Insurance firms are facing a unique set of challenges as we approach year end, with the pandemic continuing to cause both economic and non-economic impact.

Looking back to half year disclosures, as expected, most firms considered it too early and too uncertain to make changes to long-term mortality, longevity or lapse assumptions. Some firms chose to hold reserves to cover potential short-term impacts – for example, excess mortality claims in H2 2020 or lapse rate spikes if government intervention measures were switched off. Others considered the overall impact on the balance sheet – including offsets between annuitant and assurance mortality, for example – and concluded that an immediate change was not required overall. While others performed additional stress and scenario testing to ensure their balance sheets were robust to further changes.

With mounting uncertainty around the course of the pandemic over the winter months, and a second wave well underway, the same uncertainty is likely to exist at year end. With most firms’ year-end assumptions now largely set, this short article considers both how these impacts were factored into that process and what triggers insurers may monitor to ensure the assumptions remain appropriate.

Economic impacts

Earlier in the year, as the pandemic was declared and its effects caught hold, we saw a huge increase in volatility, widening credit spreads, a further drop from already low interest rates and a huge fall in the equity markets. While we have seen some recovery in the market, as cases rise across the UK and worldwide, it is not clear how the market will react over the next few months and, ultimately, what the long term impact will be on the economy. This is causing significant challenges for setting assumptions.

Given the prescribed nature of aspects of Solvency II, there is limited ability to “set” economic assumptions, although bulk annuity writers will be attempting to formulate their best views of future retail and consumer price inflation and to calibrate their limited price inflation curves.

The main impact is likely to be on the IFRS balance sheet where a number of insurers will need to establish their best view of credit default and downgrade allowances in order to set appropriate credit provisions. Although a number of the large insurers disclosed limited impacts from downgrade and no defaults on their credit portfolios, the likelihood that this remains the case over the next 18 months seems low.

For internal model firms where this is relevant, the Solvency II calibrations for volatility risks will need careful consideration with index highs seen earlier this year. And credit, interest rate and inflation risks will also need to be considered – but, given the instability in the market, it might still be considered too early to make any changes.

Mortality

Although not as stark as the population, there has been an impact on the death and disability claims within insured populations during 2020. Short-term mortality impacts will depend on the demographic of policyholders and the mix of business as well as operational aspects such as the lag on reporting and processing of claims. It has been important to analyse data representative of insured populations, from either internal or external sources, in order to make decisions about assumptions as we enter the second wave.

The long-term impacts on mortality and, in particular, mortality improvements will be a lot more complex and will require far deeper analysis. The long-term impacts can be thought of within three buckets:

1. Those that we know will happen or have happened and are able to quantify with appropriate data. For example, delays in cancer treatment. We will be able to get data on the delay to these services and the expected impact on mortality.
2. Those that we know will happen but are difficult to quantify. For example, the impact of the economy on mortality rates. We know this will have an impact, but it is hard to quantify both what the economic recovery will look like and the impact it will have on mortality.
3. Other changes which are hard to define and quantify. An example would be changes in behaviour of people with regards to hand washing or smoking and the potential impact of these.

Although it is possible to start to quantify potential impacts on “best views” of the above, the situation is constantly evolving. Insurers have developed their own views (which may differ between firms) on the impacts and these have been used to justify the decisions made with respect to long term mortality assumptions at year-end 2020. It is important for insurers to derive appropriate metrics to monitor how the views used to derive the assumption change over the course of the next few months and years.

New business and lapse experience

Since March nearly ten million jobs have been furloughed and unemployment rates have increased to the highest levels in three years. It is unlikely that all those furloughed jobs will have been saved by the government scheme and as the year progresses, we are likely to see increasing unemployment rates.

Across the insurance industry, contribution rates for group pensions schemes have already fallen, driven by the employers furloughing their staff on 80% of their salary. As redundancies increase, we are likely to see further reduction in contributions. Ultimately these are likely to result in contributions ceasing or policies transferring in the long-term, as individuals find employment elsewhere and are encouraged to consolidate their pensions savings and perhaps existing pension pots with their new provider. Therefore, short term provisions for potential policyholder actions have been considered by a number of insurers with group pensions business.

Reductions in potential new business levels may also impact per policy expense assumptions with lower new business volumes to spread fixed costs. The impacts will predominantly be driven by affordability where customers are experiencing difficult times economically.

Risk calibrations

Internal Model firms have the additional complication of trying to understand whether recent events have changed their view of the calibrations of risks which  to which their balance sheets are exposed. These calibrations are likely to come under the spotlight from the Board, particularly as there would appear to be more uncertainty within the best estimate. It is not obvious or necessarily appropriate that the SCR should increase due to the experience of the last 8 months, and we would expect additional validation in key areas to be performed to demonstrate this. For example, revisiting the appropriateness of any interest rate floors within the interest rate risk model; comparing the extent of modelled potential defaults, downgrades and sectoral concentrations is in line with worst case scenarios; or, checking that the operational risk scenarios considered are appropriately “out of the box”.

Trigger frameworks

As we approach year end, and the second wave of the pandemic progresses (including the latest lockdown measures for England announced on 1 November 2020), we are likely to see further market instability and a short-term increase in mortality rates.

As in previous years, insurers are seeking to finalise and sign off the non-economic assumptions in October / November. Typically this allows for the possibility that decisions on assumptions will be “re-opened” should any significant new information or experience come to light in the period to the year end – or, unless the event occurred after year end, to the period until disclosure. This has seldom been needed historically, but this year firms will need to ensure both the governance and processes which sit around their year-end reporting allows the flexibility to be able to reconsider assumptions if required.

The identification and monitoring of triggers will be more important than ever to enable the Board and Audit Committee to remain comfortable with the assumptions up to and through the year end. These triggers and metrics will need to allow for both internal and external factors. Internal factors would typically cover considerations such as experienced lapses or mortality claims significantly increasing in the second half of the year, while external factors may include changes in government intervention, changing policyholder behaviour due to the economic situation, changes in outlook for the number and demographic of deaths from the pandemic, NHS waiting times, and so on. And, in terms of credit allowances, triggers may include something as simple as the number of rating agency downgrades or simple early warning indicators such as CDS spreads or equity prices.

These are just examples of potential triggers and early warning indicators, and it is important that insurers monitor those important to them to help smooth the year end reporting process in what will otherwise be a very volatile period.

Our Life & Financial Services practice at Hymans Robertson has the expertise and experience to help you understand potential triggers for economic and non-economic assumptions and to help you to design and implement pragmatic trigger frameworks and governance processes. If these are areas you are currently grappling with, please get in touch with any of your usual contacts.