Reinsurance broker Willis Re has published an interesting report looking at the potential impact of the new RMS V11 risk model when used to calculate catastrophe exposures for European windstorm risks including storm surge. It’s interesting to read as most of the focus to date has been on the impact of the new risk model to U.S. hurricane risk zones and re/insurers and catastrophe bonds with exposure in those areas. Willis Re looked at the changes to the RMS modelled windstorm and storm surge losses and reported back on expected impacts to UK insurers.
The report found that the changes to the way RMS’ model assesses these risks could have significant ramifications for the UK re/insurance market. Increases in storm frequency, storm clustering and vulnerability together look like they will contribute to a large overall change in RMS’ view of risk for these perils in this region. The report goes on to explain that, while monoline property insurers could see a 97% increase in their net capital requirement, Willis Re says that diversification benefits can reduce this figure significantly.
Willis Re looked at the increases in risk and how they will relate to Solvency II capital requirements, a hot topic in the European re/insurance markets. They found that UK insurers could see increases of up to 97% in their capital requirement for catastrophe exposures under Solvency II rules when calculations are made under the new RMS model V11, in comparison to under V10. Similar increases are estimated for French and German insurers, although Willis Re do not expect them to be quite as steep.
Tim Edwards, the author Willis Re’s report, explained that a team of meteorologists, engineers and analysts have been analysing the new RMS model and considering the merits of the changes.
“From validating against 3rd party hazard datasets and testing against our clients’ loss experience, we see the RMS v11 modelling as being an improvement in many respects but with certain major calibration issues,” he said. “RMS is one of the key providers of vendor models, which insurers can license to help assess the exposures within their portfolio to extremes of risk. When one of these providers changes the parameters of their models, it can have dramatic implications for the insurers who rely on their outputs. This is what is happening as a result of RMS issuing the latest version of its own model.”
“We believe that RMS v11 for the UK is unlikely to adequately reflect the nuances of all clients, and we therefore see it as our role to customize the view of risk to satisfy our clients’ internal governance requirements and the increasing demands of external stakeholders,” he continued.
Robert Rogers, Willis Re Regional Director UK and Ireland, said; “Many insurers are coming to us because they are alarmed by the increases in modelled loss produced by the new model. They want our view on whether the output is reasonable, and to what degree it should drive their reinsurance purchasing decisions. Obviously, if the outputs of the models change, so does the expectation of what is a sensible amount of reinsurance to buy.”
He added; “Interestingly, reinsurers are also asking for our input. We are having conversations with them looking into the ways in which RMS v11 may be over-stating exposures.”
Willis Re say that RMS is expected to release some additional features to their current model in early July. These new features will include the ability to quantify the vulnerability uncertainty around their central view of risk released in August last year with an upper and lower view. Rogers added; ”Whilst we welcome this additional model feature, we believe that the Willis Re approach of calibrating the model to our clients’ claims history is likely to lead to the most robust view.”
Once Solvency II comes into play insurers will be required to demonstrate that their capital requirements are sufficient for a 1 in 200 year event. The impact a 97% increase in this capital requirement could have is plain to see. It could lead to insurers being required to take on extra reinsurance cover to ensure capital adequacy or could even lead some to look to the catastrophe bond market as a way to increase their ability to manage peak risks.
There are a number of outstanding catastrophe bonds with exposure to European windstorms which have used the RMS V10 model for risk analysis and calculation services. We wonder what, if any, impact this increased view of European windstorm risk has had on their risk profile or expected losses.