Corporate finance over the past 25 years - Financial Management Silver Anniversary Commemoration

Financial Management (Financial Management Association), Summer, 1995 by Michael J. Brennan

Asquith and Mullins (1983), using the now common event study methodology, found that share prices tended to fall on the announcement of a common stock issue. This raised the fundamental question of why managers should take actions that impoverished the shareholders. Myers and Majluf (1984) presented an elegant explanation in terms of adverse selection. In their model, unlike those of LP and Ross, privately informed managers were assumed to act in the interest of (old) shareholders.(10)

The underpricing of initial public offerings was recognized as early as 1973 by Logue, but it was not until 1986 that a plausible account of this phenomenon was offered by Rock in terms of the adverse selection problem faced by uninformed investors. This explanation finds empirical support in Koh and Walter (1989), and more recent attempts to explain underpricing as a signal of quality have met with less empirical success.(11) The Rock model would seem particularly applicable to the institutional arrangements in Britain and those Commonwealth countries that have adopted the British system of a formal public offering and less applicable in the book-building environment of the U.S. than the model of Spatt and Srivastava (1991), which emphasizes the incentives for potential investors to reveal their valuations truthfully, or the model of Welch (1992), which analyzes the consequences of approaching potential purchasers of an issue seriatim. Both types of models assume strategic behavior on the part of either underwriters or investors, and the differing models point to the importance of attention to details concerning institutional mechanism.(12) The recently noted apparent long-run overpricing of both seasoned issues and initial public offerings poses a continuing challenge to theorists.

Given that issuers of new securities face an adverse selection problem, it was natural to consider how firms could signal their types by the choice of securities or other means. Masulis (1980) had provided evidence that pure capital structure changes brought about by exchange offers caused stock price changes,(13) Brennan and Kraus (1987), Constantinides and Grundy (1989), and Heinkel (1982), all developed signaling models in which a firm could reveal its type by its choice of financing package.

The empirically documented information content of dividend announcements also provided a natural target for the construction of signaling rationales. Bhattacharya (1979) and John and Williams (1985) developed models that relied on the taxability of dividends, while the Miller and Rock (1985) model rested on a link between dividends and investment to provide the required signaling costs. All these models assume implicitly that the informed insider places positive weight on both the current and future stock prices.

Other contexts in which signaling models have been developed are the choice of debt maturity structure when the firm has private information about its future credit rating (Flannery, 1986) and tender offers when the bidding firm has private information about synergy gains (Hirshleifer and Titman, 1990). Brennan (1990) and Stein (1988) have also shown that signaling considerations may influence not only financial decisions but also real investment decisions, if the manager is concerned about the current level of the stock price.