How do policy and performance of firms change with variations in the stakeholder approach? We compare foundation owned firms (FoFs) and family firms, with and without codetermination. As foundations have no owners, the impact on corporate governance of residual claimants might be weaker and the impact of other stakeholders stronger. We find that German foundation owned firms are more labor intensive relative to matching firms. But their wages and their hiring and firing policy are about the same. Their financing policy is more conservative. Their financial performance is slightly weaker. Apart from financing policy, codetermination has similar effects. These findings indicate a stronger impact on corporate governance of employees in FoFs, combined with long-term orientation. Given various stakeholders with divergent interests, corporate governance of firms has to deal with conflicts of interests. Two approaches are widely discussed, the stakeholder and the shareholder value approach. The latter is criticised because shareholders can impose negative externalities on other stakeholders. This creates inefficiencies which may be partly resolved by a stakeholder approach (Mayer 2013, Magill et al 2015). Critics of the stakeholder approach argue that it provides no clear guidance for the managers of a firm as they can pursue various objectives (Jensen 1986, Tirole 2001). Many firms in Continental Europe are family firms whose owners appear to pay voluntarily more attention to the interests of other stakeholders, in particular employees (Bau and Chirico 2014, Bingham et al 2011). This informal power of employees is supplemented in some countries by formal power. Commercial law may entitle employees to set up workers´ councils which decide together with management on working conditions. In Germany, given more than 500 (2000) employees, 1/3 (1/2) of the seats of the supervisory board are assigned to representatives of the employees.
An important theoretical and empirical issue is how variations in the stakeholder approach affect corporate governance and financial performance of firms. A specific variation is generated by an atypical ownership structure of the firm. Typically, a firm is owned by natural persons being residual claimholders, even in pyramid ownership structures. In a foundation owned firm (FoF) these owners are partially or fully replaced by a foundation which itself is not owned by anybody. The foundation may distribute the cash flows which it receives from the FoF to a set of beneficiaries as prescribed by the foundation charter. Given a charitable foundation, these beneficiaries must not interfere with the governance of the foundation or the FoF. The managers of the foundation may exercise the voting rights of the foundation in the FoF. As they are not residual claimholders, ownership and control are separated (Fama and Jensen 1983). Ideally speaking, the foundation managers should act in the spirit of the (deceased) founder, the initial principal. In the foundation charter, he codified the purposes and objectives to be pursued by the foundation. But, given a lack of enforceability, the foundation managers may pursue their own objectives. For example, they might use their power to benefit friends or neglect their supervisory duties. This additional agency problem poses another challenge to the theory of corporate governance. Agency theory helps to better understand the associated problems (Fama 1980) and to develop contractual arrangements to harmonize the interests of the foundation managers (agents) and the foundation (principal) (Holmström 1979, Eisenhardt 1989).
The theory of corporate governance should answer the following questions. Given a lack of powerful residual claimants in a firm, how is the vacuum of power filled? Do employee representatives get a better access to the powerful managers of the firm and strengthen their informal power (Hill and Jones 1992, Rajan and Zingales 1998)? Do firm managers benefit? What are the implications for the firm´s policy and its financial performance? What are the effects of mixed ownership structures in which non-residual and residual claimants coexist? Even more complex, what happens if corporate governance is also affected by codetermination?
This paper tries to provide some empirical evidence on these questions and fill some gaps in the theory of corporate governance. The effects of blockholdings of foundations and of code-termination are at the core of this paper. In line with Shleifer and Vishny 1997, blockholdings may safeguard the long-run existence of the FoF and, thus, of the foundation. They may help to monitor FoF-managers and to counterbalance claims of other stakeholders (Rajan and Zingales 1998, Stepanov and Suronov 2017). Also, they may facilitate the cooperation with labor representatives (van Essen et al 2013).
In this paper, we analyse German FoFs and compare them to appropriately chosen German matching firms. These are mostly firms with strong blockholders, in particular family firms (Franks and Mayer 1997). We conjecture that employees are more powerful in FoFs. To verify this conjecture, we test for differences in policies and financial performance of FoFs and matching firms. Moreover, we use the unique opportunity in Germany to analyse effects of codetermination. The similarities between the empirical findings of both tests provide strong support for our conjecture that employees are more powerful in FoFs. These findings broaden our knowledge on the effects on policies and financial performance of firms of various stakeholder approaches.
Foundation owned firms exist in various countries, in particular in European countries. FoFs also exist in the US, but foundation ownership is usually less than 20% to preserve tax privileges for charitable foundations. Big FoFs such as Bosch, ZF Friedrichshafen, Körber and Bertelsmann are located in Germany, in Denmark Maersk and Tuborg are among the big FoFs, in Sweden Ikea. As most German FoFs are not listed, we study annual statements, covering the years 2003 to 2012.
