Speed of the dollar's decline suggests American recovery may not be swift

'The largest threat to the dollar is a recession induced by a fall-off in consumer demand'

Has the dollar reached an historic turning point? Probably yes. Does that mean that in the coming months it will collapse? Probably no.

Has the dollar reached an historic turning point? Probably yes. Does that mean that in the coming months it will collapse? Probably no.

Currency turning points are notoriously difficult to call. The repeated efforts by City forecasters to explain why the euro has been so weak (and why their forecasts were so wrong) are testimony to that. The problem is partly because the relationships between economies, current accounts, interest rates and a currency's value keep changing, partly because there are always a lot of random forces that are impossible to predict and partly because markets almost invariably overshoot.

As an example of the first, 20 years ago the current account was an important factor in determining wherever a currency rose or fell: it went up when the current account was moving into the black and fell when it was going into the red.

But now the capital account is much more important than the current account. Often (as in the case of the US recently) a deteriorating current account deficit is a by-product of rapid growth and is more than covered by investment inflows chasing that growth. So the currency rises rather than falls.

As for the random factors, the euro/dollar rate provides another example, though one not yet fully understood. Part of its recent strength (though no one knows how much) seems to have been associated with the need for Europeans to move their cash holdings out of currencies that will shortly disappear into one that will endure beyond 1 January 2002. It may be that by now most of these balances have been moved, thereby clearing the way for the first stage of a recovery by the euro.

And overshooting? Well, back on 6 July, only six weeks ago, the dollar hit a 15-year high on its trade-weighted measure. It has now fallen more than 7 per cent against the euro since then, but it is still by historical standards very strong – as the graph on the left charting its progress during the 1990s against the mark shows. The speed of the decline, though, does suggest that some sort of turning point may have been reached.

What next? Whether the dollar remains a fundamentally strong currency in the next three to five years will depend on several factors.

The most easily identifiable is the relative strength of the US economy vis-à-vis Europe and Japan. Japan is clearly very weak indeed and will continue to be so for several months, but there really is not much in it between Europe and the US. The first estimate of second quarter growth in the US was still positive but there are suggestions that it may be revised down and may actually turn out to be negative. In Europe, the figures are just coming through now: Italy was negative, Germany probably also negative but we don't yet have the numbers, while France was probably still growing. (The UK, by the way, seems still to show surprising strength.)

But markets look forward and what they are trying to figure out now is whether the US can stage a swifter and stronger recovery than Europe. Six weeks ago, when the dollar hit that high, the answer was yes. Now people are less sure.

A glance at the other two graphs shows why. The key to sustaining growth is consumer demand. But consumers have taken a heavy hit from the fall in the stock market, buying into tech stocks at just the wrong time. In fact GFC Economics, which produces these charts, notes that US household financial assets have taken the biggest hit since the Second World War (centre graph). Up to now it has been possible to justify the fall in the savings ratio by pointing out that it was offset by the bull-market rise in household assets (right-hand graph). But now nearly all that rise has been wiped out. The ratio of assets to income is back to the level of the early 1990s. Maybe Americans will start to feel the need to save again and, if they do, that will mean a fall in consumption.

Probably the largest single threat to the dollar is a recession induced by a fall-off in consumer demand. At some stage Americans will have to rebuild their savings but what we need is that adjustment to be spread over several years – not to happen suddenly as consumers take fright. I do not actually expect them to panic, but were they to do so expect dollar-holders to panic too.

There are of course always two sides to a currency rate. Might European consumers also suddenly take fright? In theory they are better placed, for Continental savings levels are higher. Even Britons are still saving a bit, though only about 3 per cent of their household incomes, as opposed to a historical norm of more like 8 per cent. But consumption on the Continent is really quite weak and the possibility that Germany and Italy might be in recession will keep it so.

Random factors will also affect the dollar in the future, as they have in the past. Global political instability tends to help the dollar and it seems reasonable to expect an increasingly rough period in the Middle East in the coming months. The Japan story has some way to run too: things will get worse before they get better and that will also tend to help the dollar.

Were the euro launch to be unsuccessful, that would of course confirm the dollar as the currency of choice for the world's criminals and tax evaders. In fact it is hard to see random factors that might seriously undermine the dollar, and easy to see some that will support it.

And the whole overshoot issue? I suppose the question here is whether the dollar is fundamentally overvalued at its present levels or whether it was actually rather undervalued during most of the 1980s and early 1990s. My intuitive guess is that it is a bit overvalued now, but not by more than 10 per cent. Parity with the euro? Eventually, yes; but not for a while yet.