Spillovers of Payments for Ecosystem Services: Follow-up of a RCT in Uganda

In the treatment villages, Chimpanzee Sanctuary and Wildlife Conservation Trust (CSWCT) staff members offered an incentive contract to each household that owned forest land, under which they will receive annual payments if they meet certain terms. Landowners are required to refrain from cutting trees on their land (with some exceptions built into the contract) and also to re-forest a portion of their land. CSWCT employees monitor compliance with the contract by conducting random spot checks in the forest to look for newly cleared patches of forest or fresh tree stumps and to assess if new trees have been planted.

2011-06-01

2014-01-31

1) social preferences
2) environmental preferences
3) forest sharing behavior
4) social norms related to forest sharing behavior
5) human-wildlife conflicts
6) attitude towards chimpanzee conservation
7) hypothetical enrollment in a PES program

The primary outcomes are planned to be analyzed in separate publications. Paper 1: Outcomes 1, 3, 4; Paper 2: Outcome 2; Paper 3: Outcome 5,6; Paper 4: Outcome 7

1) elicited through three binary decisions for resource allocation between 2 participants to identify social preferences types;
2a) decision between six seedling packages with varying compositions of eucalyptus and native tree species (eucalyptus representing the less environmentally-friendly option)
2b) self-reported tree planting within last 12 months (yes, no, type of species, reasons for planting)
3a) self-reported access restrictions of forest owners and non-forest owners
3b) framed-field experiment (2 players: 1 forest owner, 1 neighbor) concerning forest access and resource appropriation
4) incentivized elicitation of social norms regarding forest access restrictions
5) frequency of wildlife sightings, crop and livestock damage by wildlife
6) agreement with statements concerning chimpanzee conservation
7) hypothetical PES enrollment under varying payment levels and take-up rates at the village level

1) Environmental attitudes
2) Locus of control (environmental conservation)
3) Perceived forest benefits and costs
4) Motivations for tree planting

All secondary outcomes are analyzed for Paper 2.

1) index of self-reported survey items
2) index of self-reported survey items
3) number of forests benefits and costs listed in open question
4) index of self-reported survey items

One hundred and thirty-six villages in the Hoima and Kibaale districts of Uganda were randomly assigned to either the treatment or the comparison group. Households in both treatment and control groups received a baseline questionnaire prior to the PES program being implemented. Additionally, pre-intervention high-resolution QuickBird satellite images were taken of the study area. Boundaries of land parcels were demarcated with handheld GPS units and netbook computers and linked to the remote-sensing images to create images of and measures of the forest cover on each household’s land. Endline data collection, including both a household survey and Quickbird satellite images, will be conducted after the intervention has been implemented for two years.
The follow-up survey collects additional outcome measures that have not been collected in the household surveys before (neither baseline nor endline). The follow-up study therefore applies a cross-sectional experimental design, harnessing the random allocation of the treatment as an instrument.

In each subcounty, public lotteries were conducted to select which of two stratified groups of villages received the PES program.

village

Yes

original RCT: 136 villages, follow-up: 121 villages

1210 land owners in total: 726 forest owners (at the start of the intervention), 484 neighbors of forest owners

121 villages: 60 treatment, 61 control

Payments for ecosystem services (PES) are increasingly being implemented worldwide as conservation instruments that provide conditional economic incentives to landowners for a prespecified duration. However, in the psychological and economic literature, critics have raised concerns that PES can undermine the recipient’s intrinsic motivation to engage in pro-environmental behavior. Such “crowding out” may reduce the effectiveness of PES and may even worsen conservation outcomes once programs are terminated. In this study, we harnessed a randomized controlled trial that provided PES to land users in Western Uganda and evaluated whether these incentives had a persistent effect on pro-environmental behavior and its underlying behavioral drivers 6 y after the last payments were made. We elicited pro-environmental behavior with an incentivized, experimental measure that consisted of a choice for respondents between more and less environmentally friendly tree seedlings. In addition to this main outcome, survey-based measures for underlying behavioral drivers captured self-efficacy beliefs, intrinsic motivation, and perceived forest benefits. Overall, we found no indications that PES led to the crowding out of pro-environmental behavior. That is, respondents from the treatment villages were as likely as respondents from the control villages to choose environmentally friendly tree seedlings. We also found no systematic differences between these two groups in their underlying behavioral drivers, and nor did we find evidence for crowding effects when focusing on self-reported tree planting behavior as an alternative outcome measure.

Vorlaufer, Tobias, Stefanie Engel, Joost de Laat, and Björn Vollan. 2023. ‘Payments for Ecosystem Services Did Not Crowd out Pro-Environmental Behavior: Long-Term Experimental Evidence from Uganda’. Proceedings of the National Academy of Sciences 120 (18): e2215465120. https://doi.org/10.1073/pnas.2215465120.

https://doi.org/10.1073/pnas.2215465120

Conservation policies and programs may trigger unintended, potentially irreversible, changes that were initially not anticipated. Concerns have been raised that the introduction of payments for environmental services (PES) fosters the privatization of natural ecosystems to the detriment of marginalized groups. We assess the long-term impacts of PES on sharing of access to natural resources, associated norms, and social preferences. The studied PES program was implemented as a randomized control trial in western Uganda. Using survey and experimental data collected six years after the last payments were made, we find that the PES program did not lead to a lasting shift in resource sharing practices but did induce stronger social norms for resource sharing. Moreover, landowners in former PES villages exhibit more egalitarian social preferences than landowners in control villages. These results highlight that despite introducing unequal conservation benefits to communities, long-lasting negative spillovers of PES could be avoided.

Vorlaufer, Tobias, Joost de Laat, and Stefanie Engel. 2023. ‘Do Payments for Environmental Services Affect Forest Access and Social Preferences in the Long Run? Experimental Evidence from Uganda’. Journal of the Association of Environmental and Resource Economists 10 (2): 389–412. https://doi.org/10.1086/721440.

https://doi.org/10.1086/721440