Introduction

The analysis so far suggests that endowments and demand conditions dictate a more machine-intensive technology in the United States, leading to higher labour productivity than in Europe. It might be expected, then, that an approach relating comparative labour productivity to comparative fixed capital per worker via a conventional production function would be successful in explaining international differences in labour productivity. However, this chapter shows that in practice differences in total factor productivity (TFP), incorporating fixed capital as well as labour inputs, are almost as large as differences in labour productivity. Although there are a number of measurement and conceptual problems, which make it likely that the role of physical capital is understated in the traditional growth accounting or levels accounting approach, it is also necessary to look more closely at the accumulation of human capital.

Here, it is important to distinguish between the different types of human capital needed to operate mass production and flexible production systems, since the standard data on years of schooling suggest little difference between the major industrialised countries (Maddison, 1987, table A12). Even if we accept that there are problems of international comparability with the schooling data, there must still be doubts about the precise links between general schooling and the jobs actually performed by the majority of workers in manufacturing (Howell and Wolff, 1992). Hence it is important to look at human capital accumulation that is more directly relevant to the skills actually used in the manufacturing sector.