Incentives conditioned on socially desired acts such as donating blood, departing conflict or mitigating climate change have increased in popularity. Many incentives are targeted, excluding some of the potential participants based upon characteristics or prior actions. We hypothesize that pro-sociality is reduced by exclusion, in of itself (i.e., fixing prices and income), and that the rationale for exclusion influences such ‘behavioral spillovers’. To test this, we use a laboratory experiment to study the effects of a subsidy to donations when participants are fully informed about why they are selected, or not, for the subsidy. We study the effects of introducing different selection rules upon changes in donations. Selecting for the subsidy those who initially acted less pro-social (i.e., gave little to start) increased donations, while random subsidies and rewarding greater pro-sociality did not. Yet a selection rule which targets lower prior pro-sociality also intentionally excludes the people who donated more initially, and only that rule reduced donations by the excluded. This shows a tradeoff between losses from excluded participants and gains from selected.