Companies are facing growing demands to evaluate the performance of their boards. The pressure comes from a variety of sources but, most obviously, it is the regulators who are setting the tone. In 2003, the Combined Code on corporate governance, revised in the wake of the Higgs Report, made board evaluation mandatory for UK companies and, the following year, the New York Stock Exchange introduced similar rules for US businesses. Elsewhere, there is a similar trend as investors, fund managers, insurers, capital markets and the media urge greater levels of evaluation and transparency.
The problem for boards is that there is minimal guidance on how to carry out an evaluation and no universal agreement as to when it should take place. The NYSE, for example, politely asks that companies “address” evaluation annually, while in the UK, the Combined Code leaves the door open to a variety of interpretations. It advises the board to “undertake a formal and rigorous annual evaluation of its own performance and that of its committees and individual directors”.
The evaluation process
There are five broad sets of issues that must be decided before embarking on a board evaluation:
1) Who should conduct the evaluation? Should the board evaluate itself or should external facilitators be involved?
2) Who is to be evaluated? Should it just be the board, or should the various committees also be subjected to the process?
3) What are the issues to be covered? For example, should the evaluation only address one-dimensional hard issues – such as the meeting schedule, time allocation on the agenda or circulation of papers – or should it also look at broader questions of culture and trust?
4) What form should the evaluation take? There are two methods that are commonly used – questionnaires and interviews.
5) How should the information be handled? In other words, what should be made publicly available and what should be kept confidential?
Who should conduct the evaluation?
Although in the UK, Higgs recommended the use of a third party for evaluation, the extent to which the process is internal or externally facilitated clearly varies. Internal evaluation usually involves structured discussions between the chairman and board members, sometimes supported by the circulation of a questionnaire to board members that is completed confidentially and then discussed by them. The process, which is typically led by the company secretary, has the advantage of being reasonably straightforward to organise and relatively cheap, but it is prone to being self-congratulatory. Turkeys, as a rule, do not vote for Christmas.
Internal evaluation remains by far the most popular approach. Research undertaken in 2004 by the consulting firm Independent Audit found that 73 per cent of UK companies use self-assessment.
At Vodafone, for example, the board, its committees and individual directors are all evaluated annually. The chairman assesses the non-executive directors; the CEO reviews the executive directors; and the senior independent director reviews the chairman. According to the company’s annual report: “Each board committee undertakes a review of its work and in relation to the performance of the board, the chairman invites suggestions from all directors as to ways in which the board and its processes may be improved.” While Vodafone is developing questionnaires for future use, it publicly states its faith in an internal process, preferring to administer without the use of external consultants. More unusually, the retailer Marks Spencer based its own performance review on an employee survey.
The second general approach is to bring in external consultants – typically to conduct interviews and provide recommendations. While this approach potentially offers greater objectivity, it also has a number of drawbacks. Board members may be inclined to view the consultants with sometimes ill-disguised scepticism. After all, board evaluation is a developing and imprecise art (or science depending on your perspective) and consultants may not fully grasp the nitty-gritty of the business. It is also expensive. One multinational company is believed to have paid £240,000 for an evaluation.
A third possibility may lie in a hybrid of the two, by combining a self-assessment tool with the involvement of an external facilitator. This is an approach now being adopted by a growing number of companies. For example, Sonae, one of Portugal’s largest commercial groups, has recently introduced this type of approach for two of its boards. Typically, the outsider is an academic or consultant – ideally, perhaps, an independent, but known, quantity.
Who is to be evaluated?
It is common practice to include the entire board of executive and non-executive directors in addition to all the main committees – audit, finance, nominations and remuneration. For example, Colgate-Palmolive, the consumer products group, established its evaluation process in 1997. Its board committees conduct self-evaluations that are then reviewed by the board and the company complements this with evaluations for individual directors.
What are the issues to be covered?
The current tendency is to focus on relatively straightforward, one-dimensional measures, such as the make-up of committees, length of tenure of board members, regularity of meetings, the background of board members, and the circulation of meeting agendas. One reason for this approach is that these metrics are easily quantifiable and provide neat boxes that can readily be ticked. This is useful, but not sufficient.
When Jeffrey Sonnenfeld, a corporate governance expert based at the Yale School of Management, examined the boards at Enron, WorldCom and Tyco – some of those pilloried for recent corporate scandals – he found no broad patterns of outward corporate governance failure. In fact, the boards of these companies exhibited some of the best governance practices in terms of structural and procedural issues, such as board size and composition, attendance at meetings, the make-up of committees and financial literacy. They also scored highly on accountability mechanisms, such as codes of ethics and conflict of interest policies.
The uncomfortable fact is that, even if there were external governance checks in place, Enron, in particular, would have passed with flying colours. As this demonstrates, there is a real danger that corporate governance checklists, including the requirements of the Sarbanes-Oxley Act, miss a more fundamental issue. The point of evaluating boards is not simply to ensure they are meeting certain formal targets, but to make them more effective.
Consider the issue of separating the CEO and chairman roles – something recommended by almost all governance codes. On this issue, there is broad international agreement on governance best practice – even though the working reality and environment varies considerably across countries.
A recent study by the governance ratings agency Governance Metrics International (GMI) found that 95 per cent of the UK’s FTSE 350 companies rated by GMI split the role. In Germany, the roles of the chairman as head of the supervisory board and the CEO as head of the management board are legally separated. In France, where the combined CEO and chairman has traditionally been a powerful force, there is now a trend towards splitting the role – with companies such as Renault and Carrefour recently making this division. A similar trend is also emerging in the US where, until recently, the two roles were commonly combined. A recent survey by GMI found that one-third of US rated companies had a separate chairman and CEO; in 2002, just 25 per cent split the role. Despite these variations, there is nothing to indicate whether or not boards are any more effective as a result of splitting the roles.
