CHAPTER 5
Do Things Right
The Process of Great Governance
A board is a work group—senior, consequential, and privileged—but a work group nonetheless. As such, the determinants of its effectiveness are well known: people, structure, and process.
Work Group E[f_f]ectiveness
People
Directors themselves are surely the most important determinant of board effectiveness. Dean Gilbert R. Whitaker Jr., my boss at the University of Michigan, once remarked that if you have a great faculty, they attract great students. Gather them together under a tree and great education will occur. In general, I think the same is true of boards. Great directors will, most of the time, be good stewards, whatever the board’s structure and process.
How important is the human makeup of the board? The late J. Richard Hackman, arguably the world’s leading expert on teams, advised that individuals destructive to a group’s work process literally must be forced off. I have certainly found this is true on boards. Hackman came to other conclusions that board chairs, lead directors, and governance committees should keep in mind. For example, small teams are generally more effective than large ones; he suggests no more than nine members. Stable and experienced teams are more effective than those newly formed or comprising many new members; this should give pause about mandatory retirement based on term limits or age. Deviants (members with critical and different perspectives) are important to challenging the team’s tendency toward homogeneity.36
Still, we can’t rely on boards comprising only great directors, and there are too many instances in which well-qualified directors have overseen (or overlooked) circumstances that led to disastrous results.
Structure always matters. Variables such as board size, committees, and leadership roles deserve attention, but as we will see in the next chapter, research relating board structure to effectiveness as measured by organizational performance is disappointingly equivocal in its findings.
If structure is important but inconclusive when it comes to board effectiveness and great directors are highly desirable but not a guarantee of great governance, we are left to consider process as a critical determinant of governance success. By process, I mean how individual directors and the board as a whole go about their work.
Process Basics
Board process is a difficult subject because it is invisible except to the participants, and most boards meet privately. Allow me to draw on my experience as a director to offer two top-ten lists of process basics.
Process Advice for Individual Directors
1. Prepare. Read meeting materials, identify issues for clarification and discussion, and come with questions and comments.
2. Attend. There is no substitute for in-person attendance.
3. Participate. Make value-added contributions.
4. Initiate. Raise consequential issues for consideration.
5. Be diligent. Work hard to understand the company’s mission, strategy, plan, and performance as well as its financial position, people, and culture.
6. Be financially literate. Learn enough accounting because it is the language of business.
7. Be both tough and helpful. Maintain high standards, don’t tolerate poor performance, assist in solving problems, and be supportive.
8. Be constructive. The goal is to get the work of the board and the company done, not win arguments.
9. Be candid and courageous. It’s a director’s fiduciary duty and your goal is respect, not popularity.
10. Don’t meddle, dominate, engage in ego contests with other directors, or stay too long.
Process Advice for the Board
1. Elect and appoint leaders (chairman, CEO, lead director, committee chairs) who are strong and inclusive. Maintain unity of command: the board is accountable to shareholders, CEO to the board, management to the CEO.
2. Design agendas—annual and for each meeting—to ensure vital business gets done while leaving white space for matters arising and discussion.
3. Delegate but don’t abdicate to board committees.
4. Decide on the board’s information needs and require management to meet them.
5. Insist on excellent staff work to support the board.
6. Act with a sense of urgency, but deliberate important matters as long as necessary.
7. Have a good mix of meetings with management present and executive sessions of directors and independent directors only.
8. Get out of the boardroom occasionally: hit the road to observe, learn, and show the flag.
9. Combat staleness with education sessions, board and director evaluations and feedback, and planned turnover and succession.
10. Don’t meddle, play politics, tolerate an incompetent director or executive, or ignore smoke (there’s usually a fire).
These process basics are time-tested and essential. However, they guarantee neither great results nor avoidance of disaster. Deeper insight is required.
Digging Deeper on Board Process
This book’s starting point was the story of two companies that have had striking success since they created formal governance arrangements over two decades ago: Gordon Food Service in 1988 and Equity Residential in 1993. Directors deeply desire that the companies they govern thrive on their watch. I can report from personal experience that it’s gratifying when this occurs.
Thriving is the hoped-for condition, but an essential pre-condition to thriving is surviving. This means avoiding catastrophic losses and setbacks. Great governance requires a board process that maximizes the probability of the company or organization both surviving and thriving.
What are the process requirements for this to occur? In my experience, the boardroom must be a place of good information, rational analysis, quality discourse, careful consideration, and wise judgment. This requires directors who are intent on perceiving, thinking, and working this way and a process that facilitates it. My purpose in this chapter is to convey what I know, suspect, and believe about boardroom process that can result in great governance. My views are informed by experience along with insights from behavioral research that are transforming the fields of economic, finance, and accounting.
I had a remarkable intellectual experience while doing research for this book and teaching a course on corporate governance at the University of Illinois. The textbook for the course was Corporate Governance Matters (CGM)37 by David Larcker and Brian Tayan of the Corporate Governance Research Program at the Stanford Graduate Business School. CGM is an excellent compendium of empirical research, primarily from finance and accounting, on the relationship between many governance variables, mainly structural (e.g., board size, whether chairman and CEO are combined or separate, the independence of directors, etc.), and the financial performance of the companies being governed. The results of this research can only be called disappointing, which the authors admirably acknowledge.
