<<<	articles
Inflation targets lose their glamour Samuel Brittan: The Financial Times 16/01/04 The distinguished Belgian economist Paul De Grauwe wrote in the Financial Times on January 8 that "Central banking is not just about keeping inflation close to 2 per cent and praying the rest will be fine." There are many straws in the wind suggesting that establishment opinion, even in the euro zone, is gradually shifting in his direction. The new head of the European Central Bank, Jean-Claude Trichet, the same day deplored "exchange rate instability" - and not just for its possible effects on the euro zone price level. In an important speech in Leicester on October 14, the Bank of England governor Mervyn King suggested that it was "easier to measure the money value of spending and output in the economy than to split it into estimates of real output on the one hand and price indices on the other." The latest data revisions, which considerably altered the picture of real UK growth over recent years "have left estimated money spending and output broadly unchanged." There are numerous signs that inflation targets alone will not be a permanent regime like the gold standard or even the post-war Bretton Wood system. One sign of fraying at the edges is the arguments that have developed about how to measure inflation. The euro area has only managed to reach a common definition by using a harmonised index which excludes housing. Eurostat is now investigating methods of bringing owner occupier into the estimate, which could disturb the continuity of the series. The British chancellor, Gordon Brown, has moved and slightly relaxed his target from RPIX, to the UK version harmonised European index, which is now to be called the Consumer Price Index. In doing so he apparently cut inflation from 2.5 to 1.3 per cent at a stroke. Nobody believes this, but it has made it harder for the Bank to explain the likely need to increase interest rates. All this is coming on top of a long-standing critique of central bank neglect of boom and bust in asset markets - apart from their delayed effect on over-the-counter prices. A crude decade-by-decade history of macroeconomic policy in the second half of the 20th century provides some context to current issues: The 1950's. The dollar standard. Whatever governments said, policy was governed by the attempt to maintain the dollar parities of their currencies. The 1960s. Growthmanship. Governments tried to boost growth rates by expanding demand. This produced:- The 1970s. The decade of strato-inflation with price increases in double digits which governments at first tried to control by "incomes policies". The 1980s. Attempted monetary targets. Inflation did come down after painful squeezes; but the money supply proved difficult to regulate or even measure. The 1990s. Inflation targets and central bank independence came into fashion. The US Fed stood out against inflation targets, even though they have influential advocates on the Open Market Committee. With hindsight, inflation targets are best suited to a period like the 1990's, when inflation has come down towards low single figures, but there is insufficient confidence that it will be maintained there. An objective of low but positive inflation helps build public confidence in a "stability culture". Inflation targets are less suitable if rapid inflation has become entrenched, as in the 1970s. A gradualist policy of reducing inflation by say one per cent per annum could lead to a very long period of depressed activity. On the other hand shock therapy could be very painful. In practice policy relied then on unplanned shocks due to world events or policy errors to bring down inflation -sometimes called "opportunistic disinflation." Inflation targets are also less suitable when inflation expectations are low and the main fears are of stagnation and recession. There is here the much discussed danger of the "zero interest rate bound" which puts a limit to how far central banks can reduce the real price of borrowing. As both Mervyn King in his Ely Lecture of January 3 and Ben Bernanke of the Federal Reserve, in statements over several months, have reminded us, economic stimulation then requires cooperation between central banks and finance ministries, for instance for the monetary finance of budget deficits, rather than the arms-length relationship suggested by the 1990's model. During the break up of the monetary consensus of the 1980s I frequently urged a switch to the nominal GDP objectives at which Mr.King is now hinting. One virtue of such a nominal target is that when the economy is stagnating or in recession, it points decisively to expansionary policies so long as we are starting from a low inflation base. On the other hand if inflation takes off, the policy automatically acquires a restrictive bias and there can be no 1960s type "dash for growth." Above all it avoids relying on estimates of the output gap which some recent studies have shown to be almost useless in the eurozone in recent years and excessively optimistic in the 1970s when they powerfully contributed to double digit inflation.* The Fed under Greenspan can be said to have followed a rough and unannounced nominal GDP objective in that it has taken explicit account of output as well as consumer prices. Why then I have said so little on the subject lately?. Lord Keynes is popularly supposed to have said "When the facts change, I change my mind." This banal misattribution drives me up the wall. What he probably said was "When I change my mind I say so, what do you do?" The main reason why I have been relatively silent are:- When inflation targets appeared to be working there was no point in knocking my head against a brick wall. Moreover, some central bankers, such as King and Trichet have always believed that a sufficiently flexible approach to such targets could make them approximate to Nominal GDP. The promoters of any new target are likely to be crucified by the armies of short term financial analysts. This is especially true for Nominal GDP, which it would be neither possible nor desirable to monitor on a month-to-month basis but could only be sensibly tracked on a one, two or three year moving average. (And here is where I have shifted a bit.) Inflation and recession are not the only sources of instability even in western type economies. We have yet to discover a simple and readily understood formula which would enable central banks to nip in the bud "irrational exuberance" in asset and credit markets. If we look at charts of the US, the UK and the eurozone, we find that in the UK Nominal GDP has followed a stable path of 4½ to 5½ pc annual growth for the last few years. In the US. it has been in a 4 to 6pc band except for a sharp temporary dip in 2001. In the eurozone on the other hand nominal GDP growth has been in a much lower 2 to 4 pc band in the last three years, with a pronounced downward trend. If M.Trichet and his colleagues were to inject enough money to raise this to 5 per cent, I do not believe that it would miraculously transform eurosclerosis, the main reasons for which are "structural" - code for the stranglehold that union influence and inflexible labour costs impose. But it is still an olive branch which the ECB could honourably offer to its critics - and would be much better than tinkering with exchange rates. The worst that can happen if it did this is that eurozone inflation might creep up to 3 per cent for a while. I am not in favour of dropping inflation targets in which so much political and intellectual capital has been invested. But I would try to achieve them over a longer horizon and would keep a weather eye on some measure of nominal demand as well. Such guidelines would then be broad enough to allow central bankers to tighten policy in the face of incipient bubbles in the equity or property markets. And here alas I can see little alternative to relying on their discretion. Postscript on interest rates As short term interest rates are the main weapon used by most central banks in normal times to influence economic activity, they can hardly serve as a target. Nevertheless it would be a great advantage to look at current interest rates in the light of probable long term behaviour. At present short term real interest rates in the group of seven countries are around minus ½ per cent. In the US they are around minus 1 per cent; and even in Germany and the UK, where they are relatively high, they are still no more than plus 1 per cent. This compares with a normal level of say 2 or 3 per cent. The message is that interest rates are abnormally low and will have to rise unless we are entering a period of secular depression. If we look at nominal interest rates, there is a gap of around 2½ per cent between prevailing international short term rates and the much higher rate based on ten year government bonds. This gap - "upwardly sloping yield curve" -is about 2.75 per cent is also a classic sign of stimulative monetary policy. Such considerations might help to bring a long term perspective to discussions of whether officially administered rates are too high or too low and thus reduce the excessive reliance on econometric forecasts of the economy which are the bane of current central bank practice. *James Mitchell: Should we be surprised by the unreliability of real time output gap estimates? NIESR, December 2003. Edward Nelson, The Great inflation of the 70s. Federal Reserve Bank of St Louis, January 2004
<<<	articles
Site designed and managed by Andrew Heavens - aheavens@ftnetwork.com