Craig A. Depken, II

In recent years, several public policy initiatives have aimed at the increasingly common red light violation. Policies ranging from public service announcements, increased penalties for red light violations and, most strongly debated, the use of red light cameras to ensure near perfect enforcement of red light violations, have been employed across the country. However, these initiatives have failed to reduce the number of red light violations to zero, a situation that has frustrated public policy crafters. This paper suggests that while some drivers may purposefully violate red lights, there are natural conditions under which drivers find themselves involuntarily running red lights. If drivers involuntarily run red lights, the traditional policy tools to combat illegal behavior will be less effective. We show that past public policy initiatives may have focused on an inappropriate source of red light violations, thereby having less success than their proponents had hoped.

In this paper we propose a simple model in which firm-specific advertising has cooperative and predatory effects. Our model is set in a static spatial market in which firms are naturally segmented into two distinct submarkets: several large firms are located in the core, with small firms operating as a fringe. We test the net effect of the offsetting market size and market share effects both fringe and core firm advertising on the advertising decisions of large firms in several U.S. consumer industries. Empirically, advertising by fringe firms leads to an increase in advertising efforts by large firms in the core of an industry, implying that these advertising expenditures are strategic complements. On the other hand, advertising by core firms in an industry decreases advertising expenditures of other core firms, indicating strategic substitutes. Our findings imply that equilibrium levels of advertising by large firms may be greater than those predicted by models with symmetric strategic relationships.

Previous research on market structure concludes that it is difficult to empirically distinguish between various forms of imperfectly competitive behavior. In this paper, we demonstrate that this `identification' problem can be overcome by analyzing the dynamic relationship between output of firms in a market. Specifically, if output levels are characterized by stochastic trends then collusive firms will exhibit positive long-run cointegration while firms engaged in non-cooperative competition will exhibit negative cointegration. The methodology is applied to OPEC, which has been the subject of numerous analyses concerning its market structure. We find that OPEC has indeed behaved as an output-based cartel (a collusive market structure). A natural extension of the empirical methodology provides interesting insight to the disequilibrium dynamics associated with OPEC's cartel behavior.