Bulls and bears have been duking it out for centuries. There will always be those who think the market is going up and those who expect it to go down.

The current incarnation of that debate, of course, rages over whether the developed world is sliding into a recession. The bulls say such fears have been overdone, creating a buying opportunity. The bears, who are suddenly in the ascendant, point out that if things pan out as badly as they have done at times in history – and the world has gone three decades without a severe recession – then things could get much worse.

As with any interesting debate, both sides have a point. However, there is another dimension to the battle between bulls and bears which does not bring in the economy. Are we in a bull market or a bear market, and how do we know?

Conventional wisdom is that the bursting of the internet bubble ushered in a brief but savage bear market. That ended, depending on which index you look at, either in late 2002, or in 2003. All seem to agree that a new bull market started in earnest once the US and its allies had launched the invasion of Iraq in March, 2003. With the S&P 500, and the MSCI World index last year finally beating the highs they had made in 2000, any argument on this topic appeared to be at an end.

However, there is an intriguing case to be made that we are in fact still lodged in a secular bear market.

That argument has received a big boost from the events of the last six months. And the economy has nothing to do with it.

Stock market cycles are not to be confused either with economic cycles or earnings cycles (although they interact with both in interesting ways).

Instead, the governing theory of Ed Easterling, head of Crestmont Research in Dallas, is that markets follow secular bull and bear cycles. These are determined by peaks and troughs in price/earnings ratios. To avoid confusion with economic and earnings cycles, which markets tend to discount, the key price/earnings (P/E) ratios for this task are cyclically adjusted, taking the multiple of prices to average earnings over the preceding 10 years.

Viewed this way, stock markets look more expensive than if we simply look at the multiple of the latest year’s earnings.

But the argument goes further than that. According to Easterling, a secular bull market is a period of generally rising P/Es that multiply growth in earnings per share, and give investors an above-average return. A secular bear market is the opposite: a period of falling P/Es that offset earnings growth and provide below average returns.

These market cycles are longer than economic cycles, and very much longer than corporate profit cycles. By his estimation, we are now in the fifth secular bear cycle since 1901. This uses the (rather limited) Dow Jones Industrial Average, for which constant historical data is available, but it is unlikely that findings for the more robust S&P 500 would be significantly different.

We may have further to fall in this episode than in earlier bear cycles. Previous bear cycles started in 1901, with a P/E (as cyclically adjusted) of 23; in 1929, with a P/E of 28; in 1937, with a P/E of 19; and in 1966, with a P/E of 21. In all bar one, multiples had fallen by more than half by the time the cycle came to an end.

The current bear cycle began in 2000, with a P/E of 42. So on this analysis, we started this decade at historically irrational valuations. Multiples need to get back at least into the teens before a secular bull market can start (and we are currently at a P/E of about 26).

It is also easy to be tricked by a succession of positive years – at one point there were three “up” years in a row during the 1966-1981 bear market, or by new highs, as happened last year.

There was a similar event in 1972, when the P/E multiple hit 18, and the Dow hit a new high. Volatility was very low – exactly as was the case during the first half of last year – and, as it turned out, the market was primed for a new, savage round of sell-offs.

If there is an external driver for these cycles (beyond investors’ animal spirits), it is inflation. Higher inflation will require investors to demand lower earnings multiples. Japanese-style deflation, where reducing prices put long-lasting downward pressure on profits, also pushes down multiples.

The economy is certainly not unimportant, as a slowdown in activity will make it harder for companies to make profits. But this analysis suggests that the current great preoccuption with the risks of a recession is misplaced – at least from the perspective of investment in stocks.

According to Easterling’s calculations, the average annual return on the Dow during bear cycles was minus 4.2 per cent during the 20th century, while the average during bull cycles was 14.6 per cent. Meanwhile, GDP growth was actually slightly higher during bear cycles (6.9 per cent) than bull cycles (6.3 per cent).

Rather, investors should be more concerned with the debate over inflation. Some fear stagflation, others see a risk of Japan-style deflation, but break-evens in index-linked bond markets suggest the market still broadly believes these risks are under control. That suggests the recent sell-off could be overdone.

Markets should also worry about earnings. The profit cycle is well established, and if corporate profits were to decline this year that would be quite in line with historical experience.

Sell-side analysts’ estimates still imply a strong rebound for earnings by the end of the year.

Finally, there should be a realisation that the bizarre event of this decade is not the recent return of volatility. Rather, it was the recovery that started in 2003.

Possibly due to the flood of cheap money from the Federal Reserve, this arrested the fall in stock prices at a point when multiples had still not got down to the levels normally needed before a strong recovery can start.