Who will pick up the thread after the great unwinding?

By Martin Wolf

Is the US going to experience a recession? Two answers must be given to this question: nobody can be sure; and it does not matter. A much more important question is whether the US economy continues to experience a “growth recession”, by which is meant a lengthy period of sub-trend growth. The answer is that it will.

The standard US definition of a recession is two quarters of negative economic growth. This demands both too much and too little: too much because it requires an absolute fall in output, which is an infrequent event in a growing economy; too little, because it is consistent with rising unemployment and declining capacity utilisation. But a lengthy growth recession is likely to be far more disturbing even than a sharp recession, provided the latter ends swiftly.

Most analysts believe that the trend rate of growth of the US economy is around 3 per cent a year. Growth at below that rate, then, is a growth recession. This year, the expectation is for growth of about 2 per cent. Next year, suggests the consensus, it will be a little above 2 per cent. That would mark a cumulative shortfall of about 2 per cent of gross domestic product over two years. So the US is already in a growth recession.

It is easy to see why US domestic demand should continue to be weak: house prices have fallen by about 8 per cent from their peak, in real terms (on the Case-Shiller index) and may fall far more, after their doubling between 1997 and 2006; Ben Bernanke, chairman of the US Federal Reserve, has put unexpected write-offs from the bad mortgage lending at $150bn, up from the $50bn to $100bn he estimated in the summer, though well below the $400bn estimated by some observers; and as bank capital shrinks, credit is likely to remain tight.

Yet, however willing it might be to help, the Fed is concerned over inflationary pressures. Justifying this concern are the record prices of oil and the tumbling dollar, now lower against the world’s important currencies than it has ever been.

What happens now depends on two things: how weak domestic demand turns out to be, and the extent to which any shortfall is offset by a stimulus from net exports. The latter is “the great unwinding”: the re-import by the US of the stimulus it imparted to the rest of the world between 1996 and 2004, when its domestic purchases grew faster than GDP and the current account deficit exploded upwards.

As is well known, the US was the principal absorber of the surplus savings of the rest of the world, principally in east Asia and the oil exporting countries. Somewhat less appreciated is how this has worked domestically. In the early 2000s, the principal domestic absorber of this glut of external credit was the government. Since then it has been the private sector, which ran a financial deficit (excess of spending over income) that peaked at 3.4 per cent of GDP in the third quarter of 2006.

The latter deficit was entirely contributed by the household sector, whose deficit reached 3.8 per cent of GDP in the second quarter of 2006. Such a deficit can be financed only by selling financial claims. In the case of households, these claims took the form of rising debt, principally against the rising value of houses. That has now come to an end.

Yet the unwinding is still modest: US households still ran a financial deficit of 2.3 per cent of GDP in the second quarter of 2007. Moreover, household savings rates remain very low, at 2.5 per cent of GDP in the second quarter of 2007. Further correction in both is probable.

So what might offset such a slowdown in spending by the household sector? Normally, businesses do not invest more when the economy is weak, even if their financial position is healthy. A bigger government financial deficit is likely, as a result of the slowing of the economy. But an aggressive fiscal boost is improbable. That leaves net exports.

As Wynne Godley of Cambridge university and co-authors point out, a sustained improvement in US net trade will offset at least a part of the likely sluggishness in domestic demand.* This is why the US authorities talk about a strong dollar, but do not mean it. They want a retreating dollar, but one that does not turn into a rout.

Already net exports contributed a quarter of US growth between the first three quarters of 2006 and the first three quarters of 2007. Without it, growth would have been only 1.5 per cent, instead of 2 per cent. But exports are only some 12 per cent of GDP. They must grow by considerably more than 10 per cent a year, in real terms, if the contribution of net trade to the rate of growth is to be as much as 1 percentage point. It is likely to be much less.

A plausible view of the future, then, is that the US will experience a lengthy period of sluggish growth in domestic private demand, partially offset by fiscal expansion and an improvement in net exports. It is via the latter effect, moreover, that monetary policy should have its principal impact, since households are unlikely to borrow much more while their houses decline in value.

This is the great unwinding. So what does it mean for the rest of the world? It means that the rest of the world will adjust either by increasing demand, relative to potential supply, or by reducing its supply relative to demand. The former adjustment is clearly the more desirable.

Will it happen? The good news is that the build-up of foreign currency reserves and accompanying desire to prevent currency appreciation, by keeping interest rates down, are themselves expansionary. The resulting “overheating” is part of the solution, not part of the problem. If this overheating becomes bad enough, governments may allow still faster currency appreciations: perhaps even the Chinese will finally realise the error of their interventionist ways.

The great unwinding is a turning- point for the world economy. The rest of the world – and the emerging markets in particular – must now become the demand engines of the world economy. Will they do so? This is the big macroeconomic question to be answered over the next few years.