To what extent does the majority opinion writer alter his initial majority opinion to obtain the votes of the other justices? Baum (1997, p. 106) tells us that “it is standard practice [for the justices] to modify the language of opinions in an effort to win colleagues' support,” and Maltzman, Spriggs, and Wahlbeck (1999, p. 55) inform us that

We evaluate these statements in this chapter by examining the data supplied by Maltzman, Spriggs, and Wahlbeck (2000).

Maltzman, Spriggs, and Wahlbeck (2000) investigated opinion coalition formation on the Burger Court and discovered that the justices who dissented at the conference vote on the merits rarely sought any changes in the majority opinion, but when they did, their suggestions were almost always ignored by the majority opinion writer. Thus, if these justices shifted to the majority and joined the majority opinion, they almost always did so without getting the majority opinion writer to change his initial opinion. If any bargaining took place on this Court, it is much more likely to have taken place between the majority opinion writer and the members of the conference coalition.

Since 1958 (Schubert, 1958) and particularly since 1989 (Marks, 1989), a host of Supreme Court scholars have advanced various strategic models in an attempt to explain why the Supreme Court justices behave the way they do.

The purpose of this book is to introduce a great deal of the Supreme Court strategic research to undergraduate and graduate students, to evaluate it, and, at times, to present an original perspective regarding the topics covered by it. Strategic scholars, for example, typically view the majority opinion as a product of bargaining among the justices. In Chapter 8 of this book, however, we use the data supplied by Maltzman, Spriggs, and Wahlbeck for the Burger Court (2000) and show that there was very little individual bargaining over the content of the majority opinion on that Court. In addition, the conventional wisdom is that the individual justices pursue an outcome-prediction strategy; in other words, they vote to grant cert when they expect to win at the final vote on the merits and vote to deny cert when they expect to lose. In Chapter 5, we examine the prior literature, present some new findings, and conclude that there is much less support for this model than many strategic scholars believe.

We have attempted to write a book that can be understood by students who may lack a good methodological background but who are willing to make the effort of following a theoretical argument.

In most cases decided by the Court, the general public does not know that the Supreme Court has decided the case and, if they did know, would not have an opinion whether the case was decided correctly or not. In this chapter, however, we are interested in cases in which the general public knows that the Court has decided the case and has an opinion regarding that decision.

More specifically, we will attempt to answer two questions. First, do the justices attempt to avoid handing down individual decisions that are likely to be perceived negatively by the general public? Second, is the Court likely to be influenced by the “public mood?”

A justice, concerned with the possible negative reaction by the general public to a given decision, might ask herself the following questions: (1) Is it likely that the general public will react negatively to a given possible decision? (2) If it reacts negatively, will such reaction result in a loss of support for the Court? (3) If the general public is less supportive of the Court because of its negative reaction to a given possible decision, will that affect the Court's effectiveness? and (4) If it affects the Court's effectiveness, am I better off deciding the case based on my preferences or should I avoid handing down the decision to avoid the harm? Our reading of Baum (2006, p. 17) influenced our formulation of these four questions.

To understand these chapters it is necessary to know something about the procedures involved in decision making on the Court. We present a short summary of these procedures.

1. The Supreme Court receives requests for review by a number of methods. The most common method is a petition for a writ of certiorari. Certiorari (cert) petitions are of two kinds: paid petitions, in which the petitioner pays the required fees, submits the appropriate number of copies, and follows the other guidelines for submission; and petitions in forma pauperis (i.e., in the form of a pauper), mostly submitted by federal and state prisoners, in which the fees and most of the other requirements are waived.

2. The cert petitions are typically filed by a litigant who has lost in a U.S. Court of Appeals or in a state supreme court.

3. The cert petitions are read by a law clerk from the cert pool and by a law clerk from Justice Steven's office (who does not read all of them). The law clerks read the petition and the lower court record, prepare a memo that summarizes the case, and recommends whether cert ought to be granted or denied.

4. With the help of his law clerks, the chief justice prepares a list of cases to be placed on the Discuss List. Any other justice can add a case to the list. Any case not placed on the Discuss List is automatically denied cert.

