Survey of the empirical evidence regarding performance measures in the public sector.
Discusses the features of the public sector that make the use of performance measures problematic – in particular:
Multiple tasks – bureaucrats must often achieve several ends, e.g. increasing both efficiency and equity in the delivery of public services.
- Multiple principals – Bureaucrats must also often serve several masters, e.g. politicians, users of services and professional organisations.
Outlines the theoretical rationale for ‘performance management’ schemes, and the instruments which they may use, including pecuniary (financially rewarding well-performing agents) and non-pecuniary (‘name and shame’ policies for failing agents) rewards.
Describes the many forms in which performance measures may come – from in-depth evaluations (e.g. HMI Prison reports) to those derived from administrative data (e.g. exam pass rates for schools).
The bulk of the paper collects the empirical evidence on performance measures in the public sector, drawing 3 main conclusions: 1. Gross outcomes or levels based performance measures do not provide a sufficiently accurate picture of the relative performance of public sector organisations. 2. A single performance measure is not sufficient, owing to the multiple aims and stakeholders of public sector organisations. Instead a range of measures should be employed. 3. The intended purpose for each measure should dictate its form and whether it is published. If the aim is to improve performance within an organisation then it does not necessarily need to be published. If the aim is to facilitate a market, publication is required.
Concludes by recommending wider piloting of rigorously evaluated performance management schemes, and suggests the possibility of using independent information sources
