Future Students, Alumni & Visitors


This blog is intended to be a governance resource and source of current governance commentary, offered by a corporate governance academic engaged in research, teaching and other ongoing academic activities. There is a very public element to the governance field, and it is hoped that this blog will contribute to the public discussion of current governance issues. It is also hoped that it will address a need in the governance field by presenting a holistic online approach to the topic. There is a rapid rate of change in the field of governance (public, private, government and not-for-profit entities) and developments in internet technology move swiftly. This governance blog offers resources for a broad variety of stakeholders including: [...more]




Executive Compensation is “Corrosive” and “Undermines Trust”: Connecting the Occupy Movements

I remember when US pay czar Ken Feinberg told a group of academics gathered at Wharton business school for a corporate governance conference to discuss the aftermath of the Global Financial Crisis that he was looking for independent compensation consultants and, to quote Mr. Feinberg, “there are no independent compensation consultants.” So he turned to academics. He wanted to study the claim by consultants that executives need to be paid extraordinarily high compensation or else they would migrate to other companies and jurisdictions, which – as it turned out – did not happen, Feinberg said, or is a “myth” as was stated in the UK this week. Addressing conflicts of interest by compensation consultants is only one of twelve reforms being urged by the “Final report of the High Pay Commission” in a scathing report released this week in the UK.

Reforms to the way executive compensation is set in the UK are forthcoming that may include significant and unprecedented changes – well beyond the structural Dodd-Frank reforms in the US. Changes that may be termed “radical” by some include: binding and forward-looking voting on compensation by all shareholders; having women and worker representation on compensation committees of boards; regulating remuneration consultants; regulating the disclosure, unnecessarily complexity and format of “fair pay” compensation; and having board of director positions advertised and applied for publicly.

A central theme throughout the compensation debate has been that boards and compensation committees – particularly in the US and UK but also elsewhere – have been incapable or unwilling to address the uncontrolled disparity between pay of CEOs compared to that of other senior management and, in particular, the pay of average workers, even throughout the financial crisis. The market is not really “free,” proponents maintain, but is in reality a “closed shop” (words of the Chairwoman Hargreaves of the High Pay Commission) (video). That is to say that pay is set by a small, heterogeneous, interlocked and self-selected group of management and directors. University of Delaware professor Charles Elson and his graduate student, Craig Ferrere, have documented an annual, compounded structural 17% increase in CEO pay over decades as a result of the way CEOs are paid at or above median and the marketing of peer group data by consultants. In some cases, exit pay packages for CEOs have been the hundreds of millions of dollars. The public outrage seemingly falls on some or many (but by no means all) tone-deaf boards and senior management teams.

All reforms are now on the table and the UK Prime Minister and Business Secretary Vince Cable have weighed in, including Mr. Cable expressing sympathy with the “Occupy” movement and calling the current system “dysfunctional” and a failure of corporate governance.

What the Occupy movements have done, it can be argued, is focus the discourse on the consequences of wealth disparity. Ted Talk by British researcher Richard Wilkinson, for example, talks about the harm to society that results from economic inequality, notably the gaps within (not between) societies, which include harms such as life expectancy, literacy, infant mortality, crime, teenage births, obesity and mental illness. (Credit goes to former York University student, Cliff Davidson, for showing me this link.) The link between wealth disparity and social harm is an “extraordinarily close correlation,” Professor Wilkinson states.

What the UK experience also shows is that regulators are prepared to step in and bridge gaps if industry proves incapable or unable to do so itself. In a speech I gave a year ago, I recommended that North American compensation consultants devise a code of conduct for consultants – independently developed and enforced – that includes consequences for breach, similar to regimes that lawyers and accountants have, or governments eventually would do so for them. John Tory was in the audience and endorsed my notion of industry leadership before government regulation. Regulation tends to have unintended consequences, and industry leadership is far superior to the former. Industry leadership unfortunately is not happening and is unlikely given vested interests. We have seen the consequences of inaction in the UK.

Save and Share
  • Print
  • PDF
  • email
  • LinkedIn
  • Twitter
  • Facebook
  • Reddit
  • del.icio.us
  • StumbleUpon
  • Add to favorites
  • RSS

Boards and Sexual Assault on Campus

“As the graduate assistant put the sneakers in the locker, he looked into the shower. He saw a naked boy, Victim 2, whose age he estimated to be ten years old, with his hands up against the wall, being subjected to anal intercourse by a naked Sandusky. … The graduate assistant left immediately, distraught.”

