The U.S. Property and Liability Insurance Industry: Firm Growth, Size, and Age

Article excerpt


The relationship between firm size, age, and growth is tested for the U.S. property and liability (P-L) insurance industry, and the determinants of firm characteristics on firm growth are analyzed. Using Heckman's two-stage methodology, this article examines the relationship between corporate growth and firm size. The relationship between firm growth and firm age is also investigated. Furthermore, to determine time-varying effects, the analysis is conducted for the different subperiods. The results of this article strongly support Gibrat's Law in the U.S. P-L insurance market for the testing periods. The results are consistent for longer time periods and for shorter subperiods. It also finds that young firms grow faster than old firms during the sample periods. Related to the determinants of firm characteristics on firm growth, insurers using less input cost tend to grow fast. Economies of scope are positively related to firm growth as well.


Over the years, it has been tested whether corporate growth rates are independent of firm size as first explored by Gibrat (1931), that is, Gibrat's Law, also known as the Law of Proportionate Effects (LPE). Various empirical studies focused on that relationship across industries. These studies provided mixed results related to Gibrat's process (e.g., Evans, 1987a; Geroski et al., 1997; Weiss, 1998; Hardwick and Adams, 2002; Calvo, 2006). Evans (1987a), for example, examined a large sample (about 20,000) of manufacturing firms for the period 1976-1982. Using data pooled across industries, he found an inverse relationship between corporate growth and firm size. On the other hand, Geroski et al. (1997) investigated large U.K. firms and found that growth rates were random factors over time and difficult to predict. Calvo (2006) tested 1,272 Spanish firms and found no support of Gibrat's Law. His results showed that small, young, and innovative firms grow faster than large, old, and noninnovative firms.

Furthermore, research in this area tests the relationship between firm growth and the age of a firm (Evans, 1987a; Variyam and Kraybill, 1992; Dunne and Hughes, 1994; Harhoff et al., 1998; Yasuda, 2005; Calvo, 2006). Generally, they found a negative impact of firm age on corporate growth. The findings are fairly consistent among different samples across countries, for example, U.S. firms (Evans, 1987a), U.K. firms (Dunne and Hughes, 1994), West German firms (Harhoff et al., 1998), Japanese firms (Yasuda, 2005), and Spanish firms (Calvo, 2006). In some cases, the negative relationship is not confirmed (e.g., Audretsch et al., 1997). In fact, Audretsch et al. (1997) found that firm growth and age were not systematically related, by analyzing retail and hospitality industries in the Netherlands.

In the insurance area, Gibrat's Law has rarely been examined and only one study, Hardwick and Adams (2002), tests the Gibrat process. Hardwick and Adams investigated the firm size and growth relationship using 176 U.K. life-insurers for the period 1987-1996. Their findings generally support Gibrat's Law, with no significant difference noted between growth rate and size for small and large life insurance firms. In addition, they investigated the determinants of firm growth using a regression model with four firm-specific variables. They found that only short-term fluctuations were key to the relationship between firm size and growth rate; however, in the long run, no specific patterns or firm-specific characteristics affected the relationship, strongly supporting LPE.

In this article, we conduct a company-level analysis for the period 1992-2001 to examine the longstanding Gibrat's Law using the U.S. property-liability (P-L) insurance cross-sectional data. In addition, the relationship between firm growth and firm age is tested. Since the P-L insurance industry has observed a different underwriting profit cycle during the sample period, time-varying effects are tested. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.