Autonomy is both simple and complex, ephemeral yet ever present. It’s rarely invoked in an insurance context outside of self-driving vehicles. Yet it’s an ethical issue that is becoming central to the debate on how insurance is to be priced. Insurers need to understand it to be part of that debate.
Let’s add some other reasons why it is important for insurers to understand autonomy. So beyond that debate about how insurance is to be priced, autonomy is important because…
- it influences how people engage with the new generation of insurance products, and so needs to be accounted for in the design and pricing of those products;
- it is an ethical issue that pervades insurance, in a similar way to fairness, with which it has a lot in common;
- its influence on regulatory opinion accounts for the emergence in recent years of what could be called ‘problem/challenge/reaction’ scenarios, which are disruptive for both insurer and regulator.
You will be familiar with ‘treating customers fairly’. You need to become as familiar with the notion of ‘respecting customer autonomy’. While the former was introduced into the market in a somewhat top down way, the latter is more likely to emerge in a bottom up way. That’s because of the nature of regulatory culture, at least here in the UK.
What is Autonomy?
Autonomy has its roots in two Greek terms: autos meaning self and nomos meaning rule. Traditionally, it refers to the capacity “…to be one’s own person, to live one’s own life according to reasons and motives that are taken as one’s own and not the product of manipulative or distorting external forces...”
You can see from this description that it is centred around the individual – all those references to ‘one’s own’ and the like. The autonomous person is seen as self-sufficient and independent, able to make decisions that are in their own interest, and free from the intrusive decisions of others. This is often summed up as about self-determination.
The last few decades have seen this traditional view of autonomy being challenged. As a result, autonomy is now recognised as a more complex concept than originally thought. So what, many insurance people may say. I would urge you to bear with me on this, for that debate about the nature of autonomy has a bearing on the current dynamic around digital and the future of insurance. That I will look at in a minute.
How Insurers Experience Autonomy
Let’s look at two ways in which insurers experience autonomy. One is in relation to privacy. It’s often said that we value privacy for the sake of autonomy. If we lack privacy, then the decisions of others intrude into our right to a private live.
Think of it this way: how do you answer a question like ‘what is the point of privacy?’ The answer is invariably framed in ways like this: ‘…an individual’s ability to determine for themselves when, how, and for what purpose their personal information is handled by others.’
So if your work brings you to ethics and data ethics from a privacy perspective (which is common in insurance), then autonomy needs to be positioned pretty centrally to that work.
Another aspect of insurance in which autonomy can feature is consent. Here in the UK, consent is defined as needing to be ‘freely given, specific, informed and unambiguous’. So if an insurer doesn’t obtain the right consent from those with whom it engages, or misuses the consent it has previously obtained, then this will be impacting those people’s autonomy.
In Need of Attention
Now some of you may think that this is not that big a thing, especially in insurance. That’s not how the public see it though. Look at the independent research published in 2020 by the ABI, into “consumer attitudes to data and insurance”. This quote is from the report’s executive summary:
“Close to nine in ten (86%) consumers say that they are concerned about organisations selling or sharing information about them when those organisations don’t have permission to do so. More than half (53%) remain uncomfortable with this even when they have given permission for their data to be shared.”
Few if any of those consumers will have talked in terms of autonomy, but that is certainly what is at the heart of their concerns. If you’re engaged in any way with data and ethics in insurance, you should be seeing autonomy as woven thought narratives like this. The message the sector needs to listen to is that people don’t think their autonomy is being respected.
Insurers aren’t alone in facing such concerns. There’s many a technology firm facing similar disquiet. This has led to some insurance people talking about ‘this is just the way things are going now’ and ‘people will just have to get used to it’. That doesn’t seem to be happening though: people are not just getting used to it. And so some of those technology firms have started to think more carefully about how they engage with people, and for what purposes. Data handling practices are being revised as a result. Less intrusive data gathering practices are being considered. Insurers should prepare for doing likewise.
Steps to Start With
So what steps might an insurer take to rethink how they ‘respect customer autonomy’?
