Oxford water lecture series 23rd Oct 2013
Matt Cullen (ABI) explained flood re, Prof Edmund Penning Rowsell delved into the economics of risk rating and subsidy.
The annual property premium in the UK is circa £9bn. Most perils are predictable, but flood losses are very volatile e.g £3bn in 2007 on commercial and domestic flooding combined. Reinsurance would seem to be in the region of £13bn.
Around 500,000 homes in the UK are at very high risk of flooding, a high proportion of these are council tax band A and B and have a high probability of not insuring either structure or contents. This causes political embarrassment when large numbers of people find out what it is not to be resilient.
Flood Re would offer a subsidised insurance scheme for such domestic risks. It is estimated that home-owners would contribute £140m annually, the property insurance market would top this up with an additional £180m by annual levy. Flood Re would be able to buy reinsurance.
The 1:200 year loss for such properties would be £2.5 bn.
Some band G houses would be subsidised by the scheme, but band H are excluded.
Domestic property insurance is not flood risk based, in effect every policy holder is already subsidising those who live in high risk areas. This is not transparent to the policy-holder. The estimate is £30 per policy, even for those who live at the top of a hill.
IF risk rated, the high risk property (around 2 million of them) premium would be circa £700. But band A and band B people wouldn’t pay that, many don’t buy insurance at all, many are tenants.
Flood resilience adaptations might cost in the region of £20k per home, but area wide defences are much more cost efficient. It is estimated that every pound spent on area wide defences saves eight pounds in insurance losses. In effect, since there is no risk pricing, every pound spent on defences puts eight pounds into the pockets of insurers. No wonder insurers insist on government spending on defences as part of the deal to set up Flood Re.
Given that there was no market failure (some small insurers were lost) after the 2007 floods why should insurers bother with Flood Re? Perhaps it is to avoid governments forcing them to insure the uninsurable, as happens in the USA, France and Switzerland among others. The volatility of these high risk homes would seem to be within the resilience band of the insurers, but any reduction in volatility would reduce reinsurance costs. It may even be cost neutral to provide £180m a year?
Insurers add their voices to the campaign in favour of man-made climate change. Perhaps this is to help support government policy, perhaps it encourages more to spent on defences. In reality, the rate of change of loss with global warming is steady and slow and easily accommodated by a well diversified insurer via experience rating.
It would be interesting to know what level of curvature in the loss curve would drive insurers out of the market. If climate modellers cannot show that there is a risk of reaching that second derivative, then it is not for direct business reasons that insurers believe in climate change.