In Chapters 5 through 11 we investigated intracourt strategies, in other words, strategies that a justice might pursue in light of how his colleagues on the Court were likely to behave. In this chapter and in Chapter 13 we will explore strategies that a justice might follow based on how outside audiences are likely to behave.

One of the Supreme Court's most important jobs is the interpretation of federal statutes. At times, it decides whether a federal statute is unconstitutional or not. More often, it merely decides what a statute means. In other words, we can distinguish between the Court's constitutional decisions and its statutory decisions. When the Court hands down a statutory decision, Congress, in turn, can enact an overriding statute. In response, the Court may have the opportunity of interpreting the new statute. The process, of course, can go on indefinitely, with neither institution having the last licks.

The separation of powers (SOP) model mainly focuses on the statutory decisions of the Court. Baum (2007, p. 147) describes this model:

In reaction to the weaknesses of the legal model, Pritchett (1948) proposed an attitudinal model. This model was the dominant model for explaining the final vote on the merits from the 1950s until the end of the 1990s. Some scholars (Segal and Spaeth, 2002) believe that it is still the dominant model for explaining this vote. It is widely accepted among Supreme Court scholars (or at least among most political scientists) that the attitudes of the justices are the most important determinants of why some justices confronted with the same set of cases vote for the liberal outcome, whereas other justices vote for the conservative outcome. Baum (2007, p. 149), for example, concluded that

Segal and Spaeth (2002) present an even more elaborate explanation of why the justices can be expected to cast sincere, attitudinal votes at the final vote on the merits.

It is common knowledge that, at times, an individual Supreme Court justice will vote to reverse the decision of the lower court at the conference vote on the merits and will vote to affirm that decision at the final vote or the converse. In the Vinson Court era (the 1946 term through the 1952 term) and during the 1957–1967 period of the Warren Court, the individual justices switched their votes in 10 percent of the pairs of votes (Brenner, 1989) and in the Burger Court the percentage was 7.5 percent (Maltzman and Wahlbeck 1996, p. 587). There is not a great deal of fluidity in voting because “judges do not treat a vote, though nominally tentative, as a hypothesis to be tested by the further research conducted at the opinion-writing stage. That research is mainly a search for supporting arguments and evidence” (Posner, 2008, p. 110).

The question that interests us in this chapter is the extent to which the fluidity in voting constitutes strategic voting. Maltzman and Wahlbeck (1996, p. 581) conclude that

Baum (1997, p. 106), however, presents a mixed picture:

Our previous discussion has focused mainly on the justices' pursuit of ideological or legal goals. Because the justices disagree regarding what ideology or interpretation of the law they favor, when an individual justice pursues her ideological or legal goals she does so at another justice's expense.

The justices, however, have shared goals as well. They want the Court to decide cases efficiently. They seek harmonious relations with each other, and they want the Court to be autonomous, influential, and respected. These shared goals are secondary goals. It is doubtful, for example, that any justice's primary goal is to have harmonious social relations on the Court or even to maximize the Court's influence. These shared goals almost certainly are not the ultimate aims of the justices. Nevertheless, the attainment of these secondary goals will often facilitate or enhance the primary goals, which are likely to be attitudinal or legal.

Strategies are often helpful in attaining these shared goals. These may be called “coordinating strategies” or, if they involve formal rules, “institutional strategies.” Supreme Court scholars have not paid a great deal of attention to these strategies, but it is important to do so to understand decision making on the Court. In this chapter we argue that the following rules may make sense in light of such goals:

1. The Rule of 4 for cert decisions

2. Independence of the cert vote and the merits vote

3. Rules regarding access to the Court

4. Outcome voting

5. The narrowest grounds rule

6. Respect for precedent (stare decisis)

7. […]

There are a host of additional questions that can be explored with strategic and other theoretical tools. Consider the following:

1. Why have recent Courts granted cert to many fewer cases than in the past? During the 1970 through the 1978 terms, for example, the Court granted cert to more than 200 paid petitions per term (Epstein, Segal, Spaeth, and Walker, 2007, Table 2-6). In contrast, in the 1992 to 2007 era it never granted cert to 100 paid petitions in any term. Is it surprising that the justices do not hear as many cases as possible?

