I say so for a simple reason. During this time, there's been a strong link between annual changes in manufacturing output and annual changes in the All-Share index. Since 1996, the correlation between the two has been 0.58, which is big by economic standards. And each percentage point fluctuation in growth has been associated with a 2.7 percentage point variation in equity returns. As the average annual price rise on the All-share has been only 2.9 per cent in this period, this is a strong link.
Granted, some of this correlation reflects the fact that both are correlated with global economic conditions. Even controlling for changes in US output, however, there's still a statistically significant link between All-Share returns and UK output.
This link doesn't exist because UK-based manufacturing firms make up a significant portion of the All-Share index. They don't. It's because manufacturing activity is an indicator of cyclical economic conditions generally, and these are correlated with the All-Share index for three reasons:
■ An economic upturn increases appetite for risk, while a downturn depresses it.
■ Economic fluctuations are associated with variations in corporate earnings, which are not fully anticipated in advance.
■ The same things that affect manufacturing output also affect share prices. Both fell in 2008-09, for example, not because one caused the other but because both were clobbered by the banking crisis.
Correlation between the All-Share index and manufacturing output
Our story, then, is clear: a slowdown will hurt shares.
No, it's not. There’s a puzzle here. The strong link between manufacturing output and equity returns has only existed since the mid-1990s. Before then, the correlation was insignificant and often zero - despite the fact that manufacturing was a bigger part of the economy and All-Share index back then. And again, there are good reasons why there might be a low or negative correlation between the two.
One is simply that if economic forecasters and equity analysts are doing their job (a big if) then variations in output and profits should be discounted in advance. This should generate a zero contemporaneous correlation between output and equity returns except to the extent that there's a certainty effect: we value a certain rise in dividends by proportionately more than the 90 per cent chance of the same.
What's more, news about output should also be news about monetary policy. Stronger-than-expected output might increase expectations about interest rates, and fears of tighter monetary policy should depress share prices.
And then there's the exchange rate. If this does its job (another if) it should act as a shock absorber for the domestic economy and so fall as manufacturing falters. This should boost share prices insofar as it raises the sterling value of overseas earnings (and vice versa if output strengthens). This too should cause output and equity returns to move in opposite directions.
A positive correlation between output and returns is, therefore, by no means certain. In theory, the correlation might be positive or negative.
This matters, because in the past few years the correlation between the two has fallen to almost zero: this is true whether we look at UK or US manufacturing. For example, equities did well last year even though manufacturing stagnated for much of the time.
There's a practical message here and a more general one.
The practical one is that a UK slowdown (assuming it happens) need not be bad for the market as a whole, although it would hit some stocks hard, not least because slower growth is usually accompanied by a greater dispersion of corporate performance and hence more profit warnings.
The more general one is that there are very few reliable and robust laws in the social sciences: investing is merely an applied social science. Instead, as Jon Elster pointed out, all we have is a box of mechanisms, some of which work in some direction and some in the other and it is often only with hindsight that we can tell which ones have been most important. This is one reason why economic forecasting is often such a fruitless pursuit.
