Physical capital

Despite this simple and intuitive reasoning, a heated debate on the link between capital accumulation and economic growth is ongoing in the literature. For example, Easterly and Levine (2001) argue that factor accumulation is not the main driver of output growth. Their results are challenged by Bond et al. (2004) who argue that investment has a positive long-run effect on both the level and the growth rate of output. Krueger and Lindahl (2001, p. 1125) find “an enormous effect” of the growth rate of capital per worker on GDP growth in their cross-section estimations but they suspect some endogeneity bias.

This chapter will build on the augmented Kaldor model outlined in chapter 2 and it will argue that the accumulation of physical capital reacts to variables like labor input, human capital, trade openness and institutional settings. Output growth is unlikely to happen without accumulation of physical capital, but accumulation is not the main or ultimate reason for output growth. Nevertheless, empirical studies should find that rapid output growth goes hand in hand with rapid growth of the physical capital stock. Output and the capital stock are cointegrated as Kaldor (1957) found (although the term “cointegration” was not coined back then). This view reconciles the opposing views in the empirical literature mentioned above.

In addition, this chapter will show that many of the results of earlier empirical studies which find a significant link between the level of the investment ratio and the level of income are sensitive to the use of investment valued at international prices instead of domestic prices. The conclusion is that there is neither a theoretical nor an empirical reason why investment ratios should correlate with output levels across countries – in contrast to the Solow model as usually applied in cross-country analyses. This chapter will also show that there are significant differences in investment ratios depending on databases and on whether real or nominal ratios are used.