Another point, and its a basic one we’ve both made in our posts on RPM here: ANY price test for RPM cant be right. You read over and over again, including in the WSJ article (which I thought was sloppy on the literature and evidence), that RPM results in higher retail prices. First, that may or may not be true. It depends on the elasticities, but one can increase the margin by lowering the wholesale price and keeping the retail price constant too. Second, and more importantly, both pro and anti-competitive stories predict a price increase. In the pro-competitive dealer services story, one gets a price increase because a demand shift. Therefore, if we were interested in a consumer welfare test for RPM we better have an output-based test, not a price test to distinguish between the stories.

Third, we may not want to have that sort of test, but instead, some safe harbors and presumptions for legality. Personally, from an error cost perspective, I think this is the way to go. In fact, I’d go so far as to say that the state of the empirical literature on RPM justifies a bright line rule of per se legality in the absence of collusion (which would be a per se violation of a different variety anyway, not really an RPM problem other than it is being used to facilitate a horizontal agreement).