Local deforestation spillovers induced by forest moratoria: Evidence from Indonesia

Moratoria on commodities produced in deforestation-risk areas have been shown to be highly effective in reducing deforestation within targeted areas. Various studies have shown, however, that such policies are prone to large local spillover effects, i.e., non-trivial changes (reductions or increases) in the amount of deforestation in areas just outside the direct scope of the moratorium. Little is known about the direction and magnitude of local spillover effects that may have been induced by the Indonesian forest moratorium, an anti-deforestation policy enacted in 2011 that covers around a third of Indonesia's terrestrial area and that is of high importance in meeting international deforestation goals. Here, we empirically assess the evidence of spillover effects near the Indonesian moratorium boundaries, using several proximity metrics and a panel dataset spanning the years 2001–2018. Based on our negative binomial fixed effects regressions, we estimate that the moratorium induced 1324 km2 of deforestation in areas located within 10 km of the targeted areas in the period 2011–2018, most of which occurred near conservation and protection forests. Evidence of spillover effects is also strong within concession areas slated for development. This suggests that companies may have shifted their planned production activities from areas targeted by the moratorium to neighbouring concession areas, resulting in additional forest loss. To minimize or halt such spillover effects, the scope of the Indonesian moratorium could be expanded to high-deforestation risk areas, such as forest areas outside mountainous regions, with relatively high GDP per capita and high agro-ecological suitability for oil palm plantations. In addition, a higher uptake of certification schemes and increased international finance would complement the moratorium, helping to reduce incentives to deforest both within and outside the moratorium areas.