Money & Finance › Economics & schools of thought › GDP: measuring the size of the economy
Business cycles and recessions
NBER dates since 1854· Concept
Fact, law or dataA rule, statute, definition or measured number.
The economy's recurring swing between expansions and recessions, with official US turning points dated by a panel of economists at the NBER.
See it on the Money & Finance map →How it works
Expansions bring rising output, hiring and borrowing until something breaks: a rate shock, a credit bust, an oil spike or a pandemic. Spending and hiring then fall before recovering. The National Bureau of Economic Research defines a recession as a significant, broad decline lasting more than a few months, not simply two quarters of falling GDP. Since 1945 US recessions have averaged about 10 months and expansions about five years. The 2020 recession lasted two months; the 2009–2020 expansion, a record 128 months.
Why it matters
Recessions set the timing of layoffs, rate cuts, defaults and rising vacancies; nearly every lending and investment decision quietly bets on where the cycle stands.
Think of it like
Like a real-estate cycle sped up: lease-up and rent growth, then overbuilding, rising vacancy and concessions, then recovery.
Facts, rules and data
NBER's published chronology lists 34 US cycles since 1854. Its committee weighs payrolls, real income, spending and industrial output, and often announces turning points a year or more after the fact.
Debates and risks
Whether cycles come mainly from outside shocks or from built-in credit booms and busts is still argued; the long 2022–24 yield-curve inversion produced no recession.
Inside this (2)
Connections (5)
▸Monetary policy: steering interest rates and credit tries to smooth Business cycles and recessionsMainstream view
The connection
The Fed leans against the cycle. In a slowdown it cuts rates, making borrowing cheaper so households and firms spend and hire. In a boom with rising inflation it raises rates to cool demand. Because policy works with lags of roughly one to two years, the Fed steers by forecast and sometimes overshoots into recession.
When and why
Since the Fed's founding in 1913, and actively since the 1951 Treasury–Fed Accord; vivid cases include the 1981–82 Volcker recession and the March 2020 emergency cuts to near zero.
How solid is the link?
Studies by Christina and David Romer and others that isolate surprise Fed moves show they shift output and unemployment, with a lag.
Debates and risks
Whether the Fed smooths cycles or sometimes causes them, as Friedman argued about the 1930s, and how long the lags run today, are debated.
▸Yield curve inversion as a recession warning has often preceded downturns in Business cycles and recessionsExperts disagree
The connection
When short Treasury rates rise above long ones, the Fed is usually holding rates high while investors expect slower growth and future cuts. High short rates also squeeze bank lending margins. Every US recession since the late 1960s was preceded by an inversion, though not every inversion was promptly followed by a recession.
When and why
US Treasury markets, from the late 1960s through the 2022–2024 inversion; studied by Campbell Harvey (1986) and by Arturo Estrella and Frederic Mishkin (1996).
How solid is the link?
Treasury yield data line up with NBER recession dates; the New York Fed publishes a monthly recession-odds model based on the 10-year minus 3-month spread.
Debates and risks
No recession followed the long 2022–24 inversion through 2025; whether the signal broke, perhaps because of QE-era distortions, or was merely delayed is argued.
▸Business cycles and recessions shows up as swings in GDP: measuring the size of the economyFact, law or data
The connection
Recessions and expansions show up most directly in real GDP, which rises in expansions and usually falls in recessions. A popular rule calls two straight quarters of falling GDP a recession, but the NBER looks wider: in 2022 early estimates showed GDP falling for two quarters while jobs boomed, and no recession was declared.
When and why
BEA's quarterly GDP series since 1947 and the NBER committee's calls, such as the February 2020 peak and April 2020 trough.
How we know
NBER's published dates line up with BEA's GDP series, and the committee explains each call in public memos.
Debates and risks
GDP is revised for years, so early readings can mislead; gross domestic income, which should match GDP, sometimes tells a different story.
▸Business cycles and recessions drives swings in Unemployment and the monthly jobs reportFact, law or data
The connection
In a recession firms cut hiring and lay off workers, so unemployment rises fast; in expansions it falls slowly as hiring builds. Okun's law, a rule of thumb, says each point of GDP growth below trend adds roughly half a point to the unemployment rate.
When and why
Arthur Okun, later a Johnson adviser, stated the rule in 1962; vivid cases are 2007–09, when unemployment rose from under 5% to 10%, and April 2020's spike to 14.8%.
How we know
BLS unemployment data back to 1948 rise in every NBER recession; Okun's relationship holds roughly, though its size shifts over time.
Debates and risks
Why some recoveries are 'jobless', with output rebounding long before jobs, as after 1991, 2001 and 2009, is debated.
▸COVID-19 crash and rescue (2020) caused the shortest recession in Business cycles and recessionsFact, law or data
The connection
Lockdowns stopped activity almost overnight: GDP fell at about a 30% annualized rate in the second quarter of 2020, and 22 million jobs vanished in two months. Reopening and massive aid made the rebound just as fast, so the recession lasted only two months.
When and why
United States, February to April 2020, as dated by the National Bureau of Economic Research in July 2021.
How we know
Rulings of NBER's Business Cycle Dating Committee, plus BEA GDP and BLS payroll data.
Debates and risks
Whether a recession caused by a virus and government orders fits the normal business cycle at all, and what it teaches about future ones, is debated.