A reevaluation of economic models of Canadian tariff determination
Date
1985
Authors
Grist, Charles Michael
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Recent empirical studies of trade barrier determination have utilized several models in attempts to explain variations in the levels or variations in changes in the levels of protection provided to industries by government. This study briefly reviews and critiques those theoretical models, proposes a more firmly grounded theoretical approach and assesses it against Canada's mid 1960's tariff structure. The analysis is organized into three major sections. The first reviews eight distinct theoretical approaches inherent in existing empirical work. These approaches fall into two categories: economic models, in which trade policy results from the interplay between voters pursuing selfish economic interests, and a democratically elected government pursuing reelection; and non-economic models in which policy seems to result from the autonomous abilities of government or perhaps from broader, less selfish conceptions of individual political behavior. The review suggests that while non-economic models receive considerable empirical support, they suffer from serious theoretical weaknesses. Economic models, on the other hand, are somewhat better developed rut lack strong empirical support.
The second section of the analysis attempts to more carefully scrutinize the fundamental theoretical background supporting economic models and to develop a more sophisticated economic approach. In this section, a simple collective decision rule model of the tariff determination process is developed. This model assumes that individual voters pursue preceived selfish economic interests regarding tariff policy as determined from a simple partial equilibrium conception of the effects of a tariff. Six hypotheses regarding Canada's cross sectional tariff structure are derived from this approach.
The third section deals with the specification of these hypotheses and their empirical evaluation. Multiple regression analysis based on ordinary _ least squares estimates is employed on a sample of 69 Canadian manufacturing industries. Overall the model receives stronger empirical support than any previous Canadian study - both in its basic specification and in three alternative specifications. Hypotheses regarding the influence of import competition, average tariffs on industry inputs and the strength of union tariff structures perform particularly well. Other hypotheses on industry representation concerning traditional concepts of interest group formation and industry voting strength, however, were not supported.