“Trade in Capital Goods”

Jonathan Eaton and Samuel Kortum
IED Discussion Paper 109, December 2000

Innovative activity is highly concentrated in a small number of advanced countries which are also the major exporters of capital goods to the rest of the world. Hence a critical determinant of a country’s productivity may well be its access to capital goods from around the world and its willingness and ability to make use of them. In their paper, Eaton and Kortum develop a multi-country model of trade in capital goods to assess its role in spreading the benefits of technological advances.

The paper draws connections between a number of important empirical observations. Empirical work in growth theory has sought to establish links between cross-country productivity differences and differences in the rates of capital accumulation. Eaton and Kortum propose that differences in the rates of capital accumulation derive much more from differences in the relative price of capital than simply from differences in savings rates. Therefore, poor countries are getting much less per dollar saved than the richer ones. This naturally begs the question of whether impediments to trade in capital goods could be at work, and leads them to explore the link between productivity and imports of capital goods. Finally, if the relative price of capital goods is indeed lower in some countries than in others, this

implies a possible link between patterns of international trade and deviations from the law of one price, and points to the importance of geographic barriers between countries. In this context, Eaton and Kortum propose new measures of differences in the costs of capital goods across countries.

Applying their empirical framework to data for 1985 across 34 countries, Eaton and Kortum find geographic barriers to be quantitatively important. Moreover, their model enables them to estimate the full cost of buying and using imported equipment in the presence of such barriers. Their trade-based measure of equipment prices falls systematically with development, being the lowest in countries like Germany and the USA and more than 3.5 times these levels in Iran and Kenya. Taken in conjunction with lower consumption goods prices in poorer nations, these differences are even more telling. Since the standard measures of equipment prices, like the ones issued by the UN’s International Comparisons Program, show no systematic linkage, Eaton and Kortum’s results indicate that these earlier measures may not be fully accounting for quality differences and other indirect costs such as learning to operate imported capital. They also estimate that differences in the relative price of equipment account for over 25 percent of the productivity differences between developed and developing nations.