Why base rates should rise again on Friday

"Part of the capacity problem could be alleviated by an investment boom, which should be forthcoming in the next two years"

A week ago, the case for a further rise in base rates was looking rather shaky. Statistics on both industrial production and retail sales were suggesting that GDP growth had ground to a halt in the first quarter of the new year, and the remaining members of the dwindling monetarist fraternity were warning of over-kill on monetary policy.

But that all changed last Tuesday morning, when the GDP figures showed that economic growth remains well above trend, and the CBI survey reported a further alarming rise in capacity utilisation. A rise of at least half a point in base rates now seems both necessary and likely within a week.

The problem is that the economy appears to be using up its margin of spare capacity much more quickly than it did in the 1980s' recovery. In terms of chronology, the present upswing has reached a point equivalent to 1984 in the Thatcher "boom". Since the economy did not overheat until 1989, it is tempting to conclude that rapid growth can be maintained for several more years without much danger. But that would be a big mistake.

This whole issue revolves around estimates of the output gap, which are much quoted, and much abused. One way of estimating the gap between actual and trend output is simply to fit a "main trend" through the level of output, "benchmarking " the trend on the mid-cycle level of output in past periods. For example, the CSO cyclical indicators show that output was at mid cycle in the 1990 fourth quarter. If we assume that trend GDP has grown at its post-war average rate since then (roughly 2.25 per cent per annum), the level of GDP in the 1995 first quarter turns out to be 3 per cent below trend.

In order to err on the safe side, we should probably assume that the initial estimate for GDP for the two or three quarters ending in first- quarter 1995 is likely to be revised upwards as further information comes in. A plausible guess might be that this will increase estimated output by about 0.5-1 per cent, in which case the current output gap is cut to 2-2.5 per cent.

The implication of this simple calculation is that the economy could grow by around 3 per cent per annum for another three years before the output gap would be eliminated. Furthermore, we get some superficial support for this guesstimate from the behaviour of the economy in the last cycle. In the course of the 1980s boom, the output gap fell to 2-3 per cent of GDP in 1985. Since inflation pressures in that upswing started to rise only in 1988-89, this might suggest that we have about three more years of above-trend GDP growth before needing to worry too much about a significant upsurge in inflation.

However, this conclusion may prove altogether too reassuring. It does not take any account of the possibility that recent growth in supply potential may have been different from that in previous upswings, including that of the early 1980s. This is in fact suggested by the behaviour of the labour market and by the recent uptrend in capacity utilisation, both of which indicate that there is less spare capacity in the system now than in 1985.

First, the labour market. The potential labour supply is rising much less rapidly, for demographic reasons, than it did in the early 1980s. In the first three years of the current upswing, the rise in the potential labour supply was 475,000 less than in the equivalent period in the 1980s. This represents a difference equivalent to 1.8 per cent of the labour force, which would imply a shortfall in potential output of 1.2 per cent.

This partly explains why unemployment has dropped much more rapidly in the 1990s than it did in the 1980s. As a result, some indicators of labour market slack suggest that labour market conditions are more similar to 1986 or 1987 than 1985. In other words, the labour market seems to be nearer the overheating zone than the simple estimate of the output gap quoted above would imply.

Turning to the margin of spare plant capacity, there has been no rise in fixed investment in the present recovery to compare with the surge that was triggered by the Lawson reforms to corporation tax in the 1984 Budget. As a result, the capacity constraints which are now visible in the CBI and other business surveys have risen very sharply, and are touching levels only observed in 1988, the very peak of the last boom.

So there may well be much less spare capacity in the economy than is implied by taking the gap between the latest estimate of GDP (probably too low), and the historic trend for GDP (which may represent an overestimate of present capacity). But there is an alternative method of estimating the output gap, directly utilising the quantities of labour and capital available to the economy.

In order to do this, we make the assumption that unemployment is at its "natural" rate, and that capacity utilisation is at its average rate for the cycle. We then estimate an equation linking output to the available inputs of labour and capital. The result is that GDP in 1995 first quarter is approximately 0.7 per cent above the normal capacity of the economy estimated from this rough and ready production function.

This analysis therefore implies that the existence of any output gap in the economy at present is dubious. The long-term trend for GDP suggests that the gap may be around 2-2.5 per cent, but that estimate implicitly assumes that the availability of labour and capital has been increasing since 1990 along its normal trend line. In fact, there seems to be some evidence that this has not been the case, and the production function approach, using estimates of the amount of labour and capital actually available, indicates that output is in fact fractionally above "capacity".

Part of the capacity problem could of course be alleviated by an investment boom, which should be forthcoming in the next two years, but companies badly need time to install the capital equipment necessary to head off supply shortages. A respite in demand growth would be extremely welcome in 1995-96.

In 1985, roughly equivalent to 1996 in terms of simple calendar comparisons between this recovery and the last, there was a fortuitous slowdown in GDP growth, partly triggered by the confidence effects of the miners' strike, and this allowed time for the capital stock to catch up with the growth of output. Capacity utilisation dropped markedly in 1984-85, scotching the build-up of inflation pressures and, in the end, elongating the upswing by about two or three years. A similar growth pause would be extremely desirable in the next two years while investment picks up.

But even with a very strong investment surge, it is unlikely that the supply capacity of the economy can expand by much more than about 3 per cent in a year. So in order to allow capacity utilisation to drop back from recent extreme levels, policy should aim to slow the growth rate to around a 2 per cent annualised rate for the next two years.

Since the necessary slowdown looks unlikely, a further monetary tightening is needed. This Friday's meeting of the Chancellor and Governor would be a good time to start.