What Claims Acceptance Rates Tell Us

Imagine if one in three trains set off but then failed to get to their destination. Or if one in seven people complained about the groceries they bought? Those markets would be in chaos, let alone under pressure to reform. Yet in 2023, manufacturers of buildings insurance policies in the UK were running with these levels of performance. And the trend relative to 2022 is running very much in the wrong direction.

Here are some figures for you to think about…

 So what does this tell us?

We can immediately see that this is not just ‘a personal lines thing’. The claims acceptance rate for motor is very different to household. And it looks like the low claims acceptance rate in household could be a factor fuelling the higher complaints rate. So I’m going to focus simply on the claims acceptance rate. Why should it be so low?

A Host of Reasons

A whole host of reasons jump out, relating to marketing, underwriting, claims and counter fraud. Are consumers buying something that is quite different to what they thought they were buying? Have household policies been hollowed out, more than consumers realise? Are policies just too complicated to understand? Are household claims more complex? Do consumers have too high expectations?

That list does get pretty long, and it would be easy to get sucked down into all that detail and all those variables. Instead, I’m going to focus on the two reasons why claims are rejected.  These are: either the loss wasn’t covered in the first place, or the loss was covered but there was some form of misrepresentation present.

Here are four principal reasons why losses aren’t covered. The peril that was the proximate cause wasn’t covered, an exclusion or warranty applied, the item itself wasn’t insured, or the premium wasn’t paid. Misrepresentation is down to deliberately withholding information, providing misleading information or having not kept material information up-to-date.

Finding the Balance

So what is the balance between these two categories of reasons for claims being rejected? I don’t have (nor can’t find online) the answer to that question, but it will exist, certainly at insurer level and quite probably at the market level as well. That’s because when rejecting a claim, the insurer must give their reason for doing so. It’s a simple data field.

I suspect that misrepresentation has been behind the downward drift of the buildings claims acceptance rate. There will certainly be lots of that ‘losses aren’t covered’ element present (there always have been), but  I think misrepresentation has been growing to be something more significant than before.  

Let’s explore misrepresentation then, to try to understand some of the key drivers.

Most forms of misrepresentation fall into one of two categories: there’s either something wrong with the claim, or something wrong with the risk for which a loss is being claimed. There’s absolutely nothing wrong with checking for misrepresentation ; the ethical side relates to the basis upon which that checking is done.

A Flat Roof

So for example, if the policy was covering a house of standard construction, and it turns out to have more than a certain amount of flat roof, then the insurer will say that there’s been a misrepresentation of the risk. So how might that difference around construction emerge? Here are six ways:

1. through a loss adjuster seeing the house and noting the flat roof area in their report.
2. by the claimant self declaring it when giving details of their claim;
3. by the underwriter, at the time of claim, using drone imagery of the house and having some analytics work out the flat roof area;
4. by the underwriter, at the time of claim, referencing satellite imagery of the area in which the house is located, and having some analytics that estimates what houses of that shape typically have as flat roofs in that area;
5. by the underwriter, at the time of underwriting, using satellite imagery of the area in which the house is located, and inferring from that there is a low chance of your house having any more than the normal amount of flat roof.
6. by the underwriter, at the timing of quoting, building an assumption into their rating about houses being quoted having no more than ‘a defined amount of flat roof’.

All of these ways enable a judgement to be made around misrepresentation, but there are differences in the basis upon those judgements are being made.

Clearly, reasons 1 and 2 involve direct data about the roof. No much to argue about there then.

Reasons 3 and 4 involve statistical assessments at the claims stage, so these output indirect data. It’s done to keep claims expenses down. Remember though that the way in which the insurer sets significance levels within their analytics will influence the conclusion being output.

Reasons 5 and 6 involve more use of indirect data about the roof, but this time at the underwriting stage. It’s done to remove questions judged to be hurdles to policy sales. Clearly however, how those significance levels and assumptions are set will influence the conclusion being output.

Weighing All This Up

Misrepresentation is clearly then a judgement based on differing levels of certainty. Where direct data is involved, that certainty is very high, but with indirect data, it’s another matter. That’s because it is the insurer who is setting the dials for significance levels and assumptions to determine the conclusions being output. The key question then becomes… when is that conclusion certain enough to warrant applying misrepresentation?

Insurers can take a ‘bullish’ stance on that question by referencing the obligation on people seeking quotes to take reasonable care not to make a misrepresentation. This narrative talks about underwriting not being able to ask about everything, about reasonable quotation assumptions having been made, and about the prospect, if in any doubt, needing to have raised the flat roof with them.

Are all insurers doing this? I believe a fair number could be and point to that earlier data as evidence;

• the low (and dropping) claims acceptance rate for buildings;
• the high (and rising) complaints being raised by claimants on buildings;
• plus data from the Financial Ombudsman Bureau showing that 41% of all complaints they receive about all forms of buildings insurance are upheld.

Underwriting at Claims Stage

What we could be seeing then is data and analytics being used to increasingly underwrite policies at their claims stage. Yet underwriting at claims stage is a practice that does not bear up well to ethical scrutiny, as I outlined back in 2019.

There’s been a lot of talk this year about ethics and claims, with big insurers like Allianz, Covea and Ecclesiastical signing up to an AI code of conduct. What we have then is an opportunity for these three insurers to illustrate that commitment and show us how they are managing misrepresentation in buildings policy claims.

That may seem like a bit of effort, yet thinking long term, there’s a very good reason for them making it.

A Vicious Cycle

Consumers, when taking out any form of insurance policy, are invariably asked a specific question: have they had a claim rejected in the past? One third (and growing) of buildings claimants now have to say yes to that. Over say five years, that could add up to an awful lot of people finding insurance either much more expensive or not available at all.  

What will emerge then is a challenge to the sector, along the lines of having to explain and justify this claims performance and the impact it will be having on an ever increasing number of people.

To Sum Up

Claimants who misrepresent are being identified and isolated by insurers. So they should be, many in the sector will shout. And there’s a lot of sense in that, but also a lot of concern, that the basis upon which misrepresentations are being judged is fair. It is time for claims acceptance rates to be thoroughly analysed and the findings put into the public domain. It is only a matter of time before a good proportion of people with rejected claims start to challenge the sector on this. And clearly, it is better for the sector to challenge itself before others do so.