Board-level codetermination: A driving force for corporate social responsibility in German companies?

Abstract

We examine the relationship between board-level codetermination and corporate social responsibility in German companies, engaging with two distinct literatures. Most quantitative studies of codetermination focus on its economic impact, with little attention to other outcomes. Studies of corporate social responsibility rarely consider the role of worker representatives. Our new measure of the strength of codetermination, the Mitbestimmungsindex (MB-ix), shows a positive relationship with ‘substantive’ policies such as the adoption of targets for reducing pollution, but not with ‘symbolic’ policies, such as membership of the UN Global Compact. We therefore shed new light on the role of codetermination and provide a more differentiated view of the spread of what has been termed ‘explicit’ corporate social responsibility in Germany.

Keywords

Codetermination corporate social responsibility sustainability worker participation

Introduction

This article examines the relationship between board-level codetermination and corporate social responsibility (CSR) in German companies, building a bridge between two almost completely separate research fields (Harvey et al., 2017). Although it is not the only country with ‘workers on the board’ (Gold et al., 2010), codetermination was introduced much earlier in Germany than elsewhere (Dukes, 2005). Furthermore, the proportion of board members who are worker representatives is particularly high in comparative perspective, reaching parity with shareholder representatives in large German companies.

Given these special characteristics, it is not surprising that German codetermination has attracted considerable research interest. At the time of writing, we are aware of 37 quantitative studies (a list can be found in our online appendix) examining the relationship between codetermination and company performance in Germany (for a review, see Jirjahn, 2010). However, these studies for the most part address a narrow set of outcomes, mainly financial variables like stock market performance and profitability. The relationship between codetermination and company policies with respect to the environment and society, referred to here as CSR, is almost completely unexplored in the literature.

The second field of research we address focuses on the determinants of CSR (sometimes referred to as ‘sustainability’) at firm level. Although this literature has expanded greatly over the past 2 decades, few studies look at the role of workers or trade unions (for an exception, see Jackson and Bartosch, 2016). Instead, most pay attention to consumers, shareholders and NGOs (Carroll and Brown, 2018; Harvey et al., 2017). This is surprising given that employees are among the primary stakeholders.

To help bridge the gap between these two literatures, we ask whether there is a relationship between the strength of codetermination and CSR in German firms. To answer this question, we report some of the results of a research project to construct a codetermination index (MB-ix). This project has three aims: first, to develop a differentiated measure of the strength of codetermination, capturing the institutional variation among German companies in the ability of workers to express ‘voice’ in corporate governance; second, to expand research on the impact of codetermination beyond economic performance and third, to examine the development of codetermination over a long time frame (initially 2006–2015). (For more information on this project, see www.mitbestimmung.de/mbix)

The next section presents this new measure of codetermination, describing its construction and providing some descriptive data. We then summarize selected aspects of the discussion on CSR, particularly those dealing with comparative research. Next, we describe an empirical test of the determinants of different types of CSR policies in German companies, and show that codetermination strength is positively related to ‘substantive’ but not to ‘symbolic’ CSR policies. The concluding section focuses on the relevance of these results for the literatures on codetermination and CSR.

Measuring the strength of worker influence: the codetermination index (MB-ix)

Strong codetermination is seen by many observers as a key institution in the German political economy (Hall and Soskice, 2001; Streeck, 1984). Codetermination is considered an insurance mechanism for protecting workers’ investments in firm-specific skills, which are essential for the production of high-quality, customized goods. Furthermore, codetermination supports a ‘bottom-up’ approach to incremental innovation based on input from the shop floor.

Despite numerous quantitative studies of German codetermination, how it affects company performance has not yet been definitively answered. An analysis of the 37 quantitative studies known to the authors shows that most researches have focused on financial variables, thereby neglecting other possible impacts (see Table 1). The most frequent topics examined are stock market performance (14 of the 37 studies), company profitability (13 times) and productivity (7 times). Recently, management remuneration has also become popular (6 times). Few studies look at core issues for worker representatives, such as wages, employment levels and changes in employment. Thus, the types of variables where one would expect worker representatives to have the greatest impact have been rarely examined. The link between codetermination and CSR in German firms has been neglected in quantitative studies.

Table 1.

Main topics of 37 quantitative studies of German codetermination.

