Abstract:
The Farrell (1957) concept of technical efficiency can be measured in two ways--denoted output and input technical efficiency. The empirical literature does not distinguish between the two measures and confusion exists regarding the required restrictions on technology for the specification of either measure. With panel data, the authors show how both types of efficiency, when treated as fixed effects, are to be specified in the context of a translog cost system. Technical efficiency enters the share equations only in the output case when homotheticity is not imposed. The authors estimate their cost systems using panel data on U.S. airlines. Copyright 1994 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association.