Ask where workers have the most say and most people point north. They are right, but they usually blur two arrangements that pull in opposite directions. The first is codetermination: a legal right for employees to sit on the board that governs the company. In Sweden, companies above roughly 25 employees must seat two or three employee representatives if the union asks; in Denmark, firms past about 35 employees provide up to a third of board seats on request; in Norway, board representation scales up through thresholds at more than 30, then 50 to 200, then above 200 employees (worker-participation.eu). In these single-tier systems, elected worker representatives sit at the same table as shareholder representatives and vote on the direction of the firm.
The second arrangement is self-management, the Buurtzorg and Semco lineage, where the nurse or the machinist decides how the work is done without asking up the line at all. Both give workers power. They give completely different power, in completely different places.
Voice at the top versus authority at the edge
Codetermination is a bargain about representation. It accepts the hierarchy and the board and inserts worker voice into the governing body. The worker on the board does not decide how any particular job is done; they influence strategy, oversight, and the big decisions, on behalf of colleagues who remain, day to day, inside a normal chain of command. It is power over the direction of the firm, exercised by a few, at the top.
Self-management is a bargain about authority. It largely ignores the boardroom and instead hands each worker real decision rights over their own work: scheduling, spending, methods, sometimes hiring and pay. It is power over the doing of the work, exercised by everyone, at the edge. You can have a company with strong codetermination and rigidly hierarchical daily work, and you can have a fiercely self-managed company with a conventional board. The two bargains are not rungs on one ladder. They are answers to different questions: who governs, versus who decides.
Ricardo Semler's crisis story shows the self-management bargain in its rawest form. As Ricardo Semler has told it, during Brazil's 1990 crisis Semco put the numbers on the wall and let the workers themselves decide: rather than layoffs, they accepted a pay cut in exchange for control over spending and a larger share of the profit when things recovered. Recovery came. His line was that families do not open their books, but companies that respect you do. Notice what this is not. No worker took a board seat. The power that moved was operational and financial control over their own unit, handed down to the shop floor, not representation handed up to the boardroom.
What the honest evidence says about codetermination
Here the research delivers a useful jolt. The instinctive assumption is that putting workers on boards must significantly change wages and outcomes. Jäger, Noy and Schoefer (ILR Review, 2022) reviewed the micro evidence and ran cross-country event studies of codetermination reforms from the 1960s to the 2010s, and found surprisingly modest effects: zero to small positive impacts on productivity, wages, and job stability, and no detectable effect on aggregate wages, the labor share, or inequality. Their explanation is telling. Existing codetermination laws convey relatively little authority to workers, and the countries that have them already had high baseline levels of informal worker voice. The board seat, it turns out, is a real right that changes less than you would expect, partly because so little decision-making authority actually flows through it.
That finding sharpens the contrast rather than dismissing codetermination, which delivers legitimacy, oversight, and a durable seat at the table that self-management alone does not. But if the goal is to change how work feels and functions day to day, the evidence suggests representation at the top is a weaker lever than authority at the edge. A worker who helps elect a board member twice a decade has a voice; a worker who decides how today's job gets done has control. The Nordic map holds both. The lesson for anyone drawn to the region's reputation is to be clear about which bargain they actually want, because they are not the same purchase, and the cheaper-looking one, a couple of board seats, may move the least.
Sources
- Jäger, S., Noy, S. & Schoefer, B. (2022). What Does Codetermination Do? ILR Review 75(4):857-890 (NBER WP 28921). Micro evidence and cross-country event studies find zero-to-small-positive effects of codetermination on productivity, wages, and job stability, and no effect on aggregate wages or inequality. https://www.nber.org/papers/w28921
- ETUI / Worker Participation. Board-level employee representation across Europe. National thresholds for employee board representation: Sweden ~25 employees, Denmark ~35, Norway staged at >30 / 50-200 / >200. https://www.worker-participation.eu/National-Industrial-Relations/Across-Europe/board-level-employee-representation