Description:
This paper examines the agricultural productivity-farm size relationship in the context of Bangladesh. Features of Bangladesh's agriculture help overcome several limitations in testing the inverse farm size-productivity relationship in other developing country settings. A stochastic production frontier model is applied using data from three rounds of a household panel survey to estimate simultaneously the production frontier and the technical inefficiency functions. The "correlated random effects" approach is used to control for unobserved heterogeneous household effects. Methodologically, the results suggest that the stochastic production frontier models that ignore the inefficiency function are likely mis-specified, and may result in misleading conclusions on the farm size-productivity relationship. Empirically, the findings confirm that the farm size and productivity relationship is negative, but with the inverse relationship diminishing over time. Total factor productivity growth, driven by technical change, is found to have been robust across the sample. Across farm size groups, the relatively larger farmers experienced faster technical change, which helped them to catch up and narrow the productivity gap with the smaller farmers