Testing Trade Theory in Ohlin's Time

An empirical tradition in international trade seeks to establish whether the predictions of factor abundance theory match present-day data. In the analysis of goods trade and factor endowments, mildly encouraging results were found by Leamer et al. But ever since the appearance of Leontief's paradox, the measured factor content of trade has always been found to be far smaller than its predicted magnitude in the Heckscher-Ohlin-Vanek framework, the so-called 'missing trade' mystery. We wonder if this problem was there in the theory from the beginning. This seems like a fairer test of its creators' original enterprise. We apply contemporary tests to historical data on goods and factor trade from Ohlin's time. Our analysis is set in a very different context than contemporary studies -- an era with lower trade barriers, higher transport costs, a more skewed global distribution of the relevant factors (especially land), and comparably large productivity divergence. We find some support for the theory, but also encounter common problems. Our work thus complements the tests applied to today's data and informs our search for improved models of trade.