WORKING PAPERS

• Kellogg, Ryan, Economics of Demand-Side and Supply-Side Climate Policies, working paper, in preparation for Brookings Papers on Economic Activity .
  - NBER working paper #35830 (September, 2026)

  [abstract]

  Should a jurisdiction seeking to cut greenhouse gas emissions use demand-side climate policies that target fossil fuel consumption? Or should it introduce supply-side policies that target extraction? My analysis points to a policy portfolio rather than an either/or choice. ``Leakage'' of emissions to unregulated jurisdictions can be mitigated by combining standard demand-side carbon emissions pricing with supply-side pricing of carbon extraction. Emissions pricing drives leakage by lowering fossil fuel prices abroad; extraction pricing counteracts this effect. Supply-side climate policies in practice, however, do not broadly price extraction but instead narrowly foreclose investments in fossil fuel infrastructure and are thus inefficient. To reduce carbon demand, an alternative to emissions pricing is support for clean energy. I emphasize reforms to transmission planning and utility governance as crucial for unlocking clean energy resources' potential, and I highlight the case for publicly supporting clean energy R&D, including green industrial policy, to reduce carbon emissions not just at home but also abroad. I close by discussing distributional consequences. Domestically, carbon pricing and clean energy subsidies have sharply different distributional profiles. Globally, carbon extraction pricing would harm low-income countries that are also fossil fuel importers, though there also exist low-income fossil fuel exporters that would benefit.

• Covert, Thomas R., Konan Hara, Ryan Kellogg, and Richard L. Sweeney, Investment, Productivity, and Selection in the U.S. Shale Boom, working paper .
  

  [abstract]

  The goal of this paper is to understand why output per well increased so dramatically during the shale boom. A core question is the extent to which production growth arose from changes in well site selection---i.e., changes in the underlying geological quality of the precise areas being drilled---or from changes in firms' operational decisions, such as changes in inputs and adoption of new technologies. We develop a joint model of well-level production and drilling decisions that allows us to address site selection that comes from firms' private ex ante signals of sites' underlying geologic quality and from firms' ability to learn about quality over time by observing drilled wells' production outcomes. The model is designed to account for not just whether but when sites are drilled. We estimate the model using data on drilling, production, and leasing from two major U.S. shale plays: the Bakken in North Dakota and the Haynesville in Louisiana. We find that site selection has been important: firms initially drilled areas for which they had initial signals of high quality, and then they later repeatedly drilled in areas that had positive production outcomes. But after accounting for selection, as well as reservoir depletion, firms' operational improvements emerge as the largest driver of output growth. These improvements occurred within firms, rather than being driven by reallocation of drilling activity to more productive firms. Price and cost data suggest that these gains were driven by improvements in the productivity with which firms brought together drilling inputs.