Return-to-Office Mandates: What the Research Says About Productivity and Retention

Return to office mandates have not been shown to make companies perform better, and the studies that follow who leaves afterwards point to a cost that headcount figures hide. A preprintpreprint: A research paper posted publicly before it has been peer reviewed or published in a journal; in economics and finance the same stage is usually called a working paper. Its findings may change, or fail to hold up, by the time it is published.Full entry in the glossary by Yuye Ding and Mark Ma of the University of Pittsburgh found no significant change in S&P 500 firms’ financial performance or market value after mandates, while employees’ job satisfaction fell.¹ A second working paper, by Ding and four colleagues, followed the LinkedIn careers of staff at 54 large US tech and finance firms and found that turnover rose after mandates, most among senior, skilled and female employees.²

The strongest evidence on flexibility and retention comes from the opposite experiment. At the travel firm Trip.com, a randomized controlled trialrandomized controlled trial: A study that assigns participants to the treatment or the comparison group at random, so the groups start out alike and a difference in what happens next can be put down to the treatment. Random assignment makes the two groups comparable; it does not make the people in the trial representative of anyone else.Full entry in the glossary reported in Nature in 2024 found that graduate staff offered two home days a week resigned about a third less often than colleagues kept in the office, and their performance grades did not differ.³

Here a mandate means an employer’s rule that staff who had been working remotely or flexibly must be on site on set days. The words remote and hybrid themselves are counted differently from survey to survey, as the guide to what remote, hybrid and distributed work mean explains. This article looks at what happens when an employer takes flexibility away, and at the conditions under which office days earn their cost.

Do return to office mandates improve company performance?

We found no study showing that return to office mandates improve company results. In their working paper on S&P 500 firms, Ding and Ma found no significant change in financial performance or firm value after mandates, and a significant decline in employees’ job satisfaction; we could read only the paper’s abstract.¹

Firm-level results are a blunt test. Profits and share prices move with markets, prices and products, so a policy that changes where people sit would need a large effect to show up at all. A null result here means no detectable change, not proof of zero effect.

The authors also studied which firms chose to mandate. They read the pattern as consistent with managers reasserting control and blaming employees for weak results, and as not supporting the idea that managers mandate because they expect firm value to rise.¹ That is an interpretation of who mandated, not a measured motive.

Managers’ hunches about remote work can shift once they see results. Before the Trip.com trial, the managers taking part expected hybrid work to lower productivity slightly; after six months of results, they expected a slight gain.³

Picture, as an illustration, a manager who feels a team drifted while remote. The feeling may be right, or it may reflect less visibility rather than less output. So before judging a mandate, name the outcomes it is meant to change, such as errors, delivery dates or how fast new staff reach full speed, and measure them before and after, rather than waiting for the share price to answer.

Where office days pay off: learning from teammates nearby

Office time helps most where people learn from teammates who sit close by. A study of software engineers at a Fortune 500 online retailer by Natalia Emanuel, Emma Harrington and Amanda Pallais found that sitting near teammates increased the feedback engineers received on their code and improved code quality, with the largest gains for newer and younger engineers.⁴

The researchers used two moments when distance changed, the office closures of 2020 and the firm’s own mandates of 2022 and 2023, and compared teams housed in one building with teams spread across sites. After a three-day office mandate, engineers on co-located teams introduced fewer bugs than those on distributed teams.⁴

The authors trace the gain to a simple mechanism: sitting nearby makes it easier to ask a follow-up question and to ask more people for help. The benefit is fragile. A ten-minute walk between two buildings cut feedback about as much as working in different states, and one distant new hire weakened feedback among the teammates who still sat together, apparently because meetings moved online. It also has a price: experienced engineers, who gave most of the feedback, wrote less of their own code when they sat near teammates. This is one firm and one occupation, in a paper since published in the Quarterly Journal of Economics; the firm shared its data.⁴

Consider a new analyst whose team all comes in on Tuesdays and Thursdays: questions get answered over a shoulder, and the analyst sees how seniors handle problems. The same analyst on a team whose members pick different days gets a quiet floor and the same video calls as at home. So agree who comes in, and on which days, before arguing about how many.

Who quits after an office mandate?

Departures after a mandate lean toward senior staff, according to two studies: one counted who left in LinkedIn profiles, the other inferred it from resumes showing shorter tenure and fewer senior titles.²⁵ Both compare firms that mandated with firms or trends that did not, which cannot rule out every other explanation.

The study

Limited evidence

Who left after 54 office mandates: a LinkedIn working paper, 2024

After their first mandate announcements, the firms’ average turnover rate, adjusted for national turnover, rose by 14 percent: for every 100 departures otherwise expected, about 114. The rise was concentrated among women, mid-level and senior managers, and employees listing the most skills. Time to fill vacancies rose by about 23 percent and hire rates fell by about 17 percent.²

The authors state that their design cannot establish cause: firms that mandate may differ in other ways, and the sample covers only the largest firms in two industries. They found no rise in turnover in the five quarters before the announcements and no effect at placebo dates a year earlier, which makes a pre-existing trend a less likely explanation.² The paper names no funder.

The resume study, by David Van Dijcke, Florian Gunsilius and Austin Wright and published in 2026 in the Review of Economics and Statistics, compared Microsoft, SpaceX and Apple with similar firms that did not mandate. It estimated that after each mandate the workforce shifted toward shorter tenure and more junior titles, which the authors read as senior staff leaving, and the shift was largest at SpaceX, whose rule was the strictest. Funding came from a Peterson Foundation pandemic research fund and the University of Chicago’s Becker Friedman Institute.⁵ Microsoft’s management said its internal data did not support the findings, Ding and colleagues note.²

1. Learning from teammates: more feedback and better code quality for newer engineers when their teammates sat nearby
2. Time spent teaching: experienced engineers wrote less of their own code when they sat near the people they helped
3. Departures: departures after mandates leaned toward senior staff in two studies, and toward skilled staff and women in one

A firm can refill its headcount and still lose experience. If a manager with eight years at the company leaves and a new hire takes the seat, the headcount is unchanged, while the knowledge of past decisions and the mentoring behind the office policy walk out together. A total turnover figure hides that; counting departures by tenure, level and gender shows it.

Myth

The people who quit over an office mandate are the least committed, so the firm is better off without them.

Fact

In two studies, departures after mandates leaned toward senior staff, and one found the most skilled employees leaving more often too.

Why home days hold on to staff

Home days appear to keep people because employees value what the days give them. In the Trip.com trial, Nicholas Bloom and his co-authors found that employees credited home days with saving commuting time and money and with leaving room for occasional personal tasks.³

Trip.com’s executives estimated that each resignation cost about US$20,000 in recruiting and training, and in 2022 they extended hybrid work to every employee. That figure is the firm’s own estimate for graduate staff in China. Co-author James Liang chairs Trip.com and holds shares in it, though the paper states that the company did not fund the research.³

Taking flexibility away is not the mirror image of offering it, and no trial has tested the withdrawal directly. A 2026 qualitative study of US federal employees after the January 2025 office order reported disrupted routines, a lower sense of control over work and new emotional and logistical strain; we read only its abstract.⁶

Bloom and colleagues call hybrid work a valuable perk for employees.³ For a manager, that means treating home days as part of what people are paid in, and expecting their removal to be felt.

Four conditions that make office time pay off

The studies point to conditions rather than a verdict: office days are most likely to earn their cost when they bring the right people together without pushing experienced people out. Each suggestion below follows from one of the studies above; nobody has tried the four together and measured the result.

1. Bring teammates in on the same days, so office time produces the feedback and learning the engineers’ study found.
2. Aim office days at newer staff and shared problem-solving, and protect time for experienced people’s own work, since mentoring costs them output.
3. Keep an option where the work allows it: the Trip.com trial tested a choice of home days, not a rule.³
4. Watch departures by tenure, level and gender, because averages can hide the loss of senior staff.

As an illustration, take a team of eight with two recent hires. Instead of “three days, any three”, the team agrees on Tuesday and Wednesday as shared days. The new hires sit beside the people who review their work, Thursday stays free of meetings so the seniors can get their own work done, and the manager checks each quarter who has left and at what level.

For an employee facing an order, the choices run from complying to asking for an exception, which is a separate question from what the research shows.

Mandates on paper, attendance in practice

A mandate is a rule, and attendance is what people do. Keycard data from Kastle Systems, a building-security company that sells access-control systems, showed weekly office entries across ten large US metro areas at about half of their early-2020 level in early September 2026, with a midweek peak.⁷

The figure needs its label. It is company data from buildings that use Kastle’s systems, counts one entry per person per day, and compares against an average of three weeks in February 2020; Kastle itself says it is not a national statistical sample.⁷ Half, in plain terms, means that for every two people who badged into these buildings on a typical day before the pandemic, about one does now.

The engineers’ study adds a telling detail from the firm’s own badge records: during its mandate periods, younger and less-tenured engineers came in more than others, especially when their teammates were based in the same office.⁴ One reading, consistent with the proximity findings, is that people come in when the office offers something home cannot.

So judge an office policy by who is there together and what they do there, not by how many badges were swiped.

What nobody has tested yet

The evidence on mandates is young and mostly indirect. No randomized trial has assigned a mandate, the two turnover and performance studies by Ding and colleagues are working papers, and a 2025 review by Wayne Cascio in Human Resource Management covers outcomes from productivity to burnout and proposes 10 best practices; we read only its abstract.⁸

Three gaps matter most. Firms that mandate may differ from those that do not in ways no comparison fully removes. Almost all the data come from large tech and finance firms, and the authors of the turnover study say results may differ in other industries and smaller firms. And among the studies of mandates, only the engineers’ study measured the work itself; the others measured departures, satisfaction or market value.

When a headline claims that mandates work or fail, three questions sort it: did the study compare against firms that did not mandate, who was in the data, and did it measure the work or only people’s reactions to it?

The bottom line

The research so far gives no sign that return to office mandates lift company performance, and two studies point to a cost that falls unevenly: departures lean toward senior staff. Office days can pay off when they put teammates, and especially newer staff, in the same room at the same time. If an office policy cannot say what those shared days are for, the evidence offers little reason to expect it to pay for itself.