Yes, return-to-office mandates cause engineers to quit, and disproportionately the senior ones. Research tracking more than 3 million workers found RTO mandates raised abnormal turnover by roughly 13 to 14 percent, hit senior, high-skill, and female staff hardest, and made open roles take 23 percent longer to fill. Gartner found that intent to stay drops most among high performers. For a startup, copying a Fortune-500 mandate imports that brain-drain math onto a team that cannot absorb it. The better move is to decide your work-location policy deliberately per role, then disclose it plainly on the job posting so the right engineers self-select in before your pipeline fills with mismatches. ## Do return-to-office mandates cause engineers to quit? Yes, and the effect is measured, not anecdotal. The most rigorous evidence comes from a University of Pittsburgh working paper (Ding, Jin, Ma, Xing and Yang) that tracked RTO announcements at 54 S&P 500 tech and finance firms between 2020 and 2023 against the employment histories of more than 3 million workers, using a difference-in-differences design on Revelio Labs and LinkedIn data. The headline: abnormal turnover rose roughly **13 to 14 percent** after mandates. And the people leaving were not random. Departures concentrated among senior, highly skilled, and female employees. Women's turnover increase was nearly three times higher than men's, and mid-to-top-level managers left at higher rates than junior staff. The authors name it directly: "brain drain ... a significant cost of RTO mandates." Gartner's research points the same direction from a different angle. In an October 2024 survey of more than 3,000 managers of knowledge workers, Gartner found that strict RTO cuts employees' intent to stay by up to **10 percent** overall, but by **16 percent for high performers**, double the average. It also increased "quiet quitting," doing the bare minimum, by up to **19 percent**. Read those two studies together and the pattern is stark. A mandate does not shed your weakest people. It sheds the ones with the most options, and the most institutional knowledge. ## Why does the pain land on senior engineers specifically? Leverage. Junior engineers tend to comply with a mandate because they need the resume line and the in-person visibility early in a career. Senior engineers have options, and they use them. Nik Froehlich, founder and CEO of Saritasa, framed this well in a July 2026 Forbes Tech Council piece arguing for "gravity over mandates": you can require attendance, but you cannot require participation, and in tech that distinction is the whole game. A senior engineer who resents being dragged in does not resign on day one. They coast, then they leave, and they take the expensive things with them. That is why the same headcount loss costs far more than it looks. When a senior engineer walks, you lose architectural judgment, code-quality standards, mentorship for the juniors who stayed, and the undocumented context about why the system is built the way it is. None of that shows up in a headcount report, and none of it can be rehired quickly. The Pittsburgh data even captured a downstream signal here: departing employees taking lateral or lower-level roles rose from 41.6 to 46.4 percent after mandates, a sign that people were leaving reactively rather than for a clear step up. For a 200-person company, losing a handful of staff engineers to a policy is painful. For a 20-person startup, it can be existential. ## The hiring tax you do not see: slower to fill, smaller funnel A mandate does not only cost you quits. It quietly taxes every new hire, on both speed and volume. The same Pittsburgh study found that after a mandate, open vacancies took **23 percent longer to fill**, from roughly 51 days to 63 days on average, and hire rates fell **17 percent**, even after adjusting for market-wide hiring trends. So the mandate empties senior seats faster and then makes them slower to refill. The volume problem is structural. LinkedIn's own data shows that remote and hybrid roles make up only about **20 percent of postings but attract roughly 60 percent of all applications**. An onsite-only requisition is fishing in a structurally smaller pond, by design. You are competing for the same senior engineers as everyone else, while advertising the one attribute that the majority of applications are flowing away from. Put the two together and a rigid onsite policy is a double tax: fewer applicants in the top of the funnel, longer to close each hire, and more seniors leaving out the back.

Watching the 2026 RTO headlines and deciding your own policy? Kit lets you state the work arrangement and salary range directly on the posting, then tracks your apply-to-offer conversion so you can see whether the policy is helping or quietly filtering out the engineers you need.

