Business Cycles and Their Impact on the Labor Market
Which U.S. recessions hit unemployment hardest?
This ranking compares how strongly U.S. recessions affected the labor market by measuring the increase in the unemployment rate around each NBER business-cycle downturn. The metric is the rise in the U.S. unemployment rate from the NBER peak month to the highest unemployment rate around the recession.
The ranking is calculated from official BLS unemployment data and NBER recession dates. FRED provides the public page for the BLS UNRATE series, while NBER identifies the recession peak and trough months used to anchor each cycle.
Unit: percentage points. Direction: larger increase ranks higher. Coverage: 12 U.S. recessions from 1948 onward with comparable monthly unemployment-rate data.
The 2020 COVID recession produced the largest unemployment-rate jump in the post-1948 monthly record.
The short 1980 recession had the smallest unemployment-rate increase in this 12-cycle comparison.
The ranking covers every U.S. NBER recession from 1948 onward with comparable monthly UNRATE data.
Unemployment values come from the BLS UNRATE series; recession dates come from NBER.
Overview: what the unemployment shock metric means
A business cycle moves through expansion, peak, contraction and recovery. The labor market usually adjusts with a delay. Employers may first slow hiring, reduce vacancies, cut hours or rely on temporary layoffs before unemployment rises fully.
The unemployment-rate increase used here is a simple way to compare labor-market stress across recessions. It does not say that unemployment is the only damage. A downturn can also reduce labor-force participation, increase underemployment, weaken wage growth, lower weekly hours and shift workers into lower-quality jobs.
The ranking should be read as a focused unemployment comparison, not as a complete ranking of the worst recessions in U.S. history. It measures how far the unemployment rate rose around each NBER recession, using the same monthly series and the same method for every cycle.
Top 10 U.S. recessions by unemployment-rate increase
The top ten show two different patterns of labor-market damage. The 2020 shock was unusually abrupt, while the Great Recession and several postwar downturns produced large increases over longer adjustment periods.
Top 10 unemployment shocks, percentage-point increase
| Rank | Cycle | Increase | Data note |
|---|---|---|---|
| 1 | 2020 COVID recession | +11.3 pp | 3.5% in Feb 2020; 14.8% in Apr 2020. |
| 2 | 2007–2009 Great Recession | +5.0 pp | 5.0% in Dec 2007; 10.0% in Oct 2009. |
| 3 | 1973–1975 recession | +4.2 pp | 4.8% in Nov 1973; 9.0% in May 1975. |
| 4 | 1948–1949 recession | +4.1 pp | 3.8% in Nov 1948; 7.9% in Oct 1949. |
| 5 | 1981–1982 recession | +3.6 pp | 7.2% in Jul 1981; 10.8% in Dec 1982. |
| 6 | 1953–1954 recession | +3.5 pp | 2.6% in Jul 1953; 6.1% in Sep 1954. |
| 7 | 1957–1958 recession | +3.4 pp | 4.1% in Aug 1957; 7.5% in Jul 1958. |
| 8 | 1969–1970 recession | +2.6 pp | 3.5% in Dec 1969; 6.1% in Dec 1970. |
| 9 | 1990–1991 recession | +2.3 pp | 5.5% in Jul 1990; 7.8% in Jun 1992. |
| 10 | 2001 recession | +2.0 pp | 4.3% in Mar 2001; 6.3% in Jun 2003. |
Values are percentage-point changes. The full formula and the 24-month unemployment window are explained in the methodology section.
Chart: unemployment-rate shocks across 12 U.S. recessions
The chart uses the same values as the ranking table. It highlights the extreme scale of the 2020 unemployment spike, the deep labor-market damage of the Great Recession and the cluster of mid-sized unemployment shocks in the 1950s, 1970s and early 1980s.
Methodology: how the ranking is calculated
The metric is the increase in the U.S. unemployment rate around each NBER recession. The ranking uses descending order, so a larger unemployment-rate increase ranks higher. The calculation uses the seasonally adjusted U.S. unemployment rate from BLS, accessed through BLS and FRED, and NBER peak-to-trough recession dates.
What was measured
The ranking compares the unemployment rate at the NBER peak month with the highest unemployment rate reached during the recession or within 24 months after the NBER trough.
How to read the rank
A value of +5.0 pp means the unemployment rate rose by five percentage points from the cycle peak month to the highest unemployment rate around that recession.
Formula
Unemployment shock = highest unemployment rate around the recession minus unemployment rate at the NBER peak month. Example: 2020 equals 14.8% minus 3.5% = 11.3 percentage points.
Why the window extends after the trough
The table looks up to 24 months after the NBER trough because unemployment often keeps rising after output has started recovering. This is a comparison method for this article, not an official recession-dating rule.
Why recessions differ
Labor-market damage depends on the shock type, policy response, credit conditions, sector exposure and whether firms cut jobs immediately or first reduce hiring, hours and vacancies.
What the metric misses
It does not measure underemployment, labor-force exits, wage losses, hours reductions, job quality, long-term unemployment or sector-specific employment damage.
Recessions before 1948 are excluded because the comparable monthly unemployment-rate series used here starts in 1948. The ranking is calculated before display rounding, and all displayed values are shown in percentage points.
BLS is the underlying source for the unemployment series. FRED provides the public series page for UNRATE. NBER supplies the recession peak and trough months used to define each business-cycle downturn.
