At a recent conference, SEC Commissioner Mark Uyeda offered remarks on the “effectiveness of the broken windows policy in the securities enforcement context.”
As Uyeda explained, “the broken windows hypothesis, as described in a seminal article published in The Atlantic by James Q. Wilson and George Kelling, posits that visible signs of disorder, when left unaddressed, create an environment that encourages more serious crime.”
Uyeda then commented on a paper titled “Broken Windows Securities Enforcement” and stated:
“The empirical literature on this theory remains contested, but its migration into securities is not merely academic. The paper focuses on a period in the past when the Commission explicitly implemented a broken windows policy, about which concerns have been expressed. Whether the broken windows theory translates effectively in the securities enforcement context is an interesting question, particularly as this theory was implemented by prior Commission leadership.
This particular paper asserts that it “provides the first empirical evidence consistent with a broken windows securities enforcement policy reducing the incidence of severe financial misconduct.” The paper does, however, recognize that such a policy may not be effective overall in a white-collar setting because enforcing minor violations could divert limited resources and attention away from pursuing the more serious violations.
I appreciate the efforts to use various metrics in empirical studies in this field. My reactions to the paper are based on nearly twenty years of service at the SEC, half of which has been spent on the executive staff. During that period, I have reviewed thousands of recommendations from the Division of Enforcement.
When it comes to “broken windows,” Wilson and Kelling identify two separate, but related, fears. The first is the fear of being a victim of crime, especially crime involving a sudden, violent attack by a stranger. The second is the fear of being bothered by disorderly people. As Wilson and Kelling describe this group, they are “not violent people, nor, necessarily, criminals, but disreputable or obstreperous or unpredictable people.”
Wilson and Kelling further describe a neighborhood as being made up of “regulars” and “strangers.” Rules regulating order were defined and enforced in collaboration with the “regulars” on the street – and not necessarily by the rules established exclusively by law.
In other words, to achieve order, there was a set of regularly expected behaviors among participants in a neighborhood. In a certain sense, that regular expectation of rules and norms exists in the securities markets as well. Leaking material non-public information to a golfing buddy with the expectation that your friend can trade profitably on such information – definitely prohibited behavior. But requiring monitoring and retention of all texts on personal mobile devices by persons working for broker-dealers and investment advisers? Is that a violation of the technical meaning of the laws and regulations? Perhaps. However, given the reaction to the SEC’s sweep effort on off-channel communications, communicating on personal devices was not the type of behavior that was viewed as unambiguously impermissible.
When it comes to regulatory enforcement, one should also be considering the incentives and metrics designated by agency leadership for that program. If the metrics being stressed focus on the number of enforcement actions being brought and the amount of dollars collected in the form of disgorgement and civil penalties, then agency personnel will act in accordance with such instructions and incentives.
The SEC’s rulebook has expanded dramatically in volume and complexity over the decades, which leaves significant ground for the Commission staff to exercise discretion in making and recommending enforcement decisions. Without limitations on the exercise of such discretion, there are many violations and novel legal interpretations that can be asserted as the basis for sanctions against virtually any asset class or market participant.
However, random acts of regulatory enforcement dressed up as broken windows enforcement are unlikely to succeed at achieving behavioral conformity with widely-accepted norms intended to protect investors, maintain fair, orderly, and efficient markets, and facilitate capital formation. Abusing our discretionary authority undermines the regulatory predictability that efficient markets require or chills market activity that may be socially valuable.
The author finds that sweep investigations opened during the broken windows era were 8-9% more likely to result in an enforcement action, which can be partially explained by strict liability violations that do not require proving negligence or scienter. The author also divides SEC enforcement actions based on the tenure of various SEC chairman. However, the paper does not appear to make adjustments to reflect that SEC enforcement investigations, particularly for complex accounting cases, can take years between initiation and the bringing of any actual action. Thus, the enforcement actions taken during the early tenure of any chairman often reflects investigations initiated under the prior chairman – and enforcement initiatives undertaken by one chairman may only come to fruition, if at all, during the tenure of his or her successor.
Nonetheless, I appreciate the efforts of researchers to study the actions of regulatory agencies. Their efforts play a role in holding the government accountable to the people.”
