By Michael Horrigan, President, W.E. Upjohn Institute for Employment Research
There is a lot of talk these days about the possibility—and, in the view of some, the high likelihood—that the U.S. economy will fall into a recession this year. Surveys of consumers and economists alike show significant increases in such a belief. To date, a recession has not been declared, and despite growing uncertainty around the potential effects of tariffs, deep personnel cuts to federal agencies, and prospective tax cuts, we have yet to see any significant signs of declining economic activity that might point to a coming recession.
However, if we are indeed standing at the water’s edge of a coming recession, it might be useful to remind ourselves: Just what is a recession? And what do economists look at in trying to say whether a recession has started?
If you were fortunate enough to take an economics course in high school or college, you may have been taught that a recession happens when there are two successive quarters of declining real gross domestic product (GDP). As it turns out, while most recessions do in fact share this characteristic, it is not the sole determinant of when a recession begins. In fact, it is possible to have a recession even when real GDP does not decline for two successive quarters, such as occurred with the relatively brief eight-month recession in 2001.
Rather, a recession is characterized by “a significant decline in economic activity that is spread across the economy and lasts more than a few months.” Here I am quoting the National Bureau of Economic Research’s (NBER) Business Cycle Dating Committee, which is a non-partisan group of professional economists from outside the federal government that determines business cycle turning points for the U.S. economy.
The NBER’s Business Cycle Dating Committee emphasizes a number of considerations as it seeks to establish whether a recession has begun. One of the most important factors to note is that the committee’s approach is retrospective, and it waits until there is sufficient data to justify a determination. The data the committee looks at are often subject to revisions, which is not at all unusual in the world of federal statistics.
Take, for example, the monthly change in payroll employment levels. The committee has noted that this statistic is particularly important in its deliberations. The level of payroll employment that is initially published for a given month is then revised twice, so the final sample-based estimates are released two months later. In addition, this “final” estimate is adjusted yet again in January of the following year to reflect the birth and death of firms in the prior year that were not captured in the original sample.
Overall, the committee looks at a number of different data series, each with its own particular revision cadence, so it may take time before there is sufficient confidence in the overall direction of economic data that a recession can be declared. For example, the committee declared on April 25, 1991, that the U.S. economy had entered a recession in July 1990. This announcement actually occurred after the recession had ended as the recovery had started in March 1991. The committee’s approach is intentionally conservative in its deliberations and the timing of its declarations of the start and ending dates of recessions so as to maintain the integrity with which their announcements are viewed.
Economists and policymakers, on the other hand, naturally try to identify the start of any recession well ahead of the committee. As a result, they will track the vast array of data coming out of the federal statistical system, giving particular attention to the leading, coincident, and lagging economic indexes published by the Conference Board, another non-partisan, non-governmental organization. The Leading Economic Index, which comprises 10 different data series, tends to turn down well before a recession starts. A fall in the index does not necessarily mean the economy is in a recession, however, as it often will recover a positive trend to reflect renewed economic strength.
An example of a statistic that “leads” a downturn in activity is initial claims for unemployment insurance. Increases in weekly claims over time may be a strong indicator of a contraction in business activity. Economists trying to forecast the start of a recession will not only look at a general indicator, such as initial claims, but will also look carefully at whether rising claims are specific to a particular industry or are more widespread.
The Coincident Economic Index tends to turn down at the start of a recession, making it particularly important to track. The components of this index consist of nonfarm payroll employment, personal income less transfer payments, manufacturing and trade sales, and industrial production. The Conference Board also publishes a Lagging Economic Index of economic indicators that generally turn down after the start of a recession.
The goal of this piece is to note that calling a recession is a complex endeavor. However, this begs the question, “Are we in a recession?” Are we headed toward a recession? Stay tuned for the next installment of this video series as I explore the evidence as to the current state of the U.S. economy. I’ll also discuss the importance of data in revealing how consumers and businesses are feeling about the future of the economy. That’s important, because changes in confidence among those who buy and those who sell can contribute mightily to the future and determine whether or not we go into a recession.