Under what conditions can Payments for Environmental Services deliver sustainable improvements in welfare? Learning from a Randomized Control Trial

Lead Research Organisation: Bolivian Natura Foundation
Department Name: Socioeconomic and Policy Research


Wilson Maldondado watched his cattle die one by one. The 2010 Bolivian dry season was the worst in Wilson's memory, killing 200 cows in Villamontes and countless others across the Chaco. But it was not the first long drought. In 2004 more than 50,000 people were affected in Gutierrez, where more than 90 percent of the corn crop failed. In Gutierrez 20 communities had no drinking water, and even Villamontes town ran out. Wilson looked at the clouds gathering on the Sierra del Aguarague and wondered, with mountains so close, how the Chaco could be so dry.
Across Latin America, the watersheds that could provide users with clean water often have to support additional and sometimes conflicting functions, such as agriculture and forestry. Existing regulatory frameworks have often proved unable to reconcile these conflicting needs. Upper watershed farmers often have no economic alternative other than to deforest their land for agriculture. Upstream Water Factories are thus destroyed-often for a pittance-and cows enter streambeds to drink, forage, urinate and defecate. The subsistence agriculture of upper watershed farmers is unproductive and susceptible to climate change. Downstream municipal water sources are contaminated, children miss school with diarrhoea, sedimentation blocks pipes and dams, and waterholes supporting farmers like Wilson Maldonado dry up.

In 2003 in Los Negros, Bolivia, Fundación Natura Bolivia (Natura) helped initiate a new incentive based water conservation model: municipal payments for environmental services (PES). These projects are based on the twin axioms that 1) protecting upstream forests will help maintain water supplies in quantity and quality, and 2) downstream water users need to contribute to such forest protection. The key attributes of these schemes are the precautionary principle and local institution building and alignment. From humble beginnings in 2003, when 6 farmers agreed to protect 465 ha, more than 30,000 downstream users are now compensating 1,140 upstream families for protecting 35,000 ha of forest.

This research will identify conditions under which such small scale Payments for Environmental Services (PES) schemes can deliver sustainable improvements in welfare. We will use a series of Randomized Control Trials (RCT) to:

1. Explore how payments for ecosystem services can support poverty alleviation in the Bolivian Chaco. The Bolivian Chaco is hot and dry, poverty is widespread, the indigenous people's land is held communally, and drought is a major agricultural constraint. Conditions are thus not typical of Bolivia, and indeed are more like sub-Saharan Africa. We thus expect that lessons we learn will be applicable to less developed countries.

2. Take lessons and tools from Bolivia to other Andean countries, to evaluate the effectiveness of municipal-led PES, and the applicability of the RCT methodology for ecosystem service and poverty alleviation interventions. Taking advantage of an project that will train 200 municipal technicians from Peru, Colombia and Ecuador in how to set up small-scale PES schemes, we will undertake a controlled evaluation of the importance of external inputs and seed capital for PES development.

3. Develop and test an RCT evaluation toolkit that can support the Colombian Ministry of the Environment. We are will work with the Government's Direction of Forests and Ecosystem Services to help develop a monitoring and evaluation program for the national PES scheme which ensure that 1% of all municipal revenues are invested in upstream watershed protection.

The fundamental objective of this proposal is to assess under what conditions Payments for Environmental Services can deliver sustainable development. A secondary objective is to develop a series evaluation tools so that they can be applied globally, and which can help project developers quickly assess the effectiveness-and hence improve the impact-of their poverty reduction interventions.

Planned Impact

One of the fundamental, albeit indirect, goals of our research is to influence and help develop government policies for forest conservation, watershed protection and sustainable development. A major stakeholder group, therefore, are government institutions, including Parks Services and Ministries for Rural Development. We will involve such institutions at all stages of the project, with the goal of providing data and research results that feed into new science-driven policies for flood mitigation and watershed management.

