Board composition and committees

When it comes to variables related to the banks’ boards, it is worth mentioning that in the instances where the board is divided into an executive and a supervisory board, we consider only the latter. It is the supervisory board that carries out the monitoring and metering function that is advocated by academics, and is consists of representatives of the principals of the firm. The executive board, on the other hand, provides insight into the operations of the firm, and bridges the gap between the employees and the owners. As such, while it may improve the functioning and effectiveness of the supervisory board, it does not represent principals vis-à-vis agents. Also, potential secretaries on boards are excluded from our statistics.

At the most basic level, we consider board size and director turnover on the level of the supervisory board. Although free-riding is facilitated on a larger board, more directors facilitate oversight and maintenance of the fiduciary duty as it limits the risk of collusion between directors and company managers. Further, a larger board makes it easier for minority owners and other stakeholders to get

35 direct representation on the board. In summary, when kept within reasonable bounds, a larger board should better represent a wide group of stakeholder interests. Thereby, it potentially mitigates agency issues and supports stakeholder influence over the corporation. Previous empirical research appears ambiguous with regards to the effect of board size in financial institutions, a number of papers which find negative relationship between board size and firm performance, including Guest (2009) and Conyon & Peck (1998) while a study by Adams & Mehran (2008) focused on the banking sector finds that banks with larger boards do not underperform peers (measured by Tobin’s Q).

Similarly, it can be argued that a high director turnover benefits a broad stakeholder constituency since board seats are offered for new directors at a more frequent basis, we believe that the key advantage within an agency setting of high director turnover is the limit it puts on director entrenchment. In particular, minority owners will have a more direct avenue to challenging majority owner representatives who promote individual interest when directors are replaced more frequently. Further, for our purposes, we look at director turnover as a general proxy for change within the board of directors. The variable is thought to indicate whether or not the replacement of directors on the board has accelerated as a reaction to large institutional changes.

Director independence can be defined in a number of ways. According to some, a director is not independent unless he (1) does not have any executive responsibilities in the firm, and (2) does not own stock. Yet, as most board members own shares in the company that appointed them for the board, we employ a somewhat more lenient definition and require from an independent director that (1) he does not have any executive responsibilities in the company, and (2) that he does not have any material interest in the firm. This latter requirement is highly subjective and given the high complexity of determining what such an interest entails, we have chosen to use definitions as employed under national exchange listings. In annual reports, a firm is obliged to stating whether a director is independent according to this definition, and these statements are used throughout our paper.

Irrespective of the precise definition of the term material interest, for the purposes of acting as a proxy, director independence is in the interest of owners vis-à-vis managers.

As a proxy for board diversity, we have counted the number of nationalities⁸ represented on boards at each of the year-ends. While nationality does not encompass the range of qualities sought after when attempting to establish a diverse board of directors, it is indicative and an operational alternative for our purposes. A diverse board is more likely to possess complementary competences and perspectives, and is therefore better suited to supervise and advise management on behalf of the owners. Further, it is more likely to represent a broader set of interests, which means that it more broadly represents the stakeholder society.

8 Nationality is defined as the director’s citizenship

36 A further characteristic that we looked for among board members is previous experience from the banking sector. For each of the banks, we counted in each year how many board members were currently or had previously been employed by a bank. The variable reflects an ambition to create a board with a thorough understanding of the complex dealings of a modern, systemically important bank. With experience from the sector, a directory would, all else equal, be better able to understand and assess the risks that a bank assumed, and how these can affect profitability and solvency. First and foremost, compared to employees, equity holders have to take much of the cost incurred when a bank becomes subject to a negative risk outcome. Therefore, a better understanding among board representatives of the bank operations makes it easier for the board to notice and correct opportunistic behaviour. However, if risk can be better managed and volatility reduced, this benefits also creditors⁹ who face a lesser risk of default. Finally, in the event of default of a systemically important bank, large costs will be shifted to taxpayers and depositors. Therefore, indirect stakeholders also benefit when boards have a more thorough understanding of the riskiness of the activities that a bank engage in.

