Date of Award


Degree Type


Degree Name

Doctor of Philosophy



Major Professor

George Philippatos

Committee Members

Robert Bohm, Ramon DeGennaro, Rudy Santore, Victor Stango


The relationship between capital structure and product market competition is examined using a theoretical model and two econometric analyses. In an extension of Glazer (1994), a theoretical model is derived that allows a quantity leader and follower to issue debt and then twice play a sequential product market game, after which each firm must either repay its debt in full or go bankrupt. It is demonstrated that the follower maximizes operating profit irrespective of capital structure but that the levered quantity leader in every period produces more than the Stackelberg profit-maximizing level of output. As such, the industry characterized by a financially levered leader and follower is more competitive than it otherwise would be.

Simultaneous equations models consisting of a demand and supply relation are used to analyze monthly data from the domestic steel industry so as to test whether the industry’s increased reliance on debt finance over the period 1958 to 1981 affected competition in the market for steel. The supply relation, which follows from the assumption that firms simultaneously select output in order to maximize profit, is augmented with a sales-weighted debt to market value ratio. Two-stage least squares (2SLS), weighted two-stage least squares (W2SLS), and iterative weighted three-stage least squares (IW3SLS) regressions are estimated. Results from all regressions show a statistically significant and positive relationship between the sales-weighted debt-value ratio and the price of steel, which suggests that increased debt finance served to reduce competition in the domestic steel industry over the sample period.

In light of the fact that U.S. Steel’s market share over the sample period was significantly higher and less volatile than that of any other integrated producer, the second econometric model tests the null hypothesis of quantity leadership, using insights from the theoretical model. Two supply relations, one for the leader and one for the follower, are derived and estimated. To account for the possibly endogenous decision on the part of U.S. Steel to issue debt, a binomial probit is estimated and its fitted probabilities are included as a predetermined variable in the leader’s supply relation. Results show that one must reject the null hypothesis of quantity leadership and that U.S. Steel’s decision to issue debt had a positive but statistically insignificant effect on the composite steel price.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Economics Commons