The most demoralizing part of many jobs today is not the work. It is waiting. Global employee engagement has fallen to 20 percent, the lowest since 2020, with Gallup estimating roughly 10 trillion dollars of lost productivity, about 9 percent of world GDP (Gallup, 2025). Managers are not spared. Their engagement has slid from 27 to 22 percent since 2022, with the sharpest annual drop between 2024 and 2025 (Gallup, 2025). When overextended managers become approval routers, decision queues lengthen, and the organization starts to bleed speed and will.
The reflex in uncertainty is to centralize. It feels safer. Yet in turbulent conditions the safer bet, counterintuitively, is to move authority closer to the facts. The strongest field evidence we have from a major recession points that way.
What the data says when the storm hits
In a study spanning large datasets of management practices and plant-level decisions in the United States and other OECD economies, firms that had delegated more decision-making to local plant managers before the Great Recession outperformed more centralized peers in the sectors hardest hit by the downturn (Aghion, Bloom, Lucking, Sadun and Van Reenen, 2021). The authors’ research design is quasi-experimental. Using difference-in-differences across sectors with varying turbulence, they show that prior decentralization was associated with better growth in bad times.
The proposed mechanism is not ideology but information. Under turbulence, local knowledge becomes more valuable. The study corroborates this “localist” view with turbulence measures based on product churn and stock market volatility, and it shows the result is robust to alternative explanations such as managerial fear of bankruptcy or shifts in coordination costs (Aghion et al., 2021). The practical read is straightforward. When the environment moves quickly, decisions made by those who see changes first tend to be better and faster than those escalated to a distant center that sees them late.
This is not a claim that decentralization always wins. The same work implies that in more stable settings, the benefits of centralized coordination can loom larger. The point is contingency. In the economy firms face now, with shocks arriving in clusters, betting against local information is a luxury.
People speak up before they step up
Moving authority outward is a structural choice. Making it work is social. Teams only use new decision rights if it feels safe to do so. The foundational research on psychological safety defined it as a shared belief that a team is safe for interpersonal risk taking, and showed in a field study of 51 teams in one manufacturing company that psychological safety was associated with higher learning behavior. Learning behavior, in turn, mediated the relationship between psychological safety and performance, while team efficacy did not explain learning once safety was accounted for (Edmondson, 1999). The study is cross-sectional and confined to a single firm, so it does not settle causation across all settings. It does, however, offer a clear design implication. If you want people to act on local information, you must reduce the interpersonal cost of speaking up and experimenting.
Distributed decision-making without psychological safety invites paralysis. People wait for permission. Speak-up climates are the enabling layer that turns delegation from a paper policy into action.
How flat is flat enough
Some firms go further and remove the manager-subordinate relationship altogether. Self-managing organizations decentralize authority in a radical, formal, and systematic way across the whole organization, and their distinctive feature is the elimination of the hierarchical boss-subordinate tie (Lee and Edmondson, 2017). The scholarly synthesis is careful about scope. Managerial hierarchy works well for reliable execution of known tasks in stable environments. It tends to inhibit complex, non-routine problem solving that crosses functional boundaries, and it can harden status differences that stunt people’s development (Lee and Edmondson, 2017). The empirical base on the outer limits of self-management remains thin, so this is not a universal prescription.
What it offers leaders now is a compass. You do not need to abolish hierarchy to gain most of the benefit. The direction of travel is consistent. Move decision rights closer to the work when tasks are interdependent, information is local, and the world is choppy. Retain central coordination where integration and standardization create more value than local adaptation.
Make decision latency a management KPI
The simplest audit is to measure decision latency, the time from when a decision-relevant fact emerges to when a reversible action is taken. Most organizations do not track it. They should. In a low-engagement context with managers under strain, compressing this interval both reduces bottlenecks and signals trust (Gallup, 2025). The following moves are a practical starting point, consistent with the evidence on the value of local information and the need for a climate that supports learning:
- Map decisions to information hotspots. For each recurring decision, name the person or team closest to the facts. Default authority there. Escalate only when clearly defined thresholds are crossed. This aligns with the local-information mechanism identified in recession performance data (Aghion et al., 2021).
- Preauthorize small bets. Set explicit financial and risk limits under which local teams can act without further sign-off. Require post-action review to enable learning, which is the pathway by which psychological safety relates to performance (Edmondson, 1999).
- Publish decision service levels. For routine approvals that must stay centralized, commit to response times and report actuals. Visibility shortens queues and reveals where centralization truly adds value.
- Redesign the manager role. Shift from approver to context setter. Managers define guardrails, clarify priorities, and coach teams on trade-offs. The structural flattening this implies aligns with the conditions where hierarchy can hinder complex problem solving (Lee and Edmondson, 2017).
None of this claims a guaranteed uplift in every metric. Context matters. But the weight of the best available evidence argues that in turbulent periods the performance upside of distributing authority to where information resides is real, and that teams will only exploit it if they can ask questions, flag risks, and try new moves without social penalty.
The irony of centralization is that it often aims to protect the organization, yet in volatile markets it exposes firms to a slower, more damaging risk: decisions that arrive after the world has moved on. Treat decision latency as strategy. Put authority in the same zip code as the facts, and build the social conditions that let people use it.
Sources
- Aghion, Bloom, Lucking, Sadun and Van Reenen (2021). Turbulence, Firm Decentralization, and Growth in Bad Times. Firms that delegated decisions to local managers 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
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams. Defines team psychological safety and shows its association with learning behavior and performance via learning in a field study of 51 teams. https://journals.sagepub.com/doi/10.2307/2666999
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines SMOs and synthesizes when hierarchy helps or hinders, highlighting benefits and limits of radical decentralization. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Gallup (2025). State of the Global Workplace 2025. Reports global engagement at 20 percent, a cost of about $10 trillion in lost productivity, and a sharp drop in manager engagement. https://www.gallup.com/workplace/659610/state-global-workplace.aspx