IWQW. Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung. Diskussionspapier Discussion Papers. No. 01/ PDF

IWQW Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung Diskussionspapier Discussion Papers No. 01/2009 Worker Directors: A German Product that Didn t Export? John T. Addison University

Please download to get full document.

View again

of 21
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 23 | Pages: 21

Extension: PDF | Download: 0

IWQW Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung Diskussionspapier Discussion Papers No. 01/2009 Worker Directors: A German Product that Didn t Export? John T. Addison University of South Carolina, Columbia Claus Schnabel University of Nuremberg ISSN Friedrich-Alexander-Universität IWQW Institut für Wirtschaftspolitik und Quantitative Wirtschaftsforschung Worker Directors: A German Product that Didn t Export? John T. Addison and Claus Schnabel Abstract: Despite its lack of attractiveness to other countries, the German system of quasiparity codetermination at company level has held up remarkably well. We recount the theoretical arguments for and against codetermination and survey the empirical evidence on the effects of the institution, tracing the three phases of a still sparse literature. Recent findings hold out the prospect that good corporate governance might include employee representation by virtue of the monitoring function and the reduction in agency costs, while yet cautioning that the optimal level of representation is likely below parity. And although the German system may be better than its reputation among foreigners, it might have to adapt to globalization and the availability of alternative forms of corporate governance in the EU. Zusammenfassung: Trotz seiner geringen Attraktivität für andere Länder hat sich das deutsche System der quasi-paritätischen Unternehmensmitbestimmung als bemerkenswert stabil erwiesen. Wir erörtern die theoretischen Argumente für und gegen Mitbestimmung und bieten einen Überblick über die empirische Evidenz zu den Auswirkungen dieser Institution, wobei wir drei Phasen einer eher spärlichen Literatur nachzeichnen. Jüngere Erkenntnisse deuten darauf hin, dass zu einer guten Corporate Governance auch die Beteiligung der Arbeitnehmer (aufgrund ihrer Überwachungsfunktion und der Verringerung von Agency-Kosten) gehören könnte, wobei jedoch das optimale Ausmaß der Mitbestimmung unter 50 Prozent liegen dürfte. Auch wenn das deutsche System besser sein mag als sein Ruf im Ausland, muss es sich wohl an die Globalisierung und die Verfügbarkeit alternativer Unternehmensformen in der EU anpassen. Key words: codetermination, worker directors, board-level employee representation, firm performance, Germany JEL classification: J50 1 Europeanization is likely to encourage new representative structures and processes and to create some opportunities for changed outcomes which may not however be grasped (Hyman, 1997, p. 322). 1. Motivation Germany is the world s biggest exporter of goods. One of the few products made in Germany that has not been exported successfully is the German system of codetermination at company level (Unternehmensmitbestimmung) with representatives of employees sitting on company supervisory boards (sometimes called worker directors ). In contrast to employee representation via works councils at establishment level (betriebliche Mitbestimmung), which is found in many European countries in various forms and which has also played a role as a template in the formulation of legislation on European Works Councils (94/45/EC) in the EU, 1 Germany has not been able to convince its neighbours or the EU to adopt its system of (quasi) parity board-level representation. In short, although there do exist systems of board-level employee representation in most EU member states, these are usually not as comprehensive as the German system (for a comparative analysis, see Carley, 1998; Schulten and Zagelmeyer, 1998). What is more, competition has arisen among the various European systems of codetermination since the European Company Statute (Council Regulation 2157/2001 and Council Directive 2001/86/EC) adopted by the EU in 2001 gives companies the option of forming a European Company (Societas Europaea, SE) which may operate on a European-wide basis. Under the legislation, a German business establishing an SE can choose between the current two-tier system of corporate governance in Germany (with its separation of powers between a management board and a supervisory board) and alternative, one-tier systems common in other EU member states (such as the U.K.) where there is a single board 1 Discussions of this form of codetermination and its effects are provided by Addison, Schnabel, and Wagner (2004) and Gumbrell-McCormick and Hyman (2006). The practice of German codetermination at establishment level has also guided a number of other European-level initiatives featuring employee participation or having a participation component such as Community legislation on collective redundancies/mass layoffs (98/59/EC), transfers of undertakings (2001/23/EC), and national systems for informing and consulting employees (2002/14/EC), inter al. 2 of directors. In the latter case, companies would not have to adhere to German codetermination laws (whereas an existing German public limited company converting itself into an SE registered in Germany would have to stick to its current form of codetermination). Further, in the case of SEs formed via cross-border mergers, or the creation of a joint holding company or subsidiary, a fall-back solution in the law stipulates that the most extensive form of codetermination should apply to the merged company. 