Return-to-office mandates, budget freezes, and a thicket of approvals are back. So is something leaders rarely admit out loud: many managers look exhausted. Gallup reports manager engagement fell to 22 percent in 2025, down nine points since 2022, while the global cost of low engagement is roughly 9 percent of world GDP, about 10 trillion dollars in lost productivity (Gallup, 2025). If the people paid to sign off on decisions are the least engaged group, adding more gates is the wrong reflex.
There is a better move. Make the manager smaller. Turn approvals into guardrails and push more routine and context-rich calls to the teams closest to the work. The evidence says this pays off in hard times, and the organizational science is clear on the conditions that make it safe.
What the data actually show about decentralization
A multi-country study by Aghion, Bloom, Lucking, Sadun and Van Reenen examined firms that had already delegated decision rights from headquarters to local plant managers before the Great Recession. Using two large micro datasets, including the World Management Survey across ten OECD countries and the US Census Bureau’s Management and Organizational Practices Survey, they compared performance through the downturn. In the sectors hit hardest, the firms that had decentralized earlier outperformed their more centralized peers (Aghion et al., 2021).
The authors argue the mechanism is simple and specific. Under turbulence, local information becomes more valuable. Factory and regional managers see product churn, supply hiccups, and customer shifts first. When authority sits where that information lives, firms adapt faster. The results were robust to alternative stories such as managers taking more risks because they feared bankruptcy or broad changes in coordination costs (Aghion et al., 2021). This is not a call to flatten for fashion. It is evidence that, when conditions are noisy, moving decisions closer to the edge produces better outcomes.
The middle layer is tired. Redesign the job, not the people.
Gallup’s global survey also shows the gap between common practice and better practice is large. In top-quartile organizations, 79 percent of managers report being engaged, almost four times the global average of 22 percent (Gallup, 2025). That is an organizational design story, not a macroeconomic one. Structures that trap managers as gatekeepers of every budget tweak and customer concession are structures that turn good people into bottlenecks.
Right-sizing authority does not mean eliminating managers or wishing away coordination. Michael Lee and Amy Edmondson define self-managing organizations as those that decentralize authority in a radical, formal, and systematic way, eliminating the manager–subordinate relationship as the default (Lee and Edmondson, 2017). Their review is theoretical rather than causal, and it draws a sharp boundary: traditional hierarchy works well for reliable execution of known tasks in stable conditions, but it inhibits cross-functional problem solving and learning when work is complex and non-routine (Lee and Edmondson, 2017). Most companies straddle both worlds. The design task is to keep tight control where work is programmable, and to shrink managerial gatekeeping where novelty and local knowledge dominate.
Safety is the enabler of speed
Shifting decisions outward raises the stakes of everyday conversations. Teams need to surface problems early, admit uncertainty, and revise course without fear of blame. The seminal field study on psychological safety defined it as a shared belief that the team is safe for interpersonal risk taking, then tied it empirically to learning behavior and, through learning, to performance (Edmondson, 1999). The study covered 51 teams in a single manufacturing company and was cross-sectional, so it cannot claim causation across settings. Even so, the mechanism it documents is directly relevant. Distributed decision rights without psychological safety produce silence and second-guessing. Safety plus clear rights produces faster cycles of test, learn, and improve.
A practical redesign for smaller managers and faster calls
What gets measured still matters. What changes is who gets to decide first, and what gets audited after the fact. Four moves help.
-
Write decision-rights charters. Spell out which decisions teams own outright, the thresholds that trigger escalation, and the metrics used to review decisions later. Align rights with where information lives. The Aghion evidence suggests this alignment pays off when the environment is volatile (Aghion et al., 2021).
-
Pre-set ranges, not ad hoc approvals. Give teams authority within defined financial and risk bounds. When the bounds are exceeded, escalate. When they are not, move.
-
Standardize interfaces, not every choice. Hierarchy excels at repeatable execution, which is where rules belong. Codify handoffs and data definitions so teams can act locally without breaking the system. This preserves reliability where it is needed, consistent with Lee and Edmondson’s distinction between routine and non-routine work (Lee and Edmondson, 2017).
-
Make the manager a coach and context broker. Shrink approvals to the minimum, then hold managers accountable for building psychological safety and clarity. Safety is not about comfort. It is about enabling the learning behavior that Edmondson’s study linked to performance through improved team learning (Edmondson, 1999).
Leaders will ask about risk. The control does not disappear. It moves. You review after the fact against explicit guardrails instead of before the fact through subjective gates. You also get an insurance policy against disengagement. If managers are tired, the surest way to lose them is to make them arbiters of every small decision. If teams are tired, the surest way to lose them is to make them ask permission to do the obvious.
The trade-off is visible in the numbers and the literature. A downturn benefits the firms that already shifted authority toward where information is richest (Aghion et al., 2021). Engagement gaps at scale point to design, not talent, as the root cause (Gallup, 2025). And when teams feel safe to learn in public, they try more, notice more, and improve faster, which is the only durable advantage when plans meet reality (Edmondson, 1999).
The practical implication is blunt. In volatile markets and with waning managerial energy, flatten within guardrails. Formalize who decides what, at what threshold, and how decisions will be reviewed. Invest in managers as coaches who create safety and context. Keep the rules where work is repeatable, and put judgment where the information lives. Decisions will move faster, and so will your firm.
Sources
- Aghion, Bloom, Lucking, Sadun and Van Reenen (2021). Turbulence, Firm Decentralization, and Growth in Bad Times. Firms that delegated authority before the Great Recession outperformed centralized peers in the hardest-hit sectors, consistent with the value of local information under turbulence. https://www.aeaweb.org/articles?id=10.1257%2Fapp.20180752
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines SMOs and synthesizes evidence that hierarchy suits stable, known tasks while inhibiting complex, cross-boundary problem solving, framing when to decentralize. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams. In a field study of 51 teams in one manufacturing firm, psychological safety was associated with learning behavior, which mediated performance, highlighting conditions for safe distributed decisions. https://journals.sagepub.com/doi/10.2307/2666999
- Gallup (2025). State of the Global Workplace 2025. Reports manager engagement at 22 percent in 2025, 79 percent in best-practice organizations, and a global cost of low engagement near 9 percent of GDP, indicating a design problem that distributed decision-making can address. https://www.gallup.com/workplace/659610/state-global-workplace.aspx