تجارت کردن بین بهره وری نیروی کار و انباشت سرمایه در بخش انرژی ایتالیایی

This work provides an explanation for the puzzling trade-off between labor productivity and capital accumulation, occurred in Italian energy sector from the late 1980s onwards. By using a vector autoregressive model, we decompose labor productivity into technological and non technological shocks. We find that: (1) labor productivity responds positively to technological shocks, leading to a transition from one equilibrium to another; (2) capital accumulation shows a persistent decline in response to a positive technological shock, revealing that, in energy sector, technology and capital stock are substitutes. From our analysis we get some policy lessons. The obtained results point out the importance of a comprehensive strategy aimed at increasing technological progress through research, innovation and human capital investment in energy sector. Conversely, our findings state that institutional reforms and changes in regulation can only have a transitory effect on labor productivity in energy sector, without permanent gains in the future.

This work provides an explanation for the puzzling trade-off between labor productivity and capital accumulation, occurred in Italian energy sector from the late 1980s onwards. Our aim is to give a rationale of this stylized fact providing some new evidences to measure the impact of technological and non technological shocks on the growth rate of labor productivity and capital accumulation. The obtained results show that, over the last decades, the rising of labor productivity in Italian energy sector is mainly due to technical progress rather than capital accumulation. Hence, energy markets need policies aimed at strengthening technical advances to increase both productivity and profitability in the long run. Readily available aggregate national accounts data are frequently used to describe relative movements in labor productivity and capital accumulation. In Italy, they show that the long-run pattern of rising labor productivity in energy sector has an opposite image in declining trend rates of real capital growth. Fig. 1 indicates the existence of this conflicting pattern since the mid of 1980s. Since then, trend value added per worker in Italian energy sector has been rising; but, the recovery in labor productivity was accompanied by a correspondingly negative trend in capital accumulation. The opposite movements of these variables suggest that higher productivity growth in Italy could be related to some further sources. Our aim is to improve our understanding of the ultimate causes of the trade-off, identifying technological and non technological shocks at the origin of the phenomenon.From a methodological point of view, decomposing labor productivity in its sources requires an appropriate proxy of technological progress. We extract such a series by controlling for non technological effects in labor productivity: varying capital accumulation, crude oil price, interest rates and unemployment. An extensive literature on productivity and technological progress in energy sector, however, exists. Prior to the 1990s, especially in US, a large number of contributions have been made to investigate how regulation and technological advances affected the generation of electricity and the associated level of productivity. For a good review of the literature, on methodologies and applications, see Abbott (2005). Traditionally, the standard procedure is to compute index number of partial or total factor productivity (TFP). Kendrick (1961) was the first to assess productivity change measurements in the electricity industry, using labor hours and capital stock. He estimated, over the period 1904–1953, a global growth rate of 5.5% of the electricity utilities in the US. This pioneering contribution has been refined in subsequent papers (Kendrick, 1973 and Kendrick and Grossman, 1980). Later models have, however, questioned about Kendrik's results. Basically, the main criticism was that energy sector is a capital intensive industry with partial productivity depending strongly on technology. This would imply that empirical research should investigate directly upon the contribution of technological progress to control the effects of technology change on partial productivity. Using growth accounting, Cowing, Small, and Stevenson (1981), Nelson and Wohar (1983) and Callan (1991) found that for the US electricity industry, over the period 1970–1990, the TFP growth was the main component of labor productivity. Its deceleration (2.3% on average in the 1970s, versus 1.4% in the 1980s) explains the productivity slowdown recorded in the US electricity industry during that period. More recently, alternative methods have been suggested. One of this method is the data envelopment analysis (DEA) pioneered by Charnes and Cooper (1978). A range of works has been conducted in a number of countries to measure TFP in energy sector ( Whitman & Bell, 1994), and, recently, Abbott (2006) estimated that, after 1990, the acceleration in productivity growth of the Australian electricity supply industry has been driven by a marked increase in the level of TFP. But, the drawback of DEA – applied to the energy sector – is that it gives a good indication of the degree to which there is scope to improve productivity to world's best practice, but it does not give any indication of the reasons why the energy sector has improved its performance over the past decade (Whitman, 1999). In addition, works based on DEA do not provide any indication of sources (shocks) moving energy sector away from some potential level of productivity. As a consequence, caution should be taken when comparing productivity among countries because the pattern of productivity can be influenced either by economic cycle, shaping its short-run movements, or by technological advances driving its secular component. Nonetheless, even the most updated empirical investigations fail in exploring this topic in depth (Ang and Zhang, 2000 and Ang et al., 2004). For example, the recent paper by Wang, 2007a and Wang, 2007b decomposes energy productivity into several components, with technology as the most important source of growth, but it does not provide any adding information about the possibility that shocks that move the economy at business cycle frequencies may also affect the economy in the long-run. By contrast, in the present paper we view labor productivity movements as arising from mixture of shocks. Our goal is to disentangle these shocks. Then, we ask if the effects of these shocks are transitory or permanent. We start from the observation that, in principle, a wide range of changes (shocks) can explain the movement of labor productivity in energy sector, including changes in the rate and direction of technological progress, changes in capital accumulation, changes in incentives and regulation, fluctuations of the oil price, changes in aggregate demand, or any combination of these. But, rather than simply examining the correlation among labor productivity and nominal/real variables, our purpose is to identify the shocks which induce movements in labor productivity, and to measure the productivity response to each shock separately. This approach is similar to the one taken by Shapiro and Watson (1988), Blanchard and Quah (1989) and Galì (1999). On this basis, we find that (1) labor productivity in energy sector responds positively to technological shocks leading to a transition from one equilibrium to another; further, (2) capital accumulation shows a persistent decline in response to a positive technological shock, revealing that, in energy sector, technology and capital stock are substitutes. Yet, we do not assume that all fluctuations in labor productivity are attributable to technological shocks. Specifically, non technological shocks can affect permanently capital accumulation, but play a minor and transitory role in explaining productivity growth. The empirical findings presented here enlarge our knowledge of the changing effects of technology shocks in the long and short run. A clear understanding of the link between labor productivity and its technological and non technological components in energy market may help policy makers to implement appropriate industrial policy. The rest of the paper is organized as follows. In the next two sections, we use economic theory about long-run impacts of shocks to identify our empirical model. Section 3 gives the precise econometric specification. In Section 4 we present our estimates. Section 5 concludes.

