Authors: Carolyn Fischer, Richard Newell, and Louis Preonas

Over the last decade, concerns about global warming, local air quality, and energy security have led to a plethora of actual and proposed initiatives at the federal and state levels, particularly in the power sector. These measures aim to reduce emissions, promote electricity generation from renewable sources, and encourage energy conservation. However, little attention has been paid to whether these myriad policy efforts work together or at cross purposes. Research on policy instrument choice in the context of multiple interacting policies and market failures has been identified as an important area of further investigation (Goulder and Parry 2008). In other words, it is important to recognize that the whole of our energy policy mix is going to be quite distinct from the sum of its parts—and possibly less than that sum (Fischer and Preonas 2010).

Fischer and Newell (2008, henceforth FN) assessed different policies for reducing carbon dioxide emissions and promoting innovation and diffusion of renewable energy, with an application to the electricity sector. The stylized model represents two stages, one in which investments in R&D and LBD are made, and a second stage in which the resulting innovations are applied. The article revealed that, due to knowledge spillovers, optimal policy involves a portfolio of different instruments targeting not only emissions, but also learning and R&D. Despite those spillovers, however, the most cost-effective single policy for reducing emissions is an emissions price, followed by (in descending order of cost-effectiveness) an emissions performance standard, fossil power tax, renewables share requirement, renewables subsidy, and lastly an R&D subsidy. In this paper, we extend and update the FN model in several important ways. First, we distinguish between conventional renewable energy sources (like wind or biomass) and advanced technologies (like solar), which have different costs and learning or innovation potential. In this way we can better assess the performance of overlapping policies in terms of the kinds of technological change they induce.

Second, we incorporate a richer representation of electricity demand over time, including short and long-run investments in energy efficiency improvements. As a result, we can incorporate demand-side policies for improving energy or fuel efficiency. We also allow for imperfections in the demand for energy efficiency, as well as in the market for innovation. We analyze how these different imperfections affect optimal policy combinations and also the relative cost-effectiveness of single or otherwise suboptimal policies. Third, we expand our representation of the nonrenewable generating sectors, in order to better evaluate proposals like a Federal clean energy standard (CES). This requires differentiating between natural gas turbines and combined cycle generation, as well as recognizing greater long-run potential for nuclear energy. Finally, we update the entire parameterization based on more recent data, particularly for renewable energy supplies.