Agglomeration payments for catchment conservation and improved livelihoods in Malawi
Lead Research Organisation:
International Food Policy Research Institute
Department Name: Env and Production Technology Division
Abstract
A popular retail technique, the 'Groupon', is a system in which returns to an individual consumer are enhanced if he or she can convince others to participate as well. Analogs to the Groupon are possible in land management, where bonus payments based on the participation of neighbors can be employed to achieve contiguity of land use, prevention of land degradation, enhancement of biodiversity and other ecological services. Such payments - termed 'agglomeration payments' in the ecological economics literature - may also offset some program costs by reducing moral hazard and encouraging sustained adoption. This study applies agglomeration payments as part of an encouragement design for land conservation practices in Malawi's Shire Valley basin. In partnership with the Malawi Department of Land Resources and Conservation (DLRC) and the National Smallholder Farmers' Association of Malawi (NASFAM), our research will evaluate the impacts of agglomeration payments on the adoption of agricultural conservation technologies being promoted currently by the Government of Malawi, and the positive externalities for the Shire Valley basin that may accrue from the resulting spatial contiguity of adopting farms.
The adoption of sustainable agricultural practices in this area is still modest, making it challenging to evaluate impacts. Our two-pronged research strategy includes first a pilot study with a 4-treatment encouragement design to evaluate strategies for improving adoption of conservation agricultural (CA) technologies under DLRC-led programs in the Shire Valley. The treatments will compare the roles of extension services, conventional payments, and agglomeration payments in encouraging adoption of sustainable agricultural practices such as CA.
Second, we propose to develop an agent-based model (ABM) of the Shire Valley basin system to evaluate consequences of improved adoption of sustainable agricultural practices for the enhanced provision of ecosystem services such as improved water quality and runoff regulation, or increased natural predator and pollination services. Agent-based models treat actors in the system (such as farmers) as individual agents whose decisions and interactions lead to emergent landscape-level outcomes such as land cover, water quality or ecosystem-level impacts. Data on social interactions and decision making from our pilot study will inform this regional-scale ABM which, coupled to soil-water assessment models already developed for Sub-Saharan Africa and to literature models for provision of predator and pollination services, will allow assessment of the landscape-scale consequences of the different incentives evaluated in the pilot study.
A challenge in evaluating impacts from projects focused on sustainable agricultural practices is that while some impacts (such as reduced costs and labor) accrue rapidly, others (such as shifts in yields or water quality) may take years of consistent CA implementation to emerge. The design proposed here overcomes this limitation by combining field data collection with modeling, aiming to address three key questions:
Q1) How do agglomeration payments shift interactions among farmers, as well as rates/patterns of adoption of practices such as CA?
Q2) Can agglomeration payments lead to enhanced landscape-scale ecosystem service provision?
Q3) Do agglomeration payments facilitate cost-effective ecosystem service provision, relative to conventional incentives?
The adoption of sustainable agricultural practices in this area is still modest, making it challenging to evaluate impacts. Our two-pronged research strategy includes first a pilot study with a 4-treatment encouragement design to evaluate strategies for improving adoption of conservation agricultural (CA) technologies under DLRC-led programs in the Shire Valley. The treatments will compare the roles of extension services, conventional payments, and agglomeration payments in encouraging adoption of sustainable agricultural practices such as CA.
Second, we propose to develop an agent-based model (ABM) of the Shire Valley basin system to evaluate consequences of improved adoption of sustainable agricultural practices for the enhanced provision of ecosystem services such as improved water quality and runoff regulation, or increased natural predator and pollination services. Agent-based models treat actors in the system (such as farmers) as individual agents whose decisions and interactions lead to emergent landscape-level outcomes such as land cover, water quality or ecosystem-level impacts. Data on social interactions and decision making from our pilot study will inform this regional-scale ABM which, coupled to soil-water assessment models already developed for Sub-Saharan Africa and to literature models for provision of predator and pollination services, will allow assessment of the landscape-scale consequences of the different incentives evaluated in the pilot study.
A challenge in evaluating impacts from projects focused on sustainable agricultural practices is that while some impacts (such as reduced costs and labor) accrue rapidly, others (such as shifts in yields or water quality) may take years of consistent CA implementation to emerge. The design proposed here overcomes this limitation by combining field data collection with modeling, aiming to address three key questions:
Q1) How do agglomeration payments shift interactions among farmers, as well as rates/patterns of adoption of practices such as CA?
Q2) Can agglomeration payments lead to enhanced landscape-scale ecosystem service provision?
Q3) Do agglomeration payments facilitate cost-effective ecosystem service provision, relative to conventional incentives?
