Self-management is usually pitched as a choice, a progressive upgrade adopted by firms secure enough to experiment. Its more revealing origin is the opposite: the moment the normal hierarchy fails entirely and someone still has to keep the lights on. Crises do not just test power structures. They redistribute them. When the boss walks out of a bankrupt plant, the question of who decides gets answered by whoever is left.
Recovered firms as a natural experiment in horizontal power
Argentina after the 2001 to 2002 collapse is the sharpest case. Pizzi and Brunet Icart (2014), in a peer-reviewed study of empresas recuperadas in Buenos Aires city and province, document workers occupying failed factories and reconverting them into cooperatives that they ran themselves. The old command structure did not evolve into a flatter one. It vanished, and the workers rebuilt authority from the floor up through assemblies, collective decision-making, and networks of mutual support between recovered firms.
This is decentralization under maximum stress, which makes it worth attention. The comfortable version of self-management can lean on slack, on a healthy balance sheet that forgives slow decisions. The recovered firms had none of that. Their horizontal structures had to keep production going and payroll met with the original owners and managers gone. Pizzi and Brunet Icart (2014) treat this as a phenomenon of social mobilization as much as management, and the evidence is qualitative, a set of documented cases rather than a controlled comparison. But it establishes something a laboratory could not: horizontal control can carry a real operating business through the failure of its hierarchy.
The assembly is the load-bearing institution in these firms, and it is worth pausing on why. When the owner leaves, the firm loses its default answer to who decides. An assembly supplies a new one that does not depend on any single person: decisions of consequence go to a vote of the members, each with an equal say. This is slower than a boss, and its defenders do not pretend otherwise. What it buys is legitimacy and durability. A decision that the whole floor made is a decision the whole floor will defend, which matters enormously when the firm is fighting creditors, courts, and the temptation for each worker to walk away. Horizontality here is not a productivity trick. It is the mechanism that keeps a fragile coalition together.
The theory behind the reflex
Why would crisis push toward horizontality rather than toward a strongman. Paul Singer, the leading theorist of Brazil's solidarity economy, offers part of the answer. Singer (2001) defines the solidarity enterprise as one governed by democratic administration: leadership is elected by the members, each with one vote, and strategic decisions go to an assembly. In his account, solidarity makes cooperation between enterprises natural, in contrast to the destructive competition of the conventional market. When an economy fractures and firms can no longer count on the market to coordinate them, mutual support becomes a survival strategy rather than an ideology.
Singer (2001) is careful about the limits, and so should anyone citing him be. His is a theoretical and conceptual argument, not an empirical proof, and he warns that small solidarity ventures need scale and consolidation, invoking Mondragón, to overcome poverty and stay competitive. Horizontality keeps a firm alive; it does not automatically make it large or efficient. That caveat matters, because the romance of the occupied factory can obscure how hard the next decade is.
What crisis teaches conventional firms
The 2020s gave ordinary companies a milder taste of the same dynamic. When the pandemic scattered workforces overnight, decisions that had always climbed the chain suddenly had to be made locally, by whoever was closest to the problem, because the chain was too slow. That is the recovered-firm logic in a gentler key: under shock, authority migrates toward the people with the information, and rigid hierarchy becomes a liability.
Semco's own turn is a reminder that this can be deliberate rather than desperate. Semler (1989) describes remaking the firm around participation, profit-sharing, and open books, distributing power on purpose so it would not have to be seized in a crisis. The recovered firms did the same thing under duress, and paid a far higher price for it.
There is also a warning in the record against romanticizing the model. Singer (2001) insists that solidarity ventures need scale and consolidation to survive, pointing to Mondragón, and the recovered firms that endured were often the ones that found allies rather than standing alone. Horizontality answered the immediate question of who decides. It did not by itself answer the harder questions of capital, markets, and growth, which is why survival for these cooperatives was a decade-long fight rather than a moment of liberation. Anyone drawing lessons for a conventional company should take both halves: the structure held, and it held at a cost.
The practical lesson is not that every company should wait for bankruptcy to flatten. It is that a hierarchy's brittleness is invisible until it is tested. A structure where only the top can decide is a structure that stops deciding the moment the top is removed or overwhelmed. Building some horizontal capacity in advance, teams that can act without permission, information that everyone can read, is cheap insurance. The workers who took the keys in Buenos Aires proved that horizontal control can run a factory. They also proved it is a hard way to learn.
Sources
- Pizzi, A. & Brunet Icart, I. (2014). Autogestión obrera y movilización social: el caso de las empresas recuperadas argentinas. Peer-reviewed qualitative study of workers running recovered factories through assemblies and mutual-support networks after the 2001 crisis, the central case. https://muse.jhu.edu/article/540857
- Singer, P. (2001). Economia solidária versus economia capitalista. Defines the solidarity enterprise as democratically governed with one member one vote and argues solidarity makes inter-firm cooperation natural, with a scale caveat. https://www.scielo.br/j/se/a/Xy7BmyrV8tHfwKNVhmSXFyw/?lang=pt
- Semler, R. (1989). Managing Without Managers. Describes Semco distributing power deliberately through participation and open books, the planned counterpart to crisis-driven horizontality. https://hbr.org/1989/09/managing-without-managers