Who should control the workplace?

American debate over this question in recent decades has been limited mainly to the question of the balance of power between shareholders and managers. What has not been included in the debate is whether workers should have formal control over the places where they spend half their waking hours. This is an international anomaly — almost every country has political parties advocating for, at least, a balance between workers and owners in the running of business.

In particular, what is entirely absent from U.S. society — in contrast to the wealthiest and most equal European societies — is the practice of codetermination. This term refers to a range of practices formally including worker representatives in the running of a company, including legal rights for unions and/or elected “works councils” to veto, delay, negotiate, or be informed of company decisions. Representatives are also sometimes given powers to investigate and raise complaints surrounding health and safety, employment equality and discrimination issues. In addition to this, workers are often represented directly on the board of directors, giving them a voice in decision-making at the very highest level.

The models adopted vary widely between countries, and have often been altered, strengthened, or undermined over time. Public policy in the United States would benefit from a codetermination project, and existing European codetermination structures provide a wealth of experience to draw from. Here we look at Sweden, Germany and France.

European Models of Codetermination


The major union confederations in Sweden negotiate directly with employers on national, industrial and company levels. There is a codetermination law, but consistent with the Nordic labor relations model, it leaves the interpretation of details to direct negotiations between unions and employers. The unions are facilitated in these negotiations by having an expansive and enforceable right to information about the firm, though union officials have voiced worries that employers are increasingly ignoring their responsibilities towards unions in this regard.

Sweden has a unitary board structure much like the United States, and unions appoint two directors in firms with over 25 employees, and three in those with over 1,000. Typically this constitutes about one-third of the total number of directors, though the proportion is not set by law.

Employers must negotiate with unions over areas including changes in work organization and methods, personnel changes (including changes in supervisors), the preparation of the annual budget, structural changes, and alterations in the patterns of recruitment or dismissal. The legal framework typically only requires negotiations, and codetermination in practice occurs through wide-ranging collective bargaining on a range of levels, from the floor all the way up to national agreements, backed up by powerful unions which still represent the vast majority of the population.

In addition to this, there are union-appointed safety representatives who are empowered to investigate injuries, take measures to deal with risks, and broadly monitor the workplace environment.


France’s trade union confederations seek to organize workers under their banner in each industry and workplace, and do so through competitive elections. By law, elections must be organized in every workplace with over 50 employees every 2 to 4 years (if a company owns multiple establishments with 50 employees, separate works councils are elected for each establishment alongside a central works council covering the entire company). If an employer does not take the initiative to organize such an election, an employee or union can request that they do so, and they are legally bound to do so — and the employee in question will be protected from dismissal.

Workplace elections are the key to understanding French industrial relations, which are complex and distinctive. There are three types of legally-empowered employee representative. The first is the trade union representative, who is appointed for each union that receives over 10 percent support in the industrial election. Unions deemed to be representative in this way are officially responsible for collective bargaining (which covers almost 97 percent of French employees), and also manage strikes, demonstrations, and other conflicts with management.

A union with less than 10 percent support is also allowed to appoint a representative, but they are not included in collective bargaining (except in very rare and limited cases, such as where no other union has a representative) and their primary role is to organize and build support for the next elections.

The second type of employee representation, the works council, is elected directly at industrial elections. They are dominated by union representatives from the various confederations in almost all cases, and are responsible for positive codetermination measures: they have the right to be consulted on any proposals affecting “the strategic orientation of the company; its economic and financial situation; and its social policy and conditions of work and employment”. Any information made available to the shareholders must also be made available to the works council.

The positive codetermination rights afforded to French works councils have been progressively weakened over the last decade. This has been particularly marked in the area of dismissals and layoffs, with “flexibility” continually used as a justification to move closer to at-will employment. Where once companies were required by law to prove that mass layoffs were economically necessary, they now merely have to consult the works council on a “social plan” to cushion the impact for those being laid off before they initiate firings. These “reforms”, fiercely opposed by organized labor, were championed by ministers including Emmanuel Macron — who now seems poised to mount further attacks on codetermination in France as President.

Finally, employee delegates are elected to monitor the implementation of existing collective agreements and labor laws — they do not bargain with employers or codetermine policy. The presence of elected officials who oversee agreements and ensure that policies agreed between the employer and unions or works councils is important for ensuring compliance, and employee delegates are an important method of raising complaints about abuses and violations of agreed policies.

Of the three countries we are comparing, France has the least ambitious requirements for board-level codetermination, only requiring one employee representative in publicly-traded firms with over a thousand workers (or two where the board has more than twelve members). In addition, however, two or four non-voting representatives from the works council may attend board meetings.


All German firms have a two-tier board — a management board comprised of executives, and a supervisory board that must ratify major decisions and oversee the management board. In firms with over 500 employees, a third of the members of the supervisory board are appointed by the workers, while in those with over 2,000 employees, the proportion rises to half. Ties are broken by a chair chosen by the shareholders. A more radical form, Montan codetermination, exists in the coal, steel and iron industries where ties are broken by a director appointed by consensus between the employee and shareholder directors, giving each side full parity on the supervisory board.

In addition to this, 90 percent of those in companies with over 500 employees are represented by works councils which are directly elected and are the vehicle through which codetermination in Germany operates. Approximately three-quarters of those elected to works councils in 2014 were union members, according to the Hans-Böckler-Stiftung, and German trade unions provide training and consultation to works councils.

German works councils are required to be informed and negotiated with on almost all workplace issues, but also have “positive codetermination rights” on a subset of issues. In these circumstances, the employer cannot introduce changes unilaterally but must either agree with the works council or convince a tribunal with a neutral chair to accept their recommendation. These areas include changes to starting and finishing times, breaks, overtime, bonuses and targets, the methods of payment, and the introduction of any surveillance of employees through cameras or other devices, among numerous others.


