Principal Agustin J. Ros has authored an article published in The Energy Journal that estimates electricity demand elasticity for different customer classes as well as the price effects from unbundled wholesale generation in conjunction with retail competition in the United States during the period 1972-2009.
Brattle Principal Agustin J. Ros has authored an article published in The Energy Journal that estimates electricity demand elasticity for different customer classes as well as the price effects from unbundled wholesale generation in conjunction with retail competition in the United States during the period 1972-2009.
Dr. Ros finds that the growth in total electricity demand averaged close to 2% per year during the period with a considerable slowdown in growth occurring for all customer classes in the later years. Since 2000, the study finds that electricity demand averaged less than 0.5% per year. When controlling for factors that affect the demand for electricity (such as deflated prices, income, and population) the study finds that, on average, during the period considered residential and commercial electricity demand increased approximately 1% per year while, on average, industrial electricity demand growth decreased at slightly less than 1% per year—indicating the existence of important hard to measure factors affecting overall demand for electricity.
Additionally, Dr. Ros examines electricity prices (using average revenue per unit of output as a proxy for price) during the period and finds that the average growth in deflated electricity prices was -1.0% per year during the period. The study also estimates the combined effects of unbundled wholesale generation and retail competition on deflated electricity prices by comparing prices in those states that permit retail consumers to choose alternative electricity suppliers and those states that do not, and finds that the combined effects of unbundled wholesale generation and retail electricity competition is associated with lower deflated electricity prices. The study controls for different factors affecting price and how much leverage natural gas costs has had on changes in regional electricity prices that were bundled compared to those that were not along with other factors that caused the restructured states to start from different, often higher average costs than other regions.
Dr. Ros’ economic analysis uses utility-specific panel data covering 72 electricity distribution companies from 1972 to 2009 to econometrically estimate structural demand and reduced-form price models for residential, commercial, and industrial customers.
Based on Dr. Ros’ analysis, the study finds:
- The short and long run price elasticity of demand varies between –0.40 to –0.61 (for residential demand), between –0.33 to –0.77 (for commercial demand), and is approximately –0.60 for industrial demand, generally consistent with prior estimates. He finds that while static demand models work well for residential demand, dynamic models are more appropriate for the larger customer classes who require more time to adjust to changes in prices.
- The income elasticity of demand varies between 0.34 to 0.41 (for residential demand), 0.43 to 0.79 (for commercial demand), and 1.3 to 4.6 (for industrial demand).
- Unbundled wholesale generation and retail competition is associated with lower deflated electricity prices with the mean total impact being –4.3%, –8.2%, and –11.1% for residential, commercial, and industrial customers, respectively, with the impact diminishing over the period for residential customers, remaining relatively constant for commercial customers, and increasing for industrial customers.
- In all price models, total factor productivity is consistently the most significant explanatory factor with a 1% increase resulting in a reduction in deflated electricity prices ranging between 0.05% and 0.30%, depending on the econometric model.
The data analysis and methods used in this study to measure and quantify the changing nature of electricity demand in the U.S. can be used to help utilities better gauge and predict their changing load in the face of the many factors that are causing overall per-capita electricity growth to slow significantly.
The article, “An Econometric Assessment of Electricity Demand in the United States Using Utility-specific Panel Data and the Impact of Retail Competition on Prices,” can be accessed on The Energy Journal’s website.
Published in The Energy Journal, Volume 38, Number 4