Output and Attribute-Based Carbon Regulation Under Uncertainty

Key Findings

• When there is uncertainty about future demand for the good’s output or the value of one of its attributes (e.g., vehicle footprint), at least some amount of output or attribute-basing improves expected welfare relative to a “flat” emissions standard. Substantial output-basing, or even an intensity standard (which regulates emissions proportionately to output), can improve economic efficiency when uncertainty about future demand is large relative to the damages caused by emissions.
• Uncertainty about fuel prices alone cannot justify output or attribute-based standards on economic efficiency grounds.
• For attribute-based fuel economy standards, uncertainty about future demand for vehicle size is so small that the optimal footprint-based standard is essentially flat. Thus, accounting for uncertainty does not substantially alter the conclusions of Ito and Sallee (2018) that footprint-based fuel economy standards are inefficient.
• Uncertainty about the future demand for electricity is large, so that an intensity standard—such as the rate-based standards envisioned by the Obama-era Clean Power Plan–can potentially out-perform a fixed emissions standard. Under a flat standard, uncertainty implies that the standard has a high risk of never binding (if demand is low or the price of low-carbon fuels is low) or of imposing a very high abatement cost (if demand is high or the price of low-carbon fuels is high).
• A tax on carbon yields strictly greater expected welfare than flat, output-based, or attribute-based standards, since the tax eliminates abatement cost uncertainty while avoiding distortions to output or to goods’ attributes. The same is true of an emissions cap that is indexed to exogenous sources of uncertainty (such as fuel prices or GDP) rather than to endogenously-determined objects such as goods’ output or attributes.