The purpose of this paper was to investigate the causes of the trade-off between labor productivity and capital accumulation (growths) occurred in Italian energy sector over the 1980–2007 period. To study this issue we employed a parsimonious structural VAR, measuring the labor productivity response to technological and non technological shocks. Our estimates state that technological shocks have permanent effects on the level of productivity leading to a transition from one equilibrium to another. Further, most of the variation in productivity is due to technological shocks that account for roughly two-third of the productivity variation. For the Italian data, it appears that technology and capital accumulation in energy sector are substitutes. Additionally, favorable non technological shocks affect permanently capital accumulation, but play a minor and transitory role in explaining productivity growth, capturing only the residual fraction of productivity fluctuations at high frequencies. Thus, our findings tell us a story about the sources of shocks affecting productivity in energy sector. The data support the idea that technological progress is necessary to gain a strong and persistent advance in labor productivity. But, while the model provides a positive answer to the question of the relative importance of technological shocks in growth of energy sector, it is more prudent about the role of non technological shocks. That is, our estimates suggest that components other than technological progress are less important in driving productivity growth than markets and policy makers expect. Finally, from this analysis we get some policy lessons. Our empirical investigation confirms the importance of a comprehensive strategy aimed at increasing technological progress through research, innovation and human capital investment (Wang, 2007a and Wang, 2007b). Implementation of such a wide-range agenda would create a more dynamic business environment in energy sector, which will, ultimately, be reflected in a significant increase of labor productivity growth rate. But, the same analysis states also that institutional reforms can have only a transitory effect on productivity, without permanent gains in the future. Therefore, our study has an important implication for current energy policies. European reforms and shifts of national legislations formulated in recent years mainly to support investment in the energy sector, cannot guarantee the jump of the technological stock necessary to increase labor productivity and competitiveness in the long-run. Remedying this problem will require the development of new policies aimed at promoting the rate of technological progress in the energy sector.