Fix unemployment insurance to help young men into work

May 23, 2024
Matt Darling

Many government policies, though ostensibly neutral, have effects that vary by gender. It is important to be aware of these differences—in either direction—in policy design and evaluation.

The financing of unemployment insurance (UI) programs is one example, as I show in recently published research for the Niskanen Center. UI benefits are paid for by taxes applied to individual companies, with the tax rate based on that company’s history of workers drawing on the UI system. For the same reason your car insurer raises your rates after an accident—to encourage you to be more careful and hopefully reduce the likelihood of future claims—states raise the tax rate employers pay for unemployment insurance after they lay off a worker. This “experience rating” system is intended to encourage companies to avoid layoffs, but it has had several perverse consequences.

Most importantly, employers also change how they hire workers. Specifically, they will avoid hiring workers that they anticipate will be a future layoff risk. My new research shows that these workers are predominantly men, either because men tend to work in high-turnover industries or because men are more likely to be fired for other reasons.

Increasing the employment of men is rightly seen as an important goal for economic policy. Under current federal law, all states are effectively required to use experience rating in their UI systems, eliminating this requirement for UI would be a step in the right direction.

How unemployment insurance is paid for

The unemployment insurance system in the United States is a joint federal-state program. States operate their own UI systems within thresholds established by the Federal government. UI provides cash benefits to workers who have recently lost their jobs. While each state has its own mechanism for calculating benefits and eligibility, a typical unemployed worker can expect to have 40% of their employment income replaced for a period of several months conditional on a continued search for work.

The UI program was first established in the 1930s. Since then, there have been only relatively minor changes to the system. The core of the program is more or less the same as it was 90 years ago. One exception is how the program is financed. Initially, states had a wide degree of flexibility in how they financed their UI system. Many states used an experience rating system, others adopted a flat payroll system or their own unique variants. After a 1985 tax reform, the federal government began to effectively mandate experience rating, and all states moved to an experience rating system.

In a “fully experience rated” program, all benefits would be effectively paid for by future taxes. However, since their programs are somewhat more complex, states generally only have “partial experience rating”. For example, states typically have a maximum rate that firms can be charged. As a result, companies typically pay around 29% of their former employees’ UI benefits.

One response from employers could be to change the composition of their workforce in order to reduce the risk of higher future UI taxes. For example, instead of hiring an in-house custodial staff, a company could decide to hire gig workers or contract with another company. In that case, they can simply end the contract instead of firing workers and triggering a rise in their tax liabilities.

They could also shift to hiring workers who are lower risk, i.e. less likely to be laid off in the future. One way of doing this would be to limit hiring entry-level workers who have not already demonstrated their ability to get the job done.

Lessons from Washington State

Washington state was one of the last states to move to an experience rating system, switching only after the 1985 tax reforms. I use this switch to estimate how firms’ hiring decisions changed, compared to neighboring Oregon, which already had the system in place.


Figure 1


As Figure 1 shows, before 1985 the unemployment rate of young workers in the two states moved in parallel. Following that year’s introduction of experience rating in Washington, there was a divergence, with the unemployment rate for younger workers in Washington increasing relative to their Oregon counterparts. Using a triple difference methodology that compares changes across both state and age variables, I estimate that after experience rating was introduced, the unemployment rate for these entry-level workers increased by 2.5%.

But there is a gender story here too. This effect among young workers is entirely driven by the change in the unemployment rate for young men. For men between 15 and 24, unemployment went up by 2.7% in Washington but by just 0.8% for young women, as Figure 2 shows. Unemployment among slightly older men aged 25-34 actually decreased by 1.7%, while for men aged 35-44 there appeared to be no effect.


Figure 2

Among women, the unemployment rate went up by 0.1% in both the 15-24 and 25-34 groups. However, there was a meaningful increase for women between 35 and 44, who saw their unemployment rates increase by 1.6%—possibly indicating a similar effect for women who are taking care of children.

Why men might be riskier to hire

Why would experience rating be more likely to affect young men than women?

There is a substantial academic literature on the relationship between gender and hiring decisions, especially in connection to the gender wage gap. While there was at one point substantial and generalized labor market discrimination against women, the effects have decreased over time and have plausibly been eliminated altogether.

There has been less investigation into the effects of gender in firing decisions. Research by the Federal Reserve after the 2008 financial crisis showed that men were more likely to become unemployed during that recession. Bureau of Labor Statistics data on labor force flows shows that the rate at which men become unemployed is typically 20% greater than that of women. In fact, with the exception of 2020, males have become unemployed at a higher rate than women in every year since at least 1991, as Figure 3 shows.


Figure 3


Why are men at greater risk of unemployment? One hypothesis is that men are overrepresented in the most economically vulnerable industries. Construction and manufacturing, for example, have very volatile employment, and typically firms in those industries pay much higher experience rated taxes than others. Both of these industries are also predominantly male: according to the Bureau of Labor Statistics only 11% of construction workers and 30% of manufacturing workers are female (while women make up 47% of the aggregate workforce).

It is also possible that men are simply a higher layoff risk in general, for reasons unrelated to their specific industry. Given that men are much more likely than women to drop out of college, it’s certainly possible that there is a similar relationship with employment. The same behavioral differences between young men and women that drive the former—such as greater risk-seeking, or a lack of maturity—could certainly result in male employees being a higher layoff risk; a risk  that employers would want to avoid under an experience rating regime.

Help men get work

Over the last several decades, the male employment rate has steadily dropped, as Figure 4 illustrates. In 1979, 91% of “prime age” men (those between the ages of 25 and 54) were working. Today, that number is 86%, despite a robust labor market. This difference is substantial. For context, the male employment level today is approximately the same as it was at the lowest point of the 2001 recession.


Figure 4


There have been substantial analyses of the potential causes of this decline and a raft of potential solutions. Many of these are big lifts—new programs to help train men for new jobs, subsidizing work through a stronger Earned Income Tax Credit, or dealing with specific problems like criminal records, disabilities, or drug addictions.

While all of these are worth considering, a more straightforward option is available. By ending experience rating for unemployment insurance taxes and moving to a simpler program like a flat payroll tax (which every other country uses), we can stop punishing firms for hiring men.

Matt Darling
Matt Darling is an employment policy fellow at the Niskanen Center. Previously, Darling was a vice president at ideas42, a behavioral economics consultancy, where he managed projects in economic justice, climate change, and domestic healthcare. He was also the strategic lead on labor and workforce development. Darling’s work has been featured in the Washington Post, Vox, the Atlantic, and more. Darling has received his B.A. in Cognitive Science and Economics from Hampshire College and his M.A. in Economics from Tufts University