How the 1970s Stagflation Killed Keynesian Consensus

This essay is the development-and-history companion for readers who know the Washington Consensus story from the 1990s but want the prequel: how rich-country policy imagination was forged in the long postwar boom and then stressed in the 1970s. It connects to Milton Friedman, the natural rate hypothesis, and the Lucas critique as a methodological watershed.

The Promise

For a quarter century after World War II, Keynesian economics appeared to have cracked the code. The basic framework was elegant: governments could manage the business cycle by adjusting fiscal and monetary policy. When the economy slowed, you cut taxes and increased spending to boost demand. When it overheated, you did the reverse. The Phillips Curve — an empirical relationship first identified by A.W. Phillips in 1958 — suggested a stable tradeoff between inflation and unemployment. Policymakers could choose their preferred point on the curve: a little more inflation in exchange for lower unemployment, or vice versa.

This framework guided economic policy across the Western world. In the United States, the Kennedy and Johnson administrations explicitly used Keynesian demand management. The 1964 tax cut, designed to close the “output gap” between actual and potential GDP, was a textbook application. It worked: the economy boomed, unemployment fell, and inflation remained contained. Walter Heller, chairman of the Council of Economic Advisers, declared that the business cycle had been tamed.

The Shock

Then came the 1970s, and everything went wrong at once.

The first blow was the collapse of the Bretton Woods system in 1971-73, which removed the anchor of fixed exchange rates and eliminated the discipline that gold convertibility had imposed on monetary policy.

The second was the oil embargo of October 1973, when Arab members of OPEC cut production and embargoed exports to the United States and other countries supporting Israel in the Yom Kippur War. Oil prices quadrupled in three months — from $3 to $12 per barrel. A second oil shock in 1979, triggered by the Iranian Revolution, pushed prices to $39.

But the oil shocks were symptoms as much as causes. The deeper problem was that inflation had been building throughout the late 1960s, fueled by the fiscal strain of Vietnam War spending and Great Society programs, accommodated by a Federal Reserve that was reluctant to tighten into a political headwind.

What made the 1970s truly devastating for Keynesian economics was not inflation alone — inflation was manageable within the framework. It was the simultaneous occurrence of inflation and unemployment: stagflation. In 1975, U.S. inflation ran at 9.1% while unemployment reached 8.5%. The Phillips Curve, which posited a stable tradeoff between the two, appeared to have broken down entirely. You were not supposed to get both at once.

The Critique

Milton Friedman had predicted this. In his 1967 presidential address to the American Economic Association — one of the most consequential speeches in the history of the discipline — Friedman argued that the Phillips Curve tradeoff was an illusion. It worked only as long as workers did not anticipate inflation. Once they learned to expect rising prices, they would demand higher wages, and the economy would settle back at its “natural rate” of unemployment regardless of inflation. The long-run Phillips Curve, Friedman argued, was vertical.

Friedman’s alternative was monetarism: inflation was “always and everywhere a monetary phenomenon,” caused by excessive growth in the money supply. The Fed’s job was not to fine-tune the economy but to maintain steady, predictable growth in the money supply. Fiscal policy — the centerpiece of Keynesian management — was secondary at best.

Robert Lucas and Thomas Sargent pushed the critique further with rational expectations theory. If economic agents rationally anticipated government policy, then systematic demand management would be completely ineffective. Only unexpected policy actions could affect real variables like output and employment — and systematically tricking people was not a sustainable basis for economic policy. Lucas called it “the death of Keynesian economics,” and in 1976 his critique of econometric models — the “Lucas critique” — fundamentally changed how economists thought about policy evaluation.

The Political Revolution

The intellectual shift had political consequences. If Keynesian demand management was discredited, what replaced it?

In Britain, Margaret Thatcher came to power in 1979 on a platform of monetary discipline, trade union reform, and reduced government intervention. In the United States, Ronald Reagan was elected in 1980 promising tax cuts, deregulation, and a strong dollar. In both countries, the policy mix shifted decisively away from Keynesian demand management toward supply-side reforms and monetary targeting.

The most dramatic moment came in October 1979, when Federal Reserve Chairman Paul Volcker — appointed by Jimmy Carter, not Reagan — announced that the Fed would target money supply growth rather than interest rates. The result was a deliberate recession: interest rates surged above 20%, unemployment peaked at 10.8% in late 1982, but inflation was broken. By 1983, it had fallen to 3.2%.

The Volcker shock was painful and politically costly, but it demonstrated that monetary policy — Friedman’s prescription — could control inflation. It also demonstrated something Keynesian economics had always acknowledged but perhaps underweighted: credibility matters. Once the Fed proved it was willing to tolerate a severe recession to kill inflation, inflation expectations collapsed, and the subsequent recovery was robust.

What Survived

The story is often told as a clean narrative of Keynesian failure and monetarist triumph, but reality was messier. Monetarism in its strict form — targeting monetary aggregates — was abandoned by the Fed within a few years because the relationship between money supply measures and economic activity proved unstable. What replaced Keynesianism was not pure monetarism but a synthesis: central bank independence, inflation targeting, and the recognition that expectations matter at least as much as current policy actions.