While I taught the course and absorbed CGM, I also read Thinking Fast and Slow,38 an impressive exposition of the findings and implications of behavioral research by Daniel Kahneman. Kahneman is a Princeton psychologist who won the Nobel Prize in Economics in 2002 for his conceptual work and experimental research with a late colleague, Amos Tversky. Insights from their work have challenged the long-standing rational man assumption of economics. Kahneman was the first non-economist to win the Nobel Prize in Economics.
I found that CGM, a book about governance, offered me relatively few insights as a corporate and nonprofit director on how to improve governance due to the equivocal nature of the research findings on which it reports. By contrast, Thinking Fast and Slow, which has nothing explicitly to do with governance, was rich with insight and useful ideas! Why? Because while the research variables in CGM are primarily structural and among the least important in affecting board performance, the variables in Thinking Fast and Slow involve people and process: perception, cognition, bias, and the consequences for how individuals and groups work and make decisions. These are the elements of board process that most affect governance effectiveness. Allow me to illustrate.
Individual and Organizational Dynamics
An influential book I read as a young man was The Best and The Brightest by the late journalist David Halberstam.39 It is a riveting account of how some of the most talented men of the post–World War II era led the United States into the Vietnam War, then repeatedly doubled down until a humiliating defeat ended the nightmare. Hundreds of thousands of lives were lost, including 58,000 Americans. Billions in treasure were expended, and confidence in the veracity and competence of government was permanently shaken. The book instilled in me a career-long fascination with senior-level decision-making and an acute awareness that smart and able people, with the best of intentions, can make terrible decisions. Halberstarm’s account of President Lyndon B. Johnson, Secretary of State Dean Rusk, Secretary of Defense Robert S. McNamara, General William Westmoreland, and others planted two ideas firmly in my mind:
1. Individuals have preconceptions and preferences, rooted in their personal histories, that deeply affect perceptions, create biases, and influence judgments in ways of which they are largely unaware.
2. The organizational context in which leaders operate can distort the information on which they rely and amplify their biases in decision-making, rendering them isolated and dependent.
Individual Dynamics. The first insight—about individuals’ preconceptions and preferences and their consequences—has been reinforced for me in thirty years of senior management and board experience. The Best and the Brightest offers unforgettable illustrations. Here’s one of them.
Rusk and McNamara were hawks who fought doves like McGeorge Bundy (Johnson’s National Security Advisor) to influence Johnson on war policy. According to Halberstam, they learned to present tactical alternatives to Johnson (e.g., on strategies for bombing the north) with their preferred course of action placed squarely in the middle of two extreme options, creating the impression that it was the moderate or prudent option. This played well to Johnson, who was torn over war policy. Reluctant to escalate enough to win (if winning was possible) but unwilling to “turn tail,” Johnson was drawn to the middle course. As a result, Rusk and McNamara usually got their way. This led over time to a gradual but vast escalation in the war which would undoubtedly have appalled Johnson had it been presented in a single plan.
Thanks to behavioral research, the effects of such framing on perception and decision bias are far better understood now than then. Another phenomenon on which such research has shed light is anchoring. It played a big role in Vietnam War decision-making. Johnson and his contemporaries were steeped (anchored) in lessons from World War II and the Korean and Cold Wars. The lessons were that aggression must be stopped or it will escalate into broad domination and expanded conflict, and Communism is an aggressive and expansionary international force.
When these leaders looked at the situation in Vietnam in the 1960s, they saw, analogously, Germany and Poland in 1939, the Soviet Union and Eastern Europe following World War II, and China and South Korea in 1950. As they tried to make sense of what was happening (sense-making is a much-studied behavioral concept), what Kahneman calls their System 1 (fast) thinking kicked in immediately to give them guidance: Communist aggression must be stopped. Those opposed to the war suggested alternative anchoring (e.g., Ho Chi Minh as the George Washington of Vietnam, fighting for freedom.) But they were derided, illustrating another behavioral concept that affects perception and judgment: the recency effect. When asked to recall a list of items in any order (free recall), people tend to begin with the end of the list, recalling those items best. Thus, more recent events (like international Communism) likely had a larger anchoring effect than those in the distant past (like the American Revolution).
These illustrations demonstrate the profound ways in which framing, anchoring, bias, and types of thinking can affect judgment. Here is another, much simpler example I’ve used with my students.
Like most people, I drive exactly the same route to work every day. One morning recently, I saw that my usual route was blocked due to road construction. So instead of going east, south, and east again to get to the parking lot, I went east, north then east again. After my second turn to go east, I came, as usual, to a stop sign at the road I always cross, but this time a few blocks north. I stopped, looked both ways, then proceeded … and immediately hit an oncoming car. (It was a low-speed impact, and, fortunately no one was hurt.) Why? Because on my usual route, the intersection is a four-way stop, while the intersection to the north is two-way. Only I had to stop. The oncoming traffic had the right of way.
My mind was engaged in Kahneman’s System 1 (fast) thinking, and why not? I have commuted to work hundreds of times. On mental autopilot, I proceeded on the usual assumption that the oncoming cars would stop. Nothing about the situation triggered Kahneman’s System 2 (slow, careful, deliberate) thinking. If it had, I would have been more aware, considered, and careful, recognizing that, appearances to the contrary, this situation was novel and the risk level was elevated.
