1 991 Economic Sanctions and I nterest Group Analysis: A Reply [ 1991 SAJE v59(1) p92] W. H. KAEMPFER AND A. D. LOWENBERG *(1) IN A RECENT ISSUE OF THIS JOURNAL, Black and Cooper (1989) discuss the merits and drawbacks of a public choice approach to the study of economic sanctions, particularly the effects of sanctions on the target country. We agree with them that a great deal mo re work is needed in terms of refining interest group analysis to fit more closely with the unique historiography of the target nation's political economy. Unfortunately however, several of their criticisms levelled against our work in this area (Kaempfer and Lowenberg, 1988a, 1988b) reveal fundamental misunderstandings of the price theoretic structure of the competitive interest group model we employed. One of our purposes was to show that sanctions selectively targeted at those interest groups which benefit from apartheid policies, whomsoever those interest groups happen to be, will reduce the endogenous "level" of apartheid policy emerging out of the po litical process. Black and Cooper object that the groups which benefit from apartheid might not be the same ones that are hurt by sanctions (ibid.: 192). But this is beside the point, because we were positing a sanction that is geared specifically to affec t the pro-apartheid groups more than the anti-apartheid groups. Black and Cooper maintain that "economic sanctions are...incapable of isolating those who benefit from apartheid or have the power to change it" (1989: 192). While this is undoubtedly true of most of the sanctions currently in place against South Africa (the Kaempfer-Lowenberg approach suggests that these sanctions are a reflection of protectionist and other interest group pressures in the sanctioning countries and were not carefully designed w ith respect to their impact in South Africa [Kaempfer and Lowenberg, 1989]), it need not generally be true of all sanctions that could conceivably be applied. For example, a convincing argument can be made that an embargo on South African imports of interm ediate and capital goods, or official financial sanctions on capital inflows, would hurt white skilled workers and consumers 1 991 SAJE v59(1) p93 more than a boycott of South African agricultural exports, whose costs would fall largely on black workers. *(2) Black and Cooper also argue that, if the beneficiaries of apartheid constitute a "politically more powerful" (ibid.: 191) group than the opponents, then even a sanction that hurts the beneficiaries more than the opponents will not produce a decrease in the equilibrium level of apartheid. Instead, they claim, this equilibrium will actually increase! To see the fallacy of this, consi der our model of interest group competition in terms of a very simple supply and demand analysis. *(3) Those individuals who derive utility from policy A at a diminishing marginal rate reveal a downward sloping demand curve for A, the height of which at each point measures their willingness to pay, in the form of political pressure, for each increment of A. Similarly, those individuals who are hurt by A reveal a demand schedule for smaller quantities of A, which slopes down from right to left. The height of this curve at each point represents the supply price of A in terms of the amount that the opponents of A would be willing to pay to prevent a marginal increment of A. The "politically efficient" level of A occurs at the point of intersection of these two schedules. This equilibrium will of course, change if the curves shift, and there are two types of shift parameters which we identif y (Kaempfer and Lowenberg, 1988a). The first comprises pecuniary income effects. Suppose the pro-A group experiences a decrease in income. Since political influence is assumed to be a normal good, the ability of members of this group to purchase political influence (contribute to political campaigns, lobby legislators, organize demonstrations or strikes, etc.) is reduced if their pecuniary income falls. This shifts the demand curve for A in a downward direction and decreases the equilibrium level of A. A second type of shift parameter is the political effectiveness coefficient applying to each group (Kaempfer an d Lowenberg, 1988a). Groups differ in terms of their capacities to organize collective action (overcome free-ridership incentives) and generate political pressure. If, for example, the anti-A group's political effectiveness is increased by some exogenous e vent, then the supply curve of A will shift up (the supply price of A increases because members of the anti-A group are willing to allocate more resources to opposing A) and the equilibrium level of A falls. 1 991 SAJE v59(1) p94 Now, if the pro-A and anti-A groups are equally powerful politically, which, in terms of the model, means that each group is equally effective in exerting political influence, then the equilibrium level of A is zero. Recall that policy A is a regulation wh ich redistributes wealth (either pecuniary, nonpecuniary, or both) from the anti-A group to the pro-A group: it is like a tax on the one group to benefit the other group. But like any tax or regulatory transfer, A entails deadweight costs. The anti-A group pays more in lost welfare than the pro-A group receives in enhanced welfare, the difference being a net loss to society. Hence, if both groups are equally effective politically, the anti-A group will always be willing to pay more to avoid the policy A than the pro-A group is willing to pay for it. In other words, the supply curve lies above the demand curve at all levels of A, and a zero quantity of A is produced. A fundamental result of competitive interest group models of the political proces s is that 53