The FT led its front page on Monday with a startling headline: “Economic growth must slow, warns BIS“. That’s right, economic growth must slow. As Martin Wolf cogently argues in the FT today, simultaneous deleveraging by private and public sectors, encouraged by all wings of macro policy, could result in a prolonged slump in global demand. Yet the BIS is not alone in its thinking. More and more policymakers seem to be gravitating towards similar conclusions.
Viewed from a truly global perspective, including the booming emerging markets, this may not be such a surprising conclusion. There are some indications that the commodity price shock is feeding into wages and unit labour costs in the emerging world. Furthermore, some asset prices are in bubble-like territory as capital inflows to emerging countries continue unabated. So a recommendation to tighten policy in the emerging world is relatively uncontentious.
However, the BIS clearly believes that macroeconomic policy should be tightened in the developed world as well. Action should be taken, it says, to raise private savings ratios, and also to reduce structural budget deficits. And, for good measure, interest rates should be increased, perhaps even more rapidly than they were brought down in 2008.
At root, these conclusions are based on a super-pessimistic view of the capacity for growth in the developed world. Fortunately, in this area, the BIS spells out its view very clearly, so that it can be scrutinised by outsiders. Essentially, it believes that the supply potential of the developed world has been so badly damaged by the 2008 economic shock that there is no spare capacity left in the system – even in a country like the US, where the recovery has been both short-lived and anaemic by any past standards.
According to the pessimists, output was artificially boosted during the last boom by unsustainable growth in the financial sector and in construction, and this output has now permanently disappeared. To some extent, this is indisputable. But the critical question is, to what extent? In order to believe that there is no spare capacity left in the US and Europe today, you need to believe that the credit crunch has removed about 6 per cent from the economic capacity of these economies, in perpetuity. Note that this refers not to the actual level of GDP today, but to its potential level, now and in the future. This is stretching credulity.
The dispute actually boils down that old and treacherous issue, the estimation of the output gap. The graph below focuses on the US, where the issue is at its starkest. It shows three different measures of the output gap:
Normally, the three methods give somewhat similar results, but currently they are very different.
The simplest method is shown by the green line, which simply extrapolates the long run growth rate of the economy, and assumes that eventually GDP can return to this trendline. Although simple, this method works very well over extremely long periods of more than 100 years. It is currently optimistic, in the sense that it concludes that the economy has considerable room to expand, provided that demand grows quickly enough to exploit this potential.
The second method, shown by the blue line, is less optimistic. This is based on detailed economic estimates, prepared by the OECD, of the amount of labour and capital equipment available in the economy, along with their productivity. The OECD reckons that the 2008 shock has reduced the potential level of output by about 3 per cent, of which two-thirds comes from a reduction in capital equipment, and about 1 per cent from lower levels of participation in the active labour force. But this still leaves a margin of spare capacity which has only been exceeded twice in the last 50 years.
The third method, shown by the red dotted line, is far more pessimistic. This is based on a statistical technique called the Hodrick-Prescott filter, which attempts to separate the GDP path into its trend and cyclical components. This technique produces the very unappealing result that output has already returned to trend, leaving no scope for unemployment to fall without triggering inflation.
Although widely used, the problem with the HP filter is that it becomes highly unreliable as the estimates move towards the most recent dates, which is exactly when they are most needed. If there is a prolonged downturn in GDP lasting for several years, the HP filter will almost inevitably interpret this as a permanent decline in the GDP trend, even if this is not true, and is not validated by subsequent data.
It seems that the BIS is placing a great deal of weight on the least optimistic of the various different ways of estimating the output gap in the US (and also in Europe, where the figures are fairly similar). Indeed, its austere recommendations about monetary policy are hinged largely on this. The BIS economics staff explain their thinking on the output gap in more detail in this research report by Petra Gerlach, which is fairly balanced, and admits that estimates based on HP filters are unreliable on recent data. Yet the annual report has chosen to draw some strikingly firm conclusions about policy based on what seems to be very debatable evidence.
For those who are allergic to the output gap, of which there are many, it is worth adding that direct measures of spare capacity in the US indicate that the economy is nowhere near its full potential. Unemployment, on most estimates, is 3 or 4 percentage points above the natural rate. Capacity utilisation is abnormally depressed. And wage inflation is extremely subdued, giving no indication that the labour market is tight.
On all of these grounds, tighter monetary policy in the US would make the global predicament worse, not better.
