By Jade Lagrue, Associate Consultant and Charl Cronje, Partner, LCP.
Case reserving is imperfect
Claims handlers are responsible for setting case reserves day-by-day, and they generally aim to maintain a consistent degree of case reserving strength over time. However, even for the best claims teams, retrospective analysis tends to show that case reserve strength varies by more than you might expect.
That variability in case reserve strength doesn’t necessarily mean the claims team is underperforming. More often, it simply reflects that, despite best efforts to respond to changes in the claims environment, it is extremely difficult to reserve claims consistently over time. Once we accept this, we can start to measure the variability in case reserving scientifically, rather than allowing it to become an emotive issue.
Proactive or reactive?
A further problem is that, although actuaries attempt to monitor case reserving strength as part of their reserving work, this tends to be retrospective rather than proactive. By the time actuarial analysis reveals that case reserving has weakened materially, it is often too late to address it.
A better approach is to have leading indicators of changes in case reserving strength, enabling firms to reflect this in something closer to real time in reserving, pricing and capital modelling. A case reserving strength index can help to achieve this.
Case reserving strength index
A case reserving strength index should ideally consist of a basket of measures, each of which gives clues as to the current strength of case reserves. It’s a basket because no one metric can capture case reserving strength adequately on its own. For example, an increasing paid to incurred ratio could indicate that case reserves are weaker, but it could also just mean that we are paying claims more quickly.
Another helpful measure is the value of recently settled claims against their incurred, say, six months prior to settlement. That essentially tells you how well we are predicting ultimate settlement values a few months out (or, for longer-tailed business, a few years out).
A further example is triangle analysis using reported cohorts. These are valuable because, with reported cohorts (rather than accident or underwriting bases), once claims are reported, we’re directly tracking the case reserving journey. These are some examples from the three main families of metrics that can contribute to a case reserving strength index. In practice, a range of KPIs from these families can be basketed together to create an index that gives us an objective measure of how case reserve strength today is likely to differ from that in the past.
What’s the payoff?
In a typical quarterly reserving process, there may not be time bottom out emerging issues in case reserve development and fully reflect these in the actuarial projections in the same quarter. Instead, it often takes until the next quarter to be able to understand and factor in such issues, which means that the appropriate response to new information is a quarter out of date. But what if, on day one of the quarterly process, we have an indication from the case reserving strength index that case reserves are expected to be 5% weaker than normal? That gives us a real chance to build this into the ultimates in the same quarter.
Claims event timeline
Experienced reserving actuaries know that it is essential to understand the history of a book of business, in particular the various distortions that will have occurred to the data over time.
For example:
We may have suffered a backlog in processing claims payments, leading to a lag in the paid patterns.
We may have revised our standard case reserves, leading to a step-change in incurred claims development.
A legislative change may have led to a speeding up in settlement for all claims reported after a given date.
In any mature book of business, many internal and external changes and shocks will have occurred. A handful of experts will know about these distortions. A claims event timeline aims to get all of that information out of people’s heads and into a system where everyone can use it. The claims event timeline provides a structured record of what’s happened in a book of business over time, and that shared history allows everyone to interpret trends with greater confidence.
Monitor first, modify later
Initially, the case reserving strength index should be focused on understanding the underlying variability in case reserve strength better, rather than seeking to change what case handlers do day-to-day. Otherwise, there is a danger that we influence claims handling behaviour, distort the normal trends and make the job of analytics even harder.
The case reserving strength index is just the first step in developing your claims analytics capability. Once you have a handle on that, you can begin levelling up. The path could involve claims journey analytics, breaking the claims journey into various stages and analysing how the claims development within each stage is changing over time.
This can be taken further by introducing machine-learning algorithms trained on detailed claims transaction data, to develop a mathematical picture of what “normal” claims handling behaviour looks like. These can then flag anomalies for further investigation, giving a more holistic early warning system for any claims handling changes.
In all cases, the claims event timeline provides a backbone for the analysis, ensuring that known past distortions are taken into account, so that more accurate inferences can be made about claims handling trends.
Next steps
Best practice in the area of claims analytics is developing fast. This is an area where there is a lot of value to be unlocked. Insurance is a learning business – firms who can understand their claims experience more quickly than others will gain a competitive advantage by being able to price more accurately and improve the efficiency of claims handling.
|
