Five Numbers That Signal a Recession Before It Arrives

No indicator is a crystal ball, but a handful of data points have earned their reputation as early warnings.

Recessions are officially declared by a committee of economists — often a year after the downturn began. Markets can't wait that long, so they watch a small dashboard of early signals. Here are the five with the best track record, and the caveats that come with each.

1. The yield curve

When 3-month Treasury yields exceed 10-year yields, bond investors are betting on future rate cuts — which usually means they expect trouble. Inversions have preceded most modern U.S. recessions, typically by 6–18 months. Caveat: the lag is long and irregular, and an economy can shrug off a brief inversion.

2. Initial jobless claims

The weekly count of new unemployment filings is the fastest labor-market data that exists. A sustained climb of 15–20% off the lows has historically been serious; one bad week is noise. This series matters because layoffs are how slowdowns spread — job loss cuts spending, which causes more job loss.

3. The Sahm rule

A recession signal with an unusually clean record: when the three-month average unemployment rate rises half a point above its low from the past year, a recession has essentially always been underway. It confirms rather than predicts — but it confirms early, while official declarations take a year.

4. Manufacturing surveys (PMI)

Purchasing managers are asked monthly whether orders are rising or falling. Readings below 50 signal contraction. Manufacturing is a small slice of modern economies but swings first and hardest, making it a useful canary. Caveat: it has cried wolf during soft patches that never became recessions.

5. Consumer spending on discretionary goods

Restaurant sales, air travel, big-ticket retail. Households cut the fun spending first, and companies say so in earnings calls — often the earliest anecdotal evidence, quarters before it shows in government data.

How to use the dashboard

One flashing light is a caution; three or four together have rarely been wrong. And note what is not on the list: the stock market. Equities crash without recessions and rally into them — as the old line goes, the market has predicted nine of the past five recessions.