Association for Public Policy Analysis & Management

In February of 2018, 6.7 million American workers were unemployed. Of these workers, one in four had been unemployed for more than half a year (BLS, 2018). Unemployment has been linked to numerous negative outcomes, including increased risk of poverty and of material hardships. A major goal of the Federal-State Unemployment Compensation Program (UI) is to protect individuals and their households against the economic risks associated with unemployment. By providing weekly cash benefits to displaced workers while they search for new jobs, we expect that UI would help households to meet basic needs and act as a buffer against economic hardships. However, with a few exceptions, the prior literature has not paid a great deal of attention to the effects of UI on poverty and material well-being.

One reason for this lack of attention is that studies interested in identifying optimal benefit levels and optimal program size have primarily focused on how UI affects the behaviors of workers and firms. Another possible reason is that UI is not targeted towards the poor, and helping workers and their households reach or maintain a certain level of economic well-being is not an explicit goal. As a result, there may be less scrutiny of whether the UI program makes participants better off compared to means-tested programs such as SNAP and TANF.

Nevertheless, from our perspective, poverty and material hardship measures are particularly appealing for examining the effects of UI participation. First, determining optimal benefit levels requires identifying behavioral distortions and identifying what prior studies call the “beneficial insurance effect," such as knowing how UI receipt affects household income and household consumption of goods and services. Second, UI participation among individuals in or near poverty has received greater attention following welfare reform in the mid-90s and following historical rates of unemployment during the Great Recession. However, by focusing only on the poverty effects of UI, we would be ignoring the effects that the program might have on the economic well-being of households who are not in poverty, and we would be assuming that having a certain income level is synonymous with being able to meet basic needs.

In order to examine whether receipt of UI benefits reduces poverty and material hardships, we use data from the Survey of Income and Program Participation (SIPP) and bivariate probit regression analysis to model jointly the probability of UI benefit receipt and the probability of experiencing poverty and of experiencing housing, utility, food, and medical hardships. In order to account for unobserved differences between individuals who receive UI benefits while unemployed and those who do not, our models include state UI policies as instrumental variables. Similar to prior studies, our preliminary results suggest that UI receipt has a substantial negative effect on poverty, and that UI receipt reduces food insecurity, but not other hardships. By examining UI's effects on economic well-being this study contributes to current understanding of how the program is meeting the needs of workers in the modern economy.