Poverty and Economic Behavior: Gambling at Social Security Paydays

Abstract

The goal of this research is to explore whether actual lottery revenues are sensitive to scarcity, as measured by intra-monthly variation in financial resources. Exogenous paydays of social security benefits are employed to generate the intra-monthly variation in financial resources. Using two million observations on daily lottery revenues that cover more than 2,500 lottery outlets in Israel for two years (2015-2016), I find that gambling revenue spikes at social security paydays. The estimation results imply that on Income Support payday aggregate lottery revenues are higher by 5 percent after controlling for outlet, weekday, holidays, month and year fixed effects. However, the calculated aggregate response of lottery revenues on Income Support payday is quite small and equal 0.5 percent of the total monthly payments deposited to the bank account of Income Support recipients. In addition, the other social security and salary paydays induce a trivial impact relative to total monthly payments deposited to the bank account of the respective recipients. These results survive a list of sensitivity analyses and pass a placebo test.