Number of times studied
Share value	Profit	Productivity	Executive pay	Wages	Other
14	13	7	6	5	18

Source: authors’ own analysis.

Items add up to more than 37 because of the studies with multiple subjects.

A further limitation of the existing studies is that, with only six exceptions, worker influence in all companies with a specific type of codetermination has been treated as homogeneous. There are three types of codetermination in German companies, each of which has different rules for representing workers (see Rosenbohm and Haipeter, 2019): one-third representation in smaller firms under the 2004 Drittelbeteiligungsgesetz; in larger firms under the 1976 Mitbestimmungsgesetz, parity representation but with a casting vote for the shareholder-appointed chair and in Montan firms in the steel and mining sectors, parity representation with a neutral chair.

Most quantitative studies have assumed that worker influence is equal in all parity codetermined firms. However, qualitative research has shown substantial variation within each codetermination type. For example, a number of traditional ‘paternalistic’ firms and newer ‘high tech’ firms have resisted the formation of works councils and the election of external trade union officials to supervisory boards, which tends to weaken worker voice (Girndt, 2006). Furthermore, unions tend to focus more on the largest companies in their sectors (Raabe, 2010). Thus, the assumption of equality between firms with the same codetermination rights neglects important variation.

A second argument for a more differentiated measure is new legislation at EU level which defines procedures for negotiating rules on codetermination in specific situations. This includes the 2001 European Company Statute, the 2003 European Cooperative Society Directive and the 2005 Directive on Cross-border Mergers. In all three cases, negotiations between management and worker representatives may result in an agreement on worker participation which deviates substantially from either the ‘parity’ or ‘one-third’ models (Weiß, 2016).

To address these gaps, a project was launched at the WZB in 2014. The first ambition is to develop a differentiated measure of board-level codetermination in German companies. This index (the Mitbestimmungsindex or MB-ix), developed with input from experts and practitioners, consists of six components. Each component takes a value between 0 and 100 and the overall measure MB-ix, which is a weighted average of all six components, also varies from 0 to 100 (Scholz and Vitols, 2016, 2018).

The first component measures the number and type of worker representatives on the supervisory board, where final decisions on key policies such as dividends, large investments and mergers and acquisitions are taken. This component takes into account both the proportion of worker representatives on the board and the extent to which they are connected to a works council or trade union, since articulation with other levels of collective representation will increase the strategic capacity of board level representatives (Jürgens et al., 2008). The highest score for this component is reached if half the supervisory board members are worker representatives, all the internal employee representatives (those working for the company) are works councillors and all trade union representatives are full-time officials.

The second component relates to the deputy chair of the supervisory board, who consults the chair to prepare meetings of the full board and to make emergency decisions between meetings. In some companies, there are two deputy chairs. In companies with strong codetermination, one of these deputies is supposed to be a worker representative. The full score for this component is reached if there is only one deputy chair, and this person is either an external trade union official or works councillor.

The third component measures the extent of worker representation on board committees. Since these committees take important decisions and formulate recommendations to the full supervisory board, the proportion of worker representatives has an important influence on their output (Höpner and Müllenborn, 2010). The full score for this component is achieved if half the members on all the key committees are worker representatives.

The fourth component focuses on the degree of fragmentation of worker representation through the internationalization of employment. The lower the proportion of employees in the ‘home’ country of a multinational, the more difficult it is to organize workers’ voice, given the diversity of national industrial relations systems. However, a European or an international works council is a forum for worker representatives from different countries to coordinate their voice (Waddington, 2010). The highest score for this component is achieved if all employees are in Germany (no fragmentation across national boundaries); or otherwise, if there is a European or an international works council to coordinate workers’ voice across borders.

The fifth component measures the importance of the supervisory board in the corporate governance of the firm. The decision-making powers of the supervisory board, which are defined by law, vary quite substantially across different company legal forms (Köstler et al., 2013). If the supervisory board has limited powers relative to the shareholders’ assembly, the potential for worker representatives to influence company decisions will also be limited. The full score for this component is reached if the company has a legal form which defines extensive decision-making rights for the supervisory board.

Finally, the sixth component assesses where responsibility for personnel policy is located in the management board. If this is with the chief executive officer (CEO) or the chief financial officer (CFO), personnel policy will tend to be subordinated to the financial priorities of the company (Giertz and Scholz, 2018). This is in part because these types of managers tend to interact most extensively with shareholders. The full score for this component is obtained if primary responsibility for personnel policy is not assigned to the CEO or CFO.

