Money & Finance › Economics & schools of thought › Business cycles and recessions
The Phillips curve: inflation vs. unemployment
1958· Theory or school
Experts disagreeEconomists or investors genuinely disagree.
The observed tradeoff in which lower unemployment comes with faster wage and price inflation, at least for a while.
See it on the Money & Finance map →How it works
In 1958 New Zealand-born economist A.W. Phillips showed British wages rose faster when unemployment was low, using data from 1861 to 1957. Paul Samuelson and Robert Solow applied it to US prices in 1960, suggesting policymakers could pick a mix. Milton Friedman and Edmund Phelps argued the tradeoff fades once people expect inflation; only a 'natural rate' of unemployment is sustainable. The curve then looked nearly flat in the late 2010s, when unemployment hit 3.5% with inflation under 2%.
Why it matters
Central banks still reason with it: a very tight labor market is an inflation warning, which is why strong jobs data can push the Fed toward rate hikes.
Facts, rules and data
Phillips's 1958 paper in Economica and decades of data showing the relationship shifts over time. Phillips also built MONIAC, a water-powered machine that modeled the economy.
Debates and risks
Whether the curve is flat, steep, or bends sharply when labor markets get very tight, as 2021–22 suggested, remains unsettled.
Connections (2)
▸The Phillips curve: inflation vs. unemployment trades off inflation against Unemployment and the monthly jobs reportExperts disagree
The connection
The Phillips curve links the two: when unemployment is low, employers compete for workers, wages rise faster, and firms pass costs into prices. When unemployment is high, wage and price growth slow. Modern versions say the tradeoff holds only in the short run, around a 'natural' unemployment rate.
When and why
Phillips's 1958 British data; Samuelson and Solow's 1960 US version; the natural-rate revision by Friedman and Edmund Phelps in the late 1960s.
How solid is the link?
Wage data show the pattern in many periods, but statistical estimates of its strength vary widely by decade and country.
Debates and risks
Why the curve was almost flat in 2010–2019, then seemed steep in 2021–22, and where the natural rate sits now, are unsettled.
▸The Phillips curve: inflation vs. unemployment broke down during The Great Inflation and 1970s stagflationFact, law or data
The connection
In the 1970s inflation and unemployment rose together, which the simple curve said should not happen. Oil shocks hit supply, and workers and firms came to expect inflation and built it into wages and prices. Friedman and Phelps had predicted this: holding unemployment low with easy money produces ever-higher inflation.
When and why
United States and Britain, 1973–1982; in 1975 US unemployment reached 9% with inflation near 9%, and in 1980 inflation hit 14.8%.
How we know
BLS data trace no stable curve in the 1970s; Friedman's December 1967 address forecast the breakdown before it happened, a rare successful prediction in economics.
Debates and risks
How much of the Great Inflation came from oil shocks versus Fed mistakes and political pressure on the Fed is debated.