I have reflected that had my wife been with me, I probably would have avoided the accident. Why? First, I’m more watchful and careful when driving with others for reasons of pride and safety. Second, there would have been two brains at work on the matter instead of one. (I don’t want to say that my wife doesn’t trust my driving, but she doesn’t. So she pays attention and alerts me.) Two or more minds often are better than one.
Surely this is a reason that commercial airliners always have two pilots and why juries rather than judges alone determine guilt versus innocence in criminal matters. (Watch the classic Henry Fonda film Twelve Angry Men with this in mind.) This is why CEOs (individuals) report to boards (groups) in all public companies and, by choice, in many private companies as well.
These practices result from our understanding that the perceptions and mental processes of one individual are more likely to lead to faulty decisions than the multiple perceptions and mental processes of a group that deliberates and decides together. While there is evidence that groups, on average, make better decisions than individuals, the process is not automatic, as Halberstam illustrates in The Best and the Brightest. The group needs a diversity of perceptions and perspectives and a process to bring them together to produce good decisions. This is certainly the case for boards.
Organizational Dynamics. Boards must concern themselves not only with matters that affect individual perception and judgment but also with organizational dynamics that can affect directors’ decision-making. Two risks are noteworthy in this regard.
First is the quality of information on which directors rely. Most boards have some version of a dashboard (a visual summary) of results that supports their monitoring duty. But how reliable are the gauges? In The Best and the Brightest, Halberstam describes the hazards leaders faced through the paradox in Vietnam of the United States supposedly winning battles and body counts by overwhelming margins while consistently losing the war. Pressures and incentives for performance in the U.S. military and Foreign Service led to selective and inflated reporting and optimistic interpretation of results (i.e., gauges with faulty readings).
In this vein, I once had a direct report who had a bad habit of beginning every meeting with “Joe, I have some good news for you!” It made me uneasy.
After a while, I said to him, “That’s great, but why don’t you tell me first the things you’re concerned about since we’re here to solve problems. Then if there’s time for good news at the end, fine, or you can just shoot me an e-mail.” It was my way of encouraging more balanced and less selective reporting of what was going on in his area of responsibility.
The second risk of which directors must be aware is the tendency of organizations and their leaders to fall in love with courses of action that may not be optimal or wise. Acquisitions and mergers are a prime example, as discussed previously. This helps explain their overall poor record of value creation for the acquirer. Here’s another example: the high bay stacker.
Inventory management is a big issue in every manufacturing company. It was at Cummins when I worked there. Inventory matters because if there’s too much, it’s expensive and wasteful, but if there is too little or the wrong kind, it costs sales and hurts customer service. There has been a sea change in attitudes about manufacturing inventory in recent decades, from manage it well to minimize it. The change has been driven by the shareholder value and quality movements and facilitated by information technology and transportation systems.
When I joined Cummins, there was a lot of talk about a proposed high bay stacker for the Walesboro, Indiana, components plant that machined parts for Cummins engines that were assembled a few miles away in Columbus and elsewhere. This piqued my curiosity. I learned that the stacker was a multi-million dollar physical system in combination with a sophisticated information system to accommodate (stack) inventory by using air space (high) in one section (bay) of the plant. The stacker would accommodate far more in-process inventory than normal floor storage. In those days, manufacturing’s view was the more inventory, the better. Abundant inventory helped smooth production and minimize stock-outs that disappointed customers and distributors. The manufacturing organization had fallen in love with the high bay stacker concept and pushed Jim Henderson, the president, to secure board approval for the multi-million dollar capital investment.
Jim appeared ready to agree but then held off. Tension grew. Ultimately, manufacturing management did not get their heart’s desire. The president insisted on a fundamental change in manufacturing thinking about inventory. Too much of it was not only costly, it also undermined quality by serving as a crutch when defective parts couldn’t be used by the assembly line. Instead of storing more inventory efficiently, Jim established a goal of drastically reducing inventory through high-quality, lean production and just-in-time delivery of components to the assembly line. Manufacturing was not happy, but it’s clear, in retrospect, that the president was right.
Boards and executives must be alert that organizations do, indeed, fall in love with ideas and initiatives. Powerful momentum develops behind proposals requiring their approval. Succumbing is the course of least resistance. Doing so minimizes conflict and makes people happy, at least in the short term, but it’s not always the right thing to do.
There are organizational dynamics beyond distorted reporting and falling in love with initiatives that impair board decision-making. Functional dominance can be a problem. For example, finance organizations became too powerful relative to design (product) and manufacturing (cost and quality) in U.S. auto companies, contributing to decades of decline. Financial incentives that are too weak, too strong, or poorly targeted can also cause trouble. This may explain the infamous statement by Charles O. Prince, then CEO of Citigroup, in mid-2007 at the peak of the financial bubble. Explaining the company’s aggressive lending to private equity for leveraged buyout deals at nosebleed prices, he said, “As long as the music is playing, you’ve got to get up and dance. We’re still dancing.”40 The lending produced enormous fees that bulked up profits on which annual bonuses were based. Shareholders and taxpayers were left holding the bag when loans soured.
Wise boards need to be attentive and on guard about the effects on major decisions of these and other organizational dynamics.
