The Influence of Employee Representatives on
Executive Compensation in Germany
Master Thesis
Msc in International Business and Management
Andre Kuczynski
Student Number: s1610616
S1610616@student.rug.nl
First Supervisor: Dr. Niels Hermes
Second Supervisor: Dr. Huib Stek
Groningen, 10
thof August 2007
Faculty of Management and Organization
University of Groningen, NL
Abstract
INTRODUCTION
Due to rising levels of managerial pay since the early 1990’s executive compensation increasingly attracted attention by academics, politician and in the media (Conyon, 2006). In an internationalising world national boundaries are diminishing and products, innovation and business strategies are spreading across boarders. As a result the demand of skilled executives that are capable of effective, cross-boarder management is increasing (Sanders and Carpenter, 1998; Conyon, 2006).
The literature on executive remuneration is dominated by research conducted in an Anglo-Saxon business environment (see e.g., Sanders and Carpenter, 1998; Bertrand and Mullainathan, 2001; Bebchuck and Fried, 2003; Core et al., 2003; Conyon, 2006). The academic research concordantly assumes that: Firstly, the conflict between managers and shareholders, the Principal-Agent Conflict, is predominant in a company (Aguilera and Jackson, 2003). Secondly, the superior task of the company’s management is to enhance the owner’s wealth (Friedman, 1962). Therefore in executive compensation contracts the managers are motivated by incentives that are linking their personal wealth to the increase of shareholder wealth (Jensen and Meckling, 1976). These assumptions hold for Anglo-Saxon countries with corporate governance mechanisms that are focussed on shareholders as the primary stakeholders of a company (see e.g., Fama and Jensen, 1983).
However, in the more stakeholder-oriented societies of East Asia and Continental Europe, the conflict between Principals and Agents is considered as one conflict among others. It is criticized, that other stakeholder groups, the interaction between different stakeholder groups and shareholders, and the specific institutional setting of a country are disregarded by solely concentrating on the relationship between shareholders and the hired managers (Aguilera and Jackson, 2003). Furthermore, based on Freeman (1984) it is argued that managerial contracts that are giving incentives to executives to increase shareholder wealth facilitate managerial actions at the expenses of other stakeholders (Arora and Alam, 2005; Coombs and Gilley, 2005).
participation of labour in company decisions is seen as the most distinct characteristic of the German corporate governance mechanism (DuPlessis, 2004). Jensen and Meckling (1979) emphasize that codetermination leads to a redistribution of profits from shareholders to employees. Gorton and Schmid (2004) show that German companies with equal codetermination are traded at a discount at the stock exchange compared to smaller companies with less powerful ERs. Whereas the negative impact of codetermination on shareholder wealth is already established (see e.g., Schmid and Seger, 1998; Gorton and Schmid; 2004), the impact of labour participation on executive compensation is under researched (Chinema and Buck, 2006). In this paper ERs are recognized as important actors in the executive payment setting process. The following research question is formulated.
Are ERs a determinant of executive compensation in Germany, and if so what is the influence of ERs on executive compensation?
Here it is argued that executive compensation as a characteristic of a country’s corporate governance mechanism can not solely be explained by the Principal-Agent Theory, since it neglects the power and interests of other stakeholders, namely employees (Aguilera and Jackson, 2003). Although in a variety of research the power of German labour is illustrated in a descriptive way, no paper empirically assesses the impact of labour on managerial compensation. Empirical research on Germany is focussed on traditional Anglo-Saxon determinants of executive compensation, namely ownership dispersion and managerial performance (see e.g., Elston and Goldberg, 2003; Kaserer and Wagner, 2004). Here ERs are included in an empirical analysis on executive compensation in Germany for the first time.
This paper is structured as follows. At the beginning the literature on executive compensation in Anglo-Saxon economies is reviewed. Hereby attention is attracted to the composition of the compensation committee in the United States of America (US). Hereinafter the German corporate governance system is described and the effects of the country’s governance arrangements on the composition of the SB are outlined. Codetermination as source of labour power is explained. This part is concluded by showing the differences in the structural composition of the executive compensation setting bodies between the US and Germany.
In the next part the focus of executive compensation research on Germany is reviewed. It is highlighted that labour power is described theoretically in a variety of research on executive compensation but not included in empirical analyses. Measurement problems are identified as potential source of disregarding labour power. Subsequently ways are proposed how to incorporate ERs as a determining factor of managerial compensation in an empirical analysis. The line of argumentation results in a new approach that includes ERs when determining executive compensation in the stakeholder oriented German society. This section is concluded by graphically illustrating the line of argumentation.
Hereinafter the newly introduced approach on determining executive pay is tested. First of all, based on existing literature the hypotheses are developed. Attention is attracted to interactions and interdependencies between ERs, SRs, and managers.
In the following part the hypotheses are tested. The data collection process, the variables, and the measures are presented in the methodology. Following, collinearity tests are run and the data are described. Subsequently the results are displayed and explained.
Finally, the paper is concluded by discussing the outcomes of the empirical analyses. Implications for theory and praxis are given and the limitations of the study are summarized.
The line of reasoning in the theoretical part is instantly tested empirically. Thus based on the results of the empirical analyses the hypothesized impact of ERs on executive compensation is substantiated or invalidated.
LITERATURE REVIEW
Executive compensation and corporate governance in Anglo-Saxon economies
This view, later called the optimal contracting approach, is based on the fundamental assumption that the shareholders are setting the pay for the hired managers. Based on Jensen and Meckling (1976) elements of performance related executive pay are seen as the solution of the Principal-Agent Conflict. Stock and stock options, as performance related compensation components, are linking managerial performance to the creation of shareholder value. Hence, a financial incentive is given to the managers to act in the shareholder’s best interest, since an increase in shareholder wealth leads to a rise in managerial pay (Jensen and Murphy, 1990). The optimal contracting approach describes the rising levels of executive pay since the 1980’s (see e.g., Abowd and Bognanno, 1995) by using market-based explanations. Proponents of the optimal contracting approach cite an increasing scarceness of skilled managers who are capable to deal with more complex product and corporate strategies in an internationalising business environment as the driver of accelerating remuneration levels (e.g., Sanders and Carpender, 1998).
Furthermore, the principals link the performance of managers closer to the creation of shareholder wealth. Conyon (2006) shows that the rise in managerial pay in the US is a result of the extensive use of performance related pay, mainly stock options.
Finally Murphy and Zabojnik (2004) prove that companies are increasingly incapable to fill job vacancies at the executive boards with members of their existing workforce due to a lack of qualified managers inside the company. As a consequence firms tend to hire highly skilled external executives who need to be attracted with a wage premium compared to inside managers (Murphy and Zabojnik, 2004).
non-executive directors, since they are closely affiliated with each other (Core et al., 1999) and the non-executives might fell committed and crateful to the CEO (Bebchuk et al., 2002; Bebchuck and Fried, 2003).
Secondly, interlocking membership of directorates are increasing the interrelations, the interdependencies, and personal favouring between the actors involved in the executive wage bargaining process (Bebchuk et al., 2002; Bebchuck and Fried, 2003). Furthermore, outside directors who are serving on boards of more than three companies are less effective in monitoring the CEOs’ as a result of work overload (Core et al., 1999).
