The roles of risk and ambiguity in the adoption of the system of rice intensification (SRI): evidence from Indonesia

Given the recognized yield-enhancing potential of the System of Rice Intensification (SRI), its low adoption and high discontinuance rates in some locales are puzzling. Combining experimental measures of risk and ambiguity aversion with household-level and plot-level panel data collected in rural Indonesia, this study empirically explores factors shaping SRI's adoption and discontinuance. Employing multivariate and Heckman probit models to control unobserved heterogeneities, we find that farmers' risk aversion significantly reduces their likelihood of using all individual SRI practices. However, once the effects of risk aversion on the use of SRI in the previous year are statistically controlled, risk aversion does not significantly explain farmers' subsequent decisions to continue or discontinue SRI practices. Farmers' ambiguity preferences play no significant role in decisions to use most practices, except alternate wetting and drying, which requires proper coordination of irrigation among neighboring farmers and thus amplifies the uncertainty of effective implementation. The results also show that access to irrigation is a significant factor in the use of SRI and its continuance. Moreover, as SRI requires greater input of labor and therefore curtails time for alternative household activities, including off-farm work, family composition is a significant factor determining its adoption and continuing use. Although these findings are not necessarily generalizable, our study expands the existing knowledge of factors underlying SRI's slow diffusion.