Learning, Catastrophic Risk, and Ambiguity in the Climate Change Era
Property insurance markets in multiple US states are experiencing volatility in the form of increasing premiums, insurer bankruptcies, and complete withdrawal from some areas. This volatility appears linked to growing costs of natural disasters, with some of the most serious effects in states heavily exposed to hurricanes and wildfires. This paper illustrates how uncertainty introduced by a changing climate can produce abrupt changes in insurance market conditions as insurers (and reinsurers) integrate the possibility of changing climate conditions into their pricing and risk management.
The mechanism operates via changes to the updating process by which agents infer current climate conditions and, by extension, the distribution of weather risk. Key methodologies used for managing weather risks in both engineering and finance applications have, implicitly or explicitly, relied on the assumption that the climate (the probability distribution over weather) is not changing over time, and therefore the historic weather record is representative of current risks. Anthropogenic climate change upends this assumption by introducing the possibility, or even the likelihood that the climate distribution today is different from past experience. This effectively reduces the information available to actors and increases ambiguity in the estimated climate distribution, with associated costs for weather risk management and risk-averse decision-makers. These costs arise purely from the knowledge that the climate could be changing, may arise abruptly, are additional to any direct costs or benefits from actual climate change, and are, to date, entirely unquantified.
Using a case study of extreme rainfall-related flood damages in New York City, this paper illustrates how these ambiguity-related costs arise. Greater uncertainty over the current climate distribution interacts with a steeply non-linear damage function to greatly increase the mean and variance of the loss distribution. I show how this uncertainty can ripple through insurance markets in the form of higher and more volatile premiums and higher reinsurance costs, with limited potential for diversification within the insurance sector, impacts consistent with observed changes in the US property insurance market in recent years.