For an organisation tasked with ensuring a ‘well functioning’ insurance market that maintains public confidence, the UK regulator is being remarkably short sighted. ‘Not our problem’ is the conclusion they’ve reached about the most fundamental of changes affecting the insurance sector. So why does this regulatory mindset seem so myopic? And how should insurers respond?
This month, the UK regulator of insurance markets, the Financial Conduct Authority (FCA), issued its 2020 ‘Sector Views’ report. This presents a helicopter view of the issues that the FCA is interested in, in each of the sectors it regulates. The section on ‘general insurance and protection’ is particularly revealing.
Some background first. The most fundamental change happening in insurance is personalisation. Many of the ethical questions associated with ‘data, ethics and insurance’ are symptoms of personalisation. So most of the questions raised in an insurance context about issues like privacy, fairness and autonomy have their origins in the market’s move towards ever greater levels of personalisation.
This makes the regulator’s approach to personalisation pretty important then. Yet in their Sector Views report, they admit to not having an approach. Despite acknowledging that personalisation could have implications for consumer harm, they say that it is not within their remit.
A decision not explained
And the reason for their conclusion? That they are not sharing, for according to the FCA, personalisation isn’t happening yet. And that is why I referred to them earlier as being myopic. Perhaps after two weeks of widespread flooding across the UK, the FCA might like to think back over the past seven years or so of flood underwriting and reconsider the mindset they have so clearly adopted.
Flood underwriting evidences just how much impact personalisation can have on the functioning of the insurance market and the public’s confidence in it. Data and analytics allowed insurers to underwrite on an individual property by property basis. The consequences were huge. Long established communities found their continued existence under threat because the move from pooled risk assessment to individualised risk assessment meant that for some, premiums increased exponentially.
And the irony is that out of this situation rose Flood Re, a pooled risk reinsurance vehicle. In essence, flood underwriting had been heading towards a political cliff and was pushed away from it just in time. This sounds all rather ‘big game’ish’, the grand sweep of markets and politics. Yet the reality for consumers was quite different. If you sat down with people who had been with the same insurer for 40 years, who had never claimed in those 40 years, and whose household renewal had just gone from £300 to £5,000, you will see the financial and emotional crisis that many everyday consumers have faced.
A problem being repeated
So personalisation is happening, now, and regulators need to take off their macro outcome spectacles and see the accumulating micro outcome evidence that is out there. This has been a problem for the regulator before. The pricing super-complaint provides ample evidence of that. In 2016, they saw price optimisation as not really a problem. Now it is the biggest problem on their insurance agenda.
And so the regulator needs to ‘walk the talk’ on transparency and explain why personalisation is not their problem. It’s an explanation that needs to be opened to scrutiny, for the assumptions upon which it has been based may be less than objective. How have they been interpreting ‘a well functioning’ market? How are they gauging ‘public confidence’?
The danger is that the regulatory mindset is viewing personalisation through the lens of economic rationality. I’ve seen this in play, influencing perspectives and lowering regulatory interest. Surely higher risk policyholders should pay higher premiums, goes the argument. It’s only fair that lower risk policyholders should pay less, the regulatory mindset reminds them. And while there’s certainly mileage in that, it depends hugely on how long a perspective is adopted.
It’s a logic that works well for some people in the short term but badly for everyone in the long term. In that long term, loss events will happen, to most people, and many of them will be outside of their control. My house being struck by lightning four years ago produces evidence of a higher risk, which under personalisation would cause my premium to shoot up. After all, why should my neighbours pay for my loss?!?
The end of insurance as we know it?
And so, over time, the continued growth of data and analytics means the natural outcome of personalisation will be individualised underwriting. The pool of one – you, me, your partner, child, parent. And as explained in earlier blog posts like this and this, the outcome of that is pretty much the end of insurance as we know it. The prospect of that, and consequences that go with it, will hardly fill the public with confidence.
The FCA mindset around personalisation is also out of kilter with that of another big regulator, the Competition and Markets Authority (CMA). In May 2019, the chair of the CMA, Andrew Tyrie, had this to say…
“The rise of the digital economy has brought huge benefits to millions of people. But it has also rendered previously confident and capable consumers vulnerable to getting bad deals and poor service. This is not just people who are vulnerable on well-understood indicators: those who might be old, or on low incomes. It includes millions – perhaps even the majority – of the population, many of them ‘time poor’. They – us – are the “new vulnerable”. We are all vulnerable now.”
Personalisation and vulnerability
That ‘we are all vulnerable now’ narrative is based upon what might be called the dark side of personalisation. A logic permeating the personalisation mindset is that if someone is prepared to pay more, then what’s wrong with charging them more. So if someone in a moment of vulnerability might pay more for an insurance product, then what’s wrong with marketing it to them at that particular point of vulnerability? Examples of such moments might be as people approach marriage, house buying or childbirth. Or when they are changing jobs or breaking up with a partner.
So if data and analytics can identify such moments on a person by person basis, what’s wrong in promoting the product there and then? (more here) After all, it’s a private market. Yet just because you can, doesn’t mean that you should. The choice a firm makes here is down to its ethical culture, and the leadership its executives are giving on ethics. Exploiting vulnerabilities, macro or micro, will always raise ethical questions. Insurers have to decide what attention to give such questions, and how this might look when opened to public scrutiny.
So the problem for the regulator is that they see personalisation through too simple a ‘high risk, low risk’ lens. Why then have they decided not to see the influence and impact of personalisation on micro vulnerabilities as within their remit? And when the CMA so obviously, and strategically, have? Why do they fail to understand the implications of personalisation on the assessment and settlement of individual claims?
It is time for the FCA to turn their lens the other way round and bring in micro as well as macro outcomes.
The implications for insurers
What might be the implications of this FCA stance for insurers? Might some be tempted to label personalisation as ‘not our problem’ too, and push forward with digital strategies premised upon it? I would caution against this, for these reasons.
Firstly, the FCA have got it wrong around data and analytics before. The aforementioned lack of issue with price optimisation in 2016 and their necessary volte-face in 2018 is a lesson that should not be forgotten. Why? Because those micro financial and micro emotional crises don’t just go away because the regulator doesn’t recognise them.
Secondly, as explained in this earlier post, “the regulation of insurance is political” and Citizens Advice’s super-complaint evidences that. If civil society groups find that the regulator isn’t paying attention to their concerns, then they will simply use other, more political means by which to have those concerns heard. So chief risk officers need to think beyond compliance and see the personalisation clouds building on the horizon. Those clouds will continue to build (regulatory attention or not) as evidence of harm builds. Flood underwriting was a harbinger, not an exception.
Thirdly, an insurer’s digital strategy should not be premised on its risk side by what the regulator happens to be interested it. That strategy needs to focus on the audience that it is there to serve: the customer, both present and future. And it should reflect the trust you want those customers to have your business. Transform your business – absolutely. Do it with one eye closed – not wise.
To sum up
The FCA’s view on personalisation is short sighted and lacks transparency. It looks very much out of kilter with its obligations in relation to well functioning markets and consumer confidence. Some insurers might see this as an opportunity, but those whose digital strategies are orientated around long term consumer trust will think twice.