Thirdly, mandating compensation consultants with provisions based on the range of consultant services they provide facilitates high costs of compensation contracts (Bebchuk et al., 2002; Bebchuck and Fried, 2003).
Fourthly, a lack of disclosure of important compensation elements that are granted to managers without any performance relation are interpreted as further indicators of managerial power particularly goodbye payments and pension benefits (Bertrand and Mullainathan, 2001; Bebchuk et al., 2002).
Finally poorly designed, non-indexed stock option contracts that are not suited to efficiently link executive pay to performance are seen as an outcome of managerial power, since these contracts are resulting in a low pay for performance sensitivity. Bertrand and Mullainathan (2001) show that CEOs are rewarded for luck as a of result contracts that fail to filter out windfall profits. They show that especially in companies without a major shareholder the managers are capturing the payment-setting process and generate high levels of pay without performance relation. In contrast, executive managers in companies with a major shareholder are less likely to receive windfall profits. Similar Cyert and colleagues (2000) prove that a shareholder with a high stake in the company is reducing the level of executive compensation. Bertrand and Mullainathan (2001) conclude that managers of a company with a major shareholder are more effectively governed. Thus a major shareholder reduces the power of managers and therefore the ability of executives to negotiate suboptimal payment contracts including features such as incentives without performance, option plans that allow for windfall profits, gratuitous goodbye payments and intransparent payment structures (Bertrand and Mullainathan, 2001; Bebchuk et al., 2002; Bebchuck and Fried, 2003).
only capable of capturing the payment setting process if they are able to influence the independence of the compensation committee. Compensation committees are seen as independent if the supervisors are working in the best interests of the principals. Conyon (2006: 37) points out that “Whether the supervisor will work in the principal’s best interest, or instead collude with the agent, is dependent on whether the supervisor’s interests are more tightly related with those of shareholders (principal) or management (agents)”.
Conyon and He (2004) show that effective supervision of executives (expressed by the use of performance related payment elements and lower levels of pay) is most likely in companies with shareholders at the compensation committee. According to Conyon and He (2004) shareholders have less incentive to align with the managers they supervise, because collusion with managers would reduce shareholders’ own wealth as a result of higher agency costs. Conyon (2006) finds that the participation of shareholders at the compensation committee increased from 1998 to 2003, and describes the supervisors as independent from managerial influence. In Figure 1 it is illustrated that in 2003 around 92% of the members of the compensation committee in the US are shareholders with an equity stake in the company. The remaining members are affiliated directors with no shareholdings in the company (7,7%), and ERs (0,4%). Hence shareholders are dominating the compensation committee in the US and compensation committees are seen as independent (Conyon, 2006), since shareholders have less incentive to collude with the agents (Conyon and He, 2004). Conyon (2006) concludes that market-based explanations are the driver behind the rise in executive compensation since the US-American corporate governance mechanisms are working effectively.
However, the academic discussion seems to be continued in order to determine who is setting the pay for executives, the managers themselves or the shareholders, who are represented by the compensation committee. The common ground of both approaches is that the shareholders are seen as the primary stakeholder of a company, and that the hired managers are solely obliged to increase the shareholders’ wealth as described by Friedman in 1962.
primary stakeholders of a company. Hence stakeholder theory requires from executive managers the balancing of the interests and constituencies of various interest groups in the firm (Freeman, 1984).
In research on US-American companies Arora and Alam (2005) evidence that the interests of shareholders are predominant over the interests of customers, suppliers and employees. McGuire and colleagues (2003) show that a management style disregarding the responsibility of a company for its employees, the local community and environmental protection does lead to higher CEO compensation in terms of fixed salary and long-term incentives. The authors conclude that there is no incentive for the companys’ executives to adopt a stakeholder-oriented corporate strategy. Similar Coombs and Gilley (2005) find that stakeholder-related management activities in the US have a negative effect on managerial compensation. Thus managers jeopardize their personal wealth when putting stakeholder management on the agenda (Coombs and Gilley, 2005). The findings indicate that in Anglo-Saxon economies the compensation committee gives a strong priority to shareholders’ interests at the expenses of the other stakeholder groups in the firm. This is not surprising, since shareholders are dominating the compensation committee.
Summarized, the discussion of executive compensation in Anglo-Saxon economies in a Principal-Agent Model reflects the corporate governance arrangements of these economies. The dominance of shareholders in the corporate governance landscape (Fama and Jensen, 1983) is substantiated by the composition of the compensation committees comprising to more than 90% of shareholders with a direct stake in the company (Conyon, 2006). Thus in Anglo-Saxon economies the main conflict of interest when setting executive pay is between managers and shareholders. Other stakeholders do not have significant power to interfere the executive compensation setting process (McGuire et al., 2003; Arora and Alam, 2005; Coombs and Gilley, 2005).
However, Aguilera and Jackson (2003) emphasize that the corporate governance arrangements differ across countries and point out that management, capital and labour are important stakeholder groups in a company. Furthermore, they argue that corporate governance mechanisms explained by a pure the Principal-Agent Theory are of limited explanatory power to describe national variations in the corporate governance arrangement in East Asia and some Continental European countries.
industry, and within industrial sectors create a cooperative industry-structure (see e.g., Dore, 2000; Elston and Goldberg, 2003). In respect to managerial pay two questions are posed: Firstly, how is executive compensation determined in the stakeholder-oriented German society and secondly, to what extent out of a Principal-Agent Conflict a more complex Stakeholder-Agent relationship emerges.
Following the German corporate governance landscape is described in order to answer these questions. Stakeholder groups that have substantial power in the executive payment setting process are presented. A focus is laid on ERs, since the strong involvement of ERs in company decisions is unique (Niedenhoff, 2005).
German Corporate Governance Landscape
The German corporate governance landscape is characterized by a three-tier agency model for public traded companies (Niedenhoff, 2005). The shareholders (principals) as the owner of the company delegate their supervising authority to the Supervisory Board (SB) that controls and monitors the Management Board (MB). The MB is responsible for the strategic decision-making and the day-to-day business of the enterprise (Gorton and Schmid, 2004; Niedenhoff, 2005). The task of the SB is comparable to non-executive directors in the US, whereas the MB is comparable to executive directors (Elston and Goldberg, 2003). The SB is in charge of the appointment of the executive managers and entrusted with the negotiation of the executive compensation contracts (Nietsch, 2005; Niedenhoff, 2005). The German Corporate Governance Code recommends that if a compensation committee is elected it should comprise of the members of the SB (GCGC, 2002).
In principle the German corporate governance arrangements that are dealing with the executive compensation setting process are similar to the US, where the principals delegate power to a supervisor (compensation committee). The supervisors’ task is to monitor and evaluate the work of the executive managers (Conyon, 2006). However, the fundamental difference between the German and US-American corporate governance systems is the composition of the executive payment setting bodies.
