What Is a Recession, Really?

WTH Editorial August 15, 2026 5 min read

Everyone repeats the same definition of a recession: two straight quarters of shrinking GDP. It’s tidy, it’s easy to remember, and in the United States it isn’t the definition at all. The real one is a judgment call — and that’s exactly why an economy can be sliding for months before anyone official will say the word out loud.

The rule that isn’t the rule

The two-quarters idea isn’t random. GDP — gross domestic product — is the broadest single gauge of what an economy produces, so when it contracts two quarters in a row, that’s a clean, countable signal the economy is going backward. Some countries lean on it formally: the UK and Canada treat two consecutive quarters of negative growth as a “technical recession.” The rule has the appeal of being objective. You either had two red quarters or you didn’t.

The US just doesn’t define recessions that way. There’s no law on the books that sets the threshold, and no government agency declares one. Which raises the obvious question: if not GDP arithmetic, then what?

Who actually decides

In the US, the call belongs to a private nonprofit — the National Bureau of Economic Research — and a small committee of economists inside it. They don’t plug numbers into a formula and read off an answer. They make a judgment, weighed across three dimensions: how deep the decline is, how widely it’s spread across the economy, and how long it lasts. Depth, diffusion, duration.

That framing does real work. A brutal slump confined to one industry isn’t a recession, however painful it is for the people in it — it fails the diffusion test. A broad downturn that pulls down jobs, incomes, spending, and production at the same time is the real thing, even if any single quarter’s GDP print looks survivable. The committee is trying to capture the character of a downturn, not just its arithmetic.

Why GDP isn’t the star

Here’s the part that surprises people: GDP isn’t even the number the committee leans on hardest. In recent decades it has put the most weight on two other measures — real personal income minus government transfers, and nonfarm payroll employment. Strip the jargon and it comes down to a simple pair of questions: are people earning, and are people working?

That emphasis on jobs and income over raw output is the whole reason the two-quarters shorthand can mislead. Output and employment usually move together — but not always, and not on the same schedule. When they diverge, the committee follows the labor market.

Where the “two quarters” rule breaks

Once you see what the committee is actually measuring, the rule of thumb falls apart on contact with history. The 2001 recession never produced two consecutive quarters of shrinking GDP — not a single back-to-back pair — and it is still, officially, a recession. The 2020 downturn was the opposite failure mode: savage, but so compressed that it barely spanned the window the two-quarter rule needs, and no one seriously disputes it counted. The committee has, on occasion, declared a recession before a second negative quarter ever printed.

The pattern is consistent. The shorthand gets the easy, textbook cases right and the genuinely interesting ones wrong — which is a problem, because the interesting ones are usually the ones people are arguing about in real time.

You find out in the rearview mirror

There’s one more consequence of making the call a careful judgment instead of a formula: it arrives late. The committee dated the start of the most recent deep recession to December 2007 — but didn’t announce that determination until nearly a year afterward. For months, by the official timeline, the economy was already contracting while the data was still being debated and revised.

This isn’t really a flaw so much as a design choice. Rushing the label would mean reversing it when revisions come in, and a recession call that flips is worse than a slow one. But it does mean the official answer to “are we in a recession right now?” is almost never available when you actually want it.

What people watch instead

So in the gap between the downturn and the verdict, what do economists actually watch? Not GDP alone. They watch the labor market and incomes, because that’s what the official call leans on. They watch leading signals — most famously the yield curve, where short-term interest rates climbing above long-term ones has preceded most modern recessions. A newer gauge, the Sahm rule, flags trouble when the unemployment rate’s three-month average rises about half a percentage point above its recent low.

None of these is the definition. They’re early-warning instruments for a verdict that, by construction, only arrives later. Treat them as smoke detectors, not as the official fire report.

The honest version

Strip it all down and a recession isn’t a box that checks itself the moment two GDP numbers come in red. It’s a judgment about how deep, how broad, and how long the damage runs — built more on whether people are working and earning than on raw output, and usually made official long after it began. The two-quarters rule survives because it’s easy, not because it’s right.

Keep it as a rough mental shorthand if you like. Just don’t mistake the shorthand for the thing it’s standing in for — because the real definition is the one that actually tells you how much an economy is hurting.

Not investment advice. WTH Markets is editorial commentary, not financial guidance.