When UK gilt redemption yields on a 10-year duration bond touched 2 per cent unparallelled in more than a century we really should have asked ourselves whether conventional bonds that are not indexed to inflation can possibly play the risk-reducing role given to them in traditional balanced management.
For bonds to perform this role, you have to be suffering from money illusion, meaning you cannot see the transfer of risk from short-term volatility to long-term purchasing power outcomes. A bad trade-off to make at the best of times, it is a shocking one when bonds are priced artificially. It leaves investors defenceless in the face of even modest inflation.
This false market came about partly through Government intervention in the pricing of inflation risk, which is what quantitative easing is all about. It manipulates the price signals that markets rely on to inform the choices that investors make.
The Government is aided and abetted in its intervention in the gilt market by two other externalities:
- accounting rules that make bonds appear artificially low-risk for hedging insurance and pension liabilities
- the convention of balanced management that makes it appear rational and good regulatory practice for private-client advisers to hold more bonds to reflect a reduction in a client’s risk appetite.
Ironically, when a central bank buys bonds to drive yields down in order to signal that it is rational to take more risk, these externalities prevent the signal from having the desired effect. Liability-driven investors increase their hedging-demand for bonds to prevent accounting deficits from increasing.
Can bond yields near 2 per cent be rationalised other than by these externalities? To answer this, we need to think in terms of probabilities. Risky assets are those with an uncertain chance of meeting an investor’s expectations. In the absence of foresight, rational investment in risky assets requires a favourable distribution of probable returns.
In nominal terms, 2 per cent nominal yields on 10-year bonds can hardly fall much further but yields could rise severalfold. This is clearly not favourable. But for most private clients, it is purchasing-power outcomes that matter most they can live with volatility but they live off outcomes.
To turn this into a more symmetrical, rational range of future real yields requires equal scope for both deflation and inflation. If the distribution of probable future changes in the general price level is skewed upwards, there is no symmetry in real returns from bonds either.
The price of gold is an indicator of what some investors think about skewness in the inflation process. However, many individual investors also appear to hold the same intuitive view that the combination of paper currencies and democratic elections leads governments to take far greater risks with inflation than they will with deflation. Since they also vote, the chances are that individual investors probably share the same bias. Most households have more to lose from sustained deflation than continued inflation.
In a skewed world, turning the nominal yield distribution into a real one might shift most of the probable range of real yields to minus 6 per cent (2 per cent yield and 8 per cent inflation) as a bad outcome, to 3 per cent (2 per cent yield and 1 per cent deflation) as a good outcome.
These estimates ignore the tail risks, which are also not symmetrical. A higher rate of deflation than 1 per cent per annum over a sustained period is less plausible than a higher rate of inflation than 8 per cent per annum. The real extreme risk is hyperinflation, which has no equivalent in scale of falling prices.
If bond holdings cannot be rationalised in terms of probable nominal and real returns to an investor who has less than perfect foresight, they cannot be relied on as the basis of risk management in a balanced portfolio.
Balanced management relies on diversification to smooth the return path of a portfolio. The bond percentage is the allocation that most contributes to the reduction in volatility. This is because bonds are the only asset with reliably weak correlation with equities and property. In fact, in some economic conditions, they are inversely or negatively correlated. Yet they are also the asset with the greatest unrewarded inflation risk.
The risk-reducing role of non-indexed bonds has never looked so flawed, yet the investment industry and the FSA appear not to have noticed.