The Effectiveness of Corporate Governance in One-Tier and Two-Tier Board Systems

The Effectiveness of Corporate Governance in One-Tier and Two-Tier Board Systems
Evidence from the UK and Germany by Carsten Jungmann*
Germany and the UK are paradigms of systems in which the control of managing directors of companies either lies in the hand of a separate supervisory board (two-tier system) or is an additional task of the board itself (one-tier system). This paper provides for an empirical test of the effectiveness of both systems of corporate governance. The analysis of the financial performance and board turnover of the largest companies listed on the stock exchanges in Frankfurt and London over a total of 400 financial years establishes that both systems are effective means of control. Yet the analysis also demonstrates that it is not possible to assign superiority to either of them. Therefore, the often raised question as to whether one of the two systems will finally prevail and whether there will be ultimate convergence of both systems, has to be answered in the negative. As the discussion of the strengths and weaknesses will show, however, there is still scope for improvements in each of the two board models.
Table of Contents
ECFR 2006, 426474

427 432 432 435 437 438 438 439 440 441
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Forms of supervision of corporate management in the UK and in Germany 1. The German system of corporate control (two-tier model) . . . . . . . . 2. The British system of corporate control (one-tier model) . . . . . . . . . 3. Groups of persons entrusted with monitoring tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
III. Analysis of the relationship between board turnover and financial performance 1. Earlier comparative research . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Earlier research with relation to the effectiveness of corporate control . . . . 3. The uniqueness of the focus on different board systems . . . . . . . . . . . . 4. Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Dr. iur., LL.M. (Yale), M.Sc. in Finance (Leicester), Wissenschaftlicher Assistent, Bucerius Law School in Hamburg. The author is grateful to Christian Bochmann for his research assistance on this project.
Brought to you by | Universittsbibliothek Muenchen (Universittsbibliothek M Authenticated | 172.16.1.226 Download Date | 3/12/12 6:41 PM
ECFR 4/2006
One-Tier and Two-Tier Board Systems
427
446 448 449 458 462 463 469 473 473
IV. Testing the hypotheses: data on financial performance and board turnover . . . . . . V. Strengths and weaknesses of the concurring board models . . . . . . . . . . . . . . . 1. Advantages and disadvantages of the two-tier model . . . . . . . . . . . . . . . . . 2. Advantages and disadvantages of the one-tier model . . . . . . . . . . . . . . . . . VI. Conclusion and recommendations for the improvement of corporate control . 1. Recommendations for the improvement of corporate control in the two-tier system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Recommendations for the improvement of corporate control in the one-tier system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Reflection on the limits of corporate control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VII. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I. Introduction It is a well-known phenomenon that there are two main sets of legal rules on the supervision of corporate management: one-tier boards and two-tier boards. Within Europe, the United Kingdom is a prominent country with a single board system, consisting of executive and non-executive directors; other countries such as Ireland take the same approach. On the other hand, Germany traditionally employs the dualism of a management board and a separate supervisory board. This system is also found in the Netherlands, Austria, Finland and Denmark 1. While some countries such as Sweden have legal frameworks that cannot be classified as one-tier or two-tier systems, other countries, such as Belgium, Portugal and Spain, allow corporations to choose between the two systems 2. The same once held true for France, but France has since introduced a third option through which more than one organ can be entrusted with executive supervision 3. The European legislator also took the approach of creating a mixed system when it created the European Company, the Societas Europaea (SE). According to Article 38 of the Council Regulation (EC) No. 2157/2001 of October 8, 2001 on the Statute for a European company (SE), a SE shall be comprised of either a supervisory organ and a management organ (two-tier system) or an administrative organ (one-tier system).
1 See Klaus J Hopt, The German Two-Tier Board: Experience, Theories, Reforms in Klaus J Hopt and others (eds), Comparative Corporate Governance: The State of the Art and Emerging Research (OUP, Oxford 1998) 227, 228; Udo C Brndle and Jrgen Noll, The Power of Monitoring (2004) 5 German LJ 1349, 1353. 2 See Carsten Berrar, Die Entwicklung der Corporate Governance in Deutschland im internationalen Vergleich (Nomos Verlagsgesellschaft, Baden-Baden 2001) 3641. 3 See Michel Storck, Corporate Governance la Franaise Current Trends (2004) 1 ECFR 36, 4153.
428
Carsten Jungmann
ECFR 4/2006
There are strengths and weaknesses of the monistic as well as of the dualistic systems with regard to nearly all aspects of the organisation of a companys affairs. Many of these problems, however, lie outside the focal point of this paper. This paper will instead concentrate on the question of effective corporate control, i.e. corporate governance as the system by which companies are directed and controlled 4. Thus, the key questions are whether the German two-tier and the British one-tier systems are effective, and whether it is possible to classify one of them as the superior method of successful corporate governance? Both questions have been regarded as fundamental to the lively comparative debate on corporate governance 5. The relevance of the questions presented and the necessity of analysis becomes apparent when we consider the recent discussions concerning the reform of corporate governance rules in Germany and the UK. One of the key criticisms of the erstwhile German rules concerned the mandatory two-tier board system. Some fifteen years ago, the German dualism of a management board and a supervisory board had been criticised as inferior to the British one-tier board system with executive and non-executive directors in a single board 6, but today, this view is no longer shared 7.
4 Committee on the Financial Aspects of Corporate Governance, Report (London 1992) 15 (para 2.5); Department of Trade and Industry, Modern Company Law For a Competitive Economy (DTI, London 1998) para 3.5; see Reinhard H Schmidt and Gerald Spindler, On the Convergence of National Corporate Governance Systems: A note on Braendle and Noll (2006) 17 J. Interdisc. Econ. 83, 8485; cf. Rolf Birk, Germany in Arthur R Pinto and Gustavo Visentini (eds), The Legal Basis of Corporate Governance in Publicly Held Corporations A Comparative Approach (Kluwer Law International, The Hague 1998) 53, 67 for the German interpretation of the expression. For the broader implications inherent in the concept of corporate governance see Klaus J Hopt, Common Principles of Corporate Governance in Europe? in Joseph A McCahery and others (eds), Corporate Governance Regimes (OUP, Oxford 2002) 175, 176. 5 See Klaus J Hopt and Patrick C Leyens, Board Models in Europe Recent Developments of Internal Corporate Governance Structures in Germany, the United Kingdom, France, and Italy (2004) 1 ECFR 135; Patrick C Leyens, Deutscher Aufsichtsrat und U.S.-Board: ein- oder zweistufiges Verwaltungssystem? (2003) 67 RabelZ 57; but see Udo C Brndle and Jrgen Noll, On the Convergence of National Corporate Governance Systems (2006) 17 J. Interdisc. Econ. 57 (arguing that the question has become obsolete due to the convergence process of corporate governance systems); cf. Reinhard H Schmidt and Gerald Spindler (n. 4), 87 for a critical view on Brndles and Nolls methodological concept. 6 Manuel R Theisen, Das Board-Modell: Lsungsanstze zur berwindung der berwachungslcke in deutschen Aktiengesellschaften? (1989) 34 Die Aktiengesellschaft 161, 163165; see also Erich Potthoff, Board-System versus duales System der Unternehmensverwaltung Vor- und Nachteile (1996) 48 Betriebswirtschaftliche Forschung und Praxis 253 (highlighting the strengths of the board model). 7 But see Patrick C Leyens (n. 5) 97 (arguing that the one-tier system has a higher potential for development in the long term); cf Manuel R Theisen, Empirical Evidence and
ECFR 4/2006
429
Accordingly, the two-tier system has not been abolished under the new German Corporate Governance Code adopted in 2002 (as amended in 2006). A vital discussion of changes to the current system continues instead, which is primarily concerned with ways to improve internal control. If, however, there was to be a major reform to the German corporate governance system in the future, it is likely that such a reform would converge the current system to the British one 8. In the UK, neither the reports on corporate governance nor the recent proposals for a reform of company law called for changes to the unitary board system 9. The British system has instead been regarded as being superior to the German system 10. A decade ago, however, the advantages of the German two-tier board system were at least sporadically highlighted, and its introduction into British law was even postulated 11. However, this proposal never found approval by a majority of commentators. Thus, it appears that the one-tier system is favoured over the two-tier system, and the adoption of a
10
11
Economic Comments on Board Structure in Klaus J Hopt and others (eds), Comparative Corporate Governance: The State of the Art and Emerging Research (OUP, Oxford 1998) 259. See Reinhard H Schmidt, Stakeholderorientierung, Systemhaftigkeit und Stabilitt der Corporate Governance in Deutschland (Working Paper No. 162, University of Frankfurt Working Paper Series: Finance & Accounting, 2006) 2 <http://www.wiwi. uni-frankfurt.de/schwerpunkte/finance/wp/1150.pdf> (accessed 26 October 2006); Thomas Raiser, Unternehmensmitbestimmung vor dem Hintergrund europarechtlicher Entwicklungen in Verhandlungen des sechsundsechzigsten Deutschen Juristentages (Verlag C. H. Beck, Mnchen 2006) B 109110. See Colin Mayer, Corporate Governance in the UK (2000) 8(1) Hume Papers on Public Policy 1; Eva-Dsire Lembeck, UK Company Law Reform (2003) 6 Neue Zeitschrift fr Gesellschaftsrecht 956. Committee on Corporate Governance, Final Report (London 1998) 2627 (para 3.12); Paul L Davies, Board Structure in the UK and Germany: Convergence or Continuing Divergence? (2000) 2 ICCLJ 435. Thomas Sheridan and Nigel Kendall, Corporate Governance: An Action Plan for Profitability and Business Success (Pitman Publishing, London 1992) passim (especially at 161164); Lewis Robertson, Corporate Governance: The Lessons of Recent British Experience (1995) 3(4) Hume Papers on Public Policy 1, 4; see also Robert Ian Tricker, Corporate Governance: Practices, Procedures and Powers in British Companies and their Boards of Directors (Gower, Oxford 1984) 242249 (proposing that all public limited companies should have a Governing Body); for a review and discussion see Geoffrey Owen, The Future of Britains Boards of Directors Two Tiers or One? (Board for Chartered Accountants in Business, London 1995); see also Owen Green, Why Cadbury Leaves a Bitter Taste Financial Times (London 9 June 1992) 19.
430
Carsten Jungmann
ECFR 4/2006
compulsory supervisory board into UK law is very unlikely to happen in the foreseeable future 12. In the ongoing discussion regarding corporate governance, arguments in favour of either system were often based on theoretical approaches 13, rather than on empirical research on their effectiveness 14. In addition, at the beginning of the debate on reform, the proposals concentrated on the abolishment of one system and its replacement by the other. However in the last years, the idea of convergence although forecasted about twenty years ago 15 originated. Today there is a very significant discussion concerning convergence versus continuing divergence in the field of corporate governance 16. And the 2005 Recommendations of the EU concerning the roles of nonexecutive directors and members of the supervisory board 17 may be regarded as a preliminary result of this process 18. This discussion concerning the comparison of the board structure in the UK and Germany was led by Davies 19. However, again, empirical research on the particular question is still missing.
12 Eils Ferran, Company Law and Corporate Finance (OUP, Oxford 1999) 219; Department of Trade and Industry, Modern Company Law For a Competitive Economy: Developing the Framework (DTI, London 2000) paras 3.1403.141. 13 For an overview see Udo C Brndle and Jrgen Noll (n 1) 13621363. 14 See Peter Hommelhoff and Daniela Mattheus, Corporate Governance nach dem KonTraG (1998) 43 Die Aktiengesellschaft 249, 250251 (pointing out that empirical evidence on the superiority of one of the models is missing). 15 Knut Bleicher and Herbert Paul, Das amerikanische Board-Modell im Vergleich zur deutschen Vorstands-/Aufsichtsratsverfassung Stand und Entwicklungstendenzen (1986) 46 Die Betriebswirtschaft 263. 16 See among the important contributions Paul L Davies (n 10); Lucian Bebchuk and Mark J Roe, A Theory of Path Dependence in Corporate Ownership and Governance in Jeffrey N Gordon and Mark J Roe (eds), Convergence and Persistence in Corporate Governance (CUP, Cambridge 2004) 69; Wymeersch, Convergence or Divergence in Corporate Governance Patters in Western Europe? in Joseph A McCahery and others (eds), Corporate Governance Regimes (OUP, Oxford 2002) 230; John C Coffee Jr, The Future as History: The Prospects for Global Convergence in Corporate Governance and its Implications (1999) 93 Nw. U. L. Rev. 641; Henry Hansmann and Reinier Kraakman, Toward a Single Model of Corporate Law? in Joseph A McCahery and others (eds), Corporate Governance Regimes (OUP, Oxford 2002) 56. 17 Recommendations of the European Commission of 15 February 2005 on the Role of Non-Executive or Supervisory Directors of Listed Companies and on the Committees of the (Supervisory) Board, Official Journal of the European Union L 52 (25 February 2005) 51; for a discussion see Gerald Spindler, Die Empfehlungen der EU fr den Aufsichtsrat und ihre deutsche Umsetzung im Corporate Governance Kodex (2005) 26 Zeitschrift fr Wirtschaftsrecht 2033. 18 See Christian Frster, Europische Corporate Governance Tatschliche Konvergenz der neuen Kodizes? (2006) 27 Zeitschrift fr Wirtschaftsrecht 162. 19 Paul L Davies (n 10).
ECFR 4/2006
431
This study is a quantitative comparative assessment of the degree of managerial discipline provided by both systems. It examines the financial performance of companies in the UK and Germany and the reactions of the monitoring bodies in cases of poor performance. Typically, there are various reasons for poor performance, and there are numerous ways to control management 20. If negative performance of a company appears to be related to mismanagement, it is then the appropriate response to release management from its duties, either by not re-electing the managers or by dismissing them 21. Although this measure is frequently the last resort, it is the most meaningful task for those who are entrusted with managerial control. Thus, the functioning of internal control is a proper measure of the performance of a corporate governance system 22 and consequently used in this paper. This leads to the two hypotheses tested: Hypothesis 1: Board turnover is inversely correlated to financial performance, so that the non-executive directors (UK) and the members of the supervisory board (Germany), respectively, exercise their power to (effectively) unseat management. Hypothesis 2: It is impossible to deduce the superiority of either of the two systems from the degree of inverse correlation between the board turnover and the financial performance. The paper is structured as follows: in the subsequent part, the forms of supervision of corporate management in the UK and Germany are presented. This includes a short analysis of the legal rules as well as practical observations. Part III contains the description of the methodology used in this paper to examine the relationship between board turnover and financial performance. It explains the uniqueness of the approach taken here and gives definitions of the relevant terms used. The empirical results of this study are then presented in part IV, which also offers an answer to the question of whether the above mentioned hypotheses have proven to be true. Part V addresses the relevant theoretical explanations for the empirical results: the advantages and disadvan20 See Eugene F Fama, Agency Problems and the Theory of the Firm (1980) 88 J. Polit. Economy 288. 21 Brian R Cheffins, Company Law: Theory, Structure and Operation (Clarendon Press, Oxford 1997) 605; see Eils Ferran (n 12) 217223. 22 See Jonathan R Macey, Institutional Investors and Corporate Monitoring: A DemandSide Perspective in a Comparative View in Klaus J Hopt and others (eds), Comparative Corporate Governance: The State of the Art and Emerging Research (OUP, Oxford 1998) 903, 908909; Rudi K F Bresser, Reynaldo Valle Thiele, Annette Biedermann and Holger Ldeke, Entlassung des Vorstandsvorsitzenden und Unternehmenserfolg: Eine empirische Untersuchung der grten deutschen Aktiengesellschaften (2005) 75 Zeitschrift fr Betriebswirtschaft 1165, 1166.
