Introduction Investors hold a set of information that generates expectations concerning risk and earnings. In an efficient market, stock prices quickly reflect all of the relevant information involving a firm. Several studies have examined the type of information investors consider relevant for pricing decisions. Dividend and earnings announcements and stock splits have been extensively covered (Charest, 1978; Aharony and Swary, 1980; Asquith and Mullins, 1983). Other studies have addressed the impact of losses on firm value. Sprecher and Pertl (1983) and Davidson, Chandy, and Cross (1987) found a negative relationship between large losses and firm value. Reilly and Drzycimski (1973) found that stock prices adjust immediately following the announcement of major world events. Shelor, Anderson, and Cross (1990) examined the effect of the October 17, 1989, California earthquake on the stock value of firms in the real estate industry. They concluded that the earthquake conveyed important new information to the market that was reflected in significant negative stock returns among real estate firms operating around San Francisco (the area sustaining the most damage from the earthquake). Real estate-related firms operating in other areas of California were generally unaffected by the earthquake. The occurrence and timing of the earthquake could not be anticipated and, thus, instantaneously introduced new and relevant information to the marketplace. The large negative returns for the sample indicated that investors viewed the earthquake as a signal of unfavorable financial conditions for the real estate industry in the San Francisco bay area. The market response to the earthquake among other California firms was consistent with that observed for an event that has no impact on market valuation. These results provide evidence that the market may discriminate among firms in relation to their geographic risk exposure. Such behavior is expected in a rational and efficient market. In a related article, Shelor, Anderson, and Cross (1992) examined the market response of property-liability insurers around the earthquake. In contrast with the real estate-related firms, the property-liability industry demonstrated a significant positive response to the earthquake, indicating that investor expectations of higher demand for insurance (positive effect) may have more than offset the potential earthquake losses (negative effect). Aiuppa, Carney, and Krueger (1993) also explored the impact of the earthquake on property-liability stock values and found a similar positive response. These results suggest that the two industries were affected by the earthquake in different ways; that is, the California earthquake stimulated industry-specific responses. This conclusion, in part, provides the motivation for this study involving the examination of the effect of Hurricane Andrew on property-liability stock values. Research Question In August 1992, Hurricane Andrew struck South Florida and Louisiana with sustained winds of 138 miles per hour. The magnitude of this storm's devastation made it the costliest natural disaster in U.S. history. The property-liability insurance industry estimates that Andrew caused more than $20 billion in property damage.(1) This dwarfs the $7 billion in claims paid after Hurricane Hugo (the previous costliest natural disaster) struck North Carolina and South Carolina in September 1989. The effect of catastrophic property damage on the value of property-liability insurers is the subject of this study; specifically, does the market demonstrate an ability to discriminate by the degree of loss exposure of property-liability insurers around significant hurricanes (such as Andrew)? Although some insurers have heavy loss exposure in certain geographic areas of the country, other firms have little or no business in the affected areas. If the market is rational and efficient, then it would be surprising to find that unexposed or lightly-exposed firms experience a similar hurricane-induced market reaction to that of heavily-exposed firms. …
Read full abstract