Behavioral Research
Economists used to assume that human actors are rational and omniscient and, therefore, able to make unbiased decisions based on complete information and objective reasoning to maximize their utility. This simplified and simplistic view of human beings led to the notion of markets as perfect, with prices the result of millions of decisions accurately reflecting all relevant information, adjusted instantly and continuously.
These assumptions may be convenient for economic analysis, but they don’t square with the decision processes of real human beings. Our decisions are subject to emotions and biases, inaccurate perceptions, ignorance, and misinformation. A perfect markets view doesn’t square with the obvious manic-depressive character of markets at times and their tendency toward excesses like bubbles and busts. Yet the assumptions persisted in economic analysis for a long time because of inertia, the benefits of simplification in modeling, and the lack of a well-grounded alternative. Gradually, though, things have changed. Thanks to scholars like Kahneman and Tversky, we now have a richer understanding of how people perceive and process information and make decisions. In the real world, directors and executives can be much more aware of impediments to the boardroom being a place of good information, rational analysis, quality discourse, careful consideration, and wise judgment. Foremost among the impediments are bounded rationality and motivated reasoning.
Bounded Rationality. One of the first scholars to question the omniscient, rational man assumption and offer a useful alternative was Herbert Simon, an organizational scholar who articulated the concept of bounded rationality.41 Simon concluded after examining executive decision-making that people are not hyper-rational supermen and superwomen with unlimited deductive ability assumed by traditional economic models.
Rather, there are serious limits (bounds) to their rationality. This leads them to use simplifying assumptions and decision rules that we now call heuristics when making judgments and decisions. Many heuristics are useful and necessary, allowing us to conserve scarce cognitive resources. One scholar refers to most human information processing as fast and frugal.42 But heuristics can lead to errors and unintended biases, as illustrated by examples of Vietnam War decision-making in The Best and the Brightest and my judgment error in a novel situation that led to an auto accident.
I have noted in this book the importance of financial statements in enabling boards to monitor management and evaluate performance. Flowing directly from bounded rationality is a large stream of literature on format effects in financial reporting, which finds that more saliently displayed information receives more attention and is weighted more heavily in judgments and decisions. For example, one study43 found that explicitly disclosing information (e.g., in an earnings announcement or press release) improves market efficiency, even when these disclosures are redundant. In other words, just repackaging value-relevant information that is already available and highlighting it can improve market efficiency.
In similar fashion, experienced directors know that how management presents financial information to them can have a powerful effect on their perception of performance. Research shows that we subconsciously pay attention to information that is presented more prominently, but the most prominent information may not be the most important.
Bounded rationality reminds directors that neither they nor management can know and process all relevant information, even on important issues. This is why the best directors ask questions much more frequently than they give answers or make declarations. Questions open pathways and can bring new information and considerations to bear on consequential matters. “Have you thought about x, y or z?” is an excellent governance question for this reason.
In retrospect, I think that Irwin Miller at Cummins was making reference to the problem of bounded rationality when he once said to me, “In management, you don’t have to worry about what you already know. You’re working on it. Worry about what you don’t know.” This is true for directors as well.
Motivated Reasoning. Encouraged by the pioneering work of Herbert Simon and his Nobel Prize in Economics in 1978, intellectually adventurous scholars in economics, finance, and accounting began to examine the findings of experimental psychologists and consider what the insights might mean for their own research. The results have been transforming these fields for the last two decades.
The most important finding of which directors should be aware is the concept of motivated reasoning. It occurs frequently in boardrooms and may explain better than anything else how and why groups of smart people can make inferior and occasionally truly awful decisions.
While bounded rationality establishes that we are unable to process all available information, subsequent research has established that we selectively search for and process available information that supports our preferences and goals. This is motivated reasoning: analysis based on selective versus unbiased information and weighting that information in ways that lead to decisions consistent with our goals and preferences.
We know that individuals have preferences and goals and that these are deeply rooted in their psyches. For example, looking at the same information, the majority of white Americans were convinced that O.J. Simpson was guilty of murder, while the majority of African-Americans believed he was innocent and concurred with the not guilty verdict.44
Motivated reasoning is a generalized bias to which we are all susceptible. When information is consistent with our preferences, we tend to accept it at face value and don’t search for contradictory evidence. But when information is not consistent with our preferences, we actively scrutinize and attempt to discredit it. This happens even though we think we are being objective. For example, one scholar found that investors in long stock positions forecast higher earnings than investors in short positions when given identical information.45 (A long position benefits from a higher future stock price, and a short position benefits from a lower price and earnings presumably drive prices.) In this context, it’s easy to see why a board hungry for topline growth could be persuaded to make a big acquisition at an inflated price, especially when management musters selective evidence to support the action and price.
Here’s a boardroom-relevant example of motivated reasoning from Thinking Fast and Slow: “He likes the project, so he thinks its costs are low and its benefits are high. Nice example of the affect heuristic.” The affect heuristic means that the individual is influenced in decision-making by his feelings about the project, something we’ve all observed and experienced.
