NO ONE pretends the antics of the foreign exchange markets are based on anything more than a very short term view of the world - next week's vote, next month's interest rate prospects. But as recession deepens a growing number of economists are trying to work out whether we are actually at the mercy of a pattern of deeper, almost subterranean, forces that unavoidably shape economies over regular 10, 20, or even 50 year cycles.

Whether or not this is anything more than an economist's equivalent of astrology, there are plenty of theories to choose from. The grandfather of them all is Kondratiev's Long Wave theory which says that the world economy goes through a regular boom-and-bust cycle every 50 to 60 years - roughly 25 years up and 25 years down.

The upswing is fuelled mainly by cheap raw materials and is brought to an end by high prices and high interest rates. On the downswing there is high unemployment and increasing technological obsolescence in industry which is not corrected until new technology and production patterns are introduced during the next upswing.

N D Kondratiev was a Marxist working in Russia in the Twenties. Perhaps because his theory did not predict the end of capitalism, Stalin sent him to Siberia. Economists who give any credence to the theory generally reckon that the Wave last peaked in about 1975. This means we are well into the downswing which is unlikely to end until about the year 2000. Events like the Eighties boom are merely temporary aberrations.

Kondratiev supporters claim that economic growth rates in industrialised economies this century fit snugly into the theory. Starting in about 1900, industrialised economies grew fast for the first 25 years. In the Twenties growth rates turned down through the Depression, hit bottom in the Forties, then accelerated again until the mid Seventies.

Most currency and stock market dealers regard such theorising as hocus-pocus. Many economists are also suspicious of the theory because the evidence is unclear and there has barely been time for two complete Long Waves on reliable data - hardly indisputable proof.

There are, in any case, plenty of shorter term and more easily provable theories to choose from, such as the Kitchin (14 months), Jugler (11 years) and Kuznets (20 years) cycles. Economists have often tried to fit several cycles together to explain why you can have shorter boom-and-bust periods within the longer Kondratiev Waves.

According to some economists, the current recession is particularly bad because we are simultaneously in the downswing of a 20-year Kuznets cycle and a Kondratiev Wave - both due to hit bottom at the end of the century. (In the Eighties things were not so bad because although Kondratiev was on the way down, it was partially cancelled by the upswing in Kuznets.)

Even if such theories are not complete drivel, it is unclear what use they are. Neither markets nor governments have ever based their actions on them. If, however, it could be proved that we are in the grip of inescapable economic patterns, we would at least know what we were in for.