Additionality Violations in Payment for Ecosystem Service Programs : Experimental Evidence

Payment for ecosystem service (PES) programs incentivize farmers to implement agricultural best management practices (BMPs) with the goal of reducing nutrient and sediment runoff and improving water quality. These programs are widespread at both the federal and state level. Because some farmers adopt BMPs even in the absence of PES programs, it is natural to question the extent to which BMPs adopted by program participants are additional to the counterfactual scenario of no program. I test the assumption of perfect program additionality for a variety of subsidy programs in an experimental setting using a variant of the common public good game. I find that a large proportion of enrollments are not additional, as much as 100% for some subsidy programs. Further, I identify two sources through which additionality is violated: the pay-for-nothing (P4N) effect, in which the subsidy pays for a practice that would be done without incentives, and behavioral substitution, in which the subsidy generates substitution from an unincentivized BMP to the incentivized one, and quantify each effect for a series of subsidy regimes. I find that subsidies for “good” BMPs are largely ineffective, especially when a “great” BMP option is also available