Introduction: The importance of productivity

regarding the importance of productivity growth in raising standards of living. That this needs repeating suggests that the writers feel it necessary to emphasise the positive contribution of productivity growth. In the past, productivity growth was often seen as synonymous with job cutting mainly due to substitution of capital for labour, cost cutting or intensification of work effort. However, in the recent past, policymakers have focussed more attention on productivity growth, both here in the UK and in the EC. Simultaneously, there has been a greater emphasis on positive aspects: the realisation that, by investing and learning to do things better, society as a whole benefits. This change in attitude probably stems from the acceleration of productivity growth in the US since the mid-1990s and its link with the introduction and use of information and communications technology (ICT).

With the growing realisation of the importance of productivity growth in the EC, came a demand to understand the sources and sector location of changes in outputs and productivity. In turn this required reliable internationally comparable measures. In March this year this gap in knowledge was considerably diminished following the launch of the EU KLEMS database in Brussels. This made publicly available a data set that involved the research effort of multiple individuals and institutions across the EU, funded by the EC 6th framework programme, and often with considerable input from National Statistical Offices. Most of the papers that follow stem from the EU KLEMS project.

The first paper introduces the EUKLEMS database and discusses some preliminary results. It sets out the harmonised methodology, drawing on the pioneering work of Dale Jorgenson, discusses data sources and gives an overview of the variables and industry division available. The results section confirms the EU poor performance relative to the US in both labour and multifactor productivity growth. The difference is most pronounced in market services, highlighting that productivity growth in the US has been and still is services led (Bosworth and Triplett, 2007). This overview paper also highlights the considerable diversity of experiences across EU member states. Given this it was decided to report some of the country detail in three papers that draw directly from the EU KLEMS database.

The second paper in this issue compares the UK to France and Germany and concludes that, while nowhere near the success achieved in the US, the UK is now performing better in terms of both employment and productivity growth than our two largest EU competitors. The next paper covers Spain, another large country but at a very different stage of development. The paper highlights the success in reducing unemployment and raising the employment rate but with the downside of poor productivity growth. Finally, in respect to EU KLEMS based research, the last of the three papers considers countries on the geographic fringe of the EU, Austria and the five New Member States from Eastern Europe. The latter experienced strong productivity growth, in particular in manufacturing, but also with falling employment, mostly of unskilled workers. These countries also tended to concentrate more on high technology industries with Austria keeping to a more traditional pattern of specialisation.

While the EUKLEMS project represents a considerable achievement over previously available data, there remain unresolved measurement issues that affect international comparability. Chief of these is the estimation of output and productivity growth in that part of the economy where output is not sold on a market. These non-market services comprise health and social services, education and public administration. Here the most difficult issue to resolve is measuring the volume of output, since we are lacking information on consumers' marginal valuations contained in prices in marketed output. In general terms, the solution is to combine information on quantities of output, often activity rates, with information on outcomes to adjust for quality change. In practice this can prove to be very difficult, as discussed in the final paper in this issue, which again summarises the efforts of a large research group to measure output and productivity growth in the National Health Service. This paper suggests that before and after treatment health outcomes can be used together with activities growth, weighted by unit costs, to produce a quality adjusted output measure. The data to populate such an index are not readily available but the paper shows how available outcome information on short-term mortality rates and waiting times impact on output growth.

The papers in this issue highlight the importance of having robust measures as a basis for further research and to inform policy. The EU KLEMS database runs to 2004 but since then there have been tentative signs of an improvement in the EU's productivity growth relative to the US. Based on only two years' data it is too early to say whether this merely reflects cyclical factors or is symptomatic of more long-term changes. Nevertheless it is vital that the appropriate data exist to analyse such changes and the papers here show that it is feasible to produce these data. Similarly, as the non-market services represent a large and increasing share of national output, it is no longer sensible to put aside measurement for these sectors as being too difficult. The message from the final paper is that the difficulty is in willingness to gather the relevant data rather than a conceptual problem.

Finally, the research reported in this issue owes more than most to the effort of a large number of individuals, including those providing funding, information and research effort. Many are acknowledged in the papers below. The EUKLEMS project would not have been feasible without the considerable research effort of a large number of individuals too numerous to list individually Ð consortium members are listed on the EUKLEMS website. Thanks are also due to individuals employed by the European Commission Services, Eurostat, OECD and National Statistical Offices for copious advice and active participation in meetings, the UK Department of Health, the project steering group and individuals from the UK Centre for the Measurement of Government Activity at ONS.