Beazer Homes and the skewed signals of deferred prosecution agreements

Floyd Norris has an article in the New York Times today criticizing the Deferred Prosecution Agreement that the Department of Justice recently announced between Beazer Homes and the United States Attorney for the Western District of North Carolina.

According to the Information filed by the government, Beazer Homes engaged in a number of accounting frauds, which of course hurt home buyers and shareholders alike. Norris states that he has "no reason to believe" that Beazer's CEO "understood the crimes the company was committing" and then suggests toward the end of the article that the Deferred Prosecution Agreement sends the wrong message to corporate  boards and officers insofar as the settlement of the case leaves intact the positions held by Beazer's CEO and Board of Directors, all of whom apparently failed to detect and prevent the fraud.

The article demonstrates the problems that arise when we prosecute corporate entities.  For a number of reasons that I outline in this paper, we don't really want to indict and convict corporate entities because the collateral costs of indictment are too high and affects too many innocent parties.  So, we offer the companies Deferred Prosecution Agreements, which require the companies to make restitution and remedy the internal problems in the company that led to the problem in the first place.

Deferred Prosecution Agreements (DPA's) send fuzzy, mixed signals.  The mere presence of a settlement agreement (and often the press releases accompanying the agreement) suggests that the company's management has done something "good" while the presence of a criminal charge and investigation suggests that the company has done something deeply harmful and morally repugnant.  Even worse, since the agreement is often reached prior to an indictment, the public is privy to only the broad outlines of what happened and how various factors drove the government's decision-making.  As a result, it is not surprising that DPA's leave a bad taste in everyone's mouth.

You can also see how corporate criminal liability inevitably morphs into an often simplistic and not very helpful discussion about corporate governance.  Should Beazer Homes' CEO be fired and its directors voted off the board? Maybe. Were I a shareholder, I probably would say it depends.  On one hand, I would be pretty steamed that management and the board failed to catch the fraud.  I would probably ask what the heck happened to the company's internal controls and internal compliance program (not to mention the company's auditor and Audit Committee).  I would scream and holler about the CEO's executive compensation.  And maybe, like Larry Ribstein did two years ago, I might start to wonder if SOX-mandated compliance systems had failed to live up to their billing. 

But I probably would be forward-thinking too - analyzing whether I thought the CEO and Board members were best positioned to help the company emerge from this mess and considering whether I thought management had built a decent relationship with the government in the meantime.  I would want to hear what the CEO had to say about how he planned to change the company and how he thought the company could survive in the current market.

In sum, I would be weighing options and considering carefully what was best for the company and my investment.

But that above answer is not very satisfying if you are coming from the "throw the bums out and punish them" camp.  And if you approach a corporate governance problem with a criminal law frame, emotions like anger quickly crowd out more cautious, introspective responses.  That's the problem with corporate criminal liability and Deferred Prosecution Agreements.  The mere presence of a criminal case suggests that People Did Very Bad Things and that a good dose of punishment is in order.  But it's hard to punish companies without punishing a lot of people who had nothing to do with the Very Bad Things, and tossing CEO's out of their jobs for failing to prevent Prior Bad Things does not at all guarantee that their successors will prevent Future Bad Things.  For a prime example of that, see AIG.