5. The justices meet in secret conference to decide whether to grant or to deny the petitions of the cases on the Discuss List. The justices vote in order of seniority, with the chief justice voting first, the senior associate justice (in terms of tenure on the Court), voting second, and so on. If eight or nine justices are voting, it takes four votes to grant cert. But if only six or seven are voting, it takes three votes.

6. […]

Almost all of the cases heard and decided by the current Supreme Court come to the Court after it has granted a petition for a writ of certiorari. The Court grants this writ at its discretion.

Supreme Court scholars have identified a host of variables associated with the granting of cert by the Court. Caldiera and Wright (1988), for example, inspected the cases granted and denied cert during the 1982 term of the Burger Court and discovered nine variables that were associated with granting cert. The three most important were (1) the United States Government as a petitioner; (2) the presence of four or more amicus briefs in support of cert, indicating that the case was salient or important; and (3) the presence of an actual conflict between two or more lower courts or between the court immediately below and a Supreme Court precedent.

In addition, Cameron, Segal, and Songer (2000) examined a random sample of search and seizure cases between 1972 and 1986 and discovered that the Burger Court was likely to vote to deny cert when a conservative lower court had voted in favor of the criminal defendant. The conservative justices on the Burger Court probably reasoned that if a conservative lower court had voted in favor of the criminal defendant, the facts in the case were likely to be so strong in his favor that it was unnecessary for the Supreme Court to review that decision.

For most of the Court's history, the justices used a two-stage procedure at the conference vote on the merits. In the first stage, the justices discussed the case and did so in order of seniority, with the chief justice speaking first. In the second stage, they voted and did so in the order of juniority, with the chief justice voting last. This procedure was ideal from the perspective of the chief justice. In the first stage he was in a good strategic position to persuade his colleagues how to view the case. In the second stage, he was ideally situated to vote with the majority if he so wished and, thereby, to assign the majority opinion. Chief Justice Hughes, for example, took advantage of this two-stage procedure (Dickson, 2001, p. 12). During the later years on the Warren Court, this procedure changed (Murphy, Pritchett, Epstein, and Knight, 2006, p. 633). The two stages were merged into one, and the justices both discussed the case and cast their votes in order of seniority, with the chief justice discussing and voting first.

Chief Justice Warren Burger, who came to the Court in 1969, was stuck with the new procedure. Yet he wanted to assign the majority opinion more often and was burdened with voting first. To solve this problem, Chief Justice Burger, at times, “passed” at his initial conference vote and voted with majority after all the other justices had voted.

Why do the justices on the United States Supreme Court behave they way they do? In an attempt to answer this question, Supreme Court scholars have posited three theoretical models: the legal model, the attitudinal model, and various strategic models.

Models, according to Segal, Spaeth, and Benesh (2005, p. 20), are “a simplified representation of reality. They do not constitute reality itself. A good model serves two contradictory purposes: It accurately explains the behavior in question and it does so parsimoniously, that is, sparingly or frugally.” One can always maintain that a given model explains the behavior, but in science we expect models to be testable or falsifiable. A simple model might be easier to test and may be accurate enough especially when used to explain behavior averaged over a large number of people or for a long period of time. The legal model is the traditional, nineteenth-century explanation of why the justices behave the way they do. Advocates of this model posit that the justices decide cases based on their interpretation of the relevant legal materials. These materials include the U.S. Constitution, federal and state statutes, local ordinances, and the Court's precedents.

The legal model is insufficiently explanatory of decision making on the Court because the justices usually have to interpret ambiguous legal texts and it is uncertain how they ought to do so.

“Uneven development” in this book refers to the fact that many middle-income, developing countries, such as Thailand, succeed at structural change but not, or much less, at upgrading. Why is upgrading rare? I contend that the tasks required for economic diversification, although difficult, are less challenging than those of upgrading. This is so in part because of potential tradeoffs among the three aspects of the latter. Increasing value added with local inputs and doing so at export levels of price, quality, and delivery require that private and public actors overcome a particularly tough set of collective action challenges. Doing so in turn requires significant institutional capacities. But politics makes such capacities rare in the developing world. The creation of institutional capacities is itself a difficult collective action problem that leaders will attempt to resolve only when facing a particularly tough set of domestic and external pressures.

