I recently came across two articles on SRRN and forthcoming in the Virginia Law Review: Chevron's Two Steps by Bamberger and Strauss (Columbia) and Chevron Has Only One Step by Stephenson and Vermeule (Harvard). Both articles provide theoretical models for judicial interpretation of statutes and agency decision-making process under Chevron.
There are at least four potential parts of judicial review of agency interpretations of federal statutes:
(1) Chevron Step 0: Not usually counted as a "step," addressing whether Chevron applies at all. See Sunstein, Chevron Step Zero.
(2) Chevron Step 1: A determination of statutory ambiguity using traditional tools of statutory construction.
(3) Chevron Step 2: A reasonableness inquiry into perhaps the nature of the policy choice or perhaps the agency decision-making process.
(4) APA/Hard Look Review: See State Farm. Not counted as a "step," but may be implicated in Step 2.
After reading the papers above and their theoretical frameworks which suggest how judges should act and why, it continues to occur to me that very little data exist informing scholars how judges are actually dealing with Chevron.
The Chevron doctrine's complexity makes empirical inquiry especially difficult. The Miles/Sunstein piece Do Judges Make Regulatory Policy? An Empirical Investigation of 'Chevron' is excellent, but they do not seek to separate out the various Chevron "steps." My own paper An Empirical Investigation of Judicial Decisionmaking, Statutory Interpretation & the Chevron Doctrine in Environmental Law suggests a number of problems, at least in the environmental law context, exist in implementing Chevron. For example, some judges do turn Chevron into a single step process, and some judges see no difference between step 2 and hard look review while others do.
Thus, I'm curious to learn what other papers are in this forthcoming issue of the Virginia Law Review.... Perhaps a paper entitled, "Does Chevron Really Have Four Steps?"