Are shares really worth 3 per cent less today?

A DAY like yesterday, with the Footsie losing some 5 per cent, is one of those days when people tend to ponder questions about the valuation of equities. Are shares really worth 5 per cent less than they were a few hours earlier? Which of the many forms of valuing shares are more appropriate? Is there such a thing as a fundamental valuation?

I would not pretend to be able to answer these questions, but I have found a couple of recent papers extremely helpful in sorting out the various issues that equity valuation involves. One focuses on the UK, the other on the US.

The UK paper actually comes from Montreal, from the latest issue of The International Bank Credit Analyst. Myles Zyblock, one of the editors, concludes that the period of under-performance of UK equities is about to come to an end. Under-performance? Yes, the market has done pretty well in absolute terms but relative to the US our market has not done well during the 1990s. This is in sharp contrast to the 1980s, when UK shares did much better than US.

The argument developed here is that the great burst of share prices during the 1980s was in response to the supply-side reforms of the British economy that took place. True, similar reforms also occurred in the US, but because the UK economy started from so much lower a base, there was more scope for improvement. This was recognised by the markets, which duly marked shares up.

During the 1990s, however, it was US industry that carried out the great restructuring. There were several aspects to this: takeovers and mergers, downsizing of workforces, a surge in investment, particularly in information technology. This has improved performance, which again has been reflected in share prices.

However, there are weaknesses in the US corporate scene. Profits have not done as well as headlines suggest, as recent earnings growth has been bolstered by one-off events.

By contrast, there is more scope for increases in the earnings of UK companies. Investment is growing and the wage bill is falling. Corporate profits as a share of GDP are close to the peak of the late 1980s. Further, falling bond yields have made UK equities look cheap relative to bonds. Only on four occasions in the last 50 years have shares looked so cheap relative to gilts.

The moral of all this is that UK shares have relatively little downside risk and considerable upside potential, at least relative to US securities. Expect short-term volatility, but build modest UK positions and sell US ones.

Implicit in this view is the assumption that US shares are overvalued. That has certainly been the position of the Bank Credit Analyst team for some time. A more developed analysis of US equity market valuations comes from Sushil Wadhawani, currently at Tudor Investment Corporation, previously equity strategist at Goldman Sachs, and one of the most thought-provoking market analysts here in London.

The core ideas of his paper, "The US Stock Market and the Global Economic Crisis", to be published in next month's National Institute Economic Review, are: first, that the US equity market is indeed highly valued by historical standards; second, that there are above-average risks associated with the world economy at the moment (I suppose Brazil counts as one of those); and third, that holders of US equities are not being compensated for those risks by having low prices.

Of course none of this precludes the possibility that US shares might go higher in the short-term; nor does it mean that equities are a bad investment long-term.

Indeed one of the most fascinating aspects of Dr Wadhawani's work is the historical data that he has unearthed on just what a good investment US equities have been over a very long period. Look at the graph on the left. That shows the percentage of periods between 1802 and 1996 when stocks have outperformed bonds. As you can see, in any one year there has been 60 per cent chance that equities will do better than bonds; over any 10-year period it has been 80 per cent. And over any 30-year period it is virtually 100 per cent.

Now look at the graph on the right, which looks at the total real return on US equities over very long periods. (That is the capital gain or loss, plus or minus dividend income, adjusted for inflation and compounded.)

The astonishing thing here is that returns are extremely stable at around 7 per cent. Take the 1802-1870 period: it was exactly 7 per cent. Take 1871 to 1925: 6.6 per cent. Take 1926 to 1997: 7.2 per cent. Only if you take shorter periods do you get notably different results. If you take 1966 to 1981, returns were negative. And finally, take 1982 to 1997 and returns have been nearly 13 per cent.

The common-sense conclusion from this would, I suppose, be that the last 15 or so years have been compensating for the poor returns of the previous 15 years. Things are, so to speak, back to normal. But it also means that anyone whose memory of the market lasts only 15 years will have had what has been historically a very unusual experience: double the usual returns. One could go on and argue that the experience of the market since the end of 1997 (i.e. outside the period in the study) has further stretched share prices beyond their long-term trend.

Dr Wadhawani develops the argument much further, looking at various justifications for the present level of US share prices and testing each of these. But perhaps the big lesson of both these studies is simply that, while equities are and will remain the best place for people to hold savings over a very long period, timing entry into and exit from any share market is of overwhelming importance.

Is investment just timing, then? Perhaps. But maybe also patience.