Pick your poison: class actions or corporate criminal prosecutions?

Stanford's Joe Grundfest writes in today's WSJ that securities class actions are declining. One reason? He suggests maybe it's because there's less fraud.  And that might be because

there is a new, tougher and superior enforcement mechanism in place. The SEC and the Department of Justice now insist that any corporation suspected of a sufficiently serious fraud conduct an internal investigation that will finger the executives responsible. The corporation must also cooperate in prosecuting these executives. This enforcement technique is stunningly effective, if often overbearing. It eliminates the government's need to conduct expensive and lengthy investigations and provides the authorities with extraordinary leverage over every executive suspected of wrongdoing. Private litigation doesn't have an equivalent deterrent effect because it can't threaten executives with jail and because damages are almost always paid by corporations and insurers, not the executives who cause the fraud. As long as the government's enforcement activities remain sufficiently vigorous, the private class-action securities fraud lawsuit can be viewed as an expensive, wasteful and unnecessary sideshow that generates little deterrence and offers questionable levels of compensation. The question then is not why these lawsuits have been shrinking so rapidly in recent months, but when and whether they should exist at all.

Grundfest's explanation of the decline in private securities litigation may be correct, but his suggestion that this is a good thing is deeply troubling. As I've often suggested (see my Corporate Crime archive) there are serious problems with criminalizing the sorts of corporate misconduct that are the subject of private action.

Does the dip in securities litigation suggest that the corporate criminal prosecutions have been worth these costs?  I'd demand more evidence that we don't just have an up-market litigation cycle. Moreover, even if we do have less fraud to litigate, I'd wonder whether it's been worth the price.  Do we have less risk-taking?  A zero fraud world is not necessarily paradise.

The question here is not who one's favorite bad guy is (federal prosecutors or class action lawyers) but what system achieves the maximum deterrence at the minimum costs. (Compensation really doesn't enter into any of these approaches).

As I said in my Dabit, Preemption and Choice of Law, "the efficiency of private litigation can be assessed only in the context of all of the tools for fighting fraud." But I also noted that there's a lot we don't know about the optimal mix of approaches to deal with fraud, and that a one-size-fits all federal regime, with its lurches from one "solution" to another, is not necessarily the way to go.

Update:  In the post above I refered to "maximum deterrence at the minimum costs."  That was careless phrasing.  I meant "most cost-effective deterrence." 

While I'm at it I should note that, generally consistent with Grundfest, Karpoff, Lee and Martin, in their recent Legal Penalties for Financial Misrepresentation, find that "a recent increase in regulatory penalties has coincided with a decrease in private penalties."  But, for the reasons discussed in my post, this doesn't either establish a causal relationship between the two phenomena or justify the current mix of sanctions.