Judicial Deference to Administrative Statutory Interpretation in the Modern American Administrative State

The American administrative state of the twentieth and twenty-first centuries is defined by deference by federal courts to administrative agencies. The political science and (especially) legal literatures have long discussed how federal courts defer to agencies, but little attention has been dedicated to how to identify deference and why courts defer. This dissertation redefines deference, a term that has been topic of extensive discussion in the last forty years but that was missing a key feature: the intent of the deferrers. Using administrative courts as the proxy for agencies at large, this dissertation suggests three reasons why judges may defer. First, an Article III court might defer to an administrative court by the advice of Chevron v. National Resource Defense Council (1984), a case that provided an explicit declaration in favor of deferring to agencies on the subject matters of which they are an expert. Second, an Article III court might defer to an administrative court when the courts are staffed by co-partisans (i.e. when the partisanship of the Article III court panel and the administrative court panel match). Third, an Article III court might defer to an administrative court when the Article III court would like to communicate instructions on good judicial practice to administrative law judges. To test these theories, this dissertation utilizes a new approach to identifying deference. Using the universe of precedent decisions at the Board of Immigration Appeals, I identified each decision where there was a companion case in the various courts of appeals. This created dyads of cases (n = 116) that provide the opportunity to trace a dispute through two different judicial institutions. I coded each dyad on several variables that provided analytical leverage on all three theories. Upon explicating the research design and case selection featured in this dissertation, I introduce the four forms of deference. If a court rules in favor of an agency and there is evidence in the text of the opinion that the court is actively deferring to the agency, then we can code this instance as active deference. If a court rules in favor of an agency and there is no evidence in the text of the opinion that the court is actively deferring to the agency, then we can code this instance as passive deference. If a court rules against an agency and there is no evidence in the text of the opinion that the court is actively not deferring to the agency, then we can code this instance as passive non-deference. If a court rules against an agency and there is evidence in the text of the opinion that the court is actively not deferring to the agency, then we can code this instance as active non-deference. In this deference scheme, even non-deference has theoretical significance. When a court provides rationales directly related to the actions of the agency, it is the rationale rather than the outcome that matters. When we combine an agency\u27s outcome in court with active language either for or against the agency\u27s actions, we can confidently make conclusions about judicial intent regarding administrative behavior in a way that is impossible when solely relying on win rates