A design team, a sales team, and a payroll team do not have the same reason to be in a room, or the same need to be in one at all. A company-wide mandate treats them as if they did. It sets a single rule, three days, five days, everyone, from the one vantage point in the organization furthest from the daily texture of any particular team's work. That is the structural flaw of the mandate: it makes the coordination decision at the level with the least information about coordination.
Where the information lives
The case for pushing the decision down is not sentimental; it is about who knows what. A team knows which of its tasks genuinely benefit from shared time, when its members need to overlap, and what its members' constraints are. The executive setting a company-wide policy knows none of this at the resolution required to get it right, and so must default to a blunt uniform rule that fits no team exactly. Local knowledge is the scarce resource, and it does not sit at headquarters.
The economics of this is more than intuition. A large study of firms across the OECD found that those which had delegated more decision-making authority from headquarters to local managers before the financial crisis outperformed centralized firms in the sectors hardest hit, and the mechanism was precisely the rising value of local information under turbulence. When conditions are stable and predictable, a central rule can be good enough. When they are uncertain and fast-changing, the advantage shifts decisively to the people close to the ground. Work arrangements after a pandemic are exactly the kind of unsettled terrain where that advantage should dominate.
What a team agreement looks like
A team-level agreement is not the absence of structure; it is structure set by the people it governs. The randomized hybrid research is clear that coordination is one of the conditions for hybrid to work: teams need to be in on the same days for presence to pay off. A team agreement is simply the vehicle for producing that coordination without imposing it from above, the team decides its own anchor days, its own norms for when to come together, its own defaults for the work that can happen anywhere. It gets the coordination benefit of a mandate without the mandate's uniform cost.
The objection is that teams left to themselves will free-ride or fragment. Sometimes they will, which is why the agreement has to be explicit and the team accountable for its results rather than its attendance. But the assumption that only a central rule can produce order is often just untested pessimism. When Semco let factory workers set their own hours in 1985, management expected the line to fall apart; the workers coordinated among themselves instead, governed by nothing more than the instruction to talk to the person next to you. The order was available at the team level all along. Management simply had not asked for it there.
Accountability, not just autonomy
The deeper reason team agreements can be trusted is that teams can hold their own members accountable, sometimes more effectively than a distant manager can. Semco built this in structurally: teams, not the boss's boss, could veto a manager. The first year was uncomfortable, but managers who used to shout began to ask, and the ones who could not adapt returned to technical work without a pay cut. The point is that a group with real authority over how it works also develops real standards for how its members contribute. Autonomy and accountability arrive together, or the autonomy curdles.
That is the balance a company-wide mandate forfeits. It takes both the decision and the accountability up to the executive level, leaving teams with neither authority nor ownership, only compliance. A team that owns its working arrangement owns the outcome too, and has every incentive to make the arrangement produce. The executive's job in this model is not to set the rule but to set the goal and hold teams to their results, then get out of the way of the coordination that only the team can actually design.
Sources
- Aghion, Bloom, Lucking, Sadun & Van Reenen (2021). Turbulence, Firm Decentralization, and Growth in Bad Times (AEJ: Applied). Firms that delegated authority to local managers outperformed centralized firms in the hardest-hit sectors; mechanism is the rising value of local information under turbulence. https://www.aeaweb.org/articles?id=10.1257%2Fapp.20180752
- Bloom, Han & Liang / HBR (2024). One Company A/B Tested Hybrid Work. Coordinated (team-level) schedules are among the three named conditions for hybrid to succeed. https://hbr.org/2024/10/one-company-a-b-tested-hybrid-work-heres-what-they-found