The recession isn’t over till they say it’s over. (But who are they?)
By Neil Irwin
One year ago, a committee of economists declared that the pandemic had officially caused the United States to fall into a recession. So is it over yet?
It might seem a simple question — and yet the committee still doesn’t have an answer. That’s because it’s more of a head-scratcher than it might seem. And the issue raises some other weird questions, including: What is a recession?
Let’s back up. America’s semiofficial arbiter of these things is a committee of the National Bureau of Economic Research, a private organization based in Cambridge, Massachusetts.
Called the Business Cycle Dating Committee, it consists of eight esteemed academics who specialize in macroeconomics and business cycles.
Their definition of a recession has been “a significant decline in economic activity that is spread across the economy and that lasts more than a few months.”
Economists such as those on the committee use the term recession to refer only to the period in which economic activity is contracting, not to the entire period of bad economic times. That’s why the previous recession was ruled to have technically ended in June 2009, even though the economy still felt terrible to people during the sluggish recovery that followed.
Which brings us to the pandemic recession. There’s little argument that economic activity peaked in February 2020 and contracted in March 2020. The United States went through a period of rapidly collapsing economic activity with no modern precedent. Travel-related industries were most heavily affected, but sectors as varied as manufacturing, retail, construction and health care also took a hit.
“Significant decline in economic activity.” Check. “Spread across the economy.” Check. Ah, but now consider “Lasts more than a few months” — what about that?
It now looks as if economic activity bottomed out sometime in April, maybe six weeks or so after the February peak. The committee says it weights two monthly data series particularly heavily: personal income excluding government transfers, and payroll employment. Both of these were higher in May than in April, providing evidence that April was the trough.
That’s not to say that the economy was in very good shape for the remainder of 2020. The personal-income measure surpassed its pre-pandemic level only last month, and employment is still well below pre-pandemic levels. But the direction of change has now been positive for more than a year.
Robert Hall, a Stanford economist who chairs the committee, declined to comment Thursday on when a ruling on the recession end date might come. But let’s imagine that the end date winds up being April. If the economic peak of the previous expansion was February 2020 and the trough of the recession was April 2020, then it really lasted only two months. Or even less, if you believe the rebound actually started in mid-April.
Two months are not “more than a few months.” The previous shortest recession on record was the one that began in January 1980 and lasted six months.
In effect, the committee’s judgment is that even if the pandemic recession did not fit the usual definition of a recession, it still was one.
In its own words: “In the case of the February 2020 peak in economic activity, we concluded that the drop in activity had been so great and so widely diffused throughout the economy that the downturn should be classified as a recession even if it proved to be quite brief.”
Hall declined to comment beyond that, saying that further elaboration on the jointly approved written comments would need to be approved by the committee.
But it’s easy to imagine why the committee ended up with that conclusion. It would be awfully pedantic to refuse to classify an enormous economic contraction as a recession just because it didn’t fit a somewhat vague definition that a few economists had written down.
“I think way back in March of 2020 there was a question of: ‘Should we call this a recession?’” said Tara Sinclair, an economist at George Washington University who studies business cycles. “Or if this is something that is going to last a few weeks, then the economy bounces right back, should it be treated as a recession or as the equivalent of a natural disaster?”
Waiting to call an endpoint made sense, she said, as the committee was watching for evidence of a second wave of virus outbreak that might cause the economy to tumble again.
In fact, a wave of infections in the fall dragged down employment numbers for a single month, but by most evidence it did not cause a broad or sustained contraction in economic activity.
But “at this point, they are taking a particularly long time to call a particularly clear trough,” Sinclair said.
There’s one more wrinkle that shows how the pandemic recession is a weird one. Another common definition of recession, used especially widely outside the United States, is two straight quarters of contraction in gross domestic product, or GDP.
Even though the actual economic contraction lasted only a few weeks in early 2020, it appears in the GDP tables as having stretched over two quarters. The economy was shutting down in mid-March severely enough to cause the economy to shrink at a 5% annual rate in the first quarter that ended March 31.
Then the continued collapse of the economy into early April meant that the second quarter, which began April 1, recorded a further 31% rate of shrinkage. If the pandemic had started at the beginning of a quarter rather than the end, the data would most likely have shown only a single quarter of declining GDP.
The exact start date and end date aren’t of great importance, of course, unless you’re a chart maker focused on where the gray bars should go in an economic data visualization, or a politician looking for talking points on the campaign trail. What matters for people is how long and how severe the bad times turn out to be.
But if nothing else, what appears likely to be the shortest recession on record shows just how odd the pandemic economy has really been.