Published: December 4 2004
The dollar fell to another "record" low against the euro on Friday, currency markets chattered.
No it did not.
The only way anyone can describe the current level of the dollar as a record is if they think the world started in 1999; or that Europe had maintained a barter economy until the euro came along; or that currencies had not been traded across the Atlantic before then.
Against European currencies, the dollar was just as low eight years ago and much lower a further four years earlier in 1992.
But if the only crime committed by those purporting to shed light on events in currency markets was a dodgy sense of history, it would be small beer.
Far worse is the daily dose of predictions about where the dollar, euro or sterling is heading and the stream of explanations about the latest gyrations in the currency markets. In each case the language is precise, the reasons are cast-iron and the views about the next movement are firmly held.
Surely, if those responsible know the long and miserable history of attempts to predict exchange rates, they know they cannot make such pronouncements about the near future with any certainty. And if they do not know that they cannot know, they should not be in the business.
To borrow from Donald Rumsfeld, forecasting daily, weekly or monthly exchange rate movements is a known unknown. Studies have shown that the best forecast of today's exchange rate is yesterday's level.
That is why Alan Greenspan, the Federal Reserve chairman, said in 2002: "There may be more forecasting of exchange rates with less success than almost any other variable." Mervyn King, governor of the Bank of England, said last month: "I have no idea where exchange rates will go in the future and I have no intention of ever starting to forecast exchange rates. That's a mug's game."
It gets worse. The explanations for past currency movements defy credibility. When bad data come out of the US and the dollar falls, many analysts say that "the dollar fell on expectations of lower US interest rates."
If the dollar rises after bad US news, they say: "The dollar rose on expectations of a narrowing of the US current account."
In each case, they are attempts to rationalise something that cannot be rationalised.
Research has shown conclusively that it is almost impossible to explain short-term currency movements.
We do know that currencies slowly adjust to equalise relative prices over very many years, but that does not help in the short-term predictions so beloved of the currency markets.
So here are two early new year resolutions for currency analysts.
When the dollar goes down, they should say: "At the old price there were more sellers than buyers." When asked where it will go next, they should say: "I haven't the foggiest."
And a resolution for the rest of us: do not listen to the endless chatter in currency markets. Learn to live with uncertainty. The writer is the FT's economics editor