Political Spillovers of Worker Representation: With or Without Workplace Democracy?

2.1 The Influence of Trade Unions

The overwhelming number of empirical studies on the political spillovers of worker representation examine the role of unionization. A couple of studies have been conducted for the United States. These studies find a positive link between unionization and workers’ propensity to vote (Becher and Stegmueller 2019; Freeman 2003; Kerrissey and Schofer 2013; Radcliff and Davis 2000; Lamare 2010a, 2010b). Union members are also more likely to engage in a broad range of political and civic activities such as volunteering, participating in protests, signing petitions, and attending political rallies or public meetings (Kerrissey and Schofer 2013; Zullo 2011). Evidence from Canada confirms the existence of political spillovers of worker representation (Bryson et al. 2013). Union members in Canada are more likely to vote in elections than nonmembers. They are also more likely to participate in demonstrations, sign public petitions, attend public meetings, contact politicians and volunteer for a political party.

A series of studies consider more than one country. Analyzing data from 32 countries all around the world, Flavin and Radcliff (2011) find that union members have a higher propensity to vote. Similarly, Kerrissey and Schofer (2018) use data from roughly 60 countries to show that union members participate more than nonmembers in various political activities with the influence being stronger in democratic countries and in less corporatist countries (countries where unions and employers’ associations are to a lesser degree incorporated into state structures). Based on data from 18 Sub-Sahara African countries, Karreth (2018) finds that union membership is positively associated with the frequency of workers’ participation in political activities such as voting, demonstrating, or contacting members of parliament. Bryson et al. (2014) find a positive relationship between union membership and voting propensity for 29 European countries. Using data from 15 member states of the European Union, D’Art and Turner (2007) show that the presence of a union in the firm and the worker’s individual union membership have a positive influence on voting propensity, political interest and political engagement. Turner, Ryan, and O’Sullivan (2020) analyze data from 11 stable European democracies to confirm that union membership is positively associated with political participation. Moreover, they find that union members have more favorable attitudes towards democracy.^[3]

The available evidence suggests that unions not only influence workers’ political and civic engagement, but also shape their political preferences. Studies for Australia (Leigh 2006), Brazil (Ogeda, Orneals, and Soares 2025), and the United States (Freeman 2003, Kerrissey and Schofer 2013) find that union members are more likely to lean towards left-wing parties. Using data from 16 Western countries, Arndt and Rennwald (2016) show that union members are more likely to vote for social democratic parties and are less likely to vote for the Greens, center-right parties, and radical right-wing parties. Mosimann, Rennwald, and Zimmermann (2019) use data from 11 European countries to confirm that union members have a higher propensity to vote for a social democratic and a lower propensity to vote for a radical right-wing party.

Studies examining attitudes towards specific political issues also support the notion that unionization fosters a left-wing political orientation. Kim and Margalit (2017) find for the United States that union members are more likely to oppose trade liberalization. Arndt and Rennewald (2017) obtain for 16 Western countries that union members show stronger support for government intervention and have a more tolerant attitude towards immigrants. Mosimann and Pontusson (2017) find for 21 Western and Eastern European democracies that union members have stronger preferences for redistribution. Interestingly, union members not only support redistribution, they are also more willing to contribute their share. Booth, Lup, and Williams (2017) and Zullo (2011) obtain for the United States that union members have a higher propensity for charitable giving than nonmembers.

2.2 The Influence of Works Councils

The available evidence suggests that nonunion representation plays a role in political spillovers, too. Recent research considers the role of works councils. Works councils provide a highly developed mechanism for representative worker participation in decision making at the establishment level. While works councils play a role in corporate governance in many European countries, research so far has focused on works councils in Germany. Compared to their counterparts in most of the other countries, German works councils have acquired quite extensive powers (Mohrenweiser 2022).

