The IFoA Matching Adjustment Working Party discusses the fundamental importance of the Solvency II matching adjustment to UK insurers
With the first 12 months of Solvency II under our belts, it is clear that the matching adjustment (MA) framework is far from perfect. But was the allowance for an illiquidity premium under the previous regime any better? Given the choice, would firms go back to any elements of that regime?
With just two years until the UK exits the European Union, firms might just get this chance.
To help answer these questions, the IFoA Matching Adjustment Working Party has considered four key areas for improving the MA framework, taking into account the needs of a range of stakeholders.
We have also considered how the MA framework could be amended post-Brexit to deliver better outcomes for these stakeholders.
1 Rules versus principles-based regime
Solvency II is sometimes described as a shift away from rules-based regulation to a principles-based regime. However, the general industry view is that the MA rules are very prescriptive, and in some areas they are a step too far.
A good example of this is the requirement for individual assets to have fixed cashflows. Some assets provide a good match to annuity liabilities when considered in aggregate (for example, a portfolio of equity release mortgage loans), but they have to undergo complex restructuring in order to comply with the MA rules. The economics of the assets haven't changed, but the restructuring introduces additional costs (which may be borne by customers) as well as operational risk.
2 Approvals processes
Unlike the illiquidity premium under the previous regime, firms have to apply to the Prudential Regulation Authority (PRA) for approval to use the MA under Solvency II. The PRA has up to six months to approve an MA application. Each application requires detailed, line-by-line analysis of in-scope assets and liabilities.
Consider a firm looking to invest in a new asset class. Simple illiquid credit assets can be privately placed in as little as two weeks, potentially increasing to six months for more complex, bespoke investments. The time required to produce and process an MA approval application is therefore likely to exceed the time for which the investment opportunity is available. Adding to this, technical experts responsible for assessing new investment opportunities are now required to divert their focus to the MA application.
3 Eligibility criteria
One of the most distinctive elements of the MA framework is the concept of eligibility. To which liabilities can firms apply the MA, and which assets can be used to back those liabilities?
For assets, there is a seemingly simple requirement - the asset needs to deliver certain cashflows. However, the reality is that asset cashflows are very rarely fixed in every conceivable situation. Firms may therefore pass over some attractive investment opportunities because it is too difficult to get them to fit the requirement for certain cashflows. For example, some infrastructure projects include an initial construction phase, where the income during that phase or the completion date (or both) is unknown. Firms can earn up to an additional 100bps pa from such projects while having the opportunity to influence the project in early stages. However, the unknown cashflows in those early years are likely to render the asset ineligible for the MA.
4 Consequences of a breach of any of the MA rules
The MA rules are strictly binary. When a breach is identified, the firm only has two months to resolve it. If this cannot be done, then approval to use the MA is revoked and cannot be reinstated for a period of two years.
Suppose that, one month, a firm were to breach the requirement to have a closely cashflow matched portfolio. Most firms would be prepared to realise a material mark-to-market loss by rebalancing assets (even if the firm's investment managers would argue against doing this) in order to restore the matching position and avoid the severe hit to the balance sheet from losing the MA.
During the financial crisis, very few bonds were trading in the market, resulting in a limited supply of good-quality credit. Where portfolios had to be rebalanced, significant losses would have been suffered. If firms had been subject to the same two-month rule previously, under the old regulatory regime, then arguably insurance company balance sheets would have ended up in a much worse position.
Shortcomings of the matching adjustment framework
Severe consequences for breaches
Increased certainty over ability to earn illiquidity premia
Greater understanding of risks by both insurer and regulator
Greater transparency over assets backing annuities
Encourages greater focus on governance
Complex structures and arrangements needed in some instances to satisfy the rules
Insurers not able to take advantage of new investment opportunities
Arbitrary limits disqualify assets/liabilities which satisfy the majority of the requirements and meet the spirit of the rules
May lead to pro-cyclical behaviour
Increased cost and operational risk
In many cases missing out on c.100bps+ pick-up on illiquid investment opportunities
'Cliff-edge' eligibility for assets with prepayment risk
Inability to take on blocks of liabilities with handful of policies which have optionality
SCR to cover need to restore eligibility within just a two-month period under stressed conditions
The need to quickly restore MA eligibility post stress is also one of the reasons why capital requirements in respect of MA portfolios are very high. It means firms need a good supply of eligible assets outside their MA portfolios to plug any gaps that might arise under stress.
Therefore, apart from the potential pro-cyclical effects of having to resolve breaches of the MA rules in a very short time frame, additional capital and liquidity requirements necessitate a more costly annuity offering from insurers and, consequently, unaffordable pricing for many customers.
How could the MA rules be improved?
The MA rules are black and white in many areas, leading to significant practical constraints. We believe two changes could give firms more freedom, while still ensuring risks are being managed appropriately:
1. Introduce materiality thresholds; and
2. Allow firms to hold capital buffers to provision for risks, where appropriate.
Firms already have to comply with the prudent person principle under Solvency II. This stipulates that insurers can only invest in assets whose risks they can properly identify, measure, monitor, manage, control and report on.
So it follows that firms should be able to set appropriate capital buffers for asset (and liability) features that don't provide an exact fit to the MA rules.
Examples of where materiality and capital buffers could potentially be used are:
- Asset eligibility: Avoiding the need to restructure assets with small amounts of prepayment risk
- Liability eligibility: Allowing bulk purchase annuity transactions which include a small handful of liabilities with non-standard features
- Approvals processes: setting aside capital while you wait for the PRA to approve a new asset class, so you can add it to the MA portfolio straight away and not miss out on the opportunity.
Adopting these two changes could be a quick win following Brexit, alleviating many of the issues currently facing firms with significant volumes of annuity business.
Lessons from further afield
Supervisors in other parts of the world have been developing rules on the back of Solvency II. They have had the benefit of hindsight when picking those elements of the MA framework to adopt in their regulatory regimes.
Singapore is currently in the process of drafting its own rules, and Bermuda (which is deemed equivalent to Solvency II) also has its own version of the MA. Looking at these regimes, we see some notable differences to Solvency II:
- More inclusive liability criteria: predictability of cashflows is evaluated based on stress tests
- Less stringent hedging requirements: - for example, capital can be held in respect of currency mismatches
- More flexibility in terms of grouping assets and liabilities: for the purposes of satisfying matching tests; and
- Longer time horizons: for restoring MA compliance.
We believe that these amendments, or variations thereof, also have the potential to greatly improve the current MA framework under Solvency II. Many areas of the MA rules are too black and white. The application of materiality thresholds and the use of capital buffers to provision for certain types of risks would be big wins for insurers, customers, the UK economy and the PRA.
The Working Party sees the potential for substantial benefits if we use Brexit as an opportunity to refine those areas of the current MA framework that are badly in need of greater flexibility.
Ross Evans, Stephan Erasmus, Michael Henderson, Peter Maddern, Keith Neil, Andrew Kenyon and Ravi Dubey are members of the IFoA Matching Adjustment Working Party