Words and language are vital when discussing big changes with lots of implications, because consensus emerges from a shared reality, and are built upon acknowledged platforms of values and facts. To refer to a well known board game, having that reality and platforms puts the sector on a ladder ; not having them puts it on a snake.
At the moment, this pooling / personalisation debate feels like it is in the middle of the board for that ‘snakes and ladders’ game. Sometimes it seems to move up a ladder, sometimes down a snake, but still nowhere near the sort of consensus (aka long ladder) that would enable the sector to forge ahead without a challenge being likely from a regulator or consumer group.
“Who cares”, you may ask; “let’s just steam ahead and bring people with us”. To be honest, that’s really just day dreaming. Insurance is too important to society for the sector to be left ‘to do its own thing’ in that way. And we can see policy makers increasingly taking that position. The trust just isn’t there.
Why Words and Language are Important
Before exploring particular problem areas, it’s worth looking at why words and language are important in such circumstances. Pooling and personalisation raise big questions about fairness, and the words and language we choose to use to discuss an issue like fairness, in this sort of context, are in fact value statements about how we see the world around us.
Words and language construct and shape the realities we are engaging from. If the realities coming into that engagement do not overlap in some way (or have the potential to overlap), even initially to just some extent, then that consensus will be impossible. Everything will be talking, but too few will be understanding.
It’s worth remembering as well that many of the things that we think of as facts are actually an interweaving of facts and values. Simple examples of this would be statements like ‘that is proven’, ‘that is correct’ and ‘that is clear’. We may think of these as facts, but they each have values interwoven into them. For example, clear on whose terms? Correct on what basis? Proven according to what criteria? Is this being a bit too pernickety? Not if the basis of proof is that of the sector and not of the court. Not if that proof is not disclosed, or with no right to appeal.
So, to help steer engagements in and around the implications of pooling and personalisation, I’m going to look at some of the ways in which words need to be used more carefully.
Subsidise
Those rated as high risk are said to ‘subsidise’ those at low risk. And risk pooling is referred to as a ‘subsidising mechanism’. Yet subsidising is a word that is hugely orientated around the concept and values of personalisation. So if I am paying more than the risk transfer price for my individual risk (and I mean just me and absolutely no one else), then I must also be ‘subsidising’ someone else’s risk transfer price.
Yet this ignores the fairness of time, whereby at one point of time, my risk may be high, while at another point in time, it may be low. With personalisation, my risk transfer price would jump up and down accordingly. With pooling, my risk transfer price would be less volatile.
I prefer to think of this situation as representing the ebb and flow of two types of premium: one could be called the risk premium and the other perhaps the reciprocity premium. And this period of ebbing and flowing would reflect the claims likelihood and severity for that particular type of risk.
Tax and Levy
In circumstances where the insurer carries some risks and fronts for other risks, and that fronted risk is pooled, the premium for the fronted risk can sometimes be referred to as a tax or levy across all policyholders. Again, as with subsidies, tax and levy are words permeated with the concept and values of personalisation. So to pay something other than the risk transfer price for my own individual risk, I must be being taxed in order for others to pay less.
Clearly, those interested in personalisation see such non-individualised premiums as a tax or levy, and that’s to be expected. But equally clearly, those interested in pooling do not see them as a tax or levy, but as a form of reciprocity premium, a price for exchanging risk with others for mutual benefit over time.
Are there values inscribed into terms like reciprocity premium? Of course. Am I perhaps doing, with my words, what I’m critiquing others doing with their words? Of course. My point is that both words have values and perceptions interwoven into them, and for the pooling / personalisation debate to make progress, the words used need to be thought through.
Why am I not then proposing words that would be suitable for both sides of that debate? Well, that’s not the point of this article. Perhaps a later one!
Success
Another word associated with many a debate going on in insurance is ‘success’. Clearly, when talking about the private market for insurance, success matters. The market wouldn’t be sustainable if it couldn’t be financially successful. Yet the market also wouldn’t be sustainable if it was not trusted by consumers to fulfill the promises inherent in a policy of insurance.
So, for example, anyone who has watched the pensions market over the past 40 or so years or so will have seen its long term financial success overlain with periods of reputational disaster. The latter has shaped a lot of subsequent legislation and regulation, the weight of which all present day firms have to carry. Indeed, one could say that the long term financial cost of those reputational disasters has reduced the financial success that that line of business has experienced over the time period.
What this points to then is financial success and reputational success not being two unrelated things. The latter will shape the former ; the unsuccessful will influence the successful. Or to put it another way, you can’t really understand financial success without understanding reputational success.
Control and Accountability
Last year, I described control as…
“…like the crankshaft of the personalisation engine that many in the sector want insurance to develop into. It represents the connecting logic between the data being observed and the analytical decisions being taken. How you calibrate control will influence the outcomes being generated.”
Yet control is very much a social construct, permeated with values. Take health insurance for example. Some see a lot of health outcomes determined by the decisions we take. Change the decision and the outcome changes with it. Why therefore should the healthy pay the price for the moral hazard presented by others’ decisions, goes the logic.
Yet look at the wording of the European Health Data Space (EHDS), especially around what is meant by secondary health data. That definition takes a very different view on what level of control we have over our health outcomes. And it is also very much in line with medical research into what makes us healthy or not.
My point with control is that there are two contrasting takes on control and outcomes in relation to lines of insurance business like health. If they stay so far apart, then the pooling / personalisation debate will become very stuck, resulting in all sorts of ‘work arounds’ being attempted in relation to legislation like the EHDS.
Perceptions and Change
How we use words and language influences perceptions. What often happens in relation to the pooling / personalisation debate is that there are impressions that the actuarial fairness that underpins personalisation is something that is objective, certain, reliable, calculated and individual, while the solidarity fairness that underpins pooling is something pliable, soft, subjective and vague. Why would a business rely on the latter, when it can use the former, goes the argument.
Yet is actuarial fairness as objective as people think? Consider these words from the renowned British actuary Charles Ansell to a Parliamentary committee in 1844:
“My own opinion, for a long series of years, has been that female life, as it exists on the books of insurance offices, was very much worse indeed than male lives, amounting almost to a motive for refusing the insurance of them at all; and I believe the experience of others has been the same.”
Ansell thought he was being objective. We would disagree. What you choose to calculate actuarially, how you do that calculation, and how you judge the outcome: all of these involve social values and political choices.
A more recent example? Well, in the southern US states in the 1950s, how you calculated how full a bus was, was very dependent on how you saw its passengers.
Looking forward, I think there is a vital role to play in having a group of academic experts from philosophy, sociology and business work on a language for this pooling / personalisation debate that will on the one hand, move it forward, and on the other hand, avoid it becoming bogged down.