The rights of works councils are laid down in the Works Constitution Act (WCA). On some issues they have the right to information and consultation, on others a veto power over management initiatives and on still others even the right to co-equal participation in the design and implementation of policy. Their rights are strongest in social and personnel matters including payment methods, allocation of working hours, monitoring employee performance, and up- and down-grading. Empirical studies indicate that works councils may extend their influence even to issues that are nowhere covered by the WCA (Jirjahn and Smith 2006; Jirjahn, Mohrenweiser, and Backes-Gellner 2011).

Industrial relations in Germany are characterized by a dual structure of worker representation through both works councils and unions. Even though there are important linkages and overlaps, the two institutions have distinct functions. First, while unions have a redistribution function, works councils and employers are obliged by law to cooperate ‘in a spirit of mutual trust … for the good of the employees and of the establishment’. Second, while industrial action is the most important measure of unions to represent workers’ interests, communication and consultation play a key role in representation through works councils. Third, while unions are mainly concerned with wage negotiations and general working conditions, works council representation has a much broader scope as councils participate in almost every decision management makes. Fourth, while unions in Germany tend to represent workers’ interest at the industry level, works councils represent workers at the establishment level meaning that the councils are closer to the employees and their workplaces. Fifth, while unions particularly mobilize employees when negotiations over collective agreements occur, works council representation involves a more continuous participation in management decisions.

The institutional framework suggests that works councils have the potential to contribute to improved communication and exchange within the workforce – a potential that goes beyond the regular elections of works councilors held every four years. Once implemented the works council may fix hours for consultation. This allows workers to be in contact with the works council. Each worker has the right to propose issues to be discussed by the works council. Furthermore, the works council holds regular works meetings with the whole workforce to report about its activities and to discuss topics such as collective bargaining policy, social policy, environmental and financial matters, equal opportunities, or work-life balance. The works meeting may make suggestions to the works council and take a stand on its activities.

Of course, the functioning of works councils cannot be immediately derived from a reading of the WCA. The behavior of works councils is not completely determined by the letter of law. The institutional framework of establishment-level codetermination sets out general principles rather than specific rules. It involves indeterminacy and situational ambiguity (Jackson 2005). This implies that works councils have scope to set their own agenda and to decide which goals they pursue (Frege 2002; Jirjahn and Smith 2006). Thus, only empirical research can answer the question of whether or not works councils influence workers’ political behaviors and attitudes.

Works councils shall be elected by the workforce of establishments with five or more employees. However, their creation depends on the initiative of the establishment’s workforce. Thus, works councils are not present in all eligible establishments. This allows conducting within-country analyses comparing workers in establishments with and without a works council.

Jirjahn and Le (2024a) examine whether workers in establishments with and without a works council differ in their interest in politics. Research in political science has shown that interest in politics is strongly associated with political engagement and democratic citizenship. It is a key indicator of political engagement that is linked with more knowledge about politics, more systematic thinking about political decisions, a higher propensity to vote, and more political participation in other ways.^[4] Jirjahn and Le find that workers are more likely to have a strong interest in politics if a works council is present in the establishment.

A second study by Jirjahn and Le (2024b) analyzes whether workers in establishments with and without a works council differ in their party preferences. The study shows that the presence of a works council has a positive influence on preferences for the Social Democratic Party and The Left while it has a negative influence on preferences for extreme right-wing parties. Of course, this does not mean a simple switch of political preferences from extreme right-wing parties to the Social Democrats or even to the Left. As the authors emphasize, the dynamics of party preferences usually implies a movement from having no party preferences at all to having party preferences (or vice versa). Thus, workers who have initially no party preferences are less likely to develop preferences for an extreme right-wing party and are more likely to develop preferences for the Social Democratic Party or The Left. Altogether, the presence of a works council not only fosters workers’ political interest, but also steers the interest in a particular political direction.

The studies by Jirjahn and Le also provide separate estimations by gender. These estimations show a significant influence of works councils on political interest and party preferences for male, but not for female workers. The authors argue that this reflects the moderating role of asymmetric gender norms still prevailing in society. These roles infer that political engagement is more of a male than a female characteristic.^[5] Thus, the political behavior of women may be less responsive to circumstances encouraging more political participation. Altogether, the gender differences show that the political spillovers of worker representation depend on broader societal circumstances.

