Scheme Funding
April 2021

The Risks of De-Risking

DB regulatory focus is based on the view that schemes are carrying too much investment risk relative to the employer's ability to repair deficits.

The Pension Regulator’s standard response to a weaker covenant is to require de-risking of investments, with the objective being to minimise investment risk, reduce funding level volatility and reduce the severity of the impact of future downside events.

As a consequence, de-risking is now a key item on trustees’ agendas, with trustees’ wanting to ensure that the employer has the ability to support any investment risk (which typically means making sure the employer can make good a VaR loss over a relatively short time period).

Against this background, the purpose of most strategic investment reviews is to choose between asset allocations that minimise VaR and reduce interest-rate and inflation risks. No consideration is given to actual outcomes for members and the impact that de-risking has on pensions.

Proportion of Pension Payable

Consider a simple scheme that promises to pay pension of £100 per annum for the next 10 years. The total pension due is £1,000 (£100 per annum x 10 years). However, the sponsor defaults after 5 years and the scheme is 50% funded when this happens. The pension actually paid in this scenario is £750 (£100 x 5 years + 50% x £100 x 5 years) and the proportion of total pension paid is 75% (£750 divided by £1,000).

Trustees and employers want to ensure members receive their pensions in full when they fall due. If they can't guarantee pensions, then trustees and employers will want to make sure members receive as much of their full entitlement as possible. Trustees and employers could, therefore, consider the ‘expected proportion of pensions paid’ to measure the success of their DB strategies. That is, the sum of pensions expected to be paid divided by the sum of pensions promised.

In practice, the expected proportion of pension payable will depend on the probability of sponsor failure, what assets the trustees can recover on sponsor failure, scheme funding, investment strategy and investment risks, interest rates and inflation, corporate profitability, the ability to make deficit contributions and the inter-relationships between all these items.

Traditional asset liability modelling considers investments and funding alone and ignores covenant. To calculate the expected proportion of full pension payable, it is necessary to carry out a stochastic IRM exercise that considers the complex interplays between covenant, investment and funding and their impact on pensions.

The Impact of De-Risking on Member Pensions

De-risking will mean that an underfunded scheme is expected to remain in deficit for longer. As the scheme is in deficit for longer, it is more exposed to the risk of employer insolvency. By de-risking, the trustees may only have exchanged investment risk for covenant risk, and it’s not clear if the position for members has actually been improved.

To assess the impact of de-risking on DB pension benefits, we carry out a stochastic IRM modelling exercise that allows for covenant, investment strategy, funding, market conditions and market volatility, the interplays between all these items and the proportion of full pensions paid under each simulation.

We consider covenant risk in a similar way to the default risk for corporate bonds of different ratings. In the event of employer insolvency, we assume that the trustees’ secure pensions with an insurer. No allowance is made for any recoveries from employer after insolvency.

The results of our modelling is set out in the table below, which shows the allocation to growth that maximises the proportion of full pension paid to members.

Growth Allocation that Maximises the Expected Proportion of Pension Paid
  • As expected, a scheme that is fully funded could guarantee pensions by investing in a portfolio of gilts that cash-flow match future pensions
  • If a scheme is backed by a very strong employer, then benefits are secure with all investment strategies and whatever the funding position. It’s not that there isn’t investment risk, there is, but this risk is borne by the employer who is strong enough to ensure pensions are paid in full in all circumstances
  • Benefit security is poor in schemes that are poorly funded with weaker covenants. However premature de-risking reduces benefit security further. This is because premature de-risking transfers investment risk to a weaker covenant with poorer outcomes for members
  • A large allocation to growth should be associated with lower funding levels and weaker covenants
  • In general, benefit security is optimised with a larger allocation to growth, with de-risking only happening later and after the funding position has recovered

Conclusion

Incorporating covenant risk into traditional asset-liability modelling challenges the conventional view that the best outcomes for DB members are achieved by having a higher funding target, de-risked investments and shorter recovery plans. This is especially the case for schemes with weaker covenants.

The exact impact of investment and covenant risks on benefit security for a particular scheme is very scheme specific. It depends on scheme specific factors such as maturity and how the covenant may change with changes to market and economic conditions.

However, IRM modelling suggests that, in many cases, a larger allocation to growth assets makes sense. That is, instead of increasing the future reliance of the scheme on a (possibly weakened) employer, better outcomes for DB members can be achieved by meeting pension promises from a diversified pool of growth investments alongside the employer covenant.

Please contact us if you would like to discuss this