Abstrakt
We conduct a meta-regression analysis of the empirical literature on frontier efficiency measurement in the banking sector of transitional countries to explain heterogeneity in estimated efficiency scores. Our sample includes 119 models taken from 14 papers estimating technical, cost, and profit efficiency scores.
The results suggest that the choice of methodology affects the reported estimates in a systematic way. The definition of bank outputs is important: cost efficiency scores are lower when the intermediation approach is used.
The differences between the cost and profit scores using parametric and non-parametric approaches arise when the Fourier-flexible functional form is employed; generally, the parametric approach yields higher scores than the non-parametric approaches. The removal of scale effects using equity capital has no significant impact on the score.