Finally, a study by Pfeifer (2023) examines the factors influencing satisfaction with democracy in Germany. The study finds that workers in establishments with a works council are more satisfied with democracy than their counterparts in establishments without a works council. This holds for both male and female workers. Pfeifer’s findings can reconciled with Jirjahn and Le’s results if one takes into account that satisfaction with democracy does not necessarily imply that workers are specifically interested and engaged in politics. Workers may be satisfied with democracy without being politically active. Establishment-level codetermination increases satisfaction with democracy for both men and women while it stimulates political interest and political preferences only for men.

3.1 Workplace Democracy as a Possible Explanation

Altogether, the basic point for our topic is that both union and nonunion worker representation have the potential to foster workers’ political engagement and shape their political preferences. This gives rise to the question of how worker representation influences workers’ political behaviors and preferences. One explanation is that worker representation leads to increased workplace democracy. Democracy at work in turn fosters workers’ political activities outside the workplace. From a theoretical viewpoint, this supposed chain of causation involves two steps that need further consideration. First, we have to clarify under which conditions worker representation is associated with increased workplace democracy. Second, we need to specify the transmission channels through which the experience of workplace democracy influences workers’ political activities outside the workplace.

Turning to the first step, advocates of worker representation are sometimes quick to assume that worker representation automatically implies increased workplace democracy (Müller-Jentsch 1995, 2008). However, workplace democracy means that two basic requirements are met (Jirjahn and Kiess 2024). On the one hand, worker organizations must have the power to break with authoritarian management forms and bring in the perspectives and interests of workers. On the other hand, worker organizations must represent workers’ interests in a manner that ensures the involvement of the rank and file and, hence, leads to more democratic experiences workers make at work.

Considering the second step, the literature suggests several transmission channels through which workplace democracy fosters workers’ political and civic engagement outside the workplace (Greenberg 2008; Jirjahn and Le 2024a). Workplace democracy may strengthen workers’ sense of efficacy stimulating them to participate in political activities (Pateman 1970). Workers may develop political skills facilitating political participation (Verba, Schlozman, and Brady 1995). Workers may gain a greater awareness of social and political issues (Bryson et al. 2013, 2014; Kim and Margalit 2017). Finally, intensified communication and interaction among workers may promote values of solidarity, collective responsibility, caring and compassion (Ahlquist, Clayton, and Levi 2014; Weber, Unterrainer, and Höge 2008; Weber, Unterrainer, and Schmidt 2009).^[6]

The possible transmission channels through which workplace democracy influences workers’ political behavior outside the workplace appear to be little controversial in the literature – even though they are often only postulated without being tested. By contrast, the question of whether worker representation involves greater workplace democracy is much more subject to controversial discussion. Thus, we need to consider in more detail whether or not the two basic requirements for workplace democracy – the power to influence management decisions and the involvement of the rank and file in the decisions of worker organizations – are met.

Turning to the first requirement, there is ample evidence that worker organizations have the power to challenge autocratic management structures and to bring in workers’ interests. Studies for a series of countries show that there exists a substantial collective bargaining wage premium (Brändle 2024; Jirjahn 2025). The influence of unions is not limited to wages. For example, in Britain workers’ unionization is positively associated with employers’ use of family friendly practices (Budd and Mumford 2004). Furthermore, research on German works councils shows that the presence of a council has far reaching consequences for the personnel policy of establishments (Jirjahn 2018; Jirjahn and Smith 2018; Mohrenweiser 2022). Establishments with a works council are not only characterized by a higher wage level, lower wage inequality and more stable employment relationships. They are also more likely to provide training (Stegmaier 2012), implement equal opportunity practices (Jirjahn and Mohrenweiser 2021), use flexible working time arrangements (Ellguth and Promberger 2004), promote occupational health and safety (Jirjahn, Mohrenweiser, and Smith 2022) and invest in environmentally friendly production (Askildsen, Jirjahn, and Smith 2006).