Firstly, the German Codetermination Act of 1976 regulates the statutory participation of ERs at the SB. ERs are representing of the interests of the company’s workforce at the SB (Niedenhoff, 2005). German law requires that in public traded companies with less than 2000 employees one-third of the SB members are representing labour interests. In enterprises with more than 2000 employees half of the SB members are ERs (Sandrock, 2005). This paper focuses on the largest 80 German public traded enterprises that are employing on average more than 2000 people. Thus the further argumentation in this paper is based on equally codetermined firms.
German Codetermination law requires that around one-fourth of the ERs consist of members delegated from the on industry-level operating labour unions. The remaining seats are allocated to the firm’s employees that are organised in work councils (Niedenhoff, 2005). In praxis around 30% of the ERs are recruited from the labour unions and the remaining 70% are employees of the company (see Figure 1; Gerum, Steinmann, and Fees, 1988 in Gorton and Schmid, 2004).
The equal composition of the SB gives substantial power to the company’s workforce represented by ERs. Thus in Germany shareholders are legally obliged to share monitoring authority with ERs, whereas in the US shareholders are free to elect their representatives at the compensation committee (Dore, 2000). Although the by the shareholders elected chairman of the SB has a tie-breaking vote (his votes counts twice) in case the SB cannot agree on a majority decision (Sandrock, 2005), the power of ERs is still substantial. Gorton and Schmid (2004) report that the vast majority of decisions in codetermined SBs are made by consensus and the chairmen of the SBs do rarely use their tie-breaking vote.
Furthermore, all members of the MB including its chairman (CEO) are elected with a qualified majority of two-third of the votes of all members of the SB (Niedenhoff, 2005). Thus in order to be able to appoint executive managers with the by law required qualified majority the votes of SRs and at least some ERs are needed or vice versa. Hence the nomination of a CEO in an equally codetermined German firm is in any case a consensus between SRs and ERs.
The remaining seats in the SB are assigned to SRs. The composition of ERs and the equal allocation of seats between ERs and SRs are defined by German law and therefore are uniform among large corporations. In contrast, the composition of SRs can differ between companies since the shareholders are electing their representatives at the annual general meeting upon the suggestion of the existing SRs at the SB for an indefinite time (Niedenhoff, 2005).
(Conyon and Schwalbach, 2000; Elston and Goldberg, 2003). In 1992 around 10% of the SRs were managers of large German financial institutions (Conyon and Schwalbach, 1999). Banks and insurance companies tend to hold shares on large German corporations.
Furthermore, banks are acting as custodians for minority shareholders by exercising their clients’ proxy voting rights. German Company law facilitates the transferral of voting rights from minority shareholders to financial institutions. The procedure is as follows. Minority shareholders transfer their voting rights to the bank they keep an account with and legitimate this bank to exercise their voting rights at the annual general meeting (Elston and Goldberg, 2003). The banks’ clients are allowed to give a distinct order to their bank and therewith instruct the bank how to vote at the annual general meeting. But in fact the clients tend to authorize the bank to exercise the transferred voting right freely without any order or restriction. Thus a considerable amount of voting rights is attributed to financial institutions without any direction. In consequence banks are able to increase their control rights in excess of their ownership rights by using the proxy voting rights of minority shareholders (Gorton and Schmid, 2004). In 1996 around 36% of the voting rights of the 100 largest German companies are controlled by financial institutions (Elston and Goldberg, 2003).
After all, in the banks-based German economy, financial institutions are an important source of finance by providing loans and share underwriting (Conyon and Schwalbach, 2000). Summarized, banks are simultaneously acting as shareholders, creditors, representatives of minority shareholders, and business partners of a firm. Thus the extent of power and influence of banks goes beyond a traditional debtor-creditor relationship (Du Plessis, 2004).
Thirdly, the ownership of German companies is highly concentrated. Faccio and Lang (2002) prove that Germany has the highest concentration of ownership among 13 European countries. Their results show that 89,63% of the largest public traded companies have a controlling shareholder with a stake of 20% or higher in the company. Furthermore, in Germany cross-holdings in terms of company X holds a stake in company Y and vice verse between large corporations are popular mean of getting control, obtaining information and establishing business relationships (Conyon and Schwalbach, 1999). The degree of cross-holdings between firms in Germany is the highest in Western Europe (Faccio and Lang, 2002).
without actually acquiring an equity stake in the company. In total 18.5% of the SRs are from related businesses that have no equity interests in the firm (Gerum, Steinmann, and Fees, 1988 in Gorton and Schmid, 2004). Furthermore, consultants such as auditors, advisors and lawyers are accounting for 13.5% of the SRs (Gerum, Steinmann, and Fees, 1988 in Gorton and Schmid, 2004). In total around 32% of the SRs have no equity interest in the firm. It is common practice that some German managers serve at the SB of up to ten different companies. This is increasing personal relationships, dependencies and a high work load (DuPlessis, 2004). Critics point out that interlocking and accumulating of SB seats makes it rather difficult for the supervisor to effectively and independently control the company in the best interest of direct shareholders (for Germany: Conyon and Schwalbach, 2000; Du Plessis, 2004; for the US: Bebchuk et al., 2002;).
These four characteristics of the corporate governance mechanism are resulting in a specific composition of the SB as shown in Figure 1. One half of the SB seats are allocated to Unionists (15%), and employees of the company organised in work councils (35%) as representatives of the company’s workforce. The remaining half of the SB seats is allocated to direct shareholders (29%), SRs with no equity interests, namely consultants (6,8%) and managers of related companies (9,3%), and financial institutions (5%) as representatives of shareholders.
Figure 1
annual general meeting in order to elect direct shareholders in the SB. The reasons can be summarized as follows:
Firstly, the SRs at the SB are proposing new candidates, thus shareholders have only an indirect impact on the selection process. According to German Company Law shareholders are only allowed to accept or reject a candidate proposed as supervisor (Niedenhoff, 2005). Secondly, it is common practise that if a representative of one interest group resigns from the SB, the open position is replaced with a member of the same interest groups in order to keep the power balance between different stakeholder groups representing the shareholder interests (Niedenhoff, 2005). Thus if for instance a manager of a related business leaves another manager of a related firm is likely to be nominated for election at the annual general meeting by the SRs. Finally, the proxy voting rights of banks reduce the voting power of minority shareholders and in combination with large block holders these two interest groups are dominating polls at the annual general meetings in general and decisions about the composition of the SRs at the SB in specific (e.g., Elston and Goldberg, 2003). Thus direct shareholders, especially minority shareholders have low influence on the composition of the SB.1
Summarized, the composition of the SB shows that the German corporate governance mechanisms do not only shift power from shareholders to ERs (Gorton and Schmid, 2004), but as well from shareholders to SRs with no equity stake and banks (e.g., Elston and Goldberg, 2003). The high involvement of other stakeholder groups than direct shareholders in the executive compensation setting process in Germany is the major difference to the composition of the compensation committees in Anglo-Saxon economies and of fundamental importance for the further argumentation in the paper. Following the interests of the actors bargaining over executive pay are identified and it is shown to what extent these interests are differing.