Given the systematic importance of our sample, each bank is intrinsically linked to the broad society.

Not only will externalities be borne by taxpayers, but a default will potentially destabilize the financial system and set off a sequence of events that have a negative effect on the broad economy. As such, it can be argued that tax-payers deserve representation on the board of any systemically important bank.

We have created a binary variable that indicates whether any of the board representatives has current or past political affiliation or duties as a civil servant. A board member with direct care for the public would benefit those stakeholders who are subject to potential externalities. We are aware, however, that this interpretation should be treated with some caution. A board member with likes to the political sphere may be a shortcut for shareholders in banks to gain access to and understanding of the political domain, or to get official approval of certain actions.

When a CEO also chairs the board, a phenomenon known as CEO duality, opposing interest may exist. In traditional organization theory, CEO duality establishes strong, unambiguous leadership and legitimacy through the organization. However, from an agency perspective, it may work in favour of CEO entrenchment and weaken the ability of the board to effectively monitor the bank’s management.

We also state the number of board meetings per year for each of the banks. All else equal more frequent board meetings strengthen the ability of the board to monitor, and should be an effective deterrent to the management to act opportunistically as such behaviour would more likely be discovered. However, there may be a reason why the board has convened more frequently than

9 Equity holders, who appoint board members, will be interested in increasing the volatility above what would maximise value for the bondholders. Yet, as a further investigation within this topic is beyond the scope of our paper, we assume that equity holders are interested in running the company at a value maximizing volatility level. Taking this as given, bond holders will demand covenants that limit opportunistic shareholder behaviour.

37 planned in a given year. If so, the number does not reflect more stringent oversight, but other interests which may not be related to the first agency problem.

Finally, we have mapped the prevalence, size and turnover of three different board committees: (1) audit committee, (2) compensation (remuneration) committee, and (3) risk committee¹⁰. The audit committee is the most central committee and has the principal responsibility for a bank’s audit and control systems. While rich disclosure is essential for any of a firm’s stakeholders, the linkage is most obvious for discretionary investors and business partners vis-à-vis the insiders of the firm (i.e. agency problems one and three). Reliable audit is essential for investors and trade counterparties alike in forming an opinion about the firm, and to obtain the information necessary to prevent opportunism.

The compensation committee is responsible for designing executive compensation schemes and to align managerial incentives with those of shareholders. As discussed above, the power of those incentives may motivate excessive risk taking and such volatility may be to the detriment both of a bank’s creditors and its broader stakeholder constituency. Thereby, the compensation committee can mitigate both the first and the second agency issue, while pressures from the public make it subject also to stakeholder interest alignment. Finally, the risk committee is responsible for overseeing the risk that a bank is exposed to but also to provide an assessment of the organization and processes surrounding risk management. Similar to the compensation committee, its responsibilities have a direct effect on several groups in society. Primarily, however, if we assume that the ultimate responsibility of the risk committee is to keep the bank from going into bankruptcy, the main interest groups represented are creditors (third agency problem) and the broad stakeholder constituency.

In mapping committee compositions and membership, it should be noted that we have focused on each board committee’s raison d’etre rather than its actual title. Therefore, if a committee was set up with responsibilities that closely resembled those of any of the three committees presented above, it will be considered irrespective of its formal name.

In addition to director turnover on the board, we map for each of the banks the turnover among the directors per committee. This variable should be treated with caution, however, since changes to committee affiliation may not be indicative of malcontent. Rather, it is in many instances common practice among board members to rotate across committees (see footnote 10). Acknowledging the underlying rationale, rotation across committees is an intentional corporate governance mechanism that seeks to enhance the committee work. Within our framework, similarly to director turnover, this variable serves as a general proxy for change within the board of committees.

10 Our account of board and committee responsibilities relies on statements made in the annual reports we used in collecting our data, see 10.1 Appendix 1

38