Enron had quite an impressive board of directors. The 17-member group was almost certainly too big, but beyond that it met or exceeded most modern criteria for good corporate governance: company insiders were a tiny minority; the jobs of chairman and CEO were separated; the key role of audit committee chairman was held by a respected outsider who presumably understood accounting (at least, he taught it at Stanford). What could possibly go wrong?
There's little or no evidence that the modern criteria for good corporate governance actually lead to better-governed corporations. What's generally seen as the most important good-governance move of them all, pushing insiders off boards in favor of independent directors, may actually hurt performance. At least, that's my reading of the voluminous academic research on the topic.
Yet the anti-insider movement rolls on unchecked. The Wall Street Journal reported this week that chief financial officers are getting chucked off their companies' boards right and left as part of the push for more independent directors. And now my friend Felix Salmon has let loose with a video rant about the Goldman Sachs board that lambastes it for being populated by too many insiders and long-timers.
Why does this idea that outsiders make better board members persist despite the mostly contradictory evidence? I think it's because of a persistent misconception about what the role of a board of directors is. The common view is that, as Felix puts it, "board members are meant to keep management under control."
Boards — I'm quoting Felix again here but the view is of course quite widely held — "are meant to be representing the shareholders and stopping the managers from just running away with the whole business."
In corporate law, though, this isn't the role of the board at all. As Cornell Law School professor Lynn Stout explains here, the board is actually responsible to the corporation, not its shareholders. And no, the shareholders don't own the corporation — they own securities that give them a not very well-defined stake in its earnings, and the freedom to flee with no responsibility for the corporation's liabilities if things go pear-shaped.
So where did this idea come from that the board's job should be watchdogging management on behalf of shareholders? Well, mainly from Michael C. Jensen and William H. Meckling's famous 1976 article "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure," although the argument was floating around the University of Chicago (where Jensen and Meckling were educated) for a while before then. Jensen and Meckling proposed a model of the firm in which the managers were agents and shareholders the principals on whose behalf the agents were supposed to be acting. The main challenge of corporate governance was getting those agents to obey their principals.
The strengths of this model are that (1) it's simple and elegant and (2) in one special circumstance, it seems to accord pretty closely with reality. That is, corporations where it's clear who the principal is and what he or she wants — that is, they have a non-management shareholder who owns more than 5% of the common shares — appear to be better run than the rest of the corporate universe. But as Jay Lorsch and I wrote in our HBR article, "What Good Are Shareholders" this summer and Stout writes in her new book, The Shareholder Value Myth, when you try to impose that principal-agent model outside of such special circumstances, it doesn't work very well. Shareholders are a fickle, skittish, otherwise-preoccupied lot. And the more that executives are treated as miscreants in need of oversight, it seems, the more they act like miscreants in need of oversight.
These limitations of the principal-agent model of corporate governance are frequently discussed in academic papers and in HBR. But apart from a few opinion columns about the likes of Stout's book and my and Lorsch's article, these arguments don't seem to have penetrated mainstream financial discourse at all.
This is especially curious in the case of financial firms like Goldman Sachs, where shareholders contribute only a small portion of the balance sheet and lenders (and taxpayers) are in many ways truer owners. Multiple studies have shown that it was financial firms with the most shareholder-friendly governance and executive compensation schemes that got into the most trouble during the financial crisis. That only makes sense — shareholders pocket the gains if big risk-taking pays off, but they aren't on the hook when a bank collapses. Goldman's relatively smooth sail through the crisis was in part the product of a governance culture that doesn't put the short-term interests of shareholders first and conceives of the firm as an institution geared toward long-term survival. And yeah, having lots of friends in government didn't hurt either, but Goldman's tradition of sending retired executives to Washington is an important part of lits culture, too. Those insiders and long-termers on the Goldman board are the representatives of that culture. And that's supposed to be a bad thing?
The difficulty here, as with any attempt to supplant the simple-if-often-wrong shareholders-first view of corporate governance with something that actually works is that there's no straightforward rallying cry. Throwing insiders off of boards is simple. Putting together a truly effective board — one full of committed, expert members who generally have a constructive, supportive relationship with management but are curious enough to keep digging into the company's business and tough enough to take a stand when management begins to lose the plot — is hard. HP's board has been full of impressive outsiders for years. It has also been a disaster. Citigroup's board, which ousted CEO Vikram Pandit on Tuesday and is loaded with outsiders and newcomers (the longest-serving board member joined in 2008), may have set the course for a more prosperous new future for the company. Then again, it may have embarked on an HP-like drama of tension and dysfunction.
As for Goldman, the sad saga of Rajat Gupta has obviously tainted the board. Felix is right that lead outside director Jim Schiro probably has too much else on his plate to be effective. And the board seems stacked with people whose knowledge and experience tilt toward Goldman's investment banking side, even though trading is now where the firm makes the vast majority of its money. On the other hand, maybe that's a good thing: maintaining a balance of power between trading and banking was long a key to Goldman's success. Corporate governance is all about balancing diverse, sometimes conflicting priorities and interests. Which is why seemingly simple, straightforward attempts to improve it so often backfire.
Update: Felix Salmon responds.