Interests and interests of conflicts at the SB
It is important to notice that all stakeholder groups involved in the company are having one common interest, namely the survival of the firm. For work councils the competitiveness and productivity of the company is as crucial as for shareholders, since it safeguards employment and wages (Braun, 2002). However, no consensus exists about the redistribution of income as a result of the company’s operating activities (see e.g., Jensen and Meckling, 1979; Braun, 2002; Gorton and Schmid, 2004). Jensen and Meckling (1979) describe a conflict over the allocation
1
of profits between employee and shareholders. Assuming that SRs and ERs at the SB are insistently representing the interest of its stakeholder group, the conflict over profit allocation, is transferred from the shop-floor level to the SB and carried forward there.
SRs with an equity stake in the Company
Aguilera and Jackson (2003) point out that investment of shareholders is either motivated by financial or strategic considerations. Hence direct shareholders who occupy on average 29% of the seats at the SB comprise of either financial or strategic investors. A financial investor is following a value maximising investment strategy aiming to generate high dividends and stock price gains. These investors are predominant in an Anglo-Saxon business environment (Dore, 2000).
The investment decision of strategic investors is dominated by non-financial considerations. Dore (2000) argues that in Germany and Japan a considerable proportion of the investment decision are based on strategic considerations in order to influence the business strategy of company. Strategic investors aim to get control rights over a company in order to establish existing business relationships, securing access to markets, enhancing technological cooperation, vertical integration, and safeguarding a company’s authority (Aguilera and Jackson, 2003). Here one example:
In 2007 the German car manufacturer Porsche acquired a stake of 31% of Volkswagen. The drivers behind this investment are of strategic nature. Porsche and Volkswagen are producing cars based on the same, jointly developed platform and cooperate in other areas of research and development (Stern, 2007). Porsche was afraid that financial investors, namely foreign hedge funds could acquire a majority stake in Volkswagen and strip of the company’s assets. As a consequence the industrial corporation between the two car producers would be endangered (Stern, 2007). After the acquisition managers of Porsche hold three of the ten seats allocated to SRs at the SB of Volkswagen (VW, 2007) and therewith safeguard the industrial corporation.
Furthermore, Faccio and Lang (2002) show that cross-holdings between large public traded German firms are widely used in order to get strategic control over business partners and establish contractual relationships. Thereby managers are serving at each others SBs resulting in interlocking directorships’ (Conyon and Schwalbach, 2000).
likely to collide with purely financial interests of minority shareholders. One recent example of a conflict between shareholders in Germany again includes Volkswagen, but this time as the initiator of a strategic investment and not as the target.
In 2007 Volkswagen purchased 30% of the shares of the truck producer MAN. As a result of the investment, representatives of Volkswagen hold 3 of the 10 seats allocated to SRs at the SB. Minority shareholders of MAN accuse the chairman of Volkswagen’s SB and major stakeholder of the company to illegally collide with MAN’s ERs at the SB of the truck manufacturer. Volkswagen’s chairman committed not to strip of MAN and lay of workers in the case of a complete takeover. In turn labour representatives at MAN’s SB welcomed Volkswagen’s intention to acquire the 30% stake in the truck producer. Volkswagen plans to form a strategic alliance with MAN and its own truck business unit in South America. Minority shareholders are planning to sue Volkswagen, as they see their interests neglected. They are afraid that the SRs delegated by Volkswagen at the SB of MAN are focussed on the success of the indented joint venture and at the expenses of MAN’s minority shareholders (Handelsblatt, 2007a).
Shareholder Representatives without an equity stake and Banks
The remaining 21% of the SB seats are allocated to SRs without any equity stake in the company or banks. Financial investors are aiming to increase the wealth of their stockholdings (Dore 2000; Aguilera and Jackson, 2003). However, SRs without an equity stake have no stockholdings. Thus, although they are elected by the shareholders, SRs without an equity stake are likely to have other than purely financial interests in the firm. The attendance of business partners without an equity stake in the company is likely to be motivated strategically in order to establish a business relationship (Conyon and Schwalbach, 2000). The by Aguilera and Jackson (2003) described reasons for a strategic investment, namely control, can be applied to business partners at the SB. For instance:
With respect to banks Conyon and Schwalbach, (2000) point out that financial institutions use their seats at the SB for strategic considerations in order to get information advantage over competitors and improve their position as creditors. Here two examples:
The executive managers of the Allianz SE, Germany’s largest insurance company, are holding SB mandates in 6 of the 30 largest German companies (Allianz, 2007). The managers of Deutsche Bank hold SB mandates at 7 of the 30 largest German corporations
(Deutsche Bank, 2007). This gives both financial institutions an information
advantage compared to its competitors, since important financing decisions are presented at the SB before they are made public.
Furthermore, Krozner and Strahan (2000) argue that creditors at SB are risk averse and focussed on safeguarding their interest income and the redemption of credits handed out to the company. Maximising shareholder wealth is not their primary focus. Agarwal and Elston (2001) prove that financial institutions are influencing the financing choices of a firm and conclude that rent-seeking behaviour of banks results in a conflict between creditors and shareholders.
Summarized, the composition of the SRs at the SB indicates that the interests of SRs are dispersed ranging for purely financial, wealth maximising interest to manifold strategic interests such technological partnership, safeguarding the company’s authority, and creditor specific interests as described by Aguilera and Jackson (2003). A dominance of purely financial interest as in the US is disputable (Dore, 2000).
Unions and work councils
The interest of unions is to secure adequate compensation of their members across different enterprises in an industry. Unions favour external means of control by using their collective bargaining power in order to set standards that apply to an entire industry (Aguilera and Jackson, 2003).
In contrast, enterprise-based unions (work councils) favour internal participation on a company level. Long-term employment, regulation of internal promotion prospects and adequate compensation are their main objectives. Work councils have an incentive to improve the firm’s competitiveness in order to safeguard growth and stable employment (McDonald and Suen, 1992; Aguilera and Jackson, 2003).
A conflict between unions and work councils may arise when firm-specific interests oppose with industry wide standards set by the unions (Hassel and Rehder, 2001). Here one example:
agreements and industry-wide standards are softened. This would substantially weaken the power of unions. Consequentially a conflict between unions and work councils would arise.
In large German companies this potential conflict between internal and external labour representation is diminished, since the Codetermination Act allows unions to internally control and influence the company’s decisions as a result of their participation at the SB (Niedenhoff, 2005). Hence, in large firms external mechanism of control are getting less important and collective bargaining is replaced by internal firm-level bargaining (Hassel and Rehder, 2001). Wages in large companies are directly negotiated between the work councils in corporation with unions predominant in that company on the one side, and the firm’s managers on the other side (Hassel and Rehder, 2001; Ochel, 2005). Thus is large German codetermined firms unions and work councils are jointly using internal participation to bargain over employee’ wages. In the 1990’s wages in around 46% of the largest 120 German firms were negotiated on a company level. Ochel (2005) describes the trend of further substituting the centralised wage bargaining policy of unions by a more decentralised company level wage bargaining of unions and work councils. Company level wage agreements are increasingly used. The advantage of these agreements is that the economic situation of the company is taken into account and the competitiveness of a company can be maintained (Hassel and Rehder, 2001; Ochel, 2005). As shown work councils and unions are acting jointly in order to pursue labour interests, thereby unions are increasingly willing to consider company specific conditions. Hence, the goals of both groups of labour representatives at the SB are relatively homogenous compared to SRs.
