Finance professor Giorgia Piacentino’s recently published article “Deadlock on the Board” explores the problems that arise for firms when boards disagree on how to proceed with company policy. According to Piacentino, there is significant nuance to the issue of board deadlock, as sometimes board members are indecisive even when there is agreement on a course of action. In this Q & A, Piacentino explains that when boards hit a deadlock, it can ultimately allow CEOs to become entrenched, and offers a new view on the role of shareholders in corporate governance.

What aspect of deadlock does your research focus on?

When people disagree about what policy is best—whether to launch a stimulus package, whether to raise the debt ceiling, whether to remove Civil War monuments—political deadlock can ensue.

Deadlock also arises in corporations, when boards of directors are tasked with doing the best thing for the firm. Naturally, if directors disagree about what the right thing to do is, then the board could struggle to do the right thing, or even define what that is.

But, strikingly, sometimes boards struggle to make decisions—they are deadlocked—even when directors agree on a course of action. For example, we show that a deadlocked board doesn’t put in place a new CEO, even if all directors agree that having a new one is better than retaining the incumbent.

Uber was a salient example of this mechanism: in the summer of 2017 the board was so deadlocked that it stayed without a CEO for months, even though it seemed like everyone agreed that any alternative would have been better than no CEO at all! According to the New York Times, it was deadlock on the board that led one frontrunner for the job, Meg Whitman, to withdraw her name from consideration, saying, “It was becoming clear that the board was still too fractured to make progress on the issues that were important to me.”

I wanted to understand how deadlock could arise without disagreement about what to do.

To do so, my coauthors, Jason Donaldson and Nadya Malenko, and I developed the first model of a group of heterogeneous directors making decisions dynamically. We found that even directors who don’t disagree today can be deadlocked. The reason is that they anticipate that they could disagree in the future, and today’s decisions can influence future ones. For example, keeping a bad CEO in place today could make it easier to put your preferred candidate in place if she comes along in the future.

What are some examples of CEO entrenchment? Why would directors keep a bad CEO in place?

I have some ideas about which CEOs are entrenched, but none I want to advertise lest they take revenge and I end up in a trench myself.

More seriously, there are different kinds of entrenchment. Often, we are concerned about entrenchment due to cronyism or corruption, some of which is discovered after the fact. But one of our main points is that entrenchment can be the result of something more innocent but not necessarily more benign: disagreement among the directors. It is trickier to identify examples of this kind of entrenchment in practice, because you can never be sure whether the CEO of a struggling firm is kept in place because they are entrenched or because they actually have a lot of potential, and the firm’s underperformance was just unlucky.

But with models and data we can try to gauge whether entrenchment is a serious problem in aggregate, even absent definite examples. Indeed, only 2 percent of Fortune 500 CEOs are fired on average every year, a number other research suggests would be an order of magnitude higher if no CEO were entrenched.

What are the differences in action between an “aligned” board and a “diverse” board? Is there a preference in the context of providing a firm the proper guidance?

There are a lot of ways to try to measure diversity. Maybe people think about diverse boards as being made up of people with different sexes, ethnic backgrounds, or institutional affiliations. But I think that these attributes are really just measurement proxies for what really matters: diversity of opinions and knowledge.

What is interesting is that we find that even an ideal form of diversity, which leads to a more informed or unbiased board, can be detrimental for shareholders. The reason is that sometimes it is better to agree on the wrong thing than not to agree at all. And an aligned board does exactly that, it agrees, sometimes on the wrong thing, but it is not deadlocked. Thus, there is a trade-off between the deadlock on a diverse board that can’t decide what to do and the tyranny of an aligned board that decides purely in its own interest, which is not necessarily the shareholders’.

What does your study find about the role of the shareholder in board decisions, or in the composition of boards?

The classic theory of corporate governance is that shareholders should have the power to appoint directors, rather than directors actually doing the appointing for shareholders.

Although there are good reasons to maintain this view, we think our work could cast a shade of doubt on it. The reason is that deadlock can arise even when current directors do not disagree with each other, but just worry they will disagree with directors whom shareholders appoint in the future. Limiting shareholders’ power to do this can alleviate their worry and thereby curb deadlock.

What are the policy ramifications of your research findings?

One theme in the corporate governance literature is that short-termism is bad. Our analysis suggests that there is a cost of long-termism because, as I touched on, the anticipation of disagreement in the future can generate deadlock today.

Thus, our analysis provides a new rationale for policies that limit director terms, which have been discussed in several countries.