Still your grandfather's boiler: Estimating the effects of the Clean Air Act's grandfathering provisions

While vintage differentiation is a highly prominent feature of various regulations, it can induce significant biases. We study these biases in the context of New Source Review-a program within the US Clean Air Act imposing costly sulfur dioxide (SO2) abatement requirements on new boilers but not existing ones. In particular, we empirically investigate how the differential treatment of coal boil- ers shaped the generation landscape by affecting unit utilization, retirement, and emissions. Leveraging a novel dataset covering state-level sulfur dioxide regula- tions for power plants, we show that the differentiation continued to have a strong effect even 30 years after its passage, raising the probability of surviving another year by 1.5 percentage points for grandfathered boilers and increasing their op- erations by around 800 hours annually. We estimate exempted units would have emitted a third fewer emissions per year, had they been subject to NSR. We run back-of-the-envelope calculations to assess the societal damages associated with the delayed retirements, higher utilization and higher emission rates of the grand- fathered boilers. Focusing solely on the additional SO2 emissions, we estimate annual costs of up to $ 65 billion associated with the vintage differentiation in New Source Review.