Data challenges in the measurement of agricultural productivity: Lessons from Chile

Productivity measurement and analysis have motivated considerable theoretical and empirical work in recent decades. Models that have enjoyed noticeable expansion are stochastic production frontiers for panel data. These models have proven very useful in total factor productivity (TFP) measurement and the analyses of its components. However, the related empirical literature in Latin America and the Caribbean has been limited, and a likely reason for this gap is data constraints. This article examines the setting surrounding the measurement and analysis of productivity in the Chilean agricultural sector. The specific objectives are to (1) provide a summary of key agricultural productivity measures and recent associated methodological advances; (2) present an overview of micro studies reporting technical efficiency and TFP in Chile; (3) portray the major sources of agricultural data available in the country; and (4) discuss salient features of the agricultural data systems used in Australia and the United States. The paper ends by identifying challenges and possible improvements to the prevailing data system that could strengthen the measurements and monitoring of productivity in Chile. The analysis suggests that the country needs substantial improvements in the collection and analysis of agricultural statistics to develop TFP and related research. This line of work is a critical step to enhance competitiveness and to foster adaptations to climate change, as well as to fully participate in efforts sponsored by the IFAD, FAO and the OECD to monitor progress toward the SDGs. On the positive side, several avenues are available to move toward a more robust agricultural statistical architecture.