In 1958, the economist A.W. Phillips studied nearly a hundred years of British wage and unemployment data and found a strikingly consistent pattern: when unemployment was low, wages rose faster; when unemployment was high, wage growth slowed. Plotted as a curve, it looked like a genuine, stable trade-off — and through the 1960s, more than one government treated it exactly that way: a dial they could turn, accepting a little more inflation to buy a little less unemployment, or the reverse.
A dial policymakers thought they could turn
The appeal was obvious. If the Phillips curve held, a government facing high unemployment could stimulate demand, accept somewhat higher inflation as the cost, and expect unemployment to fall in a fairly predictable, calculable amount — the sort of clean policy lever every government wants and rarely gets. For roughly a decade, that's how it was used, treated less like an empirical observation about one country's economic history and more like a dependable, exploitable law.
The 1970s broke the graph
Then came stagflation: high unemployment and high inflation at the same time, most sharply after the 1973 oil shock, which the simple curve said shouldn't be able to happen at all — you were supposed to be able to have one or the other, not both. Milton Friedman and Edmund Phelps had already argued, before this happened, that the trade-off was only ever temporary. Once workers and businesses start expecting a given rate of inflation, they build that expectation into wage demands and prices regardless of how tight the job market is, and the curve itself shifts rather than staying fixed in place. Their "expectations-augmented" version of the curve is close to vertical in the long run — meaning no lasting trade-off exists at all, only a short-run illusion that lasts until expectations catch up.
What we're still unsure about
Even the modern, expectations-adjusted version of the relationship isn't settled science. In many developed economies over the past couple of decades, the short-run trade-off appears to have weakened or "flattened" — periods of genuinely low unemployment that didn't produce the inflation older models would have predicted. Economists actively disagree about why: better-anchored inflation expectations, globalisation loosening the link between domestic labour markets and domestic prices, and structural changes in how wages get set have all been proposed, without a consensus on which explanation, or combination, is doing the real work. The Phillips curve remains one of the more visibly, openly contested tools still in active use in real-world monetary policy debate.
This sits inside Inflation & the Phillips Curve, one of eight topics in Macroeconomics, one of five domains in Economics, one of seventeen subjects the app can quiz you on.