There are four main categories in which codetermination rights differ. The first is in the scope of who is covered: almost all codetermination laws have a range of thresholds, usually by number of employees, below which provisions are weaker or do not apply. A stronger codetermination law is one which covers a wider range of businesses with the most meaningful provisions.

A second key difference is in the method of worker representation. Germany and France use works councils which have legal rights that are clearly laid down in law. French unions are given the right to run official candidates in the first round of elections for their works councils, ensuring that they are linked to the labour movement, while in Sweden the codetermination law is a framework that regulates negotiations between employers and a union movement that is powerful enough to enforce these rights by itself.

Board-level representation is an aspect of all of these systems. Germany has “parity codetermination” with a much higher proportion of worker representatives on a supervisory board that is not responsible for the day-to-day management of the company. Sweden has a smaller number of worker representatives on a board structure more similar to the United States, and France has an even smaller number of representatives on a board structure chosen by each company.

Finally, the level of power given to worker representatives (whether in the form of unions or works councils) can vary considerably — both between countries and over time. The rollback of positive codetermination rights in France is a reminder that without confronting the question of ownership, there will always be conflict between the owners of capital and those who seek to have greater freedom within their workplace.

Codetermination in the United States

For the United States to adopt a codetermination system it will eventually be necessary to take the best aspects from various systems internationally and reconcile them into a model which will serve the necessary purposes. Clearly, anyone concerned with ensuring workers have more control over the workplace should advocate for codetermination legislation that gives workers real power over decision-making, rather than just talking shops which can be bypassed or ignored by employers.

Much as with plans for “universal healthcare”, there are good models — with parity board-level representation and expansive positive rights in decision-making; but there are also bad models, like the “comply or explain” codetermination floated by the Conservative Party in the United Kingdom. Under this model, there may be a single worker representative on boards with no powers at all — as long as employers can make up an excuse to ignore them.

Reducing the idea to a single buzzword could ensure that such proposals come to replace the substance of the idea, which is that workers in each firm should have a real influence on how it is operated.

While there will be details to work out in order to apply codetermination to the United States, it is important to point out that the best-known example, the one in Germany, has been implemented in a federal system. Thus it should not be the case that the particularities of the U.S. political structure pose any considerable barrier to the adoption of a meaningful codetermination system any more than they do to any other progressive reforms.

What does provide difficulties when considering codetermination structures in an American context is low private sector union density, at only 6.6 percent. Of course, the U.S. government should immediately repeal all of its anti-union laws and aggressively promote unionization across the country — but if it is not possible to reach Nordic levels of union density, it may be necessary to adopt a works council model with rights enforced through legislation rather than collective agreement — at least at first.

Does It Work?

The approach of supporting a codetermination system which maximizes the rights given to workers is based upon a conviction that their interests are currently massively undervalued, and empirical evidence that codetermination rights both at board and workplace level help to redress this balance.

Klas Levinson finds that in Sweden, where representation on boards is more limited but workplace-level codetermination is widespread, that “legal reforms and other forms of support have had the effect of extensively broadening and deepening local trade union activities related to management questions” but that workers on the board largely confined their interventions to questions over staffing, pay and conditions, and the working environment.

An extensive econometric study on board-level codetermination in Germany by Gorton and Schmid finds that workers in Germany, where there is parity codetermination in large companies, exercise real control over the direction of firms- they find that it changes the firm’s “objective function” in a way consistent with the aims of workers: parity codetermined firms were likely to be more diversified, reducing the risk of the firm failing, and were also less likely to undergo restructuring and layoffs. These are considered “evidence that codetermination shifts risk to capital” consistent with a real increase in the power of labor.

The study confirms that “employees have different objectives than those of the suppliers of capital and that achieving their objectives reduces the utility of the suppliers of capital” — or to put it another way, that classes continue to have different, often mutually exclusive, interests — and to struggle for those interests, even under codetermination.

Where employees are given meaningful power through codetermination, they use it. The evidence from Europe shows that workers actively participate in the running of their companies, often increasing productivity and making their business a better place to work, when they are given the ability to make changes and decisions.

We humbly suggest that American workers are no less capable of doing so.

A Step Forward

It is important to note that the concept of codetermination is not an inherently socialist or capitalist one. Despite a considerably more robust public sector and welfare state than the United States, the Swedish, German and French economies are capitalist.

What codetermination achieves is not a redistribution of ownership, but a method of ensuring that the particular interests of workers in a firm are reflected in how it operates and makes decisions. It can balance decision-making between a firm’s workers and its private owners just as well as it could between the workers and a publicly-owned wage earner fund — indeed, the latter would expand workers’ influence further if their representatives were included in committees and boards within its structures.

A codetermination law will not solve the fundamental question of control over capital — but we do believe that any system where all firms are not owned directly by those working in them — whether by private owners, the state, or any other organization, should have structures in place that grant employees direct influence and control over their daily lives and the operation of their workplace. Elizabeth Anderson has documented the way that employers behave as a private government — regulating speech, manners, clothing, and associations — almost always without any democratic input or influence from those being so governed.

For these reasons, codetermination rights deserve the support of progressives and social democrats concerned with workers’ rights and democracy, as well as those who support more ambitious measures to reform capital ownership in America.

Extending the democratic principle to the workplace has occupied the minds of politicians and thinkers from John Stuart Mill to Jeremy Corbyn. It is an idea whose time has come.

Mio Tastas Viktorsson also contributed research to this report.