Organisations have spent the last decade flattening hierarchies. Agile teams, decentralised managements, self-organising squads – for all of these, the architectural logic is consistent: distribute the power and let everyone’s voice matter equally. It’s a compelling vision. A new longitudinal study published in the Journal of Occupational and Organisational Psychology suggests that the reality might be more complicated.
What the study found
Harju & Tims (2026) studied 585 employees at a large Dutch financial organisation that had just dismantled most of its management structure by reducing managerial positions from 278 to just 25. Now these weren’t theoretical leaderless teams. These were real people, newly operating without formal hierarchy and re-learning how to work together.
The researchers tracked two things over six months: how respected employees felt within their teams (self-perceived status) and how much they spoke up either promotively (suggesting improvements and innovations) or prohibitively (raising concerns, flagging vulnerabilities, challenging practices). In the first three months, status and promotive voice fed each other in a tight loop. Employees who felt more respected spoke up more with ideas. Employees who spoke up more with ideas were perceived as more valuable. Meanwhile, lower-status members spoke up less, which further reduced their standing. The pattern was self-reinforcing.
One detail is particularly striking: the social pecking order formed fast. The window for status gain through voice was essentially open for the first three months, then narrowed sharply. By month four, the social hierarchy had largely stabilised. If someone hadn’t established themselves early, the evidence suggested that gaining influence became increasingly difficult.
The hierarchy that wouldn’t dismantle
This is where the findings become more philosophically interesting than the authors perhaps intended. Within weeks of forming, these leaderless teams had built informal status hierarchies from scratch, using the only available material: behaviour. Who spoke, how often and in what way. The titles were gone but the instinct to rank remained. This isn’t a design flaw. It may be closer to a feature of our species. Long before organisations existed, human groups developed status systems to coordinate collective action. Be it to decide whose assessment of a threat to trust, whose idea for a hunt to follow, or whose judgment to defer to under pressure. These systems predate agriculture, let alone Agile methodology.
The speed with which informal hierarchies emerged in this study suggests that such tendencies may run deeper than organisational design. Which raises an uncomfortable question that organisational psychology rarely puts this plainly: if humans instinctively recreate status hierarchies whenever they collaborate, if the informal pyramid assembles itself within weeks without a single job title in sight, is the promise of a truly egalitarian team not just difficult, but fundamentally at odds with how people actually work? The practical takeaway from the study is that teams and coaches need to intervene early, explicitly solicit voices and build psychological safety before the hierarchy calcifies.
Another finding complicates the story
The researchers expected that once team members had accumulated enough standing, they would begin using that status to raise concerns. The theoretical logic here was elegant: earn the right to dissent, then exercise it. Status would eventually unlock the harder, riskier form of contribution. But that didn’t happen. Not in any statistically meaningful way. Employees could build status by generating ideas. They could not build it by identifying problems. The social machinery that converts contribution into standing appeared strong for one kind of input and weak for the other. The surprise isn’t that raising concerns is socially risky. It’s that even high-status members, having presumably earned the latitude to challenge the group, didn’t do so. This points to something more structural than individual risk aversion. Teams appear to be genuinely better at recognising and rewarding those who help them imagine a better future than those who help them see a present danger clearly. Builders appeared more likely to accumulate standing. For diagnosticians, the study found no comparable pathway.
Which leaves us with a question worth sitting with: if the social reward system inside teams favors possibility over diagnosis, are organizations quietly selecting against exactly the kind of thinking that catches problems before they compound?
Reference
- Harju, L. K., & Tims, M. (2026). Stand out, speak out: Change‐related reciprocal relations between self‐perceived status and voice in leaderless teams. Journal of Occupational and Organizational Psychology, 99(2). https://doi.org/10.1111/joop.70116