Unraveling the Phillips Curve: Unemployment, Inflation & Economic Trade-offs
Ever since I first dipped my toes into the world of economics, few concepts have fascinated and frustrated me quite like the Phillips Curve. It’s one of those foundational ideas that seems so simple on the surface, yet, like a chameleon, it constantly shifts and adapts to the economic landscape, challenging our understanding. For decades, it offered a seemingly straightforward trade-off: want lower unemployment? Be prepared to accept a bit more inflation. Sounds neat, right? But the real world, as always, is far messier than our models.
Back in 1958, A.W. Phillips, a New Zealand economist, published a groundbreaking paper. He looked at over a century of data from the UK, specifically the relationship between wage inflation and unemployment. What he found was striking: an inverse relationship. When unemployment was low, wages tended to rise faster, implying higher inflation. When unemployment was high, wage growth slowed and sometimes even fell. It was a revelation, suggesting policymakers could, in theory, pick a point on this curve – a little less unemployment for a little more inflation or vice versa.
This initial observation became a cornerstone of macroeconomic thinking. For a while, it felt like central banks had a clear menu of choices. Say, a government wanted to boost employment; it could stimulate the economy, pushing down unemployment and just accept the resulting bump in prices. I remember my econ professor vividly describing it as a “policy menu,” a straightforward guide for steering the economic ship.
But as we all know, economic relationships aren’t static. By the 1970s, something odd started happening. We saw “stagflation” – high inflation and high unemployment simultaneously. This was a real head-scratcher and seemingly defied the Phillips Curve’s logic. What went wrong?
Enter economists like Milton Friedman and Edmund Phelps. They argued that A.W. Phillips’s original observation missed a crucial ingredient: inflation expectations. If people expect prices to rise, they’ll demand higher wages and businesses will pass those costs on. This creates a self-fulfilling prophecy. They posited that in the long run, there’s no stable trade-off between inflation and unemployment. The economy would always gravitate back to its “natural rate of unemployment” (now often called the Non-Accelerating Inflation Rate of Unemployment or NAIRU), regardless of the inflation rate.
Think of it this way: in the short run, if the central bank surprises everyone with more stimulus, unemployment might dip below its natural rate, causing inflation. But once people catch on and adjust their expectations, that initial boost to employment fades and you’re just left with higher inflation. In my years observing the markets, I’ve seen firsthand how crucial expectations are to economic outcomes. It’s not just about what is happening, but what people think will happen.
This distinction gave us the idea of a short-run Phillips Curve, which can shift depending on inflation expectations and a vertical long-run Phillips Curve at the NAIRU. Any attempt to permanently push unemployment below this natural rate would only lead to ever-accelerating inflation.
So, where does that leave us today? The Phillips Curve hasn’t gone away, but its form and reliability are constantly debated. It’s certainly not the simple, stable relationship of Phillips’s initial discovery.
One of the big questions floating around boardrooms and academic seminars is whether the Phillips Curve has flattened significantly. Meaning, even big swings in unemployment seem to have only a modest impact on inflation. Why might this be?
- Globalization and Supply Chains: Global competition can cap price increases even when domestic demand is strong. If you can import goods cheaply, domestic producers might find it harder to hike prices, regardless of how tight the local labor market is.
- Anchored Inflation Expectations: Central banks have worked hard over the past few decades to anchor inflation expectations, convincing people that they will keep inflation low and stable. If people believe this, they’re less likely to demand huge wage increases or pass on costs quickly, even during periods of low unemployment.
- Structural Changes in Labor Markets: The nature of work itself has changed. We’re seeing situations where labor market cooling doesn’t necessarily translate to high unemployment, but rather to a shortage of specific skills. For instance, recent analysis of the Russian labor market highlights that while it’s “gradually cooling,” high unemployment isn’t the threat. Instead, the “main challenge for the economy is not so much labour shortages as a lack of highly qualified specialists” (Irina Ryabova, Econs, “Economic Mirror,” July 21, 2025). This kind of structural mismatch can muddy the waters of a simple unemployment-inflation trade-off.