One aim of the project is to trace the development of codetermination strength over a longer time period. Initial data collection was for the 10 years 2006–2015 inclusive. For each year, the sample includes all companies in the major stock indexes DAX (largest 30 firms), MDAX (50 mid-size firms), SDAX (50 small firms) and TecDAX (30 technology firms), which by definition adds up to 160 companies. In addition, data were collected for the parity codetermined companies listed on the stock market, which are not in a major index; this number varied between 30 and 35 during the period. In all, data on 1924 company-years are included. As there is a certain amount of fluctuation in index membership from year to year, 285 companies were in the whole sample for at least 1 year.

As expected, there is substantial variation in the distribution of MB-ix scores. Values for the 192 companies in the study in 2015 (the latest year for which we have complete data) can be seen in Table 2. Six percent of the companies in the sample received the highest score (MB-ix = 100), whereas 30 percent had no codetermination (MB-ix = 0). There are even cases where workers’ voice is stronger in companies with one-third participation than in some companies with parity codetermination; three companies with one-third participation had MB-ix values higher than the parity codetermined company with the lowest MB-ix value.

Table 2.

MB-ix distribution in German companies, 2015.

MB-ix Values	n	%
MB-ix = 100	11	6
100 > MB-ix ⩾ 90	32	17
90 > MB-ix ⩾ 80	26	14
80 > MB-ix ⩾ 70	30	16
70 > MB-ix ⩾ 0	35	18
MB-ix = 0	58	30
Total	192	100

Comparative perspectives on CSR

Much CSR literature focuses on the determinants of policies at the firm level (Aguinis and Glavas, 2012), for example, in the role of stakeholders in pressurizing firms into adopting specific policies. However, the stakeholders that are typically considered are consumers, NGOs and responsible investors, to the almost complete neglect of trade unions and works councils (Harvey et al., 2017).

Although most CSR studies focus on one country, some researchers have used large multi-country datasets to take a comparative approach. Catering to the growing demands of investors for consideration of ‘ethical’ criteria or reputational risks in their investment decisions, ratings companies such as Sustainable Asset Management, ASSET4 and Vigeo/Eiris gather data on the CSR policies and performance of thousands of companies around the world. As CSR is understood to span a number of dimensions such as the environment, social issues (including labour practices, diversity policies and human rights) and corporate governance, these datasets include hundreds of data points for each year for each company examined.

A key question is, in what kinds of countries do firms have better CSR performance? Two theses predominate. The first is that CSR has emerged as a substitute for government regulation. As governments in the past few decades have deregulated important areas such as labour markets, firms are pressurized by stakeholders to adopt CSR policies to fill the gap, or do so voluntarily. This has been labelled ‘explicit’ CSR (Matten and Moon, 2008), as firms attempt to get public credit and legitimacy for these policies. This approach expects that CSR policies will be most common in countries with weak regulation: ‘liberal market economies’ such as the UK and USA (Hall and Soskice, 2001). Evidence has been found in support of this thesis (Jackson and Apostolakou, 2010; Kinderman, 2012).

The contrary thesis is that CSR functions as a mirror of national institutions. According to this view, CSR policies will be more prevalent in countries that have strong states and coordinated industry associations. This has been labelled ‘implicit’ CSR, as firms implement such policies mainly in response to government requirements or collective pressure. This approach expects to find stronger CSR policies in ‘coordinated market economies’, such as Germany and the Nordic countries. Evidence for the validity of this second view has also been found (Campbell, 2007; Favotto et al., 2016; Gjølberg, 2010).

Some recent comparative research has provided a more differentiated view by arguing that both hypotheses may be partially true. Looking at membership in CSR associations, Kinderman and Lutter (2018) argue that CSR will initially be more prevalent in liberal market economies, but as it becomes more established, coordinated market economies will have an advantage. Jackson and Bartosch (2016), in one of the few studies that include the strength of worker participation at the national level as an explanatory variable, show that works council rights are positively related to social and human rights elements in CSR policies but negatively related to charitable spending. Matten and Moon (2008: 416), the originators of the implicit/explicit distinction, note that ‘explicit’ CSR policies seem to be spreading from the USA to Europe. One reason they give is the ‘deregulation of labor markets and the weakening position of trade unions … In cases of redundancy … European companies increasingly assume responsibility for fulfilling stakeholder expectations rather than relying on welfare state institutions’. However, they claim that explicit policies are largely restricted to larger companies and are driven by government policies and industry associations.

