Collusion and Financial Leverage: An Analysis of the Integrated Mill Steel Industry
Lord, Richard A., Farr, W. Ken, Financial Management
We show that firms can design their capital structure to provide a publicly observable indication of compliance with a collusive agreement. We develop two empirically testable hypotheses based on this argument and test these propositions on data for seven integrated mill steel firms. Our study period covers years when prices were overtly coordinated under the basing point pricing system and after the demise of the system. Empirical tests confirm the hypotheses that leverage is positively related to both price elasticity of demand and the level of convertibles outstanding during the years after the collapse of the basing point pricing system.
Since the Modigliani and Miller (1958) study, one of the central issues in financial economics is why and how firms choose specific capital structures. Ravid (1988) reviews a number of studies that focus on product market conditions and financial decisions, yet Harris and Raviv (1991) indicate a serious lack of research relating financial leverage to the nature of the firm's product market.
Maksimovic (1988) suggests that leverage can be used to effectively maintain collusion. First, he defines the general conditions and motivations that make it possible and desirable for oligopolists that have only equity in their capital structure to engage in collusion. He then shows that when a firm is partially financed with nonconvertible debt, there is a discernable ceiling on leverage that is a positive function of a firm's price elasticity of demand. This ceiling provides a publicly observable gauge that the colluding firms can use to judge compliance with a collusive agreement. The firms that do not exceed the threshold indicate their intention to abide by the compact. Firms that exceed the ceiling send the opposite message, thus jeopardizing the continuing existence of the cartel. Brander and Lewis (1986, 1988) and Stenbacka (1994) outline similar approaches to Maksimovic.
We extend Maksimovic's model to show that the use of convertible securities in place of nonconvertible debt permits a firm to exceed the industry's debt ceiling and still demonstrate compliance with the agreement. This is possible because the conversion options attached to the debt allow bondholders to share in gains the stockholders would realize by deviating from the collusive arrangement. The extent to which debt can be increased above the ceiling is a positive function of the number of shares of common stock obtained after exchange of the convertible securities.
If we assume that firms use capital structure to indicate compliance with a collusive arrangement, the theoretical arguments imply two empirically testable hypotheses. The first is that leverage should vary positively in response to changes in the price elasticity of demand. The second is that leverage should be a positive function of the number of new equity shares created if bondholders were to exercise all their convertible security options.
We test both of these hypotheses using data that measure financial leverage, price elasticity of demand, and convertible security usage for seven domestic integrated mill steel firms. Our sample period is 1947 through 1980. These firms comprise a relatively concentrated sector of the American steel industry and are considered to have maintained some type of collusive arrangement throughout this entire period. We find the steel industry especially interesting since there is continuing uncertainty about how it was able to maintain collusion after 1959, when the US Steel Corporation's overtly collusive formal price leadership, the basing point pricing system, collapsed. This sample provides an opportunity to examine firm behavior towards collusion before and after the breakdown of this system. We anticipate that prior to the demise of basing point pricing, subtle forms of collusion would not have been necessary, given the strong position held by US Steel. However, in the following years, we anticipate that firms operating in the industry used and maintained some form of tacit collusion. …