On July 28, the United States Bureau of Economic Analysis (BEA) will release its advance estimate of second-quarter GDP growth. The impending announcement has observers on the edge of their seats, with many expecting it to confirm that the US economy slid into recession in the first half of 2022. But even if the announcement seems to say that, the reality is more complicated.
The recession prediction is based on two assumptions: first-quarter growth was negative, and a recession is defined as two consecutive quarters of negative growth. As a result, if second-quarter growth is estimated to have been negative, the stock and bond markets could react by rising in the very short run. A recession might lead investors to believe that the US Federal Reserve will ease up on its aggressive interest-rate hikes.
But there are three major flaws to this reasoning. First, growth is as likely to have been positive as negative in the second quarter. Yes, the Atlanta Fed’s GDPNow model estimates a second-quarter annual growth rate of -1.5%, based on data available through July 15. Nonetheless, some economists—including me—would argue that growth was more likely positive in the second quarter.
Even if the BEA estimate is negative, however, it does not necessarily mean that the US has entered a recession. That is because—and this is the second flaw—a US recession is not defined as two consecutive quarters of negative growth.
True, the two-consecutive-quarters rule is used to determine whether most advanced economies—particularly in Europe—are in recession. But it is not the main criterion in all countries. It certainly is not in the US, where the Business Cycle Dating Committee of the National Bureau of Economic Research makes that call, based on a variety of indicators—a role that the BEA officially recognises. (It is worth noting that institutions that are private nonprofits, like the NBER, also produce other important economic indicators, such as the consumer confidence index and the purchasing managers’ index.)
The NBER’s approach arguably produces more accurate assessments than the simplistic two-consecutive-quarter rule. This was demonstrated, for example, by the recession of 2001, an episode that would fail the two-quarter test, because GDP growth was negative in the first and third quarters of that year, but positive in the second quarter. But if one looks at a variety of indicators—especially employment—it is clear that there was indeed a recession. The NBER recognised that.
It is still natural to view output growth as the most important indicator of a recession. But even if one operated according to that two-consecutive-quarter rule, there is a third flaw in the assumption that the US entered a recession in the first half of 2022: contrary to popular belief, first-quarter growth was not necessarily negative.
There are two ways to measure output. The one that gets all the attention in the US is GDP, which is measured on the product side—that is, by adding up the sectors in which goods and services are sold. Using that measure, annual US GDP growth was negative in the first quarter, at -1.6%.
But growth can also be measured by gross domestic income, which is calculated on the income side—that is, by adding up kinds of income, like employee compensation. In theory, the two measures should be precisely equal. But in practice, there is a statistical discrepancy—one that, in the first quarter of 2022, was large, with GDI growing by 1.8%. The average of the two measures—sometimes called gross domestic output (GDO)—was also positive.
Here comes the important part. Experts—including in the US government—have long viewed GDI as just as informative in measuring economic output as GDP. So has the NBER committee, which considers GDO in determining quarterly turning-point dates.
This has two implications for the upcoming BEA release. The first is that, even if the BEA shows GDP growth to have been negative for two consecutive quarters, the NBER committee is very unlikely to conclude that a recession started in the first quarter of 2022. That conclusion would be too far out of line with GDI, employment growth, and other economic indicators from the first quarter.
Second, the GDP figures will soon be revised. This is routine, and revisions are substantial: the mean absolute revision for a given quarter—even just going from the third to the final BEA release (after mid-year “benchmark revisions”)—is 1.2 percentage points, for a sample ending in 2018. This is the main reason why the NBER committee waits so long—11 months, on average—before calling turning points.
When the BEA undertakes its comprehensive “benchmark revision” of the National Income and Product Accounts—this year, the results are set to be released in September, instead of July—it could well revise first-quarter GDP growth upward, conceivably even by enough to turn it positive. Revisions historically have moved GDP in the direction of GDI more often than vice versa. In that sense, GDI may be a more reliable measure of domestic output than GDP.
Brace yourself for headlines claiming that the US economy is in recession, with all the public and market reactions that will trigger. But do not be surprised if you are told the opposite two months later.
The author is Professor of capital formation and growth at Harvard University
Copyright: Project Syndicate, 2022