Study: Remote Work Hurts Junior Employees More Than AI
A new study published in the Quarterly Journal of Economics found that remote work harms junior employees more than artificial intelligence. The researchers tracked software engineers at an internet retail company and determined that being away from the office hurts work quality, the learning process, and the hiring of inexperienced workers, leading to increased unemployment among them. Collaborative work increases feedback by about 18%, and the main damage is a 63% drop in follow-up questions.
A new study published in the economic journal Quarterly Journal of Economics (QJE) found that remote work harms junior employees more than artificial intelligence. The researchers Natalia Emanuel (Federal Reserve Bank of New York), Emma Harrington (University of Virginia), and Amanda Pallais (Harvard University) tracked software engineers at an internet retail company included in Fortune magazine's list of the 500 largest companies between 2019 and 2024. They found that sitting physically alongside team members increases feedback on code by about 18%, while remote work harms the feedback process and work quality, with the main victims being young and new engineers. They also found that employers prefer to hire experienced and older workers for remote positions, leading to increased unemployment among inexperienced youth. The researchers calculated that remote work accounts for about 64% of the rise in youth unemployment and argued that the effect appeared even before the AI breakthrough. The study revealed that the shift to remote work dramatically reduced professional feedback: when offices were open, engineers in shared teams received 23.9% more comments. The number of follow-up questions from code writers dropped by 63%, and feedback given to women decreased by 39% compared to 13% for men. However, the researchers note that remote work makes it easier for mothers to stay in the workforce. The study is limited to one company and one profession, so it is unclear how broadly its conclusions apply.