Despite their recent success inflation targets are highly unlikely to be the last word in monetary policy

There is one prediction about the next Wednesday's British Budget which can be safely made. That is that no sooner has the chancellor Gordon Brown sat down than financial discussion will shift to the likely decision of the Monetary Policy Committee, the very next day, on UK short term interest rates.

In principle the decision must be guided by the Chancellor's directive that the inflation target is "at all times" 2½ per cent and by the threat of having to write a letter of explanation to him if the recorded rate deviates by more than 1 percentage point on either side. Indeed inflation targets have enjoyed such success in the last decade that it seems vain to query them. Yet it is just in such periods that it is most necessary to look at the weaknesses of the prevailing regime and ask what is to follow, It is not enough to go on blindly reiterating what the Chancellor said in 1997.

It was originally expected that explanations of target breaches would be quite frequent; and the deputy governor, Mervyn King even spoke about "reviving the lost art of letter writing". At his last press conference Mr King said that the probability of having to write a letter was now higher, a prospect he seemed almost to relish. The thought behind such remarks is that almost any monetary regime can be made to work if run intelligently. But already financial commentators are discussing the prospect of a letter from the Bank in terms of failure and humiliation; and so far the chancellor has done nothing to discourage such discussion. Moreover we cannot always rely on enlightened people being in charge; and who knows what the political pressures of the next few years will bring. The number of countries with their own central banks is now running into hundreds; and there is a need for some guidelines which do not require the shrewdness of a Keynes or a Milton Friedman to operate.

The Chancellor introduced a 2½ per cent inflation target as a rough and ready way of preventing the Bank of England from being too contractionary in the face of recession, as it was in the 1920s. But circumstances in which this target would be inappropriate are easy to envisage.

Suppose that inflation is low or negative, not because of a recession but because the much heralded productivity takeoff has at last occurred. Competition between firms will still put a limit on profit margins and the benefits of the breakthrough will be felt in zero inflation or even falling prices. Would it be sensible then for the Bank to have to write a letter of abject humiliation and promise and to take steps to lever up inflation to 2½ per cent?. It is not stable or falling prices which are disliked by the public but the recessions which have often - but far from always - accompanied them.

Now take an opposite situation, an oil price explosion. Would it be sensible for central banks to force the inflation rate down quickly to where it was before? Yet one reason why the European Central Bank has been reluctant to cut interest rates is that the recorded euro area inflation rate has been running above the approved corridor of 0 to 2 per cent, entirely as a result of increased oil prices.

Or suppose that there is some event such as a Wall St. crash or a collapse of an institution like Long Term Capital which induces a rush for liquidity. Should not the Fed supply this need even if there is no immediate danger of price deflation?

There was once an alternative theory of monetary policy different to both the inflation target version of monetarism and the Keynesian emphasis on stimulating output directly. This was the so-called "Austrian" school, which in its turn drew on a Swedish economist of a century ago, Knut Wicksell. The latter distinguished between two rates of interest. There was the actual market rate;and there was also the unobservable "natural rate" which would be bring intended savings and investment into line if there were no distortions introduced by the monetary system.

The aim of monetary policy should to be "neutral"; in other words to try to minimise divergences between the natural and the market rates. The "Austrians", in particular Von Mises and Heyek, emphasised that it was not sufficient to try to keep the general price level stable. Quite apart from index number problems - which they overemphasised - the main harmful effect of inflation and deflation came, they argued, from distortions in relative prices which could occur even if the overall price index remained fairly stable. They emphasised particularly the malinvestment which would result if production became either too capital intensive or not intensive enough.

One still controversial example is that of the behaviour of the Fed in the 1920s, leading up to the 1929 crash. While the monetarists believe that the Fed of those days was too tight, the "Austrians" considered, even at the time, that it was from 1927 onwards, too expansionary in promoting an investment boom which could not last. A similar controversy surrounds the period since the mid-1990's, during which US inflation has remained low and stable, but when there were many signs of a rush into fashionable investments in IT and elsewhere.

During the interwar period the Austrian school discredited itself by insisting that the depression should run its course, to allow mistaken investment to be liquidated. Lionel Robbins of the London School of Economics, who had earlier supported such views, made a well-known recantation. Even "assuming that the original diagnosis of successive financial ease and mistaken real investment was correct" to take no action against the ensuing depression, he remarked, was "as unsuitable as denying blankets and stimulants to a drunk who had fallen into an icy pond, on the ground that his original trouble was overheating."

Since the early 1970s there has however been a revival of Austrian economics, not anywhere near Vienna, but mostly in the United States where it is now a sufficiently important minority movement for respectable economic publishers to maintain sections of their lists devoted to it.

Like all such minority movements the new "Austrians" have their quota of fanatics and cranks. Indeed when one of the more sophisticated of them attempted a semi-mathematical exposition of the school's teachings, he received hate mail from some who thought that he was being disloyal to the anti-mathematical gospel of von Mises.

But at the other end of the spectrum there are "Austrians" well aware of modern techniques, as well as of mainstream economics and of what is going on in the economy. One of the best examples can be found in a book with the unnecessarily forbidding title of Microfoundations and Macroeconomics, by Steven Horwitz, (Routledge, 2000@. The emphasis is on the need to shift from monetary or inflation targets to some attempt at neutral money. He is less convincing on how to do so.

Like many other modern Austrians he has endorsed Hayek's last idea of competitive private enterprise currencies to allow market forces to adjust the supply of money and other financial instruments to the desire of the public to hold them. Unlike Hayek, Horwitz he wants these private enterprise suppliers to be firmly committed to convert on application any currency or deposits issued into actual stocks of commodities in specified bundles.

None of the apostles of free banking seem unable to confront the following problem. Commercial banks are not usually allowed to issue notes and coins; but these are now the small change of the monetary circulation. The English legal tender laws place no obstacle to contracts in any chosen medium - whether gold, cowrie shells, euros or in the liabilities of any financial institution. A spirited debate on whether the emergence of competitive private enterprise money, free from central bank control, is one day likely to emerge from the development of electronic transmission processes can be found in the July issue of International Finance. The result is a draw. Meanwhile the evidence suggests that in present conditions it needs a runaway inflation to induce people to abandon their official currencies.

Yet we do not have to give up all attempt at a neutral monetary policy. A starting point would be the research, stimulated by people such as de Anne Julius of the MPC, on what a normal real short-term rate of interest might be. This would have to be based on the historical trend, abstracting from crisis years or periods of high inflation. Such an estimate will almost certainly come to somewhere in the range of 3 to 5 per cent.

This is only a beginning. Faced with developments such as a hectic scramble to invest in high tech industries or a scramble for liquidity, it would be clear that the natural rate has moved. There is no avoiding judgment about how much, but it would surely be possible to do something better than just plough on regardless with rigid inflation targets.

For the time being, estimates of normal and natural interest rates are unlikely to be good enough to replace inflation targets altogether; and they will have to come in as a supplement. But anything that can replace the parrot like incantations about price stability at every European central bankers' gathering would be a move in the right direction