By Richard Marshall, Director in Willis Towers Watson’s Insurance Consulting and Technology business and leads the development of mortality and demographic risk models for our UK business.
For pricing new business, whether protection or annuity business (in particular), this fails to counter the risks posed by COVID-19: data distortion, potential future waves of infection and behavioural changes amongst policyholders.
Also, existing best-estimate assumption-setting processes may be unreliable due to a new year of unrepresentative data, changes in the demographic profile of new customers and lapses, uncertainty about improvements in mortality linked to economic performance, additional waves of COVID-19 and delayed impacts of the lockdown.
For insurers there is a further risk of an asymmetry of knowledge with reinsurers, who are likely to have invested more time understanding the impacts of COVID-19.
Finally, regulators and auditors may expect insurers to show consideration of the effects of COVID-19 on their business in their assessment of capital adequacy, in their business planning and in their assessment of the adequacy of their premiums.
There are also opportunities to gain a competitive advantage. In the bulk annuity market, a detailed understanding of COVID-19 deaths and the potential for select effects and other mortality impacts may make the difference between being able to win on a competitive price and losing out on a tender. There are also opportunities for product innovation for multi-line writers too, covering risks such as the risk of being furloughed or made redundant due to a future wave of COVID-19.
Pricing and underwriting
Both the base mortality and mortality improvements assumptions used for pricing will likely need to be updated to allow for the effects of COVID-19. Companies will need to:
• Remove the effect of COVID-19 from recent deaths in their experience analysis; we have developed an approach to doing this (even when the number of actual COVID-19 and related excess deaths is not known exactly) and can tailor this to the A/E analyses used by life insurers to estimate mortality rates as proportions of standard tables.
• Have an evolving view of possible excess mortality in the second half of 2020, and also into 2021 (e.g. due to second or subsequent waves) to allow an agile response to new information (vaccine/treatment progress, policy change and levels of adherence to social distancing).
• Develop a view of how non-COVID-19 mortality improvements may be accelerated or dampened by a range of economic, physiological and behavioural factors. Moreover, they will need a view as to how rapidly (or otherwise) excess mortality from COVID-19 deaths will decline, allowing for a perhaps unusual shape to mortality improvements not normally achievable through use of the CMI model alone.
In research recently undertaken for the life industry on the effects of COVID-19 in the UK, we looked at potential ‘second wave’ scenarios and considered the potential effects of some of these factors on future improvements.
Detailed analysis of the possible risks from considering a range of scenarios can make it easier for the relevant technical committee and/or the board to approve the base mortality and improvement assumptions for use in pricing in the aftermath of COVID-19. It changes an unknown level of uncertainty into something which can be described and measured. And since insurers will wish to update their persistency assumptions, the same future COVID-19 scenarios will also be able to inform this decision.
ORSA development
Many insurers have been considering ORSA scenarios to assess the impact of additional waves of COVID-19 on existing business plans. Such work can also help to optimise business planning for a post-COVID-19 world.
Alongside demographic effects, insurers will need to consider asset-side shocks and further changes to new business volumes.
Consistency between the mortality effects and the other shocks will be important if these scenarios are to demonstrate awareness to the regulator of the whole range of risks posed by different possible ‘second wave’ outcomes.
Capital modelling
The pandemic has highlighted a key weakness of the Solvency II standard formula mortality catastrophe allowance – its age-, gender- and region-agnostic strength. Internal model companies may have seen this component of their model as unimportant and been satisfied with a similar flat stress, but may now wish to investigate more granular stresses.
Dependency assumptions will no doubt be in the spotlight too: COVID-19 has brought about material shocks to equity values, credit spreads and interest rates, and may have triggered operational risk events. Many companies will have surmised so far that their capital ‘by risk’ has been adequate, but the dependencies have not, and these dependency assumptions should be revisited in light of recent experience. Companies also need to beware of adjusting too much to allow for the precise nature of the coronavirus shock: the key is to use these lessons to prepare better for the next shock, which is unlikely to be identical to what we have just seen.
Financial disclosures
Finally, there will be considerable interest from investors in the response of (re)insurers to the changes and challenges brought about by COVID-19. Companies will therefore need to consider the messages they present to the investor community about the present and future effects of COVID-19 on their business, and their preparedness for further waves of the pandemic.
Through analyst presentations or financial reports, they will need to address likely investor concerns around the effect of COVID-19 on sustainability of business models, risks to dividend paying capacity and earnings growth, and key judgments underlying the financials which may need revisiting in light of COVID-19.
|
