This is not (yet) the promised elaboration my thoughts on non-delegation and complexity last week, but rather a clarification and response to Chris’s thoughtful engagement with it.
It’s clear that I should have been clearer about what I was suggesting. Chris (reasonably) interprets my post as calling for a jettisoning of the non-delegation doctrine along with simplifying the judicial review doctrines that stand in as proxies. That would line up with Kathryn Watts’ proposal to a tee. What I was suggesting was in some ways more radical—and even less likely to get accepted by the courts. First, the courts would actually attempt to enforce the non-delegation doctrine directly, thus adding complexity on the constitutional level as they work through the intricacies of deciding how much delegation is too much. Second, and in exchange, the courts would pare down judicial review of administrative action at the APA level: an arbitrary-and-capricious approach that is closer to rationality review; holding agencies only to the APA text’s lenient procedural requirements on notice and comment; and easing up on invalidation of nonlegislative rules for evading notice-and-comment grounds. This would reduce complexity at the APA level by paring away administrative common law that stands in for delegation concerns. (What to do with Chevron is more complicated.) Again, I’m not sure if this grand bargain is a good idea; anything that encourages more constitutional litigation about the validity of legislation starts off with a strike or two against it in my mind. Nevertheless, I find it intriguing.