What is insurance? And why data analytics makes this matter

  • 29 January 2020

When does a firm become an insurer? When does it stop being an insurer? And what is insurance? These are not nebulous questions. The influence that data and analytics are having on the insurance world makes it hugely important that there be clarity around them. And this is more than just a drafting issue. From an ethical perspective, we’re talking about transparency, honesty, fairness and accountability, and that’s just to start with.

The insurance sector has always had insurers and intermediaries of various sorts. Mutuals, friendly societies, brokers, advisers, composites and the like. Nothing wrong with that. Yet it has had some serious ethical consequences. The nature of the intermediary is hugely important for whose interests are put first. Problems with that have lead to some huge mis-selling scandals.

Yet many of you will be thinking that those problems have now been laid to rest with the comprehensive regulations that firms operating in the UK insurance market must adhere to. Well, yes and no, for several reasons. The problem I’m going to focus on here is coming to the fore because of the way in which data and analytics are reconstructing insurance. And I’m going to illustrate it with reference to two significant trends well underway.

Trend 1 - Organisations that Facilitate Insurance Decisions

The first trend is the growth and influence of organisations whose sole raison d’être is to facilitate underwriting and claims decision making. Examples of such organisations would be the Motor Insurers’ Bureau (MIB), the Insurance Fraud Bureau (IFR) and the various platforms that they operate, such as CUE, MIAFTR and the IFR. Other platforms exists amongst data brokers and software houses as well.

Such organisations collected underwriting and claims data from insurers. They add in data of their own, analyse it all together and provide outputs that influence the many types of decisions that insurers make. So this will include marketing decisions, access decisions, pricing decisions, cover decisions, service decisions and settlement decisions. In other words, just about everything.

Meet the Perimeter

Yet such organisations are not considered to be insurers. They fall outside of what’s called the ‘FCA Perimeter’ and so are not regulated. Yet they are hugely influential when it comes to the outcomes that consumers experience when engaging with the insurance market.

What does constitute that regulatory perimeter then? That is what is called the Regulated Activities Order (RAO). And the RAO talks about the effecting of contracts of insurance, which is clearly something that organisations like the MIB and IFB do not do. What they most certainly do do though is provide services that affect contracts of insurance.

Does that one letter make that much difference? The FCA is increasingly thinking that it does. In their 2019 Perimeter Report, they voiced this concern: “The boundary between providing mostly unregulated technical infrastructure and providing regulated activities is increasingly narrowing.”

Organisations like the MIB and IFB, in my opinion, undertake activities that, to all intents and purposes, replicate what an insurer does, but just without the ‘effecting contracts’ bit. How significant is that little end bit then? Not a lot, in my opinion, given their position at the centre of the market, their integration into much of what the market does and the influence they exert.

Trend 2 – Selling, but not Insurance

The second trend that is raising the question of when a firm is an insurer or not, relates to the extent to which insurance features in what is being sold. So we’re talking here not of firms positioned around the FCA’s perimeter, trying to keep their activities unregulated, but of firms that are well within that perimeter, doing things that could result in them being outside of the perimeter. In other words, a regulated firm doing things that look like a regulated activity, but that aren’t.

This is very topical, for a regulated investment firm recently went into administration after selling an unregulated product while proudly proclaiming that the firm itself was regulated. So how might an insurer find themselves in such a situation? After all, isn’t insurance, well, just insurance! That’s where data and analytics comes in, in two ways.

Personalised Insurance - a Misnomer?

Firstly, data and analytics are increasingly allowing insurers to personalise policies, so that each policyholder is rated on their own individual risk. Dress that up how the market may wish, but the extent to which risk transfer is taking place in such circumstances is hugely different to when the policyholder is rated as part of a pool. If the ambitions of some insurance technologists are fulfilled, and personalisation (even hyper personalisation) becomes the norm, is it actually still insurance being sold, and not some form of short term savings plan?

The Seeds of a Scandal

And secondly, as firms currently known as insurers increasingly position themselves as personalised loss prevention service providers, will the insurance content of their offering diminish to the point that it is no longer an insurance product? (And hence the question in this post’s title: what is insurance). This ‘retreat from insurance’ is happening now, and this post from 2016 explains it in more detail.

The problem is that the public may think that firms they identify as insurers will sell them insurance, when it might increasingly be the case that insurance is minimal at best.  And so when a loss happens, the customer only then realises that all that paperwork was about a service and not risk transfer.

What we have here are the seeds of a scandal. Firms will use the insurance history associated with their brand to engage with and convert consumers to a service based offer. And again, we see here the vital influence of data and analytics, allowing such firms to pick out the customers that they don’t want to do risk transfer business with, and instead sell them loss prevention services.

And this brings us back to our original question. When does an insurance product become so hollowed out that it is no longer insurance? When is that approached? Where is the tipping point?

Clarity and Certainty

What data and analytics demands of firms in the insurance market, and the regulator of that market, is clarify and certainty, as to what is an insurer, as to what is insurance. Is a twenty year old Regulated Activities Order still fit for purpose in a digitising market, in a potentially deconstructing market? I do not think so. It’s time for more clarity, for more certainty. If it’s not provided, then the market will sail into another big scandal. More subtle than previous ones perhaps, but still one that will undermine trust.

If you have any questions about this post, please get in touch

Related Posts 

These two posts will also be of interest...

  1. Are we seeing the start of a ‘retreat from insurance’?
  2. Personalisation – could it take insurance into a digital winter?