432
Carsten Jungmann
ECFR 4/2006
tages of the one-tier system and the two-tier system are contrasted. The discussion of these strengths and weaknesses leads to the conclusion in part VI, which encompasses recommendations to enhance both of the systems of corporate control. II. Forms of supervision of corporate management in the UK and in Germany The purpose of this paper is to examine the effectiveness of corporate control. As the extent of board turnover will be used to measure the exercise of power of control, it is important to analyse who exactly is entrusted and empowered with control in both jurisdictions. 1. The German system of corporate control (two-tier model) According to the German Stock Corporation Act of 1965, it is mandatory for all German stock corporations (Aktiengesellschaften) to have two boards: the management board (Vorstand) and the supervisory board (Aufsichtsrat). The supervisory board members are either shareholder representatives or labour representatives 23. Simultaneous membership of the management board and the supervisory board is not permitted. The main tasks of the supervisory board are to appoint and dismiss the members of the management board and to monitor them. The supervisory board also represents the corporation in all affairs concerning the management board, especially by initiating court actions against the board members. In addition, the supervisory board must approve the annual accounts and can intervene in cases where the companys interests are seriously affected. For certain extensive and fundamental decisions, the by-laws must impose that authorisation by the supervisory board is required 24. Finally, the supervisory board also has some soft functions, such as networking with stakeholders 25. Notwithstanding these numerous tasks, supervision remains the core func23 This composition of the supervisory board is due to the German laws of co-determination in companies with more than 500 employees: depending on the size and the business of the corporation, up to 50 % of the members of the supervisory board are labour representatives. They are elected in a rather complicated procedure governed by the applicable co-determination act, while the representatives of the shareholders are elected in the general meeting by the shareholders. 24 See Carsten Berrar, Die zustimmungspflichtigen Geschfte nach 111 Abs. 4 AktG im Lichte der Corporate Governance-Diskussion (2001) 54 Der Betrieb 21812182. 25 Paul L Davies (n 10) 450453.
ECFR 4/2006
433
tion of the supervisory board, whereas all management issues are in principle reserved for the management board, which acts autonomously and is not bound by orders of the shareholders or the supervisory board. As mentioned above, the members of the management board are appointed and can be removed by the supervisory board. The management board does not only manage the companys affairs, but also sets up long term goals and guidelines. At least in principle, there is a clear separation between the tasks and responsibilities of the two boards 26. However, case law in the last few years has shown that the monitoring task of the supervisory board has become more and more permanent and future-oriented, as the intended business policy has also to be controlled by the supervisory board. Sec. 5.1.1 of the German Corporate Governance Code expressly clarifies that it is the task of the supervisory board to advise regularly and supervise the management board in the management of the enterprise and that it must be involved in decisions of fundamental importance to the enterprise. Thus, the members of supervisory board act, to a certain extent, as consultants of the management board which necessitates an ongoing and continuous discussion 27. There are, in addition, shareholder control mechanisms, which are often referred to as voice and exit 28. In small family companies and companies with only a small number of shareholders, the shareholder influence is significant, both practically and theoretically: few shareholders are able to coordinate their activities and act efficiently 29; in particular, few shareholders can effectively make use of their rights in the general meeting, which the German Stock Corporation Act of 1965 regards as the highest corporate organ. By contrast, in listed and publicly traded companies, which hold the spotlight in the corporate governance debate and are the focal point of this study, most investors have only a tiny proportion of the shares and are not in a position to effectively exercise control in the general meeting. Furthermore, monitoring costs exceed the benefit of being able to exert influence and nonmonitoring shareholders would gain the same advantages (free-rider problem). Thus, the incentives for shareholders to exercise control are low. Notwithstanding these general reflections on the role of shareholders in large corporations, we need to make two remarks concerning distinct particular26 Knut Bleicher and Herbert Paul (n 15) 265. 27 For a review of the latest court decisions see Hartwig Henze, Neuere Rechtsprechung zu Rechtsstellung und Aufgaben des Aufsichtsrats (2005) 58 Der Betrieb 165. 28 Albert O Hirshman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Harvard University Press, Cambridge 1970); see also Reinhard H Schmidt (n 8), 69. 29 See Rolf Birk (n 4) 65.
434
Carsten Jungmann
ECFR 4/2006
ities of the German stock market. The first concerns the ownership structure of large companies. At least traditionally, institutional investors and especially banks have held large proportions of shares in Germany 30. This was regarded as a major advantage for effective corporate control 31. However, in Germany as well as in the UK, there is a now general tendency towards a less dispersed ownership 32. Thus, the first phenomenon observed is becoming less important; at least, it is no longer a German particularity 33. Secondly, the threat of (hostile) takeovers helps to discipline the management in other jurisdictions 34, but we have to recognise a lack of hostile public takeover bids in Germany 35.
30 Theodor Baums, Corporate Governance in Germany: The Role of the Banks (1992) 40 Am. J. Comp. L. 503; Klaus J Hopt, Corporate Governance und deutsche Universalbanken in Dieter Feddersen, Peter Hommelhoff and Uwe H Schneider (eds), Corporate Governance (Otto Schmidt-Verlag, Kln 1996) 243, 247248. 31 See Mark J Roe, A Political Theory of American Corporate Finance (1991) 91 Colum. L. Rev. 10; Edward B Rock, Americas Fascination with German Corporate Governance (1995) 40 Die Aktiengesellschaft 291; see more generally Stu Gillan and Laura Starks, Corporate Governance Proposals and Shareholder Activism: The Role of Institutional Investors (2000) 57 J. Finan. Econ. 275 32 See Office for National Statistics, Share Ownership: A Report on Ownership of Shares as at 31st December 2004 (London 2005); Jean Tirole, The Theory of Corporate Finance (Princeton University Press, Princeton/Oxford 2006), 39 (Figure 1.4); Deutsche Bundesbank, Ergebnisse der Gesamtwirtschaftlichen Finanzierungsrechnung fr Deutschland 1991 bis 2004 (Frankfurt 2005). 33 Cf Paul L Davies (n 10) 455; see also Stefan Grundmann and Peter O Mlbert, Corporate Governance: European Perspectives (2000) 2 ICCLJ 415, 421 (noting converging trends in Europe with regard to the role of shareholders in corporate governance). 34 See David Scharfstein, The Disciplinary Role of Takeovers (1988) 55 Rev. Econ. Stud. 185; John W Byrd and Kent A Hickman, Do Outside Directors Monitor Managers? (1992) 32 J. Finan. Econ. 195; Jean Tirole (n 32), 4351; Klaus Gugler, Corporate Governance and Performance: The Research Questions in Klaus Gugler (ed), Corporate Governance and Economic Performance (OUP, Oxford 2001) 3241. 35 Theodor Baums, Corporate Governance Systems in Europe Differences and Tendencies of Convergence (Working Paper No 37, University of Osnabrck/University of Frankfurt Working Paper Series, (1996) 9 <http://www.jura.uni-frankfurt.de/ifawz1/ baums/Bilder_und_Daten/Arbeitspapiere/a0896.pdf> (accessed 26 October 2006)); Stephen Prowse, Corporate Governance in an International Perspective (Bank for International Settlements, Basle 1994) 4650; Andreas Hackethal, Reinhard H Schmidt and Marcel Tyrell, Banks and German Corporate Governance: On the Way to a Capital Market-Based System? (2005) 13 Corporate Governance 397; see also Ekkehart Boehmer, Germany in Klaus Gugler (ed), Corporate Governance and Economic Performance (OUP, Oxford 2001) 108 (stressing how little is known about hostile takeover bids due to the lack of publication requirements for takeover negotiations).
ECFR 4/2006
435
Consequently, in Germany, the members of the supervisory board are the predominant monitors of the corporate management 36. This also holds true in the light of the remaining shareholder control mechanisms 37, since the means by which the shareholders can exercise control are different from those that the main bodies for corporate control can use. This study incorporates board turnover as an indicator of whether effective control is exercised. The shareholders have no direct means of influencing the replacement of the management. Thus, shareholder control mechanisms are not in the centre of this study 38. 2. The British system of corporate control (one-tier model) In the UK, only one single board exists 39. All board members, i.e. executive directors as well as non-executive directors, are normally elected by the shareholders. The shareholders also have the power to remove the directors from office (Sec. 303 Companies Act 1985). However, this remedy would be the last resort and is only used in cases of serious misconduct or extreme underperformance. Usually, the board takes decisions to reorganise the manage36 This even holds true in the light of the latest judgments of the German Federal Supreme Court, according to which the most important key decisions (eg, the sale of core business units, the investment in highly risky projects or the acquisition of other firms) need the approval of the shareholders in the general meeting; see BGHZ 83, 122 (Bundesgerichtshof, II ZR 174/80, 25 February 1983); BGHZ 159, 30 (Bundesgerichtshof, II ZR 155/02, 26 April 2004); for an overview see Holger Fleischer, Ungeschriebene Hauptversammlungszustndigkeiten im Aktienrecht (2004) 57 Neue Juristische Wochenschrift 2335. 37 See Paul L Davies, Institutional Investors as Corporate Monitors in the UK in Klaus J Hopt and Eddy Wymeersch (eds), Comparative Corporate Governance: Essays and Materials (Walter de Gruyter, Berlin 1997) 47; Alexander Mann, Corporate Governance Systeme (Duncker & Humbolt, Berlin 2003); see also Colin Mayer (n 9) 67 for the situation in the UK. 38 As this study concentrates on the largest companies only and as these companies are not subsidiaries, the influence of parent companies being responsible for reviewing the performance of directors can be disregarded. See Korn/Ferry International, European Boards of Directors Study (London 1996) 32, for the monitoring role of parent companies in Germany. 39 See Paul L Davies, Gower and Davies Principles of Modern Company Law (7th edn Sweet & Maxwell, London 2003) 316321 (also pointing out that the Companies Act 1985 does not explicitly address the question whether the board is to be a one-tier or a two-tier board); but see Jonathan Rickford, Fundamentals, Developments and Trends in British Company Law Some Wider Reflections; Second Part: Current British Priorities and Wider Reflections (2005) 2 ECFR 63, 7378 (on the legal necessity of a unitary board).
436
Carsten Jungmann
ECFR 4/2006
ment, and the shareholders, as well as the creditors, have only a limited capacity to exercise pressure on the board members and thereby indirectly affect the board composition40. An increasing pressure to reorganise the board is a reaction to the financial situation of the company, and demand to remove directors only arises with regard to those who are managing the companys affairs the board. According to legal provisions, it is the board that manages the company. Nowadays, however, virtually all boards consist of not only executive directors but also non-executive directors: according to the Combined Code on Corporate Governance (as amended in 2006), it is best practice that half of the board of larger companies should be comprised of non-executive directors. Therefore, the question arises as to how the role of an executive director differs from the role of a non-executive director. Non-executive directors are not employees of the company but rather members of the board and are, accordingly, concerned with managerial problems. It is indeed one of their functions to judge on strategy, key appointments and standards of conduct. The Combined Code meanwhile expressly states in the supporting principles to its Sec. A.1 that non-executive directors should constructively challenge and help develop proposals on strategy. However, it has become apparent in the corporate governance debate during the last years that there is a tendency towards control becoming their most important task 41. It was only a logical consequence of this development that not only the Cadbury Code but also the Combined Code refers to this role of non-executive directors. Sec. A.1 of the Combined Code assigns them, as becomes apparent in the supporting principles of this section, with the duty to scrutinise the performance of management in meeting agreed goals and objectives and monitor the reporting of performance. The first studies subsequent to the adoption of the Cadbury Code revealed that the disciplinary function of the board increases if the monitoring role of non-executive directors is strengthened 42. It is also apparent that even if managerial power is vested in the board, e.g. by the articles of association, the day-to-day business of the company is
40 Eils Ferran (n 12) 118119. 41 See generally Brian R Cheffins (n 21) 602652. See also Paul L Davies, Board of Directors: the European Perspective in Geoffrey Owen, Tom Kirchmaier and Jeremy Grant (eds), Corporate Governance in the US and Europe Where Are We Now? (Palgrave MacMillan, New York 2006) 42 (arguing that the Combined Code is to push the UK towards a de facto two-tier board system). 42 See Jay Dahya, John J McConnell and Nickolaos G Travlos, The Cadbury Committee, Corporate Performance, and Top Management Turnover (2002) 57 J. Finance 461; for a theoretical evaluation see, eg, Brian R Cheffins (n 21) 642652.
ECFR 4/2006
437
managed by individuals, not by the board as a whole, which meets once a month or even less frequently. Thus, most of the tasks required to run the business are executed by senior managers 43. They do not necessarily need to be board members, but in practice, this is often the case among the most senior members. It is, in any case, not the function of the non-executive directors to get involved in the day-to-day management of the company. In this respect, it is possible to differentiate non-executive directors from executive directors. Nonetheless, in contrast to the two-tier board system, there is no black or white distinction between the functions, neither between the separate organs nor within an organ of the company itself.
