Association for Public Policy Analysis & Management

Transitory and permanent income shocks may have varying effects on recipients and their neighbors. Among these effects are risk-sharing, in which individuals use their increase in income to provide loans and related items to neighbors as a way of insuring against future setbacks, altruism, in which recipients of greater income use it to increase the consumption of others, and social connections, in which recipients of greater income use it to build social connections with friends and neighbors. Neighbors may also try to imitate consumption patterns of recipients or may feel unhappy about changes on recipient’s income. To analyze these, and particularly how they may be affected by a permanent income shock, we analyze the results of a randomized control trial in an urban area of Mexico, where pension benefits were randomly assigned to some older adults and not to others. We exploit a double randomization design with differential proportions of treated households within randomly selected city blocks (treated blocks) to understand the presence of spill-over effects to non-recipients in the treated blocks, by comparing them with non-recipients in control blocks. We also estimate spillover effects by modeling spillovers as a function of distance between recipients and non-recipients. We use rich data collected before the pension as well as 14 and 26 months after its introduction. We found that non-recipients in treatment blocks report increased food availability and health care utilization as well as improved self-reported health. This evidence is consistent with benefit recipients being altruistic and sharing their resources.