Does Working from Home Work? Evidence from a Chinese Experiment

Innovation and Economic Growth

Article Snapshot


Nicholas Bloom, James Liang, D. John Roberts and Zhichun Jenny Ying


Working Paper, 2012


This paper uses experimental data to examine the effects of permitting employees to work at home.

Policy Relevance

Letting workers do some of their work at home seems to increase their efficiency and job satisfaction, however, uncertainty about the benefits of work-from-home programs may be stymying their adoption by firms and employees.

Main Points

  • New technologies make it possible for employees to do some work from home, but good data about the effects of work-from-home (WFH) programs are scarce. This paper uses a well-designed data set from a Chinese call center.
  • Existing studies on WFH are not statistically ideal because in those studies workers are not randomly assigned to WFH programs. This limits the reliability of the findings.
  • In the call center program examined in this paper, workers randomly assigned to work from home reported increased job satisfaction after nine months, and most opted to continue working from home after the experiment’s termination.
  • The call center also benefited greatly from the WFH program. Attrition rates of home workers fell by more than half, reducing training costs, and home workers became more efficient: a measurement of their general performance increased by 12%. 
  • These results may be specific to call center employees; for example, some of the performance increase was attributed to a quieter environment at home than could be obtained at a crowded call center.
  • The slow adoption of WFH programs can partly be attributed to uncertainty on the part of both workers and employers about the programs’ effects of work and job satisfaction.

Get The Article

Find the full article online

Search for Full Article