2. Why does the Court grant cert to cases “that do not possess any political significance” (Cross, 1998, p. 559)?

3. Why does the Court grant cert to many cases they affirm (Cross, 1998, p. 560)? Is this the best use of the Court's resources?

4. If the justices cast four votes to grant cert and five to deny cert, why do the five rarely vote to DIG (dismiss it as improvidently granted) the case (Segal and Spaeth, 2002, p. 435)?

5. “Given the advantage of speaking first and voting last, why in the world did the Chief Justice even begin voting first” (Segal and Spaeth, 2002, p. 435)? Note that the change took place when Warren was chief justice.

6. Brenner and Spaeth (1988) inspected minimum winning (MW; 5–4, 4–3) conference coalitions on the Warren Court and discovered that the marginal justice (i.e., the justice ideologically closest to the dissenters) was assigned to write the majority opinion more than twice as many times than can be expected based on random assignment. Why is this justice favored in this situation?

7. […]

Although United States Supreme Court scholars have long recognized that the majority opinion constitutes “the core of the policy-making power of the Supreme Court” (Rohde and Spaeth, 1976, p. 172), only recently have a number of scholars systematically investigated the processes involved in the drafting of this opinion (e.g., Epstein and Knight, 1998; Maltzman, Spriggs, and Wahlbeck, 2000; Bonneau, Hammond, Maltzman, and Wahlbeck, 2007). But none of the recent research supplies us with a good answer to the question: At whose ideal point will the majority opinion be written? In their conclusion to their famous text, Segal and Spaeth (2002, pp. 434–435) list this question as one of the six questions that ought to be explored. And Westerland (2003, p. 2) maintains that this question is “arguably the most important outstanding question about what the Court does.”

Why should Supreme Court scholars be interested in at whose ideal point the majority opinion will be written? The main reason is that answering this question will increase our understanding of (1) the content of the majority opinion, (2) coalition formation on the Court, (3) the goals of the majority opinion assigner and writer, and (4) the consequences of majority opinion assignment. Concerning the latter, if the majority opinion writer tends to write the opinion at her own ideal point or at a point close to it, which justice is assigned to write the opinion in a given case is of great importance, but which justice is assigned to author the majority opinion is not of great importance if the majority opinion writer tends to write the majority opinion either at the ideal point of the Court median or at the ideal point of the conference coalition median.

Rohde (1972) viewed opinion coalition formation on the Court as a bargain between the majority opinion writer and the other justices on the Court wherein the majority opinion writer will alter her ideal opinion in exchange for the other justices' votes. Although the majority opinion writer will desire an opinion that reflects her ideal position, she would be willing to modify it to obtain sufficient additional votes for it so that her opinion will be an authoritative opinion of the Court (also known as a majority opinion), instead of a plurality opinion. Four votes will be needed if eight or nine justices are participating in the case. The majority opinion writer, however, would be unwilling to change her opinion after a minimum winning (MW) vote is secured, and the other justices will not join the opinion unless additional changes are made. As a consequence, opinion coalitions on the Supreme Court will be MW.

This MW model pertains only to nonthreat cases (i.e., to cases in which no outside group is threatening the Court). In threat cases, Rohde argued, the Court is likely to form a united front against the threat. In other words, some justices would be willing to vote against their ideological predispositions to protect the Court against the threat.

This model is based on a number of unrealistic assumptions. First, Rohde assumes that the majority opinion writer will seek an opinion that reflects her views.

Do the justices exchange votes or join each other's opinions? In groups like the United States Supreme Court – that are small and repeatedly engage in collective decision making – such an exchange is likely to be a beneficial strategy for all the actors. It certainly could make sense for the Supreme Court justices. Suppose that, contrary to her personal preferences, Justice A votes with other justices on some cases about which she cares less, and the other justices vote with her on some cases about which they care less. All the justices ought to end up more satisfied. Such an exchange, however, will violate the ideal of justice, according to which the outcome of each case ought to depend on its own merits. The justices who aspire to this ideal, therefore, may refuse to engage in such an exchange. In this regard, Justice Breyer (Toobin, 2005) stated:

We agree that justices do not exchange votes on cases in such a way as to affect the decisions in the cases, but the justices still may join one another's opinions reciprocally or they may leave minorities and join majorities reciprocally. We focus on these two possibilities in the remainder of this chapter.

