by Brandon L. Garrett, Nan Li, and Shivaram Rajgopal

A company facing charges will present a “Chicken Little routine” describing the dire consequences of a prosecution for the company, then-U.S. Attorney for the Southern District of New York Preet Bharara famously explained.  Yet typically, after settling the criminal case, “the sky does not fall.”  Instead, Bharara maintained, all too often “the sky brightens,” the firm is seen as having put its problems behind it, and “the CEO even gets a raise.” Other commentators have been skeptical that prosecutions of a company alter behavior of high-level officers such as CEOs. To be sure, sometimes the CEO appears to be affected by a possible prosecution of the company.  The CEO of Wells Fargo recently stepped down before any criminal prosecution was initiated, after civil enforcement and high-profile Congressional hearings brought public attention to bear on unlawful sales tactics the bank used. Perhaps the  culture of not taking responsibility at the top is changing.  Or perhaps cases like that of the Wells Fargo CEO are salient examples only because it is so rare that a CEO is made accountable, in some measure, for corporate crimes.

In a new draft titled “Do Heads Roll? An Empirical Analysis of CEO Turnover and Pay When the Corporation is Federally Prosecuted,” recently posted on SSRN, we ask: Does the criminal prosecution of a corporation affect the CEO? Or do criminal actions directed at the organization itself pose few consequences for the individuals at the top, and the CEO in particular? While CEO’s are rarely themselves prosecuted, organizations could discipline CEO’s through pay cuts, claw backs, or outright replacing the CEO in response to a criminal prosecution. We sought to examine whether and how that occurs.

Our analysis focuses on a dataset of public companies that settled criminal cases brought by federal prosecutors from 2001-2015. Our set included 109 public companies.  We compared those companies to the larger set of companies in the Execucomp database of S&P 1500 firms, focusing on CEO compensation and turnover during the same time period. The control sample consists of 2,801 firms leading to a total of 2,910 firms.  We examined the time period before and after prosecution, and the year that the company resolved the criminal charges against the company.

We found that in the year that the company settled its prosecution, through a guilty plea or a deferred or non-prosecution agreement, there was a significantly higher level of CEO turnover. Notably, the CEO turnover at the prosecution year is 23.4%, which is almost twice as high as the unconditional average.  Turnovers are also more frequent than normal in the two years prior to the prosecution, and less frequent in the two years after prosecution. This pattern should be interpreted with caution, because it is possible that both prosecution and CEO turnover are triggered by other factors unrelated to the question at hand, such as the 2008 financial crisis.   After controlling for firm characteristics and time trends, however, we found that prosecuted firms are associated with a 10.7% increase in the frequency of CEO turnover in the year of prosecution. This increase is also significant, as the unconditional turnover probability is around 12%.   Although statistically insignificant, prosecuted firms also have a higher rate of CEO turnover in the two years before prosecution.  However, in the two years following the prosecution year, CEOs are less likely to be replaced.   We also examined whether these were forced or voluntary CEO departures and concluded that the departures were largely forced.

There was little evidence of any CEO pay cut. The average total pay drops in the two year prior to prosecution (from $10,778,000 to $9,569,000), but rises in the prosecution year (to $10,198,000). Second, for the prosecuted firms that did not have CEO turnover after prosecution, there is little evidence of a reduction in compensation. Indeed, we observed a spike in CEO bonuses in the year of prosecution—confirming concerns expressed by judges, prosecutors, lawmakers, and academics that corporate prosecutions do not sufficiently impact high-level decision-makers like CEOs. For the prosecuted firms that did have CEO turnover after prosecution, there is some evidence of a pay cut, both to salary and bonus, prior to the replacement of the CEO. We also observe that relative to the CEOs of firms that were prosecuted, where both the firm was prosecuted and the CEO was individually charged, the CEO received  a significant cut in compensation.

These data describe for the first time a picture of how corporate prosecutions affect CEOs, and  adds to the literature on corporate criminal liability in several ways.  There is some literature on how corporate prosecutions affect the reputation and share price of a company, but very little literature on how CEOs or other high-level officers are themselves affected by crimes. Corporate criminal prosecutions are not frequent and they reflect, one would hope, the most serious possible violations by a company.  And yet we know very little about whether companies hold top-level officials accountable in response.  We do know that those top-level officials are rarely themselves prosecuted.  These results raise larger questions about whether federal prosecutors who aim to target the most serious corporate crimes are sufficiently incentivizing  accountability at the top.  We hope that this examination of the relative non-impact, of criminal prosecutions on CEOs will help to stimulate further inquiry into the utility of corporate criminal prosecutions and into the role of CEO’s in modern public corporations.

Brandon L. Garrett is the Justice Thurgood Marshall Distinguished Professor of Law at University of Virginia School of Law.  Garrett’s recent book examining corporate prosecutions, titled “Too Big to Jail: How Prosecutors Compromise with Corporations,” was published by Harvard University Press in Fall 2014.  Shivaram Rajgopal is the Roy Bernard Kester and T.W. Byrnes Professor of Accounting and Auditing at Columbia Business School.  Nan Li is a PhD student at Columbia Business School.