The frequency of working from home has been rising rapidly in the US, with over 10% of the workforce now regularly work from home. But there is skepticism over the effectiveness of this, highlighted by phrases like "shirking from home". We report the results of the first randomized experiment on home-working in a 13,000 employee NASDAQ listed Chinese firm. Call center employees who volunteered to work from home were randomized by even/odd birth-date in a 9-month experiment of working at home or in the office.
We find a 12% increase in performance from home-working, of which 8.5% is from working more minutes of per shift (fewer breaks and sick-days) and 3.5% from higher performance per minute (quieter working environment). We find no negative spillovers onto workers left in the office. Home workers also reported substantially higher work satisfaction and psychological attitude scores, and their job attrition rates fell by 50%. Despite this ex post success, the impact of home-working was ex ante unclear to the firm, which is why it ran the experiment. Employees were also ex ante uncertain, with one half of employees changing their minds on home working after the experiment. This highlights how the impact of new management practices are unclear to both firms and employees.