And they do, to some extent. Take gilts first. The ratio of consumer spending to the All-Share index, along with the dividend yield on the latter, has explained two-fifths of the variation in three-year returns on gilts since 1986. A high ratio of spending to share prices predicts low returns on gilts, but high returns on equities. This is because consumer spending is forward-looking: we spend more when we anticipate good times. And while a strong economy is good for shares, it's bad for gilts.
In the same way, a high ratio of spending to share prices also points to small and mid-caps outperforming the FTSE 100. This is because these are a little more cyclical than mega-caps, and so should outperform when the economy does well.
Which poses the question: do ratios of consumer spending to wealth really predict economic growth? Yes - to some extent. Since 1986, two-fifths of the variation in three-yearly real GDP growth has been explained (in the statistical sense) by the All-Share dividend yield and consumer-wealth ratios, lagged three years. High ratios of spending to share and house prices lead to stronger GDP growth.
Such ratios, however, did not predict the 2008-09 recession: yes, they told us that growth would slow – but not that it would turn negative. This suggests that recessions are unpredictable not only by professional forecasters but also by the wisdom of the crowd.
So, what are consumption-wealth ratios predicting now? At 3.2 per cent, the dividend yield is a little below its post-1986 average (of 3.55 per cent). So, too, is the ratio of spending to house prices. Both point to slightly below-average GDP growth and equity returns. And the low ratio of spending to house prices also points to below-average returns on the FTSE 250 and small-caps relative to the FTSE 100, as well as to low returns on high-yield stocks relative to low-yield ones. However, these signals are mitigated by the fact that the ratio of spending to share prices is slightly above average, which is good news for equities, economic growth and cyclical stocks.
On balance, all this tells us that if past relationships continue to hold, real GDP will grow by 1.8 per cent a year over the next three years, which is slightly below average. More accurately - but less precisely - it tells us there's a two-thirds chance of annualised growth of between 0.8 per cent and 2.8 per cent. Such growth is enough to generate reasonable returns on equities - a two-thirds chance of annualised returns between 3.3 per cent and 13.5 per cent - but not so strong as to scare the gilt market; past relationships predict a two-thirds chance of annualised returns between 2.6 per cent and 8.3 per cent. The ratios also point to the FTSE 350 high-yield index doing about as well as the low-yield index, and to mid-caps slightly outperforming the FTSE 100 - but the margin of error here is such as to imply a high chance of underperformance.
You might think all this is rather mundane, especially considering the error margins around the forecasts. But this is only to be expected; you don't need to believe that returns are normally distributed to see that they must be around average on many occasions, and the next three years looks like being one of these times.
Nevertheless, there's an implication here. All this implies that our asset allocation - not just between equites and bonds, but also between different types of equity - should be around normal. Of course, there's a chance of something extraordinary happening. But insofar as markets are predictable, this chance is not especially high so we shouldn't bet very much on it.
