The Phillips curve is an economic concept describing a tradeoff between inflation and unemployment: when unemployment is low, inflation tends to rise, and when unemployment is high, inflation tends to fall. It originated from a 1958 study by economist A.W. Phillips, who analyzed nearly a century of data from the United Kingdom and found a clear, inverse relationship between wage growth and unemployment. The idea has shaped how central banks think about interest rates and economic policy ever since, though the relationship has grown weaker and more complicated over the decades.
Where the Idea Came From
Phillips studied UK data from 1861 to 1957, plotting unemployment rates against the rate of change in wages. The pattern was striking: years with low unemployment consistently showed faster wage growth, while years with high unemployment showed sluggish or falling wages. The relationship was also nonlinear, meaning wages accelerated much faster as unemployment dropped below a certain threshold than they slowed down when unemployment climbed above it.
Other economists quickly extended this finding. Instead of just looking at wages, they connected it to overall price inflation. The logic was straightforward: when businesses pay higher wages, they pass those costs on to consumers through higher prices. This broader version, linking unemployment directly to consumer price inflation, became the standard Phillips curve taught in economics courses and used by policymakers.
How the Tradeoff Works
The mechanism runs through the labor market. When the economy is strong and demand for goods and services is high, businesses hire more workers, and unemployment falls. As the pool of available workers shrinks, employers have to offer higher pay to attract and retain people. Those rising labor costs push businesses to raise prices. The economy moves “up and to the left” along the curve: lower unemployment, higher inflation.
The reverse happens during a downturn. Weaker demand means layoffs, fewer job openings, and less pressure on employers to raise wages. With labor costs stable or falling, price increases slow down. This is the core policy dilemma the Phillips curve presents. Policymakers who want to push unemployment very low may have to accept higher inflation as a consequence, and those who want to stamp out inflation may need to tolerate a period of higher unemployment.
The Natural Rate of Unemployment
Economists eventually added a critical refinement: the idea that there’s a specific unemployment rate where inflation neither speeds up nor slows down. This is often called the NAIRU (non-accelerating inflation rate of unemployment) or the “natural rate.” If unemployment drops below this threshold, inflation doesn’t just stay elevated, it keeps climbing. If unemployment sits above it, inflation gradually fades.
The Congressional Budget Office currently estimates the U.S. noncyclical unemployment rate at about 4.3%, and this number has trended downward in recent decades. It’s not a fixed target. Structural changes in the economy, like shifts in workforce demographics, technology, or how easily workers can switch jobs, move the natural rate over time. Central banks watch this threshold closely when deciding whether to raise or lower interest rates.
Why the Curve Has Flattened
One of the biggest puzzles in modern economics is that the Phillips curve relationship has weakened considerably since the 1980s. For much of the 2010s, U.S. unemployment fell steadily without producing the inflation spike the curve would have predicted. Research published in the Journal of Economics and Business found that two forces explain most of this flattening.
The first is globalization. When companies can source labor and materials from around the world, a tight domestic labor market doesn’t create as much cost pressure. Workers competing in a global market have less individual bargaining power than they would in a closed economy.
The second is better-anchored inflation expectations. Since the early 1980s, the Federal Reserve has built a track record of keeping inflation near its target. When businesses and workers expect inflation to stay around 2%, they set prices and negotiate wages accordingly, creating a self-fulfilling stability. This contrasts sharply with the 1970s, when rising inflation expectations caused workers to demand ever-higher wages, which pushed prices up further in a spiral.
What Supply Shocks Do to the Curve
The classic Phillips curve describes what happens when demand shifts, like a stimulus package boosting spending or a recession pulling it back. But some of the most dramatic inflation episodes come from the supply side: oil price spikes, pandemic-related supply chain breakdowns, or bad harvests. These shocks don’t move the economy along the curve. They shift the entire curve.
A supply shock can produce rising inflation and rising unemployment at the same time, something the original Phillips curve can’t explain. Economists at the Cleveland Fed have noted that these “markup shocks” cause inflation and economic output to move in opposite directions, making the curve extremely difficult to estimate in real time. The 1970s oil crises are the textbook example: prices surged while unemployment climbed, a combination called stagflation that initially seemed to discredit the Phillips curve entirely.
The Pandemic Put It to the Test
The COVID-19 era gave economists a dramatic natural experiment. Unemployment spiked to historic levels in 2020, fell rapidly through 2021 and 2022, and inflation surged to 40-year highs before retreating. The relationship between inflation and unemployment weakened further during this period, according to research examining the New Keynesian version of the Phillips curve.
Federal Reserve Governor Chris Waller captured the confusion in 2023, asking publicly what had happened to the Phillips curve “that it is acting so differently now than in the pre-pandemic period.” Part of the answer is that the pandemic created simultaneous supply and demand shocks of unusual size. Supply chains broke down while government stimulus flooded the economy with spending power. Disentangling those forces from the normal unemployment-inflation tradeoff has been a major challenge.
Recent Fed research has also found evidence that the curve is nonlinear in an important way: the tradeoff between inflation and unemployment is steeper when the labor market is very tight than when it’s loose. In practical terms, this means pushing unemployment from 5% to 4% might produce a noticeable inflation bump, while pushing it from 7% to 6% might barely register. This finding suggests that policymakers face a fundamentally different set of risks depending on where the economy sits at any given moment.
How Policymakers Still Use It
Despite its limitations, the Phillips curve remains a central framework at the Federal Reserve and other central banks. When the Fed raises interest rates to cool the economy, the underlying logic is Phillips curve reasoning: slowing demand will ease labor market pressure, which will reduce wage growth, which will bring down inflation. The curve isn’t treated as a precise forecasting tool anymore, but it provides the conceptual backbone for how monetary policy decisions get made.
The practical takeaway is that the tradeoff between inflation and unemployment is real but messier than the original curve suggested. It shifts with supply shocks, flattens when inflation expectations are well-anchored, steepens when the labor market gets very tight, and weakens as economies become more globalized. Understanding the Phillips curve means understanding both the core insight, that labor market conditions and inflation are connected, and the long list of reasons that connection keeps surprising economists.