I develop this framework as follows: Section I briefly reviews two approaches – developmental state analyses and new institutional economics (NIE) – whose treatments of collective action are central to my arguments. Section II presents the core of the demand-side component of my argument by identifying (1) types of difficulty along which to differentiate among policy tasks and (2) the institutional capacities appropriate for addressing those tasks. Section III applies the framework to differentiate between the tasks of structural change and upgrading and to anticipate the challenges of more specific policy tasks. Section IV addresses the question of institutional design.

DEVELOPMENT PUZZLES

Students of economic development have commonly focused on the question of why some countries grow and others stagnate, and with good reason. Empirically, the spectacular income gap separating the world's rich and poor nations has become “the central economic fact of our time.” This gap, moreover, now extends beyond the industrialized vs. developing worlds to stark differences among developers. The rich–poor gap has defied easy theoretical explanations. Cultural explanations, for example, foundered on the unexpected success of East Asian countries whose Confucian ethic had previously been identified as an obstacle to growth. Initial endowments of assets such as human resources or manufacturing experience failed to explain the disappointing performance of countries such as the Philippines. Regime type – democracy or authoritarianism – failed to correlate with variation in development outcomes. Postwar development economists' belief that large, state-led projects could liberate entrepreneurs from traditional practices and other rigidities ran up against Latin America's state-led stagnation and Africa's institutional bloat. Finally, the fact that global income levels exhibit not convergence but “Divergence, Big Time,” has undermined the claims of neoclassical economists. As one long-time practitioner notes,

The creation of the modern Thai state in the mid-to-late nineteenth century prompted predictions that Thailand's economic growth would rival that of Japan's newly established Meiji state. Instead, Thailand emerged from World War II and moved into the late 1950s as a preindustrial economy with a healthy but relatively undifferentiated agricultural sector. Beginning in the early 1960s, however, the economy took off. Agriculture began to diversify, and new industrial activities attracted investment. By the 1990s, dynamic growth and diversification had made Thailand an “East Asian Miracle.” Yet apprehension about the shallowness of the country's industrialization grew as diversified export dynamism was matched by rising import dependence and competition from other low-wage producers. The 1997 Asian economic crisis made it clear that these concerns were justified. Upgrading became a central focus of the country's recovery efforts. But almost a decade after the crisis, the World Bank warned that Thai growth could not be sustained unless the country's technological base was improved.

The present chapter traces the institutions and politics of this evolution in order to provide both background for the subsequent sectoral analyses and an initial assessment of the book's framework. The chapter is organized into six time periods, each of which is distinguished by shifts in reform pressures and political competition. The analysis of each period begins with a demand-side focus on key policy tasks and institutional responses.

Thailand poses two challenges to students of economic growth. One is to determine the degree and nature of the country's development: “Is Thailand a case of successful development or rather a case of maldevelopment?” My answer, presented in Section I, is that neither of these options is accurate. Thailand has experienced impressive but still uneven development, whose weaknesses threaten the sustainability of the economy's growth. The second and central challenge of this book is to understand the reasons for this mixed performance. Section II establishes the background for my answer by assessing several prominent, noninstitutional approaches: investment capital, human resources, entrepreneurship, policy, and political regime type. This empirically based assessment draws on Thai and other East Asian experiences and highlights the insufficiency of these explanations, while also noting their value. In the process, it demonstrates the importance of attention to the demand for and supply of institutional capacities.

THAILAND: THE “FIFTH TIGER”?

Achievements

Thailand has, by many criteria, been an economic superstar. Its position as one of the World Bank's High Performing Asian Economies reflects the country's consistent growth and structural change. GDP growth has averaged roughly 7.6% during the 1960s and 1970s, dropping to 5.55% during the debt crisis years of the early 1980s, and then rebounding to 9% during the boom years of 1985–1995 (Table 2.1).