Start your free trial

## What just happened in California, and why founders should care The 2026 headlines are swinging toward the office, and California is the loudest example. On **July 1, 2026**, a mandate requiring state workers to be in the office four days a week, up from two, took effect. It stems from an order Governor Newsom issued in March 2025. Roughly 100,000 state workers are affected, though unions describe the reach as "hundreds of thousands." The pushback was immediate. Unions including SEIU Local 1000 protested, filed a complaint with the Public Employment Relations Board in June 2026, and are backing state legislation (AB 1729, from Assemblymember Alex Lee) to unwind the requirement. Private employers are tightening too: company trackers note Airbus moving to four days, PNC to five, and Instagram to five over 2026. Here is the trap for founders. "Everyone else is doing it" is not evidence that it works, and for a lean team the math is worse, not better, because you cannot absorb senior loss the way a state agency or a 50,000-person enterprise can. A top-down mandate at government scale is precisely the wrong template to copy onto a 25-person engineering org. The headline is a cautionary tale, not a playbook. ## "So go fully remote?" No, there is a supported middle The honest answer to the RTO debate is not "offices bad, remote good." It is that there is an empirically supported middle, and following a big-company mandate skips right past it. The strongest evidence is a gold-standard randomized controlled trial. Stanford economist Nicholas Bloom ran an experiment at Trip.com with more than 1,600 university-trained professionals, published in *Nature* in 2024. A hybrid schedule of two work-from-home days per week **cut resignations by 33 percent** with **no measurable hit to productivity or promotion rates**. The people least likely to quit were women, non-managers, and those with long commutes, exactly the groups a blanket mandate pushes out fastest. That is the disarming counter to the "so you want everyone remote?" objection. No. A deliberate hybrid arrangement measurably retained people without an output penalty, in a randomized trial. The lesson for founders is that the office-versus-remote question has a defensible answer in the middle, and reaching for a five-day mandate throws that away to import attrition your balance sheet cannot survive. ## Gravity over mandates: decide your policy, do not copy one The winning move in 2026 is not "go remote" or "follow the mandate." It is deciding your work-location policy deliberately, per role, and being able to defend it. Here is a four-part decision: 1. **Run a "why come in" audit.** Name the work that is genuinely better in a room: design reviews, incident response, onboarding a new hire's first two weeks. Then be honest about the work you would be dragging people in to do that they could do anywhere. Froehlich's "gravity" framing is the right instinct: design days people want to attend, do not mandate a seat count. 2. **Check what the talent pool for this role expects.** Engineering, and especially security and infrastructure roles, skew heavily remote and hybrid in 2026. Slapping an onsite-only tag on those requisitions is a self-inflicted funnel cut before a single candidate sees it. 3. **Decide whether you can afford the senior-attrition tax.** If a 13 to 14 percent turnover bump concentrated in your most senior people would break the team, do not copy a mandate. The startup that cannot absorb the loss has the least business imitating the companies that can. 4. **Disclose the decision plainly.** Whatever you land on, remote, hybrid with named in-office days, or onsite, the output is a defensible, stated policy per role, not a vibe. The point is not that flexibility always wins. It is that the decision should be deliberate and role-specific, not inherited from a news cycle. ## The move most founders miss: disclose the policy at the posting stage Work-location policy stops being an internal HR question the moment you publish the role. At that point it becomes the highest-leverage filter on your entire funnel, and most teams waste it. The failure mode is the hedge. "Flexible work arrangements available" reads, to a sophisticated senior candidate, as a dodge. It does not filter anyone. Instead it defers the real conversation, about commute, relocation, and exactly how many in-office days, all the way to the offer stage, where a mismatch blows up a deal both sides spent three weeks building. The fix is to state the specific arrangement in the posting itself: remote, or hybrid with the actual in-office days named, or onsite, plus the location and a one-line reason. When the arrangement is explicit, candidates self-select on it at the application step, not at the offer step. The right people opt in with clear eyes. The wrong people opt out before either of you invests. Clarity beats hedging every time, because it moves the mismatch to the cheapest possible point in the process. This is the same logic that drives [pay transparency on job postings](/blog/pay-transparency-2026-honest-salary-ranges): disclosing the deal upfront does not scare off good candidates, it filters out the ones who were never going to accept anyway, and it does it in week zero instead of week three. Location and pay are the two surprises that most often kill offers late. State both. ## How to run this in Kit Kit is built so the work-location decision gets stated where it matters, on the posting, and measured where it counts, in your funnel. Educate yourself on the policy first; then the product makes disclosure the path of least resistance. - **State the arrangement upfront.** A Kit job posting carries a remote flag plus a location field, and Kit renders a remote badge on the public posting so the arrangement is visible before anyone applies. The discipline the article argues for, put the policy in the posting, do not hedge, maps directly onto filling these in deliberately. Kit models this as remote plus location rather than a three-way enum, so the clarity lives in how plainly you write the location and description, and Kit surfaces it prominently. - **Disclose the whole deal, not just location.** Employment type and the salary range fields sit alongside location on the same posting. The transparency logic that kills arrangement-mismatch surprises at offer stage applies to pay too, and Kit lets you disclose both in one place. - **Let self-selected candidates engage without friction.** Kit's candidate portal is [magic-link based, no passwords](/blog/why-we-killed-passwords-for-candidates), so a candidate who saw the policy and opted in moves through stages without a login tax that reintroduces drop-off. - **See whether your policy is helping or filtering out talent.** Kit computes an apply-to-offer conversion rate across your postings on the dashboard. That is the honest feedback loop: if an onsite-only role's conversion is thin, or offers keep dying late, the policy is the prime suspect, not your sourcing. You measure the funnel effect of your RTO decision instead of guessing at it. - **Standardize the disclosure across every role.** Process templates let your team encode "always state the arrangement, always state pay" once, so no hiring manager ships a vague "flexible" requisition that dies at the finish line. The through-line: an RTO decision is a hiring lever, and levers should be pulled deliberately and then watched. Kit is where you state the policy and where its funnel effect shows up in numbers. ## The bottom line RTO mandates are not a neutral culture preference in 2026. The data shows they push out your most senior engineers first, slow every subsequent hire by weeks, and shrink the applicant pool you draw from, while a well-designed hybrid arrangement retained people in a randomized trial with no productivity cost. Copying a Fortune-500 or state-government mandate imports the attrition without the balance sheet to survive it. So do not inherit a policy from the headlines. Run the "why come in" audit for each role, decide the arrangement deliberately, disclose it plainly on the posting alongside your salary range so candidates self-select before your pipeline fills with mismatches, and then watch your apply-to-offer conversion to confirm the policy is helping rather than quietly filtering out the people you most need to keep. If you want a place to state and measure that policy, [start a free trial of Kit](/users/sign_up), and read the companion piece on [hiring a distributed team across time zones](/blog/hiring-distributed-remote-team-across-time-zones) once your policy is set.