3. Groups of persons entrusted with monitoring tasks This final aspect will be significant to the discussion of the advantages and disadvantages of the British one-tier system. Since the management of the daily business affairs of the company does not fall within the responsibilities of the non-executive directors, it is the role of executive directors in the UK. This distinction is mirrored in Germany with the role of the members of the management board being identical to that of the executive directors, whilst the members of the supervisory board are the counterparts of the non-executive directors 44. Accordingly, this study concentrates, as far as board turnover is concerned, on the members of the management board and the executive directors, respectively. This is in accordance with the results of previous studies, which have demonstrated that supervisory board turnover or replacement of the non-executive directors cannot be regarded as being related to financial performance 45, with the exception of cases after periods of persistently poor performance, where there is some reaction 46. This reaction of the shareholders, however, does not concern the monitoring institution itself, but rather concerns a problem arising from the failure of the individuals who are entrusted with the monitoring tasks. Accordingly, turnover of non-executive directors and members of the supervisory board is not examined in this paper.
43 See Brian R Cheffins (n 21) 603604. 44 For a sound discussion of other internal and external mechanisms of corporate governance in the UK, see Alexander Mann (n 37). 45 Julian Franks, Colin Mayer and Luc Renneboog, Who Disciplines Management in Poorly Performing Companies? (2001) 10 J. Finan. Intermediation 209. 46 Julian Franks and Colin Mayer, Ownership and Control of German Corporations (2001) 14 Rev. Finan. Stud. 943, 965.
438
Carsten Jungmann
ECFR 4/2006
III. Analysis of the relationship between board turnover and financial performance Empirical studies on the relationship between board turnover and financial performance have been conducted in the past 47, but none have focused on the distinction between one-tier and two-tier systems. Hence, this paper is an attempt to fill the gap in available literature. Most of the former studies either failed to take Germany into account, or concentrated on the shareholder structure and differentiated between insider and outsider systems. 1. Earlier comparative research In the few existing comparative studies, Germany has been regarded as the prototype of an insider system 48, i.e. a system with a small number of quoted stock companies and concentrated share ownership, while the UK has been taken as a clear example of an outsider system 49, i.e. a system with large equity markets and dispersed ownership 50. Typically, this approach was taken to verify the theoretical explanation given by Shleifer and Vishny 51 who hypothesized that concentrated share ownership mitigates free-rider problems and is thus superior to dispersed ownership, as far as questions of effective corporate control are concerned. Goergen 52 was one of the first to undertake a quantitative study with regard to the markets in Germany and the UK. However, he concentrated his study on ownership and control in initial public offerings (IPOs), which makes his findings hard to generalise. About ten to fifteen years ago, the differences in the ownership structure between the UK and Germany, in terms of the concentration of shares owned by banks, non-financial corporations, and households, were significant 53. Several
47 For an overview see Udo C Brndle and Jrgen Noll (n 1) 13631365. Important information concerning board size, structure, composition and remuneration with regard to Germany and the UK can be found in Korn/Ferry International (n 38). 48 See Klaus Gugler, Dennis C Mller and B Burcin Yurtoglu, Corporate Governance and Globalization (2004) 20 Oxford Rev. Econ. Pol. 129, 130134. 49 See Rafael La Porta, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert W Vishny, Law and Finance (1998) 106 J. Polit. Economy 1113, 1145-1148; see also Jean Tirole (n 32), 5356. 50 These differences are sometimes still stressed today, see Udo C Brndle and Jrgen Noll (n 5). 51 Andrei Shleifer and Robert W Vishny, Large Shareholders and Corporate Control (1986) 94 J. Polit. Economy 461; see also Harold Demsetz and Kenneth Lehn, Large Shareholders and Corporate Control (1985) 93 J. Polit. Economy 1155. 52 Marc Goergen, Corporate Governance and Financial Performance A Study of German and UK Initial Public Offerings (Edward Elgar Publishing, Cheltenham 1998). 53 See, eg, Stephen Prowse (n 35) 3339; Julian Franks and Colin Mayer (n 46) 946952.
ECFR 4/2006
439
more recent studies have taken this approach: Franks, Mayer and Renneboog examined the relation between board turnover and, inter alia, corporate performance and addressed the question of how the UK system of corporate ownership relates to control 54. Franks and Mayer undertook a similar approach with regard to the ownership and control of German corporations 55. The results of these studies might have been diluted in the meantime. The remaining differences notwithstanding 56, the latest statistics clearly show a recent trend that the ownership structure in the UK is no longer as dispersed as it once was 57. Consequently, whilst a couple of years ago the differences in the shareholder structures might have been a valid explanation for the results of comparative studies on the effectiveness of corporate control, with respect to recent data for the UK and Germany, this is no longer the case 58. 2. Earlier research with relation to the effectiveness of corporate control Most of the other studies on the effectiveness of corporate control have been conducted solely based on the US market. Therefore, they analysed the Anglo-American one-tier system but did not contrast it with the dualistic model. Some used Tobins Q as a performance measure 59, whilst others more traditional means 60. However, they all focused on the composition of the
54 Julian Franks, Colin Mayer and Luc Renneboog (n 45). 55 Julian Franks and Colin Mayer (n 46). 56 See Luis Correia Da Silva, Marc Goergen and Luc Renneboog, Dividend Policy and Corporate Governance (OUP, Oxford 2004), 915; Colin Mayer (n 9) 14; Rafael La Porta, Florencio Lopez-de-Silanes and Andrei Shleifer, Corporate Ownership Around the World (1999) 54 J. Finance 471; Ekkehart Boehmer, Who Controls German Corporations? in Joseph A McCahery and others (eds), Corporate Governance Regimes (OUP, Oxford 2002) 268. 57 The percentage of UK shares owned by individuals/households dropped from 37.5 % in 1975 to 14.1 % in 2004 (Office for National Statistics (n 32) para 3.8), whereas between 70 % and 80 % of shares in UK listed companies are today held by institutional investors (Jonathan Rickford, Fundamentals, Developments and Trends in British Company Law Some Wider Reflections; First Part: Overview and the British Approach (2004) 1 ECFR 391, 411). 58 See also Andreas Hackethal, Reinhard H Schmidt and Marcel Tyrell (n 35); Reinhard H Schmidt, Corporate Governance in Germany: An Economic Perspective in Jan Pieter Krahnen and Reinhard H Schmidt (eds), The German Financial System (OUP, Oxford 2004) 386. 59 Robert Morck, Andrei Shleifer and Robert W Vishny, Management Ownership and Market Valuation: An Empirical Analysis (1988) 20 J. Finan. Econ. 293; Sanjai Bhagat and Bernard Black, The Uncertain Relationship between Board Composition and Firm Performance (1999) 54 Bus. Law. 921. 60 Benjamin E Hermalin and Michaels S Weisbach, The Effects of Board Composition and Direct Incentives on Firm Performance (1991) 20 Financial Management 101;
440
Carsten Jungmann
ECFR 4/2006
board and on the compensation of the board members. These studies could not find a significant relationship between the board composition and financial performance. These results can be contrasted with a study by Byrd and Hickman 61, who concluded the opposite. Indeed, in the American studies of the 1980s and the early 1990s, there is no clear line of evidence to verify whether outside directors are valuable or whether market forces have made them superfluous. With regard to the previously mentioned studies concerning the UK and Germany, we arrive at the widely accepted conclusion that a strong relationship exists between board turnover and corporate performance. This result is supported by earlier research by Coughlan and Schmidt 62, Warner, Watts and Wruck 63, Weisbach 64 and, very recently, Bresser, Valle Thiele, Biedermann and Ldeke 64a. These studies notwithstanding, Hypothesis 1, with its emphasis on the inverse correlation between board turnover and financial performance, can be afforded stronger supporting evidence.
3. The uniqueness of the focus on different board systems The above-mentioned studies comparing the UK and Germany are of limited use with regard to the key question examined in this paper, namely whether the monistic or the dualistic system is superior (Hypothesis 2). One of the reasons is that these studies stand in contrast to earlier theoretical approaches, which state that there is little association between concentration of ownership and managerial disciplining 65. This problem, however, can be dismissed in this paper, as due to the significant changes in the ownership structure mentioned above, ownership structure should no longer be the focus of research. Current studies should instead illustrate the advantages of the corporate governance systems and must therefore focus on the distinction between one-tier and two-tier systems. It is the concentration on this distinction here which makes the approach a novel one.
Hamid Mehran, Executive Compensation Structure, Ownership, and Firm Performance (1995) 38 J. Finan. Econ. 163. 61 John W Byrd and Kent A Hickman (n 34). 62 Anne T Coughlan and Ronald M Schmidt, Executive Compensation, Management Turnover, and Firm Performance (1985) 7 J. Acc. Econ. 43. 63 Jerold B Warner, Roll L Watts and Karen H Wruck, Stock Prices and Top Management Changes (1988) 20 J. Finan. Econ. 461. 64 Michael S Weisbach, Outside Directors and CEO Turnover (1988) 20 J.Finan. Econ. 431. 64a Rudi K F Bresser, Reynaldo Valle Thiele, Annette Biedermann and Holger Ldeke (n 22). 65 See Andrei Shleifer and Robert W Vishny (n 51).
ECFR 4/2006
441
One other recent study Lucier, Schuyt and Tse 66 is worth mentioning. This study compared the percentage of performance-related and total CEO succession rates in, inter alia, Germany and the UK. While a study concentrating solely on the CEO has various conceptual weaknesses, it did prove that the performance-related successions in both countries increased in the last three to four years 66a. However, since the study shows no evidence of the number of poorly performing companies, the increase of performancerelated successions does not, by itself, prove the effectiveness of the means of corporate control in question. 4. Methodology This paper examines the relationship between the board turnover and the financial performance of companies listed on the stock exchange in the United Kingdom and in Germany in the years 1994 to 2003. In each country, a sample of 25 companies was randomly chosen from those listed on the London Stock Exchange (FTSE 100) and on the Deutsche Brse (German Stock Exchange in Frankfurt) (DAX or M-DAX). Companies that were only temporarily listed during the examined period (except during the first years) were excluded from the sample, as were companies that were party to a major merger, since board restructuring in these companies was likely to be based on reasons other than financial performance. In addition, some companies were excluded due to a lack of information on composition of the board or on financial performance. This amounted to a total of 397 financial years under review. For each of these years, board turnover and financial performance were analysed. As board members need time to digest financial news, board changes that are related to prior performance take place in the following year. Thus, this study concentrates on board turnover following a year of poor financial performance. a) Board turnover Board turnover is defined as the number of executive directors who leave the board during the financial year, divided by the total number of board members at the beginning of the financial year. If the resignation of a board member was due to death or retirement, it was not included in this total.
66 Chuck Lucier, Rob Schuyt and Edward Tse, CEO Succession 2004: The Worlds Most Prominent Temp Workers (Report) (2005) <http://www.strategy-business.com/ media/file/sb39_05204.pdf> (accessed 26 October 2006). 66a But see Rudi K F Bresser, Reynaldo Valle Thiele, Annette Biedermann and Holger Ldeke (n 22) for the contrary result on the German market.
442
Carsten Jungmann
ECFR 4/2006
Ideally, this would ensure the measurement of only forced resignations. However, the annual reports state often that a board member has retired without it being apparent whether retirement was indeed the reason or whether it was a euphemistic expression. In this context, financial newspapers are normally also ineffective means of distinguishing between forced and unforced retirements 67. Thus, in this paper, resignation due to retirement is only excluded if the director is at least 65 years of age at the time he leaves the board. Choosing the age of 65 means that a more conservative approach is taken here in comparison with other studies which used 62 as the relevant age68. It is, however, consistent with the results of the empirical research by Warner, Watts and Wruck 69, according to which the average age of resignations from the board which were declared as retirements was 65 years. Weisbach 70 also chose 65 years as the relevant limit. It is pertinent to note that the data on board turnover was collected from the annual reports, which detail the number and the names of board members. In Germany, only changes in the members of the management board were examined, and with respect to the UK, the study concentrated solely on the turnover of executive directors. b) Poor financial performance There are numerous ways to measure financial performance, and it is thus difficult to form a decision as to whether the performance was good or poor in a particular year. Consequently, it is not surprising that past studies have concentrated on different performance measures: Warner, Watts and Wruck 71 report on an inverse relationship between the probability of board turnover and a firms share price performance. Weisbach 72 concentrated on stock returns and earnings changes as measures of US firms performance. Kaplan 73, in his
67 See Michael S Weisbach (n 64); but see Chuck Lucier, Rob Schuyt and Edward Tse (n 66) as well as Rudi K F Bresser, Reynaldo Valle Thiele, Annette Biedermann and Holger Ldeke (n 22) in contrast. 68 Julian Franks, Colin Mayer and Luc Renneboog (n 45) 213. 69 Jerold B Warner, Roll L Watts and Karen H Wruck (n 63). 70 Michael S Weisbach (n 64). 71 Jerold B Warner, Roll L Watts and Karen H Wruck (n 63). 72 Michael S Weisbach (n 64). 73 Steven N Kaplan, Top Executive Rewards and Firm Performance: A Comparison of Japan and the United States (1994) 102 J. Polit. Economy 510; Steven N Kaplan, Corporate Governance and Corporate Performance: A Comparison of Germany, Japan and the U.S. in Klaus J Hopt and Eddy Wymeersch (eds), Comparative Corporate Governance: Essays and Materials (Walter de Gruyter, Berlin 1997) 195 [hereinafter Corporate Governance and Corporate Performance].
ECFR 4/2006
443
