27 Nov 2018 5 min read

DB scheme health: another way

By John Southall

The traditional way of looking at pension scheme solvency is in terms of deficits.


This leads to interesting arguments between actuaries and financial economists in terms of how large those deficits should be. From the perspective of financial economics, as explained in a hugely influential paper by Jon Exley, Shyam Mehta and Andrew Smith in 1997, pension liabilities should be valued using market bond yields. The investment strategy followed - which obviously concerns the scheme's assets - should be irrelevant to the valuation of the liabilities. Thinking otherwise amounts to saying one can conjure money out of thin-air, simply from changing investments. An analogy is pretending your ISA is now worth more because you just moved it from cash to equities. Of course, the distribution of future outcomes has changed (perhaps favourably depending on your attitude to risk) but valuations should not take credit for higher expected returns from assets without allowing for their additional risk.

The question is how, and why, do pensions actuaries get away with allowing for a risk premium in their Technical Provisions bases? The answer, I believe, is that actuaries are trying to capture not only the size of the objective deficit but also how manageable that deficit is, albeit in a way that is repulsive to financial economists.

How manageable a deficit is isn’t obvious purely from its size; it also depends on the strength of the sponsor, the size of the deficit relative to the size of the assets, the quality of the investment strategy (e.g. in terms of diversification, hedging, costs, liquidity management and appropriate return target) and the extent of economic and demographic risks in the scheme and how well these risks are managed. 

However, the Technical Provisions funding position can be rather a poor measure of scheme health and can give a misleading impression of benefit security. A 100% funding level, for example, does not correspond to a 100% chance of meeting all benefits even if there is no investment risk in the scheme. This is because of demographic risks such as longevity uncertainty. Likewise covenant strength matters: a large deficit for a strong, large sponsor is not nearly as worrying as for a small, weak sponsor.

Funding position assessments, based on deterministic assumptions, are dated, originating from a time before computing power was as cheap and widely available as it is now. We believe a useful approach is to model all the risks a DB scheme faces - economic, demographic and sponsor (covenant) - for thousands of scenarios of the future. A scheme deficit could be deemed manageable if the distribution of the final outcomes for the scheme - in terms of the chance all the benefits are met, and the severity when they are not - is sufficiently attractive in some sense.  

As an illustration of our approach we are tracking on a quarterly basis how one of our metrics for scheme health - the Expected Proportion of Benefits Met (“EPBM”) - evolves for a typical DB pension scheme in the UK. As at 30 September 2018 we estimated its EPBM to be around 93%. Such measures are no silver bullet and should not be used as an excuse to downplay the financial cost of a pension scheme. However, we believe they offer a powerful lens through which to understand scheme solvency and guide investment strategy, such as when to move into cashflow matching strategies and how to adapt the strategy to allow for sponsor risk. This can support Integrated Risk Management, ultimately helping trustees ensure they can meet all benefits as they fall due.


John Southall

Head of Solutions Research

John works on financial modelling, investment strategy development and thought leadership. He also gets involved in bespoke strategy work. John used to work as a pensions consultant before joining LGIM in 2011. He has a PhD in dynamical systems and is a qualified actuary.

John Southall