The contribution of information and communication technologies to the growth of the Spanish economy

In recent decades the branches of activity of goods and services related to information and communication technologies (ICT) have made significant technological advances, enabling final prices to be significantly reduced at the same time as the quality of the goods and services produced has been improved. There is a general consensus as to the importance of the diffusion of technical progress for economic growth. In relation to information and communication technologies three main channels of transmission should be distinguished. First, it might be expected that the ICT industries in which the technological innovation takes place will, in the initial phase of the process, see an increase in their rates of change of output and of productivity. Second, the reductions in prices and improvements in quality of ICT goods and services should boost investment in new technologies across the other branches of activity, and intensify the capital-labour ratio, thereby increasing the rate of growth of productivity in the other productive branches. Finally, positive spillover effects and improvements in the organisation of the production of the economy as a whole, linked to the use of this type of goods and services, may foreseeably emerge, with a positive impact on total factor productivity (TFP). Given how significant the development of the new technologies may be for economic growth, it is important to try to quantify the impact of the three channels of transmission mentioned above. However, this task involves some serious problems. First, for most economies, detailed data are not available for output by industry and for stocks of IT and communications capital. Second, measurement of the output of some of the components of these branches of activity involves certain difficulties. Finally, the deflators used to estimate output and the stock of capital do not taken into account the full extent of quality changes, which in recent decades have had a very marked effect on ICT goods and serv In the United States (where the aforementioned methodological problems are less serious) the numerous available studies coincide in indicating that the contribution of ICT to GDP growth has been significant, both as regards the direct impact of the industries producing ICT, which have seen their output and productivity rise, and through greater use of the new technologies as productive inputs. However, the results in terms of the effect on total factor productivity are not so unambiguous. Meanwhile, the studies existing for the European economy show that the growth of industries producing ICT and the greater use of these goods as inputs have had a positive effect on the expansion of GDP, albeit a smaller one than in the US. Nonetheless, this growth of ICT in Europe does not seem to have been sufficient to prompt significant increases in productivity, or at least to offset the effect of other factors that have had a negative impact on the efficiency of the European economy. As regards Spain, the studies available analyse the contribution of ICT to economic growth from the viewpoint of their use as inputs by the rest of the branches of activity. Daveri (2001) and Hernando and Núñez (2001) find that, as in the case of most European economies, the contribution of the stock of ICT capital to growth of output and labour productivity has been significant, especially when it is taken into account that the weight of this kind of capital good in the total capital stock is still low. Hernando and Núñez (2001) also find a tendency for this contribution to grow during the nineties. The main purpose of this paper is to try to estimate, for the Spanish economy, the direct impact of the industries producing ICT goods and services on the growth of output, employment and productivity, making a comparative analysis with the overall growth of the market economy and developments in other EU economies and in the United States. The analysis is based on the definition of ICT industries used by the OECD, which includes three kinds of activity: the manufacture of computer and communications equipment (ICT manufacturing), telecommunications services (ICT communications) and the services of computer and related activities (ICT computer activities). The main source of information used in this analysis has been the Central Balance Sheet Office of the Banco de España (CBBE), since the National Accounts information has significant limitations, both in terms of the number of variables available and the level of detail required. Moreover, it has a significant time lag (the latest available data corresponding to 1997). By contrast, the information provided by the Central Balance Sheet Office (although it also has certain methodological problems, as we shall see later) goes up to 1999, enables a large number of economic variables to be compiled and covers around 55% of the employment of the ICT sector, according to the DIRCE (national directory of corporations). Although the coverage is relatively uneven across the various industries, the CBBE data provide a sufficiently representative statistical base. Meanwhile, the information relating to other European economies has been obtained from Van Ark (2001). It should be taken into account that not all firms are represented in the CBBE and that neither are all the various branches of activity equally well represented, so that this information cannot be used directly to calculate the economic weight of the ICT industries in the economy as a whole, nor their contribution to the growth of the latter. Population variables have therefore had to be constructed for each of the variables analysed. They have been estimated by applying certain coverage ratios toices. Nonetheless, numerous recent studies try to overcome these