“Potentially, boards have three resources to use: power, information and knowledge” observes Edward Lawler, the US academic who specialises in board effectiveness. “When these three resources are present and effectively directed at, first, handling emergencies; second, making sure an effective strategy is in place; and third, truly influencing the decisions of the chief executive officer (and whoever succeeds the CEO), then we can say that the board is acting in a high-performance way. But there’s another key point that should be stressed. A board is a group, perhaps in some cases, a team. Boards need to be assessed by the same conditions and behaviours that lead groups to be effective.”
Mr Sonnenfeld agrees. The primary distinctive feature of an effective board, he argues, is the extent to which it performs as an effective team, or a “high-functioning work group”.
“We need to consider not only how we structure the work of a board but also how we manage the social system a board actually is,” he observes. “We’ll be fighting the wrong war if we simply tighten procedural rules for boards and ignore their more pressing need – to be strong, high-functioning work groups whose members trust and challenge one another and engage directly with senior managers on critical issues facing corporations.” He notes that stock ownership, financial literacy, attendance records, service of the former CEO and “independence” do not seem to correlate in a meaningful way to performance of the company or the board.
The best evaluations take a broader perspective. They look at the climate and culture of the board, its team dynamics, and the levels of trust and relationship between board members. If the point of board evaluation is to improve performance, such insights into the behaviour of board members, individually and collectively, are crucial.
More enlightened companies are already moving in this direction. Legal and General, for example, has been conducting this type of wide-ranging survey for several years – well in advance of the recommendations of the Higgs Report. At the telecommunications company Cable Wireless, according to their 2004 annual report: “The independent non-executive directors meet privately without the chairman, and as a group with the chairman, as necessary, at least once a year, and similarly with both the chairman and the chief executive officer, to consider management performance and succession issues. The independent non-executive directors also review annually the relationship between the chairman and the chief executive officer to ensure that the relationship is working to promote the creation of shareholder value.”
Generally, of course, relationships and behaviours will never be as easily measurable as structural and procedural factors. This suggests that it is wise to consider multiple measures (quantitative and qualitative) as well as a variety of contexts for feedback and discussion.
What form should the evaluation take?
There are two main methods to generate information on which to base an evaluation – questionnaires and interviews. The former are quick, cheap and provide comparable data. However, they can appear impersonal, threatening and may lack depth. The latter are clearly more personal and create a much deeper pool of data. The trouble is that they are slow, time-consuming, expensive and produce a sprawling mass of data. Good practice often consists of a combination of the two – a robust questionnaire and carefully constructed interviews – followed by a discussion.
How should the information be handled?
Clearly, there are issues of confidentiality. Without reasonable safeguards, directors are unable to enjoy candid and productive conversations with outsiders. At the same time, companies are obliged to make some of the information public. Unless the boundaries are clear, board evaluations are unlikely to generate useful recommendations. Where outsiders are used, reasonable assurances must also be provided that information gathered through board evaluations will not be used as a lever to generate additional consulting work.
Once the results have been gathered and analysed, best practice usually involves individual data for all directors being shared with the chairman and data for the executives being shared with the CEO. This is then discussed in a follow-up session.
The final stage of the evaluation is a one-to-one session for each board member with the chairman – and sometimes with the external facilitator available for separate assistance if required. In addition, there are board discussions to address collective issues – again typically with the facilitator present.
Results of an evaluation
A successful board evaluation is a highly demanding process that requires the full support of the chairman and CEO. Without their buy-in, evaluation quickly degenerates into a box-ticking exercise that sheds little light on anything. Directors should be consulted throughout the development and design of the process – indeed, the more involvement they have, the more likely they are to commit to any action-related outcomes.
A productive relationship between executive and non-executive directors is also essential for an evaluation to succeed. Indeed, properly structured, the annual board evaluation process should help to raise awareness of important issues among non-executive and executive directors, as well as improve the relationship between them.
If the board is doing its job, the end result of the evaluation process is unlikely to be seismic. Just as in any formal appraisal system, there should be no huge shocks. A well-managed evaluation should primarily encourage a board to focus and adjust its thinking and actions.
Small shifts in thinking and behaviour can produce substantial impacts. “Boards that assess their members and themselves tend to be more effective than those that don’t,” concluded the US academic David Finegold and Mr Lawler after extensive research into corporate governance.
In the longer term, it is also important to keep the evaluation process fresh. Any appraisal process fails when it becomes no more than a routine or ritual. Rob Margetts, chairman of BOC Group, the industrial gases company, follows up his board’s externally facilitated, questionnaire-based evaluation with appropriate actions. Then, in the following year’s review, he leads a focused, internal discussion using a series of structured interviews rather than a survey. Alternating in this way continually refreshes the evaluation process.
But while board evaluation is clearly useful and can improve board effectiveness, it does not necessarily provide easy answers for investors. Public reporting of board appraisal processes is typically brief and lacking in detail. Using board evaluations to compare the performance of one board with another is still all but impossible. Here – as in other areas of board appraisal – there is scope for further development as boards become more comfortable with the evaluation process and recognise the importance of sharing information in ways that benefit all stakeholders.
Rob Goffee is professor of organisational behaviour and faculty director for executive education at London Business School. He is co-author (with Gareth Jones) of “The Character of a Corporation” (Profile, 2003) and “Why Should Anyone Be Led By You?” (forthcoming). He has evaluated the performance of many corporate boards.
