Journal of Risk and Insurance最新文献

Using daily stock market data for European insurers, I investigate how a stock market contraction, as experienced during the COVID-19 pandemic, affects insurers' credit risk allocation of their corporate bond portfolio. I find that insurers shift their portfolio holdings pro-cyclically towards lower credit risk assets in the first month of the market contraction. As the crisis progresses, I find evidence for counter-cyclical, riskier investment behavior by European insurers, especially in high-yield instruments, that can neither be explained by credit rating downgrades of held bonds nor by hedging with CDS derivatives. This counter-cyclical investment behavior cannot be observed for US firms, which provides evidence for a difference in investment behavior between US and European insurers. The observed investment behavior of European insurers could be beneficial for systemic stability by attenuating price declines through insurance liquidity provision, but excessive risk-taking by insurance companies over longer periods can also reinforce systemic stress.

Variable Annuities, which comprise a substantial proportion of the retirement products sold by insurance companies, have become increasingly complex over the past decades. We investigate the drivers of the product trends. We distinguish “virtuous” innovations that expand upon the existing set of consumption paths in retirement from “obfuscating” innovations that increase complexity without clear benefits to consumers. We document a recurring pattern where, in each benefit category, obfuscating products follow the introduction of virtuous innovations. This pattern generates the overall increase in product complexity. Our results challenge prevailing perspectives on Variable Annuities in the popular press and the literature.

We analyze the relationship between stock prices and insurance accounting and compare a historical cost with a full fair value measurement approach. During our sample period, European insurers had to determine the fair value of all assets and liabilities according to the Solvency II (SII) regulation, in addition to the historical-cost-based setup of the International Financial Reporting Standards (IFRS). This alternative source of information allowed investors to update their expectations about future dividends, risks, and firm values. Comparing both frameworks, we report three findings. First, we show that the association between stock prices and SII full fair value accounting items is greater than that of IFRS historical cost measurements. Second, we find that this effect stems from unexpected news disclosed by regulatory reporting. Third, our results suggest that insurance accounting is relevant for firms exposed to lower insolvency risk and offers no additional information when the risk level is high.

We document an extreme-weather risk premium in the cross-section of stock returns. Between 1995 and 2019, stocks of domestic U.S. firms with the most negative sensitivity to aggregate storm losses earned an annual excess-return spread of more than 6 percentage points relative to those with the most positive sensitivity, a difference not explained by standard factors. Fama-MacBeth regressions confirm that more negative storm-risk betas predict higher subsequent returns. The premium concentrates in geographically exposed and historically affected firms and in institutionally held stocks, consistent with fundamental-risk and salience channels. Our results establish a link between physical climate risk, the cost of equity, and ultimately firm value.

We investigate heterogeneity in health insurance choice using data from a controlled laboratory experiment. Participants make consecutive choices from sets of insurance plans that vary in premium, deductible, and complementary coverage of illnesses. We find that there is considerable heterogeneity in how much individuals are willing to pay for certain plan attributes. To better understand these differences, we account for individual risk preferences using a rank-dependent expected utility (RDEU) model and assess the welfare effects of plan choices. At the aggregate level, we find welfare losses under both the normative RDEU model and the descriptive EV model. At the individual level, however, the results are more differentiated: for some individuals, choices are consistent with their RDEU preferences, whereas for others, choices do not fit either model, suggesting either decision errors or reliance on heuristics.

Creating new markets is a prevalent approach for implementing large social programs. Assuming firms have full information about the relevant parameters upon market inception is commonplace in the literature. In contrast, we develop an adaptive learning model with selection to study how firms' knowledge of demand and cost affects the market equilibrium. We estimate alternative learning models with data from the California ACA exchange and assess their external validity using novel data on firms' predicted costs from insurer rate filings. The learning models provide statistically significant improvements in fit relative to the standard model that assumes firms have full information. Most of the improvement results from allowing firms to learn about the relationship between demand and cost. Firms with full information can increase profit, but at taxpayers' expense. Regulation that prohibits firms from using certain consumer information to set premiums makes them react more to the information they can use.

Wildfires increased in frequency and severity over the past 30 years, raising the exposure of municipalities. We study whether municipal bond yields reflect wildfire risk and find that the municipal bond market begins pricing wildfire risk around 2000. A one-standard-deviation rise in wildfire risk leads to a 3.646 basis points increase in a bond's yield spread, increasing financing cost for affected communities. The effect on yield spreads disappears for maturities of less than 1 year or more than 15 years, supporting the view that wildfire risk is a medium-term risk. When including measures of investor attention into our analysis, we find that investor attention is associated with an additional increase in yield spreads. Fire mitigation measures, on the other hand, are associated with a reduction in yield spreads, highlighting the benefits of wildfire risk management for municipal financing.

This paper proposes a dynamic longevity risk hedging strategy for smooth survival benefit profiles of group self-annuity (GSA) schemes in the presence of population basis risk. The fund manager of GSA acts on behalf of fund participants in selecting the optimal hedge. The hedging framework is formulated as a mean-variance optimization problem, which serves as a theoretical framework for selecting the optimal hedging strategy. The hedging mechanism involves trading standardized longevity-linked securities dynamically. A semi-analytic solution to the optimal hedge ratio is derived, which enhances the numerical implementation of the strategy. Furthermore, a risk decomposition method is developed, enabling hedging of various sources of risks, such as longevity and investment risks. Numerical illustrations highlight that the hedging strategy effectively mitigates variability in survival benefits. Meanwhile, a holistic risk management framework utilizing the longevity risk hedging strategy and a target volatility investment strategy increases the fund's return per unit of risk.

We examine the systemic risk of 46 systemically important financial institutions (SIFIs), that is, 34 global systemically important banks (G-SIBs) and 12 global systemically important insurers (G-SIIs) between 2010 and 2023. We use tail risk network-based systemic risk measures for SIFIs. We find that G-SIBs' systemic risk is driven by various shocks, including the 2011–2012 Eurozone crisis, the 2018–2019 US–China trade tensions, and the 2023 US regional bank crisis. In contrast, G-SIIs' systemic risk is largely driven by the 2020 COVID-19 pandemic. Moreover, the distribution and correlation of systemic risk for G-SIBs and G-SIIs vary significantly across jurisdictions. We also find a bidirectional causal relationship between G-SIBs' and G-SIIs' systemic risk. Our findings have important implications for the tail risk independence and stability of the financial system.

We study optimal demand for insurance in a classical expected utility setting where the insured party has limited liability and has access to three different types of progressively more restrictive contracts. At one end, with no restrictions on the indemnity schedule, it is optimal to fully insure certain losses while leaving others uninsured. At the other end, if indemnity schedules and retained losses are assumed to be increasing functions of the underlying loss, the optimal insurance policies are shown to be capped deductibles. For the intermediate case when the indemnity schedule is only an increasing function of the loss, we find that optimal contracts exhibit a richer structure beyond the capped policies suggested in earlier literature. Our study extends and provides a unifying perspective on the existing literature on optimal insurance under limited liability.