Poor lane discipline cannot impede free flow conditions by definition. Poor lane discipline is irrelevant to capacity because it isn't possible to measure in near capacity conditions. Likewise it is a circular argument If you have free flow conditions no amount of lane changing makes a difference, if lane changing made a difference then you no longer have free flow conditions putting you back in the impossible to measure column.
Your premise is based on the idea that there is a system optimum that is possible but not obvious to the individual participant. That's possible for many large systems but you don't understand that the roads are designed to be optimal by having people attempt individual best outcome decisions. It isn't a perfect but you cannot fix it with KRETP because in its' extreme example you'd have an entire freeway stopped in the right lane and no one using the other lanes. What's that you say? Someone will decide to pbutt? More than one? See where that ends up? Right back at fully developed plug flow across all lanes. It still all boils down to a selfish thought experiment; "If only all these other drivers would pull to the right I could sail past this congestion."