2 This too might encourage companies to locate or relocate their new headquarters outside Germany. 3 Despite the German system s lack of attractiveness to other countries, codetermination at company level has held up remarkably well inside Germany. According to Hans Böckler Stiftung, a union-sponsored foundation monitoring codetermination, as of 2006 some two years after member states had to implement the Regulation/Directive (Germany, on this occasion, being two months late in complying) 721 companies were still covered by the German Codetermination Act of This number is slightly down on the maximum of 767 in Although some German companies close to the employment threshold for introduction of (quasi) parity-based codetermination have set up SEs with a single board of directors not including employee representatives, none of the large public limited companies in Germany that have turned themselves into SEs (e.g. Porsche, BASF and Allianz) has deviated from (quasi) parity representation of shareholders and employee representatives. 4 This raises the question of why parity codetermination at company level has survived in Germany (and whether it will be able to survive in the future). Have 2 Note that these are just two examples of the directive s potential impact on codetermination. In the case of SEs formed through mergers (or via the formation of a holding company or subsidiary), it is also possible for an agreement between the special negotiating body and central management to result in a lesser degree of board-level participation than the highest proportion that applies within the participating companies. All that is required here are the votes of two-thirds of the SNB members representing at least two-thirds of the total workforce. This option is not available in the case of a company conversion. 3 Such fears are not only voiced by trade unionists in Germany. For the U.K., Hyman (1997, p. 310) notes that regulation at the level of the European Union (EU), while in general modest, may encourage the implementation of alien representative mechanisms in Britain. 4 For details and examples, see the foundation s webpage (http://www.boeckler.de) as well as the recent analysis by Keller and Werner (2008). Somewhat in contrast, Stettes (2006) reports that in 2005 every seventh newly-established private limited company in Germany was registered according to the legal form of the U.K., thereby avoiding German codetermination laws. 3 companies learned to live with worker directors as they apparently have with works councils (see Kotthoff, 1994) and just fear the hassle of switching to SEs? Or is the system simply better than its reputation among foreigners? Since the first of these questions is difficult to investigate empirically (because surveying managers is likely to result in answers that are deemed to be politically correct), this paper will focus on the second question and survey the empirical evidence on the economic effects of employee representation on company supervisory boards (Unternehmensmitbestimmung) in Germany. 5 Our treatment proceeds as follows. We first sketch the institutional framework of codetermination at company level in Germany before recounting the theoretical arguments for and against. Next, we survey the empirical evidence on the effects of the institution. Finally, in our concluding remarks, we draw together the threads of the preceding arguments and offer a brief perspective. 2. Institutional Framework In the German two-tier system of corporate governance, the supervisory board has basically four functions (according to the 1965 Stock Corporation Act, Aktiengesetz). It approves the appointment of management board members; it monitors the management board (which has to inform it of the broad lines of business policy and corporate planning on an annual basis and of business operations on a more regular basis); it can codetermine business operations requiring its approval; and it scrutinizes the annual accounts of the company or group. Various laws and their amendments stipulate that differing shares of seats on the supervisory board be allocated to employee representatives, so that there exist three different regimes of codetermination at company level in Germany: full-parity codetermination for the coal and steel industries under the 1951 Codetermination Act, almost-equal or quasi-parity representation under the 1976 Codetermination Act for corporations having more than 2,000 employees (where the chairman of the board, elected by the shareholders, has the casting vote in case of a tie), 5 For a survey of the earlier theoretical and empirical literature, see Junkes and Sadowski (1999). 