Motivated reasoning can occur in important personnel decisions, such as hiring a CEO or evaluating candidates for the board. I remember being shocked when I first read research on the employment interviewing process, which found that subjective impressions in the first few minutes of an interview lead the interviewer to have a positive or negative disposition toward the candidate, which she then spends the remainder of the interview reinforcing with selective solicitation and interpretation of facts. I’ve certainly seen this happen in CEO and director interviews. This is not to say that chemistry doesn’t matter; people have to like and respect each other well enough to work together. But motivated reasoning in this context can lead to problems, such as excessive emphasis on weaknesses in internal candidates the board knows well and too much emphasis on positive attributes and glossing over shortcomings of outside candidates.
Another form of motivated reasoning is found in the self-serving attribution bias. When things go right, we attribute the success to our talent and hard work, but when they go wrong, we attribute the failure to things beyond our control such as the situation or bad luck. I’ve observed that golfers tend to claim ownership of their pars and birdies, feeling they are a proper reward for skill, practice, and persistence, but they often disclaim bad shots, holes, and rounds, attributing them to distractions, a new swing or anything but their own ineptitude
Self-serving attribution bias is often at work in the boardroom. When results are good, management and the board tend to take credit. When results are poor, management may point to factors beyond their control (e.g., the economy, the weather, or irrational competition). Right or wrong, good results don’t require much explanation. Boards and management accept them as confirmation of their judgment and skill. Poor results, however, unleash rationales. Recent research suggests that even experienced financial managers—professionals on whom boards rely for accurate scorekeeping and even-handed interpretation of results—give greater weight to internal than external factors in explaining past good performance.46 This leads them to be overconfident in future performance and more likely to issue optimistic forecasts.
The self-serving attribution bias poses a challenge for governance. A great board gives credit where it’s due for good results and accepts valid rationales for poor results. But directors also understand that luck and factors beyond management’s control play a role in both cases. Only the pattern of performance compared to the competition over a sustained period really tells the story of how management is doing.
The bottom line for directors on motivated reasoning is that we humans search for and process information in ways that support our preferences and goals as well as our view of ourselves as smart, good performers. We screen out or rationalize information that is incompatible with this view. Whether practiced by management and accepted by the board or practiced by the board itself, these tendencies can seriously undermine the objectivity and rationality required for directors to make wise decisions and accurately evaluate performance.
System 1 and System 2. Central to Kahneman’s work in Thinking Fast and Slow are the System 1 and System 2 constructs. System 1 thinking is fast. It is how we make most day-to-day decisions and involves intuitive judgments and preferences. System 2 thinking is slow, what Kahneman calls a more deliberate mode of operation.
Every director can relate to these constructs. Much of the time, directors must process information quickly and make judgments intuitively, based on long experience. This is System 1 thinking. As Kahneman observes, “Most behavior is intuitive, skilled, unproblematic and successful.” It suffices and gets the job done. A key issue for directors is when and for what a shift to System 2 is warranted. When and why is the more deliberate mode of operation warranted? A board getting this right is absolutely essential to great governance.
Let me illustrate. Earlier in the book, I related my difficult experience as a director of a failing company. The company had been growing rapidly by opening new stores and taking on debt to finance growth. Results were good, and the stock price was rising. In my judgment we, the board, were attentive, but we were somewhat lulled by good results just as the self-serving attribution bias would predict, so our monitoring was probably more System 1 than System 2.
For me, things began to change as I tried to make sense of an emerging pattern of events: same store growth slowed, our very capable CEO resigned, I observed sloppy housekeeping in a warehouse tour, and then I encountered a stock-out of a popular product in one of our category killer stores. What cinched my thinking that we were in trouble was a debt offering that earned a junk-bond interest rate.
I found myself alone in my study several evenings in a row, deeply engaged in what Kahneman would call System 2 thinking about the situation: slow and deliberate. I put my thoughts together and went to the chairman and other directors. At the time, I felt I failed to sell my point of view about what was happening and the destructive trajectory we were on. Today I realize more deeply that I failed to persuade my board colleagues to join me in moving from a System 1 to a System 2 approach to monitoring the company: slow down, ask hard questions, consider alternative explanations, and follow facts wherever they might lead. More time would be required of the board, the work would be intense, and conflict with management would was likely. It didn’t happen. In retrospect, I think the other directors just wanted things to be okay, so they selectively perceived information to support an optimistic conclusion—an example of motivated reasoning. I was incapable of joining that point of view and, as a result, left the board. As reported previously, the company eventually declared bankruptcy.
Here’s the point: a board is a small group charged with responsibility to keep an organization on track, avoid mortal risk, and achieve good results. The board hires management to do this work. Directors monitor and assist management. Most of the time, necessarily, the board trusts and supports management and tracks big picture results. Directors challenge and comment on management recommendations but ultimately accept most of them. It doesn’t make sense to put management in charge then relentlessly criticize and deny their recommendations and interpretation of results. But there are moments of truth in the life of every board when directors need to shift from System 1 thinking about company matters to slow and deliberate System 2 thinking. At those moments, directors must supplement their normal trust and acceptance with skepticism, scrutiny, in-depth analysis, and serious challenge. Great governance requires directors to recognize these moments of truth and make the necessary shift in thinking style and board process.
Matters requiring this shift may be obvious, such as appointing a new CEO, considering a major investment, or deliberating a strategic move such as a big acquisition or selling the company. But sometimes the cue is not so obvious. As my failing company story illustrates, it may not be singular but a pattern of apparently unrelated developments or experiences that together signal cause for board concern. Skilled sense making on the part of individual directors and the board as a whole is essential. Unstructured time for directors—white space—in board meetings, executive sessions, and over meals is vital to allow them to compare perceptions and verbalize emerging concerns with their colleagues.
