We held our latest panel debate entitled De-risking - De-rigueur or Debacle in the making? on 9th February 2022. Aoifinn Devitt kindly hosted four panellists from both sides of the Atlantic for a live hour’s discussion:
Pete Drewienkiewicz: MD Global Assets, Redington
Elizabeth Carey: Independent Investment Adviser
Eric Lambert: Independent Investment Adviser
Christopher Levell: Partner NEPC.
Should public sector pension funds de-risk?
Over the past 15 years private sector DB funds have de-risked as they have become more mature and better funded. For example, the average UK DB pension scheme equity weighting has fallen from 53% to 26%. This debate was about whether they have been wise to do so and whether public sector funds such as the LGPS should follow their path.
Our panellists started by making a big distinction between public and private sector funds: open vs largely closed; a state guarantee vs the risk of sponsor default; asset- or inflation- based rather than gilt-based actuarial discount rates.
We asked what de-risking really means: is it simply diversification of assets, or a shift in allocation as the number of active members reduces? If the latter, the panel was adamant that assets should remain return-seeking. They noted that the more successful approach in the private sector has been to de-risk (past) liabilities and accept risk on the assets side to generate return. Leverage through derivatives has proved a useful tool to help de-risking while maintaining return.
By contrast, US public sector funds are in the main substantially less well funded than the LGPS. They were fully funded in the 1990s but found themselves with targets on their backs and under political or even legislative pressure to lower contributions or raise benefits. They also tend to set return objectives in nominal rather than inflation-linked terms (e.g. 5% annualised return) and in some cases vary pension pay-outs according to the funding level.
What really matters when investing in a pension scheme
In the course of the conversation we came back to some of the bedrock facts of administering a pension scheme: they accumulate assets first and decumulate them later, and will inevitably go from cashflow positive to negative at some point; the only alternative to raising risk is to raise contributions; the promise to employees today is much more expensive because we all live longer; and the main purpose of actuarial valuations is to set employer contribution levels.
Is the funding level so crucial?
We looked at whether there was a funding level which should trigger some de-risking. Here there was criticism of the deterministic nature of actuarial assumptions which assume a single return for each asset class. There was also questioning whether they reflect demographic shifts, the need for fairness between generations, and in the LGPS between employers and taxpayers.
One panellist pointed out that private sector models where the discount rate is tightened as funding levels rise, have worked well. It may not be as appropriate for open funds, but it is possible to use different discount rates for different age cohorts and invest accordingly. Some funds in the LGPS do in fact use different discount rates for different employer cohorts, but this approach could be extended.
We agreed that stochastic (Monte Carlo) models provide a better picture, but they are sensitive to the return assumptions and (from the floor) the correlations and relationships which underpin them may be open to question. Does de-risking really mean minimising the distance between the expected outcome and reality? Or does it mean reducing the size of the cone of outcomes?
How to keep contribution levels affordable
We then looked at the sustainability of contribution levels, which is enshrined in LGPS legislation. Panellists agreed that there was a level at which contributions became unsustainable, and the focus had to be on affordability over stability. Should the actuary take into account the ‘1 in 20’ bad outcome in their calculations? How are exogenous factors (e.g. a sudden dip in market prices at valuation time) which might trigger unnecessary regulatory action taken into account? Are actuaries too prudent anyway?
We kept coming back to the example of USS, where there are proposals either to cut teachers’ pensions by 25% or to raise total contributions to nearly 50% of pay-roll. The panel agreed that both these were unacceptable outcomes which had to be avoided by better management of risk and assets.
Unlike the U.S., where pension payments can vary, panellists were clear that in the U.K. employers take the balance of risk, and any surplus should be returned to them. This could be done either by using negative secondary contributions or by creating an asset pool to reduce future contributions.
How does the back-stop affect investment decisions?
We also looked at the back-stop in the worst outcomes. In corporate schemes there is a real ‘risk of ruin’, but over the past twenty years the assumption has been that someone in authority will come to the rescue. One panellist suggested that there needs to be more harmonisation of governance, so that outcomes in these situations do not vary excessively. Another asked how in the same world was it possible for US public sector pension discount rates to be more than 7% while LGPS funds were 3 to 4%?
The LGPS has a crown guarantee behind it, but does that mean less actuarial prudence is required? One panellist suggested a gilts-based discount rate is inappropriate because there is no risk of insolvency. However, a counter-thought was that in a situation where the guarantee was called on, it was likely that a government would find itself funding not only the LGPS but unfunded schemes too, and it would not be able to stand behind its promise.
Hedging against inflation risk
Finally, we looked at hedging against inflation risk. It is not so relevant for the U.S. where the remit tends to be in nominal terms, but the LGPS has an uncapped inflation-linked liability. One panellist was clear that this is best done through purchasing suitable assets such as infrastructure, as the opportunity cost of index-linked gilts (the usual private sector route) is too high. Another suggestion was to use FX hedging to exploit the fact that sterling would inevitably fall in a more inflationary environment.
In summary, this was a dense conversation with many insights from our panellists, and we thank them all, as well as Aoifinn Devitt, and our audience. The general conclusions were that open pension funds such as the LGPS are obliged to take some risk, and that de-risking is probably better done on the liabilities side rather than the assets. The focus needs to be on keeping employer contributions affordable, which may mean engaging robustly with actuaries at the upcoming valuation.
Recommended reading
Andrew Smithers - The Economics of the Stockmarket (to be published 10th March 2022)