These results suggest a need for further research into the determinants of different types of CSR policies. As the influence of national institutions may differ across firms (codetermination legislation may apply only to specific kinds of companies), more differentiated measures of institutions at the firm level would be appropriate. Second, as types of policies differ across broad areas such as ‘social’ and ‘environment’, a more detailed examination of specific CSR policies might shed more light on the ‘substitute’ versus ‘mirror’ controversy.

A distinction in the literature between ‘symbolic’ and ‘substantive’ CSR actions is relevant for this discussion. Schons and Steinmeier (2016) define the difference as follows:

While symbolic CSR actions are any actions related to CSR topics that a firm takes to show ceremonial conformity, substantive CSR actions involve actual changes at an operational level, generally implying tangible and measurable activity that requires the use of a firm’s resources. (p. 359)

We would expect that workers as stakeholders would have little interest in symbolic CSR but would be supportive of substantive CSR actions that would improve their situation, such as less exposure to pollution at the workplace.

MB-ix and CSR in German companies

Data and hypotheses

We now describe an empirical test of the relationship between the strength of codetermination and CSR in German companies. Data on CSR policies in 96 firms for 2006–2014 were selected from the ratings company ASSET4 as indicators of different types of CSR policies (a more detailed description of the data can be found in our online appendix). The first two variables are indicators for ‘symbolic’ CSR policies, as they are oriented mainly at external legitimation, with little, if any, change at the operational level. We present the ASSET4 description of the variable in italics followed by our motivation for using the variable:

GLOB_COMP: Has the company signed the UN Global Compact? The UN Global Compact has been analysed as a high-visibility initiative which, however, makes few substantive demands on companies for implementing CSR policies. Companies are asked to sign up to 10 principles but ‘all credible and publicly available data and documentation conclusively demonstrate that the UNGC has failed to induce its signatory companies to enhance their CSR efforts and integrate the 10 principles in their policies and operations’ (Sethi and Schepers, 2014: 193). Membership can thus be considered a symbolic CSR policy.

PUB_COMM: Has there been a public commitment from a senior management or board member to integrate [environmental, social and governance] issues into the company strategy and day-to-day decision making? This indicator is purely a measure of public statements by key persons in the company, and as such it can be considered a second indicator for symbolic CSR.

A second set of variables was chosen as indicators of ‘substantive’ CSR, that is, policies which require real changes at the operational level and commitment of firm resources:

EMISS_RED: Does the company set specific objectives to be achieved on emission reduction? This can be seen as a substantive CSR policy, as a concrete goal has been defined by the company for reducing its impact on the environment, against which progress can be measured. The achievement of this goal typically requires substantial investment of company resources, for example, in more environmentally friendly machinery.

CSR_REP: Does the company publish a separate CSR/H&S/Sustainability report or publish a section in its annual report on CSR/H&S/Sustainability? Although the publication of a CSR report or section in the annual report is not in and of itself a guarantor of substantive CSR, the absence of such a report is an indicator that the company is not willing to publish information on its impact on the environment and society. Setting up systems to monitor and report on CSR impacts, such as environmental emissions and improvements in health and safety, requires an investment of the firm’s resources.

EMP_SEC: Does the company have a job security policy? The presence of a no-layoff policy can be seen as a major voluntary commitment to the workforce, as it is not required by law in Germany. Employment protection legislation requires companies only to negotiate a ‘social plan’ with the works council when mass layoffs occur. A ‘no-layoff’ policy is therefore a substantive CSR policy which potentially could mean the expenditure of considerable company resources.

We hypothesize that workers as stakeholders have no direct interest in GLOB_COMP and PUB_COMM. We thus expect to see no relationship between codetermination strength and the presence of symbolic CSR at the firm level (Hypothesis 1a). In contrast, workers have an interest in promoting the last three policies listed above, as they are substantive and result in direct improvements for the workers. In the first case, since pollution can directly affect working conditions, workers have an interest in its reduction (EMISS_RED). Second, workers have an interest in company reporting on environmental and societal impacts, as an instrument for improving their access to information and for measuring progress (CSR_REP). Third, workers have an interest in company adoption of a no-layoff policy (EMP_SEC). Therefore, we expect that the strength of codetermination has a positive relationship with the presence of these three variables (Hypothesis 1b).