limitations and analyse the contribution of new technologies to the growth of certain western economies. In the United States (where the aforementioned methodological problems are less serious) the numerous available studies coincide in indicating that the contribution of ICT to GDP growth has been significant, both as regards the direct impact of the industries producing ICT, which have seen their output and productivity rise, and through greater use of the new technologies as productive inputs. However, the results in terms of the effect on total factor productivity are not so unambiguous. Meanwhile, the studies existing for the European economy show that the growth of industries producing ICT and the greater use of these goods as inputs have had a positive effect on the expansion of GDP, albeit a smaller one than in the US. Nonetheless, this growth of ICT in Europe does not seem to have been sufficient to prompt significant increases in productivity, or at least to offset the effect of other factors that have had a negative impact on the efficiency of the European economy. As regards Spain, the studies available analyse the contribution of ICT to economic growth from the viewpoint of their use as inputs by the rest of the branches of activity. Daveri (2001) and Hernando and Núñez (2001) find that, as in the case of most European economies, the contribution of the stock of ICT capital to growth of output and labour productivity has been significant, especially when it is taken into account that the weight of this kind of capital good in the total capital stock is still low. Hernando and Núñez (2001) also find a tendency for this contribution to grow during the nineties. The main purpose of this paper is to try to estimate, for the Spanish economy, the direct impact of the industries producing ICT goods and services on the growth of output, employment and productivity, making a comparative analysis with the overall growth of the market economy and developments in other EU economies and in the United States. The analysis is based on the definition of ICT industries used by the OECD, which includes three kinds of activity: the manufacture of computer and communications equipment (ICT manufacturing), telecommunications services (ICT communications) and the services of computer and related activities (ICT computer activities). The main source of information used in this analysis has been the Central Balance Sheet Office of the Banco de España (CBBE), since the National Accounts information has significant limitations, both in terms of the number of variables available and the level of detail required. Moreover, it has a significant time lag (the latest available data corresponding to 1997). By contrast, the information provided by the Central Balance Sheet Office (although it also has certain methodological problems, as we shall see later) goes up to 1999, enables a large number of economic variables to be compiled and covers around 55% of the employment of the ICT sector, according to the DIRCE (national directory of corporations). Although the coverage is relatively uneven across the various industries, the CBBE data provide a sufficiently representative statistical base. Meanwhile, the information relating to other European economies has been obtained from Van Ark (2001). It should be taken into account that not all firms are represented in the CBBE and that neither are all the various branches of activity equally well represented, so that this information cannot be used directly to calculate the economic weight of the ICT industries in the economy as a whole, nor their contribution to the growth of the latter. Population variables have therefore had to be constructed for each of the variables analysed. They have been estimated by applying certain coverage ratios to the sample values and, like all estimates, must be interpreted with caution. Finally, certain limitations of the CBBE information for the conduct of this analysis should be noted. First, most of the variables analysed refer to nominal values, so that their transformation into real terms requires the application of deflators, which are not always available with the degree of detail required. Also, as already mentioned, the deflators used do not sufficiently taken into account the changes in quality in ICT goods and services, so that the real value added and productivity may be underestimated. At the same time, the CBBE is somewhat biased towards large established firms, which means that developments in the ICT industries according to these data are largely determined by this kind of firm. Finally, the variables obtained from individual data usually display a higher degree of variability than those observed at the aggregate level and, in this respect, it may be more appropriate to conduct the analysis for a period of years rather than confining it to specific years. In spite of these limitations, the information presented in this paper is sufficiently representative to make a general assessment of the economic development of the ICT industries and their contribution to the growth of the economy as a whole. This same information constitutes the basis of the indicators of information and communication technologies presented on the Banco de España’s website. This article is organised as follows. Section two analyses the weight of the industries producing ICT goods and services in the economy as a whole and the economic behaviour of these industries is characterised relative to the patterns observed for the market economy as a whole. Section three examines the contribution of the ICT industries to the growth of value added, labour productivity and total factor productivity of the market economy as a whole. In both sections the results obtained for Spain are subjected to international comparison. Finally, section four sets out the main conclusions.