Research

Publication

Working papers

Bad Times, Bad Jobs? How Recessions Affect Early Career Trajectories, with Parag Mahajan and Heiko Stüber. Revised and Resubmitted to Journal of Labor Economics

Workers who enter the labor market during recessions experience lasting earnings losses, but the role of non-pay amenities in exacerbating or counteracting these losses remains unknown. Using population-scale data from Germany, we find that labor market entry during recessions generates a 5 percent reduction in earnings cumulated over the first decade of experience. Implementing a revealed-preference estimator of employer quality that aggregates information from the universe of worker moves across employers, we find that 17 percent of recession-induced earnings losses are compensated by non-pay amenities. Purely pecuniary estimates can therefore overstate the welfare costs of labor market entry during recessions.

Breaking the Implicit Contract: Using Pension Freezes to Study Lifetime Labor Supply. Revised and Resubmitted to Journal of Political Economy: Macroeconomics

This paper studies the elimination of defined benefit pensions and subsequent adoption of 401(k) plans by U.S. employers. Using matched employer-employee data linked to thousands of firm-level retirement plan changes, it shows that unexpected losses in future compensation engendered by pension plan transitions induce premature retirement for some workers and delayed retirement for others. Heterogeneity in retirement behavior stems from differences in the relative strength of substitution and wealth effects. Exploiting treatment effects as estimation targets, it fits a model of retirement and uses the model to quantify welfare costs and labor supply effects of changes in workplace pension structure.

Finding Needles in Haystacks: Multiple-Imputation Record Linkage Using Machine Learning, with John M. Abowd, Joelle Abramowitz, Margaret C. Levenstein, Kristin McCue, Trivellore Raghunathan, Ann M. Rodgers, Matthew D. Shapiro, Nada Wasi, and Dawn Zinnser. Revised and Resubmitted to Journal of the Royal Statistical Society: Series A

This paper considers the problem of record linkage between a household-level survey and an establishment-level frame in the absence of unique identifiers. Linkage between frames in this setting is challenging because the distribution of employment across establishments is highly skewed. To address these difficulties, this paper develops a probabilistic record linkage methodology that combines machine learning (ML) with multiple imputation (MI). This ML-MI methodology is applied to link survey respondents in the Health and Retirement Study to their workplaces in the Census Business Register. The linked data reveal new evidence that non-sampling errors in household survey data are correlated with respondents' workplace characteristics.

The Consumption Response to Liquidity Shocks: Evidence from Delayed UI Benefits , with Michael Gelman and Zachary Orlando.

This paper estimates how consumption responds to an unexpected change in liquidity using transaction-level data on a sample of low-income individuals. The unprecedented increase in UI claims at the start of the COVID-19 pandemic overwhelmed state UI agencies leading to widespread, and plausibly random, processing delays. Workers who experienced processing delays were retroactively compensated with lump sum payments. UI processing delays constitute a liquidity shock because they caused a shift to both the timing and to the smoothness of benefit receipt, but left permanent income unchanged. We find that delay-affected individuals reduced spending while waiting for UI and sharply increased spending when lump sum payments arrived. Although net total spending remained unchanged due to the delay, we find that net durable spending actually increased. Our findings suggest that the mode of government transfers (lump sum vs smooth) can affect the composition of spending categories.

Selected work in progress

The Effects of Increasing the Full Retirement Age on Retirement Savings: Evidence from U.S. Tax Data, with Victoria Bryant, Jonathan M. Leganza, and Ellen Stuart.

What do Revealed Preference-Based Workplace Amenities Measure? Evidence from Surveys Matched with Administrative Data, with Parag Mahajan and Heiko Stüber.

Resting papers

Coming Home: The Consequences of Deployment to Iraq and Afghanistan.

First version: October 2015. Revised May 2019. 

This study estimates the causal effect of recent military service in Iraq and Afghanistan by exploiting plausibly exogenous deployment policies. Veterans transitioning to the civilian labor market from war-zone deployments exhibit unemployment rates that are at least 5.2 percentage points higher than those transitioning from non-war-zone deployments. Driven by increases in cognitive and sensory difficulties, disability rates for veterans recently deployed to war-zones are at least 2.6 percentage points higher. These findings indicate that adverse labor market outcomes within the veteran population may arise for reasons other than lost civilian experience.