One can state categorically that there has since 1980 been no reliable relationship between the dollar and:-The US trade balance; economic growth; money supply; budget deficit; capital flows; and oil prices.

This has been true both for the absolute US figures and for the US data relative to other countries, which can be measured as differentials. The purported relationships here come into the “absolutely useless” category.

Admit it: that simplifies matters a lot. Note that many of these data are thought to have a bearing on the outlook for interest yields, which is why they are avidly watched by currency observers – for the existence of links between currencies and interest yields is not in dispute. We shall return to that. But what of intervention and political events?

Political events, like the timing of sterling’s full adherence to the EMS, German reunification, Russian Revolution of ’91 et al can undoubtedly have a more than temporary impact on dollar rates and cross rates alike. Who can forget the effect of James Baker’s efforts to jawbone the dollar down in 1986– 7? So yes, there may be things to keep an eye out for there,

but not nearly as many as observers think. Baker’s efforts, for example, were highly successful. But was that just because they were in the direction of the perceived trend in the dollar anyway? And the difficulty with most political influences is the sheer impossibility of knowing to what extent they have been discounted in the market-place.

Much the same has to be said of intervention. We have seen it blown away so many times. And the times we have seen it succeed have usually been the times when there was other evidence that the tide was turning –or when it was in the direction of the trend as in 1985, when the central banks were “pushing down on a lead balloon”. So one has been forced to the conclusion that intervention –or at least the prospect of intervention is another piece of baggage that must go; we can heed it only when we can look back on a pattern of intervention which corroborates a script drawn from other evidence. And then we had jolly well better heed it.

Note here a proviso to the conclusion that US trade is part of the excess baggage: it doesn’t mean that relative trade performance among non-US partners should be ignored. Nobody much looks at that. Yet see what a good account the differential between German and Japanese trade gives of the Yen/D-Mark cross rate – a relationship that has been valid ever since currencies floated free. How so? Because no-one is looking there.

But the heaviest piece of baggage has to be interest rate forecasts. That’s where everyone is looking. Remember the mullah and his key. Forecasting interest rates is not where it’s at. I think the attitude of all the punters out there is that this is the holy grail: if only we could forecast interest yields we would know where the dollar was headed. If only we could, maybe

we could. But we can’t, and even if we could, maybe we still couldn’t! Currency forecasting, like all forecasting of financial markets is the art ofthe possible. It’s impossible to find the key out in the bright sunlight when it's hidden in the dark – in the depths of people’s hearts. Meanwhile, here are some conclusions about the valid underlying trend.

1) An emerging perception that a currency has stabilised or entered an uptrend combined with a relatively high real interest yield strongly favours an upward movement; and an emerging perception that a currency has stabilised or entered a downtrend combined with a relatively low interest yield strongly favours a slide.

2) On empirical evidence you can count on a significant time-lag before a big shift in real interest rates makes an impact on a currency: no hurry, and no need to forecast.

3) What matters with interest rates is not where they have been moving but why. There is no reliable direct link between interest rate movements and currency movements. Most of the time, you will see an inverse relationship. The reason for this is that most of the time, currencies are leading interest rates rather than the other way round. So when do interest rates lead currencies ? And when is this relationship reliable and useful? Answers next chapter.

4) We can speculate endlessly about future events – political, economic and financial. Don’t bother. That’s where the crowd is looking. All we need to do is to focus on what the crowd is expecting, and we’ll know what’s already reflected in prices. What makes prices move is surprise.

Yield differentials* drive currencies. But what matters is not the differentials themselves, but how they come about. It’s not interest rates, but people who move currencies

In the case of a Treasury bill*, the yield and the price are all one: T –bills are repaid at par 100 and issued at an appropriate discount so that the capital appreciation is the yield –the yield is in the price. The same kind of thing happens in currencies.

For example, you might be told that the pound (in June, say) is $2.00 for cash or “spot”*, $1.97 for September and $1.94 for December. These different prices reflect the difference in interest rates on sterling and on dollars, when sterling interest rates are higher. Why? Because if you want dollars in, say, 3 months time in exchange for pounds and want to pin down today’s exchange rate, you could either 1) buy them now for cash on the spot; or 2) buy them “forward*”, for settlementin3 months time. Option 2) must work out at exactly the same cost as option 1). Otherwise, Shell and Volkswagen and Citibank would arbitrage* the hell out of any difference. You can see that.

To spell out the arithmetic. Suppose 3-month sterling interest rates are 12% and 3-month dollar rates are 6% (i.e. sterling deposits return 1.5% a quarter more than dollar deposits, and sterling borrowings cost 1.5% a quarter more). If you (1) buy dollars now, you will get 1.5% interest over 3 months and forfeit the 3% that you would have earned on your pounds: you would only do that if you got 1.5% more dollars by buying now than by buying in 3 months time – which of course is just what you do get.

And if you 2) buy dollars 3 months forward you get 1.5% less than if you buy them for cash on the spot: obviously, because you still have your pounds which are earning you 3% a quarter, which is 1.5% more than you can earn on dollars. QED. This arithmetic is carved in stone.

If you bear this arithmetic in mind, you will never forget that interest rate differentials always have been and always will be lurking there at the heart of every currency parity: they are “in the price” in every forward and future transaction, just as the T-bill yield is “in the price” . Consequently

no trader in forward or futures markets can help being conscious over time of the interest rate differentials in the currencies they are trading. If you hold a currency with a strongly adverse differential, it gnaws away at your position, whittling down over time any profits you may be making and compounding losses. If it is a favourable differential it cumulates your profits, over time, and has a forgiving effect on losses. For example, assuming the interest rate structure cited above, if you are holding sterling against dollars, the value of your holding rises 1.5% a quarter with no change in the spot rate. Conversely, if you are short sterling against the dollar, you lose 1.5% a quarter with an unchanged spot rate.