Comment: Chancellor pushing at the limits on base rates

It is now believed with unshakeable firmness that the Chancellor was right to reject Eddie George's advice to raise base rates in May 1995, despite the fact that, a year later, the Government's own inflation target (2.5 per cent on the underlying rate) continues to be missed by a significant margin. Admittedly, this target was intended to apply to "the end of the Parliament", but that is a detail. The events of the past year have proved that the body politic does not accept that inflation should be the only objective of monetary policy, despite the clear marching orders the Treasury has given to the Bank on this point.

If we are going to set twin objectives for monetary policy - low inflation and output stabilisation - the least the Treasury can do is to tell the Bank about it. But even in this context, is the stance of policy about right, following last week's quarter point cut in base rates? On 28 May, this column outlined the "Taylor Rule" for setting monetary policy. This rule, which is being intensively studied in the Treasury and the Bank, produces an optimal level of base rates by reference to an automatic formula linking interest rates to inflation and the output gap in the economy, not just the former. At present, the Taylor Rule suggests base rates should be 5.9 per cent (see below), insignificantly different from the Chancellor's new setting of 5.75 per cent.

However, this bald statement conveys a false sense of precision, since one of the inputs to the Taylor Rule - the output gap - can only be measured with a wide margin of uncertainty. For every 1 per cent that output falls below the normal capacity of the economy (i.e. for every 1 per cent of "output gap" in the system), the rule states base rates should be set 0.5 per cent below equilibrium. Hence a measurement error of 2-3 per cent on the output gap can lead to serious error in the setting of monetary policy.

How likely is such a large error in estimating the output gap? This was the subject of the latest report of the Chancellor's Forecasting Panel, coincidentally published on the same day as the base rate cut. The more the panel thought about the output gap, the more elusive the term became. For the majority of the panel, the preferred definition of the output gap is the difference between actual and equilibrium output, where the latter means that level of output consistent with stable inflation and a sustainable balance of payments.

The problem comes in pinning down the concept of equilibrium output in practice. In particular, what elements of the economic structure should we hold constant when seeking to measure it? Martin Weale of the National Institute of Economic and Social Research argued that, in principle, everything in the economic structure should be allowed to vary in the very long run, in which case labour productivity in the UK could be allowed to rise to the level currently implied by global best practice - say the level achieved in the United States. This would imply that the long-run output gap is enormous, since there is no God-given reason productivity in the UK should be permanently held below that in the US. But, as Mr Weale would readily admit, this is not what most people mean when they refer to the output gap. Instead, they have in mind something much shorter-term.

After debating this for quite a while, most of us on the panel concluded that it makes sense to define two separate time periods when measuring equilibrium output. A long-run equilibrium exists in which the capital stock has been brought to its desired level, and in which the level of unemployment is at the "natural" rate required to stabilise wage inflation. It may take many years for the economy to return to this equilibrium once it has been disturbed from it.

However, a shorter-term equilibrium also exists, in which the capital stock is not at equilibrium, because companies have not yet achieved their desired level of capital equipment, but in which unemployment is at the natural rate. This distinction reflects the familiar assumption in economics that the labour input can be more readily adjusted to its desired level than the capital stock.

This then leads naturally to the concept of "speed limit" for the growth in the economy. Even when output is well below its long-term trend, policymakers need to be cautious about allowing output to expand too rapidly, since companies need to be given time to increase their capital stock to its long-run desired level. Thus inflation can rise even when output is well below trend, but only if the adjustment back to trend occurs gradually.

What does all this imply for the present situation? The accompanying table shows the estimates made by panel members for the short- and long- run output gap, and the speed at which the economy can expand over the next three to five years without generating extra inflation. For all the theoretical differences between panel members, the empirical estimates for these three key concepts did not differ much. On average, the panel thought that the short-run output gap is 1.7 per cent, while the long- run gap is 3.2 per cent. This implies the economy can grow by more than 3 per cent per annum over the next three to five years while still managing to hold inflation at or close to the 2.5 per cent target. With the economy growing at an annual rate of less than 2 per cent, the panel's conclusions would appear to give the Chancellor plenty of cover for his latest decision to cut base rates. In fact, if we plug the panel's estimate for the short- run output gap (1.7 per cent) into the Taylor Rule, we get the result (quoted above) that the "correct" level of base rates today should be 5.9 per cent. But if we plug the panel's estimate for the long-run output gap (3.2 per cent) into the Taylor Rule, then we find the "correct" level for base rates is as low as 5.1 per cent. The Chancellor is probably blissfully unaware of this fact, but his remark that UK real interest rates are still much higher than those in the rest of the EU suggests he may have something similar in mind. For example, if he wants the level of real rates in the UK to come into line with Germany's, base rates would need to fall under 5 per cent.

At anything approaching that point, I would certainly part company with Mr Clarke. We should not blithely assume that the long-run output gap can be quickly closed. For short-run policy, the short-run gap is the more relevant, and on present policy settings, economic growth could easily be exceeding the speed limit by the year-end. Furthermore, we are still supposed to be following an inflation objective, not the compromised inflation/growth objective which is implicitly built into the Taylor Rule. I have no doubt that, should he be inclined to overlook any of these points, the Chancellor will be forcibly reminded of them by the Governor in future monetary meetings.