studies on the Japanese and the US market, also examined earnings and stock returns, but added sales performance as an additional measure. According to Marsh 74, dividend cuts are powerful signals of bad news, both in respect to the current situation and future prospects. This is also the conclusion of a study on 6,000 UK dividend announcements in the years 1989 and 1992; Franks, Mayer and Renneboog report that the same result holds true for the US market 75. However, they also examined abnormal share price returns, earning losses, after-tax cash flow margins, and after tax rates of return on book equity. Franks and Mayer 76 chose earning losses, dividend cuts and abnormal returns as indicators for poor financial performance. Finally, Bresser, Valle Thiele, Biedermann and Ldeke 76a favoured a keep it simple approach and utilised three accounting measures of performance (earnings before interest and taxes to total assets, return on assets, and sales growth) and one market measure (total return to the firms shareholders). aa) Data for the measurement of financial performance In this study, four measures were taken as means of evidence for poor performance: loss for the financial year, loss of profits, abnormal negative share price performance and a dividend cut or omission. With regard to all of the above, the data collected refers to the financial years of the companies, i.e. not necessarily to the legal year ending December 31st. Whether there was a loss for the financial year was decided on the basis of the figures on net income for the financial year (i.e. profit after taxation and equity minority interests) provided in the annual reports. The loss of profits was calculated by a comparison of the net income in two consecutive years. During the time analysed in this study, new financial reporting standards were adopted. In addition, a number of the companies changed their accounting principles (e.g. from UK-GAAP to IAS) between 1999 and 2002. Furthermore, in some very rare circumstances, it became obvious that losses or losses of profits were only book losses, i.e. solely due to changing legal or accounting requirements (such as the activation of good will, patents, etc.) but not due to operational business. In all of these cases, the losses were not taken as an indicator for a year of poor performance. Finally, with regard to
74 Paul Marsh, Why Dividend Cuts are a Last Resort Financial Times (London 12 August 1992) 16; for a detailed analysis see Luis Correia Da Silva, Marc Goergen and Luc Renneboog (n 56). 75 Julian Franks, Colin Mayer and Luc Renneboog (n 45) 213. 76 Julian Franks and Colin Mayer (n 46) 959. 76a Rudi K F Bresser, Reynaldo Valle Thiele, Annette Biedermann and Holger Ldeke (n 22).
444
Carsten Jungmann
ECFR 4/2006
Germany, all data was expressed in EURO, i.e. converted from Deutsche Mark (DM) to EURO in the years prior to 1999. Share prices and dividend payments were also converted into EURO. In addition, conversions of the numbers printed in the annual reports were made: some of them were due to rights issues of equity shares, some due to changes in the nominal value of shares (frequently from 50 DM to 5 DM and/or to 1 EURO). In most cases, information concerning the dividends paid and the share price at the end of the financial year was contained in the annual reports. Alternatively, the Financial Times London and the Handelsblatt were used as data origins. A share price performance was considered to be abnormally negative if the change in the share price was less than 15 % of the change of the relevant sector index. Figures for the indices (normally sub-indices of the FTSE 350 or the M-DAX) were available in the Financial Times London (UK) and in the OnVista/IS Teledata database (Germany). bb) Definition of poor financial performance As Table 1 shows, the four measures chosen show different frequencies: a loss for the financial year occurred only in 3.52 %/5.88 % (Germany/UK) of the years, and there was a dividend cut or omission in 7.49 %/2.94 % of the years. By contrast, the incidence of a loss of profits and an abnormal negative share price performance was much higher, i.e. 27.31 %/38.82 % and 46.70 %/42.94 %.
Jurisdiction
Loss for the Financial Year
Loss of Profits
Abnormal Negative Share Price Performance 106 46.70 % 73 42.94 %
Dividend Cut or Omission
Poor Performance
Germany Total Percentage United Kingdom Total Percentage
8 3.52 % 10 5.88 %
62 27.31% 66 38.82 %
17 7.49% 5 2.94 %
44 19.38 % 36 21.18 %
Table 1: Measures of poor financial performance
For the purposes of this study, however, poor performance is defined as a financial year in which two out of the four indicators were positive, as opposed to one out of the four indicators. The reasons for this particularly restrictive approach are two-fold: firstly, it reconciles differences resulting from varying accounting and legal requirements as well as differences in
ECFR 4/2006
445
business behaviour. For example, Da Silva 77 found that dividend cuts are far more likely to occur in Germany than in the UK or the US. This is likely to be related to different business cultures, since the signalling role of a dividend cut in Germany is less strong than in the Anglo-American market. At the same time, German companies build up and reduce hidden reserves in order to smooth earnings figures 78, which is due to more liberal legal rules concerning the reporting of earnings. These observations are consistent with the data in Table 1: dividend cuts or omissions are more frequent in Germany (7.49 % compared to 2.94 %) whereas losses for the financial year and losses of profits occur more often in the UK (5.88 % compared to 3.52 %, and 38.82 % compared to 27.31 %, respectively). Secondly, the removal of directors is a severe form of exercising control and is in most cases the last resort. As the reason for the removal of a board member is rarely clearly stated, the restrictive approach taken in this study increases the probability that, in years of poor financial performance, the reasons underlying the board turnover rate are indeed related to poor performance. At the same time, this reduces the probability that reasons unrelated to poor performance had an influence on the turnover rate. This necessarily leads to a higher board turnover percentage in years of good performance, since board turnover figures which should actually be included in the statistics for years of poor financial performance, are, due to the restrictive approach of defining years of poor performance, erroneously included in the statistics for years of good performance. If, however, the data shows that the board turnover remains significantly higher in years of poor performance, despite the extra restriction, the results are even more convincing. Given the fact that it is a widely accepted view amongst commentators that a loss in a financial year and a dividend cut or omission are regarded as the most important indicators of poor performance, and notwithstanding different legal und cultural backgrounds, it is worth mentioning that the definition of poor performance used here does not diminish the importance of these two indicators: in the UK, 100 % of the years with a dividend cut or omission and 80 % of the years with a loss of profits are years of poor financial performance as defined in this study, and the corresponding numbers for Germany are 82 % and 100 %, respectively. Consequently, the two out of four approach used here respects the theoretically well-founded importance
77 Luis Correia Da Silva, Corporate Control and Financial Policy: An Investigation of the Dividend Policy of German Firms (PhD thesis, University of Oxford 1996); see also Luis Correia Da Silva, Marc Goergen and Luc Renneboog (n 56) 113123. 78 See Ray Ball, S P Kothari and Ashok Robin, The Effect of International Institutional Factors on Properties of Accounting Earnings (2000) 29 J. Acc. Econ. 1, 1516.
446
Carsten Jungmann
ECFR 4/2006
of losses and dividend cuts as performance measures but, at the same time, prevents loss of profits and abnormal share price performance from influencing the results beyond a legitimate degree. IV. Testing the hypotheses: data on financial performance and board turnover In this part, the results of the study are presented. The first apparent difference between the board structure in Germany and the UK concerns the average board size.
Germany Total Board Size (Average) 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Average 24.33 24.56 24.87 25.17 25.26 19.47 19.38 25.17 24.95 24.51 23.77 Management Board (Average) 6.77 7.00 6.96 6.91 6.65 6.90 6.02 6.39 6.17 5.73 6.55 Supervisory Board (Average) 17.56 17.56 17.91 18.26 18.61 18.78 18.78 18.78 18.78 18.78 18.38 United Kingdom Board Size (Average) 12.00 12.44 12.50 12.22 12.17 11.44 11.78 11.50 11.83 11.40 11.93
Table 2: Board size in Germany and the UK (19942003)
To test Hypothesis 1, i.e. the inverse correlation of board turnover and financial performance, the average board turnover in years following a year that did not fall into the definition of poor performance had to be compared to years following a year that did. As can be seen from Table 3, there were 44 years characterized by poor performance in Germany and 36 years in the UK.
Jurisdiction Germany UK Total Years of Poor Performance 44 36 80 Other Years 183 134 317 Total 227 170 397
Table 3: Years with and without poor financial performance in Germany and the UK (19942003)
ECFR 4/2006
447
In Table 4, the average board turnover percentage in years following a year of poor financial performance is contrasted with the board turnover percentage in other years. In the years that did not fall within the definition of a year of poor performance, in Germany, 8.88 % members on average left the management board for reasons other than death or retirement, and in the UK an average of 11.71 % of the executive directors abandoned the board. In both countries, there was an increase in the average board turnover in years of poor performance: in Germany, it rose by 3.30 percentage points to 12.18 %, and in the UK, there was an increase by 3.73 percentage points to 15.44 %.
Jurisdiction Performance in the Financial Year Not Poor Poor Not Poor Poor Board Turnover s s n 0.87
Germany
8.88 % 12.18 % 11.71 % 15.44 %
11.81
United Kingdom
13.90
1.21
Table 4: Relation of board turnover and financial performance in Germany and the UK (19942003)
The sample size for the years that did not fall into the category of poor performance is nGermany = 183 and nUK = 134, respectively. That means that the sample size is large enough so that the sampling distribution of the mean can be approximated by the normal distribution. As the standard deviation of the sampling distribution is sGermany = 11.81 and sUK = 13.90, respectively, the increase is statistically significant at the 99 % level of confidence in both countries. Thus, the board turnover was substantially higher in years following a year of poor performance than in other years, and accordingly, the data supports Hypothesis 1, underscoring the notion that board turnover is inversely correlated with financial performance. Hence, both systems, the onetier and the two-tier system, are effective means of control, and non-executive directors, as well as the members of the supervisory board, attend to their duty of removing incompetent directors. This result stands in contrast to other papers 79 and consequently offers significant new evidence capable of influencing the discussion on questions of corporate governance and the reforms of the relevant statutory provisions.
79 See, eg, Stefan Grundmann and Peter O Mlbert (n 33) 421 (arguing that in both systems the control function is deficient); cf n 62 to n 64.
448
Carsten Jungmann
ECFR 4/2006
The comparison of the differences in the board turnover between the years following a year of poor performance and other years allows us to comment on the validity of Hypothesis 2: this hypothesis has also to be accepted as the data contained in Table 4 does not allow us to judge the superiority of either the monistic or the dualistic system. Both systems show an inverse correlation between the financial performance and board turnover; however, this correlation is so statistically significant that it is impossible to come to the conclusion that one system would be more effective than the other. This is due to the fact that the difference between the degrees of effectiveness between the two systems is too small.
V. Strengths and weaknesses of the concurring board models According to the data presented in the last part, it is hardly possible to deem one board system superior to the other. This result is, intuitively, not very surprising as both systems have existed for more than one hundred years 80. Metallgesellschaft or Philipp Holzmann on the one hand and Bank of Commerce and Credit International (BCCI), Eurotunnel or Barings Bank on the other are well known examples of financial scandals and cases of evident failures of corporate control in Germany and in the UK. Surprisingly, the changes made in the statutes governing company law were rather conservative in both countries and, indeed, one of the concurring board models was never replaced by the other. However, this fact alone does not prove the equal adequacy of the systems. According to economic logic we should expect the more effective system to prevail over time and thus we should see tendencies to converge. But the development of statutory provisions is always driven by structure (path dependence) 81. Accordingly, the strengths and weaknesses of a system must always be seen in the context of both business and legal environments. Legal systems develop in accordance with their historical, societal and even cultural roots82. This view is not limited to models of corporate governance but is
80 Cf Jonathan Charkham, Keeping Good Company: A Study of Corporate Governance in Five Countries (OUP, Oxford 1995) 349366; Udo C Brndle and Jrgen Noll (n 5) 1. 81 Lucian Bebchuk and Mark J Roe (n 16); Reinhard H Schmidt and Gerald Spindler, Path Dependence, Corporate Governance and Complementarity (2002) 5 International Finance 311; Mark J Roe, Path Dependence, Political Options, and Governance Systems in Klaus J Hopt and Eddy Wymeersch (eds), Comparative Corporate Governance: Essays and Materials (Walter de Gruyter, Berlin 1997) 165. 82 Klaus J Hopt (n 4) 183184; Klaus J Hopt and Patrick C Leyens (n 5) 161162; see also Reinhard H. Schmidt and Gerald Spindler (n 4) 93.
ECFR 4/2006
449
also true with regard to, e.g., accounting standards, taxation principles, etc. Hence, only exceptional circumstances will lead to the adoption of a completely different model83. Thus, it will often be more advantageous to explore ways to make each system optimal within its own legal, historic and cultural constraints 84. As this is also the more realistic way to reach an enhanced level of corporate control, it is worthwhile to examine the advantages and disadvantages of the two systems. Consequently, a comparison will allow for an analysis of the possibilities to improve each system individually as a means of corporate control, by adopting new measures and by systematically eliminating structural weaknesses. The conclusion drawn in part VI depends therefore on the discussion of the strengths and weaknesses of both systems in this part.
1. Advantages and disadvantages of the two-tier model This discussion of the advantages and disadvantages of the dualistic system with two separate boards refers to the present state of legal rules applicable to corporate control. The key advantage of the two-tier system, however, needs to be seen in the context of its historic origins.
a) Separation of control and management as the key advantage Historically, the core idea behind the two-tier model was to strictly separate the controlling institution from the managing institution. When implementing the laws governing the stock corporation in Germany in the late 19 th century, it was the legislators intention to protect both shareholders and the public interest 85. This strict separation of control and managerial tasks was regarded as one of the major advantages of the two-tier system 86, and this argument
83 Mark J Roe, Chaos and Evolution in Law and Economics (1996) 109 Harv. L. Rev. 641, 663665; see also Reinhard H Schmidt and Gerald Spindler, Path Dependence and Complementarity in Corporate Governance in Jeffrey N Gordon and Mark J Roe (eds), Convergence and Persistence in Corporate Governance (CUP, Cambridge 2004) 114, 116. 84 See Stephen Prowse (n 35) 5556; but cf. Reinhard H. Schmidt and Gerald Spindler (n 4) 90. 85 Klaus J Hopt (n 1) 230231. 86 See, eg, Udo C Brndle and Jrgen Noll (n 1) 13591360; Uwe H Schneider, Die Revision der OECD Principles of Corporate Governance 2004 (2004) 49 Die Aktiengesellschaft 429, 432; cf Paul L Davies (n 10) 436.
450
Carsten Jungmann
ECFR 4/2006
