STEM rather than in non-STEM occupations. However, both groups experienced comparable increases in unemployment in STEM and non-STEM occupations from 2019 to 2021. The unemployment rate for workers in non-STEM occupations with at least one disability was higher in 2021 (11.2%) than in 2019 (8.5%). In contrast, unemployment rates were comparable in 2021 (5.3%) and in 2019 (3.8%) for workers with at least one disability in STEM occupations. For those without a disability, the unemployment rates for both occupation groups increased over the 3-year period.

When the STEM workforce is divided by broad occupation type, those in middle-skill occupations had the highest unemployment rate in 2021 (5.9%), which was about twice that of S&E (2.5%) and over twice that of S&E-related (2.1%) occupations (figure 5-2). Higher unemployment rates for middle-skill occupations occurred across all groups of workers except for those with at least one disability.

Although women in middle-skill occupations had a higher unemployment rate than men in these occupations, the unemployment rates for men and women in S&E or S&E-related occupations were similar.

In the middle-skill occupations, Blacks experienced the highest unemployment rate (10.6%) out of all the racial and ethnic groups. Unemployment rates for S&E workers who are Black (4.8%) or Hispanic (4.4%) were higher than those who are Asian (1.7%) or White (2.2%). In S&E-related occupations, Whites had lower unemployment rates (1.4%) than did Blacks (4.4%) and Hispanics (3.4%), but they had rates that were comparable to Asians (2.3%).

Although STEM workers with no disabilities had similar unemployment rates as those with disabilities in middle-skill occupations (5.9% and 6.1%, respectively), their rate was notably lower than that for those with a disability in S&E-related occupations (1.9% and 6.8%, respectively).

Overall, among all STEM workers in 2021, higher educational attainment is associated with lower unemployment (figure 5-3). STEM workers with a bachelor’s degree or higher had a lower unemployment rate than those employed in the skilled technical workforce (2.4% vs. 4.9%).

For both men and women, unemployment was higher if they worked in the skilled technical workforce than if they were STEM workers with at least a bachelor’s degree. However, there was no difference in the unemployment rates of men and women in the STEM workforce by education: 2.6% of men and 2.2% of women with at least a bachelor’s degree were unemployed, as were 5.1% men and 4.2% of women without a bachelor’s degree.

The pattern of higher unemployment rates for workers in the skilled technical workforce than for STEM workers with at least a bachelor’s degree persists across all racial and ethnic groups. However, in both educational groups, Blacks had higher unemployment rates than Whites or Asians, but their rates were not statistically different from rates of Hispanic unemployment. In the skilled technical workforce, Hispanic unemployment was higher than White unemployment, but it was not statistically different from the unemployment rates of Asians and Blacks.

For STEM workers with at least one disability, educational attainment had little effect on unemployment. For those with a disability, the unemployment rate among the skilled technical workforce (5.2%) was about the same as the rate among the workforce with a bachelor’s degree or higher (5.4%). Likewise, when comparing the unemployment rates of each educational group by disability status, the differences between the estimates are not statistically significant. That is, unemployment rates are comparable between those with at least a bachelor’s degree who have one or more disabilities (5.4%) and those who have no disability (2.4%). Additionally, there is no statistical difference between the unemployment rates for those without a bachelor’s degree by disability status (5.2% with at least one disability vs. 4.8% with no disability).