Longer-dated index-linked gilts now yield less than minus 2 per cent. This means that if you hold a 20-year one to maturity you will lose one-third of your money in real terms. Which poses the question: why would anyone want to hold them?
Insurance, that’s why. Not all assets should have positive returns. You wouldn’t expect to make money on your home insurance, so why should you do so on other things that insure you against bad times? And index-linked gilts do just this.
You might think the bad times they offer insurance against are falls in share prices. Not quite. Since 1998, the correlation between monthly returns on the All-Share index and monthly returns on long-dated linkers has been zero. This means linkers are a nice diversifier against equity risk, but not insurance: they are as likely to lose as to gain when equities fall.
Linkers do, however, insure us against some particular falls in equities – those associated with higher inflation.
Rising inflation is often accompanied by falls in equity valuations: since 1989 the correlation between consumer price index (CPI) inflation and the dividend yield on the All-Share index has been a hefty 0.65. Rises in inflation in 1989, 2002-03, 2007 and 2016-17 all saw shares fall and the dividend yield rise.
Unexpected inflation, therefore, is doubly nasty. Not only does it reduce our purchasing power as customers, but it also reduces our wealth as equity investors.
If you are working or have an index-linked pension this pain is mitigated by inflation-linked rises in your income. If you don’t have these, however, unexpected rises in inflation are horrible. And index-linked gilts offer insurance against such rises.
For some investors, though, there is another form of insurance: time diversification. History tells us that the dividend yield on the All-Share index predicts returns, and the longer the time horizon the more strongly it does so. This means that if the dividend yield does rise because of inflation (or anything else) we can buy at the higher yield and so earn higher subsequent returns to compensate for earlier losses.
But this strategy is not perfect, even if you have the discipline and the time horizon to implement it. There is an unquantifiable risk that history will not repeat itself and that a higher dividend will lead not to higher returns but even higher yields and falling prices. This would be the case if a rise in inflation leads to a further rise.
But how likely is this? Those of you whose formative years saw the high and volatile inflation of the 1970s can be forgiven for thinking it a significant danger. Recent history suggests otherwise. Since the early 1990s CPI inflation has been remarkably stable in the face of huge swings in commodity prices, sterling and economic activity. Since the UK left the exchange rate mechanism in 1992 and adopted inflation targeting, CPI inflation has averaged 2.1 per cent a year with a standard deviation of just one percentage point. This implies only a one in six chance of inflation rising above 3.1 per cent.
And, in theory, inflation should be stable. It depends very much on people’s expectations. If companies expect prices to rise they’ll put up prices in anticipation of costs rising and their rivals raising prices, which means that inflation will rise. Conversely, if they expect inflation to be low and stable, they’ll hold prices down. This means that inflation is like Maynard Keynes’ view of interest rates – a highly conventional and psychological phenomenon; any rate accepted with sufficient conviction as likely to be durable will be durable. With inflation having been low and stable for so long, therefore, we have reason to believe that inflation expectations will stay reasonably low and hence that actual inflation will be too.
But, of course, we could have said exactly the same thing in the 1960s, just before inflation took off. Could history repeat itself?
Granted, temporary rises in inflation are always possible because of rising commodity prices or a fall in sterling. But sustained and serious inflation requires that wage growth takes off. And we’ve recently seen a sign of this – latest figures show that it has hit an 11-year high. Few economists, however, expect this to continue. Arno Hantzsche at the National Institute Of Economic and Social Research (NIESR) foresees it falling back later this year as the economy softens. And many economists agree with the Bank of England’s Andy Haldane that, thanks to workers’ lack of bargaining power, wage growth around the world is no longer as responsive as it once was to low unemployment.
All this suggests that while there might be a case for some investors to have some inflation insurance – such as those who are retired without an index-linked pension – there is little need for very much of it.
