| Abstract: | We analyse the effects of information and communication technology (ICT) on levels and growth of per capita GDP in two different ways: (1) by treating ICT as a specific type of physical capital and as a variable that helps to correct for quality existing physical capital measures, and (2) by considering that telephone lines, personal computers and internet hosts are ‘bottleneck‐reducing’ factors that increase the productivity of labour by making easier the diffusion and processing of (non‐rivalrous and almost non‐excludable) knowledge. We compare the relative significance of the two hypotheses in level and growth estimates and find that, when separately taken, both of them improve upon the classical Mankiw et al. (Quarterly Journal of Economics, Vol. 107 (1992), pp. 407–437)/Islam (Quarterly Journal of Economics, Vol. 110 (1995), pp. 1127–1169) framework. These findings show that our approach captures dimensions of time‐varying country‐specific technological progress that previous approaches in the literature did not take into account. |
