It is close to impossible to find a parking attendant in Paris, Frankfurt or Milan, while in New York City they are common. When you arrive even in an average Hotel in an American city you are received by a platoon of bag carriers, door openers etc. In a similar hotel in Europe you often have to carry your bags on your own. These are not simply trivial traveler’s pointers, but indicate a deeper and widespread phenomenon: low skilled jobs have been substituted away for machines in Europe, or eliminated, much more than in the US, while technological progress at the “top” i.e. at the high-tech sector, is faster in the US than in Europe. Why?
This paper suggests that an important difference between Europe and the US that leads to such technological differences lies is their different labor market policies. While US labor markets have been deregulated and labor unions there were significantly weakened, most European countries have kept wage inequality low through a host of labor market regulations, like binding minimum wage laws, permanent unemployment subsidies, firing costs, etc. These policies created incentives to develop and adopt labor saving capital-intensive technologies at the low end of the skill distribution. At the same time technical progress in the US has been more skill biased than in Europe, since American skilled wages have been higher.
There are only a few ways to model differential technology adoption across countries. One is to assume that technology adoption is costly, like Parente and Prescott (1995). This approach may help in understanding gaps between rich and poor countries, but it does not fit our case, since if adoption costs in Europe were higher, we should observe less technical progress in all sectors, which is not the case. Basu and Weil (1998) suggest instead that technology adoption depends on supplies of factors of productions, as different technologies fit better different capital-labor ratios. But this approach as well is more applicable to the rich and poor countries, which differ significantly in their capital abundance, but cannot be applied to study differences between Europe and the US. We therefore resort to a third approach, following Champernowne (1963) and Zeira (1998, 2007), which models technological change as substituting labor by machines. According to this approach new technologies reduce labor costs but require purchasing machines, namely increasing capital costs. Hence, such technological innovations are invented and adopted only if wages are sufficiently high, so they reduce the cost of production.