This article first appeared on Hymans Robertson’s website and has since been republished in Professional Pensions. It argues that loading a Defined Benefit Scheme with long dated bonds is a good thing. Thank you Consultant James Mullins
Headline news in 2001 is now good news for members, says James Mullins
20 years ago, in the autumn of 2001, the Boots Pension Scheme hit the headlines because it had invested all of its assets in high quality long-dated bonds, selling over £1bn of equities in the process. I was only three years into my pensions career at the time but I remember being fascinated by this ground-breaking development.
At the time, it was a unique and surprising thing to do, because there was little focus on managing pension scheme investment risk. Indeed, in 2001, pension schemes in the UK invested 75% of their assets in equities (growth-focused, with risk) and less than 20% in bonds (low risk, matching assets). Liability-driven investment (LDI) strategies were rare and the buy-in market was non-existent, with bulk annuities for pension schemes only being used in situations where the sponsoring employer had become insolvent. This backdrop is why the Boots Pension Scheme’s move to a 100% bond investment strategy really was headline news and there was even criticism from some people in the pensions industry.
Looking back on this today, it’s hard to believe that a pension scheme actively managing down investment risk was headline news. It makes you realise how far defined benefit (DB) pension schemes have come in the last 20 years, with all the significant work to manage investment risk to reduce the reliance on sponsoring employer covenants and make members’ benefits more secure.
Today, in almost complete contrast to the position 20 years ago, DB pension schemes in the UK invest 70% of their assets in bonds (low risk, matching assets) and only 20% in equities (growth-focused, with risk). Furthermore, pension scheme LDI strategies are commonplace and now hedge an impressive £1.5trn of interest rate and inflation risks.
Cashflow-driven investment (CDI) strategies, which aim to match the payments a pension scheme expects to make to its members, have taken things a stage further. Further still, the bulk annuity market has taken off since 2007 with around £200 billion of pension scheme liabilities now having been fully insured via buy-ins and buy-outs. In addition, longevity swaps have insured the longevity risk associated with a further £100bn-plus liabilities.
In just 20 years, we have gone from a position where a pension scheme deciding to sell its equities and invest only in bonds was headline news, to a position where bonds, and more intricate matching assets, are absolutely the investment of choice and 40% of FTSE 100 companies that sponsor DB pension schemes have now even insured a large proportion of their liabilities via buy-ins, buy-outs or longevity swaps.
Projected growth in those risk transfer markets means that we expect £1trn of DB pension scheme risk to have been insured in ten years’ time. To put that into context, £1trn of insurance would be equivalent to around half of the value of all gilts currently issued by the UK government or around half the value of all of the companies in the FTSE 100.
All of this is good news for the members of DB pension schemes. Their benefits are now more secure, with less reliance on long-term support from sponsoring employers and I expect this trend to continue at pace. We should be grateful to John Ralfe and others involved with the Boots Pension Scheme in 2001 for having the foresight and courage to lead the way in significantly managing pension scheme risk.
James Mullins is head of risk transfer solutions at Hymans Robertson
This article was also recently published in Professional Pensions. It argues PWC’s view that loading a Defined Benefit Scheme with long dated bonds is a bad thing. Thank you Consultant Raj Mody.
Pension schemes have been “shoehorned” into valuing liabilities against gilts, creating a “herd mentality” that does not reflect scheme funding accurately, says PwC.
Any change to the defined benefit (DB) funding regime should therefore refrain from tying parameters to gilts and instead focus on scheme cashflow obligations, the firm said.
Responding to The Pensions Regulator’s (TPR) consultation on the principles to underly the proposed DB funding code, PwC said ditching gilts as a benchmark could open up £40bn of extra value.
The consultancy has now urged the watchdog to recognise in the funding regime that schemes can back pension commitments with a diversified portfolio of cashflow-matching bonds and low-risk income-generating investments outside of the gilt arena.
Global head of pensions Raj Mody said: “The pensions industry has been shoehorned into an undue focus on referencing everything back to gilts, as a so-called risk-free benchmark. While that doesn’t stop individual schemes doing their own thing, the trouble with this kind of reference point is that it creates a herd mentality. This then puts pressure on trustees or companies looking to follow a more bespoke approach, even if it’s a better strategy for their own pension scheme.”
Throughout the last decade, scheme investments in gilts nearly doubled from 23% to 45% of assets, while the proportion of pension funds related to paying current pensioners is 40%. PwC said this demonstrated that there was a drift towards using gilts, partly as a result of gilt-based rules and regulations, even if this may not necessarily be the optimal solution for all pension funds.
The regulator should refrain from underpinning the future funding framework with gilts, instead ensuring the parameters remain flexible enough to allow a focus on the future cashflow obligations of pension funds and the best assets to match those.
By not focusing on a “single-point valuation” and comparing liabilities to a “single-point asset number”, extra value can be unlocked to benefit pension schemes and facilitate more diverse investment strategies, the consultancy said.
Mody said previous funding regimes, including the Minimum Funding Requirement, recognised the need to treat funding of existing pensioner liabilities differently.
“If anything, the need for something like this is now more acute. Pension schemes are now doing exactly what they exist for – to pay out retirement incomes. But that makes it all the more important to get the new funding framework right, both for existing and future pensioners.”
Thanks to the excellent Professional Pensions for promoting two such widely different views on the same issue. The past 20 years have seen the existing members of DB schemes provided with increasing security. That security has had to be paid for by higher contribution rates that have meant that those not in DB schemes have suffered lower pension contributions and less security.
Has liability driven investment saved our occupational pension system or done more damage than good? I leave you to decide.
Comment from Con Keating
the attached comment could not be posted in the comments box and so has been included as a blog- appendix
Anyone wanting to participate in the equity/bond asset allocation debate would do well to read
“J. E. Woods (2020) New exercises in decomposition analysis, Journal of Post Keynesian Economics, 43:1, 36-60.” This demonstrates and examines the components of returns for conventional gilts, index linked gilts and UK equities. In particular, I would like to draw attention to the importance of the revaluation effect, which can be considered a measure of speculation in a market. I reproduce Figure 2 from that paper below. Note that these are rolling ten year averages.
In particular I would draw attention to the revaluation effect which reached a maximum in the year 2000 and its nadir in 2007-8. Note also the resultant high correlation with the total real return over this period. Speculation was rampant in the dot.com bubble and minimal during the global financial crisis. This revaluation effect metric may be considered a measure of the riskiness of the asset class.
The equivalent diagram, Figure for index-linked gilts is markedly different – and frankly alarming. Its rise since the advent of the Pensions Regulator in 2006 has been almost monotonic. Its level, at 13.3% dwarfs the total realised return of 4.6% It is also far exceeds the highest value seen for equities; 9.4% in 2000, when the equity total realised return was 12.2%.
The situation for conventional gilts is quite different, the revaluation effect is within its historic range.
The take-away from this is a question: Can this really be the time to introduce the proposed new DB Funding Code.
There are a host of other questions which might have been asked, such as: what would the performance of equities or gilts have been had the pensions world not sold equities in favour of bonds. Equities now account for less than 30% of scheme asset allocation with approaching 60% in bonds.