The major interests of labour are setting adequate working standards such as safety at work, ensuring adequate compensation and qualification, and facilitating job security (see e.g., Aguilera and Jackson, 2003). However, Hassel and Rehder (2001) show that the focus of employees differs according to the economic situation of the firm. In times of profit growth labour tries to extract large proportions of profits in order to increase their wages. By contrast, in times of corporate restructuring and economic downturn ERs are aiming to maintain staffing levels and secure jobs (Hassel and Rehder, 2001; Gorton and Schmid, 2004).
dependent on the company’s economic situation (Hassel and Rehder, 2001; Gorton and Schmid, 2004).
It is important to realise that the common goal of all stakeholders is the survival and the competitiveness of the firm. But, employees and shareholders are competing for a high share of the company’s profits. Furthermore, the goals within the group of SRs and ERs are differing (Aguilera and Jackson, 2003). However, the interests of unions and work councils are likely to be more conform than the interests of SRs, since the unions in large German companies are in favour of internal control strategies and therefore corporate with the work councils (Hassel and Rehder, 2001; Ochel, 2005). As shown the interests of SRs are more dispersed (Aguilera and Jackson, 2003).
As a result of manifold interests of SRs it is questionable if German SRs are capable to act jointly and speak with one voice. Following Coff (2005) one determinant of the bargaining power of a group is its ability to act unified. Thus the diverging goals of SRs are likely to weaken the power of SRs at the SB. In return, the power of ERs is likely to be further increased as a result of the weakness of SRs to act conjointly.
Following it is described how executive compensation research on Germany considered the country’s specific corporate governance landscape, characterised by dispersed interests of SRs and strong ERs at the SB.
The focus of executive compensation research in Germany
Aguilera and Jackson (2003) argue that the Principal-Agent Theory is focussed on the conflict between managers and shareholders. Thereby the diverse interests of other stakeholders as well as the interaction and interdependencies between stakeholder groups are overlooked.
The composition of the SB makes it unlikely that executive compensation in Germany is set solely between principals and agents as assumed by Principal-Agent Theory. As shown ERs are powerful stakeholders and involved in the executive compensation setting process. Thus ERs are likely to influence the executive compensation setting process. A more dimensional Stakeholder-Agent Model needs to be applied in order to determine managerial pay. German executive compensation research partly did so.
concentrated shareholder structure of German firms ownership structure is the central focus of German executive research.
Schmid (1997) integrated variables for bank ownership, family ownership and state ownership in order to analyse the effects of the shareholder structure on executive compensation. A significant positive relationship between bank ownership and executive compensation was established, whereas a high concentration of ownership reduces managerial pay.
In this regard Elston and Goldberg (2003) are developing a more complex approach in categorising control. According to its ownership structure and the composition of the SB the German companies are described as ‘Bank controlled’, ‘Governmental controlled’, ‘Foreign controlled’, ‘Family controlled’ and in the case that a company can not be dedicated to one of the former groups the firm is ‘Manager controlled’.2 This methodology was followed by Haid and Yurtoglu (2006). Both papers established contradictive to Schmid (1997) a negative effect of bank ownership on executive compensation. The influence of concentrated ownership on managerial pay is negative as evidenced by Schmid (1997).
Kaserer and Wagner (2004) consider owners with a stake of 5% or higher as major shareholders, and show that major shareholders reduce the level of executive compensation. As illustrated research on executive compensation focuses on SRs, namely the role of banks and strategic investors with a controlling stake in the company. However, the composition of the SB as a result of the country’s corporate governance arrangements is only partly recognized, since ERs are neglected. Thus codetermination as a factor influencing managerial compensation is still under researched (Chizema and Buck, 2006).
With respect to the methodology used by Elston and Goldberg (2003) and Haid and Yurtoglu (2006) the question arises: Why is a company that is not dominated by banks, the government, foreign investors or a family automatically controlled by its managers and not by employee representatives? This question is of specific interest considering that Elston and Goldberg (2003) illustrate the power of German ERs in the descriptive part of their paper (see also Conyon and Schwalbach, 2000). This shows that authors are aware of the power of labour representatives, but fail to include ERs in their empirical analysis. So far no paper on executive compensation in Germany empirically considers the impact of ERs on managerial pay in equally codetermined firms (Chizema and Buck, 2006).
The focus on SRs and managers can be explained by two major reasons. Firstly, the research on corporate governance in general (Aguilera and Jackson, 2003) and on executive compensation as a feature of a country’s corporate governance arrangements in specific
2
(Chizema and Buck, 2006) is highly focussed on an Anglo-Saxon business environment. As a consequence empirical research on Germany focuses on the Principal-Agent Conflict (and partly on banks) when determining managerial pay. Thereby labour representation as a potential factor influencing executive compensation is overlooked.
Secondly, Chizema and Buck (2006) refer to measurement problems as a reason for disregarding ERs. The legal requirement that in Germany in every large public traded company half of the SB is occupied by ERs and the fixed proportional representation of members of the work councils and unionists creates uniformity between large corporations. Therefore it disqualifies proportional measures of labour representation such as the amount of seats at the SB allocated to ERs in order to analyse labour’s impact on managerial pay.
Summarized, in research on executive compensation the power of ERs and SRs is only theoretically addressed. In the empirical analysis the research focuses on the impact of stakeholders who provide capital to the firm, namely shareholders and bankers on managerial pay. The impact of ERs is empirically underresearched (Chizema and Buck, 2006). In this paper the impact of employee representation on executive compensation is analysed empirically for the first time.
ERs as a determinant of executive compensation
Chizema and Buck (2006) cite the uniformity in the proportional participation of ERs as a reason for disregarding their impact on executive compensation contracts. This excludes proportional measures of labour power based on equal and one-third codetermination as applied by Gorton and Schmid (2004).
Gorton and Schmid (2004) measure the bargaining power of ERs specific to a company by dividing a sample of firms according to one-third and equal participation of ERs at the SB. The effects of strong labour representation (group of large, equally codetermined firms) on company performance are analysed compared to firms with weak labour representation (group small firms with one-third codetermination). The authors show that companies with equal codetermination are traded at 31% discount at the stock exchange compared to companies with one-third ERs and that employees in equal codetermined firms are able to extract higher proportional rents. This approach enables a comparison between large and small firms. However, it does not allow for comparisons among large codetermined companies with more than 2000 employees as carried out in this paper.
Furthermore, the source of labour power has to be determined by company specific characteristics and not on an industry level such as by the bargaining power of an industry-based labour union.
The bargaining power of the ERs in a codetermined firm is more likely to reflect the power of work councils than the power of the on industry-level operating unions because of the following reasons. Firstly, 70% of the seats of ERs at the SB are allocated to members of the work councils, the remaining 30% of the seats are allocated to unionists (Figure 1; Niedenhoff, 2005). Thus the labour fraction at the SB is dominated by company insiders and not by the unions. Secondly, wages and staffing levels are frequently bargained decentralised on a company level and not centralised between industry-based unions and employers’ associations (Hassel and Rehder, 2001; Ochel, 2005). Hence, in large codetermined firms the power of labour is more likely to be dependent on the power of the work council reflected by ERs at the SB, and not by the union dominant in that specific company. Therewith the power of ERs is predominantly determined at a company level and differs between firms.
