بیمه بازنشستگی و مالیات شرکت در صنعت بیمه زندگی در انگلستان

There are two main tax-related arguments regarding the use of reinsurance – the income volatility reduction and the income level enhancement arguments. The income volatility reduction argument contends that firms facing convex tax schedules have incentives to hedge in order to reduce the volatility of their annual taxable income and thereby lower expected tax liabilities [Smith, C.W., Stulz, R.M., 1985. Optimal hedging policies. Journal of Financial and Quantitative Analysis 19, 127–140]. The income level enhancement argument, advanced by Adiel [Adiel, R., 1996. Reinsurance and the management of regulatory ratios and taxes in the property–casualty insurance industry. Journal of Accounting and Economics 22, 207–240], is more specific to hedging via reinsurance. This perspective holds that reinsurance enhances current reported earnings via the receipt of reinsurance commissions and so increases tax liabilities. Consequently, insurance firms with high marginal tax rates should use less reinsurance than those with low marginal tax rates if tax matters. Prior studies using data on financial derivatives have produced mixed results on the validity of the first argument, while Adiel (1996) finds the second argument insignificant in his study of the use of reinsurance by a sample of United States (US) property–liability insurance firms. This study tests the two tax-related arguments using 1992–2001 data for a sample of United Kingdom (UK) life insurance firms. We find that UK life insurers with low before-planning marginal tax rates tend to use more reinsurance; in contrast, tax convexity is found to have no significant impact on the purchase of reinsurance and so the volatility-reduction argument is not supported.

The management of corporate taxes has been examined in the banking industry (e.g., Scholes et al., 1990, Beatty et al., 1995 and Chen and Daley, 1996) and in the insurance industry (e.g., Grace, 1990, Petroni, 1992 and Cummins and Grace, 1994). Research in the United States (US) property–liability insurance industry (e.g., Petroni and Shackelford, 1995, Petroni and Shackelford, 1999, Derrig and Ostaszewski, 1997, Gaver and Paterson, 1999, Ke et al., 2000 and Petroni et al., 2000) suggests that managers can manage corporate taxes and/or meet regulatory capital (solvency) requirements through reserving and asset-liability management. Other research finds that these business objectives can also be achieved by transferring underwriting risks to other insurance entities through reinsurance treaties (e.g., see Hoerger et al., 1990, Berger et al., 1992 and Adiel, 1996).3 Indeed, since reinsurance is an important contingent financing (capacity enhancing) mechanism for primary carriers of insurance (Mayers and Smith, 1990) and taxes are potentially important determinants of firms’ financing decisions (Graham, 1996a and Graham, 2003), the reinsurance-tax relation in the insurance industry is an important area for empirical research. The corporate risk management literature (e.g., Smith and Stulz, 1985, Mayers and Smith, 1990, Garven and Louberge, 1996, Garven and Lamm-Tennant, 2003a and Graham and Smith, 1999) reports that if firms face convex (i.e., progressive) tax schedules, then risk management/hedging activities that reduce the variability of taxable income can help lower the expected net present value of future taxes (hereafter described as the income volatility reduction argument). Some studies test the income volatility reduction argument using data on financial derivatives, but produce mixed results.4 For example, Barton, 2001 and Dionne and Garand, 2003 cite supporting evidence, while Tufano, 1996 and Graham and Rogers, 2002 find no support for the above view. Not surprisingly, Graham and Rogers (2002, p. 816) state that “ … (whether) companies can reduce expected tax liabilities by hedging to reduce taxable income volatility is an important but unresolved hypothesis.” In their analysis of the purchase of reinsurance by a sample of US property–liability insurance companies, Mayers and Smith, 1990 and Garven and Lamm-Tennant, 2003b did not test the income volatility reduction argument of tax due to data limitations. In this study, we test the income volatility reduction argument for corporate hedging using data on reinsurance purchases for a sample of UK life insurers over the period 1992–2001. Reinsurance provides a potentially interesting setting within which to conduct this test for two reasons. First, unlike the use of financial derivatives, reinsurance cannot be used for speculative purposes and so our study offers a potentially cleaner empirical test of the income volatility reduction argument compared with prior studies using data on financial derivatives. Second, the reinsurance-tax relation is worth investigating because, in addition to the income volatility reduction effect, there is an alternative tax-related argument for reinsurance purchases – which does not normally exist in other forms of hedging (e.g., using financial derivatives). Specifically, Adiel (1996) points out that the receipt of reinsurance commissions increases the current period pre-tax earnings of insurers that purchase reinsurance and thereby increases their expected tax liabilities. Therefore, insurers facing high before-planning marginal tax rates are likely to use less reinsurance than insurers facing low before-planning marginal tax rates in order to reduce tax liabilities (hereafter described as the income level enhancement argument). We believe the income volatility reduction and income enhancement arguments are not mutually exclusive as reinsurance can affect both the level and variability of future taxable income. Hence, which tax effect predominates in the reinsurance decisions of insurance companies should be an interesting question worthy of empirical investigation.5 Examining the reinsurance-tax relation is important not only because reinsurance as an important risk management decision affects insurers’ financial strength and solvency ability, but also because many industrial companies indirectly access reinsurance via captive insurance companies.6 Indeed, Skipper (1998, p. 660) reports that over 90 percent of the top 500 companies in the US and over 80% of the top 200 companies in the UK have established captive insurance operations. Given such a widespread (indirect) use of reinsurance in the general corporate sector through captive insurance subsidiaries, the results of this study could also be of interest to parties in sectors outside the insurance industry. If the income volatility reduction argument dominates managerial decision-making, insurance companies facing a convex tax position are likely to use more reinsurance in order to reduce the expected future tax liabilities than other insurers, other things being equal. If the income level enhancement argument is valid, an insurer with a high (before-planning) marginal tax rate is likely to use less reinsurance than an insurer with a low (before-planning) marginal tax rate in order to reduce the expected tax liabilities. Our tests provide support for the income level enhancement argument but not for the income volatility reduction effect. To our knowledge, evidence consistent with the income level enhancement argument advanced by Adiel (1996) has not previously been reported in the literature. Our finding on the income volatility reduction argument corroborates the findings of some prior studies using data on financial derivatives (e.g., Tufano, 1996 and Graham and Rogers, 2002), but is different from the findings of Barton, 2001 and Dionne and Garand, 2003. One prior study that has a close connection to ours is Adiel (1996). However, several major differences exist. First, Adiel (1996) uses data from US property–liability insurers, while we use data from the UK life insurance industry.7 Second, Adiel (1996) examines the relation between insurers’ reinsurance purchases and contemporaneous marginal tax rates (to test the income level enhancement argument) and finds no significant relation. Instead, he finds that meeting regulatory solvency requirements is an important motivation for the purchase of reinsurance by his sample of US property–liability insurers. In contrast, our study concomitantly tests both the income volatility reduction and income enhancement arguments (using before-planning marginal tax rates rather than contemporaneous rates). Third, our construction of proxies for marginal tax rates is built on recent developments in tax-related studies (e.g., Graham, 1996b and Plesko, 2003). The remainder of this paper is structured as follows. Section 2 introduces background information on the taxation requirements for UK life insurers. Section 3 develops the two tax-related arguments concerning the purchase of reinsurance and discusses other determinants of reinsurance. Section 4 describes the research design, including the model specification, the measurement of the variables and data sources. Section 5 discusses the empirical results. Section 6 provides some robustness checks and further tests, and Section 7 concludes the paper.

