Separating Anticipatory Labor Supply Response to Old-Age Cash Transfers

Cash transfers can affect labor supply through two channels: current transfers and anticipated future transfers. I develop a framework that jointly estimates both channels using a difference-in-differences model. I apply the framework to the noncontributory pension program in South Africa, which lowered the minimum eligibility age for men from 65 to 60. I find the anticipatory effect to be considerable: it accounts for 30 percent of the labor force contraction of the newly included age-eligible men. My analysis also finds that overlooking the anticipatory effect leads to bias in estimating the effect of current transfers. Moreover, I demonstrate that comparing the effects of current and future transfers can identify liquidity constraints for soon-to-be eligible people. These results show the importance of jointly estimating the two channels