has also been highlighted in the recent British discussions on the reform of company law 87. While this statement still appears to be true in general, one has to observe the impact that case law had in the last ten years on the functions of the supervisory board. As mentioned above, at present, control is not purely backward-oriented but requires an open discussion between the supervisory board and management board and also their close and trustful cooperation. Notwithstanding this development, at least theoretically, conflicts of interests between the members of the two boards should not arise: members of the supervisory board are elected by the shareholders in the general meeting. Accordingly, supervisory board members increase the likelihood of getting re-elected by fulfilling their most important task, i.e. to monitor the management board adequately and thoroughly. At the same time, the members of the management board are elected by the supervisory board and will remain in office only if the company performs well. In practice, however, the members of the supervisory board are chosen by the management board and are only formally elected in the general meeting. It is unusual for persons to become members of the supervisory board who have not been pre-selected by the management board 88. This has two consequences: only persons that are adequate in the opinion of the management board become elected as members of the supervisory board, and only those who adequately control in the eyes of the management board remain in office. In both cases, the definition of adequacy by the management board does not necessarily correspond with the shareholders definition of adequacy. Consequently, the members of the supervisory board are not completely independent when they exercise control. This becomes strikingly evident in the cases in which members of the management board (often the chairman) switch to the supervisory board (and often become chairman of this organ) after retirement 89, which was, and sometimes still is, usual in practice. While the separation of management and control the key advantage of the two-tier system somewhat dilutes the independence of the members of the supervisory board, this separation remains a strength with respect to the
87 Department of Trade and Industry (n 12) paras 3.1373.141. 88 Klaus J Hopt (n 1) 250; Patrick C Leyens (n 5) 81; see Rolf Birk (n 4) 65. 89 Bundesverband der deutschen Industrie eV and PwC Deutsche Revision AG, Corporate Governance in Germany (BDI-Drucksache No 322, August 2002) 33; Klaus J Hopt (n 1) 243244; Patrick C Leyens (n 5) 81; Udo C Brndle and Jrgen Noll (n 1) 1358. According to Korn/Ferry International (n 38) 12, 43 % of the members of the supervisory board are former members of the management board.
ECFR 4/2006
451
management board 90. Each member of the management board has the same task, namely to run the business in a way that allows further development and financial prosperity. In order to perform this task, every information is shared among all board members and is not to the exclusive knowledge of some board members. Accordingly, the decision-making process within the management board is swift and efficient.
b) The need to distinguish between structural weaknesses and problems arising from co-determination There are also a number of disadvantages to the two-tier system. With regard to Germany, a thorough review must distinguish between those weaknesses derived from the mere fact of having two separate organs of the company (structural weaknesses) and those owed to a German particularity, namely its co-determination statutes 91. The latter strongly influences the composition of the supervisory board: the circle of eligible persons, the responsibilities and the primary interests of its members, possible expert knowledge requirements, etc. clearly depend upon the co-determination laws. The distinction between structural weaknesses and problems arising from codetermination is not only necessary for a sound analysis of the strengths and weaknesses of the two-tier model, but it also helps to objectively assess the value of the two-tier model and also its potential in countries that are not governed by co-determination laws. For example, even European Companies (SE) based in the UK might opt for the two-tier system. For these companies, only the arguments regarding the structural weaknesses of the two-tier system will be relevant. Finally, the above-mentioned distinction is important for the discussion of potential improvements to the one-tier system which could adopt single features of the two-tier system.
90 Cf Klaus J Hopt (n 4) 178 (mentioning that the German understanding of independence differs slightly from the British as it is common for members of the supervisory board to be representatives of large shareholders); Alice Belcher and Till Naruisch, The Evolution of Business Knowledge in the Context of Unitary and Two-tier Board Structures [2005] JBL 443 (concluding that the German two-tier structure is likely to encourage more teamwork and thus more likely to create a competitive advantage than the UK one-tier structure). 91 See Peter Bckli, Konvergenz: Annherung des monistischen und des dualistischen Fhrungs- und Aufsichtssystems in Hommelhoff and others (eds), Handbuch Corporate Governance (Otto Schmidt-Verlag, Kln 2003) 201, 205208; Patrick C Leyens (n 5) 98; Lewis Robertson (n 11) 4.
452
Carsten Jungmann
ECFR 4/2006
aa) Structural weaknesses The key advantage of the two-tier system, i.e. the independence of the members of the supervisory board due to the separation of the control and management, is at the same time the reason for a number of structural weaknesses. At least traditionally, the members of the supervisory board, by definition, do not make strategic management decisions of their own, and with the exception that decisions of the management board need the approval of the supervisory board they are not involved in the decisionmaking process at all 92. They are not informed at the outset, nor are they asked for their opinion or advice. Instead, their role is limited to evaluating measures already taken by the management board. Consequently, the way in which the supervisory board exercises its control is always reactive and never active 93. This necessarily leads to a decrease of the quality of control: at the outset, there are normally a number of possible solutions to a strategic management problem. In a two-tier system, the supervisory board is restricted to commenting on the chosen solution in hindsight 94. It might criticise the choice and even go so far as to remove management. However, this does not enhance the current status of the company, but rather only reduces the possibility that the same mistake will be made again. If, on the contrary, the supervisory board had had the opportunity to exert influence on the decision ex ante, a better solution may have been found. The amendments to the German company law provisions by the Control and Transparency in the Corporate Sector Act of 1998 95 and by the Transparency and Publicity Act of 2002 96 were a step in the right direction. Since then, it has become an additional task of the management board to inform the supervisory board on intended business policy and corporate planning. As the words intended and planning indicate, the supervisory board should increase its involvement in strategic decisions. German courts have highlighted this additional tasks 97 and have, as mentioned above, thus initiated the process of transforming the supervisory board from a pure review body to a consultant of the management.
92 Marcus Lutter, Das dualistische System der Unternehmensverwaltung in Eberhard Scheffler (ed), Corporate Governance (Gabler Verlag, Wiesbaden 1995) 5, 18. 93 See Peter Bckli (n 91) 213214. 94 See Patrick C Leyens (n 5) 9394; Carsten Berrar (n 24) 21812183. 95 Gesetz zur Kontrolle und Transparenz im Unternehmensbereich (KonTraG) (27 April 1998, BGBl I 786). 96 Transparenz- und Publizittsgesetz (TransPuG) (19 July 2002, BGBl I 2010). 97 See Hartwig Henze, Neuere Rechtsprechung zu Rechtsstellung und Aufgaben des Aufsichtsrats (2005) 58 Der Betrieb 165.
ECFR 4/2006
453
The need for information is another weak point in the two-tier system of control. This stems from the supervisory boards non-involvement in the decision-making process and the fact that its members are not present at the meetings of the management board. In order to be able to judge on the performance of the management board, the supervisory board needs adequate disclosure of information. However, there is a strong information asymmetry between the two boards 98, since all information concerning questions of strategy, future projects, business opportunities, budgetary questions, etc. lies in the hands of the management board. The members of the management board are in direct contact with employees and junior management. They obtain all information unaltered and unfiltered. By contrast, the members of the supervisory board receive all their information from the management board. They are not permitted to collect information on their own 99, and employees are not obligated to report to the supervisory board directly. Thus, the access of the members of the supervisory board to sensitive information is limited 100. Theoretically, the situation described above should not contribute to information asymmetry since, according to Sec. 90 of the German Stock Corporation Act of 1965, the management board must regularly report to the supervisory board. This provision was changed by the Control and Transparency in the Corporate Sector Act of 1998, which led to an improvement in the flow of information 101: disclosure must be comprehensive and punctual, and the reports must contain all information needed to evaluate the work of the management board, meaning that they must include information concerning future projects and strategy. In addition, the management board must not withhold information from members of the supervisory board which is deemed to be delicate. In reality, however, we must note that the supervisory board would not be able to perform its tasks if the relevant files were not prepared, the information was not processed and the data was not conceptualised prior to the meetings of the supervisory board. This becomes evident if we consider
98 Bundesverband der deutschen Industrie eV and PwC Deutsche Revision AG (n 89) 17; Department of Trade and Industry (n 12) paras 3.1403.141; Brian R Cheffins (n 21) 619. 99 Dieter Feddersen, berwachung durch den Aufsichtsrat in Hommelhoff and others (eds), Handbuch Corporate Governance (Otto Schmidt-Verlag, Kln 2003) 441, 458 460. 100 See Geoffrey Owen (n 11) 17. 101 See Marcus Lutter, Comparative Corporate Governance: A German Perspective (2000) 2 ICCLJ 423, 426427.
454
Carsten Jungmann
ECFR 4/2006
the average working hours of the supervisory board: prior to 1998, it was common for the supervisory board to meet only 3.8 times a year on average. Each meeting lasted less than four hours, namely 3.74 hours on average 102. This amounted to a total working time of 14 hours and 13 minutes per year. Since the amendment of the statutory provisions in 1998, it has become mandatory for the supervisory board to meet at least two times in each half of the year. This sets a minimum standard, but can not be regarded as a major increase in the frequency of meetings 103. Hence, it is still impossible to reach decisions in such a short time if the members of the supervisory board are not provided with the data that is already processed. Consequently, the members of the supervisory board rely on reports which contain all information concerning the company and the managements strategy. As these reports are prepared on behalf of the management board, the information is necessarily filtered. The aspects that appear to be important in the eyes of the management board are stressed, while others are deemphasised or even left out 104. This has the following consequence: the management board members make decisions based on information which they themselves regard as important. However, a thorough control requires full transparency in order for one to be able to properly evaluate the quality of the managements decision-making process. In a two-tier system, with its typical flow of information, there is a danger that the supervisory board cannot discover deficiencies due to a lack of information. In evaluating the decisions of the management board, the supervisory board might, consequently, repeat the management boards mistakes. This is not a question of bad faith or that a management board was not willing to provide relevant information. It is rather due to the fact that the management transfers only the information that it relied upon itself 105. Furthermore, a meeting frequency of about four times a year confines the supervisory board to its traditional but weak position, in which it exercises mere ex post control. More continual monitoring and more frequent meetings are needed in order to adequately process the high amount of data 106.
102 Klaus J Hopt, The German Two-Tier Board (Aufsichtsrat): A German View on Corporate Governance in Klaus J Hopt and Eddy Wymeersch (eds), Comparative Corporate Governance: Essays and Materials (Walter de Gruyter, Berlin 1997) 9; for older data see Knut Bleicher and Herbert Paul (n 15) 273274. 103 More recent studies on the duration and the frequency of meetings of the supervisory board are not available. 104 See Brian R Cheffins (n 21) 611. 105 Cf Brian R Cheffins (n 21) 619. 106 See Marcus Lutter (n 101) 428; Peter Bckli (n 91) 214218; see also Erich Potthoff (n 6) 265.
ECFR 4/2006
455
bb) Problems arising from co-determination In addition to the structural deficiencies of the two-tier system, there are problems originating solely from the German laws of co-determination 107. As mentioned above, up to 50 % of the members of the supervisory board are labour representatives. According to the Co-Determination Act of 1976, as amended, one-half of the supervisory board must, in companies with 2,000 or more employees, be composed of labour representatives, and this 50 % rule was applicable to all companies under review in this study. In the supervisory board, the labour representatives and the shareholder representatives have equal rights and duties with the exception that the vote of the chairman, who is never a labour representative, counts twice (casting vote) in the event of a deadlock in the supervisory board. Before turning to the disadvantages of labour co-determination with regard to corporate governance and the ability of the supervisory board to control management, it is worth glancing at the historical roots and the benefits of co-determination. The first laws governing co-determination were reactions to the economically and socially difficult times after the wars in the first half of the twentieth century and reflected the significant role of the coal and steel industry in Germany at that time. As todays working environment is much more developed and governed by other laws, co-determination does not appear to be as crucial now as it was sixty or eighty years ago. It still has its merits, however, especially as an early warning system for social conflicts, thereby reducing the probability of strikes 108. Furthermore, management accommodated co-determination, which is capable of shielding companies from hostile takeovers, somewhat gratefully 109. One must keep all these aspects in mind when evaluating the German two-tier system and when subsequently mentioning the disadvantages of co-determination with respect to corporate governance. The first of the weaknesses is related to the information asymmetry mentioned above. It is apparent that the shareholder representatives on the supervisory board are reluctant to discuss certain highly confidential issues with labour representatives. In countries with co-determined supervisory boards, this has the practical consequence that certain issues are not discussed with the supervisory board as a whole, although the laws provide for the contrary.
107 See Thomas Raiser (n 8) for a fundamental review and for an analysis of the future of co-determination laws. 108 See Klaus J Hopt and Patrick C Leyens (n 5) 145. 109 Theodor Baums (n 35) 12; see Thomas Raiser (n 8) B 4957.
456
Carsten Jungmann
ECFR 4/2006
