Volume 11, Issue 2 -

In this paper, we propose an integrated approach to adjust the premium relativities in a bonus-malus system by using the information of the first claim time (expressed in terms of sub-period in a year) and the number of claims reported by individual policyholder. We provide a formal representation for the newly proposed structure and derive the analytical expressions for the adjusted premium relativities. Other things being equal, a lower adjusted premium relativity is imposed for an earlier sub-period of the first claim made, whereas policyholders with more claims are subject to a higher adjusted premium relativity.

The linear regression model is the dominant tool employed in applied risk and insurance research. Based on my 2016 APRIA lecture at Chengdu, China, I illustrate the simple geometry of the linear regression model, as well as some standard results from it: omitted variable bias (OVB), classical measurement error (CME), simultaneous equation models (SEM), and instrumental variable estimation. Instrumental variable estimation solves OVB, CME, and SEM problems by constructing similar triangles to retrieve consistent estimates. I apply these tools by estimating the determinants of the Witt and Mehr awards given annually for Journal of Risk and Insurance articles, as two examples. The Witt vs. Mehr awards also contrasts short-term scholarly recognition (Witt) versus long-term scholarly recognition (Mehr). The comments made here apply to other paper awards, such as those presented by the Asian Pacific Journal of Risk and Insurance. I also present a simple index function based on the classical Gini index (hence, this new index is denoted as the regression gini index, RGI) useful for comparing two regression models, and apply this to explain the empirical difference between the determinants of the Witt and Mehr awards.

This study gives an empirical analysis of residential insurance demand-side reactions after an earthquake disaster using survey data. The paper discussed the study hypothesis from economic analysis perspective with significance econometric tests to explain how insurance demand for residential property changed post-catastrophe. Our empirical results observe higher risk perception from those who have had prior experience of catastrophes than those who have not. This positively influence the demand for residential insurance cover in the aftermath of a natural disaster with higher demand observed in the regions with higher seismic risk. These results support the research hypothesis and are consistent with the findings in our literature. A change in insurance level is less likely when the cost of the insurance is high, when the expected loss is low, and individuals becomes wealthier. We also find evidence of effects that remain to be explained, such as the greater sensitivity to both cost of insurance coverage and risk perception than to the size of the potential loss. We observe a positive relationship between household characteristics and the degree of risk aversion and how this changes with change in level of income, change in value of insured assets and change in insurance premiums. An increase in individual’s income alone has no major effect on insurance demand if the premium rates are within the demanders’ price range. However, positive loading of premiums to reflect transaction costs and possibility of adverse selection might affect insurance coverage level if premium rates become too high. In such cases, the direction of the effect depends on whether an increase in income increases both the premium rates and the insured asset and on whether the insurance demander has increasing or decreasing absolute risk aversion.

This study examines diversification-performance relationship in the U.S. property-liability insurance industry over the period of 1996–2010. Unlike prior studies that rely on the conditional mean estimation method, we employ quantile regression, which captures the heterogeneous effects of diversification on conditional return distribution. The results show that diversification does not necessarily drive down risk-adjusted returns and its effects vary along return distribution. We find that there is a diversification discount for firms in the lower levels of return distribution, whereas a diversification premium exists for firms in the upper levels of return distribution. We provide evidence that the relations between risk-adjusted returns and other explanatory variables are not constant, but vary over the quantiles of return distribution. Our results are robust to alternative measures of an insurer’s performance and product diversification.