Return‑to‑office edicts, AI anxiety, and a visible slide in manager morale have created a brittle mix. Global employee engagement has fallen to 20 percent, its lowest since 2020, with Gallup estimating about 10 trillion dollars in lost productivity, roughly 9 percent of world GDP (Gallup, 2025). Manager engagement is deteriorating faster, down from 27 to 22 percent in a single year, while the best organizations report 79 percent of managers engaged, nearly four times the global average (Gallup, 2025). That spread is the point. It is not a pandemic hangover. It is a design gap.
Stop treating disengagement as a mood problem
Executives often reach for communications plans and perks when energy sags. The better lever is structural. Gallup’s comparison between typical and top‑quartile organizations implies that engagement is largely a function of how work is set up, not a passing sentiment (Gallup, 2025). The research on self‑managing organizations makes the same argument from the other side. Lee and Edmondson define self‑managing organizations as those that decentralize authority in a formal, systematic way and eliminate the manager‑subordinate relationship as the core unit of control (Lee and Edmondson, 2017). Their synthesis of the management literature is blunt: conventional hierarchy excels at reliable execution in stable settings, but it tends to inhibit cross‑functional problem solving, amplify rigidity, and reinforce status differences that stifle people’s development (Lee and Edmondson, 2017). In a decade defined by non‑routine, interdependent work, that is a poor fit.
This does not mean every firm should scrap managers. It does mean the default assumption that decisions must climb ladders before they move is at odds with the work many teams now face. Where to start is not with a reorg chart, but with the task itself.
Let people shape their tasks, then watch engagement move
Job crafting, the disciplined practice of employees adjusting tasks, relationships, and perceptions of their roles, has been studied in controlled interventions. A meta‑analysis of 14 such interventions found that job crafting programs produce small but reliable gains in job crafting behaviors and, crucially, increase work engagement (Hedges’ g around 0.31) and contextual performance (Hedges’ g around 0.39), with the confidence interval for performance starting near zero and widening upward (Oprea et al., 2019). The mechanism matters. Once engagement gains are accounted for, added job crafting behaviors do not further explain performance, which means the performance lift runs through engagement rather than through crafting per se (Oprea et al., 2019). For leaders, that is actionable. Design time‑boxed job crafting sprints, support people in reshaping a slice of their role, and measure engagement before and after.
The economics are not hand‑wavy. A utility analysis in the same meta‑analysis shows that job crafting interventions can generate substantial economic returns for healthcare organizations, even at small effect sizes (Oprea et al., 2019). The effect sizes are not large, and they should not be oversold, but they are consistent across studies and tied to an outcome boards care about.
The team variable you can actually manage
Empowerment without safety is theater. Edmondson’s seminal field study of 51 work teams in a single manufacturing firm introduced the construct of team psychological safety, the shared belief that the team is safe for interpersonal risk taking (Edmondson, 1999). In that setting, psychological safety was associated with learning behavior, while team efficacy was not once safety was controlled, and learning behavior mediated the link between safety and performance (Edmondson, 1999). The design caveats are important. This was cross‑sectional evidence from one company, not a randomized trial. Yet it established the mechanism leaders can influence today: make it normal to surface problems, ask naïve questions, and experiment in small loops, then track the learning behaviors that connect to results.
Blend that with the job crafting evidence and a pattern appears. The credible route to performance runs through engagement and learning, not performative autonomy. Empowerment works when people can reconfigure parts of their job and when teams feel safe enough to test and talk through the consequences in the open.
When structure is the bottleneck
Team climate and job design go only so far if authority remains centralized. The literature on self‑managing organizations explains why firms tackling complex, cross‑boundary problems often need a different scaffold. By design, these organizations push decision rights to the edges and replace supervisor approval with formal, distributed mechanisms for coordination and accountability (Lee and Edmondson, 2017). The claim is not that hierarchy never works. The claim is about fit. Where the work is stable and predictable, classic hierarchy still delivers reliable execution. Where the work is uncertain and interdependent, that same hierarchy can slow decisions and suppress development, especially when status cues make people less likely to speak up or stretch into new responsibilities (Lee and Edmondson, 2017).
Seen against Gallup’s engagement slide, the managerial middle looks less like a talent deficit and more like a design trap. Managers are less engaged than before, and the best organizations show that high managerial engagement is possible at scale (Gallup, 2025). That suggests a choice. Either keep the manager’s job as coordination bottleneck and morale cop, or redesign roles and routines so managers become stewards of context and capability while more decisions move to those closest to the work. The former invites burnout and attrition. The latter aligns with what the evidence says about engagement and learning.
The practical read
Treat empowerment as an operating decision. Three moves follow from the evidence, recognizing the limits and strengths of each study. First, run lightweight job crafting interventions at regular intervals, with clear guardrails, and track engagement and contextual performance at the team level, knowing the expected effect sizes are modest and flow through engagement (Oprea et al., 2019). Second, make psychological safety a managerial duty and a team habit, then measure learning behaviors explicitly, while keeping in mind the foundational study was one firm and cross‑sectional (Edmondson, 1999). Third, clarify decision rights and reduce approval layers where work is complex and cross‑functional, drawing on the self‑management literature’s definition and fit logic rather than assuming a one‑size‑fits‑all model (Lee and Edmondson, 2017).
If you need a board‑level case, start with retention and productivity. Gallup’s numbers frame the cost of doing nothing, and the gap between common and best practice suggests the room to move is large if leaders focus on design, not slogans (Gallup, 2025). Empowerment is not a perk. It is a choice about how work gets done, who decides, and how people learn their way to better performance.
Sources
- Gallup (2025). State of the Global Workplace 2025. Reports global engagement at 20 percent, a 10 trillion dollar productivity cost, and a sharp drop in manager engagement, underscoring a design gap between typical and best-practice firms. https://www.gallup.com/workplace/659610/state-global-workplace.aspx
- Oprea, Barzin, Vîrgă, Iliescu & Rusu (2019). Effectiveness of job crafting interventions: a meta-analysis and utility analysis. Shows small but reliable gains in engagement and contextual performance from job crafting interventions, with performance effects mediated by engagement and positive utility in healthcare. https://doi.org/10.1080/1359432X.2019.1646728
- Lee & Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines self-managing organizations and argues hierarchy fits stable tasks but constrains complex problem solving and development, framing when to decentralize authority. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams. Finds psychological safety associated with learning behavior, which mediates its link to performance in a field study of one firm, establishing a mechanism leaders can act on. https://journals.sagepub.com/doi/10.2307/2666999