Abstract
Data envelopment analysis (DEA) has the potential to make strong contributions to the formulation of public policy. In particular, DEA can be used to identify the determinants of efficiency and this information can be used to inform debate around municipal boundary reform and other matters. However, there is a dearth of empirical literature on the accurate and correct specification of DEA. We use conceptual considerations to posit an ideal specification for Australian municipal DEA before estimating a number of models to demonstrate the effect of specification on the identification of determinants. Our evidence suggests that incorrect specification may well produce spurious associations and lead to poor public policymaking.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
