Serious Money Ideas can become self-fulfilling prophecies even if they fail simple tests, writes Chris Johns.

Exchange rates are a driving force for many different asset classes. The ups and down of the dollar, yen and euro can reflect underlying shifts in economic fundamentals that feed through into profits, which drive equities, and interest rates, which drive bonds.

But currency gyrations often reflect nothing fundamental at all: the structure of the foreign exchanges means that rates can often go wildly up or down (often both) for reasons that have nothing to do with the stuff of economics textbooks.

A market that sees nearly two trillion dollars a day in trading activity is one that is often characterised by herd behaviour and countless nutty theories about why prices do what they do.

Ideas about why an exchange rate might move become important if they are believed and have little connection with whether or not they are right.

Ideas often become self-fulfilling prophecies even though they fail the simplest tests of rationality or logic.

The current fashion in global markets is to assert that the dollar must go down, more or less forever, because the US has a balance of payment deficit approaching 6 per cent of GDP.

No country, certainly not a large one at any rate, has ever run that kind of deficit and we know, with certainty, that it is not sustainable. Some commentators at the lunatic end of the profession have tried to suggest that the bulk of this deficit occurs because of the activities of US companies - firms importing and exporting to themselves - so it is as significant as the trade deficit between Arkansas and Alabama.

But the world pays no attention to such ravings and believes that the US must stop running such a large deficit and must acquire the same cost structure as the Chinese. All those cheap manufactured goods currently flooding into the US from Asia must instead be put together in Idaho. This, according to my calculations, means that the dollar has to decline for the next several decades in order to equalise the unit wage costs of US and Chinese manufacturing industry.

One of the curious aspects of all of this is that the currencies against which the dollar should decline - the Chinese renminbi in particular - are fixed, so the dollar goes down against those currencies which play a relatively small part in the US imbalances - the euro area. According to this logic, the US dollar must decline to the point where a trade surplus with Europe balances the trade deficit with China. At current rates of depreciation we will have to wait until a fifth or sixth generation George Bush is in residence at the White House to see all of this happen - and one euro will buy most of Microsoft, or what's left of it.

The rise of the euro is being accompanied by some unusual circumstances. What is not odd is that European economic growth is once again starting to disappoint. This could be because the European economy always disappoints - it's just structured that way - but the behaviour of the exchange rate is also playing its part.

What is very different is the behaviour of European equity markets. For many years now we have become used to the simple empirical regularity of a negative correlation between European stock prices and the euro: historically, when the euro goes up, Europe's equity markets go down (and vice versa). That looks logical because of the connection with growth: if a strong euro crimps the economy, corporate profits are also hit, damaging equities.

However, for virtually all of 2004, European equities have been positively correlated with the euro. Each time the currency rises, stock prices, on average, have gone up. Indeed, the most recent surge in the euro has been accompanied by some stock markets moving up to their highs for the year. Even markets traditionally regarded as "dollar-sensitive" - Germany comes to mind - have continued to do well. What is going on?

The answer, I think, lies with the increasing disconnection between European corporate life and the behaviour of the wider economy. The kinds of companies that make it as far as a stock market listing in Europe often have to face up to the facts of globalisation. Competition is the most obvious of these, but the demands of a global investor base also play a role.

Put simply, in order to survive, quoted European companies have to match the standards set by global competitors and those demanded by global investors. Given the hopeless performance of the wider economy this is a big mountain to climb and few make it - hence the relatively small size of Europe's equity markets.

But those that do succeed, by definition, offer decent returns to investors. In Germany, there are signs that companies have given up waiting for government-led restructuring of the labour market and are doing it for themselves.

As a result, the German stock market has been doing better than expected, notwithstanding the stronger euro.

More generally, the levels of profitability being reported by European quoted business, in terms of margins, have never been higher. Hence the strong performance of markets in the face of much adversity. European equities cannot remain positively correlated with the euro forever. But the reality of corporate restructuring will provide something of a cushion for a while yet.

Should the US economy start to save a bit more - this is the problem, not competitiveness - the dollar will stop falling. Sadly, this does not look likely: few commentators think that the Bush administration knows it has fiscal and balance of payments deficits, so curative action does not look likely. The way in which a weaker dollar forces higher saving is by sparking a US recession, an odd policy choice for the world economy.

The sure fire bet looks to be stronger Asian currencies over the medium term. That alone, never mind continued strong growth, makes the region an extremely attractive investment proposition.

Chris Johns is an investment strategist with Collins Stewart. All opinions are personal.