At issue is whether the second requirement is met and worker organizations represent workers’ interests in a manner that ensures the involvement of the rank and file. An involvement of the rank and file can be expected if worker organizations act as collective voice institutions. Collective voice theory assumes that worker organizations are institutions aggregating workers’ preferences (Bryson et al. 2014; Freeman 1976; Freeman and Medoff 1979; Jirjahn and Smith 2018). Aggregating workers’ preferences requires intense communication and discussion about work-related issues between worker representatives and workforce and also within the workforce. The worker organization as a collective voice institution has to bring its policy into agreement with the workforce. It helps workers find a consensus around common objectives and aligns its policy to the preferences of the workforce. The power of a worker organization to influence management decisions depends on the support by the workforce (Jirjahn, Mohrenweiser, and Backes-Gellner 2011). Democratic decision processes within the workforce increase this support (Gahan and Bell 1999; Stepan-Norris and Zeitlin 1995) and, hence, contribute to the effectiveness of worker representation (Koyi, Tembo, and Sichinsambwe 2021). Moreover, in the long run, worker representatives may be not reelected without workers’ support or the worker organization may lose its members if it does not involve the rank and file.

Of course, the term ‘aggregating workers' preferences' used in collective voice theory does not mean that representing workers’ interests is a passive or mechanic task where worker representatives simply track preferences and are completely tethered to those preferences. Representing workers’ interests involves that representatives take a leadership role. In order to coordinate workers and create a common good, worker representatives need some discretion for developing a vision, setting an agenda and also influencing workers’ views. The crucial question is how they fulfill their leadership role. Workplace democracy means that worker representatives pursue a democratic style of leadership actively involving the rank and file in their decision making. While evaluating situations and developing ideas worker representatives create at the same time a climate that allows debate, discourse, and deliberation. The basic point is that communication and influence flow in both directions – from worker representatives to the workforce and from the workforce to its representatives. The agenda set by a worker organization may be seen as a proposal made to the workforce (Darlington 2018). This proposal is based on exchange with the workforce and may be revised according to workers’ feedback.

3.2 Political Spillovers Without Workplace Democracy?

A contrasting view is that worker organizations are rather bureaucratic or even autocratic institutions (Honneth 2024) and, in the end, are subject to the ‘iron law of oligarchy’ (Michels 1915). According to this view, worker organizations act as service providers treating workers as more or less passive consumers (Morris and Fosh 2000).^[7] They act on behalf of the workforce without involving it. Worker representatives face a tradeoff between involvement of the workforce and strategic leadership (Gumbrell-McCormick and Hyman 2019). Thus, they may prefer a paternalistic leadership style with a top-down strategy, centralizing their power to increase the influence on management decisions. Moreover, there is a principal-agent problem (Bellante and Porter 1992; Kremer and Olken 2009). Worker representatives are agents of the workforce who cannot be perfectly monitored by the rank and file. This enables them to pursue their own private goals. Self-interested worker representatives may limit democratic processes within the workforce if suppressing critique and debate helps them secure their position (Taft 1944). They may provide some services to satisfy basic needs of the workforce, but primarily tend to secure their position and pursue their own goals.^[8]

The eventuality that worker organizations may not promote workplace democracy raises the question of whether there are alternative explanations for the political spillovers of worker representation. Palmieri (2024) suggests in the context of employee-owned firms that not only the governance rights, but also the economic benefits (wealth, income, and employment stability) associated with employee ownership influence workers’ political behavior. This reasoning can be applied to the context of worker representation (Bryson et al. 2014). Even if worker organizations were completely bureaucratic or autocratic institutions, they have to negotiate higher wages and better working conditions to justify their existence. Centralizing power through a paternalistic leadership style may even increase the power to successfully bargain over wages and working conditions. The basic point remains that higher wages and better working conditions can have a positive influence on workers’ political and civic engagement regardless of whether or not worker organizations contribute to workplace democracy. Thus, an alternative explanation for political spillovers is that worker representation has an influence on workers’ political behavior just because it improves the economic situation of the workforce.

