Analysis of trends in technical and allocative efficiency: An application to Texas public school districts

We present a new DEA based method to analyze efficiency trends over time and differences across subgroups in a panel data setting. We employ a result that the aggregate technical and allocative inefficiency score equals the technical inefficiency when input qualities are aggregated into a single total input cost variable, and develop test procedures to evaluate the presence of allocative inefficiency. We apply these methods to test for the presence of allocative inefficiency in Texas school districts over 1993–99, and analyze shifts and trends in both technical and allocative inefficiencies over time for different regions. Our empirical results indicate the existence of statistically significant allocative inefficiencies. While technical inefficiency increased over the six year sample period, allocative inefficiency remained relatively stable during this period. These results for the full sample obtain also when we repeat the analysis for different regions.