Sidney Weinberg and his son John Weinberg both served as longstanding chairmen of Goldman Sachs. Recently, John’s 1948 Princeton undergraduate thesis came to light. Like a 1933 memorandum and a 1949 speech by Sidney, it addressed a fundamental issue: What should be expected of boards of directors? All three documents argue that directors should advance corporate governance by asking discerning questions. In a new book chapter, we look at their argument.
The chapter deals with questions by now familiar to professors, judges, and practitioners. But the governance of large firms, generally incorporated in Delaware and publicly traded, illuminates broader fields. The development of corporate governance over the last century teaches something about the limitations of law, and the ultimate dependence of law upon political institutions, in turn dependent upon some sense of community. For reasons we discuss, there is no way to use legal rules and processes to convene a group of people and ensure that they reach substantively good decisions. Rules and processes are of course important, and perpetually insufficient, but realizing the insufficiency of law has profound consequences for how we think about not just corporation law, but constitutions, treaties, and the like.
The emergence of doctrine. Reading the works by Sidney and John Weinberg makes one think of the old joke about Shakespeare: terrible writing, full of clichés. The Weinbergs recommend regular meetings of directors, provided with quantitative written information in advance. They emphasize independent directors, especially with regard to executive compensation. They champion the diversity of boards – the list goes on. Aspects of corporate governance that a contemporary business associations teacher might consider obviously right and good were not ordinary for much of the twentieth century.
Writing in the 1930s and ‘40s, the Weinbergs responded to and furthered a then-recent tectonic shift in how corporate governance was conceived. Since the inception of the business corporation, directors were expected to direct the business of the firm. In many cases, they were also investors and employees, and in almost all cases friendly with senior management. That is, the interests of the board of directors and the business, in effect meaning its management, were assumed to be closely aligned. Implicitly, a business was assumed to be a tight-knit affair.
In the years after the Civil War, however, both the U.S. economy and individual businesses grew and grew. Firms like General Electric embodied a new business model: centralized management pooling the resources of vast numbers of widely dispersed shareholders, the nominal owners of the firm. This model continues to dominate the public imagination of “the corporation.” The Crash of 1929 and the ensuing Great Depression demonstrated that stockholder confidence in such corporations can evaporate quickly. Massive selloffs resulted in plummeting prices. Stockholders across the economy, many of whom had no other connection to the firm in which they were invested, were badly hurt.
The law responded in three ways that bear mention. First, the collapse of Swedish Match gave rise to shareholder lawsuits claiming that directors were negligent, which occasioned Sidney’s 1933 memo. Under what circumstances might directors be liable? Under what circumstances would directors be able to argue that, notwithstanding unfortunate events, their actions were not negligent? More generally, what do we expect of directors, particularly independent directors? How should boards be structured so that directors might fulfill society’s expectations?
Second, the first New Deal Congress sought to protect shareholders through passage of the Securities Act of 1933 and the Securities Exchange Act of 1934. The new measures imposed obligations directly upon directors: signing the registration statement, liability for misstatements or omissions of material facts, reporting purchases and sales of the company’s securities and short-swing profits.
Third, in 1933, Adolph Berle and Gardner Means published The Modern Corporation and Private Property. The corporation, Berle and Means argued, was distinguished from other forms of business organization by the separation of ownership from control. The beneficial owners, the shareholders, had no effective control over the business. On the other hand, a corporation’s managers had not only legal control over the business but economic reasons to divert the company’s profits to themselves. Berle and Means urged that the corporation be understood along the lines of the trust, with managers bearing fiduciary obligations to operate the business in the interest of the “beneficiaries,” the shareholders.
These efforts to address the vulnerability of shareholders changed the position of the board of directors vis-à-vis management. Where the board had been expected to be supportive, perhaps occasionally offering advice, directors were now expected to take responsibility for the direction of the firm, and even to remove managers if necessary. Failure to do so meant that directors could be personally liable for the actions of managers, depending on the circumstances. Since directors qua directors do not actually run businesses, the threat of liability in effect required directors to know what managers were doing, and approve of it, or not. Directors were expected to ask what John Weinberg called the “discerning question.”[1]
The 1930s efforts to protect shareholders tended to separate: the board from management; “running the business” from “day to day operations;” and the directorial function from the managerial function. Then as now, it would be too much to call the relationship between the board of directors and managers adversarial, except perhaps in extraordinary circumstances like takeovers. But the board of directors was and still is expected to ensure that the business is run in the interest of the corporation and its shareholders.
The persistent failure of governance. As a practical matter, how are directors to hold managers to account, and on that basis, formulate judgements about where to direct the firm? Sidney and John proposed a number of aids to institutional navigation: regular meetings, financial accounts, audit committees, and so forth. Sidney even set out “commandments” for boards. It is difficult to disagree with any of these. So informed, boards might be able to ask the discerning question, to steer the firm around the icebergs and into harbor.
Yet Sidney himself sometimes failed to ask the discerning question. For almost 30 years, he was a director of McKesson & Robbins. For some 13 years, the company was looted by Frank D. Coster, the alias (!) of Philip Musica. Sidney simply did not see it. There is no evidence that Sidney’s failure stemmed from breaking one of the commandments that he himself had laid down for directors. The commandments just didn’t work, not even for their author.