In Chapters 5 through 14 of this book we have explored whether the justices on the Court behave strategically. What have we discovered? We first examine the evidence that the justices are not behaving strategically:

1. In cases granted cert by the Court, reverse-minded justices do not vote in accord with the outcome-prediction strategy. Affirm-minded justices in the aggregate do so, but only about half of the individual affirm-minded justices pursue this strategy. In cases denied cert by the Court, the justices do not vote in conformity with this strategy (Chapter 5).

2. Even though Chief Justice Burger, at times, voted insincerely and with the majority at the conference vote on the merits, there is no evidence that the associate justices on the Court voted this way (Chapter 6).

3. Although we identified some possible strategic reasons for justices changing their votes between the conference vote and the final vote, the justices almost always shift their votes for nonstrategic reasons, for example, because they have changed their mind regarding which outcome they favor or because they were in the minority at the conference vote and wish to conform to the majority (Chapter 7).

4. Successful bargaining over the content of the majority opinion between the majority opinion writer and the other members of the conference coalition took place in less than 6 percent of situations on the Burger Court (Chapter 8).

5. Majority opinion coalitions usually are not minimum winning (MW) (Chapter 9).

6. The majority opinion is usually not written at the ideal point of the median justice on the Court, even when that justice is a member of the conference coalition (Chapter 10).

7. […]

If the justices are “strategic actors,” as Epstein and Knight (1998, p. 10) maintain, what is their goal? According to Baum (2006, p. 6) a strategic justice will want the Court and the government as a whole to adopt a legal or public policy that is as close as possible to the policy she favors. A strategic justice, for example, might vote to grant certiorari (i.e., vote in favor of the Court hearing and deciding the case), in part, because she expects the Court to adopt her preferred policy when it hands down its decision (see Chapter 5). In addition, a strategic justice might vote insincerely and with the majority at the conference vote on the merits because she believes that such voting might enable her to influence the content of the majority opinion and, thereby, obtain some damage control (see Chapter 6). Finally, a strategic majority opinion writer might be willing to modify her ideal opinion to attract four additional votes for it, so that she can hand down an authoritative Opinion of the Court (also called a majority opinion), instead of a plurality opinion (see Chapter 9).

In contrast, a sincere justice will vote in accord with his preferred legal or policy position without considering the influence of his vote on the legal or policy position that will be adopted by the Court or the government as a whole.

Supreme Court scholars disagree regarding the extent to which the justices on the Court are strategic actors.

This book mainly focuses on the use of strategies by the justices of the United States Supreme Court. In this chapter, however, we preface this endeavor with a consideration of why a losing litigant might seek review at the Supreme Court level.

Why would anyone take a case to the United States Supreme Court? If one weighs costs and benefits, except for the unpaid or in forma pauperis petitions, the financial costs of submitting a petition for a writ of certiorari are large, and the chances of getting one's case heard are slim. In the 2004 term of the Court, for example, only 69 of the 2,041 paid petitions (3 percent) were granted cert (Epstein, Segal, Spaeth, and Walker, 2007, Tables 2–6). The percentage of the unpaid petitions was even lower (0.2 percent). Even if cert is granted, there is a good chance that the petitioner will lose on the merits. Epstein and Knight (1998, p. 27), for example, tell us that in the period from 1953 through 1994, the Court affirmed the decision of the lower court 38.7 percent of the time. It is certain that the financial costs will increase dramatically if cert is granted.

Despite these percentages, we can suggest why it might be rational for a litigant who has lost in the lower court to file a petition for a writ of certiorari. There has been little research on this topic to date, so for the most part we will reason theoretically.