Main ranking: U.S. business cycles by unemployment-rate shock
The table can be searched by recession name, filtered by cycle type and sorted by unemployment-rate increase, year or name.
Ranking by unemployment-rate increase, percentage points
| Rank | Cycle | Increase | Data note |
|---|---|---|---|
| 1 | 2020 COVID recession | +11.3 pp | Feb 2020: 3.5%; Apr 2020: 14.8%. |
| 2 | 2007–2009 Great Recession | +5.0 pp | Dec 2007: 5.0%; Oct 2009: 10.0%. |
| 3 | 1973–1975 recession | +4.2 pp | Nov 1973: 4.8%; May 1975: 9.0%. |
| 4 | 1948–1949 recession | +4.1 pp | Nov 1948: 3.8%; Oct 1949: 7.9%. |
| 5 | 1981–1982 recession | +3.6 pp | Jul 1981: 7.2%; Dec 1982: 10.8%. |
| 6 | 1953–1954 recession | +3.5 pp | Jul 1953: 2.6%; Sep 1954: 6.1%. |
| 7 | 1957–1958 recession | +3.4 pp | Aug 1957: 4.1%; Jul 1958: 7.5%. |
| 8 | 1969–1970 recession | +2.6 pp | Dec 1969: 3.5%; Dec 1970: 6.1%. |
| 9 | 1990–1991 recession | +2.3 pp | Jul 1990: 5.5%; Jun 1992: 7.8%. |
| 10 | 2001 recession | +2.0 pp | Mar 2001: 4.3%; Jun 2003: 6.3%. |
| 11 | 1960–1961 recession | +1.9 pp | Apr 1960: 5.2%; May 1961: 7.1%. |
| 12 | 1980 recession | +1.5 pp | Jan 1980: 6.3%; Jul 1980: 7.8%. |
All values are percentage-point changes in the unemployment rate. Ranking order is based on the numeric increase, descending.
Insights from the business-cycle labor-market ranking
Key insight
The 2020 recession stands apart because unemployment rose with unusual speed. The NBER recession was short, but the labor-market shock was historically large.
Notable pattern
The Great Recession ranks second, showing that financial crises can produce deep labor-market damage even when unemployment rises more gradually than in a sudden shutdown.
Why cycle type matters
Inflation shocks, credit stress, policy tightening and demand contractions can all raise unemployment, but they affect hiring, vacancies, hours and layoffs through different channels.
Outlier
The 1980 recession ranks last by this measure, but it was followed closely by the deeper 1981–1982 recession, when unemployment rose much further.
What this ranking means for readers
The ranking shows that business cycles affect labor markets through both depth and timing. A downturn can be officially short while still creating a large employment shock if firms stop activity suddenly or if layoffs are concentrated in a short period.
It also shows why unemployment is often called a lagging indicator. In several cycles, including 1990–1991 and 2001, the unemployment-rate high came after the NBER trough. Output can begin recovering before hiring, hours and job openings fully return.
A larger unemployment-rate increase generally signals weaker demand for labor, but it should not be read as the full human cost of a downturn. Participation declines, underemployment, reduced hours, weaker wages and long-term joblessness can move differently across recessions.
FAQ
What is a business cycle?
A business cycle is the movement of economic activity through expansion, peak, contraction and recovery. Recessions are the contraction phase of the cycle.
How do business cycles affect the labor market?
During contractions, employers often slow hiring, reduce vacancies, cut hours or lay off workers. During recoveries, hiring usually improves, but unemployment can remain elevated for some time.
What does this unemployment shock metric mean?
It measures how many percentage points the unemployment rate rose from the NBER peak month to the highest unemployment rate around the recession. A larger value means a larger unemployment-rate shock.
Why does unemployment often peak after a recession ends?
Employers may wait for demand to stabilize before hiring again. That delay means the labor market can keep weakening after output has already begun to recover.
Why does the 2020 recession rank first?
The unemployment rate rose from 3.5% in February 2020 to 14.8% in April 2020, a jump of 11.3 percentage points in this comparison.
Is this a ranking of the worst recessions overall?
No. It ranks recessions by one labor-market metric: the unemployment-rate increase. It does not rank recessions by GDP loss, duration, wages, wealth losses, financial damage or household hardship.
Why are pre-1948 recessions excluded?
The monthly U.S. unemployment-rate series used here starts in 1948. Earlier recessions are excluded because they cannot be compared with the same monthly BLS/FRED unemployment-rate series.
What else should readers consider besides unemployment?
Labor-force participation, underemployment, hours worked, wage growth, long-term unemployment and job quality can all change the way a recession is experienced by workers.
Sources
U.S. Bureau of Labor Statistics — Labor Force Statistics from the Current Population Survey
Primary statistical source for the U.S. unemployment-rate series used in the calculation.
FRED — Unemployment Rate, UNRATE
Public page for the monthly U.S. unemployment-rate series sourced from BLS.
NBER — U.S. Business Cycle Expansions and Contractions
Source for U.S. business-cycle peak and trough months used to define recession episodes.
https://www.nber.org/research/data/us-business-cycle-expansions-and-contractions
NBER — Business Cycle Dating
Context source explaining the role of the Business Cycle Dating Committee and the interpretation of recession dates.
Access snapshot: BLS, FRED and NBER source pages reviewed on July 9, 2026. BLS is the underlying source for the unemployment series; FRED provides the public page for the same series.
