The bull case for the dollar has three strands - that it is undervalued on purchasing power parity, that US economic growth is likely to slow, thereby reducing inflationary pressures, and that the US current account is likely to improve in the months ahead.
The PPP case is the standard starting point for anyone wanting to make a fundamental judgement on an exchange rate. True, currencies can be out of line with their purchasing powers for years on end. Countries with efficient export industries but with severe restrictions on imports, such as Japan or the Soviet Union before its market reforms, will tend to have a currency that trades above its purchasing power parity.
Countries that are particularly open to imports but have suffered disruption in their exports, like Russia and the rest of Eastern Europe now, will tend to have currencies trading below their PPP.
Given the Japanese barriers to imports - cultural as well as administrative - one would expect some undervaluation of all currencies against the yen, just as all currencies are overvalued against the rouble. So for a rational judgement of the dollar's general PPP, one should look at levels against the open economies of Europe. The result, taking parities calculated by Citibank at the beginning of this year, shows the dollar at about DM1.72 against the mark and dollars 1.55 against sterling. This compares with rates of around DM1.58, and dollars 1.55 against the pound.
In other words, the dollar is already undervalued against the mark but within 10 per cent of the PPP, and it is on the nose against the pound. The really wacky numbers are for the yen, where the PPP is around Y148 and the close yesterday below Y100. As far as currencies are seriously out of line, the yen is the one with the problem, not the dollar.
Now for point two, the US economy. This will remain the most over-analysed economy on earth, and the constant temperaturetaking results in a stream of ( often misleading) information.
But, in a nutshell, the fears of the first three years of recovery - that it was a slower rise out of recession than in any previous cycle - have given way this spring to fears that growth has been unsustainably rapid. This has led to criticism of the Federal Reserve for not meeting this growth with sufficiently large increases in short-term interest rates.
What no one knows, though, is how the US economy will respond to the rises in long-term interest rates that have taken place. The weakness of the bond market must have some effect on demand. The effect will be all the greater if equities in the US continue to fall.
The markets have compensated for the tardiness of the Fed in tightening policy by tightening it themselves. This must slow growth and restrain inflation.
Finally, the current account. The popular wisdom is that at last the string of current account deficits of the US through the 1980s is catching up and foreign investors are no longer prepared to offset these deficits by transferring investment funds into the US. There is some truth in this, as there usually is in such views, and the fall in bond prices this year will have unnerved many foreign investors.
But after some months of widening there are now signs that the US current account will slowly improve. Slower growth will tend to cut import demand. Service earnings have been very strong and account for a growing proportion of world trade - invisibles are now more than 40 per cent of the total - and the deficit may be overstated because of the difficulty in distinguishing between current inflows from the export of services and capital inflows. The deficit, even on published figures, is in any case only 2 per cent of GDP.
Could this case for a modestly higher dollar become credible to the markets? The main case against it is that strong growth will continue and will not be met by a sufficiently sharp rise in interest rates to allay inflationary fears.
Were this to happen the malaise in the bond market could become seriously destabilising and a dollar crash - pushing it to a silly undervaluation rather that its present modest one - could occur. The PPP argument only works in the long term, for while PPP rates pull they may take a long time to do so.
Until it is clear that the US bond market has turned, which could take anything up to two years, the exchanges will be cautious. But just as dollar weakness has struck so suddenly, a recovery could be equally rapid.