Asia-Pacific Journal of Risk and Insurance

A Catastrophe Insurance System for the European Union

Milton Nektarios — Thu, 28 Jul 2011 15:06:52 PDT

This paper proposes an integrated risk management plan for catastrophe risks in the European Union, consisting of three layers. The private markets would have the first layer of responsibility, while the National Catastrophe Insurance Organizations would represent the second layer. This layer would in turn be supported by the European Group of National Catastrophe Organizations (EUROCAT), a new organization operating under the auspices of the European Commission. An approach that utilizes a pan-European reinsurance program is proven to be the most efficient solution for minimizing the total cost of catastrophe risks in the European Union. EUROCAT would be a reinsurer of last resort and provide reinsurance to qualified state or regional catastrophe insurance funds. Member-state funds would be required to adopt adequate disaster response and management mechanisms and enforce reasonable building code, land use, and mitigation efforts to minimize the amount of insured losses. As the reinsurance premiums charged by EUROCAT would be risk-based, the pricing mechanism would be used to encourage active development and enforcement of these standards.

Self Insurance and Insurance Demand under Self-Deception

Ram Ranjan — Thu, 28 Jul 2011 15:06:50 PDT

A dynamic model of self insurance and market insurance demand against uncertain natural hazards is developed where agents incur emotional costs when the true information about potential future catastrophes becomes known. Agents purposefully ignore the incoming signals about future hazards in order to avoid finding out the true information. Faced with the emotional costs associated with true information revelation, self-deception through true information avoidance effort may be the optimal strategy to maximize current and future expected utility. Government interventions for complementing private risk mitigation and adaptation through incentives and regulations may have a varying influence on such agents exhibiting emotional costs.

Going Undercover: The Paradox of Political Risk Insurance

Belinda Spagnoletti et al. — Thu, 28 Jul 2011 15:06:46 PDT

Political risk insurance is an insurance product sold to firms as a means to protect them against some political risks associated with investing abroad. This paper argues that political risk insurance may in fact be a source of political risk. Looking at cases in the Asia-Pacific region, we argue that investing firms should exercise caution when considering political risk insurance. We conclude with several recommendations for reform of the political risk insurance industry.

On Robust Testing and Modelling of Threshold-Type Non-Linearity in ASEAN Foreign Exchange Markets

Wai-Sum Chan et al. — Thu, 28 Jul 2011 15:06:43 PDT

There are many situations in which insurance companies operating in the Asian region encounter foreign exchange risk. This paper considers a stochastic threshold approach to modelling foreign exchange risks in some Association of Southeast Asian Nations (ASEAN) markets. The class of self-exciting threshold autoregressive (SETAR) models is used. These models are able to capture the exchange rate band phenomenon that has been discussed in the monetary economics literature. Both ordinary least squares and robust modelling procedures are considered.

Longevity Risk and the Econometric Analysis of Mortality Trends and Volatility

Carolyn N. Njenga et al. — Thu, 28 Jul 2011 15:06:40 PDT

Longevity risk and the modeling of trends and volatility for mortality improvement have attracted increased attention driven by ageing populations around the world and the expected financial implications. The original Lee-Carter model that was used for longevity risk assessment included a single improvement factor with differential impacts by age. Financial models that allow for risk pricing and risk management have attracted increasing attention along with multiple factor models. This paper investigates trends, including common trends through co-integration, and the factors driving the volatility of mortality using principal components analysis for a number of developed countries including Australia, England, Japan, Norway and USA. The results demonstrate the need for multiple factors for modeling mortality rates across all these countries. The basic structure of the Lee-Carter model cannot adequately model the random variation and the full risk structure of mortality changes. Trends by country are found to be stochastic. Common trends and co-integrating relationships are found across ages highlighting the benefits from modeling mortality rates as a system in a Vector-Autoregressive (VAR) model and capturing long run equilibrium relationships in a Vector Error-Correction Model (VECM) framework.

