The United Kingdom is one of the first countries to report output data for the third quarter of 2009. Other countries will report shortly. The fall of 0.4 per cent in GDP announced by the Office for National Statistics on Friday 23 October was, for some, a shocker.
It is the sixth quarter in a row that we have seen GDP fall, the first time this has happened since record-keeping for this measure began, 60 years ago. Output has now dropped by 6.1 per cent from its peak. Contraction in services has been around 4.6 per cent. Though substantial, this is still less than the 14.7 per cent drop in manufacturing output and 15.6 per cent in construction. These drops suggest that any possible improvement in the labour market situation is a long way off.
To put the drop in output in context, it is less than that experienced by Finland, Germany, Ireland, Japan, Spain and Sweden, but more than in Denmark, the Netherlands and the US, for example. The countries that have experienced the biggest drops in output have been those most exposed to world trade (Germany, Japan and Sweden), those that have experienced rapid increases in house prices (Ireland, Spain and the UK) and those with large financial sectors (the UK and the US). The UK was particularly exposed to this global shock because of its relatively large financial sector and substantial house-price bubble, considerably larger than was observed even in the US.
Blurring past and future
The news on Friday caused consternation among City economists and commentators, especially those who had been seeing green shoots all round and had expected positive growth of at least 0.2 per cent. I was especially taken by a comment made by Kevin Daly of Goldman Sachs, reported in the Financial Times. He argued that preliminary GDP figures for the past decade contained "no statistically useful information about growth".
That seems too strong. I think there is every prospect that the figure will be revised down rather than up, as has happened in previous quarter releases in this recession. But what has happened is a good illustration of the "Gieve law", which operates in a downturn, and is named after my friend Sir John Gieve, a former deputy governor of the Bank of England. The Gieve law goes like this. The most recent data release is considered to be the worst it is going to get, so this is the trough until the next data release, which may be worse again. So we have reached a new trough, until the next release when the data is worse, and so on.
The Monetary Policy Committee of the Bank of England, in its most recent forecast, wrongly predicted that there would be positive growth in the third quarter of 2009. The consequence of this is that the MPC will have to introduce more quantitative easing at its next meeting, if it is to have any hope of getting inflation back on target within the period it forecast.
So what is going on? To me, it is clear that all of these forecasters continue to make the error typical of economists in a recession, of extrapolating from the relatively benign recent past to make projections. The result is that their forecasts tend to be overly optimistic. In fact, the latest fall in GDP should not have been that much of a surprise, given that banks aren't lending and few firms are investing or hiring. What was a surprise was that most commentators were surprised.
The question everyone is asking is: how long will this recession go on for? The chart (above) presents trends in this and previous recessions, based on data kindly provided to me by Martin Weale of the National Institute of Economic and Social Research (NIESR). What it shows is that the drop in level of output is greater than observed in the recessions of the 1970s and 1990s, but broadly comparable to that observed in the 1980s. It is not as deep as the 1930s.
What stands out from the chart is just how long it can take to emerge from a deep recession. In both the 1980s and the 1930s it took
just over four years for output to return to pre-recession levels. But unemployment is another matter altogether. Unemployment takes a long time to recover in recessions, hence the claim that it is a "lagging indicator". For example, the unemployment rate was 5.3 per cent in July 1979 but didn't get back to that level again until June 2000, more than 20 years later. At its high point, unemployment reached 11.9 per cent in April 1984 but then fell to a low of 5.2 per cent in April 2008; it stands at 7.9 per cent at present. If the unemployment rate were to hit 11.9 per cent, that would imply that 3.7 million people were out of work.
Important research by Carmen Reinhart and Kenneth Rogoff*, who have studied a large number of recent financial crises around the world, shows that recessions have "deep and lasting effects on asset prices and employment". The effects include the following. First, the real house price declines by an average of 35 per cent stretched out over six years, while the equity price collapses by 55 per cent on average over a downturn of about three and a half years. Second, the unemployment rate increases by an average of 7 per cent over the down phase of the cycle, which lasts four years on average. Output falls (from peak to trough) by an average of 9 per cent-plus and lasts on average two years. Third, the real value of government debt tends to explode, rising by 86 per cent on average; certainly this has been the case during the major recessions since the Second World War. The collapse in tax revenues in the wake of deep and prolonged economic contractions is crucial in explaining the large budget deficits and increases in debt that follow a crisis.
Any recovery rarely runs smooth. We may yet see one or more negative quarters of growth before this is all over. My advice to the Chancellor would be to postpone the proposed VAT increase, from 15 to 17.5 per cent, due in the New Year, for at least another six months. This is no time to raise taxes.
David Blanchflower is professor of economics at Dartmouth College, New Hampshire, and at the University of Stirling