Why So Many Unemployed Workers Leave Benefits on the Table

In summer 2023, Stanford PhD candidate Brendan Moore joined the Stanford Impact Labs PhD Fellowship, an opportunity for graduate students collaborating with real-world organizations to put social science to work for society. In partnership with the Washington State Employment Security Department, Brendan was focused on designing and testing solutions that support workers. With results now in hand, Brendan offers striking insights on how state agencies can approach unemployment insurance.

Even during times of economic resilience, employers routinely lay off workers as part of the normal churn of business activity. During 2024, when the aggregate U.S. unemployment rate hovered at 4.0 percent, 1.7 million workers were laid off on a monthly basis. For many, with job loss comes large earnings losses, adverse health outcomes, and a sense of instability and precarity about the future.

In the United States, the primary program providing temporary income support for workers who lose their jobs through no fault of their own – as in mass layoffs, plant closures, or seasonal layoffs – is Unemployment Insurance (UI). However, only half of eligible workers claim UI benefits when they are laid off. The average person on UI receives more than $7,000 during their unemployment spell, suggesting that many people potentially forfeit substantial sums to which they are legally entitled.

Why is it that so many people facing unemployment forgo critical income support when they are without a paycheck? What policies would increase UI take-up? If we increase take-up, how might this impact workers’ return to the labor force? These are the questions I set out to answer in my recent study co-authored with Casey McQuillan and supported in part by Stanford Impact Labs. 

Our research revealed that many unemployed individuals do not claim UI due to burdensome application steps, stigma associated with UI, and worries about having to comply with work search requirements. The study also found that reducing these barriers (and thereby increasing take-up) does not prolong joblessness, because work search requirements encourage newly enrolled workers to search more actively for jobs.

To better understand why eligible workers fail to claim UI – and what happens when these barriers are reduced – we partnered with the Washington State Employment Security Department (ESD) to study the application decisions of those who had recently been laid off. Working with the agency, we identified 50,199 individuals who had likely lost their jobs and were eligible for UI but had not yet applied. We then mailed letters, on official state letterhead, to a randomized subset of that group explaining how the unemployment insurance system works and how to apply. The letters, drafted in conjunction with ESD staff members, provided the phone number, application URL, and offered a Spanish language alternative. Half of the letters also sought to destigmatize UI benefits by clarifying that UI is an earned benefit and is not based on financial neediness.

Prior to the experiment, we knew that only about 43% of eligible unemployed workers in Washington had filed for UI benefits. In our study, workers received the informational letters once they had been out of work for an average of three months, a point that is often marked by a person’s savings running low. Two months after mailing the letters, UI take-up among those who received the letters rose to 45%, a statistically significant increase driven by a less complicated application process. The workers who were most likely to receive UI benefits as a result were those displaced from lower wage jobs – paying under $25 per hour.

We also wanted to test whether directly addressing the perceived stigma of UI could further increase take-up. Along with the basic informational letter, half of the workers received a brief note explaining that UI is an earned benefit funded by payroll taxes – not a means-tested benefit. We found that adding this message increased take-up by about one-third as much as the informational letter alone, a statistically significant uptick. Notably, these gains were driven entirely by higher-wage workers displaced from jobs paying between $25 and $40 per hour. This pattern is consistent with the idea that higher-wage workers may be less likely to view themselves as “the type” to claim public benefits, but become more receptive once they understand that UI is a social insurance program they have paid into over time.

Finally, we examined whether increasing UI take-up prolongs job search. A large body of research shows that more generous unemployment benefits can lengthen unemployment spells, so it is natural to worry that expanding access to UI might delay re-employment. In this setting, however, we found no evidence of delay. If anything, workers who received letters – who were more likely to receive UI – returned to work slightly faster.

We attribute this pattern to the role of work search requirements. Before claiming UI, unemployed workers were not required to document their job search. Once enrolled, however, they must regularly verify search activity to remain eligible for benefits. We found that this requirement meaningfully changed behavior: treated workers are more likely to actively search for jobs and attend career services workshops at public employment offices. Importantly, we show that these faster returns to work are not driven by the public employment services themselves, but by the incentives created by the search requirements.

At the same time, these search requirements are far from costless. Many of the workers induced to apply ultimately do not verify search and never receive a UI check. In this sense, work search requirements play a dual role: they screen out many people induced to apply, while accelerating job search among those who do receive UI. This helps explain why expanding take-up in this context does not come at the cost of longer unemployment spells.

Taken together, our experiment’s findings show that a modest, low-cost intervention can meaningfully expand access to unemployment insurance without discouraging work. All U.S. states are required to collect earnings records on UI-covered employees, which could be used to identify workers who lose their jobs but haven’t claimed benefits. Therefore, our approach of outreach is readily scalable beyond Washington State. By partnering directly with a state agency and testing simple interventions in real-world settings, this project illustrates how rigorous evidence can inform practical policy design, helping public programs better reach workers when they need support most.