Culture Matters: Long-Term Orientation and the Demand for Life Insurance

Sojung Park et al. — Thu, 28 Jul 2011 15:06:36 PDT

A large body of literature addresses the determination of the economic, financial and institutional factors that significantly influence variations of life insurance demand across countries. Chui and Kwok (2008) included four cultural variables developed by IBM psychologist Hofstede and demonstrated that culture has a profound impact on life insurance consumption. We extend Chui and Kwok’s work by analyzing the fifth Hofstede cultural dimension: Long-Term Orientation, a variable that scores countries based on adherence to Confucian principles such as perseverance and thrift, respect of tradition and family values, and honoring of parents and ancestors. After building a database that includes values of 17 variables for 27 countries over a period of 9 years, we apply an unbalanced panel GLS regression model to prove that Long-Term Orientation has a strong positive influence on life insurance demand. Additionally, two new variables, not used in previous life insurance literature, are also found to impact life insurance demand: a modified Herfindahl index and the use of a Common Law legal system. Several robustness tests confirm the importance of Long-Term Orientation, leading to the conclusion that life insurance consumption is bound to increase rapidly in Asia, as its GDP per capita increases.

Who Uses Physiotherapy Services for Motor Vehicle-Induced Whiplash-Associated Disorders? Interrogating Motor Accident Insurance Data for 2006-2009

Karen Grimmer-Somers et al. — Tue, 01 Mar 2011 11:48:14 PST

Objectives: Little is known about who uses physiotherapy services for whiplash injuries sustained in motor-vehicle accidents. Compensable agencies around Australia are grappling with ways to identify early claimants who are likely to consume different types of health services. The objectives were to identify key characteristics of a typical user of physiotherapy services for motor-vehicle accident-induced whiplash-associated disorders.

Method: A dataset was provided by the Motor Accident Commission (MAC) in South Australia which included demographic, socioeconomic and accident information on WAD claimants from 2006-2009. Univariate and multiple logistic regression models were developed to test the association between physiotherapy service use, claimant demographics, socioeconomic status, and accident descriptors.

Results: A typical consumer of physiotherapy services for WAD was female, aged between 40 and 79 years, living in upper middle or high socioeconomic suburbs, with legal representation, and who has been assigned at least three MAC injury codes (denoting severe injury). She was the driver of the car. She was wearing a seatbelt at the time of the accident, and the accident occurred on hard dry road surfaces. Her accident was unlikely to be angular.

Conclusion: A clear profile of a WAD claimant who would consume physiotherapy services was established.

Testing for Adverse Selection in China's Auto Insurance Market

Feng Gao et al. — Tue, 01 Mar 2011 11:48:09 PST

This paper empirically examines the adverse selection models by using the data from China’s personal automobile insurance market through the standard positive-correlation test. Results show that higher-risk insurance policyholders do tend to buy insurance with a lower coinsurance rate, which supports the existence of adverse selection in China's automobile insurance market. This is quite a contrast to the findings in mature insurance markets.

A Real-Options Approach to Post-Hurricane Loss Valuation of Damage Property: Rebuild or Repair?

Risa Kumazawa et al. — Tue, 01 Mar 2011 11:48:04 PST

In this paper, we investigate the rebuild or repair decision that property owners face after damages caused by catastrophic hurricanes such as Katrina in New Orleans. In particular, we consider how the degree of risk aversion and uncertainty affect the decision-making process. A theoretical model is developed using the real-options framework of Dixit and Pindyck (1994). According to the model, the decision to rebuild a property is reached much later when there is a high degree of uncertainty over future social costs and a high discount rate. We demonstrate these effects using simulations with actual numbers from Hurricane Katrina.

Shareholder Value: The Case of Japanese Captive Insurers

Yuji Maeda et al. — Tue, 01 Mar 2011 11:47:58 PST

We examine how a Japanese firm can derive the most value from a captive insurance company. We use actual data on losses, captive operating costs, interest rates and foreign exchange rates to simulate cash flows from a hypothetical pure captive. We examine three domiciles popular with Japanese firms: Bermuda, Hawaii and Guernsey. We find a high likelihood that the captive generates economic value for its Japanese parent firm especially when it operated over multiple years. We find that when the captive reinsures its entire book of business thus acting as a pass-through vehicle to the global reinsurance market, it normally delivers economic value but near to break-even. We also find that the captive can generate high levels of economic value but only by taking on higher levels of operating risk. We find that the value-maximizing strategy is for a Japanese corporation to establish its captive in Bermuda while the risk-minimizing strategy is to establish its captive in Guernsey. We also find that Bermuda would be a better choice than Hawaii for Japanese captives from the viewpoint of risk-return tradeoff.