In order to test for a relationship between the strength of codetermination MB-ix and the five CSR policies, we conducted a multivariate logistic regression analysis (see appendix for details). To address the problem of the direction of causality, the MB-ix and ownership variables were lagged by one year. A number of control variables that could also be expected to influence the presence of these practices were also included:

Size: The size of the company (measured by the logarithm of sales), as larger companies generally have higher visibility and thus experience greater pressure to implement CSR policies. Matten and Moon (2008: 417) state that ‘explicit’ (i.e. voluntary) CSR in Europe is restricted for the most part to large firms. We would thus expect that size is positively related to CSR policies (Hypothesis 2).

Owner: The dispersion of ownership (measured by the percentage of shares held by small shareholders). The literature has conflicting expectations of the role of shareholders: some argue that large shareholders have a longer-term interest in the future of the company than smaller shareholders, since they cannot exit (sell off their shares) as easily and possess the resources to influence company policies; others argue that pressure for CSR mainly comes from smaller investors, such as ethically oriented mutual funds or government pension funds. Thus, we have conflicting expectations with regards to ownership, on the one hand predicting a positive relationship between ownership dispersion and CSR policies (Hypothesis 3a) and on the other hand predicting a negative relationship (Hypothesis 3b).

Profit: The operating margin (difference between sales and costs, expressed as a percentage of total sales). As symbolic CSR has no costs, we would not expect a significant relationship between profitability and the first two indicators (Hypothesis 4a). However, as substantive CSR has real costs, we would expect a positive relationship between profit (as an indicator of the firm’s ability to finance CSR investments) and the three substantive CSR policies (Hypothesis 4b).

Construction, Manufacturing, Transport + Utilities, Trade, Finance and Services: Variables describing the main industry in which a company operates (measured by the Standard Industrial Classification code at the 1-digit level). The industry of operation defines important parameters for CSR, such as exposure to customers and the types of environmental and employment concerns the company faces (Knoppe, 2015). One hypothesis is that industries with the most direct contact with consumers (such as retail trade and services) should have the greatest interest in introducing CSR policies, particularly of the symbolic type. Therefore, there should be a positive relationship between CSR policies and the industry variables Trade and Services (Hypothesis 5a). Further, as transport and public utilities generate the most pollution and experience the greatest pressure for emissions reduction, we would expect a positive relationship between CSR policies and Transport and Utilities (Hypothesis 5b).

Time trend: As some CSR practices have become more prevalent in the recent past, we have included dummy variables for each year between 2007 and 2014 to identify any trends. The general expectation is that the trend should be positive (Hypothesis 6).

These hypotheses are summarized in Table 3.

Table 3.

Hypotheses.

1a	Codetermination is not related to symbolic CSR
1b	Codetermination is positively related to substantive CSR
2	Size is positively related to all CSR policies
3a	Ownership dispersal is positively related to CSR policies
3b	Ownership dispersal is negatively related to CSR policies
4a	Profit is not related to symbolic CSR
4b	Profit is positively related to substantive CSR
5a	CSR policies are positively related to Trade and Services
5b	CSR policies are positively related to Transport + Utilities
6	Time trends for CSR policies should be positive

Results

Table 4 reports the results of the full models that we ran on the five CSR policies. Each model measures the impact of different variables on the probability that the firm has a specific type of CSR practice. The first two models are for symbolic CSR indicators, while the others are for substantive policies.

Table 4.

Determinants of CSR Policies in German firms, 2007–2014.

Independent Variable	Dependent Variable
	GLOB_COMP	PUB_COMM	EMISS_RED	CSR_REP	EMP_SEC
	(1)	(2)	(3)	(4)	(5)
MB-ix			++	++	++
Owner
Size	+++	+++	++	+++	+++
Profit					+++
Construction
Manufacturing
Transport + Utilities			–
Trade	–	-		–	+++
Finance
Services		–
Time trend	increasing	increasing	increasing	increasing	no trend
Observations	595	595	595	595	595
Clusters	96	96	96	96	96

+++ = positive and significant at .01 level, ++ = positive and significant at .05 level, + = positive and significant at .1 level, – = negative and significant at .05 level, - = negative and significant at .1 level.