As an example of how ESPA research might lead to policy change, the state Government of Santa Cruz is currently debating PES scale-up both as a state law, and as a mechanism to protect the watersheds above the city of Santa Cruz de la Sierra (1.5 million inhabitants). Our research on what works at a municipal level (as opposed to communities) can help provide a legal basis, and finance, to develop municipal PES throughout Santa Cruz State (an area ~50% larger than the United Kingdom). Other Andean countries also offer significant potential for research uptake: for example Colombia's soon-to-be passed Ecosystem Services Law will annually provide $250 million for land purchases, compensations and RWA. Having helped set up RWA in the Colombian municipalities of Roncesvalles and Guasca, we have advised the government's Direction of Forests and Ecosystem Services on how to best define the national program. Similarly, the Government of South Africa sent a delegation to visit Bolivia and has since expressed interest in applying the lessons of small-scale reciprocal agreements to its Working for Water Program.

Given that we have already replicated our local PES model in two different ecological and cultural zones in Bolivia, have helped set schemes up in the different socio political situations of Ecuador and Peru, and have had colleagues from South Africa adopt parts of the model, we believe that this model of conservation of ecosystem services really can help alleviate poverty. The reseach we propose here-across four countries-will further elucidate whether there is potential for small-scale PES to be implemented globally, in any such areas where upstream land managers can influence the quality and quantity of water available, and downstream users are able to contribute to the provision of watershed services. Two of our project partners, Kelsey Jack and Julia Jones have extensive research programs in Malawi and Zambia and Madagascar respectively. If our research suggests that our model of municipal PES might well be useful outside Bolivia, our next step will be to promote such schemes in eastern Africa.

At the international policy level, Bolivia's Climate Change Negotiating Team is promoting small-scale, "non market based approaches, such as joint mitigation and adaptation approaches for the integral and sustainable management of forests" through the UNFCC Durban COP17 resolution "Outcome of the work of the Ad Hoc Working Group on Long-term Cooperative Action under the Convention"

Unlike REDD, which focuses on market based mitigation, the Bolivian proposal ("Sustainable Forest Life") is a non-market based mechanism for simultaneous mitigation, adaptation, restoration, and economic development. Small-scale PES are the best concrete Bolivian (and for that matter global) example of the "Sustainable Forest Life" concept. The results of our impact evaluation-on how well small-scale conditonal transfer schemes can simultaneously mitigate and adapt to climate change, and provide local development-will therefore feed directly into Bolivia's attempt to negotiate grassroots pro-poor alternatives to REDD into future international climate treaties.


10 25 50

publication icon
Wiik E (2020) Mechanisms and impacts of an incentive-based conservation program with evidence from a randomized control trial. in Conservation biology : the journal of the Society for Conservation Biology

Description The major insights were that:
• Incentive based conservation schemes can work: RWA had positive impacts on water quality within 24 months
• Such schemes have slow initial uptake, but once confidence of landowners is gained, the win-win nature of the concept lead to uptake rates of ~70%
• RWA can be remarkably cost effective to implement, with local service users contributing almost 80% of costs.

A description of the RWA model can be found at:
(http://revista.drclas.harvard.edu/book/reciprocal-agreements-water) and (http://revista.drclas.harvard.edu/book/investing-latin-americas-water-factories).
Asquith N.M. 2013. Investing in Latin America's Water Factories: Incentives and Institutions for Climate Compatible Development. Harvard Review of Latin America 1: 21-4.
Asquith N.M. 2011. Reciprocal Agreements for Water: An Environmental Management Revolution in the Santa Cruz Valleys. Harvard Review of Latin America 3: 58-60.
Exploitation Route The research is already being acted upon by 35 municipal governments in Bolivia. Jointly with Natura Bolivia, these municipalities are trying yo scale up and out the reciprocity-based conservation model.

The scaling up process observed so far in Bolivia-a simple horizontal diffusion and modular replication of a small-scale model-is in stark contrast to the centrally led scale up of Payments for Environmental Services seen in countries such as Ecuador and South Africa. In Bolivia there is as yet little central government buy-in for RWA, nor is there significant central government financing. Rather than continue to inefficiently expand municipality by municipality, Natura has therefore been required to both take advantage of existing regional structures, and where they don't exist, create them.

However, Natura cannot continue to replicate the RWA model one watershed at a time. This is neither cost effective nor sustainable. Natura has neither the human nor the financial resources to invest significant staff time in the hundreds of municipalities where RWA could be a useful tool. Natura therefore believes that the best way to scale up the Bolivian experiences is to synthesize the lessons of the last ten years into a didactic learning tool. This tool can then help interested parties develop their own RWA and municipal water funds, with minimal direct and limited indirect input from Natura or other "experienced" NGOs. Our research was an important first step in this process, and the investigation played a pivotal role in redefing Natura as an institution.