4 one-third representation in companies with between 500 and 2,000 employees under the 1952 Works Constitution Act. The 1951 Act on the Codetermination of Employees in the Supervisory and Management Boards of Companies in the Coal, Iron and Steel Industry (or Montan- Mitbestimmungsgesetz, as it is also known) established supervisory boards ranging in size from 11 to 21 members, according to share capital, comprising equal numbers of shareholder and employee members and one neutral member. Further, the appointment of a Labour Director (who serves on the management board) requires the agreement of the employee representatives. In 1976 under the Codetermination Act (Mitbestimmungsgesetz), equal but not full-parity representation (hence quasi-parity representation) was extended from coal, iron and steel to corporations of all other industries where there are as a rule more than 2,000 employees. The number of seats on the supervisory board is a function of employment: 12 members if the employment total does not exceed 10,000, 16 if it exceeds 10,000 but is less than 20,000, and 20 where it is greater than 20,000. Election of the chairman and vice-chairman of the supervisory board in each case requires majorities of two-thirds of the votes. If neither gains the necessary votes, the shareholder (employee) representatives elect the chairman (vicechairman). This procedure ensures that the chairman is always a shareholder representative and he/she has an extra, tie-breaking vote (unlike the situation in the coal, iron and steel industries). The law also made provision for the inclusion of managerial employees, who were given one seat on the supervisory board. The 1952 Works Constitution Act (Betriebsverfassungsgesetz) introduced a weaker form of codetermination by providing for one-third representation of employees on the supervisory boards of large and medium sized corporations with more than 500 employees. The sections of the 1952 Works Constitution Act dealing with supervisory board membership in companies with 500 to 2,000 employees were amended in the so-called Third Part Act (Drittelbeteiligungsgesetz) of It should be noted that codetermination legislation has generated fierce and ongoing employer resistance, and companies (as well as unions) have engaged courts at all levels on codetermination issues. For example nine corporations and 29 employers associations challenged the 1976 Act on constitutional grounds, as infringing the property rights of shareholders. The Federal Constitutional 5 To summarize, the proportion of worker representatives on company boards varies from one-third, in companies with between 500 and 2,000 employees, to onehalf, in companies with more than 2,000 employees. In the latter, the chair in effect represents the shareholders and has the casting vote. The exception is the larger coal or iron and steel companies where the chair is independent; hence the expression full-parity representation. The number of members of the supervisory board is determined either by the share capital or employment of the company or group. The election procedure for employee representatives is complicated and varies by type of company and type of codetermination (for details, see Addison, 2009). Empirical analyses of the effects of board representation have either exploited differences between codetermination and no codetermination or between the various types of codetermination. As we shall see, the early literature revealed few effects of board representation while the subsequent financial literature proved more pessimistic. Latterly, with the German national innovation debate, codetermination has received modest support from several innovation studies preceded by a panel study of productivity. Before reviewing these findings, however, we must first briefly discuss some theoretical arguments on (board-level) codetermination. 3. Theoretical remarks In (continental) European countries, codetermination is usually justified by traditional political and social arguments such as the democratization of the employment relationship and by notions of stakeholder value, all of which imply that the interests of all relevant groups should be represented in a company s board. However, even economic reasoning focusing on orthodox notions of corporate governance centred on shareholder value admits of arguments favouring codetermination. 7 The basic orthodox economic starting point is that codetermination may be a safeguard for the employee side against opportunistic behavior on the part of employers. Absent some Court in its decision of March 1, 1979, upheld the constitutionality of the law, arguing that shareholder rights were protected because the supervisory board chairman still had the casting vote, while noting that the private property rights enshrined in the constitution had also to serve public welfare as might obtain from heightened industrial peace and thence improved economic performance. 