Dairy farm technical efficiency and milk total factor productivity growth in three South American countries

Date of Completion

January 2006


Economics, Agricultural




The decomposition of productivity growth is important because technical efficiency can be interpreted as a relative measure of managerial ability given technology, while technological change leads to increases in productivity that arise from the development and subsequent adoption of new production practices. This dissertation contributes to the understanding of productivity growth in dairy production in three Southern Cone countries: Argentina, Chile and Uruguay.^ A meta-regression analysis was developed from a total of 65 frontier studies that report technical efficiency measures at the dairy farm level, using studies from the literature in English and Spanish. The effect of methodology on technical efficiency measures as well as the lack of data independence due to the presence of multiple observations is analyzed. Thus, to examine these effects, eight alternative models were estimated. Several tests indicate that the two most robust models are a fixed effects specification with dummy variables for the most significant studies without geographical effects, and a specification where the multiple observations are averaged and that includes geographical effects.^ The main results from the Stochastic Production Frontier analysis indicate that, for Argentina, Chile and Uruguay, the translog is the most appropriate functional form, the inefficiency effects display a half-normal distribution, technical efficiency is time-invariant, and technical inefficiency is statistically significant. All models exhibit increasing returns to size. The computed Total Factor Productivity Growth rates per year are -0.62% for Argentina, 1.81% for Chile, and 1.93% for Uruguay.^ The null hypothesis that the three countries share the same production frontier is strongly rejected, which implies that the meta-frontier approach is the appropriate framework to estimate and compare technical efficiency across the sample of dairy farms from the three countries. The highest average Technological Gap Ratio is for Uruguay (91.4%), and the lowest is for Chile (79.6%) while Argentina exhibits an intermediate value (83.8%). The average technical efficiency estimates from the preferred country specific Stochastic Production Frontier models are 87.0%, 84.9% and 81.1% for Argentina, Chile and Uruguay, respectively; however, these technical efficiency measures cannot be compared directly with each other. To make valid technical efficiency comparisons across the three countries it is necessary to do so with respect to the meta-frontier. In this case, the technical efficiency averages are substantially lower and take the following values: Argentina 72.8%; Chile 65.8%, and Uruguay 73.4%. ^