Using data from a sample of UK life insurance companies, this study tests two tax-related arguments – the income volatility reduction and the income level enhancement arguments – in relation to reinsurance. The former argues that firms facing convex tax schedules have an incentive to use reinsurance in order to reduce earnings volatility and thereby expected tax liabilities. The latter argues that reinsurance increases current reported earnings via the receipt of reinsurance commissions and so life insurance firms with high (before-planning) marginal tax rates are expected to use less reinsurance than others if tax matters (Adiel, 1996). The primary merit of using reinsurance, an important risk management technique for insurers (and industrial companies owning captive insurance companies), is that unlike derivatives, reinsurance cannot be used for speculation. We regress incremental reinsurance on two proxies of before-planning marginal tax rates and tax convexities, together with six other independent variables. We find that UK life insurers with low (before-planning) marginal tax rates tend to use more reinsurance than insurers with high (before-planning) marginal tax rates; in contrast, tax convexity does not seem to induce reinsurance, so the income volatility reduction argument is not supported by our data. The empirical evidence on the income level enhancement argument advanced by Adiel (1996) is not previously reported in the literature. Our results suggest that for corporate hedging via reinsurance, tax matters in the sense that UK life insurers with a high marginal tax rate seem to choose to reinsure less in order to lower expected tax liabilities. Graham and Rogers (2002) conjecture that while hedging can reduce the volatility of taxable income, other methods (e.g., accounting policies) may also suffice for the same purpose and so tax convexity may not necessarily induce more corporate hedging activities. This is particularly the case in the life insurance industry where managers/actuaries have considerable discretion to smooth taxable income (e.g., through reserving) over a long contract term. This also suggests an interesting area for future research – i.e., the interaction between accounting-based earnings management mechanisms and corporate hedging policies. We note that our results might be tempered by the data limitations inherent in the current study and therefore the results need to be interpreted with some caution. For example, the proxies that we use to represent tax convexity may be subject to measurement errors. As a result, further research could explore ways of improving the precision of marginal tax rate measures, utilizing where possible simulation techniques on large cross-sectional/time series data sets. Additionally, the life insurance industry as a research setting may increase our ability to test the tax-related income level enhancement argument, given the higher level of commissions (compared with the property and liability insurance industry). However, this setting potentially limits our ability to test the tax-related income volatility reduction argument, given the generally lower level of reinsurance in life insurance, compared with property–liability insurance. Future studies may thus seek to test further the reinsurance-tax relation in the property and liability insurance industry.