Hamish McRae: The bears kept their claws sheathed in August but there's still September

Maybe August won't be the wicked month after all; maybe it will be September – or even October.

Maybe August won't be the wicked month after all; maybe it will be September – or even October.

During these dog days of August the world's financial markets have doggedly been regrouping after the mayhem of July. The US market in particular has been buoyed by the growing view that the States will do well by comparison to the eurozone in the coming months. This has also helped push the dollar back up against the euro.

So the great bear market is over then? Of course no one can know until long after the event but in the last few days just enough cheer has crept into the markets for some of the professionals to risk suggesting that this might be so. It is certainly true that bear markets rarely last longer than two years. This one, depending on which index and which market you take, has already been nearly, or a little over, two and a half years long.

But before you get too excited – or relieved – I thought it might be helpful to put the case for caution; or rather, two of the less common such cases, both of which focus on possible problems over the next couple of months.

Most of the comment on the markets, here as elsewhere, has been about the level of the market when measured by conventional valuations: the value of company price/earnings ratios, for example, or the market's bond/equity yield ratio. But while these can tell you whether shares are expensive or cheap by historical standards and so are valuable on a five or 10-year time horizon, they are less use in the short-term.

So here are two ways of looking at the next few months. One is to look at the different months of the calendar and see how shares perform. Most people are aware of the old adage "sell in May and go away" and more recently Goldman Sachs has drawn attention to the January effect: the way shares tend to go up in the first month of each year. But I am grateful to Don Straszheim, of Straszheim Global Advisors, for pointing out that September is measurably the worst month of the year for shares, while November, December and January are the best ones. You can see the results of his research in the table. This shows what has happened on average each month to the Dow and the S&P 500 since 1950, and to the Nasdaq since 1971.

As you can see, both the "sell in May" and the year-end surge rules are generally correct. So one should indeed sell in May (or even April) and buy back in October. And it is best to be out of the market in September. Were you to follow those rules you would on average gain three or four percentage points a year, even allowing for some transaction costs, on the index. Over 10 or 20 years this is huge.

So why don't more investors follow these rules? None of the obvious explanations stack up – shares being sold for tax purposes for example – and Mr Straszheim wonders whether we just lack imagination. But the record does cast a certain cloud over the currently more sunny mood. There have been four "bottoms" to the US market already, each lower than the previous one (left-hand graph). The first three were clearly false bottoms; there is no reason why the fourth should not prove a false one too, though if it were, that would be pretty unusual.

The other reason for caution is the view of the chartists. The chartists, or "technical" analysts as they prefer to be called, rely on the patterns that charts make to predict where markets might go. To anyone with a scientific background this form of analysis is not technical at all, more black box with its "double bottoms", "neckties" and "head and shoulders" patterns. But the charts do seem to capture human behaviour and as such provide a helpful guide to short-term price movements.

So what are the chartists saying? Here I am grateful to the HSBC technical team. Their view, looking at world equities is that the present rally should last until the end of the month but then it will be time to sell again.

"This is not the final low," they write. "Expect new lows in late October."

This goes for the UK market as well as the US one, though the UK market has been improving relative to the world since the middle of 2000. We have gone down but not as much as the rest of the world. Not fully understanding quite how these people reach their conclusions, I can't fully explain why they are so certain. But it sort of makes sense. Intuitively it feels as though there has not yet been the bleak despair, the "capitulation", that typically characterises the bottom of a bear market.

True, most of these go-go New York analysts who puffed the telecoms shares during the bubble years have now been rusticated. Most of the book-cookers of large US corporations have been nabbed. And in Europe, the heads of the biggest borrowers, Vivendi and Deutsche Telekom, have duly rolled. But most people seem to feel that there is something big still out there.

As far as the UK is concerned, the HSBC technical team feels that the Footsie could reach 4,950 by the end of August. (There is a small reverse head-and-shoulders pattern, see right-hand graph.) Then the market faces a severe fall to 3,475 by late October, when the real bargains arrive. Then everything gets better.

So there. What, standing back a little from all this, should the rational investor deduce?

The first thing, I suggest, is to note the tone of all this. This is a mature bear market; that is accepted. All the debate, therefore, is about how this market compares with previous examples of the breed: how long, how deep, what are the characteristics of the turning point and so on.

The second is that few people, even in the US, now expect a rapid recovery either in share prices or in the world economy. In continental Europe the mood, if anything, is worse because growth prospects appear worse. European consumers seem to be spooked by the introduction of the euro and have been cutting back their consumption, particularly in Germany. So Europe is dependent on external demand, in practice US demand, to crank up the economy.

Third, the looming threat of a Middle East war will depress both markets and the world economy (the latter through higher oil prices) through the autumn.

But if you stand back, all three conclusions are helpful, in that they reflect depressed expectations. The less we expect, the less chance there is that we will be disappointed. The prime problem of the past two years is the repeated, relentless string of disappointments. It is only when the last vestiges of optimism have been battered out of people that they can begun to look up instead of down. The job of markets at a time like this is to look forward though gloomy valleys and glimpse the sunlit slopes beyond. We just can't see them yet but that does not mean they aren't there.