The first result is that the strength of codetermination (MB-ix) is not significant in the first two models, but is positively significant in models 3–5. In other words, codetermination is positively related to substantive but not to symbolic CSR, thus confirming Hypotheses 1a and 1b. It seems that worker representatives use their voice to promote substantive CSR policies in which workers have an interest, including targets for reduction in emissions, CSR reporting and employment security, but appear to have little interest in promoting symbolic CSR policies.

The second result is that firm size has a significant positive relationship with all five CSR policies. This confirms Hypothesis 2, based on the observation by Matten and Moon (2008) that explicit CSR seems to be concentrated in large firms, which face more public scrutiny and pressure than smaller firms. This is the case for both symbolic and substantive CSR; large firms respond to higher visibility by implementing both types.

The third result is that ownership concentration is not a determinant of any of the five CSR policies; hence, neither Hypothesis 3a nor Hypothesis 3b is supported. In other words, there does not seem to be a systematic difference between large and small shareholders regarding pressure exerted on companies for CSR. However, it may be that this measure is not capturing important differences in investors’ orientation to CSR within each shareholder group. A further test of the influence of investors would involve coding shareholders based on their CSR orientation and including this in the analysis.

The fourth result is that profitability is not related to the symbolic CSR policies included in models 1 and 2. This is not surprising, as we expected in Hypothesis 4a that profit should not be relevant for CSR actions that involve no commitment of firm resources. In contrast, our expectation with regard to the relationship between profitability and substantive CSR policies is only partially confirmed, as employment security is the only one of the three substantive policies tested which is positively related to profitability. In short, profitability does not appear to influence symbolic CSR and also influences only some forms of substantive CSR. Thus, Hypothesis 4a is confirmed, whereas Hypothesis 4b is only partially confirmed.

The fifth result is that, contrary to our expectations, the trade sector appears to have weaker CSR policies, including both symbolic policies and one substantive policy (CSR reporting). One possible explanation is that companies in this sector may expect to have limited returns from CSR policies, as for the most part they sell products of other companies and have no influence on the reputation of those products. Furthermore, they may be reluctant to publish CSR reports, given their lack of control over the supply chains. However, firms in trade are more likely to provide employment guarantees. One possible explanation is that retail and wholesale firms experience heavy seasonal fluctuations in sales, thus they experience pressure to provide employment guarantees to retain experienced workers over the annual cycle. Firms in the service sector seem to have CSR policies in line with the average, with the exception of less frequent public commitments to CSR. Thus, Hypothesis 5a is not confirmed. Hypothesis 5b is also not confirmed, as firms in the transport and public utilities sectors have CSR policies in line with the average, with the exception of emission reduction targets, which they are less likely to adopt. It may be that these firms find it difficult to set such targets, as many transportation firms are expanding (particularly in the airline industry) and utilities such as electricity generation may have limited capacity to reduce emissions without implementing drastic measures such as shutting down plants. A final result is that, with the exception of no-layoff policies, the prevalence of both symbolic and substantive CSR policies is increasing.

Discussion and conclusion

This article contributes to the literatures on both codetermination and CSR, thereby bridging these two largely separate research traditions. With regard to the first, we expand our understanding of the impact of codetermination beyond financial outcomes. In order to do this, we utilized a new measure of codetermination strength, the Mitbestimmungsindex (MB-ix) which overcomes the restrictive assumptions made underlying conventional measures. With regard to CSR, this is one of the few quantitative studies that includes measures of collective labour representation as a determinant of policies.

The analysis we present shows that there is a differentiated relationship between codetermination and CSR. Symbolic CSR is designed to influence external stakeholders’ perceptions of the firm but does not involve any commitment of resources. Strength of codetermination is not related to the symbolic forms we analyse, membership of the UN Global Compact and public commitments by top managers. By contrast, substantive CSR involves real changes to operations and requires the commitment of organizational resources. Codetermination strength is strongly and positively related to all three of the substantive types of CSR we examine, the adoption of targets for emissions reduction, the publication of a CSR report and commitment to employment security. This suggests that worker representatives are selective with regard to the policies they support: they appear less likely to support symbolic than substantive forms of CSR.

