Most postwar recessions did not exceed 10 months

The recent national payroll employment report for April, which showed a decline of 20.5 million jobs and a 14.7% unemployment provides additional evidence a recession is underway.  Coupled with peaks in other coincident indicators and a 1Q-2020 GDP contraction, it appears the macro-economy peaked in February, ending the nation’s longest expansion.

A formal announcement by the National Bureau of Economic Research’s (NBER) Business Cycle Dating Committee is unlikely for months. A 2008 memo explains, “The committee’s approach to determining the dates of turning points is retrospective. We wait until sufficient data are available to avoid the need for major revisions. In particular, in determining the date of a peak in activity, and thus the onset of recession, we wait until we are confident that, even in the event that activity begins to rise again immediately, it has declined enough to meet the criterion of depth. As a result, we tend to wait to identify a peak until many months after it actually occurs.”

The NBER panel didn’t announce the start of the Great Recession (December 2007) until a year later in late 2008 The trough–the start of a new expansion–occurred in June 2009 but was not announced until September 2010.

Last September, I noted “more attention is usually lavished on the (monthly) unemployment rate,” despite its status as “a lagging indicator, not a good measure of the economy’s current status.” By contrast, payroll employment “is one of four coincident, or real-time indicators followed by markets.  The others are real income less transfer payments, manufacturing and trade sales, and industrial production.”

Industrial production measures the physical output of the nation’s mines, factories and utilities, and is an excellent indicator for understanding cyclical domestic industries such as autos and steel. Economist Ludwig von Mises, in his 1949 treatise Human Action (Yale University Press) observed, “The editors of the financial and commercial chronicles were right when–for more than a hundred years–they looked upon production figures of these industries as well as of the construction trades as an index of economic fluctuations.”

Federal Reserve data shows industrial production peaked in December 2018, the same month that durable goods also peaked. But the other three coincident indicators and GDP grew until 1Q-2020. U.S. Bureau of Labor Statistics records suggests payroll employment peaked in February, followed by a preliminary loss of 870,000 jobs in March and April’s decline. Income and sales indicators also appear to have peaked in 1Q-2020.

National authorities have responded to the economic damage left in COVID-19’s wake with an abundance of monetary and fiscal measures.

Some find it counterintuitive to consider the possibility of a recession when the economy is expanding. Likewise, how to contemplate – given current economic conditions – that this recession will end? The NBER’s chronology records 11 business cycles in the postwar era.  The average contraction lasted 11.1 months from peak to trough. Three lasted more than a year: 1973-75, 1981-82, and 2007-09.

By contrast, six lasted less than the average, ranging from six to 10 months. These occurred in 1953-54, 1957-58, 1960-61, 1980, 1990-91 and 2001. The shortest (January to July 1980) lasted six months.

At the national level, stability will first emerge among several coincident indicators. In Arkansas, a handful of counties generated above-average employment or income growth in the last recovery and can be expected to lead again.

Editor’s note: Greg Kaza is executive director of the Arkansas Policy Foundation, a think tank founded in 1995 in Little Rock. The opinions expressed are those of the author.