Insurers must prepare for big changes around genetic testing

We will soon see some significant changes to the way in which the use of genetic information by UK insurers is regulated. And these changes will result in insurers having to tighten their monitoring and control some emerging practices. It will be a different world, and insurers need to start preparing for it now.

The ‘Code on Genetic Testing and Insurance’ is an agreement between the Association of British Insurers (a trade body) and the UK Government. It has its origins in a moratorium agreement signed in 1998 and superseded in 2018 by this code. The main difference between moratorium and code was that while the former had an in-built lifespan of three years, the latter is more open-ended.

So Why the Change?

On the face of it, it seems odd that a code agreed only two years ago is about to be reviewed. Yet dig around and five reasons for this change become clear.

Firstly, the nature of that preceding moratorium. It was created as a temporary measure, a stop gap until something more detailed and embedded was in put in place. That was the reason for the moratorium’s three year lifespan.

What actually happened was the very opposite. Rather than spur the insurance sector and the Government to work on that more detailed and embedded solution, it actually resulted in them, to all intents and purposes, kicking the issue into touch. It was easier to just renew the soft regulation (six times), rather than work out something more lasting.

The second reason lies in the imminent confluence of two trends: in relation to genetics, the reduced cost and speed of genetic testing, and in relation to insurance, the increased appetite of the sector for data, and the increased capacity of its analytics to make use of it. Add in the market logic behind personalisation and the pressure on insurers to use genetic data was significant.

Insurer defined Goodwill

And the third reason was that both code and moratorium had significant weaknesses. Both are voluntary, applying only to ABI members. This certainly represents a large part of the market, but with still a significant minority of capacity outside of it.

Furthermore, the ABI has not always had an easy relationship with the life and protection market. Big players like Aegon, the old Skandia and Legal & General have all left the ABI, although those three have agreed to abide by the code as if still continuing members. Overall, it added up to a lot but not all of the market. And it was still voluntary, relying on goodwill more than anything else. Yes, it was a requirement of ABI membership, but bear in mind that that was a condition determined by the insurers themselves.

The fourth reason were signs that some insurers were looking for loopholes in the code. And, unbelievably, they were finding them. The most glaring one was the code’s almost singular focus on underwriting. This allowed law firms working for insurers on large injury claims to weigh up opportunities to bring genetic test data into settlement agreements.

Now you might think this is just one of those things. Actually, it is a bellwether for something altogether more serious. Voluntary agreements work when the collective works together to keep wayward thinking individuals in line. No one wants to be seen to be the one to break the agreement. And that works, until that is, most of the collective finds itself thinking the same way. Something along the lines of… ‘perhaps if we all jump at the same time, for the same reason, then we’ll change the market’s thinking in one big move.’ A consensus is nice while it works, but dangerous when it doesn’t.

Catastrophic Failure Ahead

If you bring together reasons two, three and four above, the picture that emerges is of a code progressively diminishing in value, and in danger of crumbling altogether. It’s a design susceptible to what engineers call ‘catastrophic failure’. And this brings us to the fifth and final reason for this change.

Over the 22 years of the moratorium and code, a great many people have taken genetic tests. It’s popular with family historians for example. Such tests were obtained under the understanding that insurers wouldn’t be able to use the results. If the code was to collapse (by for example a collective rejection or undermining of it by the market), the situation for those tested individuals would turn pear shaped overnight. Without the code, the default is existing law, which allows insurers to expect to see whatever data they think reasonable.

So what, some insurance people might ask. Most of those will already have policies signed and sealed. Yet some of those policies now have built-in mechanisms for adjusting cover and premiums. Think of those Vitality type policies for example.

Why the Market Thinks Change is Important

It’s worth spending a few minutes looking at a significant and persistent view within the market as to why access for insurers to genetic test results really does matter. It is likely to be this that is pushing insurers to collective reconsider their voluntary adherence to the code.

It’s centred around the notion of ‘self selection’. This says that people who take a genetic test will find out if they are susceptible to certain medical conditions. They will then use that information to actively seek out life insurance. Meanwhile, without access to genetic test results, insurers will be taking on policies at rates that do not reflect the underlying risk. It’s a form of information asymmetry which insurers believe, left unchecked, would undermine the whole life sector.

There’s some mileage in this argument, but it also has its weaknesses. Firstly, to what extent does self selection actually happen? I’ve watched a reinsurer present research that showed self selection to not be happening, and then dismiss the research with a ‘but we know it does’ comment. So I think the whole question of the extent to which self selection happens is still an open one. More and better research please.

Clear Test, Unclear Outcomes

Secondly, the ability to delivery cheaper and quicker genetic testing has not been matched by the ability for those test results to be interpreted in terms of a person’s mortality. And furthermore, the more scientists seek ways to better understand this, the more complex they find it to be (take epigenetics for example).