A second major disadvantage arising solely from co-determination laws is the size of the supervisory board. As can be seen from Table 2, the average size of the supervisory board of the companies examined in this study was 18.38 members. These numbers are not surprising, as the Co-Determination Act of 1976 stipulates that it is a requirement for the supervisory board of companies with more than 20,000 employees to have 20 seats. The prevalence of these larger seat numbers becomes even less surprising in light of the fact that the relevant provision of the Co-Determination Act of 1976 setting the seat requirement was permanently applicable for 69.56 % of the companies under review in this study 110. It is common knowledge that a supervisory organ of this size must be inefficient 111. Nevertheless, the proposal to lower the size to a maximum of twelve seats which was discussed prior to the enactment of the Control and Transparency in the Corporate Sector Act of 1998 never became law 112. Assuming that the average time of a supervisory board meeting is 3.74 hours, each member of a supervisory board with 20 members has the theoretical opportunity to speak for less than 12 minutes in a single meeting. In practice, however, the presentation of reports and other formalities are time consuming 113 and, the chairman will speak most of the remaining time. Accordingly, most members of the supervisory board do not have the opportunity to speak. There exists a third shortcoming to the two-tier system governed by codetermination laws: co-determination requirements complicate the introduction of mandatory qualification standards for all members of the supervisory board 114. The installation of such standards is suggested in the OECD Prin110 Cf Korn/Ferry International (n 38) 10 (reporting an average of 13.25 members for a sample of 59 companies, which, however, also includes smaller companies than in this study). 111 See, eg, Bundesverband der deutschen Industrie eV and PwC Deutsche Revision AG (n 89) 3334; Thomas Raiser (n 8) B 5960; Steven N Kaplan, Corporate Governance and Corporate Performance (n 73) 198; Martin Lipton and Jay W Lorsch, A Modest Proposal for Corporate Governance (1992) 48 Bus. Law. 59, 65; Eberhard Schwark, Corporate Governance: Vorstand und Aufsichtsrat in Peter Hommelhoff and others (eds), Corporate Governance (Verlag Recht und Wirtschaft, Heidelberg 2002), 75, 102 103; Manuel R Theisen (n 7) 261262; Patrick C Leyens (n 5) 8182; see also Erich Potthoff (n 6) 261; see finally Korn/Ferry International (n 38) 8 (reporting on the self-assessment of German supervisory boards). 112 See Klaus J Hopt (n 4) 177. 113 Martin Lipton and Jay W Lorsch (n 111) 64. 114 Klaus J Hopt, Unternehmensfhrung, Unternehmenskontrolle, Modernisierung des Aktienrechts Zum Bericht der Regierungskommission Corporate Governance in Peter Hommelhoff and others (eds), Corporate Governance (Verlag Recht und Wirtschaft, Heidelberg 2002), 27, 4445.
ECFR 4/2006
457
ciples on Corporate Governance of 2004 115, and recent studies highlight the importance of qualification and experience for the choice of the members of the supervisory board 116. In contrast, Point 11.1 of the 2005 Recommendations of the EU 117 as well as the German Corporate Governance Code take a rather diffident approach. Sec. 5.4.1 of the German Corporate Governance Code states that members need to have the required knowledge, abilities and expert experience to properly complete their tasks. At least every individual member of the supervisory board needs to have a certain degree of knowledge, whereas the corresponding provisions of the German Stock Corporation Act of 1965, as interpreted by the courts, only require that the supervisory board as a whole has this level of knowledge 118. The introduction of even advanced mandatory qualification standards would certainly not bar shareholders from finding adequately qualified representatives. Such standards would instead be regarded as a hindrance to employees which prevent them from freely choosing their representatives. Without common standards concerning qualification and professional experience, however, it is extremely difficult to ensure the quality of work performed by the members of the supervisory board. The ex post review of decisions by the management board is already a difficult task. It is unlikely, however, that an organ consisting of members with little or no expertise in economics, finance, corporate strategy, etc. would be able to enhance the quality of the decisions made by the highly paid, carefully selected and well-educated members of the management board. Thus, it is even more challenging to involve the supervisory board more deeply in the decision-making process.
cc) Other uniquely German characteristics The consequences of the German co-determination laws and the influence of labour representatives are not the only particularities influencing the standards of corporate governance and the effectiveness of the control by the supervisory board. The tight-knit relationship between investors and lenders also affects the manner in which control is exercised over the company and its managers. The influence of German banks on the composition of the super-
115 For an overview see Uwe H Schneider (n 86). 116 See Rdiger von Rosen (ed), Wertorientierte berwachung durch den Aufsichtsrat (Deutsches Aktieninstitut, Frankfurt 2005). 117 See n 17. 118 See Uwe Hffer, Die Unabhngigkeit von Aufsichtsratsmitgliedern nach Ziffer 5.4.2 DCGK (2006) 27 Zeitschrift fr Wirtschaftsrecht 637, 638.
458
Carsten Jungmann
ECFR 4/2006
visory board is immense 119. About 12 % of the shareholder seats in supervisory boards are taken by members of private banks 120. At first glance, this number might not appear to be terribly significant. However, this 12 % represents only part of the influence that private banks have. Their total influence is derived from the aforementioned seats in the supervisory boards, from bank proxy votes 121, and notwithstanding a slight trend to the contrary from their still significant blocks of shares of the largest German stock corporations 122. The role of financial institutions in Germany is certainly not the sole explanation for the quality of corporate control. Neither is it directly linked to the two-tier system. However, any evaluation of the two-tier system as a means of effective corporate governance and control must take this German feature into account. This means that the adoption of this system in other countries, or the introduction of elements of this system by way of converging the one-tier system and the two-tier system, might lead to other results than in Germany.
2. Advantages and disadvantages of the one-tier model A number of the advantages of the one-tier model corresponds with the weaknesses of the two-tier model and vice versa. The one-tier model in the United Kingdom does not face comparable national particularities, the way
119 Klaus J Hopt (n 4) 186188; for details see Stefan Prigge, A Survey of German Corporate Governance in Klaus J Hopt and others (eds), Comparative Corporate Governance: The State of the Art and Emerging Research (OUP, Oxford 1998) 943; Jeremy Edwards and Klaus Fischer, Banks, Finance and Investment in Germany (CUP, Cambridge 1994). 120 Klaus J Hopt (n 102) 11 (citing a study of 1995); see also Jeremy Edwards and Klaus Fischer (n 119) 207210 (analysing the relationship between proxy voting power and supervisory board representation). 121 See Theodor Baums (n 30) 505-508; Theodor Baums, Corporate Governance in Germany System and Current Developments (Working Paper No 70, University of Osnabrck/University of Frankfurt Working Paper Series, 1999) 12 <http://www jura.uni-frankfurt.de/ifawz1/baums/Bilder_und_Daten/Arbeitspapiere/paper70.pdf> (accessed 26 October 2006); Oliver Rieckers and Gerald Spindler, Corporate Governance: Legal Aspects in Jan Pieter Krahnen and Reinhard H Schmidt (eds), The German Financial System (OUP, Oxford 2004) 350, 377379; Luis Correia Da Silva, Marc Goergen and Luc Renneboog (n 56) 13; for empirical evidence see Marco Brecht and Ekkehart Boehmer, Ownership and Voting Power in Germany in Fabrizio Barca and Marco Brecht (eds), The Control of Corporate Europe (OUP, Oxford 2001) 128. 122 For recent data, see Deutsche Bundesbank (n 32).
ECFR 4/2006
459
that the two-tier model in Germany does. Thus, the arguments made in favour of and against the British one-tier model might appear to be more universal than the remarks on Germanys system of corporate governance. Nevertheless, the following comments refer to the United Kingdom, unless the contrary is explicitly stated. a) Structural strengths of the one-tier model It is the board that leads and controls the company. Thus, all members of the board are entrusted with the same tasks and are obliged to perform the same duties 123. Consequently, both the responsibilities of monitoring and strategy-setting are incorporated into the same body 124. Despite the fact that some members of the board are executive directors, while others are nonexecutive directors, they all have direct access to the same information. At least in theory, the flow of information is a vast improvement to that of the two-tier model 125. As the non-executive directors are involved in the decision-making process, they have more incentives to supply themselves with all relevant information, since they cannot argue afterwards that they were limited to an ex post control of decisions made by other persons. At the same time, the decision-making process is swifter with regard to farreaching questions. In a two-tier system, there is typically, but not necessarily, the need for the management board to receive the approval of the supervisory board. This approval can often be obtained only after a couple of weeks, due to the low frequency of meetings of the supervisory board. By contrast, in the one-tier system, board meetings take place more regularly 126. This higher frequency of meetings has, simultaneously, an impact on the understanding of the board members of the business, as it increases their knowledge of day-to-day business 127. Meanwhile, the knowledge of the nonexecutive directors helps the board as a whole, which demonstrates the potential for better informed decisions in one-tier systems. This holds especially true with regard to strategy formation 128.
123 Committee on the Financial Aspects of Corporate Governance (n 4) 20 (para 4.3); for a critical view see Owen Green (n 11) 19. 124 See Jonathan Rickford (n 39) 75. 125 Paul L Davies (n 10) 448; Udo C Brndle and Jrgen Noll (n 1) 1358; Brian R Cheffins (n 21) 615 (pointing out other ways for non-executive directors to obtain information); but see Eils Ferran (n 12) 221223 (discussing the remaining difficulties with regard to access to management information). 126 See Knut Bleicher and Herbert Paul (n 15) 273. 127 Geoffrey Owen (n 11) 16; see also Martin Lipton and Jay W Lorsch (n 111) 65. 128 See Department of Trade and Industry (n 12) para 3.140.
460
Carsten Jungmann
ECFR 4/2006
These advantages are certainly typical for the one-tier model with its unitary board. In practice, however, it is doubtful whether the flow of information is as perfect as in theory. It can be observed that the information available to the non-executive directors is prepared by the management 129. Consequently, non-executive directors face the problems of obtaining independent and unfiltered information that are comparable to those of the members of the supervisory board in the two-tier system. In the two-tier system, this leads to an information asymmetry between the two boards; in the one-tier system, there is an information asymmetry within the board. The non-executive directors might not ask for additional facts, numbers, figures, etc. if they are confronted with the well-informed executive directors in the board meeting, while in the meetings of the supervisory board, all board members face the problem of insufficient or lacking information. A single non-executive director or a group of non-executive directors in a meeting of the board may therefore be reluctant to ask for further information 130, whereas the members of the supervisory board might, collectively, be less reluctant. These observations notwithstanding, non-executive directors have, contrary to members of the supervisory board, direct access to information. This is a clear advantage of the unitary system. In this system it is, in fact, rather more a question of filling the board with capable non-executive directors than a structural problem of guaranteeing an adequate flow of information 131.
b) Weaknesses of the one-tier model In this paper, it was argued that the separation of control and management is the key advantage of the two-tier system. By contrast, the members of the unitary board fulfil both managerial and supervisory roles. Thus, they face a dilemma: they should make decisions and, at the same time, monitor these decisions. While this problem does not exist in the two-tier system due to the inherent formal separation of control and management, it is necessary to obtain this separation artificially in the one-tier system. The mere fact that there are executive and non-executive directors is not sufficient to guarantee the adequate execution of the monitoring role of the board. Non-executive directors are, as a survey confirms, well aware of their strategic role but less
129 Shann Turnbull, Why Unitary Boards Are Not Best Practice: A Case for Compound Boards (2000) 34 <http://ssrn.com/abstract=253803> (accessed 26 October 2006). 130 See Martin Lipton and Jay W Lorsch (n 111) 6566. 131 See Patrick C Leyens (n 5) 9193.
ECFR 4/2006
461
so of their monitoring role 132. It has also been stated that non-executive directors are not active in disciplining mismanagement in the UK 133. The results of this study, however, cast a shadow of doubt over the accuracy of the previous statements. Nonetheless, there is a need for some board members to be neutral and to concentrate their efforts primarily on the monitoring task. This has led to a further class of board members: within the group of the non-executive directors, only some are deemed independent. Emphasising the importance of having independent directors on the board and narrowing the definition of independence were core elements of the debate on the reform of corporate governance in the UK 134. According to Sec. A.3.1 of the Combined Code, a director is considered to be independent if there exist no relationships or circumstances which are likely to affect, or could appear to affect, the directors judgment. Independence is deemed to be a necessary precondition for the ability to handle the combination of two tasks in practice. It thus remains a problem of the one-tier system to find ways to guarantee that a certain number of board members are independent and to name criteria for independence. The independent non-executive directors face the dilemma of being colleagues with the other board members but also having to monitor them at the same time 135. This monitor-colleague-dilemma could be diminished, however, by clarifying the roles and responsibilities of each director. It is mainly the responsibility of the chairman to hold meetings in an environment in which there is a clear understanding of the different tasks of the board members and in which problems and questions can be discussed frankly and openly. In addition, it is the task of the shareholders to find suitable persons to fill the board 136. The non-executive directors must have a breadth of experience in order to understand the business and the sufficient confidence
132 See Department of Trade and Industry (n 12) para 3.134; but cf Brian R Cheffins (n 21) 604605; Jonathan Rickford (n 39) 75 (stating that the great majority of board activity in typical British public company boards is taken up with supervising and monitoring individual executive members and more junior managers); see also Korn/Ferry International (n 38) 26. 133 Colin Mayer (n 9) 8; see also Julian Franks, Colin Mayer and Luc Renneboog (n 45) (stating that non-executive directors have an advisory role rather than a disciplinary role in the UK); but cf Brian R Cheffins (n 21) 604605. 134 Klaus J Hopt and Patrick C Leyens (n 5) 152164; see also Eva-Dsire Lembeck (n 9) 958960. 135 See Brian R Cheffins (n 21) 622623; Geoffrey Owen (n 11) 16. 136 But see Brian R Cheffins (n 21) 9899, 609 and 632633 (stressing that shareholders select directors only in theory but in practice still ratify a list of candidates pre-selected by the management).
462
Carsten Jungmann
ECFR 4/2006