I apologize to all readers for quoting this alleged abhorrently heinous criminal conduct from the Grand Jury report to what is reputed to be several young boys.

Universities are historic institutions, steeped in tradition. Many however have sorely outdated governance practices. Penn State is a good example. What can we learn?

Penn State prides itself on not changing the size or composition of its board since 1951. What this means is that the entire organization is not keeping up with the times.

Thirty-two directors is not a board: it is a theatre. A board this large means management dominates and decisions are made in advance rather than at the table.

The board of trustees should immediately disestablish the Executive Committee chaired by the President. An executive committee means a “real” board where management controls rather than the board and its committees.

The board size should be reduced to half: sixteen directors maximum and preferably fewer. Multi national corporations have fewer directors.

The university president, or any other member of management, should have no influence whatsoever into director selection.

Penn State does not even have an audit or risk committee. What good board does not have an audit committee? The audit/risk committee should oversee conduct and compliance reporting. Where is this obligation overseen by a committee of the Penn State board, I wonder? No committee charters are available, which is another red flag.

A nominating and governance committee should also be established. So should a human resource committee. It is remarkable that audit, nominating or HR committees do not exist and this again suggests undue influence by management who does not want this oversight.

Penn State’s governance statements are verbose, pompous, self serving and ineffective, as are those of many colleges and universities, deliberately so and written by management who write for a living. Key governance documents are missing, such as the competencies and skills of each director linked to their responsibilities; the code of conduct; compliance procedures for the code; whistle-blowing provisions; a position description for the president; and position descriptions for the board and committee chairs.

These are now requirements for publicly listed companies all over the world and leading not-for-profit institutions. Is Penn State or are other universities immune from such best practices?

If these governance and ethics oversight practices exist, they should be documented and accessible on Penn State’s website. That they are not leads me to believe they are ineffective or non-existent. (Note: the Penn State website appears to have changed slightly as of Sunday, November 13, 2011, to include backgrounds of 32 (was 35) directors.)

Next, more to the alleged sexual assaults on campus property by football coach Sandusky.

There needs to be greater rotation and succession planning at many universities and Penn State is no exception. The same director, employee, coach, dean, or otherwise at the helm for 20-30+ years – regardless of performance or money or donations being attracted – is wrong governance. Joseph Paterno was coach for 45 years and is 85 years old.

Inadequate succession planning like this would never fly in public companies, where CEO tenure is 4-5 years and good board tenure is 9. People don’t have time to get comfortable and start capturing people but need to do their job. On boards, retirement age is 72+ and good tenure is 9. In professional service firms, it is even earlier, from late 50s to early 60s to make way for the next generation of leaders.

No one is irreplaceable or larger than an institution. Incumbents create power and fiefdoms, currying favors – such as free sports tickets and equipment to young boys (as was alleged) – or protecting colleagues (also being alleged) – where they become so dominant they cannot be resisted, within pockets of toxic culture and risk – with management and even boards of trustees acquiescing instead of governing.

All allegations have yet to be proven, but if true this is likely what happened here: People become afraid to speak. If they speak, they will suffer enormous reprisals, even loss of their jobs or banishment. The board is at fault if this is the case as a result of a flawed structure (see above) and decisions it took or did not take.

At least half of the Penn State board should be businesspeople with clout. The board should have the same transparent recruitment that companies how have, with directors who are independent, have run businesses and can tell colleges who are behind the times, or who resist reform, that this is what has to happen. Having alumni, the governor, or even agricultural societies (likely a historical artifact) appoint or elect directors does not necessarily result in competent directors being at the table or staffing key committees. There needs to be a greater link – clear and transparent – between directors, their skills, and what is required to govern. The days of ceremonial appointments should be over. Clearly they are not.

Next, all colleges should have whistle-blowing procedures at the same level or above as companies are now obliged to do. This puts the heat under management to have proper procedures, as employees can go directly to an external ombudsperson or the regulator to get protection.

A code of conduct should be developed by all colleges and universities, as is the case for any leading organization. It should be signed off on by each and every trustee, employee and key supplier and be a condition of serving and employment, including for the president. Code compliance should be part of the president’s contract. Everyone has to sign that they do not know of any wrongdoing, directly or indirectly, anywhere on campus, every year. The sign-off statement should include obligations on how to report, protection mechanisms, and assurances of a proper independent investigation.

