Nobody Rings a Bell: Who Dates a Recession, and How Late

Nobody rings a bell.

The Only Body That Dates Recessions

The National Bureau of Economic Research’s Business Cycle Dating Committee is the acknowledged arbiter of U.S. business cycles. It does not declare recessions in real time. Instead, it identifies the exact month that a peak was reached—the economy’s high point—and the trough that followed. According to NBER’s published definition, a recession begins in the month after a peak and ends in the month of the trough. That means the committee is always writing history, never reading a live gauge.

Three Criteria That Are Somewhat Interchangeable

The committee’s approach is both methodical and surprisingly flexible. It looks for “a significant decline in economic activity that is spread across the economy and lasts more than a few months.” The three pillars are depth, diffusion, and duration, which NBER says it treats as “somewhat interchangeable.” A sharp, broad drop can be enough without a drawn-out slog; a shallow but persistent slide can also qualify.

The real work happens with a range of monthly measures of aggregate real economic activity published by federal statistical agencies. Employment, output, income, and sales data all feed into the picture. The committee points out that two important measures for determining quarterly peaks and troughs are expenditure-side and income-side real GDP, but those are not available every month. The monthly indicators carry the early weight before the quarterly accounts can confirm what happened.

Why the Calendar Gap Is Built In

NBER explicitly says it waits until “sufficient data are available” to avoid having to revise the chronology later. Quarterly GDP from the Bureau of Economic Analysis shows why. The advance estimate arrives about a month after a quarter ends, the second estimate about two months after the quarter, and then annual revisions can still shift the numbers years later. BEA’s own FAQ notes that revisions can change the measured depth of a downturn without changing its apparent timing. The committee wants to sidestep that churn; it lets the statistical smoke clear before it draws a line on the calendar.

There is also the matter of GDI—gross domestic income—which sometimes tells a different story from the expenditure-side GDP that gets the headlines. Because both versions matter for dating turning points and neither is final in the first release, the committee’s caution is built into its mandate.

The Archive of Delayed Verdicts

The lag is not theoretical. NBER’s public announcement archive shows exactly how long the committee has waited in recent cycles. It placed the February 2020 peak on June 8, 2020—four months after the month the peak occurred. The April 2020 trough, marking the end of the two-month pandemic contraction, went unconfirmed until July 19, 2021, fifteen months after the economy had already turned back up.

The Great Recession was even more stretched. The December 2007 peak was called on December 1, 2008, a year after the descent began. And the June 2009 trough was finally placed on September 20, 2010, more than fifteen months of post-recovery data later. The 2001 recession followed a similar script: the March 2001 peak was announced November 26, 2001, an eight-month wait. The November 2001 trough—the economy’s low point—was dated July 17, 2003, twenty months after the fact, a long lag in this set.

GDP Is Not the Same as a Recession Call

The gap between the GDP reports that drive cable-news chyrons and the NBER chronology is more than a matter of timing. BEA explicitly says quarterly GDP estimates are revised as new and revised source data become available. Advance numbers can turn a modest decline into a small gain, or vice versa, when the annual benchmark revisions run. NBER’s chronology, by contrast, does not shift once set. The committee’s judgment rests on the weight of multiple monthly indicators, not on a single threshold crossing in the national accounts. When someone on television declares a recession because there were two negative GDP prints, the technical arbiter that the Fed and Treasury actually care about is likely still gathering evidence and will not issue a ruling for months.

That long lag is the feature, not a flaw. The committee’s whole reason for existing is to provide a definitive, no-regrets chronology. It will never issue a call that later GDP revisions might embarrass.

The latest turning point in the archive—the April 2020 trough called in July 2021—captures the deal. The committee does not name a peak or a trough until enough evidence has piled up to make any later revision unthinkable. That discipline is why the NBER chronology carries weight. It also means that if you suspect a recession is starting right now, the only honest answer is that we will not know for certain until the economists in Cambridge tell us, months later, that the top is already behind us.