Behavioral Spillovers from Targeted Incentives: Losses from Excluded Individuals Can Counter Gains from Those Selected

Abstract

An increasing number of policies condition transfers upon the taking of socially desired actions − such as donating blood, departing conflict or mitigating climate change. Many such incentives are targeted, i.e., they exclude individuals based on potential recipients' characteristics or actions. We hypothesize that: pro-sociality can be reduced by exclusion despite no price or income changes; and, further, the rationale for excluding people can itself influence whether any such undesirable side effects of pro-social incentives arise. To test for such 'behavioral spillovers', we use a laboratory experiment to study a subsidy to pro-social donations in which subjects are fully informed about why they are selected, or not, for the subsidy. We introduce three selection rules and track changes in donations. Selecting for the subsidy those who initially acted less pro-social (i.e., gave little to start) increased donations, while rewarding greater pro-sociality did not. Yet the selection rule which targeted those with lower prior pro-sociality also intentionally excluded the people who donated more initially and that selection rule reduced the donations by the excluded. This shows a tradeoff between losses from excluded and gains from selected individuals.

Keywords: monetary incentives, conditional payments, economic experiments, behavioral economics

JEL Classification: C91, D03