Effects of information and seedling provision on tree planting and survival in smallholder oil palm plantations

Abstract: Oil palm expansion in Indonesia is associated with a reduction in biodiversity and ecosystem services, as well as livelihood improvements for smallholder farmers. While this dichotomy highlights the importance of sustainable management options, empirical evidence on which policies are effective in stimulating biodiversity-friendly plantation management is relatively scarce. This paper addresses this gap by presenting results from a randomized controlled trial implemented in Jambi province, Sumatra, in 2016. We focus on native tree planting in oil palm plantations as one sustainable management option. To test whether information and input provision affect the number of trees planted by smallholders two treatments were designed: the first provided information about tree planting in oil palm, while the second combined information and seedling delivery. We model adoption in a double-hurdle framework where farmers first decide whether to adopt or not and then how many trees they plant per hectare. Our results suggest that both interventions are effective in stimulating tree planting in oil palm. Seedling provision in combination with information leads to a higher probability of adoption, but farmers plant on average relatively few trees per hectare. In contrast, in the informational treatment, few farmers adopt, but they plant more trees per hectare than farmers who received seedlings. Furthermore, we observe that the survival rate of trees planted is lower for farmers who received seedlings in comparison to farmers who only received information. Since we cannot find evidence for farmer and plot selection effects, it is likely that species choice and seedling quality are the underlying drivers of this difference. (JEL Codes: Q12, Q16, Q57, D04, C9)