Landscape, welfare, and distributional trade-offs from smallholder agroforestry contracting: An agent-based model approach

Agroforestry value chains have been extensively promoted as potential ‘win-win’ vehicles for Sustainable Intensification in tropical upland landscapes. Coffee contract farming in particular can support landscape environmental policy objectives through improvements in soil erosion and biomass carbon, while also providing smallholders with higher agricultural profits. This paper explores the extent in which coffee agroforestry contract farming influences household welfare and landscape environmental outcomes using a case study of the peri-urban uplands of Bandung, Indonesia. We develop an agent-based simulation model to test the welfare and landscape effects of coffee contracting interventions. The model incorporates household survey data with granular land cover and soil maps. Smallholder farmer agents interact with coffee value chains and other agricultural and labour markets, and make land use decisions that have landscape-level environmental and household welfare implications. The modelling simulations show that coffee contracting can achieve improved incomes for farmers, and increased agroforestry landcover. However, transaction costs can impede many households from accessing contract coffee markets, resulting in trade-offs between higher incomes and increasing rural income inequality. Contract coffee displaces lower-input agroforestry where opportunity costs are low, resulting in significant landscape declines in biomass carbon and increases in soil erosion – an outcome akin to a Jevon's Paradox. Agroforestry value chain interventions deliver trade-offs when seeking to address multiple and complex landscape challenges. Complementary non-market interventions that better align smallholder incentives with environmental policy objectives are necessary for these value chain interventions to be effective tools for Sustainable Intensification. • We present an agent-based model of coffee contract farming for Bandung, Indonesia. • Agroforestry value chains are often promoted to incentivise sustainable land uses. • The simulations highlight complex economic-equity-environmental trade-offs. • Transaction costs impede households with less land from contracting. • Other agroforestry land uses are displaced, leading to worse environmental outcomes.