What Does Codetermination Do?

In liberal market economies such as the United States, firms are controlled ultimately by their shareholders or owners, and (under the dominant legal doctrine of “shareholder primacy”) are governed with the exclusive purpose of maximizing the welfare of those shareholders and owners. By contrast, large firms in many European countries are jointly governed by their shareholders and workers, through “codetermination” arrangements mandated by law. Under “board-level” codetermination, workers elect representatives who fill a certain share (usually 20-40%) of the seats on their company’s board. Under “shop-floor” codetermination, workers elect shop-floor representatives who have rights to information, consultation, and sometimes co-decision-making over decisions about working conditions. Recent legislative proposals in the United States, including the Reward Work Act and Accountable Capitalism Act, and recent campaign proposals in the United Kingdom and Australia, would introduce board-level codetermination requirements for large firms in those countries. In addition, the possibility of amending the National Labor Relations Act to allow for experimentation with shop-floor codetermination in the US has been raised repeatedly in recent years (Liebman, 2017; Andrias and Rogers, 2018; Block, 2018).

In our recent NBER Working Paper, “What Does Codetermination Do?” (Jäger, Noy, and Schoefer, 2021), we provide a comprehensive overview of codetermination, covering the institution’s background, implementation, and the best available evidence on its economic effects. A priori, codetermination could have either positive or negative effects on both workers and firms. Proponents of shared governance argue that codetermination helps correct for imbalances of power between workers and employers that leave workers vulnerable to exploitation, and that codetermination raises productivity and improves worker satisfaction by facilitating information sharing and enabling worker investment in firms-specific human capital (Strine Jr., Kovvali, and Williams, 2021). By contrast, opponents of codetermination warn that worker influence over firm governance impedes efficient decision-making and leads to “hold-up” problems that discourage capital formation because potential investors know that the fruits of their investment can be captured by workers (Jensen and Meckling, 1979; Dammann and Eidenmueller, 2020).

We adjudicate between these theoretical arguments by surveying the available empirical evidence on the economic effects of codetermination. Recent studies estimate the economic impacts of board-level and shop-floor codetermination by exploiting “natural experiments” in which the institution was abruptly introduced or abolished in some firms but not in others, enabling “difference-in-differences” comparisons that uncover the causal effects of codetermination. These studies include Jäger, Schoefer, and Heining (2021), studying board-level codetermination in Germany (summarized here), and Harju, Jäger, and Schoefer (2021), studying both board-level and shop-floor codetermination in Finland.

The evidence we survey indicates that existing codetermination institutions have either zero or small positive effects on workers and firms. On the worker side, the introduction of codetermination in an individual firm leads to a 0 to 1.5% increase in wages; it does not seem to reduce the rate of voluntary turnover; it may slightly increase job security by reducing the frequency of involuntary layoffs, perhaps at the expense of hiring; and it may lead to increases in subjective job satisfaction. From the firm’s perspective, codetermination has zero (if anything, positive) effects on capital formation, productivity, and profitability. Overall, the existing evidence does not support the warnings of opponents of codetermination, but also indicates that codetermination does not have strong positive impacts on the welfare of workers.

Existing studies estimate the “partial-equilibrium” effects of the introduction of codetermination in a “treated” group of firms, compared to a control group of firms in the same economy that are not exposed to codetermination. However, codetermination laws could have economy-wide “general-equilibrium” effects not picked up by partial-equilibrium studies. For example, codetermination laws could influence a country’s overall industrial relations, or shift the competitive equilibrium of the labor market. To evaluate this possibility, we study the aggregate effects of a series of codetermination reforms in Western European countries between 1960 and 2015, using a difference-in-differences design with synthetic control groups. We find no evidence that codetermination laws shift aggregate outcomes such as wages, the labor share, GDP, capital formation, or the intensity of strikes, though we find suggestive evidence for increases in union density as a result of codetermination laws.

From the existing evidence and our cross-country estimates, we conclude that existing codetermination arrangements have largely zero or incremental positive effects. We devote the last sections of the paper to exploring reasons for these limited impacts, and offer three potential explanations. First, existing codetermination laws convey relatively little power to workers. Board-level codetermination arrangements give workers a minority of seats on company boards, meaning workers can always be overruled by unanimous shareholders, and shop-floor worker representatives typically have limited and narrowly defined powers. Surveys confirm that a majority of worker representatives feel they lack substantive decision-making power. The strongest existing codetermination arrangement is “parity codetermination,” under which large firms in the German coal, iron, and steel sectors must allocate 50% of their board seats to workers. Credible evidence on the economic impacts of parity codetermination is scarce, and it remains possible that arrangements like parity codetermination, which convey more power to workers, have commensurately stronger effects.

Second, European countries have widespread cultures of worker participation in decision-making that operate even outside of formal codetermination. Survey evidence indicates that workers in European firms without formal worker representation are informally involved in decision-making to almost the same degree as worker representatives are in firms with formal codetermination. It is possible that these cultures of worker participation are a consequence of the historical implementation of formal codetermination laws, but our cross-country estimates fail to find evidence that codetermination reforms improve the cooperativeness of industrial relations, and we argue that a reading of the historical evidence suggests instead that codetermination laws arose in countries with pre-existing cultures of industrial democracy and worker mobilization.

Third, codetermination may interact with other European labor market institutions, such as widespread union representation, industry-level collective bargaining frameworks, and labor market regulations. Countries with codetermination laws, such as Germany, Norway, and Sweden, also have very widespread firm-level union representation and have strong collective bargaining agreements negotiated at the industry level that set standards for wages and working conditions. In addition, the labor markets of these countries are tightly regulated by global standards. These institutions may already capture the low-hanging fruit when it comes to protecting workers and securing good working conditions, leaving little room for codetermination to have a marginal impact.

We close the paper by discussing implications of these explanations for recent codetermination proposals in the United States. If existing codetermination arrangements have limited impacts because they intrinsically do not convey much power to workers, then the minority board-level codetermination proposals in the Reward Work Act and Accountable Capitalism Act are likely to have similarly limited effects. By contrast, if codetermination’s limited effects are explained by features of European industrial relations or European labor market institutions that are comparatively absent in the United States, then codetermination could have larger effects (either positive or negative) if implemented in the US. Moreover, we discuss important complementarities between codetermination, union representation, and industry-level collective bargaining which may be relevant for the introduction of codetermination in the US, where union density is low and industry-level collective bargaining does not exist.

The complete paper is available here.