Managing Pension Risk: The Need for More Accurate Cash Flow Analysis
by James Mushin, Director of Professional Services, PensionsFirst Analytics
Defined benefit (DB) pension scheme sponsors are waking up to the need to be able to calculate their scheme’s risk position more precisely in order to carry out risk-transfer solutions. Yet their ability to do so effectively is hindered by the traditional methods used to value their scheme’s liabilities.
The topic of DB pension risk has fast moved to the forefront of treasury debate. Stricter accounting and funding rules, low interest rates and volatile stock markets have been key drivers of this, as has the much-reported increasing life expectancy in the western world and the associated longevity risk that this poses.
The potential impact that these key risks can have upon the funding status of a DB pension scheme and the financial viability of its sponsoring company should not be underestimated. As such, more and more scheme sponsors are entering into discussions with third parties – such as insurance companies or banks – in search of ways to remove or hedge as much risk as possible. When considering these potential de-risking solutions, however, many sponsors – as well as pension scheme trustees – still rely on imprecise actuarial valuation information.
Fortunately, modern risk management platforms are now providing sponsors with the means to measure their DB pension liabilities with the regularity, precision and transparency typically available for other balance sheet exposures. And it is the development of this technology that has so starkly illustrated the inaccuracies of formal actuarial valuations which, in some cases, can miscalculate by hundreds of millions of pounds.
The traditional approach
Logically, it follows that DB pension scheme valuations should be at their most precise at the point of the formal actuarial valuation. At this time, actuaries should be in a strong position to advise on reasonable assumptions to use and hence calculate an accurate valuation that considers all available evidence. After this point, of course, the valuation may lose accuracy as changes in circumstances to individuals within the scheme, changes in investment portfolio valuations and changes in economic outlook render figures obsolete.
However, this theory does not always play out in practice. In fact, DB pension scheme valuations are often inaccurate at the very point in time that they are signed off by the actuary.
This starting point is then updated over time – in a series of ‘roll-forwards’ – yet these are merely approximate adjustments to the last formal triennial valuation.
And to top all this off, this inaccurate technique tends to be used again year after year to underpin investment strategy, accounting disclosures and other financially significant decisions, until the results of the next ‘accurate’ valuation are finalised.
The root of the problem: inefficient models
Undoubtedly, some of the inefficiencies stem from the length of time taken to complete the actuarial valuation. In many cases, by the time the valuation is finalised – sometimes up to 15 months later – the input data used can no longer be considered relevant and therefore the valuation is no longer accurate. Yet this is only part of the problem.
A key component of the formal actuarial valuation – and one that is critical in defining its precision – is the model the actuary uses to project the scheme’s future cash flows. However, actuaries have traditionally used archaic models that can no longer cope with the significantly more intricate demands of modern pension scheme risk management. In many respects, actuaries’ hands are tied by the rigid nature of these tools, forcing them to squeeze data into unsuitable models.
DB pension scheme valuations should be at their most precise at the point of the formal actuarial valuation.
For example, most of the models rely on rounded ages. As such, a pensioner aged 75.4 years must be treated as a 75-year old for purposes of modelling. In theory, it could be suggested that because, on average, half of scheme members’ ages should be rounded up and half should be rounded down, the approximations will even out over a data group. A theory that may hold true for the younger generation, perhaps, but one that does not for people of pensionable age. The difference in life expectancy between a member aged 75.4 and one aged 75 is greater than the difference for a member aged 75 and one aged 74.6, so the rounding does not cancel out. In fact, it can actually bring a bias into the calculations.
Most actuaries would accept that this is a limitation of the model but would argue that it is a limitation that they are well aware of and one that would be too difficult to eradicate. But for the close management of longevity risk, such inbuilt inaccuracies can create major distortions.