Climate Risk in Actuarial Work

Climate change introduces non-stationarity into historical loss data, which violates assumptions underlying many traditional models. Warmer temperatures increase the frequency and severity of hurricanes, wildfires, and flooding. Actuaries must adjust catastrophe models, trend assumptions, and reserve projections to account for shifting risk profiles over multi-decade horizons.

The fundamental problem is that actuarial science has traditionally relied on the assumption that the past is a reasonable predictor of the future. With climate change, this assumption breaks down for weather-sensitive lines of insurance. A homeowners insurer in Florida cannot simply use the last 30 years of hurricane data to price policies for the next 10 years, because the underlying climate conditions are shifting. Sea surface temperatures are rising, which provides more energy for hurricane formation, and sea levels are increasing, which amplifies storm surge losses even when storm intensity remains constant.

Physical Risk vs. Transition Risk

Climate risk is divided into two broad categories that the exam expects you to distinguish. Physical risk refers to the direct impact of changing climate conditions on insured losses. This includes increased frequency and severity of extreme weather events, rising sea levels, changing precipitation patterns, and prolonged heat waves. Physical risk directly affects property insurance, agriculture insurance, and life insurance (through heat-related mortality).

Transition risk arises from the economic and policy changes associated with moving toward a lower-carbon economy. This includes regulatory changes such as carbon taxes or emissions caps, technological shifts like the decline of fossil fuel industries, and market shifts in asset values as investors reprice carbon-intensive assets. For insurers, transition risk primarily affects investment portfolios and liability exposures for directors and officers insurance. On the exam, be prepared to give examples of each type and explain how they interact.

Approaches for MAS-II

Representative Concentration Pathway (RCP) scenarios from climate science provide frameworks for projecting future conditions. Actuaries translate these into loss distributions using adjusted catastrophe models. Trend analysis with changepoint detection helps identify when historical patterns shift. For the exam, understand how to incorporate climate scenarios into pricing and reserving, recognize the difference between physical and transition risks, and explain why simple extrapolation of historical trends may understate future losses in climate-sensitive lines.

Climate Scenarios and RCPs

The Intergovernmental Panel on Climate Change (IPCC) defines Representative Concentration Pathways that describe different trajectories of greenhouse gas concentrations. RCP 2.6 represents aggressive emissions reduction, RCP 4.5 represents moderate mitigation, and RCP 8.5 represents continued high emissions. Each pathway implies different temperature increases, sea level rises, and changes in extreme weather frequency over the coming decades.

Actuaries use these scenarios to stress test catastrophe models. For example, under RCP 8.5, hurricane wind speeds in the Atlantic might increase by 5 to 10 percent by 2050, which translates into a disproportionately larger increase in insured losses because wind damage scales approximately with the cube of wind speed. A 10% increase in wind speed can produce a 30% or greater increase in damage. Understanding these nonlinear relationships is important for exam questions about climate risk amplification.

Adjusting Catastrophe Models

Standard catastrophe models (such as those from AIR, RMS, or CoreLogic) use historical climate data to simulate thousands of potential event years. To incorporate climate change, actuaries can apply adjustment factors to the frequency and intensity of simulated events. For hurricanes, this might mean increasing the annual frequency of Category 4 and 5 storms by a factor derived from climate projections. For wildfire, it might mean expanding the geographic footprint of fire-prone areas based on projected changes in temperature and precipitation.

A simpler approach is trend analysis with changepoint detection. If historical loss data shows an upward trend that cannot be explained by exposure growth or inflation alone, a changepoint analysis can identify when the trend accelerated. Methods like the CUSUM test or Bayesian changepoint detection can pinpoint the year when the loss pattern shifted. Post-changepoint data may be more relevant for pricing than the full historical record.

Impact on Reserving

Climate change also affects loss reserves. If claim severity is trending upward due to more intense weather events, traditional reserving methods that assume stable loss development patterns will understate ultimate losses. Actuaries may need to apply explicit trend adjustments to development factors or use stochastic reserving methods that incorporate changing trend assumptions. For long-tailed lines where reserves extend many years into the future, the impact of climate change on reserve adequacy can be substantial.

Common Pitfalls

A major pitfall is dismissing climate risk as too uncertain to model. While significant uncertainty exists, the exam expects you to work with that uncertainty rather than ignore it. Scenario analysis is the appropriate tool when point estimates are unreliable. Another common mistake is focusing exclusively on catastrophe lines. Climate change also affects workers compensation (through heat-related illness), health insurance (through changing disease patterns), and life insurance (through temperature-related mortality). Think broadly about which lines are exposed.

Study Tips

For MAS-II preparation, focus on understanding the conceptual framework rather than memorizing specific climate projections. Practice explaining how you would modify a traditional actuarial model to account for climate change, what data sources you would use, and what limitations your approach would have. Review the distinction between physical and transition risks and be ready to classify specific scenarios into one or both categories. Understand how scenario analysis differs from traditional best-estimate pricing and why it is necessary when dealing with non-stationary risks.

Climate Risk and Long-Term Actuarial Projections