Consider the insightful work of Mauricio Ulate, a Senior Economist at the Federal Reserve Bank of San Francisco. His research, including a forthcoming paper in the Journal of Political Economy, delves into how “downward nominal wage rigidities” – the stickiness of wages, particularly when they resist falling – can impact unemployment and welfare, especially in the context of events like the “China shock” (Mauricio Ulate - Home). He and his co-authors found that while the China shock led to average welfare increases in most U.S. states, these nominal rigidities “reduce the overall U.S. gains by around two thirds,” with “18 states that experience welfare losses” (Mauricio Ulate - Home). This is a prime example of how labor market frictions, distinct from simple aggregate unemployment, can significantly alter economic outcomes and make the Phillips Curve relationship far more complex. It’s not just about how many people are employed, but how flexibly wages respond to shocks.
Another factor making the Phillips Curve less predictable are supply-side shocks. These are events that impact the cost of production directly, like sudden spikes in energy prices or disruptions to global supply chains, rather than changes in demand. These shocks can push up inflation without any corresponding fall in unemployment.
Take the current discussions around tariffs. Joel Prakken, co-founder of Macroeconomic Advisers and past Chief US Economist of S&P Global, has been observing these dynamics closely. He noted on July 21, 2025, that by June, recent “Trump Administration announcements” of a “baseline 10% tariff on most imported goods and additional ‘reciprocal’ tariffs” had already “increased the import-weighted average” (Joel Prakken - Haver Analytics). This is a direct cost-push inflation mechanism, independent of the unemployment rate, illustrating how non-labor market factors can drive prices.
Then there’s the concept of the output gap – the difference between an economy’s actual output and its potential output. A positive output gap (economy running hot) usually signals inflationary pressures, while a negative one (economy below potential) suggests disinflationary forces. Recent research on Colombia’s economy after COVID-19, for example, estimated a “significant 20% decline in the output gap but with a faster recovery compared to previous crises” (Camilo Granados & Daniel Parra-Amado, “Output gap measurement after COVID for Colombia,” July 21, 2025). This shows how tracking potential output and the gap can provide insights into inflationary pressures, sometimes independently of the unemployment rate.
To really grasp the labor market’s complexity, economists often look beyond just the Phillips Curve. The Beveridge Curve, for instance, offers a different lens. It plots the relationship between the unemployment rate and the job vacancy rate (unfilled jobs as a proportion of the labor force) (Beveridge curve - Wikipedia). Like the Phillips Curve, it typically slopes downward and is hyperbolic, meaning a higher unemployment rate usually comes with a lower job vacancy rate (Beveridge curve - Wikipedia).
Why bring this up? Because shifts in the Beveridge Curve can tell us about structural changes in the labor market – like improved matching between workers and jobs or inefficiencies. If the Beveridge Curve shifts outwards, it implies that for any given unemployment rate, there are more vacancies, suggesting mismatches or less efficient job searches. This kind of insight complements the Phillips Curve by providing a richer understanding of labor market health, which in turn influences wage and price dynamics.
So, is the Phillips Curve dead? Absolutely not. It’s just… complicated. Central banks and economists still pay close attention to it, but they recognize its limitations. It remains a valuable framework for thinking about the relationship between labor market slack and inflationary pressures, even if that relationship has become less direct and more prone to shifts.
Policymakers understand that while a very low unemployment rate can signal potential inflationary pressures, they also need to consider other factors: global supply chains, commodity prices, fiscal policy and, crucially, inflation expectations. It’s less about a simple trade-off and more about understanding the complex interplay of forces shaping our economies.
After years of watching economic data unfold and engaging with the brightest minds in finance, my takeaway on the Phillips Curve is this: it’s not a rigid rule, but a powerful lens through which to view the economy. It reminds us that labor markets and inflation are deeply intertwined, even if the nature of that bond changes over time. We’ve moved beyond a simplistic “menu of choices” to a nuanced understanding that acknowledges the role of expectations, supply shocks and the evolving structure of labor markets. The Phillips Curve, in its modern incarnation, is a testament to the dynamic nature of economics – always challenging, always evolving and always pushing us to refine our understanding of how the world really works.
References
What is the Phillips Curve?
The Phillips Curve illustrates the inverse relationship between inflation and unemployment rates.
How has the Phillips Curve evolved over time?
It has shifted from a simple trade-off to a more complex relationship influenced by expectations and structural changes in the economy.