with: Esteban Rossi-Hansberg

Time & Location

9 Mar, 12:00 pm – 1:00 pm
e61 Institute, Level 3/17-21 Bellevue Street Surry Hills NSW 2010, Australia

There are limited spaces available.

About the event

The productivity of cities and the welfare of their residents depend on the value of workers’ interactions. Although remote work offers clear advantages by removing the need to commute, it also limits gains from direct and serendipitous interaction in the workplace. The global pandemic crisis demonstrated that a dramatic increase in remote work is feasible. Consequently, central business districts lost foot traffic, and housing demand away from city centers grew across many large cities. Even as public health concerns recede, steady progress in the technology to work from home continues to extend the scope of tasks that workers can perform at home. This paper investigates the impact of more effective remote work technology on city structure, productivity, and welfare. We formulate a theory of the equilibrium composition of labor delivery mode in a dynamic city model. Workers choose every period their labor delivery mode (commute or remote) and their residence location based on i) the current wage, ii) expected continuation values, and iii)idiosyncratic preferences for each mode. Crucially, the current wage for in-office work depends on the aggregate number of current commuters, an assumption that captures the value of in-person interactions. We prove the existence of stationary equilibria and provide sufficient conditions for unique and multiple equilibria. Multiple stationary equilibria can exist when agglomeration externalities are strong relative to congestion externalities. In this case, one stationary equilibrium features workers mostly in-office, high commuting costs, and high aggregate productivity; another stationary equilibrium sees workers mostly at home, low commuting costs, and low aggregate productivity. Which of these equilibria is most beneficial is a priori ambiguous, because of differences in the levels and trade offs between agglomeration externalities and commuting costs. We study how changes in some parameters may affect a city’s structure. A “mostly remote” equilibrium could be reached via a steady improvement in the remote work technology, but also after large temporary shocks to the number of remote workers, like a lockdown forcing agents to work remotely for a period of time. We estimate the key parameters of the model using a combination of longitudinal and cross-sectional data, and calibrate it to several cities in the United States. Our model allows us to study whether a short-term “lockdown shock” puts any city on path to the “mostly remote” stationary equilibrium, or whether cities will revert to the “mostly in-person” stationary equilibrium after the shock fades out. We also study the long-run welfare consequences of improvements in the remote work technology across cities in the United States.