The impact of labelled seeds on the technical efficiency of rice farming in Indonesia

One of the governmental strategies to enhance the national rice production involves improving farmers' access to high-quality seeds through subsidies and the distribution of labelled rice seeds. This research aims to assess rice cultivation's technical efficiency using labelled seeds compared to those without. Earlier investigations into the impact of labelled seeds on rice farming performance did not account for potential biases carried on by variables that are both observed and unobserved variables. This study utilized household-level data from 14 districts situated in the rice production centre of Indonesia. Propensity score matching was employed to mitigate observed variable biases, while selectivity-corrected stochastic production frontier models were applied to address biases associated with unobserved variables. The findings revealed that rice farming with labelled seeds demonstrated greater technical efficiency than farming with unlabelled seeds. Meanwhile, if no correction was made to the observed and unobserved variables, the TE of the rice farming with labelled seeds was not statistically different from the unlabelled seeds. The policy implication underscores the necessity for public policies to ensure the quality of labelled seeds used by farmers and to encourage adherence to the prescribed cultivation practices for labelled seeds.