Abstract
Ingersoll (1977a) and Brennan and Schwartz (1977) show that, in perfect markets, managers maximize common stockholders' wealth by calling a convertible security as soon as its conversion value exceeds its call price. Nevertheless, Ingersoll (1977b) and Mikkelson (1985) report that managers tend to delay calling their firm's convertible securities until conversion value is well in excess of call price. I investigate two explanations of delayed calls of convertible preferred stock. First, Jaffee and Shleifer's (1990) call failure hypothesis states that managers delay calls to assure forced conversion. When a firm must give advance notice of a call, conversion value can fall below call price during the notice period. In this case, the call fails and the firm must quickly raise cash to pay the call price. Delaying calls "guarantees" forced conversion and avoids the costs of call failure. I test whether factors that increase the probability of call failure can explain excess conversion values on call announcement dates. I find very little support for the call failure hypothesis. Next, I test the Dunn and Eades' (1989) passive investor hypothesis. According to the passive investor model, firms do not follow the perfect market call policy because some investors rationally deviate from the perfect market voluntary conversion strategy. Investors who do not convert to receive large common dividends create cash flow savings for the firm. These savings can be had only if the firm delays calling. Tests offer weak support for the passive investor model. I show that convertible preferred stock often is priced less than conversion value, but not when the passive investor model predicts. I also find positive, but insignificant, cash flow benefits for firms that delay calls. Finally, I find that preferred stock that should be converted has less systematic risk than common stock.
Dye, Richard Timothy (1993). Why are calls of convertible preferred stock delayed? : tests of the passive investor hypothesis and the call faliure hypothesis. Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -1526965.