The literature on bargaining power of unions agrees that labour power is highly dependent on the institutional environment and the industry a company is operating in. Both characteristics are setting a frame for the capability of the workforce to act unified (see e.g., Mishel, 1986; McDonald and Suen, 1992; Paci et al., 1993). Therefore much attention is paid to the different institutional and industry specific settings in order to establish proxies for the bargaining power of unions (see e.g., McDonald and Suen, 1992; Paci et al., 1993).
However, this study focuses on equally codetermined German firms; consequentially uniformity in the institutional setting can be assumed. In all companies work councils are in operation, all companies are equally codetermined, and the same law applies to all firms (Niedenhoff, 2005). Industry specific settings are neglected here, since here it is argued that the bargaining power of labour in large firms is determined on a company level as a result of internal labour control and a decentralised bargaining policy (Hassel and Rehder, 2001; Ochel, 2005). 3
In the literature a variety of proxies are developed in order to measure union bargaining power (see e.g., Armstrong et al., 1976; McDonald and Suen, 1992; Paci et al., 1993). Following three common measures are introduced and their appropriateness to proxy bargaining power on a company level is discussed.
3
Unionisation refers to the degree a workforce is organised (Armstrong et al., 1976). If a high proportion of a company’s workforce is organised in unions, labour is more effectively in pursuing their goals and increases their ability to strike. Applying the proxy of unionisation to German work councils one would be faced with the problem that the entire workforce of a company is allowed to elect the work council (Niedenhoff, 2005). Data on the proportion of employees that actually use their right to vote and therefore could be seen as actively involved and interested in work council issues are not available. Thus, participation at the work councils does not allow for statistical analyses.
Another proxy for union power is the wage differential between union and non-union members of the workforce as described by McDonald and Suen (1992). If the wages of employees who are organised in unions are much higher than the wages of unorganised employees this would indicate that the union has the power to extract high rents for its members. Thus varieties in wage differences between union and non-union members are seen as an indicator for the power of unions. Applying this to German work councils it has to be stated that all members of the company’s workforce are organised in work councils. Thus in large German enterprises there are no unorganised employees. Furthermore, German law requires that the wages bargained centralised between unions and employers’ associations as well as wages bargained decentralised by the work councils have to be paid to all employees of the company (Niedenhoff, 2005). Thus no measures based on wage differentials can be established, since there are no wage differences between organised and unorganised employees within a German company.
A third measure of labour power is the share of wages in revenues allocated to the company’s workforce (McDonald and Suen, 1992). Paci and colleagues (1993: 66) see the union power represented “in the share of any ‘rent’ captured by the union”. Similarly, Chizema and Buck (2006) argue that in a company where a high proportion of the revenues are allocated to labour, ERs are powerful actors. A company with a high involvement of labour in the production process is seen as more dependent on its employees (Chizema and Buck, 2006). Thus the united workforce of a firm that heavily relies on the factor labour in the production process is more effectively in bringing the production to a halt (Freeman and Medorff, 1984). This increases the bargaining power of the workforce since “the power of the union derives from its ability to present the employer with a credible and sustainable threat, such as the threat of strike action” (Paci et al., 1993: 66).
Suen, 1992; Paci et al., 1993; Chizema and Buck, 2006). Furthermore, the rents distributed to labour can be determined on a company level and are likely to be different between equal codetermined corporations. As described by McDonald and Suen (1992) and proposed as a measure for empirical research by Chizema and Buck (2006), the share of personnel expenditure in turnover that is allocated to labour (hereinafter called labour intensity) is used as a proxy for the power of ERs.
Summarized, high proportional rents distributed to employees indicate high labour power (McDonald and Suen, 1992; Paci et al., 1993; Chizema and Buck, 2006). This measure is simplified since it neglects other factors influencing the power of ERs such as the competitive environment, individual firm characteristics, and the institutional setting a firm is operating in (for more complex measures see for instance McDonald and Suen, 1992; Paci et al., 1993). However, the institutional settings as the most important factor influencing labour power (McDonald and Suen, 1992; Paci et al., 1993) can be neglected, since the German Company Law and the Codetermination Act create uniformity in a variety of labour power influencing factors. More direct measures such as labour employment contracts or notes of the SB meetings are confidential and therefore not available to the public. Furthermore, the uniformity in the participation of employees and unionists at the SB does not allow for a direct proportional quantification of labour power. Consequentially, the measure labour intensity is a proxy and rather indirect. However, labour bargaining power is by its nature unobservable and difficult to quantify and therefore has to be determined by using proxies (Paci et al., 1993). Thus the choice made here is seen as the best applicable.
Following the impact of employee representation on performance relation of managerial pay is discussed.
ERs and performance relation of executive compensation
companies that were affected by the Codetermination-Act in 1976 reduced by 19%. Summarized, there is sound evidence that the workforce of a codetermined company is able to alter the objective of a company by reducing shareholder wealth resulting in a reallocation of the company’s surplus to employees (e.g., Jensen and Meckling, 1979; Dilgner, 2002; Gorton and Schmid, 2004). Thus, the Codetermination Act provides German labour, represented at the SB by ERs, with the power to influence company decisions.
But just like SRs, ERs have limited direct influence on the day-to-day business of a company, since operating activities are executed by the managers appointed by the SB. Aguilera and Jackson (2003) describe that in a stakeholder-oriented society the management is obliged to the interest of different stakeholder groups. In Germany the management of a company is seen as successful when it is able to balance between the different interests of capital and labour (Hartmann, 2002). If managers are obliged to shareholders and employees, and both groups have substantial power in the executive pay bargaining process, the performance-based compensation elements of executive payment contracts are likely to comprise shareholder and labour specific performance incentives.
So far research on performance relation of managerial pay in Germany focuses on measures of company performance used in Anglo-Saxon economies. In the literature on executive compensation in Germany it is not empirically recognized that ERs might have an influence on performance relation of managerial pay.
In the Principal-Agent Theory it is stated that shareholders link the manager’s pay to their personnel wealth increase, namely to changes in stock prices (e.g., Jensen and Meckling, 1979; Jensen and Murphy, 1990). By share price performance based compensation elements the interests of managers are linked to those of the shareholders. In the US, there is no substantial difference between the ownership structure of a company and the composition of the compensation committee that comprises to 92% of direct shareholders (Conyon, 2006). Ownership of the company and the right of bargaining over executive compensation is reserved solely to shareholders. Thus performance relation of managerial pay is analysed in a Principal-Agent Model by focussing on share price performance based measures such as total shareholder return (TSR) in order to judge managerial performance (see e.g., Jensen and Murphy, 1990; Conyon and Schwalbach, 2000).
satisfies the interests of 92% of the compensation committee members when consistently pursuing a strategy focussed on the increase of shareholder wealth. A German executive, following the same strategy, is satisfying the primary interests of only 29% of the SB members and is likely to neglect the goals of the remaining 71% of his supervisors.
