This could be a once-in-a-lifetime turning point for the dollar

If the downward trend has been reversed, those levels of two years ago will never be reached again. Could this be right?

It is very difficult to catch a cyclical turning point until after the event. Thus you have to wait at least a year to be clear that, for example, a recession is over, a currency has reversed its path, or that the next movement in interest rates will be up rather than down.

But if it is hard enough to catch a cyclical turning point, how much harder it is to catch a long-term, secular one. Here you are not dealing with a change that will take place every three to seven years, but rather one that takes place every 50 or 100 years. Anyone who claims to have identified such a turning point needs to hang the large public health warning: "This idea may be very wrong."

Apply this to the notion that it is not just a cyclical change in the value of the dollar which has taken place in the last two years but also a secular change.

Since the middle 1950s the dollar has been a weak currency. The weakness was not fully revealed until the Bretton Woods system of fixed exchange rates began to collapse at the end of the 1960s (the trigger was the devaluation of the other reserve currency, sterling, in 1967). But once post-War Europe began to recover and the dollar shortage of the 1940s was replaced with a dollar surfeit, it gradually became clear that sooner or later the dollar would start a long decline. The principal beneficiaries were, of course, the mark and the yen.

Now both the mark and the yen are weak for cyclical reasons, but if the long-term trend of the dollar is intact then in another five or so years the dollar will be down in the dumps and at the bottom of that cycle it will be lower than it was two years ago.

If, on the other hand, the downward trend has reversed, those levels of two years ago will never be reached again. The dollar will never - or at least not in our lifetime - be back at 80.

Could this be right? Here are five reasons why it might, and one why it might not.

First, we are moving into a world of zero inflation, or as near zero as makes no difference. In the past one of the main reasons for currency adjustments has been differential inflation. In a world of high inflation some countries will inflate at, say, 10 per cent and others at 5 per cent, so that there will need to be adjustments every couple of years to keep them in line. But in a world of near-zero inflation you might have one country inflating at 1 per cent and another at 2 per cent. Maybe, once every 15-20 years you might need to have a change in parity, but it would not need to be very often, and it might well be that during this period the underlying performance of the higher-inflation country would improve enough not to need a currency adjustment at all.

Result: even if the US does have slightly faster inflation than Germany and Japan (as it does at the moment) the dollar may still be able to rise. It is not far from its average position in recent years (see graphs) so even after this recent rise there is still some leeway for it to move further. In fact it is probably still a bit undervalued.

Two, the fiscal position of the US, much criticised through the 1980s for the excessive deficit, is now under better control than the deficits of either Japan or Germany. True, the US retains the problem of low savings, but if you allow for the substantial private sector pensions of the US and the lack of these in Germany and Japan, the position is much more balanced. Looking ahead, in five years' time the US fiscal position may appear very favourable compared with that of almost all other developed countries.

Three, while the US population is ageing it is doing so at a slower rate than any of the Group of Seven industrial countries. Meanwhile, Japan and Germany are ageing fastest. The demographic pattern of a country has a profound impact on its economic vibrancy, quite apart from its effect on savings and investment.

Four, there are signs that the endemic current account surplus of Japan and the endemic deficit of the US may be about to reverse themselves. (The long-term surplus of Germany has disappeared thanks to the costs of unification, and the country may well stay in deficit for another generation until the reconstruction of the former East Germany is complete.) Evidence for a move by the US back into surplus is, admittedly, pretty thin at the moment, but there are some signs that Japan may be moving back towards deficit. BZW examined the impact of a disappearance of the Japanese current account surplus in its latest monthly economic digest, noting that the peak in the surplus as a percentage of GDP was reached as far back as 1986.

Five, the relative decline in the importance of the Japanese economy and the possible disappearance of the mark as a currency if European monetary union goes ahead, leave the dollar clear as the only reliable global currency. The euro might eventually become a reliable currency, but it is unlikely to be trusted by global savers until is establishes a track record. Were EMU to go ahead on a wide basis, then most investors would expect it to be weak.

There is one powerful counter-argument. It is the cultural one: that the low-savings culture is so embedded in the US, and the high-savings culture ultimately so secure in both Japan and Germany, that the value of the yen and the mark will remain solid. Provided the mark still exists, there will certainly be people who will want to hold it, and Japan's enormous net asset position will underpin the value of the yen. The US net deficit position, meanwhile, will continue to undermine the dollar. The US remains the world's largest debtor, and until there is some indication of a turning point in the net asset position it is a bit early to be confident of a turning point in the currency.

And where, might you ask, does the other chronically weak currency, the pound, fit into all of this? Oh, it becomes a strong currency too - but I am afraid that story needs another article.