Embodied Technological Change and Technological Revolution

At the beginning of 2000s the sector of Information and Communication Technologies (ICT) has been considered of fundamental importance in the explanation of the economic performance of several countries. In particular, the strong productivity growth in the computer sector has led some analysts to conclude that the era of a "New Economy" has begun, while more sceptic analysts consider this phenomenon as nothing more than a stock market bubble. For all these reasons a great attention has been devoted to the study of the "ICT Revolution" and of its effects on the economy. On the empirical side, the main studies outline the strong productivity growth in the computer sector, but also problems of measurement of the real contribution of ICT to the growth and productivity of the economy, together with the fact that the productivity growth in the computer sector has not been accompanied by spillovers from this sector to the rest of the economy. On the theoretical side, the most important contributions underline the importance of embodiment of technological progress, but also the fact that the ICT revolution has been accompanied by some "puzzling phenomena" (in particular an initial strong decrease in the productivity of the whole economy, followed only later by a rise). Boucekkine and de la Croix (2003) argue the possibility of explaining the essential characteristics of the ICT revolution in the framework of endogenous growth theory. They consider the effects of positive supply shocks (especially in the hardware and in the R&D sectors), and they find that only a positive productivity shock in the R&D sector has long term growth effects. As a consequence, only if the ICT-driven growth episode is based on an increase in the productivity of R&D it is possible to conclude that this expansion is likely to have permanent effects in the economy. The model presented in this paper is based on the contribution of Boucekkine and de la Croix, and tries to explain some characteristics of the ICT revolution that emerge from the data. It is a multi-sectoral endogenous growth model with discrete time, infinite horizon, horizontal differentiation and embodied technological progress. The crucial differences with respect to Boucekkine and de la Croix concern the composition of the workforce (that is assumed to be homogeneous) and the specification adopted for the R&D sector (the so-called "lab-equipment", first introduced by Rivera-Batiz and Romer). In this model the optimality conditions that hold at the equilibrium are derived, then the balanced growth path and the steady state are obtained, and in this way it is possible to find some analytical results concerning the effects on growth of different shocks that can interest the economy. It is then possible to consider the numerical simulation of a calibrated version of the model, that allows to obtain interesting results concerning the short run response of the system to the shocks and the robustness of the model. These results are also compared with the available data concerning the US, in order to verify the ability of the model to reproduce the real situation. The first result is that the "lab-equipment" specification assumed for the R&D sector allows growth as a consequence of productivity shocks in all sectors (while without such specification only a shock on the productivity of the R&D sector influences the growth of the economy in the long run). A second result is that the shocks on the productivity of the final good sector and on the cost of R&D on the one hand, and the shocks on the productivity of the equipment sector and of the intermediate good sector on the other hand, affect differently, in the short run, the economy, and influence the growth with different intensity. The model also turns out to be sufficiently robust, since when some parameter is significantly modified with respect to the benchmark case, both the qualitative and the quantitative implications remain valid. Finally, an extension of the model that takes into account the presence of learning and spillover effects is able to reproduce empirically the behaviour of US productivity in the recent years. The general conclusion is that if the ICT revolution can be interpreted as a permanent shock on R&D or as a spillover (on the final good sector), it will have long run effects on the economy. On the contrary, if the ICT revolution is interpreted as a shock on the equipment sector or on the intermediate good sector (i.e. as the possibility of producing easily new softwares), it will not have strong long run effects.