Here it is argued that the described dominance of other stakeholders than direct shareholders at the SB will influence the performance relation of executive contracts in Germany. Similarly, Chizema and Buck (2006) argue that TSR as a relatively narrow and shareholder focussed measure of success does not reflect the different interest represented by SB members.
That may explain why broader measures of performance, not solely concentrated on stock price changes, are frequently used in Germany. However, so far neither performance measures are consistently used nor are the choices for specific measures of performance sufficiently explained. For instance, Conyon and Schwalbach (2000) use the traditional Anglo-Saxon measure of TSR in order to determine company performance. Schwalbach (1999) uses operating margins as a measure of performance. Elston and Goldberg (2003) and Haid and Yurtoglu (2006) use return on assets (ROA) and return on equity (ROE) to describe performance. Kaserer and Wagner (2004) use TSR, Earnings before Interests and Taxes, and ROE to measure managerial performance.
In this paper the measures used in order to describe managerial performance are derived from the interests of the SRs and ERs. It is argued, that a supervisor perceives an executive manager as successfully if the executive pays attention to the goals of the stakeholder group the supervisor is representing at the SB.
The interest of SRs can be of financial or strategic nature (Aguilera and Jackson, 2003). However, strategic interests in a firm are difficult to identify and manifold as described in Section II. For simplicity the interest of SRs are measured in purely financial terms by applying the widely used and accepted measure of TSR (see e.g., for US: Jensen and Murphy, 1990; Hall and Liebmann, 1998; for Germany: Conyon and Schwalbach, 2000; Kaserer and Wagner, 2004). TSR measures the changes in share prices including dividend payments, and therefore reflects the increase or the reduction in personnel wealth of the shareholders (Conyon and Schwalbach, 2000). TSR is used as the first performance based measure in the empirical analysis representing the interests of SRs.
executives pursue the goals of the company’s workforce. Thus ERs at the SB aim to link employees’ welfare to the wealth of the management by compensation contracts.
The physical well-being of employees is difficult to measure, and in the developed economies such as in Germany high social standards are put into action through legal regulations. Thus, here attention is paid to financial benefits. As described before the strategy of labour representatives is very much dependent on the economic situation of the company, either job security in times of corporate restructuring or increases in wages in times of economic growth are their priorities (Hassel and Rehder, 2001; Gorton and Schmid, 2004).
According to the Handelsblatt (2005) the largest German companies undertook restructuring activities in the years 2001 to 2003 after a global economic downturn. As a result of the restructuring activities the profitability of German enterprises increased significantly in the years 2004 and 2005 (Handelsblatt, 2005). Thus in 2004 and 2005, the period under observation in this paper, the German companies benefited from the restructuring activities, an economic upturn, and consequentially recorded a growth in profits.4 After years of wage restraints and focussing on safeguarding jobs ERs are likely to expect from the managers to increase the wages of the workforce.
Consequently, changes in wages and other financial benefits such as pensions (hereinafter called changes in employee return) are used as a second measure of company performance out of a labour perspective in this paper. So far in the literature on executive compensation in Germany the interests of labour and their impact on performance relation of managerial pay has been not researched at all.
In research on the US Arora and Alam (2005) use the job creating capacity of a company as an alternative measure of employee-oriented managerial performance. However, this measure has one substantial weakness. The ERs are representing the interest of the existing workforce of the company and not of potential future employees. Paci and colleagues (1993) point out that it is not the primary goal of internal unions to create new employment. The creation of new jobs is likely to be a sign of company growth but not of a management style focussed on employee’s welfare. Although it can be important to create new jobs in order to avoid overtime hours and therefore improve the physical well-being of the existing workforce.
Hence, applying the measure of changes in employee return (CER) is seen as a more adequate ratio to quantify the interests of the company’s workforce in times of economic growth than the job-creating capacity of a company as described by Arora and Alam (2005).
4
Figure 2
performance to their group specific interests. The interests of shareholders are represented by applying the measure of TSR, whereas CER is used as a ratio that is reflecting the interests of employees in times of economic growth.
By deducting performance measures directly from the individual interests of the SRs and ERs involved in the executive payment setting process new insights on the performance relation of executive compensation in Germany are expected. Furthermore, by applying both measures, namely TSR for the representatives of the principals and CER for the representatives of the company’s workforce the importance of ERs on executive pay can be substantiated. If ERs are powerful enough to link managerial compensation to labour interests they are consequently influencing executive payment contracts. Thus, the fundamental claim of this paper, that labour representatives have to be considered when determining executive compensation contracts in Germany could be substantiated.
The line of argumentation of this first part of the paper is illustrated in Figure 2. Based on the composition of the SB, the importance of ERs in the executive payment setting process is shown. It is outlined why bargaining power of ERs differs between companies and the likelihood that ERs influence performance based compensation elements of executive remuneration packages is expressed.
DETERMINANTS OF EXECUTIVE COMPENSATION IN GERMANY
In the first part of the paper the constraints of research on executive compensation in Germany were described with a focus on ERs as important actors who are capable of influencing company decisions and performance. Furthermore, two ways were proposed how to measure a possible impact of ERs on executive compensation. That part aims to empirically prove the impact of ERs in order to answer the main research question:
Are ERs a determinant of executive compensation in Germany, and if so what is the influence of ERs on executive compensation?
Power of Shareholder Representatives
Wagner (2004) show that a wide dispersion of ownership leads to higher levels of executive compensation. These results are in line with research focussed on Anglo-Saxon economies (see e.g., Shleifer and Vishny, 1986; Bertrand and Mullainathan, 2001). With respect to participation in the compensation committee Conyon and He (2004) show that the involvement of a large shareholder in the executive compensation process significantly reduces CEO pay. Concordant with the literature the relationship between the power of SRs and executive compensation is expected to be negative. Thus, a low dispersion of ownership increases the power of SRs and therefore reduces the level of managerial pay and vice versa.
H1 The power of SRs has a significant negative influence on executive compensation.
Power of Employee Representatives
So far no study has determined the impact of ERs on managerial pay. As already established the power of German ERs is high, since codetermination shifts substantial influence on company decisions to ERs (see Jensen and Meckling, 1979; Sandrock, 2005). In contrast, the bargaining power of SRs is weak due to the low participation of direct shareholders at the SB (Conyon, 2006), the diverse interests of SRs (Aguilera and Jackson, 2003), and thereupon their inability to act unified (Coff, 1999).
In the German corporate governance system with powerful ERs and relatively weak SRs the question arises how managers interact with the stakeholder groups in order to pursue their own goals, namely the maximisation of their personal welfare. Dore (2000: 182) sees managerial attention redirected from shareholders by German law and describes the following situation: “Through…co-determination in the firm… the system balances the rights and powers of employees against those of shareholders, and thus constrains managers to be as concerned with the welfare of employees as with their duties to shareholders, if not more so”. In general Aguilera and Jackson (2003) describe three axes of interaction between stakeholders:
Following it is judged based on already established results of research on executive compensation and on codetermination which scenario is most likely to be predominant in German firms. Thereupon a relationship between the power of ERs and the level of managerial pay is established.
