Abstract:
At a time when protectionist tendencies are on the upswing world wide once again, it is important to gain evidence on the cause of such movements. Proper diagnosis of the determinants of protection may provide a more promising avenue to avoiding discriminatory economic policies than reiteration of the welfare losses of protection has been able to do. For this purpose Germany constitutes a potentially fruitful case study. First, historically, economic and political regimes have varied widely. Second, institutional conditions at the present are unique enough to warrant and require a closer look at the mechanisms which provide protection. Third, Germany constitutes a large and still relatively liberal market for the products of the less developed and newly industrialized countries. Hence, trade policy developments in Germany are likely to lead to serious repercussions in those countries. The general framework chosen for analysis of German protective mechanisms posits the existence of a political market for protection, and follows from the work of Downs (1957), Buchanan and Tullock (1962), and Ohlson (1965), as well as the application of this principle to international trade and protection issues by a number of authors. Very briefly, the demand for protection on the part of producing units is derived from firms' opportunity costs in eliciting protection; and the supply of protection is derived from governments' opportunity cost in granting protection. In applying this guiding hypothesis to Germany, explanatory variables are sought which messure or proxy those opportunity costs, both on average over time, and across industries at a point in time. The specific hypotheses tested must in part be derived from institutional features peculiar to Germany. In the next section of this paper, this framework is applied historically to the period 1880 - 1978, and in the section thereafter, it is applied to explain the structure of protection in manufacturing industries in 1974. Developments since then are briefly discussed.