Yesterday at the Illinois Corporate Colloquium Steve Choi presented his paper (with Pritchard and Weichman), Scandal Enforcement at the SEC: Salience and the Arc of the Option Backdating Investigations. Here’s the abstract:
We study the impact of scandal-driven media scrutiny on the SEC’s allocation of enforcement resources. We focus on the SEC’s investigations of option backdating in the wake of numerous media articles on the practice of backdating. We find that as the level of media scrutiny of option backdating increased, the SEC shifted its mix of investigations significantly toward backdating investigations and away from investigations involving other accounting issues. We test the hypothesis that SEC pursued more marginal investigations into backdating as the media frenzy surrounding the practice persisted at the expense of pursuing more egregious accounting issues that did not involve backdating. Our event study of stock market reactions to the initial disclosure of backdating investigations shows that those reactions declined over our sample period. We also find that later backdating investigations are less likely to target individuals and less likely to accompanied by a parallel criminal investigation. Looking at the consequences of the SEC’s backdating investigations, later investigations were more likely to be terminated or produce no monetary penalties. We find that the magnitude of the option backdating accounting errors diminished over time relative to other accounting errors that attracted SEC investigations.
As readers of this blog, and Ideoblog before it, will appreciate, this paper particularly resonated with me. As I wrote in a large number of posts (e.g.) backdating was a molehill the media blew up into a mountain. Now come Choi et al with evidence that while the SEC was spending its scarce resources on this overblown molehill it was ignoring real mountains (e.g., Madoff).
I found the paper overall quite persuasive. I wasn’t entirely convinced by the evidence that the backdating cases were getting weaker. In particular, stock price reactions may just indicate the market was learning about the which companies were involved before the investigations were brought, and was gradually figuring out that backdating was not such a big deal. But I was convinced of the evidence of the opportunity costs of the SEC’s backdating obsession — the otherwise inexplicable decline in investigations of serious non-backdating accounting problems.
As we discussed in the Colloquium, the paper reveals that there are agency costs not just in the backdating companies that were investigated but also in the agency that was doing the investigating. Although it’s not clear exactly what moved the SEC to follow the media, there is at least some doubt about whether the SEC’s resource allocation decisions were in the public interest.
This calls attention to another set of agents — the ones in the media. Why did the media love backdating so much? As discussed in my Public Face of Scholarship, there are “demand” and “supply” explanations: the public demands stories about cheating executives and/or journalists like to supply these stories. David Baron, Persistent Media Bias, presents a supply theory emphasizing journalists’ anti-market bias.
Whatever the cause of media bias, when the media is influential its bias can result in bad public policy. SEC enforcement isn’t the only example. As I discuss in my article (at 1210-11, footnotes omitted):
Where interest groups are closely divided, the outcome of political battles may depend on how much voter support each side can enlist. This may depend on how journalists have portrayed the issue to the public. For example, the press is an important influence on corporate governance. One factor in the rapid passage of the Sarbanes-Oxley Act, the strongest federal financial regulation in seventy years, may have been the overwhelmingly negative coverage of business in the first half of 2002: seventy-seven percent of the 613 major network evening news stories on business concerned corporate scandals.
It’s not clear what can be done to better align SEC enforcement policy with the public interest. Incentive compensation for SEC investigators? Perhaps the only thing we can do (as with corporate crime) is to try to keep in mind when creating regulation that even if corporate agents may sometimes do the wrong thing, people don’t stop being people when they go to work for the government.