However, at a closer look, it can be doubted that the economic benefits provided by worker organizations alone can explain the full pattern of evidence put together by the various empirical studies. Worker representation not only has an influence on the political engagement, but also on the political preferences of workers. It may appear reasonable to assume that higher wages and better working conditions enable workers to more effectively participate in civic society and political processes outside the workplace. One might even imagine that workers economically benefiting from autocratic worker organizations support labor-friendly parties to ensure that worker representation and, hence, their economic benefits are not challenged by policy makers. Yet, there is no clear reason to assume that higher wages and better working conditions influence political preferences in a way that workers prefer generally more state interventions, a generous welfare state and redistributive policies toward the poor.

Importantly, from an empirical viewpoint, some studies on political spillovers of worker representation control for the workers’ economic situation (Booth, Lup, and Williams 2017; Jirjahn and Le 2024a, 2024b; Kerrissey and Schofer 2013; Leigh 2006; Mosimann, Rennwald, and Zimmermann 2019; Zullo 2011). These studies confirm an influence of worker representation on the political behavior of workers. If studies show a link between worker representation and workers’ political engagement and preferences despite including controls for earnings and working conditions, this means that the economic benefits provided by unions or works councils are not the whole story. Economic benefits cannot completely explain the political spillovers of worker representation.

On the one hand, this may be seen as indirect evidence that worker representation influences workers’ political behavior through establishing workplace democracy. On the other hand, it may simply reflect that there are transmission channels other than wages and working conditions through which autocratic worker organizations influence the political engagement and possibly even the political preferences of workers. Information provision could be such channel. Even autocratic worker organizations need to provide some information on political and social issues to workers in order to ensure support and some mobilization of the rank and file. They will provide information favoring labor-friendly parties which are usually more sympathetic towards worker representation. However, one may question whether information provision without involving workers in decision processes can be effective in shaping political engagement and preferences. Only if information provision does not require worker involvement to be effective, worker representation could have political spillovers without fostering workplace democracy. Otherwise, the political spillovers of worker representation can only be explained by a positive influence on workplace democracy. If an autocratic style of worker representation reinforces workers’ apathy, workers are less likely to be responsive to the information provided by a worker organization. By contrast, if worker representatives involve workers in their decisions and contribute to a more active citizenship at work, workers will be more interested in the information provided by the worker organization.

A further channel could be that worker organizations exert pressure on workers or reward them with social recognition to be politically more active (Bryson et al. 2014). Such influences have been discussed in social custom models to explain how worker organizations overcome collective action problems (Booth 1985; Corneo 1995; Naylor and Cripps 1993). These influences may also play a role in workers’ political engagement. However, it can be questioned whether worker organizations are able to monitor workers’ political attitudes and behaviors outside the workplace to such a degree that it can explain the whole evidence on political spillovers we have. Moreover, workers’ responsiveness to social influences will be stronger if they develop solidarity and a sense of oneness. This requires communication and interaction within the workforce and between workers and worker representatives. It cannot be forced in an autocratic manner.

4.1 Unions

Buhlungu, Brookes, and Wood (2008) provide evidence from the Congress of South African Trade Unions – South Africa’s largest and most active union federation. A survey of union members shows that a large majority of the members attend union meetings at least monthly and believe that union representatives must consult members. The authors suggest that this indicates internal union democracy.

Of course, union leaders can differ in the degree to which they promote workplace democracy. Thus, workers may differ in the democratic experiences they make with unions. Some studies use this variation to examine whether democratic experiences with unions make workers more active union members. Buttigieg, Deery, and Iverson (2008) find for Australia that union members are more willing to take industrial action if they see union leader as being responsive to members’ needs in situations of perceived workplace injustice. Another Australian study by Gahan (2012) examines how union members respond to dissatisfaction with their union. The study shows that union members are more likely to voice their dissatisfaction directly to the union and are less likely to remain silent if they perceive the union as being responsive to worker’s voice.