More generally, it is worth thinking about why, despite all our teaching, corporate governance so often fails, sometimes spectacularly. For example, Enron and the other “accounting scandals” revealed fundamental flaws in corporate accounting. Congress responded rather forcefully, with the Sarbanes Oxley Act. With presumably better-informed directors, a host of financial institutions levered up their exposure to opaque investments, which ended badly just a few years later in the Global Financial Crisis. Our chapter argues that although corporations are often well governed, the legal and intellectual problem of corporate governance is insoluble. Rocks are always in danger of rolling back down their hills. Today’s good decisions are commendable, but no guarantee that tomorrow’s decisions will also be sound. The law may guide, may even teach, but cannot solve. The problem of government is perpetual.
Size, complexity, risk, and uncertainty. In concluding his thesis, John casts doubt on the entire enterprise: Suppose a well-constructed board, with well educated, well informed, and even well paid but otherwise independent directors, from a variety of backgrounds, simply isn’t enough? “As a result of the increased size and complexity of corporations, we see a general inconsistency arising: The board of directors is practically unable to carry out the tasks it is theoretically supposed to.”[2] “Size and complexity” are clearly issues, but hardly most fundamental.
A, maybe the, core purpose of the corporation is to pool capital and allow managers/entrepreneurs to take risks with that pool. And risk, by definition, entails the possibility of adverse outcomes. More generally, the function of limited liability is to free market actors from responsibility for the unknown consequences of their activities. “Risk” is a bit of a misnomer: Risks can be priced because the parameters of events are known, and probabilities can be reckoned. In contrast, at issue in the management of a business moving into the future is uncertainty: What will happen? Will there be a financial crisis, a pandemic, the development of new technologies, or the overnight imposition of sweeping tariffs? Something else unthought? And how would we begin to think about specific rules that would tell directors how to cope with the unthought as it emerges, while trying to run a profitable business?
General understandings and particular judgments. When we try to address a situation, we lawyers generate processes. Risibly enough, we usually insist that we are focused on substance, not form. The problem, of course, is there is no way to insist on “real substance.” A legal rule, by its nature, cannot talk about a particular board member, in a specific context, or a business in a time or place, with all its competing interests and outright conflicts. Legal rules are not valid “for this train and this day only.” While a legal text can be substantive, it is the substance of ideas that is at issue. This is obviously true for rules and regulations, but it is also true for the common law. The moment one attempts to say, for example, what “business judgment” means – in answer to a shareholder who has lost money, perhaps – one will have to do so vis-à-vis an understanding necessarily distinct from the particulars at hand.
Kant strove to articulate the logical limits of such conceptual thought. In life, however, we don’t encounter pure concepts. Nor, for that matter, do we encounter the law as such. In life, we encounter facts, events, and we try to think through what such phenomena mean, legally and otherwise. We try to use generalities to organize particulars. But what allows us to understand this particular phenomenon as an instantiation of this idea, rather than some other?
Kant argued that we know how to think about things because we feel in some sense at home, because it is common sense (sensus communis) to look at X in this or that way. Common, that is, in this community. Judgment thus turns out to be intimately wrapped up in politics. Particular judgments are not logically necessary because they are not conceptual (where logic operates), but humans still reach judgments all the time, as they must, and the substance of judgments is fundamentally informed not by logic but by experience in society.
Social Context. For our purposes, the problem of corporate governance is ultimately grounded in social context, not demonstrable through logical citation of doctrine or even implementation of best practices. John implicitly understands asking the discerning question as the director’s role at board meetings: “The manner of inquiry is very important. The question should be asked pleasantly, discreetly but firmly. It is definitely not a cross-examination, because the director is a participant seeking to guide and evaluate results.”[3]
Governance cannot be complete. Tomorrow will require attention to its own decisions. At tomorrow’s board meeting, one hopes that discerning questions will be asked and wise consensus will be reached, as it often is. But not always: While social groups “know” certain things, they may well be wrong. For an obvious example, going into the Global Financial Crisis, many institutions had a great deal of trust in their own financial engineering and governed themselves accordingly. The “wisdom of crowds” came to be revealed as the “wisdom” of a herd. More generally, the law can improve governance, but nothing done today will make tomorrow’s mistakes impossible. Whatever we teach, sometimes boards will fail, occasionally spectacularly, but more often in small ways. Sometimes, the rock will roll down the hill. When that happens, Sisyphus is fated to roll it back up.
ENDNOTES
[1] Weinberg Thesis, Section III.A.2.
[2] Weinberg Thesis, Section VII. Emphasis in original.
[3] Weinberg Thesis, Section III.B.3.
This post comes to us from professors Amy Deen Westbrook at Washburn University School of Law and David A. Westbrook at the University at Buffalo Law School. It is based on their recent chapter, “Sisyphus the Director: Why is the ‘Discerning Question’ Rarely Asked and Never Finally Answered?” in Boardroom Legacy: The Weinbergs of Goldman Sachs and the Evolution of Corporate Governance (Lawrence A. Cunningham ed.) forthcoming in 2026 and available here.