Casualty Actuarial Society

Research

Risk Assessment Database

Home | About RPP | Suggest New Item | RPP Results | RAD Database

RPP Results

Management Summary | Thematic Categories | Detailed RPP II Report [pdf] | Presentation [pdf]

Management Summary

Our literature review covers 961 references. The opinion of 51 colleagues from academia and practice was incorporated in the review document. As a main result we find that actuarial and financial views of how to price risk are still converging, but additional factors have been added to the discussion such as new risk measures, new valuation techniques, behavioral aspects, or emerging risks. In the aftermath of the financial crisis systemic risk, liquidity risks, and implications from the crisis are discussed. Throughout this report the five conclusions from RPP I are revised and five new conclusions are added to the discussion. Furthermore, five areas for future research are identified.

Revision of key conclusions from RPP I

(1) Financial vs. actuarial approaches: There is an ongoing consolidation between financial and actuarial literature with regard to pricing of insurance contracts. Both fields acknowledge the role of systematic and non-systematic risk in the pricing of insurance contracts.

(2) Fair value of the insurance premium: Theoretical models as well as empirical tests have confirmed that given the real-world market imperfections, the price of insurance should be a function of the (1) expected cash flow with adjustments for systematic risk, (2) production costs (i.e. expenses), (3) default risk, and (4) frictional capital costs. By-line adjustments should be integrated depending on the cash flow pattern of the liabilities.

(3) General finance: It is accepted that the single beta CAPM cannot adequately price financial contracts. Asset pricing models were systematically expanded to account for new aspects, e.g., liquidity risk or behavioral aspects. Empirical validation is still ongoing. All these aspects are of high relevance for the insurance industry, but have not been investigated in an insurance context yet.

(4) Capital allocation: Capital allocation is still controversially discussed in literature. More than 20 new approaches were proposed in recent literature and critically reviewed in the light of economic and mathematical principles. Some authors consider the Myers and Read (2001) model as a benchmark model, while others believe that this model is inaccurate. Capital allocation remains a field of active discussion in academia and practice.

(5) Risk transfer: Various papers have theoretically and empirically confirmed the assertion that default risk is recognized in pricing risk transfer to the policyholder.

Extension of key conclusions from RPP I

(6) Use of market consistent valuation techniques: Practitioners are increasingly using market consistent valuation techniques, for example in the context of regulation (Solvency II, Swiss Solvency Test) and public disclosure (IFRS, MCEV). The new valuation techniques reflect the theoretical conclusions on the price of insurance (see, e.g., conclusions number (1) and (2)).

(7) Increasing importance of holistic risk management involving classical techniques as well as new product categories: Market consistent valuation reveals the volatility of the insurer's business model and calls for holistic risk management. In this context we see an increasing role of both classical risk management techniques (e.g., risk mitigation) as well as new means (e.g., reinsurance and alternative risk transfer) to manage risk in a world of market consistent values.

(8) New risk measures and new risk categories: The last decade has seen the success story of quantile based risk measures (value at risk, expected shortfall) and generalizations of these (spectral, distortion). New risk categories (operational risk, systemic risk) have been introduced in academic literature and their limitations are discussed.

(9) Emergence of behavioral insurance: First steps have been taken towards behavioral insurance, a new area of literature that may bridge the gap between theoretical models and real world outcomes. Many researches address default risk and complement findings of theoretical models.

(10) Reinsurance and alternative risk transfer: The convergence of (re-) insurance and capital markets through alternative risk transfer (ART) has been one of the most important economic developments of the past decade. The market for ART is, however, still behind the expected capacity and has suffered from several setbacks. Recent literature analyses reasons for market failures (e.g., diversification trap) and alternative product innovations (e.g. hybrid cat bonds) to increase volume of the ART market.

Future Research

(1) Pricing and cost of capital: It is well known that the classical CAPM is insufficient to estimate costs of capital and Fama and French and Rubinstein-Leland are better models for this purpose. But much more research has been done in financial economics in recent years, with unclear implications for pricing of insurance. Are there other factors that we need to take into consideration, e.g. liquidity risk, credit risk, operational risk, or behavioral aspects such as time varying risk aversion? A systematic analysis of asset pricing theories in an insurance context could thus constitute a major empirical research agenda.

(2) Capital Allocation: Dozens of capital allocation approaches are discussed in literature and adding another one will only be of very limited value. It might rather be helpful to empirically validate the usefulness of different capital allocation approaches. Some authors see the Myers and Read (2001) approach as a best practice, while others think that this model is inaccurate. Which model is the best?

(3) ERM, modelling of risk, and dependencies: Various empirical questions surrounding ERM need to be analyzed. First, the value added by ERM is an empirical question that is unanswered yet. Second, there are many models for depiction of dependencies, but no empirical evidence for their validity. Third, the robustness of risk measures should be tackled empirically. Finally, the consistency in risk management must be addressed.

(4) Financial crisis and systemic risk: The recent financial crisis has raised important questions that need to be answered by researchers. First, does existing regulation accelerate a crisis? Second, what is the role of insurers in the highly connected financial services industry? And is an insurance run possible or not?

(5) Analysis of new insurance markets and products: Impressive potential, but in reality a rather small market. How can we eliminate the market failure in ART? What is the capacity of the ART market? Finally, emerging insurance markets are future growth markets, but we still do not know enough about insurance business in these markets.