# Phillips Curve The **Phillips curve** is the proposed inverse relationship between unemployment and inflation: as labor markets tighten, wage and price pressure builds. Named for A. W. Phillips, whose 1958 study of British wage data found the pattern, it became the empirical backbone of the last two steps of the [[Monetary Policy Transmission Flow]] — the assumption that suppressing demand reliably suppresses prices. It has repeatedly failed in ways that reshaped macroeconomics. Milton Friedman and Edmund Phelps argued in the late 1960s that any stable tradeoff would vanish once workers adjusted their expectations, and the stagflation of the 1970s — high unemployment *and* high inflation together — confirmed it, producing the expectations-augmented version and the concept of a natural rate. The relationship then went nearly flat from the 1990s through the 2010s: very low unemployment produced almost no inflation, and the 2008–09 collapse produced almost no disinflation. The curve's flatness is a large part of why the [[Expectations Channel]] has come to carry so much of the analytical load. The policy implication is uncomfortable — if the link between slack and prices is weak, hitting an inflation target through demand alone requires very large employment costs. ## Lineage | Year | Who | Claim | |---|---|---| | 1958 | A. W. Phillips | UK wage inflation and unemployment, 1861–1957, move inversely | | 1960 | [[Paul Samuelson]] and Robert Solow | A U.S. version, read as a "menu of policy options" | | 1967–68 | [[Milton Friedman]] and Edmund Phelps | Any tradeoff is temporary; once expectations adjust, unemployment returns to its [[Natural Rate of Unemployment]] | | 1970s | — | [[Stagflation]]: average inflation rose from about 2.5 to about 7 percent while unemployment rose too | | 1990s–2010s | — | The curve flattens | | 2021–23 | — | Debate over whether it re-steepened | The 1960s "menu" reading is an intellectual root of the [[Great Inflation]]. Friedman (in his 1967 American Economic Association address, published 1968) and Phelps argued that only *unexpected* inflation lowers unemployment, so exploiting the tradeoff just raises expected inflation. The 1970s bore them out, and the expectations-augmented curve became the textbook version. ## Why it flattened Three explanations competed in the 2010s. Anchored expectations may themselves flatten the curve, since firms stop passing temporary slack or tightness into prices. Globalization and weaker worker bargaining may have loosened the wage link. Or the curve may never have been as flat as it looked: a central bank that offsets demand shocks before they show up in inflation hides the very relationship it relies on, a point made by Michael McLeay and Silvana Tenreyro. ## Did it come back? The post-pandemic inflation reopened the question. Pierpaolo Benigno and Gauti Eggertsson argued in 2023 that the curve is nonlinear: flat when labor markets are slack but steep once job vacancies outnumber the unemployed, as they did in 2021–22. Paul Beaudry, Chenyu Hou and Franck Portier replied in 2025 that the nonlinearity is fragile and largely disappears once household and firm inflation expectations are controlled for; on their reading, [[Supply Shock]]s de-anchored short-run expectations. The steep-curve view implies disinflation is cheap once tightness eases; the expectations view implies its cost depends on credibility — the [[Expectations Channel]]. ## See also - [[Monetary Policy Transmission Flow]] · [[Interest Rate Policy Timeline]] - [[Dual Mandate]] · [[Expectations Channel]] · [[Supply Shock]] · [[Inflation Expectations]] - [[Great Inflation]] · [[Stagflation]] · [[Natural Rate of Unemployment]] · [[Monetarism]] · [[Keynesian Economics]]