Common mistakes about recessions

The common belief: Leading indicators give a dependable, repeatable signal that a recession is coming.

What the data shows: Most indicators are calibrated on small samples — fewer than a dozen US recessions since 1960 — so a single miss materially changes their record. Across 2022-2024 the yield curve, the Sahm rule and the Conference Board’s leading index all warned, yet the NBER dated no recession through 2024. They measure pressure, not destiny.

→ Complete explanation: How accurate are recession indicators?

The common belief: When the unemployment rate’s three-month average rises 0.50 point above its twelve-month low, a recession is already underway.

What the data shows: The threshold held across every US recession since 1960, yet it also fired in November 1976 without a recession, and again in July 2024 — at 0.53 point — with no NBER-dated downturn following. Its creator, Claudia Sahm, calls it a historical pattern, not a rule of nature.

→ Detailed explanation: Is the Sahm rule a reliable recession indicator?

The common belief: Once short rates exceed long rates, a recession is locked in within a year.

What the data shows: The curve inverted before all ten US recessions since 1955 (San Francisco Fed), with one false positive in the mid-1960s — a record strong enough to feel deterministic. The underlying series is published in our 2s10s spread data. But the 2s10s inversion that began in July 2022 became the longest in modern history and reversed in 2024 with no recession dated. Inversion signals a coming credit contraction; when that channel is neutralised, the signal detaches from the outcome. Contrary to the textbook framing, it is a conditional warning about credit, not a countdown.

→ In-depth explanation: Why does the yield curve invert before recessions?

The common belief: Corporate bond spreads are a coincident or lagging measure — useful for description, not prediction.

What the data shows: High-yield spreads tend to widen ahead of downturns as lenders reprice default risk, often before equities fully adjust. Noisier than the yield curve, they capture the funding-cost channel directly. In 2022 spreads widened, then tightened as the feared default wave did not arrive.

→ The full explanation: Do credit spreads predict recessions?

The common belief: Oil is one commodity among many and largely irrelevant to the business cycle.

What the data shows: James Hamilton’s NBER research documents that 10 of the 11 postwar US recessions before the pandemic were preceded by a sharp rise in oil prices, the sole exception being 1960, with a lag of about three quarters. The link is contributing, not mechanical — the 2003 increase brought no recession.

→ Full breakdown: Why do oil prices spike before recessions?

The common belief: Survey data on lending standards is too soft and slow to matter for forecasting.

What the data shows: The Fed’s Senior Loan Officer Opinion Survey captures the credit-supply channel before it reaches growth or employment. Tightening standards have preceded most postwar downturns, because restricted credit constrains investment and hiring with a lag — a directional signal, not a timing tool.

→ Fuller explanation: Do bank lending standards predict downturns?

The common belief: An official body announces a recession as soon as GDP turns down.

What the data shows: In the US, recessions are dated by the NBER’s Business Cycle Dating Committee — a private academic group — which weighs production, employment and real income, not GDP alone, and works retrospectively. There is no real-time government declaration; it waits until the evidence is unambiguous.

→ Extended explanation: What is the NBER recession dating methodology?

The common belief: The start of a recession is obvious in real time.

What the data shows: Recessions are confirmed only after they begin, sometimes by a wide margin. The NBER dated the December 2007 peak in December 2008 — about a year later — and the February 2020 peak in June 2020, considered unusually fast. Much of a recession is often over before it is official.

→ The complete explanation: Why are recessions only confirmed after they start?

The common belief: Sustained monetary tightening inevitably produces a hard landing.

What the data shows: Most tightening cycles have ended in recession, but not all. The benchmark soft landing is 1994-1995, when the Fed roughly doubled the funds rate to 6% without a downturn; Powell has also cited 1965 and 1984. Entering 2023, about 85% of economists surveyed by the Financial Times expected a recession the NBER has not dated.

→ Full account: What is a soft landing and has it ever been achieved?

The common belief: The definition of a recession is two consecutive quarters of falling real GDP.

What the data shows: That two-quarter rule is a journalistic shorthand, not the NBER’s definition. In the first half of 2022, US real GDP contracted for two straight quarters, yet no recession was dated because employment and real income kept rising. The technical rule and the official chronology can diverge.

→ Complete breakdown: How do technical recessions differ from NBER recessions?

The common belief: Downturns come from external events — an oil embargo, a pandemic, a war.

What the data shows: Hyman Minsky’s financial-instability hypothesis describes how stability itself breeds fragility: long calm periods encourage leverage, until a “Minsky moment” — coined for the 1998 Russia crisis — tips into forced selling. The 2008 crisis is the canonical example of an endogenous unwinding.

→ Complete explanation: What causes a Minsky moment?

The common belief: Sluggish growth after a recession is temporary and self-correcting.

What the data shows: Secular stagnation — a concept from Alvin Hansen in 1938, revived by Larry Summers in 2013 — describes a persistent shortfall of demand and a low neutral rate that can keep growth and inflation subdued for years, not quarters. It reframes a weak recovery as structural, not a delay.

→ Detailed explanation: What is secular stagnation and why does it matter?

The common belief: When the Conference Board’s Leading Economic Index declines persistently, a recession must follow.

What the data shows: The LEI fell for more than fifteen consecutive months across 2022-2023 — historically abnormal — without a recession. It overweights goods and excludes services, over 70% of US output, so it can flag a goods slowdown while a services-led economy keeps growing. It signals direction, not magnitude.

→ In-depth explanation: Why does the Conference Board LEI give false signals?