Output-based carbon regulations ignore critical types of efficiency
I’m sure relentless efficiency was the intent, but in fact it is very much a way of picking winners, of rewarding one particular type of efficiency at expense of others. The idea is that within industries, a standard will be set for maximum emissions per useful BTU delivered. So if you are heating tomatoes as part of making tomato paste, the standard would apply to your emissions per BTU used to raise the temperature of a tomato. The problem is that while this rewards delivering those BTUs more efficiently, it does not reward heating the tomatoes less, perhaps by substituting a filtering process for some of the heating.
When I brought this up in comments, Sean argued that the second method still rewards by lowering fuel bills. But then, so does the first. If delivering BTUs more efficiently needs an incentive over and above fuel saving, then so does finding a way to use fewer BTUs in the first place.
We’ve been talking a lot about sticks and carrots. We build houses out of sticks all the time; carrots, not so much. If you are trying to construct a policy that uses both, you can’t handle the carrots exactly the way you handle sticks. If you want to penalize emissions, the standard is simple: penalize any emission, with large emissions incurring large penalties and emissions near zero incurring penalties near zero. But if you want to do “carrots” and reward emission reduction, you have to ask, reductions compared to what? Sean tried to find a rule-based answer, but I don’t think you can answer that question with a simple rule. If you want carrots, you will have to go through the messy process of political choice to decide who gets them.
Maybe I’m wrong. Maybe there is such a rule. If there is, it is going to be a lot more complicated rule than one Sean has proposed.