Performance improvement and performance dysfunction

Sinopse

The literature on the use of performance measurement in government has featured prominent attention to hypothesized unintended dysfunctional consequences such measurement may produce. We conceptualize these dysfunctional consequences as involving either effort substitution (reducing effort on non-measured performance dimensions) or gaming (making performance on the measured performance dimension appear better, when in fact it is not). In this paper, we examine both performance impacts and dysfunctional consequences of establishment in the British National Health Service of a performance target that no patient presenting in a hospital accident and emergency department (emergency room) wait more than four hours for treatment. Using data from all 155 hospitals in England, we find dramatic wait-time performance improvements between 2003 and 2006, and no evidence for any of the dysfunctional effects that have been hypothesized in connection with this target. We conclude by discussing when one would expect dysfunctional effects to appear and when not.