Asian Review of Financial Research Vol.26 No.1 pp.67-100
Hedging with Property-Liability Insurance and Firm Value : Evidence from Korean Firms
Key Words : Property-Liability Insurance,Hedging,Firm Value,Financial Constraints,Investment Expenditures
Previous studies on risk management, particularly on the relationship between hedging activity and firm value, have used derivatives tradings as a proxy variable for hedging. Yet, the results from theses tudies have shown some inconsistency, weakening the argument that such type of activity as a form of risk management can actually boost firm value. To the best of our knowledge, Zou (2010) is the first paper of such kind that examines the impact of hedging activity on firm value; in particular, it analyses a property-liability insurance as a firm's hedging activity in light of risk management. Using a unique set of property-liability insurance data from China, Zou (2010) examines whether propertyliability insurance can enhance firm value. Zou is one of very few in this line of research thus far. Such lack of research on the effect of hedging, especially using property-liability insurance, on firm value can be best explained by the limitation of data availability on firms'insurance in the U.S. and Europe. In China, on the other hand, for the past decade, one of the interesting developments in the global economy is the liberalization of China's insurance market. Despite much change in the industry, many Chinese insurance companies still do not take fully into account the level of risk exposure in setting insurance prices (Zou, Adams, and Buckle, 2003). In other words, insurance premiums paid by firms in China don't reflect the level of hedged risk. Evidently, this calls for more compilation of different sets of data from global firms to further reinforce the hedging effect of insurance on firm value. Therefore, this paper applies Zou (2010)'s model to a sample of Korea's publicly listed non-financial firms. Unlike the case in China, Korean insurance companies determine the insurance prices mainly based on the firms' level of risk exposure, there by fully reflecting the level of hedged risk of each firm they insure. It is reasonable to expect that, therefore, using a sample of property-liability insurance from Korea will provide accurate and reliable results in examining the hedging effect on firm value. In measuring the effects of hedging activity on firm value, this study uses Pooled OLS, Fama and MacBeth (1973) cross-sectional regression, and fixed effects panel regression. In all three regression models, Tobin's Q as a proxy variable for firm value has a positive and statistically significant relationship with the level of property-liability insurance use. This result indicates that firms with more insurance use for hedging activity have higher firm value. This study also adopts the instrumental variable approach to address the endogeneity of insurance purchase. This result is consistent with the findings of three previous estimation methods. Based on the pooled OLS result, the average hedging premium is estimated at about 2.68%. This average hedging premium is much higher than the one reported in Zou (2010), which is about 1.47%. This study also runs regression models, considering the lag effects since the insurance use in current year as well as in previous year can affect firm value. As expected, firm value in current year has a positive and statistically significant relationship with the level of insurance use in previous year. This paper also tests whether financially constrained firms have a higher average hedging premium. Several studies (e.g., Froot, Scharfstein, and Stein, 1993; Smith and Stulz, 1985; Stulz, 1996; Ross, 1997; Leland, 1998) have argued that as firms reduce the variability of future cash flow through their risk management strategy, they can overcome their financial constraints and consequently enhance their firm value. Assuming the validity of these arguments, we can expect that financially constrained firms have higher value premiums associated with hedging than financially unconstrained ones do. In our paper, as a proxy variable for internal financial constraints, we use a dividend dummy variable (e.g., 1=if a firm pays dividends less than median firm does; 0=otherwise) and a firm size dummy variable (e.g., 1=if a firm size is less than median one; 0 =otherwise) for external financial constraints. We then run regression models with these dummy variables. As expected, we find that financially restricted firms have a higher value premium associated with hedging than financially unrestricted ones. Therefore, our finding contributes to the risk management literature in providing clear-cut direct evidence that hedging activity via property-liability insurance can enhance firm value. In addition, this study tests whether firms with large investment expenditures have higher average hedging premium than ones with small investment expenditures. According to several studies (e.g., Froot, Scharfstein, and Stein, 1993; Stulz, 1996; Ross, 1997), firms can reduce their taxes and overcome an under investment problem through the hedging activities. In this paper, we use a capital expenditure dummy (e.g., 1=if a firm's capital expenditures are more than median firm does; 0=otherwise) as a proxyvariable for firms with large capital expenditures and a R&D and advertising spending dummy variable(e.g.,1=if a firm's R&D and advertising spending are more than median firm does; 0=otherwise) for firms with large R&D and advertising spending and then run regression models with these dummy variables. According to the results, a firm with large expenditures has a higher value premium associated with hedging than one with small expenditures. Our paper contributes to research in risk management by providing richer empirical evidence and practical application of risk management theories. This paper ultimately aims to narrow the gap in findings between theoretical and empirical researches.