What this adds up to is a far from clear picture of just how much an impact genetic testing is actually having on the performance of life portfolios. What I suspect is really driving insurers’ interest in genetic test results is culture and data. In other words, ‘if we don’t know everything we’d like to know, because it might in some way influence risk, then we have a problem.’ Agree with this or not, the culture driving this thinking is powerful, and will influence big strategic decisions in relation to things like the code.

We have then a code in a precarious position. It’s more exposed now than at any point in its 22 year history. On the surface, the code’s open ended nature seemed to give it a more reliable future. It looks however to be more like a sticking plaster for the serious cracks that were emerging.

Preparing for this Change

The insurance sector has a lot on its table at the moment, with the pandemic and its associated consequences. So how should insurers prepare for what some might see as a bit of structured conjecture? The answer lies in two key documents released in recent months by the Financial Conduct Authority.

Their ‘Sector Views’ report issued in February 2020 said that “developments in bio-metrics and genetic testing… could result in the reduction of risk-pooling across customer groups and make insurance unaffordable or inaccessible for many”. And then a month later, in their 2020/21 Business Plan: “using bio-metric or genetic data for risk modelling could make some consumers uninsurable, removing their access to the pooling of risk”.

Note two things about these references. In both documents, they’re not positioned in relation to the code. And there’s no mention of genetics in the two preceding business plans.

What this signals is the movement of genetic testing from the perimeter of the regulator’s radar of insurance priorities, towards its centre. So the obvious question is why? What has triggered this change at the regulator?

I believe it down to three reasons, each based upon the consumer perspective. Firstly, the regulator’s interest in vulnerable consumers and in access to insurance. Secondly, those consumers who over the preceding 20 years have had a genetic test under what they thought of as ‘moratorium terms’ and are now seeking cover for both new and increasingly large amounts of cover. And thirdly, the weak provision in the code for consumer redress mechanisms.

Over the last 20 years, the insurance sector and the UK Government may have been able to ‘muddle through’ with auto-renewal on the moratorium, followed by a tidied up code. I believe the FCA see the next 20 years as more problematic.

What to Expect

Here are some thoughts on what I expect to see over the next few years:

the code will in some way be absorbed within the regulatory framework;
this will make it compulsory for all regulated firms, including intermediaries;
this will also make it applicable to all regulated activities, so more than simply underwriting;
consumer redress mechanisms will be enhanced;
it will then feature in regulatory reviews, data ethics being the most obvious one.

And it’s likely to be implemented as a two stage process: an initial announcement that the code is deemed to now form part of the regulatory framework, and then a more structured incorporation of it into that framework, with enhancements added as well.

How can Insurers Prepare for this.

Here are some thoughts on how a typical insurer can start to prepare for this change:

make sure your perimeter thinking is correct. Genetic testing is no longer confined to life and protection insurers / intermediaries. It’s an ethical issue for all regulated firms.
take your purpose and values, and make sure that your thinking on how your firm handles ‘consumers and genetic testing’ fits comfortably within them.
review your digital strategy so that possible friction points in relation to ‘consumers and genetic testing’ are identified. In other words, is your firm on a digital journey that could collide with regulatory and consumer perspectives around the use of genetic testing?
review your existing and projected use of genetic test data across all functions. Claims is the obvious place to start asking questions.
replicate 4 above for proxies of genetic testing. They certainly exist in overall terms, so do they exist within any of the types of data your firm holds?
replicate 4 and 5 above for suppliers of data, and with partners holding / using data on your behalf.
start preparing an evidence trail for your firm’s use of genetic test data and how that use is overseen and controlled. Expect the regulator to start asking to see such evidence sometime over the next three years.
Educate senior management functions and board members on the reputational implications for how your firm is using genetic test data. This is necessary because when/if a problem emerges, it is likely to immediately make media headlines. Senior people need to understand the issues and be able to respond to questions.

It would be very easy to run the above points through as a quick internal tick box exercise. Behavioural ethics would signal this to be a dangerous approach. In regulatory speak, it sends out a big ‘culture problem’ signal. Instead, bring in some form of independent voice into your review. It will help your people to make judgements that are more balanced across the interests of firm and consumer.

Summing up

Genetic testing is important. How the insurance sector handles genetic testing and the data that emerges from it is also important. The two have to be in sync for confidence in the sector to continue.

The moratorium was well past its ‘use by’ date. The code is open ended, but didn’t radically change the situation. Unless the sector prepares for change, it will find itself in a reputationally exposed position.

A move by the regulator to take over control of the code is therefore hardly a surprise. Insurers might not like some of the constraints they will enforce, but that comes from the issue having been left to drift into dangerous waters. Steering a clear course towards safer waters is good for everyone.