The Administrative Procedure Act instructs federal courts to invalidate agency decisions that are quot;arbitraryquot; or quot;capricious.quot; In its 1983 decision in the State Farm case, the Supreme Court firmly endorsed the idea that arbitrariness review requires courts to take a quot;hard lookquot; at agency decisions. The hard look doctrine has been defended as a second-best substitute for insistence on the original constitutional safeguards; close judicial scrutiny is said to discipline agency decisions and to constrain the illegitimate exercise of discretion. In the last two decades, however, hard look review has been challenged on the plausible but admittedly speculative ground that judges' policy preferences affect judicial decisions about whether agency decisions are quot;arbitrary.quot; This study, based on an extensive data set, finds that the speculation is correct. Democratic appointees are far more likely to vote to invalidate, as arbitrary, conservative agency decisions than liberal agency decisions. Republican appointees are far more likely to invalidate, as arbitrary, liberal agency decisions than conservative agency decisions. Significant panel effects are also observed. Democratic appointees show especially liberal voting patterns on all-Democratic panels; Republican appointees show especially conservative voting patterns on all-Republican panels. Our central findings do not show that judicial votes are dominated by political considerations, but they do raise grave doubts about the claim that hard look review is operating as a neutral safeguard against the errors and biases of federal agencies. Because judicial policy commitments are playing a large role, there is a strong argument for reducing the role of those commitments, and perhaps for softening hard look review