Planned Impact
A central premise of this work is that careful engagement of policy stakeholders will lead to uptake of the research products, and an influence on the structure of agricultural programs in Malawi. Though our direct role ends at uptake, it is implied that such influence will lead to improved livelihoods in rural communities in Malawi through the provision of well-designed incentives that encourage ecosystem service-enhancing land-use practices. The ASWAp program through within which we are operating has itself 19000 intended beneficiaries - we hope that the results of our pilot research within ASWAp may have reach at this scale in future programs. Beyond this key set of direct stakeholder impacts, we anticipate a range of other broader impacts from the project.
First, while the project is not aimed at providing studentships, our team at Lilongwe University will recruit current graduate students to the enumerator team, possibly providing opportunities for publication and thesis development, but more directly building skills in field methods, randomized control trials, and impact evaluation among the next generation of Malawian agricultural research professionals.
Second, this project will be an important case study showing a route to which landscape-wide impacts can arise from individual choices and therefore an important way into protecting "the commons". Thus, there is another class of indirect user - the global community of people with an interest in encouraging sustainability (sensu stricto). Through publication in the academic and grey literature, we expect our work to build upon the global understanding of harmonizing livelihood provision with ecosystem function.
First, while the project is not aimed at providing studentships, our team at Lilongwe University will recruit current graduate students to the enumerator team, possibly providing opportunities for publication and thesis development, but more directly building skills in field methods, randomized control trials, and impact evaluation among the next generation of Malawian agricultural research professionals.
Second, this project will be an important case study showing a route to which landscape-wide impacts can arise from individual choices and therefore an important way into protecting "the commons". Thus, there is another class of indirect user - the global community of people with an interest in encouraging sustainability (sensu stricto). Through publication in the academic and grey literature, we expect our work to build upon the global understanding of harmonizing livelihood provision with ecosystem function.
Publications
Bell A
(2016)
Opportunities for improved promotion of ecosystem services in agriculture under the Water-Energy-Food Nexus
in Journal of Environmental Studies and Sciences
Bell A
(2018)
Transformative change through Payments for Ecosystem Services (PES): a conceptual framework and application to conservation agriculture in Malawi
in Global Sustainability
Bell A
(2016)
Scaling up pro-environmental agricultural practice using agglomeration payments: Proof of concept from an agent-based model
in Ecological Economics
Bell A
(2018)
Do As They Did: Peer Effects Explain Adoption of Conservation Agriculture in Malawi
in Water
Bell AR
(2018)
Smart subsidies for catchment conservation in Malawi.
in Scientific data
Ward P
(2016)
Heterogeneous preferences and the effects of incentives in promoting conservation agriculture in Malawi
in Agriculture, Ecosystems & Environment
Description | Our baseline survey included an experiment (a discrete choice experiment) to evaluate farmers' willingness to adopt conservation agricultural practices in response to different incentives. We used the results of this experiment to i) set appropriate incentive levels for our pilot and ii) build a short research paper explaining what appear to be key factors that quantifiably shape farmers' perceptions of conservation agriculture and their willingness to adopt it as a practice. Key findings from our first year of intervention are that i) agglomeration payments do influence the willingness to mulch crop residues and ii) the adoption of CA appears to be at least a 2-part decision of adopting mulching+zero tillage, and separately adopting intercropping. We observed a doubling of program participation (from 1450 to 2870) from year 1 to year 2; analysis has not yet begun on compliance among registrants and patterns in that compliance. Modeling findings identify clear cost-effectiveness gains at scale for agglomeration payments. On simple basis of change in land area under CA in treatment vs control villages, and direct costs in the form of vouchers, marginal cost of reducing sediment via land practices appears to be on order of ~$7 / ton, given current conditions and levels of adoption |
Exploitation Route | These have been published/are in review, and we are beginning to conduct workshops, conference presentations, etc. to promote them. |
Sectors | Environment Government Democracy and Justice |
Description | informed in part on our findings, three components were included in the next major Shire River Basin development programme, titled MWASIP: 1) A pilot payments for watershed services program 2) a research trial for group incentives in payments programs 3) a biophysical and socioeconomic monitoring program |
First Year Of Impact | 2020 |
Sector | Agriculture, Food and Drink,Environment |
Impact Types | Societal Economic |
Description | BASIS AMA CRSP |
Amount | $800,000 (USD) |
Organisation | United States Agency for International Development |
Sector | Public |
Country | United States |
Start | 08/2014 |
End | 09/2017 |
Title | Ethnographic decision tree modeling dataset |
Description | Interview transcripts of 96 participants, identification of 26 criteria involved in the decision to adopt conservation agriculture, and assessment of these 26 criteria within our Endline study (1800 respondentS) |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | First applied in a machine-learning analysis of farmer CA decision-making, published in Water, 2018 |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Title | Farm Adoption Model |
Description | An agent-based model of agricultural adoption, employing an expected utility decision model for risk-averse, boundedly rational, future-discounting farmers. |
