At Law & Liberty poll sci professor James Rogers tackles the titular question, pointing out:
Simple, or naïve, attitudinalism posits Supreme Court justices vote their policy preferences in each specific case they decide. Yet if justices indeed pursue policy outcomes in their decisions, the belief that justices vote their policy preferences in each specific case makes no sense. Justices who seek to maximize achievement of their substantive policy preferences in their judicial decisions will not necessarily vote for the substantive outcome they prefer in a particular case.
In recent years the U.S. Supreme Court hears about 80 cases per term. The actual legal decision in these few cases binds only the specific parties to the dispute. While some cases are more significant than others, that’s still not necessarily a huge amount of political influence relative to the influence the precedent will have on scores, or even hundreds, of related cases that are not heard by the Supreme Court. Even thinking of actual cases ignores the policy impact Supreme Court precedents can have: Precedent influences not only actual cases, but affects behavior that never rises to the level of litigation. Competent legal counsel can advise clients in light of the precedent to behave in a way that leaves them without risk of litigation.
This broader policy impact of precedent can swamp the policy impact of resolving a legal dispute between the two litigants in a case actually before the Court.
As longtime readers may recall, Mitu Gulati and I did on this issue a while back. See Stephen M. Bainbridge & G. Mitu Gulati, How Do Judges Maximize? (The Same Way Everybody Else Does—Boundedly): Rules of Thumb in Securities Fraud Opinions, 51 Emory L.J. 83 (2002).
Our work suggests that Rogers' account of judicial incentives is incomplete because he makes two moves common to the debate over how judges decide cases. First, he focuses on the Supreme Court, despite the fact that that court decides an infinitesimally small amount of American law. In addition, he focuses on Supreme Court cases that have partisan policy implications, which make up an even smaller fraction of judicial work. In a typical Supreme Court term, "approximately 80 percent of votes are in support of the majority opinion, and only about 20 percent of cases are determined narrowly. The 5–4 cases that get national attention are in fact somewhat anomalous." A small sample of anomalous cases is not going to tell us much about how judges in general maximize. (And notice that Roger's post is titled :How Judges Maximize" not "How Supreme Court Justices maximize in the small percentage of their cases that have closely divided partisan policy implications." There is at least the suggestion that his analysis had broad application.)
Second, Rogers implicitly invokes the two basic modes of judicial decision making, which we identified as the Herculean model in which the judge has full information and full knowledge and, generally speaking, gets it right; and (b) the Wannabe model, in which the judge seeks to be Herculean, but errs because he or she is mortal. Two sub-variants of each model depend on what the scholar in question believes that judges seek to maximize: social welfare or personal policy preferences.
Gulati and I pointed out that these accounts fail to account for the fact that judges are agents with incentives to shirk in a variety of ways. We began "by assuming a nonexpert federal judge faced with an overwhelming caseload and limited time and resources with which to decide those cases. We add[ed] the assumption that most judges do not find securities law interesting. From these institutional characteristics, we infer[ed] an explanation for the development and popularity of the heuristics premised on limited cognitive capabilities, resource constraints, and a judicial desire to move cases off the docket in an acceptable fashion."
Instead of leaving [determinations of materiality and intent] for trial, as we show herein, judges are using substantive heuristics to dispose of securities cases at the motion to dismiss stage. In contrast to prior commentary, however, we argue this result reflects not a pro-defendant bias but rather institutional constraints that give judges incentives to eliminate securities cases from their dockets with minimal effort.
Second, our focus on substantive heuristics highlights a previously unobserved link between institutional constraints and the evolution of substantive doctrine. When judges invoke procedural heuristics that enable them to avoid tackling a substantive issue, there is no effect on the evolution of substantive law (except that no law is created). When judges invoke substantive heuristics, however, the use of such heuristics channels and even dominates the on-going evolutionary processes of the (quasi-common) law of securities regulation. As we demonstrate, for example, the development of substantive securities law heuristics has dramatically affected the evolution of the law on both materiality and scienter. At the same time, however, other issues, such as the scope of different duties to disclose, are largely ignored (except perhaps to say why they were not deserving of attention).
Importantly, this is true even at the Supreme Court level.
There is general agreement that the Supreme Court has not done a very good job in the securities area, especially in recent years. Scholars operating in a wide range of paradigms have criticized the court’s recent securities opinions. Supreme Court securities law decisions typically lack a broad, consistent understanding of the relevant public policy considerations. Worse yet, they frequently lack such basics as doctrinal coherence and fidelity to prior opinions.
Why doesn’t the Supreme Court do a better job in securities cases? Our model offers an answer. When deciding securities cases, the Court is faced with hard, dry, and highly technical issues. Supreme Court justices and their clerks arrive on the court with little expertise in securities law. One reasonably assumes that neither the justices nor their clerks have much interest developing substantial institutional expertise in this area after they arrive. (Former Justice Powell being the exception that proves these rules.) Accordingly, it would be surprising if the Court’s securities opinions exhibited anything remotely resembling expert craftsmanship.
Under such conditions, we would expect the justices to take securities cases rarely, typically when there is a serious circuit split, which is in fact what we observe. When obliged to take a securities issue, the Court will seek to minimize the amount of effort required to render a decision. This observation is not intended pejoratively. To the contrary, in terms of our model, the justices are acting rationally. …
Bounded rationality implies that Supreme Court justices (and their clerks) have a limited ability to master legal information, including the myriad complexities of doctrine and policy in the host of areas annually presented to the court. Specialization is a rational response to bounded rationality—the expert in a field makes the most of his limited capacity to absorb and master information by limiting the amount of information that must be processed by limiting the breadth of the field in which he develops expertise. Supreme Court justices will therefore need to specialize, just as experts in other fields must do. Specializing in securities law would not be rational. The psychic rewards of being a justice—present day celebrity and historical fame—are associated with decisions on great constitutional issues, not the minutiae of securities regulation.
The debate over judicial incentives is important. The overemphasis on 5-4 Supreme Court decisions in that debate, however, means that the vast majority of both the academic and public debate--including the colloquy between Rogers and McGinnis--grossly distorts the real picture of how (and what) judges maximize.