The MPC lives in an ivory tower but does use its escape hatches

Is the strong pound a big enough problem for the economy that the Bank of England ought to have reacted to it?

Many industrialists struggling to export to the rest of Europe would have no hesitation in saying that interest rates should have been much lower to help limit the rise in sterling,

It is also one of the arguments made in a new publication on the role of asset prices in monetary policy *. This publication is of more than passing interest as one of the coauthors is Sushil Wadhani of the Monetary Policy Committee.

In setting interest rates, most central banks do not take explicit account of the prices of most assets, like share prices, exchange rates or housing and land prices. Indeed, an increasing number are setting interest rate policy with reference to a target for some measure of consumer price inflation, and nothing else.

They might take account of asset price measures informally in making judgements, or in forecasting inflation, but for more and more banks the explicit rule for setting interest rates is cast in terms of a pure inflation target.

There are get-out clauses, however. The Federal Reserve is charged with maintaining a high level of employment, and the Bank of England can let inflation diverge more than one percentage point from the 2.5% target as long as it has a good reason to give in the explanatory letter that the Governor must send to the Chancellor.

The existence of 'escape hatches' indicates that inflation is not, in fact, the only thing that matters; so do output and employment, and not just their levels but also their volatility.

On the inflation front too, stability is as important as keeping the rate low.

The fact that central banks have more than one target - not just low inflation, but steady inflation, and steady growth too - is an important reason why setting interest rates is a matter of judgement as well as technical expertise. This is where asset prices come in. For one thing, they can certainly be useful as a means of improving an inflation forecast. Often, they are the earliest available economic indicators, and they are forward-looking. But perhaps more important, they can in principle be used as a device to reduce volatility in inflation and output.

This is, at any rate, the argument put forward in the report by Mr Wadhwani and his colleagues. They show that by actually incorporating targets for asset price growth as well as consumer price increases, central banks can, in theory, attain the same level of inflation and more stability too. In other words, a simple inflation target is not ideal when stability matters.

The theoretical result, confirmed by simulations using computer models, chimes with a common criticism of the MPC's policies. This boils down to what could be called the 'Ivory Tower' critique - essentially that the remote academics setting interest rates ought to take more account of the impact of their decisions on growth and jobs as well as on inflation.

The gut instinct of the critics is that rates would have been lower if the MPC had to moderate fluctuations in growth as well as hit an inflation target.

Certainly, the strong exchange rate has been a powerful argument for keeping interest rates lower than they might otherwise have been. The report argues that when an overvalued exchange rate is due to a portfolio shift - and does not simply reflect stronger demand growth in the country affected - the level of loan costs should be set according to an indicator of monetary conditions that in effect takes a weighted average of interest rates and the exchange rate.

If the MPC were using a monetary index, a stronger pound would point to lower interest rates. The catch is in the caveat. To set an ideal policy, the central bank has to know what has caused the exchange rate appreciation, and be sure that it is not mainly a reflection of the business cycle.

There must be a good pragmatic argument for making some allowance for a fundamentally overvalued pound, but that is not the same as a good argument for using a combined interest rate/exchange rate policy at all times, regardless of the reasons for the currency's strength.

A separate section of the report looks at share prices, and argues that central banks also set a better policy when they raise interest rates to pre-empt stock market bubbles.

Taking account of the future effects of share price movements also contributes to greater macroeconomic stability. Of course, central banks have to be able to estimate when there is a bubble, but then they have to estimate all sorts of things anyway. This is a problem in practice rather than in principle.

However, it is not obvious what the ideal interest rate ought to be when asset prices send different signals, as they have in the USA and UK in recent years. Stock market bubbles (perhaps) have gone hand-in-hand with exchange rate overvaluations, and indeed it seems likely that there is a causal link.

The currencies have almost certainly risen in good part because investors wanted to buy equities and other assets in dollars and sterling.

The separate asset prices alone could not have guided policy decisions, and interest rates would probably not have been very different with an asset price-- augmented inflation target. Making judgements about the particular context has been unavoidable in recent years.

If anything, the members of the MPC and of the Federal Reserve Open Market Committee have been much easier-going on the level of rates than they would have been if they had been using an economic model to do their job for them. The academic rules have pointed to a need for interest rates to be both higher and more volatile than has actually been the case. Central bankers are far more cautious in practice than they ought to be in theory.

And their judgements seem to have paid off. Both the USA and the UK have enjoyed many years of stable growth and low and steady inflation. The balance of growth could definitely improve, but as macroeconomic , performances go, it could have been a lot worse.