Calling Convertible Bonds too Late Can be Rational

Abstract

In this paper, we offer a theoretical explanation for the empirical puzzle that convertible bonds are called too late or too early. Contrary to the literature, we do not assume frictions to explain the observed call behavior. Instead, we assume that the firm has issued in addition to stock and convertible bonds, also straight bonds. This seemingly innocuous extension has the important consequence that a conversion can cause a wealth transfer from the stockholders to the bondholders. This wealth transfer changes the analysis and results of the seminal papers by Ingersoll (1977a) and Brennan/Schwartz (1977) substantially. First, the value of a convertible bond can no longer be determined by option-pricing methods. The strategic behavior of the convertible bond holders requires a game-theoretical analysis. Second, the negative effect of an enforced conversion on the stockholders can be avoided by calling the bonds too late. Third, a positive wealth transfer can also be produced by calling the bond too early.