آیا رشد خدمات مانعی برای رشد بهره وری است؟ تجزیه و تحلیل مقایسه ای

The relationship between economic structure and productivity growth has been a subject of increasing interest over recent decades. The innovative focus of this paper concerns the role of the service sector in this relationship. Services play a core role in advanced economies, both from a quantitative and a strategic point of view. However, empirical research in this area lies considerably behind the research into the agricultural and manufacturing sectors. This paper focuses on the impact of tertiarisation on overall productivity growth, using a sample of 37 OECD countries in the period between 1980 and 2005. The results partially refute traditional knowledge on the productivity of services. Contrary to what conventional theories suggest, this research demonstrates that several tertiary activities have shown dynamic productivity growth rates, while their contribution to overall productivity growth plays a more important role than was historically believed.

The relationship between economic structure and productivity growth has been subject to much attention over recent decades. Although there has been a growth in interest and recent attempts at modelling, this is not a new line of research and the role of structural change in economic growth and productivity has been recognized since the studies of A. Smith and D. Ricardo. Previous studies have traditionally focused on two processes1: ‘deindustrialisation’ which started with the economic crisis of the 1970s ( OECD, 1975, Blackaby, 1978 and Gemmell, 1982); and ‘tertiarisation’ or the creation of a services society ( Chenery and Taylor, 1968, Bell, 1974, Fuchs, 1968 and Lanciotti, 1971). Some of these authors considered that the changes involved in a transfer of labour from sectors with low productivity to other more dynamic sectors constituted one of the main reasons for overall productivity growth (see, for example, Kuznets, 1966). In this regard, from the first work of Fourastié (1949) and, particularly in Baumol (1967), a part of economic literature has maintained the thesis that the growth of services in the development processes, together with the low productivity in these types of activities as compared with the manufacturing industries, results in a clear disadvantage for future growth ( Baumol et al., 1985, Borjk, 1999, Wolff, 1985 and Bonatti and Felice, 2008). The novelty of this paper is the focus on the role of the services sector in the relationship between economic structure and productivity growth. Despite the quantitative and strategic weight of service activities in advanced economies, empirical work on services seems to lag behind the research on other industries. The literature focused mainly on the shift from agriculture to manufacturing or the impact of specialisation and structural change within manufacturing on productivity growth. The main hypothesis of this paper is that structural changes, and particularly the growth of services, plays a role in productivity growth. The second hypothesis is that the scope for advances in productivity differs significantly across services branches. The first hypothesis is based on the transfer of surplus labour from less productive sectors to more productive sectors, which boosts the growth of overall productivity. To the contrary, structural changes could also slow down productivity growth if the reallocation of resources is towards those branches where productivity growth is low. In relation to the second hypothesis, recent studies have demonstrated that several services branches are not unproductive and they do contribute positively to the overall productivity growth of advanced economies.2 This paper concentrates on the analysis of the relationship between tertiarisation, structural change and productivity growth between 1980 and 2005. After initially contemplating the theoretical approach to the relationship between structural change, the services sector and productivity (Section 2), the remainder of the paper is organised in the following manner. In Section 3, labour productivity growth in the European Union countries and the United States is analysed by means of a conventional shift–share analysis. Although the results are largely consistent with other contributions in the literature at a more aggregate level (Van Ark, 1995, Peneder, 2003 and Havlik, 2005), the examination of individual contributions through diverse branches of the services industry provides some interesting details. Section 4 then provides panel econometric estimations of cross-country growth regressions in a broad sample of OECD countries, together with indicators of services growth. The final section provides a brief summary and conclusions.

This paper has focused on the impact of services growth over overall productivity growth and was inspired by two facts. Firstly, previous papers in the last two decades, showing that structural change has played a role regarding overall productivity growth. Secondly, the quantitative, qualitative and strategic position of the services sector in advanced economies. The opening hypothesis was that, although the conventional theories on the relationship between the service sector and productivity continue to have some validity at a highly aggregated level, they may clearly be questioned at a disaggregated level by taking into account the empirical evidence presented, which agrees with some recent contributions. Many current works have analysed the relationship between structural change and productivity growth, but the added value of this paper is the comprehensive analysis of the most imperative structural change in recent decades: the growth of services. Additionally, an in-depth breakdown of the heterogeneous service industries has been implemented. The results reported here, based on a wide sample of OECD countries between 1980 and 2005, indicate that structural change still matters. The direct contribution of structural changes to overall growth of labour productivity has been measured by a conventional shift–share analysis for the European Union and the US. Data is consistent with most of the literature (widely cited throughout the paper). Moreover, this technique is uniquely implemented in 11 disaggregated services industries. Structural change generally had a positive effect on labour productivity during these years, although it appears to be basically dominated by the effects within productivity growth. Most growth has been due to the increase in productivity within each industry or activity, not to the reallocation of resources from some branches to others. Furthermore, the dynamic component is negative in most of the cases analysed, and an argument is put forward for the subsistence of the so-called structural frontier. Additionally, disaggregated analysis of the different service industries shows certain types of services which systematically achieve higher rates of productivity growth, even creating employment simultaneously. However, many of these likely positive effects are levelled out due to the fact that other types are still characterised by stagnant activities. Econometric panel estimations of Section 4 are applied to test for the impact of the growth of services on overall productivity growth. The data panel refers to 37 OECD countries and the period analysed is from 1980 to 2005. The regressor set includes the growth of the services sector (in terms of percentage over total employment), the initial weight of services and starting productivity levels. These are complemented by other auxiliary variables, such as physical and human capital levels, the change in demographic composition and international openness. Results show that the relationship between the growth of services and overall productivity growth is positive and statistically significant. Moreover, its impact appears to be important. Additionally, the reaffirmation of two processes of economic interest has been observed. Firstly, there is a process of convergence and the reduction of differences in terms of productivity between the economies which initially started with higher productivity levels and those which were trailing behind. Secondly, countries which were initially more specialised in services are those which demonstrate more positive dynamics in their productivity growth. Finally, differentiating between market and non-market services, it can be concluded that the impact of the former is considerably greater than those services which operate outside the market.