Natura's ability to scale up its impact was a direct result of: 1) the institution's specialized focus on implementing and expanding the RWA model, which is now based on more than a decade of experience; 2) "scientific" trial and error learning within an NGO culture that stressed a process of "test-evaluate-learn" rather than "diagnose-implement-monitor"; 3) the development of strict research protocols that simultaneously mechanized the RWA implementation process; and 4) an academic focus on changing social norms and helping landowners feel that they are contributing to a public good, as opposed to the neoclassical economic focus of PES. Through these four research pillars this research project facilitated an institutional transformation within Natura.
Sectors Agriculture, Food and Drink,Communities and Social Services/Policy,Environment

Description Reciprocal Watershed Agreements (RWA)-locally financed rainforest conservation-were pioneered in Bolivia. Water users compensate upstream forest owners with development projects such as honey production and improved cattle management, in return for a commitment to forest conservation. RWA grew very slowly-from 6 beneficiaries protecting 1100 acres in 2003 to 265 beneficiaries and 12318 acres in 2010-until NERC funded research to assess how such schemes could be improved. Four years later, an additional 2000 families were protecting 230,000 acres in Bolivia, with almost 70,000 water users contributing £200,000 annually (80% of operating costs), the State of Santa Cruz (50% larger than the UK) designed a new RWA law, Bolivia's delegation to the UNFCCC presented RWA as the poster child of non-market alternatives to global REDD programs, and the RWA model had spread to 50,000 acres in Colombia, Ecuador and Peru.
First Year Of Impact 2014
Sector Environment
Impact Types Economic,Policy & public services

