Each term 4,000 or more cases arrive at the doorstep of the Clerk of the Supreme Court of the United States. The Court narrows the field of decision from all possible choices to a smaller set of the most plausible ones; in a typical term, the Conference places 20 to 30 percent (around 500) of the cases filed on its “discuss list,” only 150 or so of which will be selected for plenary review. Here we investigate the composition, sources, and implications of the Court's discuss list. What criteria do the justices use in creating the list of cases for discussion in conference? Do these criteria differ from those ultimately applied in the decision to grant or deny a writ of certiorari? What, if any, implications does the operation of the discuss list hold for the composition of the agenda and the outcome of decisions on the merits in the Supreme Court?
We test two chief sets of hypotheses: (1) the justices weigh the various formal and informal criteria differently across the two stages of agenda building, and (2) despite differences in weighting, the justices rely on briefs amicus curiae as well as ideological predispositions to help them both to identify logical candidates for discussion and to decide whether to grant certiorari. We explain the variation in the weighting of the criteria as being largely due to the cost and accuracy of information and the different risk of errors during the two phases of choice.