FTAdviser.com

Advertising

The idea that markets move in long cycles has a lot of instinctive appeal. Over a quarter of a century in financial journalism, I have lost count of the number of times I have seen charts that “prove” markets have patterns over four, 14 or 56 years.

It always reminds me of those religious thinkers who obsessively analysed the bible in an attempt to predict the end of the world. There are undoubtedly economic cycles. But a cursory reading of recent history shows they are not predictable. The last recession happened in the early 1990s. When it ended, no one would have forecasted such a long wait for the next one.

Perhaps periodicity was more predictable when agriculture was the dominant economic sector. Poor harvests may have resulted from over-farming or from cycles of sunspot activity. But is there evidence for longer cycles such as the Kondratiev wave? The problem is that modern economic history really started with the Industrial Revolution. So there has not been enough time to make any long cycle patterns the result of anything more than chance. Kondratiev pointed to downturns in the 1820s, 1870s and early 1930s. But those who waited for an equally big crash in the 1980s were disappointed.

There are signs of supercycles in commodity prices and this may actually make sense. It takes time and a lot of capital expenditure to search for mineral deposits and then to dig mines, build oil rigs or whatever. Entrepreneurs need an extended period of high prices before they will commit to such an investment. These days they also face a lot of local opposition to new developments; who wants to live next to a refinery? At the same time, the elasticity of demand is quite low. It takes a big rise in petrol prices for us to give up our cars or even to decide to take the bus, rather than the Volvo, to Sainsbury’s.

Chris Watling of Longview Economics reckons that there have been 11 super-cycles in commodity prices since 1788. The average length of the bull markets has been 20 years, but the shortest cycle was the most recent, at 12.3 years. So although some people talk of high prices until 2020, the current cycle could end by 2013. And, as the recent fall in commodity prices has shown (July saw the worst monthly decline since 1980), these cycles can be very volatile. As for the scale of the increase, the gain in commodity prices as of end June was 129 per cent; previous bull runs have ranged from 135 per cent to 689 per cent.

What about equities? It seems less likely that stockmarkets would display a regular rhythm. After all, the nature of the businesses that comprise the market change over time; from railroads in 1900 through industrials in the 1930s to technology and financials in the 1990s and 2000s. These industries will have their own cycles.

Russell Napier, in his book Anatomy of the Bear, cites four great buying opportunities on Wall Street; August 1921, July 1932, June 1949 and August 1982. In other words, the cycles have varied from 11 to 33 years. That is not of much practical use. If we look at the UK market, the Barclays Capital equity-gilt study shows a peak in 1906 that was not surpassed, in real terms, until 1959, followed by another peak in 1968 that was not surpassed until 1993. The UK market is still below its 1999 level in both nominal and inflation-adjusted terms. Again, there is no obvious pattern.

Stock markets may well reach a point this year or next when it makes sense to buy aggressively. But it will have nothing to do with the patterns on a chart. It will be because they at last offer genuine value.

Philip Coggan is Buttonwood columnist for the Economist