It is well known that disability insurance (DI) enrollment is countercyclical. But less is known about why DI is countercyclical. Understanding this point is crucial given the rapid rise in DI caseloads in recent decades combined with the widely publicized forecast that the Social Security Disability Insurance Trust Fund will be exhausted by 2016. However, no systematic evidence describes how or why caseloads have changed during the Great Recession. In this paper, we compare DI applications and awards during the great recession to other recent recessions. We find that changes in the caseload from 2007 to 2010 are not unique compared with other recessions. We then use individual data on older U.S. workers from the Health and Retirement Study to analyze two hypotheses for why DI applications rise during recessions. Based on research suggesting that job loss and recessions more broadly have deleterious effects on health, we test whether the number and/or severity of health shocks during recessions can explain elevated DI application rates. Second, we test whether changes in the opportunity costs of applying for DI can explain higher DI application rates during recessions. Although we find evidence that severity of health shocks and measures of the opportunity cost of DI application both predict DI application among older workers, we find no support for either the health shocks or opportunity cost hypothesis. Alternative explanations for the countercyclicality of DI applications are required to describe recent recessions, including the Great Recession.