So what, exactly, is a recession anyway?

A common rule of thumb is that two consecutive quarters of declining gross domestic product counts as a recession. The U.S. could already be on that path: GDP in the first quarter looks likely to have been weak, and when it gets reported at the end of this month there is a chance it might have even shrunk. With fallout from tariffs hitting, GDP in the second quarter could be challenged.

But GDP isn’t the criterion used by the National Bureau of Economic Research, the longstanding arbiter of U.S. recessions among economists, government officials, policymakers and news organizations, including The Wall Street Journal.

NBER recession dates are determined by its prosaically named “business cycle dating committee," a group of eight academic economists, some of whom have been members of the committee for decades. What they look for in order to make a recession call is “a significant decline in economic activity that is spread across the economy and that lasts more than a few months."

The main indicators they watch are employment, inflation-adjusted (or real) personal income, real consumer spending, real manufacturing and trade industry sales, and industrial production.

Some of those measures have softened in recent months. Real spending in January and February was soft, and trade and manufacturing sales in January—the most recent available data point—were down from a month earlier. Meanwhile, the economy has kept on adding jobs this year, incomes are up, and industrial production has risen. By these measures, at least, there is no indication that a recession has begun.

While the NBER does pay attention to GDP, that measure has drawbacks as a recession indicator. One is that it can be pushed around by swings in inventories and trade. Those factors, along with declines in government spending as pandemic relief efforts tapered off, were in play when the Commerce Department reported in July 2022 that GDP had contracted for two consecutive quarters in the first half of that year, even as the U.S. added millions of jobs and consumers kept spending.

Another problem: GDP can be subject to substantial revisions, often long after the fact, as 2022 also showed. The Commerce Department that July reported that second-quarter GDP had contracted 0.9%, at an annual rate, from the previous quarter. The next month it would change that to a 0.6% contraction. This past fall there were more revisions, and now GDP in the second quarter of 2022 shows a narrow gain of 0.3%.

The main numbers the NBER looks at are subject to revision, too, which is one reason why the NBER tends to take its time when making recession calls.

A ‘Now Hiring’ sign at a job fair in Fletcher, N.C. earlier this month.

The NBER committee is focused on creating a chronology of when recessions started and ended, to help other economists study in the future, said NBER President James Poterba, who is also a member of the committee.

As a result, by the time the NBER says there is a recession, it is fair to say everybody already knows it. When the NBER announced in December 2008—when the economy was getting hammered by the financial crisis—that a recession had started a year earlier, it wasn’t exactly news. Recessions end at the point that the indicators the NBER watches turn upward, which isn’t the same as saying the economy is healthy.

“Even if things are getting better, it can still feel pretty bad," said Poterba.

According to Poterba, the very notion of “recession" as an economic term appears to have been started by NBER research director Wesley Mitchell in 1927.

In those early days, the NBER—now a wide-spanning group of mostly academic researchers—was primarily engaged in economic data collection and analysis. It first published dates for the beginnings and endings of U.S. recessions in March 1929, based on research conducted by Mitchell with the assistance of Simon Kuznets.

Kuznets would later win a Nobel Prize in economics, in part for his work developing gross national product—the predecessor to GDP—as a measure of the economy. The U.S. became the first country in the world to produce regular statistics on GNP in the 1940s.

Put otherwise, the NBER was weighing in on recessions well before either GNP or GDP were a thing.

Write to Justin Lahart at Justin.Lahart@wsj.com