to hinder the other board members from making mistakes 137. In addition, shareholders must bear in mind the fact that they need persons on the board that represent their interests 138. If these requirements are met, it is possible for the independent non-executive directors to simultaneously act as colleagues in matters of strategy-setting and decision-making, as well as supervisors. This double function was traditionally integrated into the one-tier system and it has been, with a focus on non-executive directors, highlighted in the recent debate: non-executive directors participate in setting the corporate strategy and in monitoring the management. If necessary, they are obliged to seek the removal of the executive directors from office 139. What remains, however, is the structural weakness of the one-tier system, in which the effectiveness of corporate control depends not only on the personality of the non-executive directors, but foremost on the personality of the chairman. Thus, a minimum formal requirement should be that the chairman is not also the CEO and that he is independent. Both are recommended in Sec. A.2.1 and Sec. A.2.2 of the Combined Code. These requirements are key elements of effective corporate control in the one-tier system, as there is otherwise a danger that too strong an individual, possessing the double title of chairman and CEO, would dominate the board, constraining the boards monitoring function and restricting the representation of shareholders interests.
VI. Conclusion and recommendations for the improvement of corporate control The empirical part of this study led to the conclusion that it is impossible to consider one of the two concurring systems for corporate control as being superior to the other. This is consistent with a view broadly (but not unanimously) shared amongst the academic commentators 140. The purpose of the analysis of the advantages and disadvantages of the two systems in the last part was two-fold: firstly, it brought the theoretical explanation in line with the empirical results, as there exist sound arguments in favour of both systems. Secondly, and more importantly, the last part formed the basis
137 See Geoffrey Owen (n 11) 16 and 19. 138 However, the Combined Code states in Sec. A.3.1 that being a representative of a significant shareholder is a relevant factor to determine whether a director is independent. 139 Paul L Davies (n 10) 446447; Department of Trade and Industry (n 12) para 1.133. 140 See Bundesverband der deutschen Industrie eV and PwC Deutsche Revision AG (n 89) 16.
ECFR 4/2006
463
for the following conclusion, which contains recommendations for the improvement of corporate control in both systems. Keeping the limitations due to path dependence in mind, every endeavour should be made to enhance the systems by implementing single features of each that have proved to be an important element of effective monitoring. Consequently, positive elements have to be stressed and weak features have to be abolished. This approach will necessarily lead to a convergence of the systems. However, convergence remains a concept that leads to an enhanced standard as opposed to simply creating minimum standards. Both systems will keep their uniqueness and their fundamental difference: separation of managerial and control functions in the two-tier system and the combination of the two in the one-tier system. Thus, convergence and continuing divergence are not mutually exclusive 141. The recommendations given in this part keep in mind that the Combined Code in the UK as well as the German Corporate Governance Code are valuable means of enhancing the standard of corporate governance in both jurisdictions. Rule 9.8.6 of the Listing Rules of the UK Financial Services Authority and Sec. 161 of the German Stock Corporation Act of 1965, respectively, require each listed company to include in its annual reports a narrative statement of how it has applied the principles of the relevant code and a statement as to whether or not it has complied with the code provisions throughout the accounting period covered by the annual report (comply or explain philosophy) 142. Although the provisions of the codes are accordingly not binding, any non-compliance appears to seek justification. Thus, any code provision should be as accurate as possible and it is very worthwhile to propose amendments, notwithstanding the soft law character of the code provisions. 1. Recommendations for the improvement of corporate control in the two-tier system In countries with two-tier systems of corporate control such as Germany, being a member of the supervisory board was, for many years, considered to be merely an honorary position. Decisions were normally let through on the
141 See also Lucian Bebchuk and Mark J Roe (n 81) 72 (highlighting that corporate rules can very well be path dependent and efficient). 142 For an exhaustive description and comparison see Tom Kirschbaum, Entsprechenserklrungen zum englischen Combined Code und zum Deutschen Corporate Governance Kodex (Carl Heymanns Verlag, Kln 2006); see also Simon Goulding, Lilian Miles and Alexander Schall Judicial Enforcement of Extra-legal Codes in UK and German Company Law (2005) 2 ECFR 20.
464
Carsten Jungmann
ECFR 4/2006
nod, and there was often a lack of real effort toward effective control. However, this study shows that the control mechanisms were, at least in the ten years under review in this study, nevertheless effective. Nonetheless, in the current changing economic environment and with regard to businesses that are becoming ever more international and complex, the time of regarding supervisory boards as social circles to meet old friends and cement old relationships is over. A couple of years ago, it was not uncommon to be a member of more than twenty supervisory boards. The number of mandates is now limited to ten by Sec. 100 of the German Stock Corporation Act of 1965, and being the chairman of a supervisory board counts as holding two seats. Furthermore, the amended German accounting standards provide for the publication of all seats held by members of the supervisory board in supervisory boards of other corporations.
a) Strengthening the position of the chairman of the supervisory board The introduction of these new rules can be regarded as the first step in the right direction. What is necessary, however, is the strengthening of the chairmans role. He plays the key role in the supervisory board, as he is normally the only member dealing with the management on a more permanent basis. Being the chairman of a supervisory board should therefore become a full-time job143, at least with regard to publicly listed corporations. This should be accompanied by an adequate compensation. Nonetheless, the chairman should remain strictly independent. Independence in this sense means, inter alia, that the chairman should not have had a business relationship with the company for the three to five years prior to his appointment 144. Thus, the normal practice of retiring from the management board and switching to the supervisory board as its chairman 145 should be abolished an approach which can be found in Point 3.2 of 2005 Recommendations of the EU 146. Meanwhile, Sec. 5.4.4 of the German Corporate Governance Code contains a similar, though not too rigid, provision 147. As an accompanying measure, a compulsory upper limit for his term of office should be intro-
143 See Marcus Lutter, Der Aufsichtsrat: Konstruktionsfehler, Inkompetenz seiner Mitglieder oder normales Risiko? (1994) 39 Die Aktiengesellschaft 176, 177; but see Eberhard Schwark (n 111) 104105. 144 See Uwe Hffer (n 118), 643 (favouring a period of two years as the maximum in order to avoid negative discontinuity effects). 145 See n 89 and accompanying text. 146 See n 17. 147 See Uwe Hffer (n 118), 642.
ECFR 4/2006
465
duced 148. After about five years, the danger of the chairman losing his independence becomes too large 149. Furthermore, a chairman should not hold seats on other supervisory and management boards. At the same time, the chairman of the supervisory board should have informational rights that go beyond the present status 150 and beyond the rights which the supervisory board are presently allowed. It should be ensured that the chairman has unlimited access to all sources of information. He should also independently gather the information and should not rely on information provided by the management. Proposals of this kind, made with regard to other jurisdictions, have been heavily criticised in Germany 151. However, this criticism must be seen in the context of the existing German statutory provisions 152. There is no convincing reason to delay changing these rules. The advantage of a strong chairman as proposed in this paper is obvious: within a short time, he will gain a sound knowledge of the corporation. Even if his term of office is limited and he is not eligible for re-appointment, he has a strong incentive to fulfil his tasks thoroughly: as he is appointed by the shareholders and not by the management board, it is important for him to build a reputation. If he becomes renowned for his work as a monitor he is likely to be appointed as chairman in another company. Thus, being the chairman of a supervisory board would become a profession in its own rights. b) Professionalising the role of the other members of the supervisory board The roles of the other members of the supervisory board also need to become more professionalised 153. As their task is clearly defined and limited to the control of the company, it is not necessary for them to have the same degree of independence as the non-executive directors on the British board.
148 There is no upper limit in the Recommendations of the EU (see n 17) or in the German Corporate Governance Code. See also Gerald Spindler (n 17) 2043 (questioning the sense of an upper limit). 149 The Combined Code is less restrictive: According to Sec. A.2.2 and Sec. A.3.1, serving for more than nine years on the boards affects the status of independence of a director. 150 See Christian Schlitt, Der aktive Aufsichtsratsvorsitzende (2005) 58 Der Betrieb 2007, 2008. 151 See Dieter Feddersen (n 99) 461464. 152 See Christian Schlitt (n 150) for a review. 153 Walter Oechsler, Qualifikation und personelle Besetzung des Vorstands und des Aufsichtsrats in Hommelhoff and others (eds), Handbuch Corporate Governance (Otto Schmidt-Verlag, Kln 2003) 305, 310317.
466
Carsten Jungmann
ECFR 4/2006
Nonetheless, it is a step into the right direction that Sec. 5.4.2 of the German Corporate Governance Code refers to an adequate number of independent members of the supervisory board 154. Unless a conflict of interests becomes apparent, the other members also do not have to be as independent as the chairman. On the contrary, it might be an advantage for the supervisory board as a whole if these members have special expertise and experience due to a former position in the company or due to former or present jobs in the same sector of the economy.
aa) Increasing the frequency of meetings It is necessary to further increase the frequency of meetings 155. Nothing hinders the supervisory board from meeting on a monthly basis. This would have a number of positive consequences: even though being a normal member of the supervisory board would not become a full-time job, work load and time pressure would prevent individuals from holding too many seats in supervisory boards. Thus, it is not necessary to further reduce the maximum number of seats by law. At the same time, monthly meetings would allow the supervisory board to monitor management on a more continual basis. This is of crucial importance with regard to the involvement of the supervisory board in strategic or fundamental decisions by the management board. If the list of decisions requiring the approval of the supervisory board grows, it will become increasingly necessary for the supervisory board to react in appropriate time.
bb) Installation of committees Some exceptions notwithstanding, the German Stock Corporation Act of 1965 allows for the formation of committees, and Sec. 5.3 of the German Corporate Governance Code expressly recommends such formations. Nevertheless, German corporations still do not make regular use of this right 156. Audit committees, personnel committees, finance committees, etc. should be
154 Due to the role of the supervisory board in a two-tier system and due to German co-determination rules, the adequate number of independent members is not necessarily very high. In large companies with 20 members of the supervisory board, for example, two or three independent members of the supervisory board are sufficient according to Uwe Hffer (n 118) 641. 155 See Knut Bleicher and Herbert Paul (n 15) 284. 156 See Patrick C Leyens (n 5) 8284; but see Manuel R Theisen (n 7) 261262.
ECFR 4/2006
467
set up in large companies 157. Committees allow for the utilisation of a number of the above-mentioned advantages, as members gain special expertise in certain areas of the company. Meetings can be organised more easily and meeting times can be flexibly adapted to the companys needs. As the size of each committee is smaller by definition, the work can be performed more efficiently. However with the exception of audit committees 158 there should be no mandatory list of committees for each company 159. The question of whether to install committees depends upon the nature and size of the company. Some have a high fluctuation of staff, some rely on the highly specialised technical expertise of the decision-makers, some invest and produce in exotic countries and need decision-makers on site, etc. Accordingly, some types of committees appear to be vital for some companies but not for others. To a certain extend, this view is reflected in the approach taken by the 2005 Recommendations of the EU 160. According to Point 5 of the Recommendations, nomination, remuneration and audit committees should be created within the supervisory board. However, Point 7.1 grants some flexibility in setting up these committees, as it states that companies may group the functions as they see fit and create fewer than three committees. One should bear in mind that the installation of committees is hardly the universal remedy to overcome structural weaknesses in the control process. Some issues are, however specific they may be, reserved for the board as a whole. The board cannot delegate all of its responsibilities to committees. Committees can sometimes be final decision-makers but should normally engage instead in preparing board meetings 161. In this context, their role is important in that they contribute to a more flexible approach and allow for a more efficient use of the time available to the members of the supervisory board. cc) Liability and mandatory qualification standards Members of the supervisory board have a demanding and important responsibility. In large corporations, especially in those listed on the stock exchange, numerous anonymous shareholders rely on their work, as the shareholders
157 Peter Hommelhoff and Daniela Mattheus (n 14) 254255; see also Eberhard Schwark (n 111) 110112 (discussing advantages and disadvantages of committees). 158 But see Owen Green (n 11) 19. 159 But see Marcus Lutter (n 101) 427428; Marcus Lutter, Defizite fr eine effiziente Aufsichtsratsttigkeit und gesetzliche Mglichkeiten der Verbesserung (1995) 159 Zeitschrift fr das gesamte Handels- und Wirtschaftsrecht 287, 298299 (proposing three mandatory committees). 160 See n 17. 161 See Klaus J Hopt (n 1) 261262.
468
Carsten Jungmann
ECFR 4/2006
do not have the ability to efficiently monitor the management themselves. According to Sec. 116 of the German Stock Corporation Act of 1965, members of the supervisory board are held liable to the corporation for any breach of their duties. Recent legislation 162 introduced the possibility for shareholders to take legal action against the members of the supervisory board, however, numerous requirements have to be met. In addition, the action of the shareholders is a derivative action 163. It would be even more advantageous if the shareholders themselves were able to hold members of the supervisory board responsible for ineffective control of the company. In such a regime, the company would not necessarily have to suffer a loss; the loss suffered by the shareholders would be sufficient. This would mean that a decrease of the share price would be regarded as a loss, and the members of the supervisory board would be liable if such a decrease was due to a failure on their part to properly monitor business operations. An additional method of increasing the quality of corporate control by the supervisory board would be the introduction of mandatory qualification standards 164. Such requirements should be applicable for all members of the supervisory board in order to guarantee a uniform level of effectiveness. In any case, diversity of experience and knowledge is beneficial for the work of the supervisory board. Thus, it is neither necessary nor desirable for the supervisory board to consist solely of experts in business administration, corporate finance or the like. Expertise in the relevant industrial sector, technical knowledge, access to important external information, good business relations, etc. would also serve as evidence of qualification. Only a board consisting of persons who, on the whole, have the expertise to fulfil the necessary level of supervision over the company, serves as an effective means of control. Qualification standards, in this sense, would also minimise the potential conflicts with co-determination rules. It was already mentioned above that
