How your insurer comes up with the price it quotes you has always been a bit of a mystery. Comparison sites ask you all sorts of questions when you’re applying for insurance – but how these are used to deliver your price is hidden to the customer.
And it’s not just the information you give them which goes into the black box. Some insurers also use information about you that you didn’t give them – for example, data from your credit file or your supermarket loyalty card.
There’s no transparency in any of this – and most in the insurance industry argue that there shouldn’t be. What goes into each insurer’s pricing algorithm is their intellectual property – their secret sauce. And as long as the market is competitive, why should it matter? If one insurer uses your credit file, occupation and nectar card data to price your insurance – you’re still only likely to buy from them if their quote is competitive.
But with no transparency around what goes into the pricing algorithms – how can we know that insurers are treating customers fairly – and are not unfairly penalising certain groups of customers? Insurers are no longer allowed to price based on your gender.
And the Equality Act should also prevent them from looking at other protected characteristics such as your sexuality or ethnicity. But with no way to scrutinise their pricing formulas, it’s hard to have confidence that these rules are being abided by.
Last year, Citizens Advice published a controversial report suggesting that there is an ethnicity premium in insurance – with Black and ethnic minority groups paying significantly more for their insurance than white people.
The strong insinuation was that – deliberately or otherwise – insurers are discriminating against minorities.
In reality, I don’t think their research does prove the allegation they are making – but the lack of transparency in the market has left insurers struggling to put up a credible defence.
The fundamentals of insurance pricing are about risk. Insurers want to use data to work out how likely it is that you will have a car accident, or get burgled. In pursuit of this impossible goal – of providing a precise price to fit each individual’s risk level – insurers use ever greater sources of data. For example, if their claims history shows that people who work in leisure centres crash their cars more often than childcare workers, they will price the former higher than the latter.
But there can’t be any credible causal link between either of those jobs and the propensity to crash your car – so in using these broad brush statistical correlations, they penalise people for whom the association is completely untrue.
The best example of this is how insurers treat customers who have been involved in an accident that wasn’t their fault. Many will charge those customers more next year – because statistically you are more likely to be involved in another accident. It’s not that hard to work out why this is the case. People with better reflexes, for example, might have a better chance of avoiding being in accidents – and so those who get hit by another driver are on average those with slower reflexes.
But what about the accidents where fast reflexes would have made no difference – the excellent drivers who get ploughed into while they are sitting stationary, for example. They may also be charged more next year when the statistics are drawing insurers towards the wrong conclusion.
It’s time to draw some boundaries around what insurers can use to create a price – and what they can’t. As well as banning use of non-causal data – such as your occupation – we also need to look at how we price lower income families. In many cases, families in poorer neighbourhoods have to pay more for their insurance because the risk of theft or vandalism may legitimately be higher.
Similarly, young drivers pay much more for their insurance because they have more accidents. But is it fair to charge most to those with the least? The origins of insurance were all about pooling of risk – and perhaps the rest of the pool needs to pick up some of the cost of some of the higher risk customers to ensure they can continue to access insurance.
When it comes to ethnicity, Citizens Advice are undeniably right that ethnic minorities pay more for their insurance – but they are wrong to insinuate this is down to discrimination. In reality, the higher prices that non-white customers pay are more to do with the correlation between ethnicity and income. A recent study by the Institute of Fiscal Studies found that more than half of Black African households in the UK were in the bottom 20 per cent in terms of household income, living in poorer areas where claims rates are higher. The same is true of many other ethnicities.
Rather than making their argument about race, Citizens Advice might be better placed to line up alongside the Barrow Cadbury Trust’s Fair by Design campaign – which works to eliminate the poverty premium.
The debate around how to fix these problems needs to start with our politicians. My hope is that the recent rise in car insurance prices might be the catalyst for a wider conversation about how we can keep insurance priced fairly and accessible for all.
James Daley is the managing director of the consumer group Fairer Finance