Return to Office and the Tenure Distribution
Authors: David Van Dijcke, Florian Gunsilius, Austin L. Wright
Abstract: With the official end of the COVID-19 pandemic, debates about the return to office have taken center stage among companies and employees. Despite their ubiquity, the economic implications of return to office policies are not fully understood. Using 260 million resumes matched to company data, we analyze the causal effects of such policies on employees’ tenure and seniority levels at three of the largest US tech companies: Microsoft, SpaceX, and Apple. Our estimation procedure is nonparametric and captures the full heterogeneity of tenure and seniority of employees in a distributional synthetic controls framework. We estimate a reduction in counterfactual tenure that increases for employees with longer tenure. Similarly, we document a leftward shift in the seniority distribution towards positions below the senior level. These shifts appear to be driven by employees leaving to larger firms that are direct competitors. Our results suggest that return to office policies can lead to an outflow of senior employees, posing a potential threat to the productivity, innovation, and competitiveness of the wider firm.
Reading Notes
Objective
To understand the effects of return to office mandates on the organizational structure/tenure distribution of several large tech firms
Importance
Builds on relatively new literature on the effect of RTO mandates using large-scale resume data instead of surveys.
Explores shift from remote work to (partial) office attendance, rather than static remote/hybrid/in-office regime
Background
Microsoft, Apple, and Space X account for over 2% of employment in the tech sector. Relatively early RTO mandates and no major layoff events within 1 quarter of RTO
Data & Key Variables
People Data Labs via the Dewey Data platform - global resume data for ~1 billion individuals. Start and end date in company and role/title
Outcome variables: tenure distribution, job titles to assign seniority (10 ordinal values from unpaid to C-Suite)
RTO dates for 55 large tech firms based on news reports and leaks on anonymous employee forum (Blind).
Flex Index by Scoop Technologies for ~8,000 companies
Layoffs from layoffs.fyi website
Methodology
Distributional synthetic controls estimator
Treated firms: Microsoft, Space X, Apple.
Focus on Microsoft - largest, RTO date best defined
Control firms - no RTO mandate within the study period, no layoffs around the date of treated firm RTO, matched on characteristics
Results
Return to office led to significant employee outflows, with more senior staff leaving at higher rates.
Tenure at Microsoft decreased by 2 months at top deciles
Departing employees were more likely to go to large direct competitors than the typical worker who left Microsoft before RTO
No increase in flows to unemployment or jobs that would be a demotion, indicating workers had good outside options.
No gender differences in employee outflows
Key Table/Figure
