In this paper we have focused our attention on the relation between the configuration of protection-seeking and of rent-seeking interests and the choice of the protection instruments. In doing so, we have abstracted from the problem of nonequivalence of tariffs and quotas. Recent writings on product differentiation (Rodriguez, 1979, Donnenfeld, 1984) have thrown new light on this old issue; it may well be that the discussion is of key importance to the understanding of industry's predilection for quantitative restraints. We also failed to deal with the question on nonequivalence of restrictive measures under conditions of uncertainty. Finally, we touched only peripherally on the interaction among state. It is, perhaps, more realistic to regard trade-barrier formation as a game among nations as in Baldwin and Clarke (1985). Every "player" seeks at once to protect its import-substituting industries, and to promote its exports, by putting up barriers and by demanding that others lower theirs. This approach is more suited, however, to explain the severity of restraints than it is to explain the nature of restraints, which is the problem at hand. Trade restrictions have a "protective" and a "revenue" effect. In less-developed countries, with ineffectual internal revenue-collecting systems, revenue-seeking governments may ally themselves with protection-seeking, emergent industries and institute high tariffs. Tariffs and quotas may coexist if the trade-restriction-seeking alliance includes traders. Voluntary export restraints, the newest of the protective devices, may be popular because they circumvent international treaties and short-circuit cumbersome procedures associated with changes in tariffs or quotas, but they also have the virtue of inviting foreign exporters to act like price-setters. The harm to the foreign exporters is thereby mitigated — in fact, exporters may be gainers. Hence the danger of retaliatory action on the part of foreign governments is lessened. Domestic-content legislation is a unique device in that the same domestic factor benefits from protection and also collects the protection-created revenue. We show the formal identity between the "South African" situation, in which the maximum black/white-worker ratio is a binding constraint, and regulations requiring that a given percentage of the value of the product be domestically produced. We also derive formally the maximum permissible foreign-content ratio which is optimal from the point of view of the protected interests. In our analysis we have said virtually nothing about the victims of protectionism. A discussion of the impact of the various measures on those who bear the cost of protection would, doubtless, deepen our understanding of the problem. We should thus consider the magnitude of the deadweight loss and of the income transfer, and the concentration or dispersion of the losers. Last, but not least, we should consider the question of awareness. Thus the cost impact of a tariff is easier to perceive than that of a quota, with the cost of a VER or of domestic-import content rule the most difficult to calculate of all. It may be that the principle of the hiding hand is as important to the new political economy as the Benthamite principles were to the old.