Contagion Effects in the Insurance Industry

Abstract

ABSTRACT This article investigates the valuation effects of equity issuances by insurance companies on rival firms. The literature has documented the existence of contagion effects associated with the revelation of information in industries characterized by asymmetric information, i.e., both the banking and insurance industries (see, e.g., Slovin, Sushka, and Polonchek, 1992; Avila and Eastman, 1995; and Fenn and Cole, 1994). We argue that because of high monitoring costs investors holding insurance company securities are rationally uninformed about the quality of insurance company portfolios. In this environment characterized by asymmetric information, if the market interprets an equity issuance by an individual insurer as a signal regarding the quality of insurance company portfolios in general, then we would expect to observe cross-firm effects associated with insurance company equity issuance announcements. Our results are consistent with the hypothesis that the announcement of an equity offering reveals informati on about the quality of both the announcing firm's portfolio and the quality of rival firms' portfolios. In addition, when we split the sample by industry grouping, we find that announcements by life insurance firms generate significant cross-firm effects. We argue that these results are a consequence of the different types of state-contingent contracts offered by insurers in different segments of the industry. Investors holding securities of companies that issue short-term state-contingent contracts find it economically rational to be relatively well informed about the quality and riskiness of an insurance company. Equity issuance announcements by insurers that offer short-term state-contingent contracts do not produce cross-firm effects. Before the well-publicized asset quality problems in the life insurance industry in 1990, it appears that the perceived risk of holding insurance company securities was low and the perceived quality of insurance company assets was high. Because of these perceptions regardin g asset quality and risk, investors were relatively uninformed. When we examine insurance company equity issuance announcements before and after 1990, we find that it is the announcement of equity offerings made by life insurance firms before January 1, 1990, which results in significant contagion effects. INTRODUCTION The passage of time has seen the evolution of a considerable literature investigating the market's reaction to securities issuance announcements. The empirical results documented in the literature are consistent with the models of Myers and Majluf (1984) and Miller and Rock (1985), which hypothesize that securities issuance announcements are viewed by the market as a signal of private information about expected future cash flows by the management of a firm. The reported announcement period effects for equity issuances of non-financial firms are approximately -3.1 percent (see, e.g., Asquith and Mullins, 1986; Barclay and Litzenberger, 1988; Mikkelson and Partch, 1986; and Masulis and Korwar, 1986). Comparable announcement period returns for utility equity offering announcements are approximately -0.75 percent (see, e.g., Asquith and Mullins, 1986; Masulis and Korwar, 1986; and Pettway and Radcliffe, 1985). The literature documents that the announcement period returns for exchange traded commercial banks are b etween the comparable announcement period returns for utilities and non-financial firms. Polonchek, Slovin, and Sushka (1989), Wansley and Dhilon (1989), and Keeley (1989) report announcement period returns for bank equity offerings of approximately -1.2 percent. Polonchek and Miller (1996) investigate securities offerings by insurance companies. They argue that private information associated with an insurance company's asset portfolio increases the market's expectation of adverse selection by the firm's managers. Consistent with this expectation, they find statistically significant announcement period returns of -3. …