Evidence from the United States points into the same direction. Sadler (2012) finds that a democratic leadership style of union presidents induces workers to voluntarily participate in union activities. Johnson and Jarley (2004) use workers’ perceptions of union justice (the degree to which union representatives treat workers honestly and fairly) as a proxy for granting members control over decisions within the union. The authors find that workers are more willing to participate in union activities if they perceive a higher degree of union justice.

The findings of the studies have an important implication for the theoretical interpretation of the political spillovers of unions. Political spillovers require that workers overcome apathy and actively participate in decision making within the sphere of work. The available evidence shows that more democracy within unions stimulates an activation of workers within the sphere of work. This suggests that increased workplace democracy underlies the political spillovers of unions. Of course, there is heterogeneity in the degree to which unions promote workplace democracy. Thus, a careful interpretation is that the political spillovers documented by a series of studies indicate that unions on average stimulate workplace democracy.

4.2 Works Councils

Further evidence comes from Germany. Jirjahn and Kiess (2024) examine the influence of works councils on workers’ experience of democracy at work. The authors find that workers in establishments with a works council are more likely to have a higher degree of collective efficacy – workers’ perception that they have joint control over what happens at work. Furthermore, workers in establishments with a works council are more likely to report an open organizational climate allowing them to even discuss sensitive topics that otherwise may be suppressed by management.

Moreover, Jirjahn and Kiess show that unionization plays both a direct and a crucial moderating role. The influence of works councils on workers’ collective efficacy and perceptions of an open organizational climate is much stronger among union members than among non-members. Moreover, for union members, the presence of a works council is even associated with increased self-efficacy – a worker’s belief that their personal engagement can make a change to improvements at work.

The findings by Jirjahn and Kiess fit political spillover theory well. Political spillover theory emphasizes that increased efficacy is one transmission channel through which workplace democracy stimulates workers’ political engagement outside the workplace (Pateman 1970). The findings by Jirjahn and Kiess provide evidence of exactly this transmission channel. They indicate that worker representation has an influence on workers’ political engagement by fostering their sense of efficacy at work.

5.1 Basic Implications and the Need for Further Research

Altogether, theoretical reasoning and empirical evidence suggest that worker representation has a positive influence on workplace democracy. This supports the view that worker representation involves political spillovers by promoting workplace democracy. Of course, this does not mean that the economic benefits associated with worker representation, simple information provision beyond workplace democracy or social pressure and recognition do not play a role. It appears to be likely that all of the various transmission channels are at work. Nonetheless the basic point for our topic remains that workplace democracy is one channel and that this transmission channel is likely to reinforce the other transmission channels.

Future research could fruitfully examine the relative strength of the various transmission channels and their interactions in more detail. In particular, it would be interesting to fully examine the chains of causation using a mediator analysis. On the one hand, such mediator analysis would jointly examine the influences of worker representation on workplace democracy, information provision, social factors and economic benefits. On the other hand, the mediator analysis would investigate the influences workplace democracy, simple information provision, social factors and economic benefits have on workers’ political engagement and preferences.

Of course, there are many other aspects future research could address. On a broader scale, a fruitful avenue for future research could be to examine the specific transmission channels through which workplace democracy has a particular influence on workers’ political behaviors and attitudes. As outlined in Section 3.1, a stronger sense of efficacy, improved political skills, greater awareness of political issues and increased solidarity are possible transmission channels highlighted by political spillover theory. A comprehensive mediator analysis would also account for these transmission channels.

