page-banner

What do planes
landing and reserving processes have in common?

Our viewpoint

Not a lot, at first glance! But both can be affected by “plan continuation bias” – the tendency for us to continue with a pre-existing plan, even when the situation changes and makes that plan unsuitable.

Back in February, a video of planes landing at Heathrow during storm Eunice became a viral YouTube sensation. Planes that arrived before the storm landed safely, and planes that arrived after it was in full force were safely diverted. But planes that arrived just as the storm was breaking landed in an increasingly precarious manner – ending up on YouTube in the process.

Each successive pilot faced stronger winds and heavier rain than the last one when coming to land, but had also witnessed the previous successful landing just minutes earlier. These mixed messages led to pilots continuing to land in increasingly adverse conditions. This illustrates that the risk of poor decision making is much higher when conditions are changing fast and we are getting mixed signals (like worsening weather but earlier pilots landing successfully).

The same logic applies to actuaries, and right now we run the risk of attempting to bring our reserving processes in to land precariously as the inflationary storm is breaking.

LCP inflation survey highlights this risk

We recently surveyed actuaries and other insurance professionals about their inflation expectations over the next 5 years. We asked for predictions of CPI and also for 4 major classes; motor damage, motor injury, property, and casualty.

We restated these results as the implied excess inflation (ie the difference in inflation predictions between each insurance class and CPI).

We expected to see a chunky allowance for excess inflation above the CPI index predictions. In particular, we expected this on Casualty – where social inflation has driven high claims inflation, even in the last decade of low price inflation.

Instead, respondents predicted negative excess inflation over the short term, and just 1.0 - 1.5% pa excess inflation in the longer term. For classes such as casualty and motor damage, this is lower than actual excess inflation over recent years, which is counterintuitive at best.

This survey, and our wider anecdotal experience, suggests that the insurance industry might currently have a blind spot to high inflation. Whilst our survey results showed that respondents could recognise that the CPI index was going to be higher in the near future, they struggled to translate this into commensurately higher claims inflation.

Cognitive and emotional bias vs inflation expectations

The hard truth is, even actuaries aren’t always good at forecasting or decision making. Decision making is complex. We like to think that we’re rational, but a wide range of emotional or behavioural spanners can get in the way as the cogs are spinning in our brains.

Inertia

The current bout of high inflation is a sudden problem that calls into question 10+ years of status quo. We can be biased in favour of the status quo – our existing behaviour has inertia. This means the evidential threshold for changing our approach is higher than it should be.

It takes effort to overcome our inertia. We can take a lot of persuading to switch bank accounts, shop in a different supermarket chain, or try a different route on our commute when there’s travel disruption.

Similarly, it can take more evidence than it should to make us really grasp the nettle when it comes to inflation. Our inertia leaves us at risk of underestimating the problem and of making a small process change, when the evidence really calls for much more.

The impact of emotion

Inflation affects us personally - if we predict higher inflation then we’re predicting a drop in our own living standards. Further, if we’re responsible for reserving processes, then we’re creating additional work for ourselves to properly incorporate materially higher future inflation than we have seen in the past.

Because inflation affects us personally, this can lead us to an optimism bias – a tendency to say “it won’t be that bad”, or to over-focus on one particular piece of evidence that gives the answer we’re hoping for (an example of anchoring).

Group-think

As a rule, we feel much more comfortable raising issues in a group setting when we expect them to be well received. This contributes to the well-known phenomenon of conformity bias, or group-think. We see extreme examples of this every day on social media.

Group think is a particular challenge in the face of a rapid shift of evidence away from the status quo. Even if we overcome our own cognitive inertia and emotional biases around inflation, we then have to overcome other people’s in conversation.

Challenging the status quo can be a daunting prospect and lead to self-censorship – ie the fear of not putting forward your point of view because of fear of repercussion or embarrassment. If lots of committee members do this, difficult decisions can be diluted by the desire for “consensus”.

Coming in to land

There’s a risk that we’re under-reacting to inflation – that we’re bringing our reserving processes in to land when the inflationary winds are too strong, and we would be better off changing our approach more radically.

We have a wide range of techniques available to help us manage inflation in our reserving process, but the various cognitive and emotional factors may mean we don’t make as much use of them as the evidence says we should. As such don’t necessarily need a completely new inflation-busting reserving technique to help (though that would be nice!). We just need more impetus to change our flight plan and use the tools that we already have.

As published in The Actuarial Post, June 2022. Original article here.