Money: To be - or not to be - in equities?

`You thought this was a raging bull market? You are right. It is looking as highly valued now than it has done in almost any previous period'

Each year, when BZW publishes its annual gilts/equity study, it makes the long-term case for investing in equities just about as plainly as it can be done.

Over the period since 1991, it quantifies how consistently shares have outperformed the other main asset classes, gilts and cash.

The long-run annual return of 8.11 per cent in real terms comfortably exceeds the 1.87 per cent available on gilts and the 1.48 per cent annual return on cash. In both cases, over any four-year period, history suggests that there is more than a three-in-four chance that the stock market will produce a higher return than either cash or gilts.

Long-term investors, so the headlines report, are more than amply compensated for the extra risk they take on in the stock market.

So much is well known. But what makes the study most interesting, to my mind, are the secondary details that most often get overlooked when the results are reported each January.

First are BZW's observations about the overall level of the stock market. Drawing on a number of different valuation techniques, including the pioneering approach of the American financial economist Robert Shiller, BZW calculates a "fair value" band for where share prices should be by historic standards, and then compares this with the actual level of the stock market.

For at least five years now, the BZW study has been flagging up the fact that share prices today are way above their historic "fair value" range.

You thought this was a raging bull market? Well, you are right. In fact, as the chart shows, it is looking as highly valued now - after yet another strong year last year - than it has done in almost any previous period, including the so-called golden age of the 1950s.

According to BZW, the current bull market phase, from 1982 to the present, has not only produced consistently higher returns (an average of over 12 per cent a year in real terms, more than 50 per cent above the long- run average) but done so with lower risk than in the past (measuring risk by the standard academic measure of volatility of returns).

Michael Hughes, head of economics at BZW, quite rightly observes that just because a stock market is overhauled on a long-term basis does not mean it is going to stop going up next year. It takes time for bull and bear market phases to work through their cycles.

It is the direction, not the absolute level of the market, which matters most. It so happens that the prevailing conditions since the early 1980s - falling real interest rates, steady economic growth and declining inflationary expectations - have been unduly favourable not just to shares but to almost all types of investment, including gilts.

Real returns on gilts have also been higher since the early 1980s than in any previous period, with the single exception of the between-war years. That, of course, was a period when inflation was negative for part of the time (which greatly favoured any fixed-interest investment such as gilts).

It also pre-dated the start of the great "cult of the equity", the post- war realisation by pension funds and other investment institutions that, in an inflationary world, equities were not only a - but the most - suitable investment for organisations with long-term liabilities to meet.

The old stock market saying, "Let the trend be your friend," is therefore still very much in force. It is perfectly consistent to say that the stock market is overvalued and cannot continue this way for ever, while at the same time acknowledging that it may indeed go on doing well until there is a fundamental change in the direction of interest rates, the economy or inflation expectations. It may well be (as I am inclined to suspect) that we are still some way away from the absolute bottom of the long-term interest rate cycle, notwithstanding the likely increase in short-term UK rates next year.

The BZW study is just as interesting - but less convincing - when it moves on to analysing risk and optimal portfolio mixes. If you accept that standard deviation is a good measure of risk, then the case against gilts and in favour of equities is clear-cut. The historical data shows that the stock market has generally been one-and-a-half times more volatile than gilts, and three- and-a-half times more volatile than the returns from cash. As the returns from shares are anywhere between three and six times as great as those on gilts, the case for filling your portfolio with equities is overwhelming in a statistical sense - the extra risk is amply compensated.

But this is where I think the BZW analysis starts to go off the rails, on the standard "garbage in, garbage out" principle. Standard deviation is one measure of risk, but it is not one that accurately captures the full dimensions of what risk means to most real-world investors. In practice, the risk of losing money, which is what many investors ultimately care about, from gilts during the recent phase in the level of long-term interest rates has been very small indeed.