CAP Reforms and Total Factor Productivity Growth: An Analysis of Belgian Farm-level Data

Have the 1992 and 2000 CAP reforms had any discernable effect upon agricultural productivity? In this study we derive detailed information on the total factor productivity (TFP) growth of arable farms in Belgium over a 16-year period from 1987 to 2002. Calculations are based on a carefully constructed high-quality detailed farm-level data set containing 1728 observations, involving over 100 farms in most years. Three output variables (cereals, other crops, other outputs) and four input variables (land, labour, capital and other inputs) are constructed, using multilateral Fisher index numbers where crop aggregation is required. The TFP measures are calculated using a Malmquist TFP index relative to a series of data envelopment analysis (DEA) frontiers. Annual average TFP change of 1.0% per year is estimated for this industry. This rate does not compare well with rates of over 2% that are commonly reported in studies involving other developed countries. The pattern of TFP growth over the period indicates that the two CAP reforms (in 1992 and 2000) have had no discernable effect upon TFP trends. An inspection of shadow shares (derived from the DEA frontiers) indicates that substantial distortions remain in this industry, especially with regards to the excess use of labour and constrained use of land, relative to other inputs. Finally, tabulation of results by region and farm size shows fairly uniform TFP change across regions, but poor performance in small farms, where TFP actually levels actually fell over this period.