The events of 2011 had a significant impact on property renewals at January 1, 2012. The approach to the property market, particularly property catastrophe business, has evolved significantly since Hurricane Andrew with the gradual incorporation of technical and model-based underwriting. There is evidence at this renewal that another subtle shift has taken place.
The two triggers, loss activity and model change, are not new to the industry, but there were important differences in 2011. First, loss activity was substantial - the second worst year on record. The number of events, their location and size were all factors. A fairly major insured loss in Thailand, an area previously considered as “incidental exposure,” highlighted again the need to anticipate the unexpected.
However, losses were spread throughout the globe, and reinsurers are better capitalized than during previous heavy loss periods. The impact to capital was apparent at mid-year, but capital rebounded by year end. The year 2011 did not ultimately end up being a scenario where substantial losses drained supply and increased demand creating a market reaction based on simple economics.
In many cases, model version change follows on the heels of significant losses and therefore significant new data and corresponding market reaction. In 2010 and 2011 there were several new model version releases, as discussed later in the report. These did not necessarily go hand in hand with a triggering event. Also, many of the changes addressed in the new model versions had already been taken into account by insurers and reinsurers based on previous loss experience. Therefore, in an environment of sufficient capital, and absent physical evidence of an increase in magnitude of risk, companies were in a position to step back and assess the degree to which these model changes should or should not impact their business.
Under these circumstances, many reinsurers took on a re-evaluation of how models are being used within their underwriting process. While most have been making proprietary adjustments to the modeled output in the past to reflect their own research, generally results were still maintained within a fairly predictable band from reinsurer to reinsurer.
This gap in the perspective of risk is growing as reinsurers are becoming increasingly sophisticated in deconstructing models, making adjustments and even creating their own models so as not to be dependent on someone else’s changing view of risk. The ultimate result of this is a market that is becoming more fragmented and more customized in its view. A thorough understanding of a company’s risk profile from a variety of perspectives, analysis as to the most appropriate representation of a company’s risk and very candid dialogue with reinsurers on understanding the risk and its drivers are absolutely essential in order to navigate this shifting environment.