The big question is when, not whether, to cut UK interest rates

The reasons are twofold. First, there is no need for higher interest rates. Inflation is, and seems likely to remain, below or close to the government target of 2 per cent, a target that is supposedly symmetrical. And second, the present level of rates, while low by the standards of the past half-century, seems to be high enough to curb any excessive growth in the economy.

You can see this in the two charts. On the top is what has been happening to prices, both at the formal retail sales level and at the reported shop price level. Both have been close to, or below, zero for most of the past five years. The British Retail Consortium's statistics gives an early estimate of prices and though it can diverge from the official figures (it did report a spike in late 2004) the latest number is minus 0.3 per cent for June. So there is no inflation at a retail level.

Even allowing for rises in the charges for services, the overall level of inflation for the first five years of this century has been 1.3 per cent, which by a curious co-incidence is the same as the level of inflation during the first five years of the last century.

The bottom chart shows why the present level of interest rates is sufficient to contain demand. ABN-Amro's economic team has done some calculations which show that the total debt burden is still just below the level of the late 1980s and the interest gearing far below - the top two lines. But look at the bottom two lines: the repayment of unsecured debt, credit card stuff and the like, and the repayment of mortgage debt. They make up a much bigger proportion of the total debt burden.

In a nutshell, the pressure in the late 1980s and early 1990s came from high interest rates; now it comes from repayment of principal. Debts have to be serviced - but they also have to be repaid. The Bank can cut interest rates and reduce the debt burden that way. But it cannot cut repayment of principal. The ABN-Amro team calculates that base rates would have to come down to about 2.5 per cent to give householders significant help in coping with their debt burden.

What does this say about the world of finance in 2005, except that it is more like the world of 1905 than anything in our adult experience?

Several points. The first is that people's responses probably still lag reality. I suspect - and there is no way of knowing for sure - that most borrowers reckon that inflation will reduce the real burden of their debt. They think that incomes will rise in both money and real terms, which is hopefully true but has not been the case in Germany or Japan in the past few years. They think too that most asset prices - in particular property prices - will ultimately rise, though again that has not happened in those two countries. And they assume growth will continue at a fair clip, sufficient at least to ensure that there is no return to mass unemployment.

In other words they assume that the experience of the long boom since the early 1990s will continue. So even if their debts are a bit high for comfort they can earn their way out of trouble. These are not silly assumptions. But by working on this basis, people leave themselves little margin for error.

If most people implicitly expect that there will be a bit of inflation over the next few years, the financial markets are assuming quite the opposite. Bond yields are close to the levels they were at the beginning of the last century. The only sensible assumption behind lending to the UK government for 10 years at a bit more than 4.2 per cent is that inflation will be about 1.5 per cent or less. For Germany the corresponding yield is 3.2 per cent, carrying the assumption that inflation will be close to zero.

Now it may be that the markets are so awash with money created by the central banks in an effort to puff up the world economy post the early 2000s recession, that the bond market investors are simply making a mistake. Many of us think they are, particularly those investors in Japan and China who have been piling into US securities, carrying the exchange-rate risk. But, currency risk aside, if there is virtually no inflation over the next decade it won't be a huge mistake - or at least not as huge as borrowers who at the back of their minds assume that inflation will save them in the end.

Back to the Bank's decision pathway. The evidence of the slowdown is beyond dispute. The figures are a bit bumpy, with industrial production weak, service industries doing OK and retailing recovering a touch but pretty soft. But figures almost invariably lag reality and the task of the Monetary Policy Committee members is not to look at statistics and make a decision on that basis. Any of us could do that. What they have to do is understand the feel of the figures and meld that into the non-statistical data that comes into the Bank, plus their own intuitive feeling for where the economy might be in six months.

If the lags in the figures are long, the lags in the impact of policy are longer. So what should the Bank do?

I don't think it needs to do anything now. Yes, the CBI wants a cut but then it has a history of calling for cheap money. Businesses in general borrow and they want to do so at the cheapest rate. There are two complications which suggest caution. One is the current weakness of sterling - not a dramatic decline, but enough to beware particularly when US interest rates are heading up. You don't want sterling to weaken too much against the dollar at a time when nominal oil prices, denominated in dollars, are at an all-time high. There is an argument for waiting until it is clear that US rates will not rise any further before starting to pull ours down.

The other is the fiscal position. Growth this year is clearly going to be well below the number the Chancellor first though of. It will not be above 3 per cent, and may be quite close to 2 per cent. But we don't know how much damage this does to revenues and spending. If public finances are heading into a real mess, lower rates will be needed to offset the rise in taxation that is needed to correct that mess. But we won't have much of a feel for that until the autumn.

That all suggests another couple of months wait: maybe September, more likely October. But then when rates do start to fall there may be a case for giving a sharp signal: cutting by half a point, not a quarter. Not sure about that, because it may be that to signal a change in direction will be enough. But it is often better to wait and move decisively than move early and wonder whether you have done the right thing.