Queries about sterling's future are the biggest problem facing the Bank of England in its monthly interest rates decisions

Few people who follow the actions of the Bank of England's monetary policy committee can have agreed with all of them. But I have never been able to feel strongly about my own occasional disagreements - which have usually been on the more "doveish" side. Sometimes the more dubious actions have not been serious enough to make too much difference. And on other occasions there has been plenty of time to reverse them before they could do real harm.

Decision time for the monetary policy committee is due again next Thursday, As usual the decision looks finely balanced. As a starting point, let us look at what the UK economy is doing. Enough of the year has gone by for an estimate to be made of the 1999 outcome. A good mainstream estimate to take is that of Goldman Sachs, which shows an output rise of 1½ per cent.

There are many indicators of underlying inflation - such as the official RPIX (which excludes mortgage interest), the Bank's own RPIY (which also excludes indirect tax effects), and the GDP deflator. These are nearly all at around 1½ to 2 per cent.

If you add together the estimates of the growth rate and of underlying inflation you find that gross domestic product in nominal terms is rising by less than 3½ per cent a year. But that does not account for all the increase in demand. For quite a lot of the spending surge is going into increased imports - as is the case in the US. What matters here is not the actual current balance of payments deficit - which does not have to be zero - but its rate of increase. Goldman Sachs expects this deficit to rise from nil last year to nearly 1½ per cent this year, measured as a percentage of GDP.

All the items just mentioned give a total increase in nominal spending of under 5 per cent. This is in line with what the Bank of England considers the economy can sustain.

But this is not all that needs to be said even without a formal forecast. There are all the usual conflicting indicators of whether demand on resources is speeding up next year or not. Here judgments will differ depending on the relative weight attached to forces such as booming house prices, the still-depressed manufacturing sector and controversies on the supposed "new paradigm" in the labour market.

Taking into account all the domestic indicators, there would in my view be a strong case for keeping UK interest rates on hold for now. But a further factor, which can never be ignored in an open economy like Britain's, is the exchange rate. If this were to fall it would add to inflation.

One reason given by Sushil Wadhwani, the newest member of the MPC, for voting with the minority against last month's increase in UK base interest rates to 5.25 per cent was that he did not share the majority assumption that sterling would fall. The August MPC projection assumed that the pound would drop over two years from an equivalent of DM2.95 to DM2.75. So far it has risen to well above DM3. If, however, sterling were to remain around recent levels - which Mr Wadhwani considers likely "on the assumption that current economic conditions broadly persist" - inflation in two years' time would be nearly half a percentage point less on the Bank's own model. In that case, interest rates could be well over one percentage point less than they would otherwise be.

Mr Wadhwani has since explained his view of the exchange rate in a lecture he gave on September 16 to the London School of Economics. In it, he discussed matters far more important than whether the Bank should raise interest rates again next week.

He starts from the conventional basis for currency forecasts, shown in the chart, called uncovered interest rate parity or UIP. This means is that if you take the difference between say British and German 12-month interest rates - now nearly 3 percentage points - this gives the expected annual depreciation of sterling against the D-Mark - or the euro. While I have never believed that UIP gave anything like a correct forecast of the exchange rate, I did believe that it was hard to do any better.

But as the chart shows, forecasts on this basis have performed badly. An alternative crude assumption is that the path of the currency is a random walk - which Mr Wadhwani interprets to mean that the best forecast is that sterling will remain the same - has performed slightly less badly. But it has not performed well.

The author has his own "intermediate" model in which the exchange rate depends on a number of common sense factors that affect market perceptions but which does not attempt to estimate sterling's true underlying long-term value.

In the light of this model he examines the 27.6 per cent rise in the sterling-DM exchange rate in the two years up to February 1998. Some two-thirds of the rise is accounted for by a simple move back by the real exchange rates to its trend. This gives the surprisingly high rate of around DM3 to the pound.

The second most important factor is the lower rate of unemployment in Britain than in Germany - and for that matter in the rest of the euro-zone. He believes that this has a direct effect on investment flows, perhaps because the country with fewer unemployed is regarded by investors as having a better underlying economic performance. The factor going in the other direction has been net foreign exchange assets, where the trend has favoured Germany - but not by nearly enough to offset everything else.

This intermediate market-perceived sterling rate in the neighbourhood of DM3 is not only much higher than most manufacturers think they could live with, but also higher than most economists' estimates of equilibrium exchange rates or purchasing power parity, which are much nearer DM2.5. Mr Wadhwani himself emphasises that his model may be useful to help set interest rates for a floating pound, but does not provide any indication of a the rate at which Britain should join economic and monetary union.

What, short of the prospect of early Emu membership, could initiate a fall in sterling to more comfortable levels? One trigger that Mr Wadhwani mentions is a "steep fall on Wall Street". He suggests that this is because, in a period of investor risk aversion which would accompany a falling Wall Street, investors would be less willing to finance UK current account deficits. These are not important now, but could suddenly become so in changed circumstances.

A Wall Street crash does not sound a very happy way of reducing sterling to a level corresponding with its relative purchasing power. But there is a slight silver lining even here. For a worldwide plunge in equity values would reduce people's wealth by so much that spending would be depressed and the MPC would not have to raise interest rates to offset the fall in the pound. It might even have to reduce interest rates to offset deflationary forces.

Meanwhile, what should the MPC do? There is one option that Mr Wadhwani mentions, but does not develop or endorse. This is that the MPC should assume that sterling will stay where it is and react to a fall in the exchange rate if and when it happens. There is no free lunch; and the price of following this alternative would be that interest rate changes might be fewer but more drastic. It is still the better course.