Academic journal article Journal of Regional Analysis & Policy

Residential Electricity Demand in El Paso

Academic journal article Journal of Regional Analysis & Policy

Residential Electricity Demand in El Paso

Article excerpt

(ProQuest: ... denotes formulae omitted.)

1. Introduction

Classical demand theory systematically includes income as a primary determinant (Barten, 1968). Most of the existing empirical literature on the residential demand for electricity reports a positive relationship between income and consumption, suggesting that electricity is a normal good (Dergiades and Tsoulfidis, 2008). However, some studies indicate that electricity is an "inferior good" with sales that are negatively correlated with incomes (Roth, 1981). That includes evidence that residential electricity may behave like an inferior good in the United States as a whole (Contreras et al., 2009). Research conducted at the regional level may shed additional light on the responsiveness of residential electricity demand to changes in income.

This study examines the residential demand for electricity with data for the low income metropolitan economy of El Paso, Texas. Most utilities do not serve native loads that encompass entire states and, therefore, are more interested in localized service regions. Another reason to study electricity demand at the metropolitan level is that different regions across nations and within individual states may exhibit considerable economic and customer base heterogeneity (Provenzano and Walasek, 1981; Cebula, 2012). Income heterogeneity may lead to biased estimators, especially in large states such as Texas where earnings disparities are quite large (Wooldridge, 2009; Ayllon, 2013).

El Paso Electric Company provides electricity service to El Paso County, Texas. The company is an investor-owned utility providing electric energy to approximately 380,000 retail customers in a 10,000 square mile area of the Rio Grande Valley in west Texas and southern New Mexico. El Paso Electric has a net dependable generating capability of approximately 2,010 megawatts (MWs) and a 2015 native peak demand of 1,787 MWs. New generation is already under construction to allow the company to keep pace with an estimated 2.9 percent compound annual growth rate in MW peak demand during the 2011 to 2020 forecast period (Patton, 2012).

An autoregressive distributed lag (ARDL) bounds testing approach is used to analyze El Paso residential electricity usage. This approach is relatively attractive because it allows estimating both long-term and short-term consumption relationships within a single modeling framework (Halicioglu, 2007). Another useful feature of this general approach is that its informational requirements are typically not very extensive. For many regions in the United States and elsewhere, county and municipal data are much less abundant than macroeconomic variables, making it difficult to carry out econometric analyses that require substantial statistical detail. For El Paso Electric, annual frequency data from 1977 to 2014 are available to complete the study.

A three-year out of sample forecast is used to evaluate the rates of growth in demand that can be expected given a continuation of recent historical trends in the explanatory variables. Correctly forecasting the demand for electricity consumption is crucial for electric utilities throughout the United States. The construction of a new generating unit to serve increases in native load demand almost always requires several years to complete. Failure to file far enough in advance for a new permit to build generating plant additions will force utilities to have to engage in expensive off-system power purchases. Knowledge of the demand for electricity and the accurate estimation of future demand growth have important economic and regulatory repercussions and are critical elements in the planning process for all electric companies (Dortolina and Nadira, 2005).

2. Literature review

The first econometric studies of the demand for electricity can be traced back to Houthakker (1951). Results in that study point to relatively strong sensitivity of British electricity usage to changes in price and income. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.