Our contributions to the theory of corporate governance are as follows. First, given weak residual claimholders, we hypothesize that managers and/or employees are more powerful and effectively promote their interests. We find that production in FoFs is more labour-intensive than in matching firms. Codetermination reinforces this effect. Labour intensity may be raised by producing more labour-intensive products and services or by vertical integration. But we neither find a higher average income of managers or employees in FoFs relative to matching firms nor in codetermined firms. We also do not find significant differences in the hiring and firing policy between FoFs and matching firms. Also, codetermination does not have a significant effect. However, employees in FoFs are protected against layoffs in takeovers as foundations usually are not allowed to sell their ownership stakes. Thus, FoFs cannot be taken over. Second, financing policy of FoFs is more conservative relative to matching firms. Profit payouts are lower in FoFs, perhaps because foundation managers are only mildly interested in profit payouts since they are not residual claimholders. Leverage is also lower in FoFs. A conservative financing policy mitigates vulnerability of FoFs to negative external shocks. This is also in the interest of FoF-managers and employees. Codetermination, however, has no significant effect on financing policy in our sample. This is in contrast to Fauver and Fuerst (2006) who find a higher leverage and a higher probability of dividend payments in codetermined corporations.
Third, the theory of corporate governance should relate variations in governance to variations in financial performance. Financial performance is measured by average returns on assets (RoA) and risk, proxied by the standard deviation of RoA, σ(RoA). In a panel analysis, we find that the average RoA is lower in FoFs when we linearly control for risk. Since codetermination has a similar effect on RoA as foundation ownership, FoFs likely attach a higher weight to employee interests. This conjecture is supported also by the observation that both, foundation-ownership and codetermination, lower risk to a similar extent. The RoA-findings suggest that more employee orientation weakens financial performance.
Fourth, joint ownership of a foundation and other owners appears to improve financial performance. Surprisingly, listing at a stock exchange weakens financial performance as does foundation ownership. But a positive interaction effect between listing and foundation ownership eliminates both negative effects. Apparently “capital market control” eliminates some weaknesses of foundation ownership in corporate governance, and vice versa. We also derive a Sharpe ratio for each firm, dividing the average (RoA minus risk-free rate) by σ(RoA). This ratio is neither significantly affected by foundation ownership nor by codetermination, controlling for other factors. This finding contrasts with the negative effects on financial performance found in the panel analysis. We conclude that foundation ownership and codetermination weaken financial performance, but this effect is fairly small.
These findings are important for the theory of corporate governance as they highlight the effects on corporate policy and financial performance of several variations in the stakeholder model. The effects of foundation ownership are similar to those of codetermination, except for financing policy. Hence, also foundations likely put or tolerate a higher weight on employee interests. This weakens financial performance so that it costly to residual income-owners. Employees apparently are not interested in maximising their benefits over the short/medium term by extracting money from the firm, but pursue a long-term policy of safeguarding the existence of the firm, thereby promoting job security. They can achieve these goals in a competitive environment only if financial performance is not impaired or only to a small degree. Hence, competitive financial performance strongly constrains the impact of conflicting interests (Fama 1980). Conflicts of interest may be stronger regarding financing policy. Residual income-owners may insist on high profit payouts to secure their financial claims (Shleifer and Vishny 1997), in contrast to foundation managers. At least two organizational channels may explain why the strong separation of ownership and control associated with foundation ownership has only limited effects on financial performance of FoFs. The foundation managers, a special stakeholder group, may act in the spirit of the founder to preserve the long run existence of the firm and their reputation as supervisors. Thus, they may put pressure on the FoF-managers to safeguard profitability. Alternatively, managers and employees of the FoF themselves pursue profitability to preserve their jobs in the long run in a competitive environment so that monitoring of foundation managers may be largely irrelevant. As long as the foundation managers do not insist on policies which are detrimental to FoFs, FoF-managers may act independently.
Our conclusion that foundation ownership weakens financial performance is in contrast to earlier studies. Thomsen (1996) investigates accounting based returns of Danish FoFs. He does not find inferior returns relative to the largest Danish non-FoFs. Thomsen and Rose (2004) compare Danish FoFs with other Danish firms listed at a stock exchange while Thomsen and Hansmann (2013) compare them to firms with a traditional ownership structure. Both studies also use stock market data and arrive at the same conclusion as Thomsen (1996). Hansmann and Thomsen (2013) find that greater managerial distance between the board of the foundation and that of the FoF improves the return on assets. In an early study, Herrmann (1996) and Herrmann and Franke (2002) analyze a small sample of German FoFs over the short period of 1990 to 1992. Relative to German firms listed at a stock exchange, they find slightly higher returns on assets, higher labor intensity, but lower salary levels in FoFs. Risk is ignored.
The paper is organized as follows. In the next section, we provide more details about the motives of founders of foundations and the regulation of foundations in Germany. Then we derive some hypotheses on corporate policies and performance of FoFs. The subsequent section shows descriptive statistics and the following section presents our empirical findings. After discussing some robustness results the paper concludes.