I recently read a really interesting paper by Brett McDonnell, Claire Hill, and Aaron Stentz. It struck a particular cord because I had just watched a rerun of the Law & Order episode House Counsel, in which the great Ron Liebman plays a mob lawyer who crosses the line from zealous representation to participation. I queried Brett about whether their paper spoke to the episode and, in light of that discussion, I invited Brett and Claire to do a guest post about their article. It follows:
Professor Bainbridge has very kindly asked us to post about our new article, Bad Agent, Good Citizen? Law and economics have related but distinct concepts of “agent.” In economics, when someone acts as an agent, she must act for the benefit of her principal, not for her own or another’s benefit. In law, this is also so – except that there is an override when an agent is acting for the benefit of her principal, but is helping the principal break the law. For instance, a law-breaking board breaches its fiduciary duties to the corporation even where the law-breaking benefited the corporation.
But what about cases where action may help the principal but hurt society without breaking the law? Our paper considers four possible cases: good agent/good citizen, good agent/bad citizen, bad agent/good citizen and bad agent/bad citizen. The first and the last are not that interesting analytically.
Our interest lies in the other two categories. The good agent/bad citizen lawyer might help Greece design financial instruments that would conceal the extent of its debt. The bad agent/good citizen might exaggerate possible perils to the client of environmental non-compliance in order to get the client to comply more maximally as per the lawyer’s own preferences.
In the less interesting cases, we know where we stand, both from a legal point of view, and from an economics point of view. The mixed cases are murkier, with law introducing a tension, and another party in interest, into the agency relationship—something we conclude we favor, and suggest ways to promote. That is, legal ethics should leave some (though limited) space for lawyers to act as better citizens at the cost of being less good agents.
While the context for our paper was a symposium on corporate law, Professor Bainbridge asked us how it applies to criminal law, drawing our attention to an episode of Law and Order, House Counsel. The lawyer aggressively uses procedural devices to shield his crime boss client from being held criminally liable. Eventually, the DA is able to counter those devices, and indeed, obtain a conviction of the lawyer himself, who has crossed the line and participated in heinous crimes in furtherance of his client’s crimes. At that point, the lawyer was clearly a bad citizen, but also a bad agent in the legal sense, since legal ethics forbids helping clients break the law. (The lawyer was, though, a good agent in the economic sense.)
But what of a typical representation, where a lawyer defends a crime boss against allegations based on the client’s (past) behavior? We will assume, to make it a harder case than the one in House Counsel, that a lawyer suspects the client may be guilty but does not have direct knowledge of the client’s guilt. What makes such a lawyer not a bad citizen when the law is working as intended is that he is opposed by an equally aggressive advocate on the other side. Of course, in the real world, the law often does not work as intended. Prosecutors will often do better than law intends them to: Poor defendants will not have advocates whose aggressive advocacy equals that of the prosecutor, and hence are at a disadvantage. And they may do worse than law intends them to, insofar as (well-resourced) defendants may be able to take advantage of certain structural features designed to ‘let a thousand guilty men go free to avoid convicting an innocent one.’ Thus the problem remains: criminal defense lawyers should zealously advocate for their clients, consistent with law's imposition of various ‘good citizenship’ requirements on lawyers, who, after all, are ‘officers of the court.’ So, our bottom line is perhaps to duck the question of whether aggressive use of procedural devices to get someone who is probably guilty off the hook makes the ‘good agent’ a bad citizen.
