Is the United States in recession? If one looked solely at the adverse shocks that have hit the economy over the last year, one would infer an unusually high probability of a recession. If one consulted some of the most import economic measures over the last year, one would say we clearly entered a recession last January. If one gauged the popular mood, one would hear, “Of course we are in recession!”
The one criterion that has been missing is the one criterion that people most commonly have in their minds as the definition of a recession: two consecutive quarters of negative growth. This morning, October 30, the Commerce Department released the preliminary estimate of GDP in the 3rd quarter. It showed a decline. The decline was small: just 0.3 per cent at an annual rate, and it is only quarter. But at this point there can be little doubt that we are really truly in recession.
The adverse shocks include the most severe housing bust in more than 70 years, an oil shock as big as those of the 1970s, the greatest financial crisis since the Great Depression, and the worst fiscal outlook ever. Any one of these developments would normally be enough to send an economy into recession. Leading economists from Martin Feldstein to Larry Summers have since the start of the year been warning that the downturn has arrived.
And sure enough, many of the most reliable statistical indicators have suggested all year that we are in recession.
The most important statistical criterion besides GDP is employment. Jobs peaked in December 2007 and have declined steadily ever since. The cumulative loss is 760 thousand (or 0.55%) as of September. My personal favorite among indicators is Total Hours Worked in the economy, because it combines both employment (number of people working) and average length of workweek (are they working 40 hours a week? Overtime? Part-time?). Total Hours Worked shows a similar pattern as employment, but with an even steeper decline since December: 1.4%. (The Bureau of Labor Statistics is the agency that releases these numbers, on the first Friday of the subsequent month.)
The index Leading Economic Indicators, which is designed to try to warn of turning points in advance, turned down more than a year ago. Not only that, but also the index of Coincident Economic Indicators, which is supposed to move contemporaneously with the real economy, appears clearly to indicate that a recession started toward the end of 2007.
Housing prices as of August are down 27%, relative to their peak in July 2006 (Case-Shiller composite of 20 cities). Consumer confidence, an important determinant of household spending, fell to an all-time low in September, according to the October 28 release from the Conference Board. The version collected by the University of Michigan is also looking quite bleak. Retail sales are down, especially autos. The trend in industrial production has been downward for a year, and accelerated in August and September. Corporate profits are down.
But it is still not yet officially a recession! Why not? The most important criterion for dating business cycles is real growth. The rate of change of real GDP, surprisingly, was above zero in the first quarter of 2008, and was even moderately strong in the second quarter: 2.8%. (The revised “final” estimate of GDP in the fourth quarter of 2007 did turn out to be below zero, but just barely.) It is quite a mystery why output pointed up during the first half of the year, while everything else pointed down. Clearly the demand for US goods received some boost in the 2nd quarter from tax rebates and exports, both of which are expected to diminish subsequently.
But perhaps there is some measurement problem with GDP. Gross National Income (GNI) has as much claim to measure growth as Gross National Product does. In theory the two are supposed to be virtually the same: the value of goods and services sold is conceptually the same as the value of income earned. Real GNI did in fact turn down in the 4th quarter of 2007 and the first quarter of 2008, though it rebounded in the third quarter as real output did. Real personal income – one of the indicators that the NBER Business Cycle Dating Committee looks at – has been declining almost throughout the year.
The weight of evidence is overwhelming: we are currently in recession.
Did it start at the end of 2007, when employment and the other indicators peaked? Or was the stimulus from the government and from exports enough to hold off the turning point, and did the recession thus only start towards the end of the summer, when the financial crisis intensified very sharply? I am afraid that we need to wait for some more data and some more (regularly scheduled) revisions before we will know.