Introduction

Policy makers and others often wish to compare the performance of public organisations, agencies and programmes and Chapter 5 discussed principles that can underpin such comparisons. In particular, it argued that public bodies often have multiple goals and serve disparate groups of clients and that any comparison should reflect these realities. Chapter 5 described several approaches that are commonly used in such comparison, including benchmarking, and the use of rates and ratios to allow fair comparisons when there are structural differences between the bodies being compared. Extending the latter, it also introduced some of the basic concepts of data envelopment analysis (DEA). This chapter provides a more thorough coverage of DEA and includes some case studies of its use. Its technical level is higher than Chapter 5, and is the most demanding of any in this book, though the general argument is at a level that most interested readers should be able to follow if they can cope with some algebra.

In public services, the usual aim is to transform inputs and resources into outputs that cannot be measured on the same scale as the inputs. In many private sector organisations, profits provide a partial measure of efficiency, which is possible because the outputs have prices paid by the people buying the service or goods produced. Because the prices are determined by a market, we can compare the revenues produced by a for-profit organisation with the costs of producing the goods or services, using cash as a measure for both.