Impact of agricultural productivity changes on poverty in developing countries
MetadataShow full item record
Agricultural productivity growth has long been recognized as pro-poor and as a crucial determinant of poverty reduction, but empirical estimates of this relationship are still limited. Earlier studies primarily focus on partial productivity measures, such as land and labor productivity, to explain poverty reduction. A few recent studies use frontier based total factor productivity (TFP) measures while examining its impact on poverty. The frontier based TFP measures are calculated using distance functions and are relative to the most efficient country’s TFP. Theory on agricultural productivity growth, however, emphasizes the impact of productivity growth within a country over time on poverty in that country. This study compares the impact of single factor productivity growth, as well as frontier and non-frontier TFP growth estimates, on poverty reduction in developing countries. We estimate multiple measures of agricultural total factor productivity growth employing frontier approach for 108 developing countries. We then make alternative groupings of countries to allow for the possibility of different production frontiers for countries with different income level and countries based on different regions. We compare these various measures of agricultural TFP with TFP measures obtained by Fuglie (2011) using a non-frontier growth accounting approach. Results from the TFP analysis show that TFP change estimates by income groups differ from those estimated using all countries in a pooled model. This indicates that agricultural technology and production frontiers may differ across countries based on income levels. For most of the countries, TFP measures from the pooled model, from income groups and groups based on regions are found to be notably different from those obtained using growth accounting approach. We then use these various measures of agricultural productivity growth in a poverty model. Using the Two-Step System Generalized Method of Moments estimation technique in a dynamic panel data framework, we find that single productivity measures as well as TFP measures based on growth accounting approach are significantly poverty reducing. The point estimate of growth accounting TFP growth is found to be higher than that of land productivity, but lower than that of labor productivity on poverty reduction. Most of the frontier based TFP measures are found to be both ambiguous in sign and weaker in the sense that they are not significant.