Type Of Material | Computer model/algorithm |
Provided To Others? | No |
Impact | Nothing yet. \ We were in contact for potential integration and crossover modeling with the ESPA funded ASSETS project model, developed for the same region, but the linkage has not materialized up to now. |
Title | Voucher redemption dataset |
Description | Complete set of voucher values, registration of redemption, and log of items purchased for intervention Years 1 (1450 participants) and 2 (2800 participants) |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | No |
Impact | Outputs in preparation as of March 2017 |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Title | Year 0 Baseline study |
Description | Clustered random sample of households within villages (60 villages with 30 hh per). |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | This dataset so far has been used to generate one journal article in Agriculture Ecosystems and Environment (2016), and contributed data toward one masters' thesis |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Title | Year 1 Registration and monitoring follow-up dataset |
Description | Dataset of registrants in our intervention across 48 treatment villages, and follow-up visits to check on compliance |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | This dataset has currently led to one journal article in Land Use Policy, 2018, on the nature of early adoption under our program |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Title | Year 2 Endline dataset |
Description | Agricultural household survey of 1800 respondents, panel study to complement baseline dataset |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | Applied to a publication in Water, 2018, and an article for the Journal of Global Sustainability (resubmitted as of March 2018). |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Title | Year 2 Intervention Registration and Monitoring Dataset |
Description | Detailed plot-level data collection, plot boundaries, and compliance measure for 2800 households in Year 2 of our intervention |
Type Of Material | Database/Collection of data |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | Outputs in preparation as of March 2018 |
URL | https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GTSEWP |
Description | Participation at the 16th National Conference and Global Forum on Science, Policy and the Environment (NCSE) on the Food-Energy-Water Nexus in Washington DC during January 19-21 |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Other audiences |
Results and Impact | The session at the NCSE conference was titled "Ecosystem Services for Nexus Solutions". The session abstract is pasted below. The session was linked to a special issue paper in JESS (Opportunities for improved promotion of ecosystem services in agriculture under the Water-Energy-Food Nexus) co-authored by Andrew Bell, Nate Matthews and Wei Zhang who co-organized the session. The paper is listed under EPSA outputs. The water-energy-food nexus is one important, but as of yet insufficiently explored entry point to improve ecosystem services for people and the environment. The session seeks to explore how the nexus approach can help to secure the sustainable provision and equitable distribution of ecosystem services in various agricultural landscapes across the globe. It aims to examine how increased resource use efficiency through solutions in the water, energy and food sectors ensure that ecosystem services will continue to generate services for future generations and particularly for the poor. Moderator: Nathanial Matthews, CGIAR Program on Water, Land and Ecosystems and International Water Management Institute Speakers: ?Andrew Bell, Assistant Professor of Environmental Studies, New York University* ?Anthony Janetos, Director, Frederick S. Pardee Center for the Study of the Longer-Range Future and Professor, Earth and Environment, Boston University ?Tracy Rouleau, Deputy Chief Economist, Office of Program Planning and Integration, National Oceanic and Atmospheric Administration ?Wei Zhang, Research Fellow, International Food Policy Research Institute |
Year(s) Of Engagement Activity | 2016 |
URL | http://foodenergywaternexus.org/s-d3-ecosystem-services-for-nexus-solutions/ |
Description | Participation in ESP Africa Annual Meeting 2016 (Nairobi, Kenya) |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Other audiences |
Results and Impact | Two talks given in two separate panels as part of ESP conference. ~30 attendees in each session. |
Year(s) Of Engagement Activity | 2016 |
Description | Project planning workshop Summer 2015 |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Professional Practitioners |
Results and Impact | 15 project members and stakeholders met to discuss current findings and improve planning for 2015-2016 project year. The group identified 'data collection and management' as a key topic for planned capacity building workshops in 2016 |
Year(s) Of Engagement Activity | 2015 |
Description | SSCCM End-of-project Workshop |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | We convened regional experts in Conservation Agriculture (from both social sciences and natural sciences fields), as well as local NGOs (e.g., Total Land Care) and other Shire Basin stakeholders (ESCOM, USAID, World Bank, Ministry of Agriculture) for a 2-day workshop synthesizing our project work with other efforts in the region and identifying next steps. |
Year(s) Of Engagement Activity | 2018 |
URL | http://massp.ifpri.info/2018/01/31/why-is-adoption-of-conservation-agriculture-still-a-hard-sell-in-... |
Description | Training workshop on social data collection in rural areas |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Professional Practitioners |
Results and Impact | We conducted two three-day workshops in Lilongwe and Blantyre on improved rural data collection. The workshops included curricula on i) impact evaluation, ii) data analysis in Excel, iii) data collection using Android ODK, iv) spatial analysis using QGIS, and v) choice modeling using R. We had ~100 participants across both workshops, with participants from universities, NGOs, government, and the private sector. |
Year(s) Of Engagement Activity | 2016 |