and

This paper examines the effect of rainy day funds on state savings behavior. We find that states with rainy day funds have higher total balances than states without such funds and also have higher balances after adoption than before adoption. Furthermore, rainy day fund deposits increase total balances dollar-for-dollar. While we cannot rule out that states planning future savings may adopt rainy day funds, our findings are robust to the inclusion of measures of savings preferences. In sum, these funds appear to belong to the growing set of fiscal institutions with real fiscal and economic consequences. 1Introduction During the last two decades, virtually all of the U.S. states have adopted budget stabilization funds, often called "rainy day funds, " that allow them to save for unexpected revenue shortfalls. Prior to 1981, few states had such funds (Gold, 1981). By 1984, 18 states had enacted rainy day funds, and by 1994, 45 states had them (ACIR, 1995). These accounts are designed to help state governments stabilize public spending over time by saving during booms and using the balances to cover revenue shortfalls during recessions. In 1996, rainy day fun