IFRS Convergence: The Role of Stochastic Mortality Models in the Disclosure of Longevity Risk for Defined Benefit Plans

Yosuke Fujisawa et al. — Tue, 01 Mar 2011 11:47:53 PST

In recent years, the International Accounting Standards Board (IASB) and its International Financial Reporting Standards (IFRSs) have made great strides toward achieving global accounting convergence. Various countries, including Japan and Canada, are either adopting or converging their national standards with IFRSs. The IASB is now undertaking a comprehensive review of the accounting standards on post-employment benefits, an important part of which is about quantitative disclosures of longevity risk. In this paper we examine how stochastic mortality models may assist with such disclosures. Specifically, we present three concepts that can help defined benefit plans identify the materiality of their longevity risk exposures: (1) longevity value-at-risk, (2) probability of longevity deficit, and (3) the probabilistic corridor rule. We illustrate these concepts with a hypothetical pension plan in Japan.

History of Insurance, Market Development and Regulation in Seven Least Developed Countries in Asia: Afghanistan, Bangladesh, Bhutan, Cambodia, Laos, Myanmar and Nepal

W. Jean Kwon — Tue, 01 Mar 2011 11:47:41 PST

This paper examines the history of market development and regulation in the insurance markets of seven Least Developed Countries in Asia. Historical analysis shows that most of the markets were first developed to service expatriates and the interests of expatriates then operating in those countries but disappeared or were nationalized soon after their independence as a sovereign state. All countries have adopted or are in the process of adopting market-oriented economic policies and insurance business has begun to re-emerge. Findings from the investigation of insurance regulation and supervision show that all governments apparently apply principle of national treatment in the markets, but the application is not perfect as they commonly not only impose no statistically justified rules but also apply politics-oriented guidelines and unclear capital, investment and accounting regulations on licensed companies. The reinsurance markets are either undeveloped or are subject to compulsory cession to government insurer or its designee. They also lack clear exit (insolvency) regulation and policyholder protection guidelines.

A Simple Metric for Gauging Risk Aversion

Joseph G. Eisenhauer — Wed, 14 Jul 2010 08:04:50 PDT

The underlying rationale for insurance purchases and other forms of risk management is aversion to risk, yet the measurement of risk aversion has been inappropriately focused on risks of little or no consequence. The present paper demonstrates that a quantitative measure of large-scale risk aversion can be constructed with elementary mathematics, facilitating the solution of numerical problems. We illustrate its use with hypothetical examples and empirical survey data.

Survival Mixture Model for Credit Risk Analysis

Leo S. F. Mo et al. — Wed, 14 Jul 2010 08:04:49 PDT

The survival mixture model, which is an extension of the ordinary survival model that allows the existence of a fraction of the borrowers to be risk-free, is applied to credit risk analysis. In a regression setting, the effect of borrowers' characteristics on both the risk-free probability and default risk can be assessed simultaneously. Using the C statistic as a measure of accuracy, the survival mixture model shows improved power to discriminate between ‘good' and ‘bad' customers, when compared with other commonly used statistical models for credit risk analysis. A simulation study is conducted to assess the performance of the proposed numerical estimation method. The survival mixture model not only concentrates on the time-to-default of the borrowers, it also predicts the probability of being risk-free. It provides additional information about the borrowers' default risk in relation to their characteristics, which assists the lending institutions to better manage credit risk.

