I am excited to present “The Wheel of (Over)Time” (joint with @Rob_McDonough_) at NBER Fall Labor Studies this Friday. Full program here: https://t.co/Qk6rksvWYO. The paper is relatively new so I thought it might be worth an #econtwitter thread. 1/N

We are hopeful this setting allows us to complement survey work in this area with observational data, and answer: what do workers want? Draft (soon to be revised) is here: https://t.co/nXxA9lZWp8 12/N

Example: If Bob traded a 4 hour shift at the Oscars at 7pm on a Sunday to Sally for a 12 hour shift on Monday at 4pm at the farmer market, this tells us a lot. The wheel means Bob and Sally were endowed exogenously. 11/N

1. We can now use more transparent, revealed preference arguments to identify preferences. 2. We can identify the components of preferences: do workers care about hours, commute time, the supervisor, the team size, the day of the week, the time of day, prestige? 10/N

We just obtained initial assignments for each specific special event, and the handwritten roster of who actually showed up. This is not yet in the paper, but it allows us to extend our results in two ways. 9/N

An officer at the 75th percentile values an extra shift $94 more than an officer at the 25th. We show that b/c access costs are smaller than preference differences, informal trade achieves 94% of the maximum allocative efficiency. But we think there is more to be said! 8/N

Trade is known to occur through informal networks. We construct a proxy for this informal network, show predictive power, and disentangle access from preferences. We show that individual time-invariant preferences for additional work are the dominant form of heterogeneity. 7/N

The wheel generates “even” endowments. But officers can freely trade their initial endowments with little management intervention. Whoever shows up works and gets the money. If the market was completely free, we could read preferences from data, but there is a twist. 6/N

The City of LA assigns traffic direction at special events using “the wheel,” which is essentially an always-rotating list of workers. For those on the wheel, assignment to a particular event is exogenous. Events have value b/c of 1.5x pay and prestige (the Oscars). 5/N

Our solution: study an almost free market for labor with quasi-random endowments. Crucially, this is a market among workers with little to no manager/firm intermediation, and work is on top of a normal schedule. Does such a setting exist? Enter: Los Angeles Traffic Officers. 4/N

Answering the question poses 3 challenges. 1. Managers rather than workers often decide hours. 2. When workers can choose (i.e. gig work) there is selection. 3. Asking workers to consider counterfactual hours is difficult (especially large deviations from reality). 3/N

The key question we are after is simple: how much do workers want to work? Lots of existing and proposed policies should depend in part on the answer (for example: 996, 4-day work week, 1.5x FLSA, etc.). 2/N

I'm back, for better or worse. Many opportunities to join UNC this year. Economics hiring 3 tenure track and 1 teaching track faculty: https://t.co/f6okFHEkoR School of Data Science and Society (@UNCSDSS) hiring economists: https://t.co/QcygvWao2f Please share EconTwitter!

Very excited to visit!

UNC Econ is hiring 3 tenure-track positions ( https://t.co/KkvYrLA6He) and 1 teaching-track position ( https://t.co/4XRnOnyVG9). We are interested in new PhDs as well as advanced assistant professors/associates. Please apply and feel free to reach out!

Fantastic group with high concentration of micro/macro labor economists in Chapel Hill, and the day before in Durham as well! 👇

Forthcoming in the AER: "Supply, Demand, Institutions, and Firms: A Theory of Labor Market Sorting and the Wage Distribution" by Daniel Haanwinckel.

https://t.co/mcZdqUiHQ5

Why are wages in Paris or NYC higher than in other cities? In a new WP with @PaulineCarry & @BennyKleinman , we decompose spatial disparities btw “location effects” and the local composition of workers and establishments. New data on firm mobility + double-mover design. 🧵⬇️

Our study shows that people behave more morally in uncertain situations than in certain ones. This may be because they simply feel less anxious by acting morally when faced with uncertainty, or believe in karma where moral actions lead to better outcomes. https://t.co/eG74J2d5Q0