Measuring the Efficiency and Productivity of British Universities: An Application of DEA and the Malmquist Approach

This paper uses data envelopment analysis to examine the technical efficiency (TE) of 45 British universities in the period 1980/81 to 1992/93. This period was chosen primarily because it was characterized by major changes in public funding and in student : staff ratios. To shed light on the causes of variations in efficiency, TE is decomposed into pure technical efficiency (PTE), congestion efficiency (CE) and scale efficiency (SE). The analysis indicates that there was a substantial rise in the weighted geometric mean TE score during the study period, although this rise was most noticeable between 1987/88 and 1990/91. The rising TE scores are attributed largely to the gains in PTE and CE, with SE playing a minor role. The Malmquist approach is then used to distinguish between changes in technical efficiency and intertemporal shifts in the efficiency frontier. The results reveal that total factor productivity rose by 51.5% between 1980/81 and 1992/93, and that most of this increase was due to a substantial outward shift in the efficiency frontier during this period. Introduction In the early 1980s, over 80% of the recurrent income of universities in the UK was obtained from the Treasury (Johnes & Taylor, 1990, p. 38) and, as a result of this dependency on public funding, universities came under scrutiny. Although they are now substantially less dependent on public funds, the efficiency of British universities continues to be the subject of political and economic debate. In 1984, the Committee of Vice-Chancellors and Principals responded to the government's wish for an efficiency study of universities by setting up the Jarratt Committee. This committee recommended wide-ranging changes in the management of universities and, inter alia, the introduction of appropriate performance indicators (Cave et al., 1997, p. 4). Furthermore, a White Paper in 1987 proposed radical new arrangements for the distribution of public funds to universities (ibid., p. 6). Instead of obtaining grants to cover their costs, universities would need to enter into contracts with the Universities Funding Council to provide specific academic outputs in return for the resources granted 2 to them (ibid.). The government urged higher education to provide wider access to its services, and be more responsive to the needs of industry and commerce, less dependent on public funding and more cost-conscious in managing its resources (Johnes & Taylor, 1990, p. 12). With the aim of making universities more responsive to students' demand for higher education, and to encourage them to exploit any spare capacity, an increasing proportion of public funds would henceforth be provided in the form of fee income rather than as block grants (ibid., pp. 42–47). Given these calls for greater efficiency, it is of considerable interest to examine how well universities responded to the increased pressure put upon them. This study uses data envelopment analysis (DEA) to measure the relative efficiency of 45 universities in the UK over the period 1980/81 to 1992/93. This approach makes it possible to measure the degree of variation in efficiency across the sector as a whole, and to identify possible sources of inefficiency. An important feature of the variant of DEA employed in this paper is its ability to identify ‘congestion’ inefficiency, which is inefficiency arising from negative marginal productivity of inputs. More specifically, one of the hypotheses to be examined here is whether British universities' efficiency was impaired by the exceptionally rapid expansion in the number of students that began in the late 1980s. The focus of most DEA studies of UK universities has been on the relative performance of individual departments in a given discipline. A notable exception to this is the interesting study by Athanassopoulos and Shale (1997), who apply DEA at the institutional level. However, their study of 45 institutions pertains to a single academic year, 1992/93. By contrast, the primary aim here is to examine trends in efficiency over time, along with the underlying causal factors. The period 1980/81 to 1992/93 was chosen because it was characterized by major changes in public funding and in student : staff ratios. In addition, University Statistics has a consistent set of data for this period on the key variables required. A final point is that the exclusion of any new universities created after 1992 means that the sample employed here is relatively homogeneous. The standard DEA approach has the disadvantage that it cannot distinguish between changes in relative efficiency brought about by movements towards or away from the efficiency frontier in a given year and shifts in this frontier over time. To capture these two sources of change in efficiency, Malmquist indices are computed. In the next section, the theory underlying the measurement of technical efficiency is examined. This is followed by a discussion of the variables used. The DEA and Malmquist results are then