All code compliance should be reported directly to the audit committee of the Penn State board (note: non-existent at Penn State), and independently assured. The code must include conflicts of interest statements, treatment of assets, fair dealing and harassment. Training and education should also occur, for each employee. The code should be paramount and override defensive union agreements or guises of academic freedom.

Lastly, Penn State’s internal audit charter – if it exits – should be available on its website. The design and effectiveness of internal controls, including approvals, access to restricted rooms, campus security and lighting, keys, locks, areas of vulnerability, and potential for override – most of which were likely deficient in this case – should be reported directly to and overseen by the audit committee.  The audit committee should be able to insist upon independent assurance for any risk, based on the audit report. Good audit committees know and do all this. They direct the president, CFO and finance and risk personnel to comply with best practices.

Why would Penn State management do all this, under this resistance? Simple. The board tells them to. Or they get fired. This is why a strong board is so essential. The tone at the top starts – and stops – with the board. Sandusky is not a rogue any more than a rogue trader is at a bank. He is operating within a defective system, put in place by defective management and overseen by a defective board.

Conclusion: Reform to collegiate governance

Educational institutions are complex organizations, with interdependent stakeholders and many moving parts. They are sometimes more complex to run than a large company. In the vast majority of cases, they are staffed by committed and well-meaning people. They are however, hard to manage and especially difficult to govern, given defensive unions, historic tradition and tenured, specialized academics and staff. They are however taxpayer-funded entities from which leadership and accountability are expected. Indeed, they are supposed to set the example and practice what they teach.

It is very important that governance standards and practices be current and not myopic, and this is why colleges need strong, proper, effective independent boards to counteract resistance, have the clout to direct management and staff, and impose proper governance, risk management and internal controls are is being done for public companies.

Here, Penn State, and perhaps many other universities have much to learn.

Save and Share
  • Print
  • PDF
  • email
  • LinkedIn
  • Twitter
  • Facebook
  • Reddit
  • del.icio.us
  • StumbleUpon
  • Add to favorites
  • RSS

How to do a Proper External Governance Review

Several memories stand out in my mind. One was the dominant CEO who kept interrupting me to tell me how effective he and “his” board were, despite the results of my assessment (the board was in trouble with regulators and the CEO later resigned). Another was a very rich and famous director asking me to leave his office when we debriefed on the peer review, which indicated how his colleagues thought his contribution was (he too resigned). Another was a director who actually resigned from the board ahead of time when he found out I was going to do the review. Another was a chair of a governance committee who viewed my questions and then decided against the review (he was later in the news for conflicts of interest).

I have seen all kinds of data – directors who fall asleep at meetings. CEOs who pound tables. CEOs who funnel information. Directors who are out of their depth. Directors who despise other directors. Bullying, cliques, factions, sexism, conflicts and denial. The most fascinating to see is the body language, process, relationships and dynamics – this can make or break a board regardless of structure and protocols. An intransigent CRO or ineffective risk committee chair can take a company down. I have also assessed some of the best performing boards – led by terrific leaders who spend enormous effort at getting the directors and process just right. Some have won national awards over the years they are this good.

So I have learned a thing or two.

First, why would any board do an “external” board review to begin with? Simple. Self interest. Boards assessing themselves are analogous to students marking their own exams. It’s an inherent conflict to assess your own work. It’s even worse if management facilitates (which happens most of the time a “self” review occurs) as they have the most to gain by a soft review.

Mainly, however, external board reviews are often poorly done. Why? What makes for a good external board review? Four key elements.

Who is your client?

Your client is not the CEO. Your client is not the general counsel, nor the corporate secretary, nor any manager for that matter. Management should not unduly influence you. If you conduct a governance review properly, the board’s interests could – and sometimes should and will – be adverse to those of management, whom the board controls. So you cannot act for simultaneously opposing interests. If you or your firm does business with management, or seeks to, or has done in the past, you are conflicted and should not do the review, no matter how you try and justify it.

Your client is also not the chair of the board. A good board review will assess the chair’s performance and he or she is likely the problem if the board is poor.

Your client is the chair of the governance committee (or its equivalent) and the committee as a whole, who ultimately reports to shareholders, similar to the auditors being accountable to the audit committee. If you do not work directly with and for the governance committee, the review will likely suffer or fail.

