Political scientists have conducted much work examining a court's decision on the merits of a case. We have concluded that ideology has a strong influence on the outcome on the merits of a decision. Furthermore, courts seek to render a decision that is closest to their own policy preferences. However, federal circuit courts within the judicial hierarchy are constrained by other actors according to the strategic model. There is an abundance of evidence showing that superior actors constrain courts' ideological preferences when such courts render decisions on the merits. However, there is a dearth of scholarship regarding judicial decision making on threshold issues. I argue that federal circuit courts set their judicial agendas by transforming their mandatory appellate jurisdiction into one that is discretionary. They achieve this goal by controlling the type of litigants who gain access to the courts by deciding cases on threshold grounds. I also argue that federal circuit courts are responsive to changes in Congress's ideology because Congress has power to control threshold issues through various mechanisms. I seek to establish that the grounds upon which as case is dismissed -- jurisdictional, justiciable, and procedural -- defines the parameters that constrain federal circuit courts.

This dissertation examines the relationship between three sets of variables: corporate governance and monitoring, the quality of disclosure in annual reports and securities class action litigation. In the first section, I present a game-theoretic model in which shareholders select from ex ante monitoring or ex post litigation mechanisms available to them in order to mitigate the agency problem. Firm characteristics determine the choice of which of these two mechanisms is appropriate for a particular company. I then test predictions from this model and find that firms with poor monitoring are much more likely than those with good monitoring to be sued even after controlling for the common determinants of a lawsuit. The second section of the dissertation relates the quality of disclosure in annual reports to litigation. I use a dataset containing annual reports filed electronically with the SEC in the period 1996-2005. Using two content analysis software programs that analyze the categories of words used in these annual reports, I find that firms that use more numbers, past and future words, and other informative words are much less likely to be sued, even after controlling for the common determinants of lawsuits. In order to avoid subjectively choosing categories, I use principal components analysis to identify the major components of annual report disclosure. When these components are used as regressors to identify causative factors of lawsuits, one component named 'informativeness' has significant power to explain subsequent lawsuits. In head-to-head comparisons of the 'informativeness' principal component with Standard & Poor's Transparency and Disclosure score, my informativeness measure is more effective than the S&P score in predicting the likelihood of a lawsuit. Finally, in cross-sectional tests, I find support for the theory that firms with good boards and managers who are not entrenched have better disclosure practices. Further, monitoring by institutional investors, independent boards and analysts appears to induce better corporate disclosure. / text