Ever since the Chancellor sprang his first surprise, critics of Bank of England independence have been muttering that it is a recipe for disaster putting responsibility for meeting the inflation target in the hands of the central bank - for heaven's sake, it's full of inflation hawks, they argue. Chief economist Mervyn King is seen by these parties as a particular killjoy, probably because it is his duty to present the Inflation Report, which has tried to explain why growth of 1 per cent a quarter will eventually lead to higher inflation.
The critics are a menace. They are like the kind of youths who race past you in a souped-up old Ford Escort with go-faster stripes towards the lights just as they are turning amber. They want to keep their foot down on the accelerator but will end up having to slam the brakes on at the red light.
The yob element among the commentators means the Bank will have to carry on explaining, slowly and patiently, why it is better to run a "stitch in time" policy. It is not that Mr King and his colleagues are "sado-monetarists" who want to keep the nation in permanent recession to guard against a minute risk of inflation. Rather, they understand, as the hot-headed hedonists do not, that it is better for business and consumers to put up with a handful of early small interest rate increases than face much bigger rises a bit too late. It might mean pounds 10 or pounds 20 a month extra on the mortgage now but will save home-owners thousands of pounds over the life of the loan.
There is new evidence for the benefits of running a cautious and stable macroeconomic policy in research published recently by the International Monetary Fund*. It shows that a lot of unemployment is created when the monetary authorities have to engineer a recession to reduce inflation when the economy overheats. This exceeds the unemployment that remains if the economy is prevented from growing enough to reach the "non-accelerating inflation rate of unemployment" or Nairu.
In other words, it makes sense not to squeeze unemployment absolutely as low as it can go without triggering inflation because the costs of going too far are so high. In the words of the authors: "There can be significant gains from preventing an overheating of the economy." There is an asymmetry in the results of monetary policy that will bias a competent central bank towards caution. "A macroeconomic policy that avoids boom and bust cycles can in fact raise the average level of employment and output."
To understand this it is helpful to go back to the original relationship between inflation and unemployment, known as the Phillips curve after the economist who devised it in 1958. Plotted as a graph, this curve slopes down - the lower unemployment, the higher is inflation.
Most economists believe that in the long run, there is no trade-off between inflation and unemployment. If the government stimulates the economy there will be a temporary gain in jobs at the price of higher inflation. But as people adjust to the increase in inflation and bid for higher wages to compensate, unemployment will rise again and the economy will be back to the same jobless rate but a higher inflation rate.
There is a short-run move along a Phillips curve, but the curve will shift out. The long-run Phillips curve will be vertical at a rate of unemployment - the Nairu - which is determined by supply conditions like the degree of flexibility in the jobs market, the productivity of the workforce, the availability of unemployment benefit and so on.
The development of this view explains the growing policy emphasis on controlling inflation that has emerged in most OECD countries since the late 1970s. The new IMF research emphasises that not only is there no long-run trade-off, but there is also a good reason to avoid trying to exploit the short-run trade-off. Unemployment will be lower the less variable the economic cycle.
For most economists have treated the Phillips curve as a straight line: the inflation cost of reducing unemployment is set equal to the inflation benefit of increased unemployment. In fact, there are good theoretical reasons and empirical evidence for regarding it as a convex curve, where the rise in unemployment needed to reduce inflation by 1 percentage point is bigger than the fall in unemployment achieved by letting inflation rise by 1 percentage point.
The paper shows that experience since the early 1970s supports this view. It could be caused by, for example, the tendency for bottlenecks to emerge in some segments of the jobs market which would prevent wages from falling below a certain floor. The economy will tend towards a normal - or "natural" - rate of unemployment which will be higher than the Nairu if demand is volatile. Booms will trigger wage and price inflation quite quickly, whereas busts will not achieve a big reduction in wage and price inflation as unemployment rises because some bits of the jobs market will still have bottlenecks.
It is not clear quite how big the costs of a boom and bust policy might be, but the paper concludes that unemployment will certainly be higher. The Bank of England's preference for a stitch in time, expressed again by Eddie George yesterday, will help to reduce the unemployment rate on average, even if it prevents as big a drop in joblessness as could be achieved right now.
It must be said, too, that the Bank is not unduly gloomy about the inflationary dangers. The forecast it presented in last week's Inflation Report, which made the technical assumption of unchanged interest rates, was lower than many independent forecasts, which do assume there will be further rate rises.
At long last, interest rate policy is in the hands of a steady, middle- aged driver with his foot touching the brake, his eyes on the amber light and nothing to distract him. As long as they go ahead as billed, Mr Brown's reforms will yield lower inflation, lower interest rates and lower unemployment for years to come. The Chancellor has decided that the classic British handbrake turns are not for him. Thank goodness the Bank of England is not keen on them either.
*'Phillips Curves, Phillips Lines and the Unemployment Costs of Overheating', Peter Clarke and Douglas Laxton, IMF Working Paper February 1997.