We also shed light on the debate in comparative CSR literature regarding the adoption of CSR policies in coordinated market economies like Germany. All five policies examined are of the ‘explicit’ variety, adopted voluntarily by companies. They are often supposed to be most prevalent in liberal market economies like the USA and the UK where the need for business legitimacy is greatest. Matten and Moon (2008) note that explicit policies appear to be spreading to the coordinated market economies but that they are largely confined to larger companies and are for the most part encouraged by governments and industry associations. Our results suggest that worker representatives are also an important factor in explaining the spread of some types of explicit CSR policies to coordinated market economies.

A promising field for future research would be to supplement the quantitative research presented here with a set of case studies on how worker representatives decide on which CSR policies to support and how they use their influence. This could draw on the qualitative literature cited earlier (Hadwiger et al., 2017; Preuss et al., 2015; Vitols, 2011). Furthermore, the type of CSR policies examined could be expanded to include substantive CSR policies in which it is not obvious that workers would have a direct interest. Finally, as the research here is limited to German companies, it would be informative to examine the relationship between codetermination and CSR in other countries.

Supplemental Material

Scholz_and_Vitols_online_supp_(1) – Supplemental material for Board-level codetermination: A driving force for corporate social responsibility in German companies?

Supplemental material, Scholz_and_Vitols_online_supp_(1) for Board-level codetermination: A driving force for corporate social responsibility in German companies? by Robert Scholz and Sigurt Vitols in European Journal of Industrial Relations

Footnotes

Acknowledgements

We thank participants in a workshop organized by the Berlin Study Group on Corporate Responsibility as well as the editors and reviewers for valuable comments on an earlier version of this article. Vera Fabinyi provided excellent research support.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Our research was supported by the Hans-Böckler-Stiftung under the project grant ‘Mitbestimmung und eine zukunftsweisende Unternehmensführung: Messung von Stärke und Einfluss durch einen Mitbestimmungsindex (MB-ix)’.

Supplemental material

Supplemental material for this article is available online.

Author biographies

Robert Scholz is a postdoctoral researcher at the Wissenschaftszentrum Berlin für Sozialforschung.

Sigurt Vitols is a senior researcher at the Wissenschaftszentrum Berlin für Sozialforschung and an associate researcher at the European Trade Union Institute, Brussels.

References

Aguinis

Glavas

(2012) What we know and don’t know about corporate social responsibility: A review and research agenda. Journal of Management 38(4): 932–968.

Campbell

(2007) Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of Management Review 32(3): 946–967.

Carroll

Brown

(2018) Corporate social responsibility: A review of current concepts, research, and issues. In: Wasieleski

Weber

(eds) Corporate Social Responsibility: Business and Society 360, vol. 2. Pittsburgh, PA: Emerald, pp. 39–69.

Dukes

(2005) The origins of the German system of worker representation. Historical Studies in Industrial Relations 19: 31–62.

Favotto

Kollman

Bernhagen

(2016) Engaging firms: The global organisational field for corporate social responsibility and national varieties of capitalism. Policy and Society 35(1): 13–27.

Giertz

J-P

Scholz

(2018) Strategische Personalarbeit ohne eigenständigen Personalvorstand? WSI Mitteilungen 71(2): 140–149.

Girndt

(2006) Die Betriebsratsgründer: Alles, was Recht ist. Arbeit 11: 40–45.

Gjølberg

(2010) Varieties of Corporate Social Responsibility (CSR): CSR meets The ‘Nordic Model’. Regulation and Governance 4(2): 203–229.

Gold

Kluge

Conchon

(2010) ‘In the Union and on the Board’: Experiences of Board-level Employee Representatives across Europe. Brussels: ETUI.

10.

Hadwiger

Hamm

Vitols

Wilke

(2017) Menschenrechte im Unternehmen durchsetzen: Internationale Arbeitnehmerrechte: Die UN-Leitprinzipien als Hebel für Betriebsräte und Gewerkschaften. Bielefeld: Transcript.

11.

Hall

Soskice

(2001) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press.

12.

Harvey

Hodder

Brammer

(2017) Trade union participation in CSR deliberation: An evaluation. Industrial Relations Journal 48(1): 42–55.

13.

