Abstract
This dissertation evaluates the extent to which alternative methods of estimation vary from one another in measuring technical efficiency (TE). Using data from the New Zealand dairy industry for the year 1993, this paper calculates farm-specific technical efficiency estimates and mean technical efficiency estimates for each estimation method. The methods adopted include the stochastic production frontier (STOC), corrected ordinary least squares regression (COLS) and Data Envelopment Analysis (DEA). The results derived show that the mean technical efficiency of an industry is sensitive to the choice of the production frontier method. In general, the STOC and DEA frontiers resulted in higher mean technical efficiency estimates compared to the COLS production frontier. The resulting mean TE estimate from the STOC production frontier was significantly higher than that of DEA, except under the variable returns to scale DEA model. The results from the DEA and STOC frontiers also indicate that New Zealand dairy farmers are operating nearer to or at the efficiency frontier. All three methods are consistent in ranking individual production units in terms of technical efficiency.