Kahneman’s fundamental insight, that our minds are capable of System 2 thinking but default most of the time to System 1, leads to other findings that are relevant to boards and directors.
Attribute Substitution. In his 2002 Nobel Prize lecture, Maps of Bounded Rationality, Kahneman proposed that many errors and biases arise from attribute substitution—substituting an easier judgment for a more difficult one. Two examples are hiring and investment decisions. Both are important for boards.
Hiring decisions should be based on candidate skills and experience related to the job to be done, but that’s difficult to determine. So we sometimes make judgments based on how good a speaker or presenter someone is, how attractive the person is, where he or she went to school, or personal chemistry. These are impressions based on System 1 thinking. They may be relevant to the judgment at hand. However, substituting easy and familiar criteria for harder-to-determine qualities and fit can lead to Type 1 and Type 2 errors (i.e., acting on false positive and false negative signals).
Here is Kahneman on just this situation in Thinking Fast and Slow: “The question we face is whether this candidate can succeed. The question we seem to answer is whether she interviews well. Let’s not substitute.”
Investment decisions should be based on an unbiased assessment of fundamental value and future cash flows, but that’s hard and uncertain. So we may base a decision to invest on our feelings about the brand, products, or CEO. Some research suggests that even physical proximity can have a bearing on investment decisions (i.e., we feel more comfortable investing in companies that are geographically proximate to us than those that are more distant). These factors may be relevant to good investment decisions, but when making consequential judgments, we need to guard against System 1 thinking that urges us to substitute simple and convenient criteria for the more difficult and relevant criteria required by System 2 thinking.
And More. Thinking Fast and Slow is replete with examples of how our perceptions and mental processes can make the boardroom less than objective and rational.
• Anchoring: “The firm we want to acquire sent us their business plan, with the revenue they expect. We shouldn’t let that number influence our thinking. Set it aside.”
• Availability Cascades: “This is an availability cascade: a non-event that is inflated by the media and the public until it fills our TV screens and becomes all anyone is talking about.”
• Representativeness: “The lawn is well-trimmed, the receptionist looks competent, and the furniture is attractive, but this doesn’t mean it’s a well-managed company. I hope the board does not go by representativeness.”
• Hindsight: “He’s learning too much from this success story, which is too tidy. He has fallen for a narrative fantasy.” “Let’s not fall for the outcome bias. This was a stupid decision even though it worked out well.”
• Causes and Statistics: “We can’t assume that they will really learn anything from mere statistics. Let’s show them one or two representative individual cases to make our case.”
• Judges v. Formulas: “Whenever we can replace human judgment by a formula, we should at least consider it.”
I have observed all these phenomena in boardrooms. Here is an example of causes and statistics from my experience with the board of trustees of the University of Illinois to whom I reported.
When I became president of the university, I found an $800 million backlog of deferred maintenance on our three campuses. It was growing by nearly $100 million per year. I knew if we didn’t get on top of the problem, it could reach the point of no return (i.e., certain buildings would be beyond repair and have to be razed). My predecessor and I arranged for many data-based presentations to the board on the deferred maintenance problem but to no avail.
We did not get action until I invited the Dean of the Literature, Arts and Sciences College on the Urbana campus to make a personal report on Lincoln Hall, the cornerstone building of higher education in Illinois, where she had her office and taught courses. Sarah described having to prop open windows with chairs and interrupt her lecture while a squirrel ran across the stage of the auditorium. She showed color slides of Lincoln Hall’s terrible disrepair. The board was horrified. This carried the day! We proposed—and the board approved—a program to arrest and whittle down deferred maintenance at U of I. Toward the end of my presidency, the board approved a comprehensive renovation of Lincoln Hall. As Kahneman observes, showing representative cases with stories and pictures can motivate far more learning and action than presenting statistics.
Another concept that is important in business and, therefore, for boards is what Kahneman calls judges versus formulas. Simply stated, standardized, formulaic approaches can often produce better results than thousands of individual human decisions. Take, for example, yield-management systems in the airline and hotel businesses that adjust prices in order to maximize revenue through volume (filled airline seats and hotel rooms) times price (per seat or room) depending on supply and demand conditions. It would be naive to think that gate agents and hotel clerks could do as well. The book and movie Moneyball is an entertaining portrayal of a statistical versus human (talent scout) approach to player selection in baseball.
Board Process: Brainstorm versus Debate
Insights about board effectiveness can come from unexpected places. A hotly debated New Yorker article by a controversial writer, Jonah Lehrer, “Groupthink,”47 produced a thrill of recognition in me as an experienced director even though it was not about governance.
Lehrer notes the popularity of brainstorming as a technique to improve the creative performance of human groups. He traces the origin of the term and technique to a book titled Your Creative Power by Alex Osborn, a partner in the advertising firm of B.B.D.O. in the late 1940s:
… Osborn’s most celebrated idea was the one discussed in Chapter 33, “How to Organize a Squad to Create ideas.” When a group works together, he wrote, the members should engage in a “brainstorm” which means “using the brain to storm a creative problem—and doing so in commando fashion, with each stormer attacking the same objective.”
