The case for doing so is simple. In recent years there's been a strong correlation between annual changes in manufacturing output (a good cyclical indicator) and annual returns on the All-Share index - of 0.61 since January 1996, with one percentage point above-average growth associated with 2.9 percentage point above-average returns. There's also been a negative correlation between output growth and annual gilt returns of minus 0.17 during this time. Stronger growth, then, is more often than not good for shares and bad for bonds.
One reason for this could be that investors are prone to a projection bias. They fail to foresee that their future tastes will change, so they don't anticipate that a stronger economy will increase their appetite for risky assets. As a result, share prices do not rise fully in anticipation of an economic upturn, but instead rise while the economy strengthens.
The prospect of recovery, then, means we should shift from safer assets into shares.
Or should we? There are good reasons not to do so. For one thing, economic activity is inherently unpredictable. Of course, we can say that growth is more likely than not. But this is simply because history tells us this much. In betting on growth, we are like the poker player playing the probabilities - and she's not always right.
However, even if we get growth, we might not get rising share prices. The strong link between economic growth and equity returns is a recent phenomenon. In the 10 years from 1985 to 1995 the correlation between manufacturing output and All-Share returns was actually slightly negative. For example, shares fell in 1994 even though the economy did well.
One reason for this is that growth is ambiguous for shares. Sure, it increases our appetite for risk, but on the other hand it also increases the likelihood of tighter monetary policy, which is bad for shares.
This warns us that the correlation between growth and equities might break down. A shift into shares is a bet this won't happen. Again, it's a bet.
Let's put these problems aside, though, and assume that share prices do rise as the economy grows. Would this be a reason to buy?
Not necessarily. Higher prices - and especially higher prices relative to dividends - mean lower equity returns. And the longer your time horizon (within reason), the truer this is. For example, since 1986 the correlation between dividend yields and All-Share returns in the following month has been 0.15. For the following 12 months it has been 0.49. And for the following three years it has been 0.71. This tells us that if you wait and see whether the economic recovery will lift share prices, you might end up simply buying before the market falls.
These lower expected returns are merely the counterpart to a greater appetite for risk. The more we feel able to take on risk, the less is the compensation we require for doing so. Expected returns therefore naturally fall as risk appetite rises.
And herein lies a reason why many of us need never change our asset allocation. Quite simply, changes in expected returns and risk appetite cancel out for the average investor - and someone must be the average investor.
For example, in bad times we don't feel like taking on risk, but expected returns then are often high. And in good times we're happier to take risk, but share prices are higher and expected returns are lower. On balance, it's not obvious that the average investor should change his asset allocation in either case.
The case for changing asset allocation in response to macroeconomic developments is, therefore, not at all clear-cut.
This doesn't mean you shouldn't review it from time to time. It's just that changes in it are more justified by changes in your personal circumstances - for example, if you retire, change job or get married. The main thing is to have an asset allocation that you are comfortable with. Chasing the market or trying to predict the future are both probably fruitless pursuits.