Finally, it should be noted that most of the empirical studies discussed in this article are of cross-sectional nature. This gives rise to the question of whether the relationship between worker representation and the political behaviors and attitudes of workers can be interpreted as causal. One may suspect that politically active and progressive-leaning workers are more likely to be union members and sort into workplaces with worker representation. Some studies address the issue of causation by applying fixed effect estimations (Jirjahn and Le 2024a, 2024b) or an instrumental variables approach (Leigh 2006). These studies confirm that worker representation involves political spillovers and, hence, suggest that the link between worker representation and workers’ political behaviors and attitudes is not simply driven by a possible self-selection. This increases our confidence in the results of the cross-sectional analyses. Even if cross-sectional analyses cannot dig deeper into the issue of causation, this does not necessarily mean that the associations they find do not reflect causal processes. It just means that they should be complemented by studies addressing the issue of causation in more detail. If these complementary studies provide supporting evidence, this increases the credibility of the cross-sectional findings. Of course, future research could pay more attention to the issue of causation.

5.2 Workplace Democracy is a Matter of Degree

Advocates of a far reaching democratization of work tend to dismiss existing institutions of worker representation either as being too bureaucratic (Honneth 2024) or being too weak (Ferreras, Battilana, and Méda 2022).^[9] This appears to reflect an idealization of democracy that only accepts a maximum degree of worker involvement as workplace democracy. Most importantly, it does not accord with the available evidence. Democratization of work is a matter of degree. Even though existing institutions of worker representation only imperfectly conform to the ideal of democracy, they positively influence workers’ democratic behaviors and attitudes beyond the narrow boundaries of the workplace. Theoretical reasoning and empirical evidence suggest that workplace democracy plays a role in this influence.

Of course, available studies just suggest that worker representation has political spillovers through promoting workplace democracy. So far they do not identify the degree of workplace democracy. Future quantitative research definitely has to undertake much more effort to examine the extent to which current forms of union and nonunion worker representation contribute to workplace democracy in different settings.

The optimal degree of workplace democracy is particularly important from a policy viewpoint. At issue is whether strengthening the rights of worker organizations and expanding democratic processes within these organizations can reinforce political spillovers and, hence, will help stabilize democracies in times of increasing political apathy on the one hand and globally spreading authoritarian populism and right-wing extremism on the other. There is an ongoing discussion on union renewal (Voss 2010). Expanding the degree of democracy within unions could be an important ingredient not only to revitalize unions, but also to counteract antidemocratic tendencies within society.

In a similar vein, strengthening the rights of works councils and ensuring that works councilors involve the rank and file could stimulate workers’ democratic behaviors beyond the narrow boundaries of the firm. The available evidence on political spillovers of works councils comes from Germany where the rights of works councils are strong compared to other countries that have established this institution. Thus, it could be worthwhile to strengthen works council legislation in these countries. Clearly, an implementation of works councils may be also interesting for countries where this institution is not existent so far. For example in the United States, the interest in nonunion employee representation has been spurred by a sharp decline in union density and the growth of a substantial representation gap in the workforce (Freeman and Rogers 1999; Hertel-Fernandez, Kimball, and Kochan 2022).

In the end, from a policy viewpoint, a crucial question is whether there is a tradeoff between strengthening democracy within worker organizations and strengthening the rights of worker organizations vis-á-vis employers. If worker representatives need some discretion to effectively challenge authoritarian management structures, this may impose some restrictions on the degree of democracy within worker organizations. However, to the extent an effective representation of workers’ interests requires support by the workforce, democracy within worker organizations will contribute to the bargaining power of worker organizations. In any case, policy recommendations have to consider the accountability of worker representatives and the implications for the democratic spillovers of worker representation.

Moreover, careful policy recommendations also have to take into account the economic consequences of worker representation for productivity, innovation, investment, and employment. Evidence on the economic consequences of unions is mixed (Brändle 2024; Doucouliagos et al. 2018; Laroche 2021). Research on German works councils provides a more positive picture suggesting that works councils have neutral to positive economic consequences for the economic performance of firms (Jirjahn and Smith 2018; Mohrenweiser 2022). The basic point is that the degree to which worker representation should be strengthened will depend on whether or not there is a tradeoff between democratic spillovers and economic performance. If there is a tradeoff, the benefits of democratic spillover have to be weighed up against the loss in economic performance. By contrast, if there are positive consequences for both democracy and economic performance, there is a much higher potential for strengthening worker representation. In the end, institutional details and the specific design are very likely to play a crucial role. Whatever it may be, it is important to recognize that narrow policy views focusing solely on either the democratic or the economic consequences do not appear to be appropriate as worker representation has manifold effects on society and the economic sphere.