Clearly, as I have pointed out before, those specific consumer concerns about consent need to be addressed. This is primarily something for underwriting to lead on, but a big role for claims too. I’m not convinced the sector’s preference for very generic consent is going to stand up to regulatory challenge. Insurers need to be challenging themselves on this before someone else does so.
Then the use of secondary data needs to be looked at. Think of this from a consumer perspective – it involves the insurer not asking the consumer for their data but obtaining it from other sources. This is done without the consumer’s knowledge and to a much wider extent than many consumers envisage. Again, one for underwriting to lead on, in partnership with claims.
Look at the informational boundaries being used in counter fraud. To what extent is audio and facial recognition software being used to judge fraud risk? What social media data is being captured and how is it being analysed? How is surveillance work, either in-person or digital, being controlled? These and several others are all issues that do with autonomy.
Practices like these have emerged from the notion across the sector that began with ‘it can be done’, evolved into ‘we can do it’ and then crystallised into ‘we need to do it’. At some point early on, the question that needed to be asked was ‘should we do it?’. The problem for insurance is that while the sector has tended to avoid that ‘should’ question, others are now starting to run with it, giving it a momentum that is moving insurance onto the back foot.
Influencing Many Initiatives
Here are some other initiatives happening in the sector in which autonomy plays a significant but often unacknowledged role:
- the use of generic information at both the underwriting and claims stage;
- access to and use of digital medical records;
- the scope and depth to lifestyle underwriting;
- the ways in which mental health is being handled;
- the use of emotion related data and analytics.
Autonomy is Complicated
I said at the start that autonomy is both simple and complex. ‘To be one’s own person’ is something that resonates with all of us, but we often approach it in different ways. Some see it as the capacity to govern oneself, others as a sort of personal ideal. Perspectives like this have fuelled the recent debate about what exactly autonomy is.
A key development in that debate saw the traditional view of autonomy as orientated around the individual being challenged. Some scholars have critiqued the central role played by (what is invariably) the ‘autonomous man’, seeing its focus on substantive independence as undermined by more recent research into the family, communities, institutions and traditions within which we all live and act from an early age. Alongside that old focus on substantive independence has emerged a new focus based on relations of interdependence, such as trust, responsibility and solidarity. That new focus is called relational autonomy…
“This model (of relational autonomy) recognises that autonomy is not solely an individual enterprise and that respect for the autonomy of others requires collaboration, long term reciprocity, and equitable balancing of power relationships.”
So what has this to do with insurance, some of you will be asking. Well, an awful lot, it turns out. A debate is developing around the future of insurance in an increasingly digital era. And in particular, that debate is examining the roles that personalisation and pooling will play in that future. Will the current trend towards ever more granular levels of personalisation be continued, until real time individualised underwriting becomes the norm? Or will there be some backing away from that level of granularity (as per big tech firms recently) and the maintenance of some form of pooling?
How This Links to the Future of Insurance
Let’s draw some lines to connect these debates. Advocates of insurance personalisation emphasise the savings that policyholders can achieve by not having to pay for other people’s claims. Let your premium be based around you, rather than your accident prone neighbours, goes the narrative. What they’re doing is emphasising individual autonomy.
Advocates of the need to retain some form of pooling emphasise the longer term benefits of sharing claims with others, on the basis that you never know when or how big your own claims might be. Not all claims will be within your individual control, goes that narrative. What they’re doing is emphasising relational autonomy.
Both the debate about the nature of autonomy, and the debate about the future of insurance, are evolving debates. They both involve traditional views being revisited, powerful influences being recognised and new thinking being introduced. And at the heart of each debate is the question of primacy: should it be of the individual; should it be of the community, or should it be some mix of the two? And who gets to decide, and how do they go about doing so?
I believe these two debates can each be enriched by engaging with the other. There are clear commonalities, similar interests and useful synergies. At the moment, I can find no references to insurance in academic work on autonomy. That is, I believe, bound to change, just as it has changed in relation to fairness.
Footnote: both quotes are taken from the introductory chapter 'Autonomy Refigured' to "Relational Autonomy : Feminist Perspectives on Autonomy, Agency and the Social Self" (OUP, 2000) edited by Catriona Mackenzie and Natalie Stoljar.