Description Creation of 450,000 acres of new Protected Areas In Huacareta and Machareti municipalities
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Implementation circular/rapid advice/letter to e.g. Ministry of Health
Impact Our research results supported the Municipal Governments of Machareti and Huacareta as they decided to create 450 000 acres of new Water Sanctuaries above the municipal capitals, in order to safeguard their water supplies.
URL http://www.watershared.net/?page_id=16
Description Development of national/regional/international development policies or investments
Geographic Reach National 
Policy Influence Type Participation in a advisory committee
Impact The results of the research immediately refined government policy in the five intervention municipalities, and fed into the development of a statewide PES law in Santa Cruz Department, an area 50% larger than the United Kingdom. Based on our results we have also helped other PES designers and implementers improve their schemes including the Peruvian and Colombian Ministries of the Environment as they develop and refine new small-scale PES programs internationally. We have also used our results to promote the small-scale, non-market, PES concept with the Bolivian government?s Climate Change Negotiating team, and helped Bolivia to introduce a new concept into the UNFCC Durban COP17 document ?Outcome of the work of the Ad Hoc Working Group on Long-term Cooperative Action under the Convention? ? that non market based approaches, such as joint mitigation and adaptation approaches for the integral and sustainable management of forests as a non-market alternative that supports and strengthens governance, ? could be developed.
Description We used the results of our Award to provide technical support to the Municipal Government of Santa Cruz de la Sierra for the development of a law for the protection of its water sources
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Participation in a advisory committee
Description We used the results of our Award to provide technical support to the Municipal Government of Tarija for the development of a law for the protection of its water sources
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Participation in a advisory committee
Description We used the results of our Award to provide technical support to the State Government of Santa Cruz for the development of a law for the protection of its water sources
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Gave evidence to a government review
Description Funding for new development activities or projects that utilise or have been informed by ESPA research : Grant of 800,000 GBP from the Interamerican Devlopment Bank
Amount $1,300,000 (USD)
Funding ID ATN/ME-15281-BO 
Organisation Inter-American Development Bank 
Sector Charity/Non Profit
Country United States
Start 05/2016 
End 05/2020
Description Private Investment
Amount $1,000,000 (USD)
Organisation The Coca-Cola Company 
Sector Private
Country United States
Start 04/2017 
End 03/2021
Title Baseline and endline socio-economic data from a Randomised Control Trial of the Watershared intervention in the Bolivian Andes 
Description ###This document explains the socio-economic data collected as part of a Randomised Control Trial investigating the effectiveness of a Payment for Watershed Services intervention known as Watershared (Acuerdos Reciprocos por Agua, or Reciprocal Water Agreements is the name in Spanish). ###The baseline data (2010) was collected under a grant from espa (Ecosystem Services for Poverty Alleviation) to Fundacion Natura Bolivia ìWhat types of investment can most cost-effectively ensure ecosystem service provision? A randomized program evaluationî (NE/I00436X/1) ###The endline data (2015/2016) was collected with funding from two projects: an Ecosystem Services for Poverty Alleviation funded project to Natura ìUnder what conditions can Payments for Environmental Services deliver sustainable improvements in welfare? Learning from a Randomized Control Trialî (NE/L001470/1) and a Leverhulme Trust funded research project to Bangor University ìCan payment for ecosystem services deliver environmental and livelihood benefitsî (RPG-2014-056). The latter project funded the cleaning and archiving of the dataset. ###The data consists of a) primary data and b) accompanying documentation. |--ReadMe.txt |-- data |-- HOUSEHOLD.xlsx |-- CHILD_DIARRHEA.xlsx |-- CONSERVATION_CONTRACT.xlsx |-- docs |-- METADATA_HOUSEHOLD.xlsx |-- METADATA_CHILD_DIARRHEA.xlsx |-- METADATA_CONSERVATION_CONTRACT.xlsx |-- CONSENT_ES.pdf |-- SURVEY_EN.xlsx |-- SURVEY_ES_ODK.xlsx #### (a) Primary data (within 'data' folder) - Primary data is arranged in three datasheets + *HOUSEHOLD.xlsx*: This contains all the information which was collected at the household level. Some variables were not collected in the endline; they were dropped following piloting as we found that they did not appear to work well in the field and respondents struggled to give reliable answers. These are not included in this datasheet. Some variables were added in the endline. Some data were processed to generate derived variables. The sheet therefore contains all variables which were collected in both the baseline and the endline, additional variables which were added in the endline and some additional calculated variables. 2623 households were surveyed in the baseline and there are entering here for all these households. In the endline we attempted to reach all2623 households but some had moved away, been disbanded due to the death of the household head, or could not be refound despite up to 5 attempts (1164 could not be re-found). The endline survey therefore includes only 1459 households. + *CHILD_DIARRHEA.xlsx* : This is a sub-table of the HOUSEHOLD table which provides information on each incidents of diarrhea in the last 12 months for each child under 16 years old belonging to each household. The identical data was collected in the baseline and endline. The children are linked to a specific household via the unique identifier ID_HOGAR. + *CONSERVATION_CONTRACT.xlsx* : This sheet provides information (from the endline only) on the management of a selected number of conservation contracts per household. It is structured according to the conservation agreements (each entry corresponds to one agreement). Each household in the HOUSEHOLD sheet (linked here through the unique identifier ID_HOGAR can have several contracts). This information was collected on a maximum of three contracts per household (if a household had more than three contracts, their level 1 contracts were prioritised). #### (b) Accompanying documents (within 'docs' folder) - Brief description of each document (name in parentheses) is provided below: + **METADATA_HOUSEHOLD.xlsx**: Metadata explaining each variable in the HOUSEHOLD datasheet. + **METADATA_CHILD_DIARRHEA.xlsx**: Metadata explaining each variable in the CHILD_DIARRHEA datasheet. + **METADATA_CONSERVATION_CONTRACT.xlsx**: Metadata explaining each variable in the CONSERVATION_CONTRACTS datasheet. + **CONSENT_ES.pdf**: The initial consent form used in the baseline. In the survey oral consent was gained before each survey began (all enumerators were trained to ensure they understood the importance of informed consent). + **SURVEY_EN.xlsx**: The endline survey (in English) showing the coding for each variable. NB Only those variables included in both baseline and endline survey are included in this archived dataset. + **SURVEY_ES_ODK.xlsx**: The survey was conducted in Spanish with enumerators using ODK (Open Data Kit) on android devices to enter the data. The initial Spanish language ODK code is provided here. ###The data was organised and prepared for archiving by David Crespo and Patrick Bottazzi (Bangor University) working with Julia P G Jones, Edwin Pynegar and Emma Wiik (Bangor University). ###The data collection was managed by Nature Foundation Bolivia (with input for the endline from Bangor University). 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Impact None so far 
URL http://reshare.ukdataservice.ac.uk/
Title Socioeconomic, water quality and biodiversity databases 
Description Socioeconomic, biological and water quality data for the 130 communities in the research area 
Type Of Material Database/Collection of data 
Provided To Others? No  
Impact We have researchers from various other institutions using the database for their own investigations. 
Description Watershared 
Organisation Nature and International Culture
Department Naturaleza y Cultura Internacional
Country Peru 
Sector Charity/Non Profit 
PI Contribution 185 municipalities across the Andes have now committed to learn lessons to expand incentive based conservation
Collaborator Contribution Facilitation, research and implementation
Impact A launch pad event at the Global Landscapes Forum at the Paris Climate Change Conference
Start Year 2015
Description Capacity strengthening events/workshops 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We have held a series of capacity building workshops in Bolivia, and Peru, to explain the research and its results to municipal and NGO staff.

