Are We in a Recession?

In the United States, recessions are dated not by the government but by the Business Cycle Dating Committee of the National Bureau of Economic Research (NBER), a group of academic economists. They define a recession as a significant decline in activity spread across the economy and lasting more than a few months — judged on depth, diffusion, and duration rather than any single statistic.

Crucially, they announce it with a long lag, often six months to a year after the downturn began, because they wait for revised, reliable data before making a call they rarely have to reverse. By the time a recession is officially confirmed, it is frequently already over. That is why "are we in a recession right now?" can, strictly speaking, only be answered in hindsight.

The rule everyone quotes — two consecutive quarters of falling GDP — is a handy rule of thumb, not the official definition, and 2022 showed exactly why the distinction matters. US real GDP shrank in both the first and second quarters of that year, which by the popular rule meant a recession had arrived.

Yet the NBER never declared one, for a simple reason: the labour market was booming, with unemployment near a fifty-year low and payrolls growing every month. An economy adding millions of jobs is not in the broad, deep decline the committee looks for. The episode is the cleanest illustration available that GDP alone — especially in its early, heavily revised estimates — can badly mislead.

No single chart has a better recession record than the yield curve. When short-term Treasury yields rise above long-term ones — an inversion — it has preceded every US recession for more than half a century, with very few false alarms. The intuition is that an inverted curve reflects a market expecting the central bank to cut rates in response to a weakening economy.

Its great weakness is timing: the gap between inversion and recession has ranged from roughly six months to two years, an eternity for anyone watching in real time. The 2022–2024 inversion became the longest on record without a confirmed recession following it, prompting a genuine debate about whether the signal still works as cleanly. The yield-curve and 10y–3m charts let you watch the spread directly.

Where the yield curve leads by an unpredictable margin, the Sahm Rule — devised by the economist Claudia Sahm — aims to catch a recession just as it begins. It triggers when the three-month average unemployment rate rises half a percentage point above its low of the previous year. Historically, once joblessness starts rising that fast it keeps rising, so the rule has flagged past recessions with little delay and few false alarms.

It is not infallible. The rule tripped in 2024 even as the broader economy kept growing, partly because a surge in the labour force — more people entering the job search — pushed the unemployment rate up for reasons unrelated to layoffs. A useful reminder that even the best real-time gauge depends on the mechanism behind the numbers. The sahm-rule chart shows the trigger level explicitly.

The deeper problem is that economic data arrives late, noisy, and subject to heavy revision. The strong jobs report or shrinking GDP figure you read today may look entirely different after two rounds of revisions a year later — and turning points are exactly when revisions are largest, because the economy is changing direction faster than the statistics can keep up.

Recessions are also defined by breadth, not by one falling number. A manufacturing slump alongside a strong services sector, or weak output alongside solid employment, can leave the economy in a genuinely ambiguous state for months. This is why no honest answer rests on a single indicator, and why the dashboard approach beats any one gauge.

If you had to watch one area, watch jobs. Recessions are ultimately about people losing work, and the labour-market indicators — the unemployment rate, initial jobless claims, and the Sahm Rule built on top of them — tend to turn decisively once a downturn takes hold. Jobless claims in particular are weekly and barely revised, making them one of the timeliest signals available.

Output measures like GDP, industrial production, and retail sales then confirm the story with more lag. Watching claims and unemployment alongside the production data is closer to how the NBER itself reasons — looking for a broad deterioration across the real economy, not a single bad print.

Sometimes the hardest part is that how the economy feels and what it measures pull apart. In 2022 and 2023, consumer sentiment fell to levels normally seen only in deep recessions, even as employment and spending stayed strong — a gap wide enough to earn the nickname "vibecession." High inflation made people feel poorer and gloomier than the activity data implied.

That gap is itself worth watching, because sentiment can become self-fulfilling: anxious consumers and businesses eventually pull back. The consumer-sentiment and high-yield-spread charts capture the mood and the market's appetite for risk respectively — the soft signals that sometimes lead the hard data, and sometimes simply diverge from it.

The honest answer at any given moment is almost always "the data doesn't say yet." What you can do is weigh the evidence: an inverted yield curve and a triggered Sahm Rule lean toward danger, while falling jobless claims, growing payrolls, and rising retail sales lean against it. A recession is confirmed only when the deterioration is broad, deep, and sustained — and confirmed officially only well after it begins.

Use the charts above as that weight-of-evidence dashboard rather than hunting for a single yes-or-no number. MacroRadar presents all of these as historical indicators and context, not as a forecast or investment advice — the aim is to help you read the cycle, not to call it for you.