Osborn outlined the essential rules of brainstorming including the absence of criticism and negative feedback. As Lehrer reports, “Brainstorming enshrined a no-judgments approach to holding a meeting.”
Anyone who has been around the business and education worlds for several decades can attest that brainstorming is something of a religion. Generations of students and corporate employees have been indoctrinated through exercises like the desert (or moon landing) survival problem and the Tinkertoy tower-building competition. Brainstorming is so simple and its benefits deemed so great that even kindergarteners learn it.
Brainstorming begins with a problem for the group to consider. The first and most crucial step is encouraging everyone to offer ideas on how to solve it with no premature evaluation or, especially, critical judgment of any of them, no matter how loony or ill-informed. (Osborn wrote, “Creativity is so delicate a flower that praise tends to make it bloom while discouragement often nips it in the bud.”) Once the idea generation stage has run its course, the group uses a democratic process, such as rating or voting on the alternatives, to determine the best course of action. In structured exercises like the desert survival problem, groups learn that the average of individual scores is almost always worse than the group’s scores when compared to the best solution of an expert panel.
The gist of Lehrer’s article is that these results are impressive but suboptimal. If our goal were to teach people how to attack problems in groups, he says, we would educate them on how to debate rather than simply brainstorm.
Lehrer describes a 2003 experiment by Charlan Nemeth, a Berkeley psychology professor, who gave 265 undergraduate students divided into teams of five the same problem: “How can traffic congestion be reduced in the San Francisco Bay Area?” Some teams were given brainstorming direction. Others were told to brainstorm, but with this addition: “…most studies suggest you should debate and even criticize each other’s ideas,” which they were encouraged to do. The rest of the teams received no instruction and were free to collaborate as they wished. Teams had twenty minutes to come up with as many good solutions as possible. The results:
The brainstorming groups slightly outperformed the groups given no instructions, but teams given the debate condition were the most creative by far. On average, they generated nearly 20 percent more ideas. After the teams disbanded, another interesting result became apparent. Researchers asked each subject individually if she had any more ideas about traffic. The brainstormers and the people given no guidelines produced an average of three additional ideas; the debaters produced seven.
Another experiment found that people are more creative after hearing an incorrect answer to a question. Nemeth observes, “Authentic dissent can be difficult, but it’s always invigorating. It wakes us right up.” Dissent, of course, requires dissenters, group members with a genuinely different perspective who are willing to share it.
To illustrate a kind of diversity that can be of real value to work groups, Lehrer recounts a study by Brian Uzzi, a Northwestern University sociologist, of hit Broadway musicals. Uzzi asserts that musicals are a model of group creativity, requiring a composer, lyricist, and librettist as well as director, choreographer, and so on. He wondered whether results were better with a group of strangers or close friends who had worked together before. He found that the relationship between the makeup of the group and the quality of the result was curvilinear, like many in social science. Networks with an intermediate level of social intimacy—a measure of the density of connections he called Q—produced the best Broadway shows.
Uzzi’s favorite example of “intermediate Q” is West Side Story. … The concept was dreamed up by Jerome Robbins, Leonard Bernstein and Arthur Laurents. They were all Broadway legends, which makes West Side Story look like a show with high Q. But the project also benefitted from a crucial injection of unknown talent, as the established artists realized that they needed a fresh lyrical voice. After an extensive search, they chose a twenty-five year old lyricist who had never worked on a Broadway musical before. His name was Stephen Sondheim.
I noted earlier in the book that diversity of perspectives, experience, career stage, and age is vital for boards to excel. It is for the same reason that Uzzi cites to explain his findings:
The best Broadway teams, by far, were those with a mix of relationships. … These teams had some old friends, but they also had newbies. This mixture meant that the artists could interact efficiently—they had a familiar structure to fall back on—but they also managed to incorporate some new ideas. They were comfortable with each other, but they weren’t too comfortable.
Similarly, the best boards, in my experience, comprise a core group of directors who know each other well and have worked together effectively for a long time in full collaboration with newer, usually younger board members, who unsettle the status quo and effectively challenge established views and ways of doing things. Over time, the original core group retires, novice directors become the new core of veterans, new directors are recruited and the board is renewed.
Toward Better Board Process
Great governance requires a board to do the right things and to do things right. Substance and process both need to be excellent.
Some aspects of doing things right are straightforward, as reflected in the top-ten lists for directors and boards at the beginning of this chapter. Doing these things won’t guarantee quality governance, but they lay a strong foundation.
Doing things right also means ensuring that the boardroom is a place of good information, rational analysis, quality discourse, careful consideration, and wise judgment. Of the three elements that will determine if this is the case, board structure, while always important, is the least potent. Far more consequential are people—capable and conscientious directors who are intent on working this way—and a process that facilitates it.
There are daunting impediments to boardroom rationality identified by behavioral research. From Simon’s bounded rationality to Kahneman’s System 1 and System 2 thinking and their consequences, organizational realities and the human thought process can confound even well-intentioned directors in their quest for quality governance. Many boardroom follies are best understood not as the product of evil, negligent, or inept directors and executives but as the result of perfect storms of inadequate or incorrect information, distorted perceptions, biases, and misplaced confidence about judgments that require careful, deliberate System 2 thinking.
