Poster #69 - The Effects of Payment for Environmental Services on Environment and Socio-economic Well-being

• In Event: Saturday Poster Luncheon

Abstract

A central question in environmental policy is whether conservation interventions impose a trade-off on economic well-being, and whether their effects on the environment endure beyond the period of active compensation. This study estimates the short- and long-run impacts of two Payment for Ecosystem Services (PES) programs on both environmental and socioeconomic outcomes, drawing on randomized controlled trials (RCTs) conducted in Uganda and Bolivia.

Both RCTs were initially implemented in 2011 and randomized at the village level. The Uganda program compensated participating households with cash payments of $28 per hectare for forest conservation, while the Bolivia program (Watershared) offered in-kind compensation for cattle exclusion and forest conservation practices. We estimate intention-to-treat (ITT) effects and, instrumenting for household-level participation, local average treatment effects (LATE) on a range of outcomes. For short-run socioeconomic outcomes, we use ground survey data to examine the value of crops sold, the value of forest products, the area of land cultivated, net PES benefit, net worth, and livestock holdings. For long-run environmental outcomes, we extend existing tree cover classifications using deep learning applied to high-resolution satellite imagery in Uganda, and rely on Mapbiomas land cover data in Bolivia. We additionally proxy for long-run economic well-being using the High Resolution Electricity Access (HREA) indicator and the share of building area to village area, drawn from satellite-based datasets spanning up to eight years post-treatment.

Our findings are as follows. First, PES did not generate a trade-off between conservation and poverty in either country. In Uganda, PES reduced deforestation during the program period without significantly decreasing agricultural income, forest income, net worth, or livestock holdings. In Bolivia, PES had no statistically significant effect on deforestation or on any measure of agricultural or fruit production. Second, the conservation benefits of PES were not sustained in the long run. In Uganda, the positive effects on tree cover disappear when measured over the 2011–2025 window, suggesting that participants resumed deforestation at rates similar to the control group after compensation ended. In Bolivia, we find no significant effect on forest cover at any post-baseline time horizon examined, contradicting earlier conjectures about potential long-run improvements through behavioral channels. Third, satellite-based proxies for economic well-being - electricity access and building density - show no significant treatment effects in either country across the eight years following the intervention.

Taken together, these results suggest that while PES can achieve short-run environmental gains without harming participants' livelihoods, it falls short as a vehicle for sustained conservation or meaningful poverty reduction. These findings have important implications for the design of conservation policy in low-income settings, where PES programs are increasingly proposed as tools to jointly advance environmental and development objectives.

Authors

• Sacha Siani, University of Notre Dame
  Presenting Author

• Seungmin Lee, University of Notre Dame
  Non-Presenting Co-Author