Statistical Estimation of Technical and Allocative Inefficiency Using DEA Rajiv D. Banker*, Hsihui Chang* and Ram Natarajan** *University of California, Riverside, **University of Texas at Dallas Abstract We present a new method to estimating technical and allocative inefficiency using Data Envelopment Analysis (DEA). We show that the DEA technical inefficiency measure using total revenues as the single output variable equals the aggregate technical and allocative inefficiency. We employ this result to estimate allocative inefficiency and construct statistical tests of the null hypothesis of no allocative inefficiency analogous to those of the null hypothesis of no scale inefficiency. We illustrate our method using revenue and personnel data for the top U.S. public accounting firms over 1995-98. Our empirical results indicate the existence of statistically significant allocative inefficiency in the public accounting industry. Key words: Data envelopment analysis, Technical inefficiency, Allocative inefficiency, Public accounting 1