Höpner

Müllenborn

(2010) Mitbestimmung im Unternehmensvergleich: Ein Konzept zur Messung des Einflusspotentials der Arbeitnehmervertreter im mitbestimmten Aufsichtsrat. Industrielle Beziehungen 17(1): 7–29.

14.

Jackson

Apostolakou

(2010) Corporate social responsibility in Western Europe: CSR as an institutional mirror or a substitute? Journal of Business Ethics 94(3): 371–394.

15.

Jackson

Bartosch

(2016) Corporate Responsibility in Different Varieties of Capitalism: Exploring the Role of National Institutions. Gütersloh: Bertelsmann Stiftung.

16.

Jirjahn

(2010) Ökonomische Wirkungen der Mitbestimmung in Deutschland: Ein Update (Arbeitspapier 186). Düsseldorf: Hans-Böckler-Stiftung.

17.

Jürgens

Lippert

Gaeth

(eds) (2008) Information, Kommunikation und Wissen im Mitbestimmungssystem: Eine Umfrage unter Arbeitnehmervertretern im Aufsichtsrat. Baden-Baden: Nomos.

18.

Kinderman

(2012) Free us up so we can be responsible: The co-evolution of corporate social responsibility and neo-liberalism in the UK, 1977–2010. Socio Economic Review 10(1): 29–57.

19.

Kinderman

Lutter

(2018) Explaining the Growth of CSR within OECD Countries: The Role of Institutional Legitimacy in Resolving the Institutional Mirror Vs. Substitute Debate (Discussion Paper 18/2). Cologne: MPifG.

20.

Knoppe

(2015) CSR im Retail Management: Wertschöpfung oder Marketing-Gag? In: Knoppe

(ed.) CSR und Retail Management: Gesellschaftliche Verantwortung als zukünftiger Erfolgsfaktor im Handel. Heidelberg: Springer, pp. 9–18.

21.

Köstler

Müller

Sick

(2013) Aufsichtsratspraxis: Handbuch für die Arbeitnehmervertreter im Aufsichtsrat. Frankfurt: Bund.

22.

Matten

Moon

(2008) ‘Implicit’ and ‘Explicit’ CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review 33(2): 404–424.

23.

Preuss

Gold

Rees

(2015) Corporate Social Responsibility and Trade Unions: Perspectives across Europe. London: Routledge.

24.

Raabe

(2010) Die Mitbestimmung im Aufsichtsrat: Theorie und Wirklichkeit in deutschen Aktiengesellschaften. Berlin: Erich Schmidt.

25.

Rosenbohm

Haipeter

(2019) German board-level employee representation in multinational companies: Patterns of transnational articulation. European Journal of Industrial Relations 25(3): 217–230.

26.

Scholz

Vitols

(2016) Der Mitbestimmungsindex MB-ix: Wirkungen der Mitbestimmung für die Corporate Governance nachhaltiger Unternehmen (Report 22). Düsseldorf: Hans-Böckler-Stiftung.

27.

Scholz

Vitols

(2018) Der MB-ix in börsennotierten Unternehmen: Verankerung der Mitbestimmung im letzten Jahrzehnt. Düsseldorf: Hans-Böckler-Stiftung Mitbestimmungsreport 43.

28.

Schons

Steinmeier

(2016) Walk the talk? How symbolic and substantive CSR actions affect firm performance depending on stakeholder proximity. Corporate Social Responsibility and Environmental Management 23: 358–372.

29.

Sethi

Schepers

(2014) United Nations global compact: The promise-performance gap. Journal of Business Ethics 122: 193–208.

30.

Streeck

(1984) Codetermination, the fourth decade. In: Wilpert

Sorge

(eds) International Perspectives on Organizational Democracy: International Yearbook of Organizational Democracy, vol. 2. Chichester: John Wiley & Sons, pp. 391–422.

31.

Vitols

(2011) Nachhaltigkeit Unternehmensverantwortung Mitbestimmung: ein Literaturbericht zur Debatte über CSR. Berlin: Edition Sigma.

32.

Waddington

(2010) European Works Councils: A Transnational Industrial Relations Institution in the Making. London: Routledge.

33.

Weiß

(2016) Arbeitnehmermitbestimmung und die Europäische Aktiengesellschaft (SE): Eine empirische Analyse auf Länderebene. German Journal of Human Resource Management 30(2): 84–107.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.46 MB