When people learn about such follies, it is natural for them to believe that “it would have been different if I’d been there” or “it couldn’t happen to me.” I’d caution against such hubris.
Boards intent on great governance need to take action to minimize the probability of going off the rails because of organizational and psychological impediments to rationality in the ways they go about their work. It would be nice to think that forewarned is forearmed. Surely a director reading this chapter will come away with better awareness of bounded rationality and the problematic consequences of attribute substitution, anchoring, and the rest in board deliberations and decisions. Presumably knowledge is power, so boards and directors who are informed about these matters should do a better of job of self-monitoring and taking preventive action.
But Kahneman is cautionary. He reminds us that the way we think, especially our fast-thinking System 1, is hard wired and, as a result, difficult to change even with awareness. For example, while I’ve been more cautious and attentive while driving in novel situations since the accident I described earlier, I know that over time, vigilance fades and System 1 thinking is the default in a routine task like driving. Risk rises.
Kahneman has written recently on executive decisions.48 He asserts that awareness of cognitive processes that can produce judgment errors does not, in itself, constitute prevention of those errors. He and his coauthors suggest that executives (and presumably boards) use a decision quality-control checklist prior to making major decisions (e.g., pricing changes, capital outlays, and acquisitions). The checklist has a dozen items, including
• Is there any reason to suspect motivated errors or errors driven by the self-interest of the recommending team?
• Have the people making the recommendation fallen in love with it?
• Were there dissenting opinions within the recommending teams?
• Could the diagnosis of the situation be overly influenced by salient analogies?
Separately, four steps are recommended to adopt behavioral strategy to de-bias strategic decisions49:
• decide which decisions warrant the effort
• identify the biases most likely to affect critical decisions
• select practices and tools to counter the most relevant biases
• embed practices in formal processes
I understand the checklist and recommendations. But as an experienced director, I can’t imagine boards or executives using them mechanistically, and maybe that’s the point. In governance and senior management, part of being seasoned is having a deep, intuitive understanding of the impediments to rationality in organizational dynamics and individual decision-making. The best, seasoned executives and directors have the habit of questioning, challenging, and guiding others to overcome these impediments, especially when the stakes are high.
Perhaps it’s the reason that great leadership, often attributed to an individual, is so often the result of a great team: Sam Zell and Bob Lurie of Equity Group Investments, the organization of which the EQR business was originally a part; Paul and John Gordon of Gordon Food Service; Henry Schacht and Jim Henderson and, more recently, Tim Solso and Joe Loughrey of Cummins; Warren Buffett and Charlie Munger of Berkshire Hathaway; Larry Page and Sergey Brian of Google; and, the most unforgettable duo I ever met in the nonprofit sector, the late Father William Cunningham and his parishioner Eleanor Josaitas, co-founders of Focus: HOPE in Detroit. Such partners tend not to be identical but complementary. With a foundation of shared goals and values, but with somewhat different mental habits and biases, I suspect that the partners substantially increase each other’s rationality by comparing perceptions and, through discussion and debate, reducing bias.
I know Sam Zell, but regrettably, I never had the opportunity to work with Bob Lurie. People tell me that Sam and Bob were, in some respects, as different as night and day. But together, over thirty years, they made business decisions that created enormous value for their investors. I’ll never forget Sam describing his and Bob’s mutual trust this way: “We lived out of the same checkbook for thirty years!” I know what a loss it was to Sam, personally and professionally, when Bob tragically died of cancer at age forty-eight. How rare and wonderful is a partner who can improve one’s thought processes and the quality of decisions simply by working together day in and day out, year after year, decade after decade?
Obviously, there is no easy way to ensure that a board is doing things right, so I offer these practical suggestions:
• Recruit to the board people with a variety of experiences and viewpoints who are willing and able to bring them to bear on vital board decisions.
• Set a goal of making the boardroom a place of good information, rational analysis, quality discourse, careful consideration, and wise judgment. In the board and director evaluation process, assess whether that’s the case. Do what’s required in terms of tone, leadership, membership, and process to make and keep it that way.
• Combat arrogance and cultivate humility. Overconfidence leads to autopilot (System 1) thinking, which is especially risky in novel situations.
• Consider carefully which matters that come before the board deserve a conscious shift from System 1 (fast) to System 2 (slow, deliberate) thinking. They include big personnel, policy, and investment decisions. But things that are out of the ordinary, such as a requested exception to policy, or any pattern that seems unusual or troubling, should cause a board to slow down and apply System 2 thinking.
• Build plenty of white space in and around board meetings so directors can, in an unpressured way, compare perceptions and try out on each other the sense they are making of what they see, hear, and feel and concerns they have about it.
• Make sure that there is plenty of debate in the boardroom. It should be about consequential things. Debate is essential for the board to look at matters from multiple angles, generate new ideas, and come to quality, considered judgments.
Conclusion
Great governance requires directors not only to do the right things but also to do things right. This entails effective work processes by individual directors and the board as a whole. Behavioral research is shedding much light on impediments to rational decision-making and means of combating them. Effective stewardship by the board depends more on good process than good intentions.
A large body of thought has emerged in recent years about what is commonly called good governance. We turn next to practices that have real merit and can make boards more effective.