Management Science · 2025 · David H. Hsu、Prasanna B. Tambe
中文摘要
疫情后管理性岗位设计的一个重要方面涉及远程工作。在多样性重新受到重视的时代，雇主可能想知道多样化的（性别与种族）和有经验的求职者如何回应远程职位招聘，尤其是高技能的技术和管理岗位。既有研究表明，虽然远程工作为员工提供了灵活性，但可能限制职业晋升前景，因此从求职者兴趣的角度看，将职位指定为可远程的净效应并不明确，尤其是在招聘女性和代表性不足的少数族裔（URM）时。我们分析了来自一家领先初创企业招聘平台的求职者数据，该数据涵盖2020年3月COVID-19疫情引发的停工前后较长时间窗口。为应对远程职位指定可能与不可观测的职位和雇主特征共同决定的实证挑战，我们采用匹配方法（以及利用突发停工的替代方法）来估计在其他方面相似的远程与现场职位发布中求职者特征的差异。我们发现，提供远程工作吸引了更有经验和更多样化（尤其是URM）的求职者，在多样性较低的地理区域效应更大。职位发布向远程状态的离散变化（保持其他一切不变）与女性求职者增加约15%、URM身份求职者增加33%、求职者经验增加17%相关联。利用申请数据，我们估计该劳动力市场中远程工作的平均估计补偿性工资差异约为发布薪资的7%。
Abstract
A significant element of managerial post-COVID job design regards remote work. In an era of renewed recognition of diversity, employers may wonder how diverse (gender and race) and experienced job applicants respond to remote job listings, especially for high-skilled technical and managerial positions. Prior work has shown that while remote work allows employee flexibility, it may limit career promotion prospects, so the net effect of designating a job as remote-eligible is not clear from an applicant interest standpoint, particularly when recruiting females and underrepresented minorities (URM). We analyze job applicant data from a leading startup job platform that spans long windows before and after the COVID-19 pandemic-induced shutdowns of March 2020. To address the empirical challenge that remote job designation may be codetermined with unobserved job and employer characteristics, we leverage a matching approach (and an alternative method which leverages the sudden shutdowns) to estimate how applicant characteristics differ for otherwise similar remote and onsite job postings. We find that offering remote work attracts more experienced and diverse (especially URM) job applicants, with larger effects in less diverse geographic areas. A discrete change in job posting to remote status (holding all else constant) is associated with an approximately 15% increase in applicants who are female, 33% increase in applicants with URM status, and 17% increase in applicant experience. Using the application data, we estimate an average estimated compensating wage differential for remote work that is about 7% of posted salaries in this labor market.