The Relevance of a Post-Pandemic Phillips Curve

The following is from Arjun Iyer, a sophomore at Kelley in Knall-Cohen and the Investment Management Workshop. Arjun previously contributed to a blog post titled "AI in Healthcare Providers"

From Arjun Iyer: The Phillips Curve is a staple in any high school macroeconomics class. Interestingly, it has made a reappearance in recent years, but its revival also highlights how modern economic dynamics challenge its original formulation. When we look at the pandemic from a very high level (and what came after), we can sort of get an idea of how the curve’s mechanics are more situational than universal. This “situational” thesis isn’t new, as prominent economists like Olivier Blanchard have long questioned the relevance of the Phillips Curve in modern policy frameworks. Blanchard (a former IMF chief economist) argued that inflation’s sensitivity to labor conditions has weakened over time due to globalization, automation, and central banks' success in anchoring expectations. His work in the 2010s predicted the flattening we observed before COVID-19, where unemployment could fall without causing meaningful inflation.

However, COVID disrupted every rational assumption, including Blanchard’s view of the Phillips Curve. The tight labor market during the recovery (where the vacancy to unemployment ratio hit 2:1) led to rapid wage growth in industries like healthcare and hospitality. These labor-driven inflationary pressures confirmed, for a time, that wage-price dynamics could still fuel inflation under the right conditions. This is visually captured in the labor market tightness indicator graph below, which shows the sharp increase in the vacancy-to-unemployment ratio during the pandemic recovery, far exceeding the pre-pandemic average of 1.24. This trend substantiates how labor market mismatches, rather than general labor availability, drove inflation during this period.

The crisis also exposed something fundamental about inflation: aggregate labor conditions weren’t the only factor; specific bottlenecks and localized labor shortages were just as important. This realization ties into the thesis of why the Phillips Curve, which assumes a linear relationship between unemployment and inflation, struggled to capture the nuances of post-pandemic inflation. The Beveridge Curve graph below illustrates how structural shifts in the labor market (visible in the misalignment between vacancies and unemployment) contributed to inflation even when traditional labor market slack was minimal.

For example, Ben Bernanke (the former Federal Reserve chair) and Olivier Blanchard published a 2023 paper where they explored the recent inflation surge (paper can be downloaded here). They noted that while supply shocks (like energy disruptions) initiated the inflationary burst, it was tight labor markets that sustained it, particularly in sectors with persistent worker shortages. They emphasized that central banks needed to act decisively to prevent inflation expectations from becoming unanchored, which could otherwise push economies into a self-reinforcing inflation spiral. This is precisely what the Fed did by raising interest rates aggressively throughout 2022 and 2023, which ultimately calmed inflation to less than 3% in 2024.

Blanchard’s analysis aligns with what we saw in the trucking industry, where supply chain disruptions and driver shortages drove wage increases and freight costs, amplifying inflation across the economy. However, Bernanke’s point about inflation expectations is key since wage growth alone wouldn’t have fueled inflation without the fear that it might spiral. The Fed’s aggressive tightening re-anchored these expectations, thus cutting off that potential feedback loop before it became entrenched.

The parameterization table provides further insights into this issue by modeling how small decreases in the unemployment rate require strong assumptions about labor market behavior. Specifically, the table breaks down the effects of different labor market parameters (complements vs. substitutes) and shows how shifts in labor market elasticity affect unemployment and inflation outcomes. This helps explain why inflation persisted despite minor changes in unemployment post-COVID, labor markets had become structurally tighter, requiring greater slack or more dramatic changes to curb inflation effectively.

Still, we need to ask: is the Phillips Curve now back for good, or was this just temporary? Economists like Ricardo Reis (teaches at LSE) argue that the pandemic period was exceptional leading to a temporary misalignment of supply and demand rather than a lasting shift in the inflation-unemployment relationship. He suggests that automation and technological investments will likely re-flatten the Phillips Curve moving forward. With companies increasingly turning to AI and software solutions (projected to grow by 7.5% annually), labor markets may not play the same role in driving inflation that they once did. This suggests that wage-driven inflation, while real during the pandemic, might be less likely to reoccur in the long run.

In addition, remote work has expanded labor market flexibility, allowing companies to hire from global talent pools. This limits domestic wage pressures and further weakens the traditional wage-inflation link that the Phillips Curve assumes. Even with tight U.S. labor markets, inflation may not follow the old playbook when firms can outsource tasks to cheaper labor markets abroad.

So, where does the Phillips Curve stand now? I would say that it’s still useful in explaining localized inflation scenarios (like what we saw with hospitality and logistics). To be frank, it falls short as a broad macroeconomic tool. Central banks are increasingly relying on the vacancy-to-unemployment ratio (the Beveridge Curve) for more nuanced insights into labor market dynamics. This metric, which captures mismatches between job openings and available workers, has shown a stronger correlation with wage growth during the post-pandemic period than the Phillips Curve ever did.

In the end, the Phillips Curve’s brief reappearance was more like an emergency flare than a genuine sustained revival. It worked in the unusual conditions of the pandemic recovery where supply disruptions, fiscal stimulus, and labor shortages all collided but it’s unlikely to remain a reliable guide in the coming years. Structural shifts in technology, globalization, and labor markets will continue to weaken the link between wages and inflation, pushing policymakers to look elsewhere for inflation-forecasting tools. While the curve isn’t entirely irrelevant, its days as the dominant framework for understanding inflation are likely behind us. Central banks will need to be nimble, drawing on multiple indicators rather than relying on a single, outdated model.