Class conflict
In a class conflict management and SRs align their interest against ERs. For Germany this scenario seems to be unlikely, since: Firstly, the German Codetermination Act is seen as a way to diminish the class conflict by allocating power to ERs (Hartmann, 2002; Niedenhoff, 2005). A labour dispute is considered as a typical characteristic of a class conflict. In the period from 1995 to 2006 the working days lost as a result of strikes were the lowest in Germany compared to 15 other European countries, and for instance three times lower than in the Netherlands and around 60 times lower than in Spain (WSI, 2007).
Secondly, as described SRs tend to have financial or strategic interests weakening their bargaining power compared to ERs (Coff, 1999; Aguilera and Jackson, 2003). This is likely to affect the strategic behaviour of a firm’s management. In order to be able to successfully cooperate with SRs the executives have to judge, whether SRs are predominantly focussed on strategic or financial interests. In case strategic goals outweigh the interests of financial investors, the goals of strategic investors have to be identified. Additionally, the shareholder structure can change over time, and hence the interests of SRs may change. A strategic investment may turn into a financial motivated investment. Thus, the power relations between different SRs are difficult to determine for the company’s management. In consequence, SRs are less attractive partners for executives to cooperate with.
Summarized, it seems to be unlikely that managers align with the SRs at the SB in order to jointly pursue a strategy at the expenses of ERs. A dominance of shareholders and managers over employees is more likely in a corporate governance system with weak employee power in the firm and shareholders with predominantly financial goals such as in Anglo-Saxon economies (Aguilera and Jackson, 2003)
Accountability conflict
the interests diverge too sharply the coalitions are likely to break down and managers would be able to increase their remuneration at the expenses of other stakeholders (Aguilera and Jackson, 2003).
If managers would be closely monitored by SRs and ERs the question remains, to what extent the interests of SRs and ERs are congruent and how long a coalition of ERs and SRs is operating cooperatively. The interests of SRs and ERs do substantially differ and a conflict is likely to arise when it comes to the allocation of the firm’s surplus (Jensen and Meckling, 1979; Braun, 2002). The fact that an increase of ERs at the SB by 50% is resulting in a shareholder wealth reduction by 31% (Gorton and Schmid, 2004), makes it improbable that the interests of shareholders at the SB and ERs are compatible.
However, based on the assumption that lowest common denominator of all stakeholder groups, the prosperity and survival of the firm, is strong enough to tie stakeholders together and jointly supervise the management, one would expect that executive compensation levels in Germany are significantly lower compared to other economies. Because these close monitoring activities of managerial actions, would be substantially reduce the ability of executives to capture the residual loss and thereby increase their welfare (Jensen and Meckling, 1976).
Although it is argued that codetermination in Germany a reason for lower executive compensation compared to Anglo-Saxon economies, especially to the US (Conyon and Schwalbach, 2000; Elston and Goldberg, 2003) a significant lower level of compensation to Continental European and Japanese companies could not be established (see e.g., Abowd and Bognanno, 1995; Conyon and Schwalbach, 1999). Moreover the level of executive compensation in Germany corresponds approximately with the European average (Abowd and Bognanno, 1995; Manager Magazin, 2005; Manager Magazin, 2006). Thus German managers are not likely to be free riders and turn a potential power vacuum between ERs and SRs into personal profits as theoretically described by Aguilera and Jackson (2003).
Summarized, there is little evidence that SRs and ERs are jointly monitoring the managers, since both groups tend to have conflicting interests in terms of the redistribution of the firms surplus. Furthermore, the corresponding levels of executive pay compared to other European companies indicate that German managers are not able to gain extraordinary profits as a result of a power vacuum at the SB.
Insider-outsider conflict
insider aim to assure their position in the company by delaying or impeding restructuring activities, building up barriers for hostile takeovers and favouring internal diversification. Gorton and Schmid (2004) show that restructuring activities in equal codetermined firms are less radical than in one-third codetermined firms, resulting in higher labour cost. Corporate restructuring often results in job losses, and therefore downsizing activities are not in the interests of ERs (see e.g., Gorton and Schmid, 2004). Similar hostile takeover often lead to job losses and are therefore not in the interests of ERs.
The manager’s position can be also substantially threatened by hostile takeovers, since he might become redundant. Restructuring activities bear the risk that the company size is reduced as a result of a lay off of workers and a sale of non-core business activities. These downsizing effects are not in the interests of managers, since company size is a major determinant for executive compensation (Schwalbach, 2001). Furthermore the costs of restructuring activities are negatively influencing the profit situation of a company in the short time. The profits from a successful restructuring activity are likely to be capitalised on the long run (Schwalbach, 2001). Thus downsizing activities are likely to negatively affect the pay of managers who tend to act in a short-term horizon (Schwalbach, 2001). Hence, risk-averse managers are likely to prevent hostile takeovers and avoid restructuring activities.
Thus in every firm there is a joint intersection between ERs and executives. Both groups are likely to prevent hostile takeovers by competitors and delay restructuring activities (Aguilera and Jackson, 2003). Following it is explained why the joint interests of ERs and managers are even higher in Germany.
all ERs. In contrast it is likely to be more difficult for a potential candidate to get support of all SRs as a result of their different priorities.
If managers fail to get the recognition of ERs, labour representatives are unlikely to back their business strategy and support a further tenure.
The most recent example is the appointment of Mr. Großmann as CEO of the second largest German utility company RWE. The contract with the existing CEO, the Dutch manager Mr. Roels was not elongated. Although market participants describe Mr. Roels achievements during his 4 year tenure as impressing: the share price quadrupled, the company debt was significantly reduced, and RWE was focussed on its core businesses (Handelsblatt, 2007b). But ERs and strategic investors did not honour this shareholder value oriented management style and installed Mr Großmann. Mr Großmann is considered as a very successful steel manager, focussing on good employee relationships and having close connections to German politics and labour unions (Handelsblatt, 2007c). In Handelsblatt (2007b: 1) the appointment of Mr Großmann was commentated as follows “Shareholder Value does not have the highest priority at the SB of RWE. Mr Roels offended ERs and some municipal shareholders with his shareholder value oriented management approach during his four years in office”.
Secondly, ERs are not only involved in the selection process of executive managers. There are also interdependencies between the management and ERs when bargaining over employee’s and manager’s wages. In large companies work councils (conjointly with unions predominant in the firm) are negotiating the company’s staffing level and the wages of the firm’s employees with the executives (Hassel and Rehder, 2001; Ochel, 2005). In turn ERs, who are predominantly members of the work councils (see Figure 1), are involved in the executive payment setting process (Nietsch, 2005; Niedenhoff, 2005). Thus labour representatives have a central position in the wage bargaining process that allows them to influence managerial pay as well as labour pay. Here as described by Aguilera and Jackson (2003) important interdependencies between ERs and executive managers are established as a result of the German corporate governance structure (Dore, 2000; Hassel and Rehder, 2001; Ochel, 2005). The fact that both parties are influencing each other wages facilitates a situation where managers and employees enrich at the expenses of shareholders.