Garbage in, garbage out

Next, the review itself. The vast majority of approaches are superficial, do not reflect best practice, use the wrong scale, and are heavily biased. Unfortunately, these “surveys” get perpetuated and become the lowest common denominator. They are highly imprecise and lead to misleading results. There is a false sense of reality when the board “agrees” or “strongly agrees” to a majority of ill-defined performance metrics. The analogy to “happy-face” questionnaires is “pat-on-the-back” interviews. The tough questions are not asked. Garbage in, garbage out.

The review becomes a one-sized-fits-all, check-the-box pain, no better than those of external proxy advisory firms and rating agencies. Directors are skeptical, rightly so because what really matters is not being measured and what doesn’t, is.

The board of a bank is not the same as the board of a hospital or high tech start up. Reviews need to be best practice, highly customized and rigorous. They need to span silos. Lawyers, accountants and HR advisors don’t have competencies and market permission in cross-silos and default professionally to their own. The result is even more compliance box ticking.

Importance of 360 feedback

If a board is stale, management knows. If a director is not performing, directors know. Each director and reporting management should opine on other directors. The learning and self-development here is tremendous and counteracts a board or director thinking it, he or she is better than others do. It is very difficult if not impossible to do a peer assessment internally. Directors default into collective group-think (read: denial) because their responses will be seen internally. A good peer review will include self, peer and board assessment, so each director knows how he or she performs relative to his or her own perception, other directors and the board as a whole. The peer review becomes a developmental and recruitment tool.

Reporting and debriefing

The vast majority of reporting to shareholders by listed companies on the results of board reviews is superficial, wholly inadequate and boilerplate. Companies have a vested interest in making it this way. This will not change without greater prescription and enforcement by market regulators and shareholders.

Perhaps the weakest link for a board review however is internal reporting, implementation and action taken based on the review. Feedback should be provided to each individual director and debriefings should occur to discuss development, priorities, committee and board reform, and tough discussions – such as leadership and behavioural change and director retirement. A good review effects behavior. Some of the areas include knowledge of the business, judgment, communication, integrity, constructive challenge, willingness to act, thinking skills, financial acumen, interpersonal/teamwork and commitment. These are very specific skills that each director needs to possess for the board to be effective. One director can unduly effect an entire board dynamic and decision. Governance committee chairs and I have had tough but candid discussions and coaching with directors based on these outputs. It is not easy, but without proper data and leadership, directors on the receiving end either refuse to acknowledge or do anything about it.

Conclusion: Canada ahead of US and UK but they are catching up

To Canada’s credit, we have had to explicitly recruit and assess individual directors on the basis of competencies and skills since 2005. We have also not had enormous financial failures or bailouts. The US and UK have. Imagine that – up until the financial crisis, you could sit on a risk committee of a US bank and not know anything about risk or banking. The US changed its rules since then, mandating that directors need to disclose their qualifications, which is not quite the same as directors being assessed. (Lawyers will skillfully list bullet points about past job titles.) In the UK, their Code has changed stating that for all listed companies, external board reviews have to occur at least every two or three years. This is a step in the right direction.

Save and Share
  • Print
  • PDF
  • email
  • LinkedIn
  • Twitter
  • Facebook
  • Reddit
  • del.icio.us
  • StumbleUpon
  • Add to favorites
  • RSS

The Boardroom of the Future: Changes that will reshape corporate governance

A global “mega-cap” company recently asked me to submit a briefing on how a boardroom of the future will look. This is an abridged summary of my report.

Democratization of governance

Your shareholders will nominate and elect your directors by electronic voting directly on your website. They will base their vote on the accomplishments of each director and track record of acting in the best interests of shareholders and the company overall.

Electronic registries and meetings will be the primary basis upon which shareholders select directors to your board. Director competencies will be fully disclosed.

Diversification of boardrooms

Your board will be 40% to 50% women and have far fewer CEOs on it in the next five to seven years. Your directors will be independent experts within their relevant strategic domains, will be quick studies, and will have access to the best learning of the company. They will request an Office of the Board be established. Board tenure will not exceed 9 years.

Corporate reporting

Reporting to shareholders will be fully integrated and online. Non-financial risks and internal controls will be independently assured. All reporting will be accessible, complete, accurate and independently validated.

Technology

Your board will be paperless and directors will have access to any piece of information they need to oversee and advise management. Technology will be used to attract and communicate with international directors. Risk appetite frameworks, established by the board, will translate into clear incentives and constraints using integrated firm-wide information systems.

Executive compensation

Executive compensation will be established by shareholder-directors. Professional standards will be imposed on any consultants retained by these directors. All compensation will be fully risk-adjusted and linked to performance. Current models and methods will change significantly.

