Research

Abstract: We use the high-frequency, decentralized implementation of Stay-at-Home orders in the U.S. to disentangle the labor market effects of SAH orders from the general economic disruption wrought by the COVID-19 pandemic. We find that each week of SAH exposure increased a state's weekly initial unemployment insurance (UI) claims by 1.9% of its employment level relative to other states. A back-of-the-envelope calculation implies that, of the 17 million UI claims between March 14 and April 4, only 4 million were attributable to SAH orders. We present a currency union model to provide conditions for mapping this estimate to aggregate employment losses.

Abstract: We provide a guide to using autoregressive distributed lag models for impulse response estimations with an identified structural shock or an external instrument for the shock. We illustrate how specifications widely used in practice can lead to inconsistent and inefficient estimators. We further review empirical results from previous papers and show that some results appear to be statistical artifacts. We propose a simple method to avoid such a false conclusion from biases or large standard errors and obtain consistent and precise estimates.

Abstract: This paper shows that more procyclical returns to scale in recent decades have flattened the Phillips curve when conventional activity measures, such as the output gap, labor gap, and labor share, are used as forcing variables. In contrast, the marginal cost Phillips curve remains steep. Using a simple, intuitive model with a translog production function, we illustrate a novel channel linking input complementarity and procyclical returns to scale to the identified slopes of the Phillips curves. By utilizing the estimated production functions and quantitative models, we emphasize the importance of our mechanism for rationalizing the US inflation data.

Abstract: We investigate how unexpected changes in inflation influence real economic activity by decomposing inflation surprises into demand-driven and supply-driven components using high-frequency stock market reactions around CPI releases. Employing a Bayesian structural VAR with sign restrictions, we find that demand-driven inflation surprises boost industrial production, investment, and retail sales while reducing market uncertainty (VIX). In contrast, supply-driven surprises decrease real activity and sharply increase uncertainty. The uncertainty mechanism explains why identical inflation news has opposite economic effects, with corresponding patterns in financial markets: demand-driven surprises generate negative stock-bond correlation while supply-driven surprises produce positive correlation. Results replicate in UK data.

Abstract: How do agents form their macroeconomic expectations and how do they incorporate them into their economic decisions? Do they think about each variable independently or do they revise expectations jointly? Using experimental evidence from the U.S. online labor market, we show that when people receive one relevant piece of information, they simultaneously update their expectations about multiple macroeconomic variables. For example, when people receive information about the price inflation rate, they revise not only their price inflation expectations but also their aggregate wage growth and unemployment expectations. Exploiting exogenous variation in expectations arising from randomized information provision, we document that such simultaneous revision of expectations has important implications for the likelihood of wage-price spirals. Specifically, we show that, after controlling for wage growth and unemployment rate expectations, higher price inflation expectations result in a downward revision of reservation wages, implying that households perceive inflation as a bad signal about the economy. These results suggest that the risk of wage-price spirals was limited in the U.S. in 2022, despite the high inflation rates.

Abstract: The costs of business cycles are not evenly distributed in the economy. Nevertheless, monetary policies have relied only on “aggregate” labor market variables, ignoring this heterogeneity in the incidence of economic fluctuations. How can monetary policy be more inclusive and benefit people who are more vulnerable to economic fluctuations? To shed light on this question, which is at the core of policy discussions, I study heterogeneity (“types”) in labor market arrangements and implications of this heterogeneity for welfare and optimal monetary policy. I document that the experiences of regular and irregular workers over the business cycle differ considerably. For example, the share of irregular workers in employment rises during recessions, suggesting that firms actively adjust labor composition over the business cycle. I develop a tractable New Keynesian model with regular and irregular labor types that reflect the cyclical nature of labor composition. I find that workers, who are marginally attached to either the regular or the irregular labor market, face larger volatilities in their consumption and disutility from labor supply and hence suffer larger welfare losses over the business cycle. I find that optimal monetary policy rule should react to employment dynamics in specific segments of the labor market than the overall stance of the labor market. When a central bank follows that rule, it benefits not only people who are more vulnerable to economic fluctuations but generate higher economy-wide welfare.

Abstract: Recent work suggests that when central bank releases information, they could alter expectations and spending in directions that potentially reverse the expected effect of policy changes. In this paper, we estimate the extent to which expectations changes depend on explicit information given by central banks. We compare impulse responses to high-frequency monetary policy surprises during announcements when the Bank of England also releases a detailed inflation report to those where a simple press statement is released. We find that when a simple press statement is released policy has conventional signs: output and inflation fall following a surprise tightening. However, when a detailed inflation report is released, surprise tightening raise GDP and inflation suggesting the information effect can be controlled by central banks.