With tropical deforestation a major contributor to greenhouse gas emissions and biodiversity loss, the land use decisions of small-scale farmers at the forest margins have important implications for the global environment. In some tropical forests, such as the Eastern Brazilian Amazon, farmers practice a shifting cultivation system that maintains large amounts of land under forest fallow. I examine whether local benefits of fallowing such as soil restoration, erosion mitigation and hydrological regulation are of sufficient value to farmers to stem the expansion of permanent cropland at the expense of forest.

I quantify the value of ecosystem services provided by fallow to agriculture and test whether local forest externalities are economically significant, using farm survey and GIS data from the Eastern Amazon. I estimate a production function to determine the contribution of on-farm and upstream fallow to income, using an instrumental variables approach to address endogeneity. I find that on-farm and upstream fallow are both associated with higher farm income. This result both confirms the agronomic evidence that fallow boosts yields and suggests that fallow provides positive hydrological externalities to downstream farms.

I also examine whether farmers respond strategically to their neighbors' land use, taking advantage of ecosystem services provided by upstream farms. I use a spatial econometric model to estimate the effect of upstream farms' fallow on downstream land allocation. I find no evidence that farmers alter their fallowing based on land use upstream.

I then investigate whether market failures encourage fallowing. If farmers cannot purchase inputs used in cultivation due to liquidity constraints, they may keep more land under fallow than optimal. I use the estimated production function parameters to determine whether each farm's allocation of land between cropping and fallow is efficient from an individual perspective. I then estimate the effect liquidity indicators on land use efficiency. I find that over-fallowing is negatively associated with commercial credit use and off-farm income, suggesting that liquidity constraints do hinder agricultural intensification. Because I find evidence to support the existence of positive externalities to fallow, the loosening of liquidity constraints that encourage fallowing has ambiguous implications for community-level welfare.