FTfm 'Feedback' column, why BT trustees should not just switch from equities to long bonds

by Dr. Ros Altmann

(All material on this page is subject to copyright and must not be reproduced without the author's permission.)


Your 18 February 'Talking Head' column raises some very interesting issues.  It is absolutely right that changes in pension accounting do not change the underlying economics of pensions and that understanding both the real costs and real risks of funding UK defined benefit schemes is critical.  However, investing in long-dated bonds, rather than equities will not, in my view, solve the problem of funding scheme deficits such as BTs nor will it remove the balance sheet and profit risks.

There are several reasons why holding long-dated bonds will not remove the asset/liability mis-match.  Pension liabilities may be 'bond-like' but they are not bonds.  UK final salary pensions are linked to limited price inflation – and often to salary inflation - but index-linked gilts are tied to the retail prices index.  Pension liabilities have durations far longer than almost all bonds in issue.  Using corporate bonds rather than gilts carries credit and default risk, but without the upside potential of other assets.  And there is also the problem of longevity.  So, bond investments still entail asset/liability mis-matching and suggesting that trustees should avoid assets like equities - that can return far more than long bonds - will virtually guarantee underperforming UK pension liabilities over time. 

Crucially, the investment landscape has moved well beyond just equities and fixed income.  Pension funds can access alternative risk premia that can deliver better returns than long bonds, by investing in a wide range of assets other than equities.  A diversified portfolio of return-seeking assets, together with swaps which better match long-term pension liabilities, should provide more efficient management of risks and returns than just holding bonds.

This is a key point.  Bonds do not necessarily reduce risk relative to pension liabilities.  They reduce volatility of returns, but that is not the same thing.  UK pension fund risk management should focus on paying promised pensions over the long-term – not on standard deviation of returns.

Modern methods of money management can help pension fund trustees match or outperform their liabilities better than the traditional investment approach using just equities and bonds.  Even in the case of a mature scheme like BT, holding bonds is not necessarily the best way to meet ongoing liabilities. 

Suggesting that BT trustees should hold £28billion of bonds, rather than the current £8billion, in order to meet the £1.4billion annual cost of pensions in payment, ignores some crucial points.  Firstly, the assumption of 5% bond yields implies accepting some credit and default risk in corporate paper, rather than long-term gilts.    Secondly, equities also deliver an ongoing income stream which can fund pension payments.  Thirdly, other assets such as real estate and infrastructure could also meet BT's income requirements.  Property and dividend yields are close to or even above the 5% level, but they additionally offer potential for higher long-term returns than bonds. 

So, even for regular income requirements trustees need not rely only on bonds.  Lack of supply and burgeoning demand has driven down yields with no sign of relief at the long end.  Indeed, as interest rates have fallen significantly in recent years, trustees need to look for other income sources.   Some radical new thinking is required.  For example, especially after recent falls, global property – hedged or even unhedged - could produce both income and capital growth superior to long gilts from current levels.  A portfolio of properties with top-quality tenants on long-term leases with upward-only, inflation-linked rent reviews could deliver income over time at least on a par with high quality bonds, while offering better risk/return characteristics relative to pension liabilities. 

To sum up, the investment management of pension schemes must not be driven by a misguided desire to dampen 'volatility' relative to changing accounting measures.   UK pension fund trustees should be encouraged to think beyond the traditional 'equity-bond split' when managing both the returns and the risks of their funds.  This could entail a diversified portfolio of alternative assets to capture new sources of beta returns with lower volatility than relying just on equities, while using hedging or swaps to protect the downside and better match liabilities.  Diversification, downside protection and managing both returns and risks is complex and costly but surely more likely to succeed than just locking into bonds.


© Dr. Ros Altmann  |  Home  |  Profile  |  Disclaimer