The Measurement of Welfare Loss from Tariffs : Some Further Explorations

Wide currency has been gained in recent years by the so-called "triangles method" for estimating the welfare impact of the imposition of a tariff or its removal. While this approach faces a few well-knnown conceptual problems, it seems that no better substitute has yet been devised for this purpose. In the present paper we shall adopt this method, overlooking its difficulties, and attempt to widen its scope by applying it to issues which seem to have hitherto been subject to only little exposition. Three topics will be dealt with. The first, that of the welfare impact of a customs union, relates the measurement of the resultant gains or losses to the theoretical concepts of trade creation and trade diversion. The second, that of the welfare impact of tariffs imposed on the import of a capital good, incorporates into the analysis the treatment of the dynamic process of adjustment: it will be shown that though protection of a capital good does not affect the eventual size of the capital stock, it causes a welfare loss through its effect in the adjustment path. Finally, we analyze the welfare loss in an industry where protection results in a negative value added, so that the estimated effective rate of protection is negative. It will be shown that while the effective rate itself does not appear in the loss measure, all the components of which it consists participate in determining the size of the loss. First, however, to provide a suitable framework, we shall briefly reconstruct the basic skeleton of the method under consideration