Knowledge that Transforms

Thanks to standardized work protocol and technology of McDonald's restaurants, the hourly wage of McDonald's Basic Crew enables wage comparisons under near-identical skill inputs and hedonic job conditions. McWages capture labor costs in entry-level jobs, while the Big Macs (earned) Per Hour (BMPH) index measures corresponding purchasing power of wages. We document large and growing geographical wage differences in standardized jobs using data covering most U.S. counties during 2016-2023. Before the Covid-19 pandemic, there was no BMPH growth where minimum wages stayed constant, but the pandemic wage increase, which diminished the importance of minimum wages, was stronger in these areas.

We study self-organized teams in dynamic contests. Using data from Kaggle, we document that teams outperform solo players, but few players choose to form teams. Every new team alters the composition of players, discouraging less productive solo players to make submissions. We estimate the structural parameters of a dynamic contest model, including the team formation and submission costs. We find that team formation incentives diminish with the number of teams, as do the incentives to make submissions. We empirically evaluate the productivity-discouragement tradeoff caused by teamwork and discuss implications for contest design, including facilitating teamwork and hosting open competitions.

Previously, identification of triangular random coefficient models required a restriction on the dimension of the first stage heterogeneity or independence assumptions across the different sources of the heterogeneity. This note proposes a new identification strategy that does not rely on either of these restrictions but rather assumes conditional means have a conditional linear projection representation in order to construct “correction functions” to address endogeneity and gain identification of the average partial effect. This identification strategy allows for both continuous and discrete instruments. Finally, the proposed identification method is illustrated in estimating the returns to education.

We study the effects of uncertainty on time use and their macroeconomic implications. Employing data from the American Time Use Survey and the Bureau of Labor Statistics, we document that heightened uncertainty increases housework and reduces market hours, mildly impacting leisure. We then propose a model that quantitatively accounts for these estimates. We show that substitution between market and housework provides self-insurance to households, weakening precautionary motives. However, it also reduces aggregate demand, ultimately amplifying uncertainty's recessionary impact. Time reallocation can lead to higher inflation, particularly when uncertainty is coupled with policies redirecting time use towards housework (e.g., lockdown restrictions).

The effect of the full treatment is a primary parameter of interest in policy evaluation, while often the effect of a subset of treatment is estimated. We partially identify the local average treatment effect of receiving full treatment (LAFTE) using an instrumental variable that may induce individuals into subsets of treatment (partial compliers). We show that partial compliers violate the standard exclusion restriction, necessary conditions on the absence of partial compliers are testable, and partial identification holds under a double exclusion restriction. We identify partial compliers in four applications and estimate informative bounds on the LAFTE in three of them.

Libraries deliver the majority of US book consumption, and library costs are kept low by the first sale doctrine's guarantee that libraries can lend print books without additional compensation to publishers. However, publishers' decisions to restrict ebook access may threaten libraries. We measure impacts of holdings on circulation; and library holdings choices suggest that library managers value instances of ebook circulation more than print. We use a structural model of patron demand and library holdings to show that more restrictive print book terms would have ten times the negative patron welfare impact of an equivalent restriction on ebooks.

In 1966, Southern hospitals were barred from participating in Medicare unless they discontinued their longstanding practice of racial segregation. Using data from five Deep South states and exploiting county-level variation in Medicare certification dates, we find that gaining access to an ostensibly integrated hospital had no effect on Black postneonatal mortality. Similarly, there is little evidence that the campaign contributed to the trend towards in-hospital births among Southern Black mothers. These results are consistent with descriptions of the hospital desegregation campaign as producing only cosmetic changes and illustrate the limits of anti-discrimination policies imposed upon reluctant actors.

Current Editorial Board:Treb Allen, Dartmouth CollegePierre Azoulay, Massachusetts Institute of TechnologyWill Dobbie (Co-Chair), Harvard UniversityRaymond Fisman (Co-Chair), Boston UniversityBenjamin R. Handel, University of California, BerkeleyPeter Hull, Brown UniversityBrian A. Jacob, University of MichiganScott Kominers, Harvard UniversityTavneet Suri, Massachusetts Institute of TechnologyStephen Terry, University of MichiganRecently Retired from the Editorial Board:Olivier Coibion, University of Texas, AustinDaniel Xu, Duke UniversityTable 1—Manuscripts Submitted and PublishedThis table shows the trends in papers submitted and published over the past five years. The journal saw increased growth in submissions of papers from 2019 to 2020. Submission numbers remained steady from 2020 to 2021, declined by 10% in 2022, and then remained steady from 2022 to 2023. The number of published papers increased by 25 in 2022 and an additional 17 papers in 2023 due to the publication of larger issues.Table 2—Status of Manuscripts by Year of Submission, 2019–2023This table shows the status of manuscripts submitted over the past five years. The number of papers summarily rejected has remained steady, ranging from 54% to 61% of papers submitted. The percent of papers sent out for review but ultimately rejected has also remained steady, ranging from 30% to 35% of papers. Acceptance rates range from 2% to 11% of papers submitted. Papers submitted in 2023 have not had enough time to reflect the acceptance rate.Table 3—Decision Time for Manuscripts Sent to Referees for ReviewThis table reports the decision time for manuscripts sent to referees for review over the past five years. The average decision time for papers that were sent to referees has remained steady, ranging from 82 to 95 days.Table 4—Distribution of First Decision Times, by Submission YearThis table includes all first-round paper submissions and shows the distribution of decision times in monthly increments for the past five years. This table also reflects the improvement in the journal's decision turnaround times. In 2023, 82% of manuscripts received a decision within three months and 99% within six months.Table 5—Subject Matter of Published Manuscripts, 2023This table shows the distribution of subject matter of papers published in 2023. These papers covered a variety of subjects, but the most common published topics were microeconomics; mathematical and quantitative methods; and health, education, and welfare. Table 1.Manuscripts Submitted and Published, 2019–2023YearSubmittedPublished20191,3306620201,6666520211,6746520221,4809020231,445107

Positive assortative matching increases both the wages of more productive workers and wage dispersion. We study the effect of immigration on positive assortative matching using French employer-employee data from 1995 to 2005. We find that increases in the share of immigrants, driven by historical networks across local labor markets, generated stronger positive assortative matching between workers and firms. We present evidence suggesting that this effect was associated to higher wages for more productive workers and that the findings are consistent with increased workers' screening by firms.

We examine how changes in the salience of workplace risk affect police behavior and public safety. Specifically, we investigate cases of police officer deaths while on duty. Officers respond to a peer death by decreasing arrest activity for one to two months, consistent with heightened fear. Reductions are largest for low-level arrests and are more pronounced in smaller cities. Crime does not increase on average during this period, nor do we observe crime spikes in cities with larger or longer arrest declines. While shocks to perceived fatality risk generate substantial enforcement responses, officer fear is unlikely to harm public safety.