Phillips curve

In a 1958 paper, New Zealand-born economist A. W. Phillips examined British data from 1861 to 1957. He found that unemployment tended to be lower when money wages were rising more rapidly and higher when wage growth was slower. The paper's subject was the rate of change of money wages, not the modern consumer-price inflation measure. Economists later broadened the idea to discuss the relationship between price inflation and unemployment; this later usage is what is most often meant by the Phillips curve today.

During the 1960s, many economists interpreted the observed negative association as a stable policy trade-off: stronger demand could reduce unemployment, but at the cost of higher inflation. Edmund Phelps and Milton Friedman challenged the idea that policymakers could permanently hold unemployment below its equilibrium level by accepting ever-higher inflation. Phelps emphasized that workers and firms form expectations about future prices and wages. If expected inflation adjusts upward, nominal wage and price increases no longer deliver the same lasting change in real wages or employment. In this view, monetary or fiscal policy may affect employment in the short run, but there is no dependable long-run exchange of permanently higher inflation for permanently lower unemployment.

Modern Phillips-curve models make the relationship conditional on expectations and other influences. Demand changes can initially move inflation and unemployment in opposite directions, while adverse supply disturbances can push both upward at once. The curve's measured slope is therefore not constant, and an observed association does not by itself prove that one variable caused the other. In a 2024 Federal Reserve Board analysis using regional US and European data, the authors found that the curve had flattened since 2000 under more normal labour-market conditions but remained steeper in tight labour markets. That result describes their data and method rather than a universal rule.

Central banks and researchers use Phillips-curve estimates as one input when analysing inflation, employment, and the effects of policy. The framework helps organize questions about how much spare capacity remains in an economy and how strongly prices or wages may respond as labour markets tighten. It is not a mechanical forecasting rule: expectations, supply shocks, the way inflation and slack are measured, and structural changes can all alter the estimated relationship. Economists consequently compare Phillips-curve estimates with other evidence rather than treating a single curve as a precise menu of policy choices.