5.3 Challenges

Any policy recommendation to strengthen workplace democracy has to recognize that institutions of worker representation face very serious challenges. Workers’ unionization and the collective bargaining coverage of firms are in decline in many countries (Schnabel 2020). In Germany, not only the share firms covered by collective bargaining, but also the share of firms with a works council is in sharp decline (Addison et al. 2017). The decline in worker representation may be one reason for the demise in democratic participation outside the workplace we can observe in many countries. Thus, the most important political issue appears to be to counteract the tendency of declining worker representation.

While there are a series of possible factors contributing to the decline, the resistance of employers to worker representation plays an important role. This has been documented very clearly for the United States (Cooke 1985; Hatton 2014; Schmitt and Zipperer 2009). The United States even have a union avoidance industry; i.e., specialized law and consultancy firms dedicated to defeating union organizing campaigns and keeping their clients union free (Logan 2006). Nowadays, union avoidance law firms located in the United States have internationalized their operations and provide advice to a growing number of multinational companies (MNCs). They have expanded their reach to Latin America, Europe, and Asia (Logan 2019, 2020). The evidence on employer resistance to worker representation is not confined to the United States. In Germany, owner-managers appear to show strong opposition to works councils (Behrens and Dribbusch 2020; Jirjahn and Mohrenweiser 2016). This does not necessarily reflect economic reasons. Owner-managers seem to oppose codetermination because it reduces the personal utility they gain from being the ultimate bosses within the firm.

Globalization and particularly the activities of MNCs are a further challenge to workplace democracy (Jirjahn 2024). On the one hand, there is evidence of a race to the bottom among countries. Countries compete for inbound foreign direct investment (FDI) and the ability to attract FDI depends among others on a country’s industrial relations system. Thus, countries may weaken institutions of worker representation to undercut their competitors and attract FDI. On the other hand, to the extent foreign MNCs invest in a country, they tend to engage in activities to avoid worker representation and, hence, challenge the host country’s industrial relations system from within.

Finally, the political spillovers of worker representation depend on the broader socio-political context. While worker representation increases the resilience of a country’s democratic political system, the flourishing and the democratic functioning of worker representation depends on the democratic socio-political environment. Kühne and Sadowski (2008) provide evidence of a biased media coverage of codetermination in Germany. Mass media disproportionately report about negative instances of codetermination. Biased media coverage is likely to undermine public support for worker representation. The cross-country study by Kerrissey and Schofer (2018) shows that political spillovers of unionization are less pronounced in less democratic countries. Boudreau et al. (2024) find for Myanmar – a country with a highly autocratic history (Horwitz and Cooke 2021) – that union leaders tend to crowd out workers’ speech and do not build consensus around workers’ views. Instead they align workers’ views with those of the union. Another example of the functioning of worker representation in an autocratic country comes from China where unions are controlled by the State Party (Zhu, Warner, and Feng 2011).

For democracies, the basic point is that the populist tendencies observed in many countries (Guriev and Papaioannou 2022) may reach a tipping point at which worker representation largely forfeits its ability to produce democratic spillovers. In Germany, there have been recent attempts by radical right-wing groups to nominate candidates for works council elections and to ideologically indoctrinate workforces (Dörre 2018; Kim, Greef, and Schroeder 2022; Schroeder et al. 2019). While these groups have not been very successful so far, it is an open question of whether or not they will gain more influence in the future. If these right-wing group were able to infiltrate works councils, they would use this institution for spreading their ideologies. This could entail that works councils lose their function to contribute to a stabilization of the democratic system.