RTO edicts keep multiplying, yet morale is not rebounding. Global engagement slid to 20 percent in 2025, with Gallup estimating about 9 percent of world GDP lost to poor engagement, roughly 10 trillion dollars (Gallup, 2025). Managers are not spared. Their engagement fell sharply from 27 to 22 percent in a single year (Gallup, 2025). At the same time, 70 percent of people now hold an insular mindset, hesitant to trust those with different values or backgrounds, and only 32 percent believe the next generation will be better off (Edelman Trust Institute, 2026). The paradox is that employers are also the institution best placed to act as “trust brokers,” able to scale fair interaction and conflict skills across divides (Edelman Trust Institute, 2026). The implication is plain. Treat trust as infrastructure. Write it into roles, routines, and how decisions are made.
The trust deficit is an operating problem
The engagement slide is not a vibe issue. It is a design problem visible in the numbers. Gallup’s global survey reports only one in five people engaged and a large productivity cost attributed to disengagement (Gallup, 2025). The drop in manager engagement is especially telling. When those who administer process and performance feel detached, the system signals oversight and compliance, not mutual commitment (Gallup, 2025). Layer on the broader social pattern that seven in ten hesitate to trust across differences, and you have a workplace where disagreement easily becomes avoidance and where policy hardens to control by default (Edelman Trust Institute, 2026).
Employers have a lever. Edelman finds they are the most credible “trust brokers,” with a smaller gap between what people expect and what they deliver, and with the ability to scale training for conflict resolution and structured interactions across value differences (Edelman Trust Institute, 2026). This is not a plea for town halls. It is a brief for building trust into how work is organized.
Default trust requires formal decentralization
Decades of management theory say hierarchy excels at known tasks in stable settings, but it constrains cross-functional problem solving, creates rigidity, and amplifies status barriers that stunt development (Lee and Edmondson, 2017). That is a poor fit for work that crosses boundaries and changes often. The alternative is not chaos. Self-managing organizations decentralize authority in a radical, formal, and systematic way, and their hallmark is the elimination of the manager–subordinate relationship as the basic building block (Lee and Edmondson, 2017). The point is clarity, not anarchy. Decision rights are explicit, roles are dynamic, and escalation paths are designed rather than implied by rank.
Trust by default becomes visible when a team can answer three questions in writing without calling a meeting:
- Who has the right to decide this, and on what basis?
- What information is open by default, and what is the narrow exception?
- How do we surface and resolve a principled objection?
A company that can answer those questions has already reduced the need for supervisory gatekeeping. It has turned trust into structure. The conceptual payoff matches the pattern in the literature. When hierarchy is applied to novel, cross-boundary work, it slows learning and invites status plays. When authority is decentralized in a formal way, people can act without waiting for permission, and routine friction drops (Lee and Edmondson, 2017). The evidence here is a synthesis, not a randomized trial, so causality should be read with care. Still, the design logic is consistent and actionable.
The team-level mechanism is psychological safety
Trust at the system level is hollow if teams punish candor. Psychological safety is the shared belief that the team is safe for interpersonal risk taking, a construct defined and tested in a seminal field study of 51 work teams (Edmondson, 1999). In that study, psychological safety was associated with learning behavior, which in turn mediated the relationship with team performance. Team efficacy did not explain learning once safety was accounted for (Edmondson, 1999). The caution matters. The data were cross-sectional and from one company, so we cannot claim universal causation. But the mechanism is robust enough to guide practice.
Translate that into design moves:
- Make written operating agreements explicit. These should set norms for asking for help, raising mistakes, and challenging assumptions without penalty. The goal is to make the expected risks visible and acceptable, which fosters the learning behaviors linked to performance in the field study (Edmondson, 1999).
- Replace status signals with role clarity. If contribution is tied to a role with defined decision rights, not to a title, people speak up more and defer less to hierarchy. This aligns with the critique that status differences inhibit development and problem solving (Lee and Edmondson, 2017).
- Measure what you mean. Use brief pulse questions about comfort admitting errors or asking for input, then pair the results with lightweight experiments to improve team learning routines. This treats psychological safety as a system variable, not a poster.
Employers as trust brokers: make it teachable and scalable
Edelman’s latest survey argues employers can broker trust directly, with an explicit callout to conflict resolution training and structured interactions among people with different values (Edelman Trust Institute, 2026). There is a simple reason this belongs in the operating model, not HR’s calendar. Without a shared method for disagreement, decentralized authority gets jammed by avoidance or personal rivalry.
A practical playbook fits on one page:
- Train for principled conflict. Teach and rehearse a common method for surfacing reasons, testing assumptions, and ending with a documented decision. This is exactly the kind of capability Edelman flags as scalable by employers (Edelman Trust Institute, 2026).
- Default to open information. Publish decisions, rationales, and metrics by default, with narrow exceptions. Openness reduces status games, a point aligned with the literature’s warning about the developmental and problem-solving limits of hierarchy (Lee and Edmondson, 2017).
- Redesign the manager role. Where you keep managers, make their primary output the conditions for learning and engagement. The spread in manager engagement from 22 percent globally to 79 percent in best-practice firms signals that design choices, not fate, explain a large share of the gap (Gallup, 2025). That is not causal proof, but it is a strong hint that structure changes behavior.
The close is financial as well as cultural. If disengagement is consuming a measurable slice of the world’s output and if polarization is eroding willingness to collaborate, then the cheapest lever most firms control is how they write trust into work. Formal decentralization, explicit safety norms, and teachable conflict skills are not perks. They are the operating system.
The next RTO memo will not fix a system that presumes mistrust. A redesign will. Start with decision rights and default-open information. Test team safety and train for conflict. Then hold leaders to the only standard that matters in a low-trust world: people can see how to act without permission, they can challenge without fear, and they can tell that their employer takes the brokering of trust as seriously as any budget line.
Sources
- Edelman Trust Institute (2026). Edelman Trust Barometer 2026. Finds a 70% insular mindset and only 32% optimism for the next generation, and identifies employers as the best-positioned trust brokers who can scale conflict skills and cross-value interaction. https://www.edelman.com/trust/2026/trust-barometer
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing, Research in Organizational Behavior. Defines SMOs and synthesizes evidence that hierarchy suits stable tasks but constrains cross-functional problem solving and reinforces status, motivating formal decentralization. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams, Administrative Science Quarterly. Shows team psychological safety is associated with learning behavior, which mediates performance, grounding the team-level mechanism for trust by default. https://journals.sagepub.com/doi/10.2307/2666999
- Gallup (2025). State of the Global Workplace 2025. Reports global engagement at 20%, an estimated $10 trillion productivity cost equal to 9% of GDP, and a sharp drop in manager engagement, highlighting the scale and design nature of the problem. https://www.gallup.com/workplace/659610/state-global-workplace.aspx