Office of the Board

An Office of the Board will be established. It will house independent staff and resources available and accountable to the board and paid by the company.

 

Regulation of corporate governance

The unprecedented intrusion into the governance of companies will continue until most or all of the above reforms are implemented.

Conclusion

The above changes are significant and will fundamentally change the way directors are selected and how boards control management.

Save and Share
  • Print
  • PDF
  • email
  • LinkedIn
  • Twitter
  • Facebook
  • Reddit
  • del.icio.us
  • StumbleUpon
  • Add to favorites
  • RSS

There needs to be greater scrutiny of director independence – what it means and who decides

This will get legal and technical very quickly in the first few paragraphs, but please bear with me. Director independence is an important area and worth blogging about.

Currently, if a director has no direct or indirect “material” relationship with the company, that director will be independent. Whether a relationship is “material” is a relationship that could, in the view of the board of directors, be reasonably expected to interfere with the exercise of that director’s independent judgment.

A board determining director independence is a subjective view, i.e., the view is based on the board’s judgment. The board’s judgment need not be objectively reasonable, which is otherwise known as a “reasonable person” standard. Also, there may not be a line of analysis that could reasonably lead a board from the factors it considered to the conclusion it reached.

After the 2008 financial crisis, the Ontario Securities Commission (OSC) attempted to introduce a reasonable person standard for director independence, by stating that boards should “have regard to all relevant circumstances” in their analysis, and that a director should not have a relationship with the company or its management that could be “reasonably perceived” to interfere with that director’s independent judgment.

These amendments were strenuously resisted on grounds it would limit the director pool, and that outsiders lacked the collective experience and specific knowledge of the director in question.

The OSC has yet to address this issue of director independence.

If a director is conflicted or his or her independence is compromised, the board should not be the sole arbiter of this. Directors are a small group, subject to group dynamics and peer pressure, and should not be immune from external scrutiny of perceived conflicts or of compromising of independence that incorporate best practices. Conflicts of interest are handled within an organization by people independent of the decision.

Take two good examples of how director independence could be better scrutinized, from an objective point of view: (i) director tenure and (ii) director relationship to a significant shareholder or control person.

Director Tenure

There is a reasonable belief that exceeding a certain number of years on a board promotes greater closeness to management and less objectivity in performing an oversight role of management. The UK for example has a “nine year rule” for all its listed companies, beyond which a director’s continued independence is called into question. There is also a reasonable view that board renewal and diversity of skill sets is a good thing, in that a younger director with a more relevant skill set need not be blocked by a senior director who refuses to retire.

When the UK rule is applied to Canada, several directors fail. Of five chartered banks, for example, thirty (30) directors exceed nine years with several directors on bank boards for almost 20 and up to 25 years. Directorship should not be a lifetime appointment. These directors need to let go and make room for the next generation of directors.

The debate on whether long-serving directors are reasonably perceived to be independent from management, and if so, how, and how the views of their fellow directors have been canvassed, is one worth having, juxtaposed with objective views of scholars, rating agencies, diversity candidates and shareholders on this topic. The UK incorporates both the subjective and objective views.

Director Independence from a Significant Shareholders or a Control Person

Right now a control person or significant shareholder is not disqualified from being independent.

It is equally reasonable that an appropriate proportion (subjectively and objectively defended) of a board be comprised of directors who are independent –subjectively and objectively– from both management and a significant shareholder or control person, if either exists. Numerous Canadian companies have a significant shareholder or control person. (A significant shareholder or control person could own a majority of voting shares and control the directors’ appointment to a board.)

It may also be reasonable to consider the control person or significant shareholder’s nature and degree of involvement or influence over management for purposes of assessing independence.

A related party (insider) transaction can occur such that directors elected by and from the non-significant shareholder or control person may also be reasonable, to oversee such transactions.

Will the debate occur?

Long-serving directors and significant shareholders or control persons have vested interests and their view is worth hearing. It should not be the case however that the debate does not occur or other views are excluded from an unbiased assessment of director independence, what this means and who decides.  It is important to get director independence right as much of corporate governance effectiveness hinges on the independence and ultimate effectiveness of directors.

Save and Share
  • Print
  • PDF
  • email
  • LinkedIn
  • Twitter
  • Facebook
  • Reddit
  • del.icio.us
  • StumbleUpon
  • Add to favorites
  • RSS