New Keynesian economics — developed by economists like Greg Mankiw, Olivier Blanchard, and Janet Yellen — incorporated the rational expectations revolution while preserving a role for demand management. Prices and wages were “sticky” in the short run, meaning monetary policy could still affect real output. The Phillips Curve was rehabilitated in a modified form: the tradeoff existed, but only between unemployment and unexpected inflation.

The 1970s did not kill the idea that government policy could influence the economy. They killed the idea that it could do so easily, predictably, and without consequence. The lessons — about the dangers of accommodating inflation, the importance of central bank credibility, and the limits of fine-tuning — remain central to how policymakers think about macroeconomics half a century later, at least in rich-country policy debate.

The Trente Glorieuses Context: What Made the Breakdown So Salient

Jargon note: Trente glorieuses (“thirty glorious years”) is the French label for the roughly 1945–1975 stretch of very rapid reconstruction and catch-up growth across Western Europe, built on fixed investment, rising productivity, expanding welfare states, and—crucially—stable macro frameworks once wartime controls receded. The American parallel is not identical—the U.S. had no invaded homeland to rebuild—but the feel of the era was similar: high trend growth, widening middle-class consumption, and a professional belief that technocratic macro could deliver predictable outcomes.

That belief was never universal. Labor radicals, conservative gold-standard nostalgists, and a few academic monetarists always dissented. Yet the center of gravity in ministries of finance, central banks, and economics departments was Keynesian in a loose sense: automatic stabilizers plus discretionary fiscal action; monetary policy as backup; exchange-rate regimes that varied but often traded off flexibility against credibility; and a willingness to interpret inflation spikes as “cost-push” problems to be managed with incomes policies or selective controls rather than as pure monetary phenomena.

When stagflation arrived, the shock was not merely empirical—two bad numbers at once—but civilizational for a generation trained on Phillips-curve menus. If inflation could rise alongside unemployment, then the menu was not just shifting; it might be useless for the kind of social bargain policymakers thought they were running.

Supply Shocks, Expectations, and the Difficulty of “Parsing” the 1970s

Modern readers sometimes reduce the decade to oil. Oil mattered: energy is an input into nearly everything, and a sudden relative price increase can raise the price level while depressing activity—classic “stagflationary” logic even before you open a model. But a complete story also includes:

Loose nominal anchors: After Bretton Woods frayed, the monetary regime had to find a new public understanding of what “stable money” meant. Readers can follow the international angle in our Bretton Woods piece; the domestic angle is that without a clear anchor, expectations can drift upward even when policymakers dislike inflation.
Slow productivity and profit squeeze: Not every country experienced the 1970s identically, but many rich economies saw slower total factor productivity growth after the exceptional postwar sprint. Slower supply-side expansion makes demand stimulus more likely to show up as inflation rather than real output gains.
Indexation and wage–price spirals: As inflation persisted, contracts and institutions adapted. Indexation means tying wages or payments to a price index so purchasing power is protected. That can be socially fair and economically stabilizing in some designs, but it can also make inflation more persistent once a shock hits—each round of price increases triggers automatic wage increases that validate the next round.
Political constraints on disinflation: Central banks faced legislatures, labor movements, and banking systems that were not culturally prepared for the Volcker-style medicine. Disinflation is a deliberate reduction in inflation; it often requires a period of tight money and elevated unemployment. The lesson many drew from the late 1970s and early 1980s is that credibility is purchased with visible pain—a theme that echoes through later debates on inflation targeting.

Intellectual Forks: Monetarism, New Classicals, and the Keynesian Renaissance

The decade’s policy failures fed multiple research programs simultaneously—history is not a single linear “replacement” story.

Monetarism emphasized stable rules, money growth, and skepticism about fine fiscal tuning. Even if pure monetary targeting later proved operationally awkward, the monetarist insistence that inflation is a nominal phenomenon moved the Overton window for central banks.

New classical macro, associated with rational expectations, pushed a harder line: systematic policy might be neutralized by anticipation; empirical Phillips curves might reflect mis-specified learning dynamics. The Lucas critique asked modelers to distinguish reduced-form correlations from structural behavior when policy regimes change.

New Keynesian work later reassembled a role for stabilization policy by modeling micro-founded reasons prices and wages adjust slowly—preserving short-run non-neutralities without denying that expectations matter.

If you want the Keynesian “textbook” spine before the synthesis era, our General Theory in plain English and multiplier primers bracket the tradition that stagflation challenged but did not erase.

Lessons Without Caricature

Three durable takeaways help readers avoid myth on both sides.

First, stagflation is coherent in theory: supply shocks plus imperfect credibility can produce rising prices and weak activity without requiring economists to “throw out demand.”

Second, the Phillips curve was never a law of nature; at best it summarized outcomes under particular monetary-fiscal-labor institutions. When those institutions changed, so did the curve.

Third, politics lagged economics: electorates experienced the 1970s as lost competitiveness, energy anxiety, and conflict over distribution. Macroeconomic ideas succeeded when they matched a political appetite for nominal restraint—even when the intellectual packaging oversold how clean the new consensus would be.