Evaluating the Introduction of a Quasi-Market in Community Care: Assessment of a Malmquist Index Approach

It is often the case that only routinely collected, highly aggregate administrative data are available for the evaluation of major public policy initiatives. Such data clearly fall far short of the information advocated by statisticians for policy evaluation. This paper examines the extent to which methodologies developed in the productivity literature can nevertheless extract useful evaluative information from administrative data. A non-parametric Malmquist Index approach is used which seeks to decompose changes in efficiency into technical efficiency changes, scale effects and technological change. We apply the methods to the system of “community care” introduced in 1993 for people affected by problems associated with ageing, mental illness, learning disability or physical disability. Non-parametric methods are used to estimate productivity changes amongst 39 English county councils over a four year period from 1992 to1995. The results suggest a steady annual increase in productivity of about 1.4% per annum over the period under investigation, but it is not possible to determine the extent to which this progress can be attributed to the introduction of community care. Our conclusion is that the methods deployed can exploit readily available data sets at low costs, and are a valuable form of exploratory data analysis, if interpreted with discretion.