Elsevier

Land Use Policy

Volume 82, March 2019, Pages 283-291
Land Use Policy

Payment for ecosystem services programs in the Brazilian Atlantic Forest: Effective but not enough

https://doi.org/10.1016/j.landusepol.2018.11.054Get rights and content

Highlights

  • The first PES impact evaluation in South America with a counterfactual approach, separately evaluating forest regeneration and deforestation.

  • PES programs in the Atlantic Forest incentivized forest regeneration, but its overall conservation impact was low.

  • Achieving robust forest restoration gains with PES programs in the Atlantic Forest would require approximately 180 years.

Abstract

Payment for ecosystem services (PES) are economic incentives intended to generate conservation benefits, principally on private properties. The global portfolio of PES programs is estimated to represent an annual investment of more than USD $36 billion. Despite this substantial investment, the continued lack of systematic and rigorous impact evaluation of PES has contributed to uncertainty regarding the effectiveness of this tool for conservation. Here, we assessed the ability of two watershed PES programs to promote native forest conservation in the Brazilian Atlantic Forest. Those two focal programs are among the most well-established PES programs in Brazil, and form part of a larger network of PES programs intended to protect the drinking water supply of over 19 million people in the São Paulo metropolitan region. Using a counterfactual approach, we examined if enrollment in a PES program contributed to the conservation of on-farm native forest cover. With propensity score matching, we identified a set of neighboring, non-enrolled ‘control’ properties, with similar size, altitude, soil type, demographic density, presence of water sources and forest cover. We then estimated forest cover on enrolled and control properties before and after PES implementation with a differences-in-differences method, modeled as the probability of an observed change in on-farm forest area as a consequence of PES enrollment. We found that PES has a positive effect on forest cover, with PES enrollment over a five-year period associated with an additional 2.8–5.6% of farm area coverage in native Atlantic Forest, through forest regeneration. PES enrollment was associated with a non-significant trend toward decreased loss of vegetation. We discuss the implications of these results for understanding the contribution of PES to additionality in forest conservation. While positive, the relatively slow impact of PES on forest regeneration suggests that environmental managers should not count exclusively on PES mechanism to achieve conservation goals.

Introduction

Payment for ecosystem services (PES) programs are economic incentives for environmental conservation. They are generally defined as “a transfer of resources between social actors, which aims to create incentives to align individual and/or collective land use decisions with the social interest in the management of natural resources” (Muradian et al., 2010). Globally, PES have became a popular complementary strategy to existing conservation strategies like protected areas (Schomers and Matzdorf, 2013; Ezzine de Blas et al., 2016; Salzman et al., 2018). Despite the extensive investment in PES as a conservation tool, the correlates of conservation success in PES programs remain poorly understood (Ferraro and Pattanayak, 2006; Pattanayak et al., 2010; Naeem et al., 2015; Ezzine de Blas et al., 2016; Velly and Dutilly, 2016).

One reason why clear links between PES implementation and conservation outcomes are rare may be study design. Many studies either measure forest conservation outcomes in ways that obscure important underlying socio-ecological processes, or use designs that fail to demonstrate additionality (i.e. conservation outcomes that would not have happened without the PES intervention). For example, a large number of studies that evaluate PES impact consider net forest change as the main indicator of avoided deforestation (Sanchez-Azofeifa et al., 2007; Honey-Rosés et al., 2011; Arriagada et al., 2012; Robalino and Pfaff, 2013; Yang et al., 2013; Goh and Yanosky, 2016; Jayachandran et al., 2017). This is problematic, as changes in net forest cover may be driven by avoided deforestation or alternatively through forest regeneration - processes that are incentivized by distinct PES program activities. Furthermore, a lack of net change in forest cover does not necessarily mean a lack of change in the landscape, since it is possible to swap between forested and deforested states without an overall change in land use proportions (Pontius and Millones, 2011). Evaluations of PES programs for forest conservation outcomes should thus ideally be based on spatially explicit tracking of the distinct land use transitions linked to both deforestation and regeneration processes.

A second key component of PES-mediated forest conservation evaluations is the demonstration of additionality, particularly given the challenge of distinguishing PES-mediated outcomes from outcomes generated by broader ‘background’ socio-economic-ecological trends (Velly and Dutilly, 2016). Such background trends include inter alia environmental legislation, and patterns of national or regional economic growth and natural resource use. To distinguish conservation outcomes caused by conservation interventions from those that occur because of economic slow-downs, or enforcement of environmental regulation requires a counterfactual approach capable of inferring causality to interventions, particularly where randomized treatments are not possible (Engel et al., 2008; Ferraro, 2009; Jones and Lewis, 2015; Wunder, 2015; Velly and Dutilly, 2016). Despite this importance, remarkably few PES impact evaluations incorporate counterfactual approaches (Salzman et al., 2018). For example, Calvet-Mir et al. (2015) recently reported that only seven of 26 peer-reviewed evaluations of the effectiveness of 24 different PES projects incorporated some kind of counterfactual approach. These studies often employ matching methods that use strategic subsampling from among treated and potential control cases to create treatment and control assignments that are similar in all respects except the intervention of interest (Rubin, 1973; Rosenbaum and Rubin, 1983; Morgan and Harding, 2006). A smaller group of studies combine counterfactual approaches with the differences-in-differences method, that allows further comparison before and after the intervention (Honey-Rosés et al., 2011; Arriagada et al., 2012; Clements and Milner-Gulland, 2014; Costedoat et al., 2015).

