Inflation matters for equities, but not in the way you might think – and it is telling us to be wary of the market now.
First, though, let’s consider the way in which inflation does not matter, or at least has not done so in recent years.
It hasn’t mattered in the sense that there’s been no significant link between annual consumer prices index (CPI) inflation and annual equity returns. Since 1988 (when current CPI data began) there has been zero correlation between annual inflation and total real returns on the All-Share index. High inflation has been as likely to be good for shares as bad. Recent months continue this pattern. Inflation was 4.2 per cent in the year to October – it’s highest rate for almost 10 years – but the All-Share index posted a real return of over 25 per cent in this time.
Inflation, then, is unrelated to equity returns. One reason for this is that it is sometimes the result of strong demand and economic recovery. And this benefits cyclical stocks sufficiently to offset the adverse impact of higher interest rate expectations.
Another reason is that high inflation is often only temporary, the result of higher commodity prices or short-lived mismatches between the patterns of supply and demand. When investors believe this is the case – that is, when inflation expectations don’t change much – they’ll infer that current inflation tells them little about the longer-term outlook for the economy and so won’t change their long-term earnings forecasts. Which means that share prices won’t change much.
There is, however, a caveat here. All this has been true only since the late 1980s, which has been mostly a time of low and stable inflation. It’s likely that the high inflation we saw in the 1970s would indeed hurt shares – as it did then. But this is not what markets expect. They are pricing in retail price inflation falling from its current 6 per cent to around 4 per cent by 2024 – which implies CPI inflation being below 3 per cent then. It’s no wonder, therefore, that equities have held up in recent months as inflation has risen.
There is, though, a different sense in which inflation matters. The level of consumer prices, relative to the level of share prices, predicts returns.
My chart shows the point. When consumer prices are low relative to the All-Share index – as they were in 2000, 2006 and 2017 – equities subsequently fall. And when consumer prices are high relative to share prices, as they were in the mid-90s, 2003 and 2009, shares subsequently do well.
The simple ratio of consumer prices to the All-Share index on its own can explain half the subsequent variation in three-year real returns on equities. In this sense, consumer prices matter a lot for equities.
The story here, however, has less to do with consumer prices than it has with the behaviour of investors. Sometimes, they overreact to good or bad news and drive share prices up too much (as in 2000) or down too far (as in 2009). On these occasions the ratio of the All-Share index to almost anything that has a stableish upward trend will predict subsequent returns. This something can be dividends: the dividend yield has also been a good predictor of medium-term returns. But it can also be the consumer price index or simply a time trend. Whatever, the CPI can predict returns.
Which brings us to two nasty problems.
Problem one is that the ratio of the CPI to All-Share index predicts returns so well only because real share prices have had no upward trend: in statistical jargon they have been stationary.The All-Share index relative to consumer prices is no higher how than it was in 1998. To put this another way, an investor in UK equities who had spent his dividends would be no better off today than he was 23 years ago. All the returns on equities we have seen since the late 90s have come from dividends, not from capital gains. In this important sense, shares have gone ex-growth.
Problem two is that, despite this, real share prices are now slightly above their average of the last 20 years: the CPI/All-Share ratio is quite low. This is because while consumer prices have risen a lot in the last 12 months share prices have risen even more. Which predicts lacklustre returns in coming months. If the post-1988 relationship continues to hold, total real returns will be less than 6 per cent over the next three years, or less than 2 per cent a year. And all this will come from dividend income, implying that the All-Share index will fall in real terms.
Of course, there is an error margin around this prediction. But the point is that the likeliest outcome is for modest returns.
All the talk of inflation is, therefore, distracting us from a bigger problem – that equities are somewhat overvalued.
