Corporate codetermination, which allows workers to elect representatives to a board of directors, is in the news again. Proposals for codetermination were, for example, a prominent part of this past presidential campaign, with senators Sanders and Warren disagreeing over what portion of a board workers could choose. Multiple bills in the Senate would implement codetermination and have sparked accusations that, if enacted, they would harm American economic growth.
To test these and other claims, economists and others have looked to real-world examples of codetermination, mainly in Europe, but their empirical studies have provided little information that’s relevant to the United States. The results have been notably inconsistent, and the studies that are most credible have for the most part shown that codetermination has small if any effects on the economy. Important questions have remained unanswered. How would codetermination requirements interact with existing U.S. corporate governance laws? What would the political reaction be to these corporate governance changes?
A theoretical perspective can provide some of the answers. Game theorists have developed a commonly used method for studying elections called the electoral accountability framework. In this framework, voters are viewed as the principal and elected officials as their agents in a principal-agent problem. In a new paper, we apply the electoral accountability framework to corporations, mathematically analyzing a context stripped of all specific institutional features: corporate regulations, political-economic institutions, etc. We then ask, what effect does a policy change allowing workers to elect corporate directors have?
Our results are straightforward. In comparing a context in which a manager unilaterally controls wages, hiring, or investment with one in which workers elect a manager who makes these decisions, introducing elections changes nothing about firm behavior except for the distribution of profits between corporate executives and workers. When the firm makes economic profits (in the absence of a perfectly competitive market), the election of managers by workers allows workers to obtain a share of profits greater than they otherwise would.
The logic of the formal model is straightforward as well. In the basic environment that we analyze, workers and managers have a shared interest in the firm doing well, making profits through efficient hiring and wage decisions and so on. However, workers and managers have conflicting interests over the division of firm profits. Hence, granting workers more power within the firm by giving them votes for the board of directors increases the share of profits going to workers without affecting other aspects of firm behavior. This conclusion is robust to complications we add, including the potential to hire and fire workers or invest in productive capacity. The idea is that, even while electorally accountable, the manager remains a residual claimant who has an incentive to maximize firm value, analogous to results in political science.
This formal model has important implications for understanding empirical research on codetermination in Europe. Since the mathematical environment that we analyze is so simple, it effectively holds other variables constant. Inconsistent results in studies of Germany, France, and Scandinavia suggest that idiosyncratic features of each country have an impact on codetermination policies. Therefore, the effects shown in these studies are not the effects of codetermination per se. The theoretical model shows that codetermination is consistent with efficient firm behavior. The observable differences in firm behavior in codetermination regimes across countries must then be explained by other institutional features.
Our analysis also holds implications for policy debates over codetermination. When thinking about policy proposals for codetermination in the U.S., in some ways the least important element on which to focus is codetermination. As we have argued, our model implies strong theoretical reasons to believe that codetermination will shift profits to workers without needing to alter other aspects of firm behavior. But each of the codetermination bills in Congress is really a bundle of many different policies. And these various policies might have positive or negative effects, either independent from or in interaction with a codetermination policy.
For proponents of codetermination, our analysis brings good news. It is always possible, in principle, for a codetermination policy to be consistent with efficient corporate governance. Of course, no one can say whether American politics would cause codetermination to work poorly if implemented alone. Yet our model shows that it is always possible to implement codetermination in a way that enhances worker economic benefits while remaining neutral for other aspects of firm behavior. If there is reason to believe that a specific proposal would not do that, then there are other aspects of that proposal (other rules, regulations, etc.) that proponents of codetermination could change while preserving codetermination itself.
For skeptics of codetermination, our analysis points to particular areas of concern. The analysis involves internal firm decision-making and the distribution of profits among actors within a firm (managers, workers, investors, etc.). It does not, however, address external firm factors, such as lobbying. This concern is raised by Gatti and Ondersma (2021), who worry that codetermination could increase the influence of firms pursuing bad policies from government.
Yet our analysis also suggests a way around this. While we focus on workers as the main constituency to be enfranchised through codetermination, the mathematical analysis is agnostic as to whether firm constituents granted a vote for corporate boards are workers or other stakeholders in firm policies. To address concerns about workers and corporate executives working together to pursue socially harmful policies from government, one might consider adding representatives from community, environmental, human rights, or other groups on corporate boards.
This post comes to us from David Foster and Joseph Warren at the University of California, Berkeley. It is based on their recent article, “Electoral accountability in the workplace,” available here.