Document Type


Publication Date



Center for Contract and Economic Organization


Program in the Law and Economics of Capital Markets


To understand what is and is not wrong with plea bargaining, one must understand the relationship of bargains to trials. Unsurprisingly, we disagree with much of what Judge Frank Easterbrook and Professor Stephen Schulhofer say about that relationship.1 Most of those disagreements need not be rehearsed here; readers attentive enough to wade through their essays and ours will pick up the key points readily enough. But there is one point where the dispute is at once sharp and hidden. It has to do with the fact that both trials and bargains are flawed.

That fact might seem obvious, but the literature on this subject persistently manages to ignore it. And while there are many good insights in Easterbrook's and Schulhofer's comments, we think they both fall prey to the tendency to imagine perfection in one part of the system or the other. That tendency inevitably leads to one of two conclusions-plea bargaining is just fine the way it is, or it ought to be junked. These are, of course, the conclusions that Easterbrook and Schulhofer reach; indeed, they are the conclusions that nearly all scholarship on plea bargaining reaches. Notwithstanding their popularity, we think both polar positions are wrong.

Consider first the bargaining process. Easterbrook argues that bargains are better than trials-that plea bargaining efficiently allocates punishment even if (as he believes) the trial process is riddled with deception and governed by uninformed and sloppy decisionmaking.2 If Easterbrook is right, we should dispense with trials, and let prosecutors decide whom to send to jail and for how long. But bargains are not better than trials, for a reason common to all contracts: no system of consensual allocation is better than the dispute resolution process (legal and extralegal) that backs it up. If courts deciding contract disputes persistently (but randomly) got the parties' intentions wrong, the parties could not simply be trusted to solve the problem on their own, for they must contract in the shadow of the enforcement regime. So too here: prosecutors and defendants bargain against the backdrop of the trial process that will decide their dispute if bargaining breaks down. If the trial process is flawed-if trials sometimes convict innocent defendants-bargaining will not fix the mistakes.

That much is hardly damning, but plea bargaining has deeper flaws. Prosecutors, like insurers, are charged with finding the occasional deserving claim in a sea of frauds. Claims granted too easily only encourage more frauds. And too readily accepting3 a defendant's claim of self-defense, or lack of intent, or mistaken identity--even when the claim has some ambiguous evidence to back it up4 -only encourages copycat behavior by guilty defendants and their lawyers.5 Indeed, defense lawyers are in almost the same position: they must assume innocence claims by their clients are false unless proved, because such claims usually are false-criminal defendants, unlike clients in almost all other settings, have essentially nothing to lose by making false claims. This dynamic is not a problem in cases where bargaining takes place after investigation and evidence-gathering are complete, for then the parties can sift claims in precisely the manner that Easterbrook suggests. But bargaining usually happens earlier, often much earlier, when only the bare outline of the case is known to either side.6