If it is accepted that protection is a form of redistribution that generally entails substantial net cost for the society at large, an important conclusion follows. There cannot be any rational, systematically applicable criteria for deciding which industry needs or deserves protection against any other, how much protection it should receive, in what form, and for how long. And it follows further that in the absence of such rules, the redistribution that protection effects cannot have any basis in justice. This is the essence of the political and constitutional problem posed by protection. The decision as to who will get it can only be an ad hoc value judgment, and constitutional logic requires policy judgments of this kind to be made by representative assemblies. Should the power of decision rest with the executive, the absence of preordained criteria would make it an unreviewable, in fact an unlimited power--a constitutional contradiction in terms. The danger that such a power would be used for buying political support would be great. Decisions concerning protection affect other countries as well, however. Some of them are capable of retaliation, and national legislatures cannot negotiate with each other. If import policies of all major countries were legislative policies, the result would be a more or less general and continuous commercial war. In other words, the problem arises from this dual aspect of protection--that it is domestic redistribution but has a direct effect on international relations. How the problem has been dealt within the past century and a half is the theme of this chapter.
Until the First World War, two solutions were in effect, one generalizable, the other a special case. The European states maintained among themselves a commercial treaty system in which tariffs were contractually bound for long periods, usually ten years. The bilateral