Research                                                                                                                 

Job Market Paper:

• Reconciling Wage and Price Phillips Curves

This paper uncovers a significant discrepancy between the Wage and Price Phillips Curves. The Price Phillips Curve steepens beyond the Wage Phillips Curve during tight labor markets in aggregate and cross-sectional U.S. data as well as in country panel data. This discrepancy presents a puzzle to the    conventional wage-driven Phillips Curve view. I resolve this puzzle with firms responding to transitory upward wage pressure under downward nominal wage rigidity by pricing out demand rather than excessively increasing wages, which would lead to higher future marginal cost. Additional empirical evidence from market concentration and firm survey data supports key implications of the channel. I embed the mechanism in a New Keynesian model with monopsony power, yielding the first DSGE model incorporating Kudlyak’s User Cost of Labor. My channel predicts almost painless disinflation after transitory demand and labor supply shocks and challenges the notion that ever-tighter labor markets are unambiguously good for real wage growth.

Working Papers:

• Geopolitical Risk and Import Dynamics - Evidence from US Customs Data (with M. Khalil & F. Strobel)

In recent years, major export economies have experienced rising geopolitical risk. Taking the perspectives of the US and the euro area, we employ a detailed product data panel to study the consequences of trading partner geopolitical risk shocks on domestic imports. We find that, on average, trading partner geopolitical risk shocks lower import volumes and raise import prices. The decline in imports is particularly strong when geopolitical risk shocks hit countries that exhibit a greater geopolitical distance to the US and the euro area, or when geopolitical risk shocks hit countries which are under US sanctions. A case in point are large effects for geopolitical risk shocks in China. Thus, increasing geopolitical risk triggers dynamics that may eventually lead to a fragmentation of global trade.

• Uncertainty and the Dislike of Inflation - Evidence from Retirement Insecurity

Survey evidence shows that people dislike inflation due to the short-term loss of real income and uncertainty about future living standards. While previous research has mainly focused on the small short-term income effect, this study also examines the impact of inflation uncertainty using data from the Michigan Survey of Consumers. It finds that both expected inflation and inflation uncertainty negatively affect future living conditions, with uncertainty playing the more significant role. The effect of inflation uncertainty is comparable to expecting a substantial real income drop, a 21 percentage point increase in the probability of job loss, or a 7 percentage point rise in expected inflation. These findings highlight that inflation uncertainty significantly influences expectations of long-term living standards, and should be considered when assessing the public's dislike of inflation.

• Sectoral and State Evidence on the Pandemic Beveridge Curve Shift (with N. Kodua & J. Zheng)

The national Beveridge curve shifted outward significantly during the COVID-19 pandemic, but the underlying determinants remain debated. This paper investigates these drivers using cross-sectional variation across states and industries. We introduce a novel geometric measure to quantify shifts, defined as the distance along the 45-degree line between pre-pandemic and pandemic-period curves. Industry-level analysis shows larger shifts in sectors with lower wages and limited remote-work options. Through a Diamond-Mortensen-Pissarides framework decomposition, we demonstrate that changes in matching efficiency and separation rate correlate with these industry traits respectively. State-level shifts are strongly associated with industry composition and COVID-19 infection rates. Our findings suggest that expanded unemployment insurance, limited telework capacity, and COVID-19 severity jointly drove the aggregate Beveridge curve shift during the pandemic.

 Work in Progress:

• International Evidence of Non-Linear Phillips Curves - The Roles of Market Concentration and Migration 

This paper is the first to test labor market tightness-based Phillips Curves in a cross-country panel estimation. I find evidence for the non-linearity of both Price and Wage Phillips Curves. The non-linearity of the Price Phillips Curve is crucially driven by Market Concentration. The non-linearity of the Wage Phillips Curve is also heavily influenced by market concentration, albeit in the opposite direction. Finally, I find that international migration into a country also flattens the otherwise steeply rising Wage Phillips Curve. 

• Reconciling Pandemic Inflation Narratives

I reconcile the diverging pandemic inflation narratives of Bernanke & Blanchard (2023) and Ball, Leigh, & Mishra (2022). One key aspect is allowing for the direct impact of labor market tightness on inflation. Thus, I highlight the importance of the labor market in the recent inflation bout. A proper assessment of the relative importance of demand and supply shocks under inclusion of the direct impact channel is currently in progress. 

• State Level Inflation Measures

This paper provides a methodology for constructing state-level Consumer Price Indices (CPIs) that addresses the limitations of existing approaches, particularly the use of national consumer expenditure weights and disregard of the CPI sample stratification by Hazel et al. (2020). By leveraging data from 75 Core Based Statistical Areas (CBSAs) that are representative in both the Consumer Expenditure (CE) survey and CPI sampling, I develop appropriately weighted state level CPIs. The project entails two main steps: generating state-level CE estimates from CBSA data and employing Entry Level Item (ELI) prices to compute state-level CPIs using these local expenditure weights. My methodology makes full use of the stratified nature of CE and CPI sampling, thus conforming to the methodology used to generate national CE and CPI measures. This allows me to compute CPI measures for all fifty states and D.C. instead of the limited range of data available from existing methods. Finally, I developed a bootstrap method to assess the standard error bands of low-coverage states to allow for inverse-weighted panel regressions.