162 Corporate Integrity and Modernisation of the Right of Contestation Act of 2005 (Gesetz zur Unternehmensintegritt und Modernisierung des Anfechtungsrechts (UMAG) (22 September 2005, BGBl I 2802). 163 For details see Gerald Spindler, Haftung und Aktionrsklage nach dem neuen UMAG (2005) 8 Neue Zeitschrift fr Gesellschaftsrecht 865. 164 See Peter Hommelhoff and Daniela Mattheus (n 14) 255256; Peter O Mlbert, Die Stellung der Aufsichtsratsmitglieder in Dieter Feddersen, Peter Hommelhoff and Uwe H Schneider (eds), Corporate Governance (Otto Schmidt-Verlag, Kln 1996) 99, 122123; Thomas Raiser (n 8) B 104; Theodor Baums (ed), Bericht der Regierungskommission Corporate Governance: Unternehmensfhrung, Unternehmenskontrolle, Modernisierung des Aktienrechts (Otto Schmidt-Verlag, Kln 2001) 94 and 317319; Frank Wardenbach, Interessenkonflikte und mangelnde Sachkunde als Bestellungshindernisse zum Aufsichtsrat der AG (Otto Schmidt-Verlag, Kln 1996) 178183.
ECFR 4/2006
469
co-determination law might hinder mandatory qualification requirements, and the same might hold true with respect to liability rules. But labour representatives typically have a sound knowledge of the company and can thus help the supervisory board to control management with regard to a number of questions that are unique to that company. The liability of the labour representatives might also be restricted to cases of gross negligence. This appears to be consistent with co-determination rules and labour law, according to which no employee is permitted to act recklessly. c) Involvement of the supervisory board in management decisions Finally, the supervisory board should become more involved in management decisions. This does not mean that the separation of management and control in the two-tier system should be abolished. Day-to-day management should remain the responsibility of the management board. Indeed, there are very significant differences between the daily responsibility of the members of the management board and the periodic responsibility of the members of the supervisory board to evaluate plans, decisions and results. The supervisory board should not be completely limited to an ex post control. Consequently, all fundamental decisions and especially those that set or change the corporate strategy should require the approval of the supervisory board 165. This is an expansion of the present statutory provisions and of the rules already existing in the by-laws of some companies, since the current law only requires the by-laws to provide a list of decisions that need approval. This, however, is merely a formal requirement, as the form of the list and its content are not prescribed. The proposed mandatory approval list is more than a mere expansion of the existing legal rules: it should be seen in the context of the above-mentioned professionalisation of the supervisory board as an organ consisting of experts permanently monitoring the company. In such a reformed supervisory board the flow of information is more complete and timely, and due to the higher frequency of meetings, decisions can obtain the necessary approval in a reasonable time. 2. Recommendations for the improvement of corporate control in the one-tier system Some of the recommendations made with respect to the two-tier system are of equal importance to one-tier systems and need not be repeated. This holds true especially with regard to the installation of committees and liability
165 Theodor Baums (ed) (n 164) 7677; but see Carsten Berrar (n 24) 21842185.
470
Carsten Jungmann
ECFR 4/2006
rules 166. Subsequently, this paper concentrates on the proposed changes that are essential for the unitary board. It has been stressed that the central issue in each unitary board is, firstly, the lack of separation of management and control and, secondly, as a consequence thereof, the monitor-colleague-dilemma, which the non-executive directors are facing. Hence, the key to successful control of the management of the company is to reduce the potential conflict inherent in this dilemma. As was proposed, particularly in the American literature on corporate governance 167, this can effectively be done by choosing independent non-executive directors. Thus, the recent reforms in the UK to strengthen the role of independent directors were necessary and essential. While this is a widely accepted view, we should bear in mind that independence does not only have its merits but also its downfalls. There are fewer incentives for a person who had not previously had and does not currently have a relationship to the company to spend an adequate amount of time controlling the companys affairs 168. Moreover, it is questionable whether such a person has the relevant knowledge of the particularities of the firm, which is a disadvantage for both tasks. a) Frequency and duration of board meetings Information concerning the company is complex. This is true especially with regard to data about long-term trends relating to financial performance, competitive position and organisational structure. Consequently, directors need time to evaluate the information and consider decisions and choices of the past. It is thus worthwhile to further increase the frequency of board meetings and to make board meetings last longer 169. Monthly board meetings
166 Derek Higgs, Review of the Role and Effectiveness of Non-Executive Directors (Department of Trade and Industry, London 2003) 5965; Brian R Cheffins (n 21) 621624. 167 See Ronald J Gilson and Reinier Kraakman, Reinventing the Outside Director: An Agenda for Institutional Investors (1991) 43 Stan. L. Rev. 863; Reinier Kraakman, Die Professionalisierung des Board in Dieter Feddersen, Peter Hommelhoff and Uwe H Schneider (eds), Corporate Governance (Otto Schmidt-Verlag, Kln 1996) 129, 135 136; see also Stuart Rosenstein and Jeffrey G Wyatt, Outside Directors, Board Independence and Shareholder Wealth (1990) 26 J. Finan. Econ. 175. 168 See Brian R Cheffins (n 21) 611. 169 Otto Graf Lambsdorff, berwachungsttigkeit des Aufsichtsrats Verbesserungsmglichkeiten de lege lata und de lege ferenda in Dieter Feddersen, Peter Hommelhoff and Uwe H Schneider (eds), Corporate Governance (Otto Schmidt-Verlag, Kln 1996) 217, 224.
ECFR 4/2006
471
are essential for continual supervision and for giving every board member the possibility to promptly ask questions. The emphasis is on the continual possibility of initiating discussions and addressing all areas of concern. Once such a culture of continual supervision has been established, the preparation of board meetings does not require the same amount of time as it does if meetings take place only every third or fourth month. Thus the counterargument that monthly meetings lead to an inefficient allocation of directors attention is diluted. At the same time, and in contrast to a minimum frequency, it does not appear to be sensible to postulate a minimum duration of a meeting. Each meeting should last until all issues are discussed. It is then a question of the particular circumstances as to how much time is indeed necessary. One exception to this last point should be mentioned: at least once per financial year, a board meeting should take place over two or three consecutive days. In the absence of a tight time schedule, but rather with an extended period including coffee breaks, lunches and one or two evenings spent together, thoughts would be shared more openly and frankly, and non-executive directors would have a better chance of getting involved in the work of the board, even if they are not as well prepared as the executive directors. In such a meeting, the long-term corporate strategy should be on the agenda 170.
b) Meetings of the non-executive directors Having meetings of the board which last several days is a way of better integrating the non-executive directors into the board. At first glance, it may seem contradictory to also propose that the non-executive directors convene in additional sessions without the management. However, in meetings of this kind, it is again very likely that cohesive bonds develop among the nonexecutive directors and that questions are discussed that might not be otherwise brought up 171. They also strengthen the position of the non-executive directors in the next regular board meetings, as the non-executive directors might feel more competent when convening later with the executive directors. As only questions of corporate control are on the agenda of the exclusive meetings, the monitor-colleague-dilemma is not as apparent as in the board meeting, since the non-executive directors are able to concentrate their efforts solely on questions of control.
170 Martin Lipton and Jay W Lorsch (n 111) 6970. 171 See Derek Higgs (n 166) 3334.
472
Carsten Jungmann
ECFR 4/2006
c) Composition of the board The smaller the number of directors, the more likely it is that each director can play an active and vital role. As a rule of thumb, in boards with more than ten or twelve members, at least some directors will feel superfluous and consequently stop preparing board meetings themselves but rather rely on the work of the others. In addition, sufficiently thorough discussions will not take place in groups that are too large 172. In a board consisting of eight to ten directors in total, there is enough diversity and a sufficient range of viewpoints in the boardroom and an exchange of thoughts is still possible 173. The Combined Code states in Sec. A.3.2 that at least half the board, excluding the chairman, should comprise non-executive directors determined by the board to be independent but does not stipulate a maximum number of seats in total. In some large companies, there might be practical need for six or eight executive directors. It is doubtful whether this legitimises a board consisting of twelve, sixteen or even more persons in total in order to meet the requirement of the Combined Code 174 and whether this could be inconsistent with the supporting principle in Sec. A.3, according to which the board should not be so large as to be unwieldy. In contrast, the 2005 Recommendations of the EU 175 take a more sensible approach, as they forbear from prescribing fixed numbers but refer in Point 4 to a sufficient number of independent nonexecutive directors that ensures that any material conflict of interest involving directors will be properly dealt with. Indeed, it seems sensible to provide for a flexible rule and to regard a maximum board size as the superior criterion. It should be guaranteed, however, that a minimum number of independent non-executive directors exists, and that they have effective voting rights 176.
172 See Klaus J Hopt (n 1) 255256; for a quantitative analysis see David Yermack, Higher Market Valuation of Companies with a Small Board of Directors (1996) 40 J. Finan. Econ. 185. 173 Cf Korn/Ferry International (n 38) 9 (arguing that the optimal size lies between six and eight). 174 See also Derek Higgs (n 166) 33 (recommending a strong executive representation on the board in order to ensure that the board as a whole is well informed). 175 See n 17; in this context, see also Jan Lieder, Das unabhngige Aufsichtsratsmitglied (2005) 8 Neue Zeitschrift fr Gesellschaftsrecht 569. 176 At present, it is likely that the majority of executive directors prevails too often, even with regard to issues raised by non-executive directors (see Brian R Cheffins (n 21) 609610). Cumulative voting rights for non-executive directors might overcome this problem.
ECFR 4/2006
473
3. Reflection on the limits of corporate control Effective corporate control means that there is an institutionalised, formal process of monitoring and reviewing the actions of the decision-makers in a company. Tasks should be clearly defined, and all stakeholders of the company should know about the responsibilities of those in charge of monitoring the management. However, every system of corporate control has its limits, and it can be made more effective if structural weaknesses are eliminated. Yet, this does not mean that the possibility that mistakes are made by the relevant people in the company can be eliminated, and mistakes can be made at both levels, at the managerial level as well as at the control level. Of course, we can hardly say that outsider directors, non-executive directors, independent directors, etc. or competent, well paid, full-time working members of the supervisory board have a degree of foresight and wisdom that makes them, in any case, superior to the executive directors or members of the management board. Both groups can be wrong in their decisions; both might take too many risks and thus lead the company into a period of financial distress. Thus, neither integrating the monitors into the decision-making process nor strengthening the role of those involved in the decision-making process as monitors would automatically bring an end to business failures. However, an enhanced level of control would reduce the probability of mistakes that occur at the managerial level remaining undetected. Consequently, there is an increased chance that wrong or harmful decisions can be corrected and that injury to creditors, shareholders, employees and the general public can be reduced.
VII. Summary In the last decade, corporate governance and questions of effective corporate control have become one of the core topics in finance, business administration and law. The UK and Germany represent prototypes of two competing systems. This study has shown that the British one-tier system, with its unitary board, and the German two-tier system, with a management board and a supervisory board, are both effective means of control. In addition, the empirical results showed that it is not possible to deem one board system superior to the other. This finding is consistent with theoretical approaches, which have highlighted the strengths and weaknesses of both systems. Although both regimes are demonstrably effective, it was argued in this paper that there is room for improvement. It was concluded that single features of one system can be introduced into the other, and vice versa. This must be
474
Carsten Jungmann
ECFR 4/2006
done carefully, however, as each system has its peculiarities which do not allow for a fundamental change. Consequently, the amendments proposed here would lead to a certain degree of convergence of the British and the German system, but would not eliminate the diverging key features. As a result, the standard of corporate governance would be enhanced and the likelihood of corporate failures and financial crises would be reduced.

The Effectiveness of Corporate Governance in One-Tier and Two-Tier Board Systems

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Effectiveness of Corporate Governance in One-Tier and Two-Tier Board Systems

Uploaded by

Copyright:

Available Formats

The Effectiveness of Corporate Governance in One-Tier and Two-Tier Board Systems

Evidence from the UK and Germany by Carsten Jungmann*

ECFR 2006, 426474

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

Loss for the Financial Year

Abnormal Negative Share Price Performance 106 46.70 % 73 42.94 %

Dividend Cut or Omission

Germany Total Percentage United Kingdom Total Percentage

Table 1: Measures of poor financial performance

One-Tier and Two-Tier Board Systems

Table 2: Board size in Germany and the UK (19942003)

One-Tier and Two-Tier Board Systems

8.88 % 12.18 % 11.71 % 15.44 %

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

One-Tier and Two-Tier Board Systems

You might also like