Reassessing the Phillips Curve: A Steeper Perspective on Inflation Dynamics

Discussions surrounding inflation's relationship with real economic activity have long been pivotal in macroeconomic theory and monetary policymaking. Central to this dialogue is the Phillips curve (PC), which quantifies the response of inflation to changes in economic conditions. The curve's steepness is particularly crucial—if steep, it indicates that inflation reacts quickly to economic growth or contraction, necessitating proactive measures from central banks to stabilize inflation around their targets. Prior research has often reported that the slope of this curve is minimal, suggesting a "flat" Phillips curve. However, coauthored research refutes this by demonstrating that the Phillips curve remains "alive and steep" when examining real marginal costs rather than standard outputs or employment measures.

The Conventional Phillips Curve Framework

The established interpretation of the Phillips curve posits a connection between inflation, future inflation expectations, and economic slack. Typically, economic slack is represented by either output gaps or unemployment gaps—indicating how actual output deviates from potential output. In essence, inflation is expected to increase as the economy approaches its full capacity and decrease during recessions. The steepness of the curve, denoted by the parameter K, is indicative of inflation's sensitivity to these fluctuations. However, a substantial body of research has concluded that the Phillips curve is relatively flat, leading to ongoing debates about the necessary aggressiveness of monetary policy to control inflation.

The Cost-Based Phillips Curve: A Different Approach

Alternatively, one can view inflation dynamics through the lens of firm-level pricing strategies responding to fluctuations in production costs. In this framework, the Phillips curve is centered on real marginal costs measured as a percentage deviation from a trend, rather than relying on output or unemployment gaps. This perspective suggests inflation is driven by rising production costs that firms face. The slope of this cost-based Phillips curve indicates the degree to which changes in production costs reflect in consumer prices.

Determining the Slope of the Cost-Based Phillips Curve

Estimating the precise slope of the cost-based Phillips curve poses challenges, especially when using aggregate data, which is often influenced by multiple concurrent shocks to inflation and economic conditions. To circumvent these complexities, researchers have turned to detailed microdata from firms. In a recent study collaboration involving economists Mark Gertler, Luca Gagliardone, and Joris Tielens, detailed price and cost data revealed how firms adjust their pricing in relation to cost changes. This methodology allowed for an assessment of nominal rigidities—where price adjustments are infrequent—and real rigidities stemming from competitive pricing strategies, shedding light on the Phillips curve's slope.

The findings indicate a significant link between inflation and real economic conditions, underscored by the costs borne by producers. On average, firms maintain fixed prices for three to four quarters, signifying considerable nominal rigidity. Moreover, firms often adjust pricing less aggressively due to a desire to follow competitors, effectively reducing the pass-through rate of cost shocks to consumer prices. This analysis estimates the cost-based Phillips curve slope to be three to ten times that of the traditional output-based formulation, illustrating a steeper curve across normal economic periods.

Insights on Aggregate Inflation Dynamics

Understanding the role of cost pass-through is essential for grasping aggregate inflation trends. By constructing a cost index that assimilates firm-level labor and input costs and integrating it into the cost-based Phillips curve, the researchers examined inflation in the Belgian manufacturing sector. Their model-generated inflation closely mirrored actual producer price inflation (PPI), with a correlation coefficient exceeding 80 percent. This reinforces the conclusion that shifts in production costs substantially account for approximately 70 percent of inflation fluctuations, reinforcing the influence of costs on inflation dynamics.

Reconciling the Cost-Based and Output-Based Phillips Curves

The stark contrast between the steepness of the cost-based Phillips curve and the flat nature of the output-based curve prompts a critical inquiry: why does the latter appear flat while the former remains steep? Research indicates that this perceived flatness arises from a weak relationship between output gaps and marginal costs. Simply put, while changes in production costs are directly reflected in firms’ pricing, the relationship between output fluctuations and inflation is tenuous.

The conventional Phillips curve remains valid only under the condition of a proportionality between marginal costs and output gaps, an assumption that typically hinges on perfectly flexible wages. When these conditions are unmet, output gaps serve poorly as proxies for real marginal costs, resulting in downwardly biased Phillips curve estimations. In empirical studies focused on the pre-pandemic period from 1999 to 2019, a low elasticity of marginal cost with respect to output was found, explaining the divergence in slope steepness between the two Phillips curve formulations.

Implications for Post-Pandemic Inflation Trends

The strong link between marginal costs and inflation underscores the relevance of the cost-based Phillips curve in comprehending inflationary dynamics. Conversely, the weak correlation between output and marginal costs elucidates the flatter appearance of the conventional Phillips curve. This low elasticity can typically be attributed to two factors: stable short-term returns to scale in firms’ cost structures, and the rigidities in wage adjustments that hinder feedback from demand to production costs.

However, the pandemic has highlighted the potential for rapid shifts in these relationships under conditions of economic stress. Substantial shocks in labor markets or supply chains have compelled firms to confront capacity constraints, resulting in sharp increases in marginal costs and subsequent inflation. Notably, the slope of the Phillips curve can also evolve. Data from the pre-pandemic era indicated stable price adjustment frequencies, while more recent observations reveal that firms have been adjusting prices far more frequently, enhancing the relationship between costs and inflation and giving rise to non-linear inflation dynamics. Further exploration of these dynamics will be detailed in an upcoming piece.

Simone Lenzu serves as a financial research economist at the Federal Reserve Bank of New York.