In this study, we use a spatially explicit and counterfactual approach to evaluate the additional contribution of PES programs to the regeneration and avoided loss of native vegetation, i.e. PES additionality, near São Paulo, Brazil. We define forest conservation outcomes in three distinct ways, as: i) net forest change for each property, ii) the percentage of each property under regeneration, and iii) vegetation loss rate. To our knowledge, this is the first evaluation of the conservation impact of PES in South America to explicitly detect vegetation loss and regeneration with a counterfactual approach, as well as the first quantitative assessment of PES program effectiveness within Brazil’s Atlantic Forest, despite more than 20 programs operating in the region over the past two decades (Guedes and Seehusen, 2011; Calvet-Mir et al., 2015; Grima et al., 2016).

Section snippets

Study region

The study region is located within the threatened Atlantic Forest, which hosts one of the world’s most diverse tropical forest biota (Mittermeier et al., 2004). Following over 500 years of European colonial history, the Atlantic Forest’s estimated original 1,500,000 km2extent has been radically transformed. An estimated 28% of native vegetation cover remains across the Atlantic Forest, mostly consisting of edge-affected or secondary vegetation small fragments disconnected from larger remnants (

Results

We found that PES enrollment resulted in an increase in net forest change, measured as either the percentage of forest per total property area (Eq. (1)) or the rate of forest change per property (Eq. (2)), resulting in net forest gain among PES farms (Fig. 3; Table 2). Average annual net forest change on enrolled farms increased from 0.43% (± 0.80) per farm before implementation to 0.98% (± 1.81) after implementation. Over approximately five years of program implementation, this resulted in an

Discussion

We demonstrate that PES has a positive effect on forest cover, through a counterfactual assessment of the effectiveness of PES programs for vegetation loss and regeneration outcomes (Ferraro, 2009; Velly and Dutilly, 2016). PES enrollment increased both the percent of forest area per farm, and the rate of increase of forest per farm, driven by forest regeneration. We estimate that PES enrollment in the study region would result in approximately 2.8–5.6% of total farm area in additional forest

Conclusion

We found that while PES has a significant additional positive impact on forest cover in the Atlantic Forest of Brazil, the overall conservation impact is relatively low and conservation gains may be considered vulnerable. Over a five-year period, we observed PES-driven forest gain through tropical forest regeneration in the Cantareira region. Despite these gains, these positive impacts of PES may be vulnerable to both a lack of legal and or temporal additionality. In either case, any reduction

Acknowledgments

This study was developed within the ‘Interface Project’, supported by the São Paulo Research Foundation (FAPESP, 2013/23457-6). PR was supported by doctoral fellowships from the Brazilian Ministry of Education (CAPES-DS 1437068 and CAPES-PROEX 1560102), and São Paulo Research Foundation (FAPESP, 2015/16587-6). JPM was supported by the Brazilian Science Council (CNPq). EN was supported by post-doctoral fellowships from the National Science Foundation Grant (1158817) and the São Paulo Research

References (60)

  • S. Schomers et al.

    Payments for ecosystem services: a review and comparison of developing and industrialized countries

    Ecosyst. Serv.

    (2013)
  • R. Sierra et al.

    On the efficiency of environmental service payments: a forest conservation assessment in the Osa Penincula, Costa rica

    Ecol. Econ.

    (2006)
  • G. Sparovek et al.

    The revision of the Brazilian Forest Act: increased deforestation or a historic step towards balancing agricultural development and nature conservation?

    Environ. Sci. Policy

    (2012)
  • S. Wunder

    Revisiting the concept of payments for environmental services

    Ecol. Econ.

    (2015)
  • W. Yang et al.

    Performance and prospects of payments for ecosystem services programs: evidence from China

    J. Environ. Manage.

    (2013)
  • ANA

    Manual Operativo do Programa Produtor de Água.

    (2012)
  • R.A. Arriagada et al.

    Do payments for environmental services affect forest cover? A farm-level evaluation from Costa Rica

    Land Econ.

    (2012)
  • C. Banks-Leite et al.

    Using ecological thresholds to evaluate the costs and benefits of set-asides in a biodiversity hotspot

    Science

    (2014)
  • D. Bates et al.

    Fitting linear mixed-effects models using lme4

    J. Stat. Softw.

    (2015)
  • P.H.S. Brancalion et al.

    Finding the money for tropical restoration

    Unasylva

    (2012)
  • T. Clements et al.

    Impact of payment for environmental services and protected areas on local livelihoods and forest conservation in northern Cambodia

    Conserv. Biol.

    (2014)
  • S. Costedoat et al.

    How effective are biodiversity conservation payments in Mexico?

    PLoS One

    (2015)
  • N. Dias

    O Sistema Cantareira e a crise da água em São Paulo: falta de transparência

    um problema que persiste

    (2016)
  • D. Ezzine de Blas et al.

    Global patterns in the implementation of payment for environmental services

    PLoS One

    (2016)
  • P.J. Ferraro

    Counterfactual thinking and impact evaluation in environmental policy

  • P.J. Ferraro et al.

    Money for nothing? A call for empirical evaluation of biodiversity conservation investments

    PLoS Biol.

    (2006)
  • D.A. Fournier et al.

    AD model builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models

    Optim. Methods Software

    (2012)
  • H. Gonçalves

    Pagamento por serviços ambientais segundo a ótica da comunidade envolvida – o caso do projeto “Conservador das Águas”, Extrema/MG

    (2013)
  • F. Guedes et al.

    . Pagamento por serviços ambientais na Mata Atlântica: Lições aprendidas e desafios

    (2011)
  • D.E. Ho et al.

    MatchIt: nomparametric preprocessing for parametric causal inference

    J. Stat. Softw.

    (2011)
  • Cited by (0)

    View full text