Co-authored by Nick Horsfall and Kate Mijakowska
The House of Lords Industry and Regulations Committee recently published a letter (‘the Letter’) to the Treasury and the Department for Work and Pensions on the use of Liability Driven Investment (‘LDI’) strategies by pension funds. The Letter questions the accounting standards employed to value defined benefit (‘DB’) liabilities and comments on regulation, investing, funding and the payment of member benefits. We share our thoughts on these topics below.
Accounting standards and discount rates
To start with, we need to consider what DB pensions are. We interpret DB pensions as contractual promises made to scheme members for defined amounts, to be paid at a specific point in the future. This promise is made irrespective of the realised success of the scheme’s investments. Any shortfall in scheme assets is to be covered by the corporate sponsor, and should the corporate sponsor fail, this obligation may be taken over by the Pension Protection Fund (‘PPF’) – although, in this situation, a potential reduction to the promised benefits may be applied. In essence, DB pensions are an obligation created by the corporate sponsor and collateralised in the first instance by the assets in the scheme.
We believe that a natural approach to valuing a contractual obligation is to use instruments that are aligned in nature to the obligation itself. Payments linked to assets that are contractual would be our starting point, and other factors might include the (credit) quality of those payments and whether they should allow for an illiquidity premium.
This thought process would tend to favour assets such as:
- bonds;
- other forms of fixed-interest debt; and
- leases.
In terms of wanting a clear, objective set of assets, we understand why there’s a gravitation towards bonds for assessing the nature and value of DB pensions. The case is strengthened if there is an aim of having a homogeneous standard for measuring liabilities across different corporate sponsors.
Asset-led discount rates
The Letter covers asset-led discounting, and we discuss this approach in the context of assets where the return is non-contractual in nature, or where contractual payments are not highly certain.
Asset-led discounting often employs assets such as equities, where much of the performance derives from variable capital returns. This variability means that expected returns cannot be estimated with a sufficiently high degree of confidence to give the appropriate assurance that the pension promise will be met. It also creates a large downside risk to both the scheme and corporate sponsor should equities underperform. Moreover, periods when equities perform particularly poorly are likely to be somewhat correlated with times when corporate sponsors are under financial stress. This issue is particularly acute for pension schemes that are so big as to pose a meaningful risk to the solvency of the corporate sponsor itself.
Path dependency is another key issue – i.e. it matters when good and bad returns happen. A very relevant example would be a pension scheme where significant levels of member benefit payments are currently being made. This is often the case for the many UK pension schemes that have now closed. Poor returns in early years can have a disproportionately negative impact on the long-term health of a scheme – unless, of course, there can be total confidence that its corporate sponsor can make good on those poor outcomes.
Investment strategy
While accounting standards and broader regulation play their role, the nature of the promise and the necessity to meet pension payments on time is not affected by the accounting standard. Therefore, we believe there’s value in considering the investment problem more broadly. Investment strategy should be linked to the aim it’s trying to achieve. We consider two possible objectives that may be adopted in the context of a DB pension scheme:
Objective one: to reduce the range of outcomes (meaning to avoid having a large unexpected cost that may be difficult or impossible to cover) of a corporate DB pension scheme.
- We believe that most of the pensions industry aligns with this aim, which implies a need to take a prudent approach to investment risk. And this, we think, was a key factor that gave rise to the heavy adoption of LDI solutions.
- Standard asset allocation analysis under an assumption that the liability is known and is sensitive to interest rates will likely lead to a high allocation to high-quality contractual assets – potentially with some leverage to leave sufficient capital free to invest in assets that can increase expected returns.
Objective two: to target a low expected cost of benefit.
- Such an approach would lead to substantial allocations to assets with high expected returns, such as equities. In general, these asset strategies have high volatility levels, which create large downside risks for corporate sponsors and schemes.
- Such a strategy would risk failing to deliver on the pensions promised should these high-returning assets perform poorly, particularly over a sustained period. This is likely unacceptable without an additional safety net.
- That said, this type of strategy could be deemed acceptable, provided further arrangements are made to ensure a sufficiently robust safety net exists. The government – or the PPF (collectively funded by other schemes) – come to mind as the most obvious candidates to deliver such a net, but the practicalities are far from simple. Would these parties be willing and able to absorb shortfalls, assuming corporate DB schemes were to move to these higher investment risk strategies? Would the role of corporate sponsors need to change? It would be reasonable to expect the PPF levy to rise materially if the investment risk taken across the pension scheme market increases. This may perhaps be palatable if the nature of the PPF guarantee changed, such that it could withstand a substantial crisis (i.e. when many schemes need rescuing at the same time). Alternatively, the government could provide an implicit guarantee, but would that be acceptable to the taxpayer? This is not an exhaustive list of considerations; we aim to show that the issue is complex and requires thoughtful consideration.
Key takeaways
The Letter discusses accounting standards and specifically calls to question the basis of the discounting methodology used to value pension schemes. In practice, the question of valuation is an important one that has much wider implications than purely accounting – it’s also fundamental in scheme funding and investment strategy, which in turn are influenced by a much broader set of factors.
We believe that any discussion around pension discounting should consider the nature of the DB pension promise itself – and, in our view, it’s a hard promise that should be funded, invested for and protected in a way that reflects this. The current pensions regulatory, investment and funding ecosystem – including the use of LDI – largely stems from the collective understanding of the nature of that promise.
In light of the debate sparked by the Letter, we believe it’s relevant to consider the underlying objectives that DB pension scheme trustees should be tasked with achieving and the practical implications of these.
Unless indicated, these are the views of the author and may differ from those of the firm.