7 On the two models, see for example Charreaux and Desbrières (2001). 6 form of protection (either institutional or contractual), so the argument runs, employees will be unwilling to undertake reliance investments such as firm-specific skills acquisition. The upshot is that in circumstances where not all coalition-specific resources are owned by one party, codetermination may provide a governance structure that is capable of dealing with maximizing agents with conflicting interests (Furubotn, 1988, p. 168). However, the codetermination structure envisaged in this hypothetical jointinvestment firm where the employees are residual claimants is voluntary. By contrast, under mandatory codetermination major control rights are ceded to employees irrespective of whether or not they have made coalition-specific investments. Further, they are given no income rights in the firm, and normally do not share directly in the residual, and cannot transfer property rights in the job to others, and so on. Politics, so the argument runs, now replace economic responsibility. Employees making decisions do not bear the full cost of their decisions. The situation is to be contrasted with a proper allocation of property rights in the joint investment firm a sharing of control rights via codetermination which assures that those making decisions bear the full cost of their actions. This incentive structure promotes both productivityenhancing incentives as well as relatively lower transaction costs. Yet, as we all know, such voluntary arrangements have not emerged. Why is this? For his part, Furubotn (1988) speculates that this is because employees can gain more from the political solution of mandatory codetermination than through private bargaining with the firm. After all, they get up to one-half of the seats on the supervisory board without any corresponding duty to invest. But the no-show result has been exploited more generally by Jensen and Meckling (1979), who argue that employee board membership must be detrimental to shareholder value because it has not been embraced by employers. Indeed, they would see the force feeding and strenuous opposition of German employers to parity or quasi-parity codetermination as testimony to their indirect argument as to the inefficiency of mandatory codetermination. Another explanation could be that the market is systematically biased against codetermination. The starting point is the argument by Levine and Tyson (1990) to 7 the effect that codetermination will be underprovided by the market on prisoner s dilemma grounds. The maintained hypothesis is that codetermination is valuable to all firms but to sustain it a compressed wage structure and dismissals protection are required. In these circumstances, any single innovating firm will suffer an externality and adverse selection: its stars will be spirited away by traditional firms, who can offer these workers higher rewards by virtue of their supposedly sharply differentiated wage structures, and it will simultaneously attract the work shy who are now protected from dismissal. On both counts, the codetermined firm will not emerge voluntarily and must be mandated. Another line of argument is more compelling because it explicitly recognizes rent seeking on the part of labour. Freeman and Lazear (1995) contend that although codetermination raises the joint surplus it raises the rent going to labour more. Employers duly resist codetermination and it has to be mandated albeit coupled with institutional limits on the ability of the employee side to extract rents. The inference of the Freeman-Lazear model (which, however, is constructed around betriebliche Mitbestimmung via works councils) is that the allocation of control rights to corporate assets may have important implications for economic efficiency but that the absence of the institution outside of a mandate is not necessarily decisive. Thus far we have assumed an identity of interest between management and shareholders. What if managers are imperfect agents of the shareholder principal? One of the few analyses to exploit such agency considerations is Jirjahn s (2003) treatment of executive incentives and firm performance. Jirjahn s treatment has a basis in two key associations: first, the relationship between codetermination (in his model it is works council presence rather than worker representation on company boards) and self-enforcing contracts; and, second, the relationship between agency problems and trustful employee relations. An agency problem may have a commitment value in making self-enforcing contracts feasible. But the introduction of profit sharing for managers may give them the incentive to break implicit contracts with the employees on behalf of profit-maximizing owners with adverse consequences for trust. Where codetermination and self-enforcing contracts are substitutes (i.e. the reputation effects mechanism is strong), the impact of codetermination on firm 8 performance will be stronger in firms with les
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks