Every return-to-office mandate is written as though the workforce were a single body that can be told to stand up and walk to a building. It is not. It is a few hundred, or a few thousand, separate people, each running a private calculation the memo never accounts for: what the commute costs, what the flexibility was worth, and whether anyone else would have them. The mandate assumes everyone does that sum and arrives at the same answer. They do not. The people with the shortest path to another job run it fastest.
The senior and skilled leave first
The pattern is measurable. In a study posted last year, Yuye Ding and Mark Ma of the University of Pittsburgh, Betty Xing of Baylor, Yucheng Yang of the Chinese University of Hong Kong, and Zhao Jin of Cheung Kong Graduate School of Business did the thing the arguing usually skips: they counted. Using data from Revelio Labs drawn from more than three million LinkedIn profiles, they tracked 54 large technology and finance firms in the S&P 500 that ordered workers back between 2020 and 2023, and measured turnover after each announcement against firms that issued no such order. The firms that imposed a mandate saw abnormal turnover rise about 14 percent. The word carrying the weight is "abnormal": it is the jump left over once you strip out the industry trend and the wider labor market, the churn the mandate itself caused and nothing else.
A rise in quitting of that size is something a company can absorb, if the people are interchangeable. They are not, and this is the part the memo never models. The departures did not fall evenly. Women left at close to three times the rate of men. Mid- and top-level managers left more readily than junior staff, and the higher-skilled more readily than the less skilled. The authors read the pattern as consistent with firms losing their best talent and their female employees, and facing greater difficulty attracting anyone to replace them. The mandate does not thin the workforce the way you trim a hedge. It prunes from the top.
Ask the question I ask of everything here: who gets to leave? The people with somewhere to go. Seniority is a network, and a skilled worker a decade in is exactly the person a competitor already has a seat for, and exactly the person who can afford to hold out for one that lets her keep working from home. The junior staff, still building the resume that would let them do the same, run the math and stay, because the risk of leaving is theirs alone to carry. So the mandate works as a filter tuned precisely backward. It keeps the replaceable and releases the ones who hold the institutional memory. None of that sits on a balance sheet, which is exactly why it is so easy to order it out the door.
The mandate does not pay for itself
Once it is gone it is slow and expensive to buy back. The same study found that after a mandate those firms took 23 percent longer to fill the vacancies that opened after the mandate, 63 days instead of 51, and their hire rate fell 17 percent even after adjusting for the national hiring slowdown. A mandate is sold as a way to tighten a company. What it does, on these numbers, is widen the gap between the experience walking out and the experience the firm can persuade to walk in. The people who stayed drew the obvious lesson: after a mandate, a growing share of those who did leave took a lower-ranked job elsewhere to keep the flexibility. People will accept a demotion in order not to come back.
You would forgive all of this if the office bought something in return. It does not appear to. The same core researchers, in an earlier study of S&P 500 firms, found no significant improvement in profitability or market value after a mandate, and found that firms tended to announce one only after their share price had already fallen. Their reading is that the mandates track managers reasserting control rather than any measured gain, a reading a companion piece in these pages has already worked through. What I want is the part that story leaves on the floor. If the mandate buys no productivity, then the brain drain it causes is not a trade. It is not experience swapped for focus, or flexibility swapped for output. It is a firm's most senior, most connected people carrying years of hard-won knowledge across the street in exchange for nothing the balance sheet can find. The loss is not the price of a benefit. The loss is the whole transaction.
I should be careful about what these numbers are. They come from LinkedIn profiles, which see the white-collar world that keeps LinkedIn current, tech and finance, and not the warehouse or the ward, where nobody was offered the choice and a mandate is just called a schedule. Profile turnover is a proxy, the causal claim rests on the comparison firms holding steady, and the years it covers were a strange stretch in which to measure anything. Take the precise figures with the caution any working paper deserves. But the direction is not subtle, and it holds across two separate studies: the mandate does not pay for itself, and the people it drives out are the ones with the most to teach.
The authors close on a mild suggestion, that firms let the high performers who do well from home keep doing so, since they are the ones who can most easily leave. It is sensible, and it is not quite the point. What the mandate sends out the door is not a headcount a recruiter can quietly backfill. It is the knowing: why a fix was left in place, which promise to which client still holds, where the workarounds are that keep the month-end from breaking. That knowledge sat in the people the mandate selected against, the senior, the skilled, the ones with somewhere to go, and on these numbers they are the ones who left, at a 23 percent longer wait to replace and a price no one wrote down. That is the cost of the mandate. It was never the commute.
