Elizabeth Baldwin

Applying climate policy in practice means considering capital stocks: some assets will pro¬duce pollution whenever they are used, and some will not. Therefore long-term abatement plans should influence current investment. Moreover, newer technologies exhibit learning-by-doing in the deployment of associated infrastructure. We investigate these ideas from both theoretical and numerical perspectives. An increasing carbon tax will reduce investments in assets that pollute, and so reduce emissions in the short term: our “irreversibility effect”. We also show that the optimal innovation subsidy increases with the deployment rate: our “acceleration effect”. Considering second-best settings, we show that, although carbon taxes achieve stringent policy targets more efficiently, subsidies to the “renewables” sector deliver higher welfare when policy targets are more mild. Revised February 2019