Looking the wrong way: Bias, renewable electricity, and energy modelling in the United States

Journal Article


The United States Energy Information Administration releases an AEO (Annual Energy Outlook) projecting future supply, demand, and resources for energy and electricity in the U.S. It is widely relied upon for policymaking. This study assesses twelve years of these projections of generation and capacity for six classes of renewable technologies. It finds consistent under projections for most renewable energy types in the mid- and long-term, due to inaccuracies, limitations, and inconsistencies in the underlying model. It identifies and evaluates five hypotheses that may explain such inaccuracy: inaccurate modelling of state renewable energy mandates, expiration of renewable tax credits, flaws in modelling structure, a biomass co-firing assumption, and capital cost projections. Unless AEO’s projections of renewable energy are greatly improved, the reliability of its sector-wide electricity projections is inherently low. Key modifications suggested by this study include: fully valuing financial and non-financial benefits of renewables, improving cost innovation expectations for renewable energy, and, perhaps most importantly, properly modelling state renewable energy mandates.


  • We examine how effectively the AEO (Annual Energy Outlook) projects renewable energy.
  • The AEO consistently under projects most types of non-hydro renewables.
  • These errors are partially due to a dynamic and changing policy environment.
  • They are also due to several methodological issues, particularly for policy and cost.
  • Until these issues are corrected, the reliability of the AEO is inherently low.