The centerpiece of Milton Friedman's (1968) presidential address to the American Economic Association, delivered in Washington, DC, on December 29, 1967, was the striking proposition that monetary policy has no longer-run effects on the real economy. Friedman focused on two real measures, the unemployment rate and the real interest rate, but the message was broader—in the longer run, monetary policy controls only the price level. We call this the monetary-policy invariance hypothesis. By 1968, macroeconomics had adopted the basic Phillips curve as the favored model of correlations between inflation and unemployment, and Friedman used the Phillips curve in the exposition of the invariance hypothesis. Friedman's presidential address was commonly interpreted as a recommendation to add a previously omitted variable, the rate of inflation anticipated by the public, to the right-hand side of what then became an augmented Phillips curve. We believe that Friedman's main message, the invariance hypothesis about long-term outcomes, has prevailed over the last half-century based on the broad sweep of evidence from many economies over many years. Subsequent research has not been kind to the Phillips curve, but we will argue that Friedman's exposition of the invariance hypothesis in terms of a 1960s-style Phillips curve is incidental to his main message.