« Back to Results

Remote Work and Productivity

Paper Session

Friday, Jan. 6, 2023 10:15 AM - 12:15 PM (CST)

Hilton Riverside, Kabacoff
Hosted By: American Economic Association
  • Chair: Tanjim Hossain, University of Toronto

How Working From Home Works Out

Nicholas Bloom
,
Stanford University
Ruobing Han
,
Stanford University
James Liang
,
Fudan University

Abstract

Hybrid Working From Home has become the dominant post-pandemic choice for graduate employees across the US and Northern Europe. This usually involves employees working from home two days a week and coming into work three days a week. But little is known about this given the pandemic has prevented the full return to the office of most firms. We worked with a major Chinese multinational, to run a randomized control trial on hybrid WFH on 1600 engineers in two divisions. We find no impact on productivity on promotions, assuaging some of the fears over negative effects of remote work. But on the positive side was a 31% reduction in quit rates and 15% drop in sick leave, alongside increased self-reported work satisfaction scores. Employees also flex their working hours, reducing time on hybrid WFH days but increasing it in the evenings of other days and the weekends. The firm judged the experiment was so successful that it rolled out the hybrid WFH scheme to the entire company in March 2022.

"Working" Remotely? Selection, Treatment, and the Market Provision of Remote Work

Natalia Emanuel
,
Federal Reserve Bank of New York
Emma Harrington
,
University of Iowa

Abstract

How does remote work affect productivity and how productive are workers who select into remote jobs? We decompose the treatment and selection effects of remote work using data from the call-centers of a US Fortune 500 retailer. The retailer employed both remote and on-site workers prior to Covid-19 and went entirely remote during the lockdown. In a difference-in-difference design around the Covid-19 lockdown, formerly on-site workers became 6-10% more productive after going remote relative to already-remote workers. However, during the lockdown, workers who originally chose remote jobs answered 18-21% fewer calls than those who originally chose on-site ones, indicating adverse selection into remote work. Our results suggest that adverse selection made remote work the exception rather than the rule in call-center jobs prior to the pandemic. We discuss implications of the Covid-19 lockdown for adverse selection and the consequent future of remote work.

Is Hybrid Work the Best of Both Worlds? Evidence from a Field Experiment

Prithwiraj Choudhury
,
Harvard Business School
Tarun Khanna
,
Harvard Business School
Christos Makridis
,
Stanford University
Kyle Schirmann
,
Harvard Business School

Abstract

This paper reports causal evidence on how the extent of hybrid work---the number of days worked from home relative to days worked from the office---affects outcomes relevant for workers and firms. Collaborating with an organization in Bangladesh, we randomized the number of days that individual employees worked from the office for nine weeks. We find that an intermediate number of days in the office resulted in greater self-reported work-life balance and lower isolation from colleagues. Furthermore, hybrid work also led to a greater volume of emails, more unique email recipients, and more unique information conveyed in the emails. Hybrid work was also linked with better performance ratings from managers.

Should We Coordinate Remote Work Teams?

Tanjim Hossain
,
University of Toronto
Elizabeth Lyons
,
University of California-San Diego

Abstract

Effective communication and technological coordination are critical for successful teamwork, but how this is best achieved among remote teams is unclear. Using a natural field experiment run on an online workplace, we investigate whether teams of translators given flexible work schedules perform differently than those given coordinated work schedules. Remote work supports more work hour flexibility than on-site jobs do, which may improve job satisfaction, autonomy and, thus, efficiency. On the other hand, flexibility in work schedule may hinder effective coordination and communication among team members. Less frequent and lower quality communication may lead to even larger performance losses over time as these teams will be less likely to overcome initial communication barriers. Thus, whether team managers would do better to coordinate work schedules among team members or allow them to work flex hours is an empirical question. In our field experiment, we exogenously manipulate whether two teammates have the same work schedule, or whether they can work during the hours that best suit them as individuals and compare the extent of coordination, efficiency, and output quality of teams across these two work arrangements. Moreover, a new task assigned to the same teams two months after the first task allows us to investigate how experience in working in the team and the communication method coordinate to affect productivity.
JEL Classifications
  • J2 - Demand and Supply of Labor
  • D2 - Production and Organizations