After trying various different training methodologies, in 2014 and 2015 we held three iterations of what we now call the "School for Reciprocal Watershed Agreements (RWA)", training more than 60 representatives of NGOs and local governments from Bolivia, Belize, Costa Rica, Peru, Mexico, Colombia, Argentina and Ecuador. The teaching is organized into an introductory background and then five didactic modules:

• Step 1: Verify the presence of the basic conditions for Watershed Agreements (i.e. bottom-up PES)
• Step 2: Create the local institutional structure: a three way fund between the program promoter (usually an NGO) the municipal government and the local water provider
• Step 3: Ensure financial and institutional sustainability
• Step 4: Present and sign contracts with upstream water service providers
• Step 5: Monitor and evaluate results

The course is organized into a six-day program that combines classroom and practical field work, culminating with each student presenting how they will take the lessons they have learned during the week back to their own communities to initiate setting up their own RWA.

We have taken the results of the student's projects and have started testing three modalities of post-school support: (1) "Light" (i.e. very inexpensive) follow up: In Pasorapa, Omereque and Aiquille municipalities we have undertaken two personal individual follow up visits by Natura staff (2) "Medium" follow-up: In Tarija, we organized one joint meeting between representatives from six different municipalities, alumnae and Natura, and 3) "Intensive" (i.e. more expensive) follow up: in Robore and San Jose municipalities we have held a series of municipal/Natura meetings.

The municipalities assigned to each modality (Light, medium and intensive) are advancing at different rates in their attempts to implement RWA. Perhaps to be expected, we have advanced most rapidly in one of the "intensively" treated municipalities, San Jose. An alumna of the school prepared the ground for Natura to present to the Mayor and municipal council, leading to a Municipal Water Fund being created after less than five meetings and within 30 days. A process that has traditionally been a six-month investment by Natura staff, was thus accomplished in less than a month thanks to the RWA School.

Unexpectedly though, the alumna's role was not to complete the entire process herself, but rather to simply do the groundwork and to validate to local authorities that Natura, and the RWA concept are "for real". This initial success of a "paving-the-way" modality will help Natura improve the RWA School processes in the coming months. As we continue to further refine the modalities, it is already clear that the training program has the potential to exponentially increase the number of municipalities, beneficiaries and hectares conserved through the RWA model.

We have held a series of training workshops in Santa Cruz city and in the towns of Moro Moro, Vallegrande, Pucara, and in 135 small communities with the goal of explaining to the public, and especially to interested parties who have been involved in the r
Year(s) Of Engagement Activity 2013,2014,2015,2016,2017,2018
URL http://www.naturabolivia.org/en/reciprocal-water-agreements-school/
Description Presentations to national/regional/international panels or committees: series of presentations at Universities, NGOS and governments 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Type Of Presentation paper presentation
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Nigel Asquith, PI has made a series of presentations about the results of the research

We have made a series of presentations about the research and our results, to academic institutions including Yale University, and the Field Museum of Natural History in Chicago, to private grant makers such as the MacArthur Foundation, to governments suc
Year(s) Of Engagement Activity 2011,2012,2013,2014,2015,2016,2017,2018