The long-run Phillips curve is vertical because, over time, workers and firms fully adjust their expectations to match actual inflation, eliminating any lasting trade-off between inflation and unemployment. In the short run, unexpected inflation can temporarily push unemployment below its natural rate. But once people catch on and demand higher wages or raise prices accordingly, unemployment drifts back to its natural level regardless of where inflation settles. The economy ends up with higher inflation and the same unemployment rate it started with.
The Short-Run Trade-Off That Disappears
The original Phillips curve, based on economist A.W. Phillips’ 1958 observation, showed an inverse relationship between inflation and unemployment. When inflation rose, unemployment fell, and vice versa. This suggested policymakers could “buy” lower unemployment by tolerating higher inflation. In the short run, this relationship holds because inflation expectations are sticky. If people expect 2% inflation and the central bank pushes actual inflation to 4%, the extra spending power temporarily boosts output and hiring before anyone adjusts.
The short-run Phillips curve shifts, though, whenever expectations change. Each time inflation runs above what people anticipated, they revise their expectations upward. The next short-run curve sits higher than the last one. The result is that any attempt to keep unemployment permanently below its natural rate requires not just high inflation, but continuously accelerating inflation. This is why economists call it the “accelerationist” model.
How Expectations Erase the Trade-Off
Milton Friedman’s key insight in 1968 was that workers care about real wages (what their paycheck actually buys), not nominal wages (the dollar amount on the check). Phillips had related nominal wage growth to unemployment. Friedman argued that once you reframe the relationship in terms of expected real wages, the long-run trade-off vanishes.
Here’s the intuition. Suppose the government stimulates the economy and inflation rises from 2% to 5%. Initially, firms see higher prices for their products and hire more workers, while workers haven’t yet noticed that their cost of living has jumped. Unemployment drops. But this only works as long as workers are fooled. Once they realize prices have risen 5%, they negotiate for higher wages. Firms’ labor costs go back up, and they shed the extra workers. Unemployment returns to its natural rate, but now inflation sits at 5% instead of 2%.
If expectations are “adaptive,” meaning people base their forecasts on recent experience, the simplest version assumes workers expect this period’s inflation to equal last period’s. A Federal Reserve analysis confirmed that this adaptive expectations assumption, combined with firms setting prices as a constant markup over costs, is sufficient to produce the accelerationist Phillips curve. No matter what happens at the individual worker level, the macro-level outcome is the same: you can’t permanently trade inflation for lower unemployment.
The Natural Rate of Unemployment
The long-run Phillips curve is vertical at a specific point on the horizontal axis: the natural rate of unemployment (sometimes called the noncyclical rate or NAIRU). This isn’t zero. It reflects the baseline level of unemployment that exists even in a healthy economy due to structural features of labor markets.
Friedman described the natural rate as the unemployment level that would emerge from the economy’s actual structural characteristics: how costly it is to search for jobs, how much information workers and employers have, how mobile the labor force is, and the normal churn of people switching careers or entering the workforce. Several measurable factors shift the natural rate over time:
- Unemployment insurance generosity. Higher replacement rates (the percentage of prior income that benefits cover) and longer benefit periods can raise the natural rate by reducing the urgency of job search.
- Minimum wages and unionization. Both can push real wages above market-clearing levels, creating a persistent gap between labor supply and demand.
- Tax structure. Payroll taxes raise the cost of hiring, income taxes affect the incentive to work versus stay idle, and investment taxes influence how much firms expand.
The Congressional Budget Office currently estimates the U.S. noncyclical rate of unemployment at roughly 4.2%, and projects the actual unemployment rate will converge toward that level by 2030. This number isn’t fixed forever. It shifts as labor market institutions, demographics, and technology change. But at any given time, the long-run Phillips curve is anchored to it.
The 1970s Proved the Point
For much of the 1960s, policymakers in the U.S. and U.K. operated as if the short-run Phillips curve were permanent. They believed they could maintain low unemployment by accepting moderately higher inflation. The 1970s shattered that belief.
U.S. consumer price inflation hit 6.4% in 1970, with the U.K. reaching 7.9%, and both countries saw inflation climb far higher as the decade progressed. Crucially, unemployment rose at the same time. This combination, dubbed “stagflation,” was impossible under the original downward-sloping Phillips curve. As one British official later admitted, the idea that inflation and unemployment moved in opposite directions turned out to be “a most damaging illusion.” Instead, inflation and unemployment “rose inexorably together.”
What happened was exactly what Friedman had predicted. Policymakers kept trying to push unemployment below the natural rate with expansionary policy. Each round of stimulus temporarily reduced unemployment but ratcheted inflation expectations higher. When the stimulus wore off, unemployment returned to (or exceeded) its natural rate, but inflation stayed elevated. The short-run Phillips curve kept shifting upward, tracing out the vertical long-run relationship.
Why This Matters for Policy
The vertical long-run Phillips curve carries a direct implication: monetary policy cannot permanently lower unemployment. It can only determine the long-run inflation rate. This realization reshaped central banking worldwide. The Federal Reserve Bank of New York has noted that “the general acceptance of the view that there is no long-run trade-off between inflation and unemployment” is one of the core reasons central banks now focus on price stability as their primary long-run goal.
This is the logic behind inflation targeting. If you can’t buy lower unemployment with higher inflation in the long run, then the best thing a central bank can do is keep inflation low and predictable. A clear inflation target also solves what economists call the “time-inconsistency problem,” the temptation for policymakers to juice the economy with surprise inflation for short-term political gains. By committing to a target (the Fed uses 2%), the central bank ties its own hands and signals to workers and firms that their inflation expectations should stay anchored.
None of this means short-run trade-offs are irrelevant. Central banks still respond to recessions by accepting temporarily higher inflation to bring unemployment down faster. The point is that these interventions are temporary by nature. Once expectations adjust, unemployment gravitates back to the natural rate. The long-run Phillips curve is vertical because the only variable monetary policy controls in the long run is the price level, not the real economy.