Asymptotic Tail Probability of Randomly Weighted Sum of Dependent Heavy-Tailed Random Variables

Yu Chen et al. — Wed, 14 Jul 2010 08:04:47 PDT

This paper investigates the asymptotic behavior of tail probability of a randomly weighted sum of real-valued heavy-tailed dependent random variables; the weights form another sequence random variable. Under some other mild conditions, the asymptotic relations obtained are further applied to derive asymptotic estimate for ruin probabilities in a discrete time risk model.

On Modeling Diversification Benefits in Insurance Portfolios—An Australian Perspective

Jackie Li — Wed, 14 Jul 2010 08:04:46 PDT

Empirical measurement of dependency between different lines of non-life insurance business remains an open problem. International regulatory requirements and accounting standards are moving towards increased disclosure of the variability of insurance liabilities. In estimating the variability, it is required to take diversification benefits into account, as it is generally believed that different lines of business are not perfectly dependent on one another. Due to data limitations and the complicated nature of non-life insurance business, however, direct quantification of dependency and so diversification benefits between individual lines is a difficult task. In this article, we set forth some practical considerations in modeling dependency and diversification benefits under Australian regulatory environment. We start with providing a summary of the underlying factors causing dependency and examining the reasonableness of some industry correlation figures based on these factors. We also study the correlations between the historical loss ratios of several lines of Australian business. Afterward we carry out extensive simulation studies to investigate the effects of applying some recently suggested methodologies for tackling the problem of evaluating diversification benefits.

An Equilibrium Analysis of the Insurance Market with Horizontal Differentiation

Mahito Okura — Wed, 14 Jul 2010 08:04:45 PDT

This article analyzes the life and nonlife insurance markets with horizontal differentiation and brand loyalties using a spatial competition model. We construct a three-stage game that includes sales promotion in the first stage, insurance product characteristics in the second stage, and premiums in the third stage. The analysis derives the following results. First, an insurance firm with greater brand loyalty realizes a higher degree of differentiation and sets a relatively higher premium. Second, introducing regulation in insurance product characteristics leads to a lower degree of differentiation and a lower level of sales promotion. Such regulation may then be desirable in terms of social surplus.

A Reexamination of the Relationship between Organizational Forms and Distribution Channels in the U.S. Property Liability Insurance Industry

Vincent Y. Chang et al. — Wed, 14 Jul 2010 08:04:42 PDT

How do property liability insurance companies choose their organizational forms and distribution channels? Prior studies have not yet provided a consistent conclusion. In this paper, we propose a reduced form approach to reexamine the relationship between organizational forms and distribution channels in the insurance industry, using cross-sectional data pertaining to U.S. property liability insurance companies in 2004. We adopt a conditional dependence test, which can overcome the sensitivity problem of the structural form setting. The results show that after we control for all explanatory variables, the relationship between organizational forms and distribution channels is conditionally uncorrelated. The result is consistent with Regan and Tzeng (1999), but contradicts the findings of Baranoff and Sager (2003) and Kim et al. (1996).

Some Comments on Catastrophe Risk Management and Insurance

Emilio Venezian — Tue, 17 Nov 2009 09:47:40 PST

These comments are a reaction to a panel discussion on Catastrophe Risk Management and Insurance. They point out that some measures intended to provide quick relief and remediation after catastrophic events may have the long term effect of increasing aggregate exposure to such events. In some cases, these adverse long-term effects can be mitigated.

The Cost of Delay in a Mortgage/Credit Loan Portfolio

Jiwook Jang — Tue, 17 Nov 2009 09:47:38 PST

Using an actuarial model, we examine the cost of delay in mortgage/credit loan payments. It is assumed that the default arrival process follows the Poisson process and the loss sizes are assumed to be independent and an identical truncated exponential. We also assume that the delay between default occurrence and partially (or fully) recovered payment is an independent identical truncated exponential random variable. For the recovery rate random variable, we simply use its expectation. Using the relationship between the shot noise process and accumulated/discounted aggregate losses process and applying the piecewise deterministic Markov processes theory, we obtain the explicit expressions for the expected value of losses and the expected value of part (or whole) of the loan recovered with the delay. Based on these moments, we define and predict the cost of delay in a mortgage/credit loan portfolio and their numerical examples are provided.