International capital movements and trade in an intertemporal setting

The following remarks concentrate on an aspect of the relation between trade and factor movements, which has been neglected in the literature so far. It is the intertemporal dimension of factor movements, which in turn is closely linked to the relation between real and financial capital movements. The focus is on capital as a factor of production - analogies with labor and other factors will only be touched upon. Capital movements in this context, are defined in the wide (Boehm-Bawerkian) sense. It is immaterial whether investment or consumption goods are traded internationally. All that is necessary to transfer "capital" is a positive balance on current account and an excess of savings over domestic investment in the capital exporting country - and the reverse in the capital importing country. This corresponds of course to net financial capital movements during the process of "real capital" transfers. It is modelled in a simple way by introducing as the one financial asset an equity share. The paper attempts a critique of the comparative static modelling of the substitution vs. complementarity relation between trade and capital movements. It therefore starts from the same Heckscher-Ohlin framework where factor price equalization is granted. It looks primarily at the conditions for efficiency in the world economy, which is attained if all the marginal equivalences are realized, especially if in addition to commodity prices factor prices are equalized across countries. If factor price equalization (FPE) is brought about by trade alone, there is thus no incentive - and no need - for international factor movements; the two are substitutes. On the other hand they are complements if because of non-FPE by trade alone factor movements are induced and are necessary to attain an efficient situation. The argument will be developed in several steps. First, the comparative static results will be reviewed using a convenient graphic illustration developed by Dixit and Norman (1980). It will then be shown that the comparative static procedure is inadequate, as capital movements in an essential way involve time. The intertemporal framework will be formulated in terms of a neoclassical growth model. Following Oniki and Uzawa (1965) it will be demonstrated that a trade pattern of specialization is particularly interesting for our problem. This scenario of growth cum trade with specialization exhibits non-FPE. If the possibility of capital movements (implying trade in securities and trade imbalances) is introduced, using a procedure suggested by Hori and Stein (1977), a factor movement will materialize, which is complementary to trade. In conclusion some open questions and possible extensions will be sketched.