The progressive liberalization of agricultural markets, along with the threat that imported products can pose to local producers, reveals the importance of productivity growth as a mechanism to improve competitiveness. Technical efficiency measurement is the most studied component of productivity because it can help to generate valuable information for policy formulation and farm level decisions focused on the improvement of farm performance. This study uses unbalanced panel data sets for dairy farms from Argentina, Chile and Uruguay, to estimate stochastic production frontier models. These frontiers are then used to estimate economies of size, technological change and technical efficiency. All estimations are based on the Battese and Coelli (1992) model, which is widely used in empirical productivity studies. The models for all three countries exhibit increasing returns to scale, which suggests that the dairy farms in the samples are operating at a sub-optimal size. The average annual rate of technological change for Argentina was 0.9%, for Chile 2.6% and for Uruguay 6.9%, while average technical efficiency was 87.0%, 84.9% and 81.1%, respectively.