Shadow Pricing in a Specific-Factors Model: Rejoinder

Abstract

In their comment on my paper (Maneschi, 1985), Tower, Pursell, and Han (henceforth TPH) correctly point out that in the specific-factors model, unlike the situation in a model where goods and factors are equal in number, the withdrawal of a unit of a factor induces changes in the prices of all factors, which may have repercussions on a policy-maker's welfare function and occasion a drain on foreign exchange in addition to value of the production loss associated with the withdrawal of the factor itself. They list a number of possible 'implicit assumptions' that could dispense with the need to reckon with these repercussions and go on to set out a model that explicitly allows for them. The model used by TPH for this purpose is based on a welfare function for the policy-maker given by the summation of changes in different individuals' incomes, YOi multiplied by distributional weights vi. That this is a more general welfare function than one that weights every individual equally is of course incontrovertible. However, for better or worse, most of the international trade theory literature is predicated upon the national welfare function's being a blown-up version of that of a single 'representative individual.' This is also true of some papers of the trade-theoretic shadow pricing literature which derive utility-based shadow prices of factors (such as Bhagwati and Srinivasan, 1981), in which a 'social utility function' is postulated containing the aggregate amounts of consumptions of commodities as arguments. Interestingly enough, even in the monograph by Tower and Pursell (1986) referred to by TPH, most of the analysis is conducted in terms of aggregate consumer preferences, and only one section of a chapter (section 9 of chapter iii) assumes 'heterogeneity of individual consuming units.' While in theory it is doubtless desirable to allow