The UK consumer is back. Official figures next Thursday are expected to show that retail sales volumes have risen significantly in recent months; in August they were 3.3 per cent up from March.
With real wages falling, it’s tempting to attribute this to a bout of irrational exuberance among consumers. Such an inference would probably be mistaken.
To see why, consider the ratio of retail sales to the All-Share index. This should be more or less stationary over time. If it were not, we’d end up either with huge retail sales and worthless equities or with a highly valuable stock market and little spending, both of which are implausible. This in turn implies that the ratio should mean-revert: when it is high it should subsequently fall, and when it is low it should rise. And this is the case. Since 1996 (when the ONS’s current data begins) the correlation between the ratio and subsequent annual changes in it has been minus 0.49.
But how does it mean-revert? If consumers were irrational, it’d be by retail sales falling as shoppers reined in their spending after earlier profligacy.
This is the case, but to only a little extent. The correlation between the ratio of retail sales to the All-Share index and subsequent annual changes in retail sales is just minus 0.15. That’s statistically significant, but weak.
Instead, what happens to a much greater extent is that the ratio mean-reverts by share prices correcting themselves. The correlation between the ratio and subsequent annual changes in the All-Share index is 0.47. This means the ratio has done almost as good a job of predicting annual equity returns as the dividend yield.
For example, high ratios of retail sales to share prices in 2003, 2009 and (to a lesser extent) early 2016 led to shares rising, and low ratios in 2000 and 2007 led to shares falling.
This implies that consumer spending is to some extent forward-looking. High spending is a sign of good times to come – which is why share prices rise – and low spending a harbinger of bad times.
You might object here that what’s really going on here is in fact a story about the stock market. Shares tend to overreact on the upside and downside. This means the ratio of share prices to any stable upward trending variable will be able to predict subsequent price changes – even a simple time trend.
Even if we accept this, however, it still leaves us with the inference that consumers, in aggregate, are more rational than share prices.
So, does this perspective tell us that consumers are optimistic now?
No. The ratio of retail sales to All-Share index is still below its post-1996 average – it’s just slightly less so than it was in March. This tells us that consumers are moderately pessimistic. And this in turn is a warning for equity investors. If the post-1996 relationship between the ratio of retail sales to All-Share index continues to hold, there’s a two-fifths chance of share prices falling over the next 12 months.
