What you need to show is that, at an exchange rate that permits balanced trade, there is not a set of voluntary trades (willingly entered into by the owners of the respective commodities) that would produce a potential pareto improvement in at least one country with no potential pareto loss in the other.
In pure production models (without relations of production), the implicit exchange rate is 1:1 in labor hours. As soon as you distinguish between hours of work and commodity price (due to variable rates of exploitation), you need to specify the exchange rate explicitly. Incidentally, it is not necessarily the capitalists who trade. The capitalists may sell their output to trading houses which then engage in trade (as was the Japanese case). If your model is properly specified, this should not make any difference. Peter Rakesh Bhandari wrote: > What I am saying is that there is the possibility that trade that > would deliver potential pareto improvement from the perspective of > social labor may not have any such benefits for the capitalists. Is > it possible to show this? > > Rakesh
