Estimating Price Elasticity using Market-Level Appliance Data

This report provides and update to and expansion upon our 2008 LBNL report “An Analysis
of the Price Elasticity of Demand for Appliances,” in which we estimated an average relative
price elasticity of -0.34 for major household appliances (Dale and Fujita 2008).
Consumer responsiveness to price change is a key component of energy efficiency policy
analysis; these policies influence consumer purchases through price both explicitly and
implicitly. However, few studies address appliance demand elasticity in the U.S. market and
public data sources are generally insufficient for rigorous estimation. Therefore, analysts
have relied on a small set of outdated papers focused on limited appliance types, assuming
long-term elasticities estimated for other durables (e.g., vehicles) decades ago are applicable
to current and future appliance purchasing behavior. We aim to partially rectify this problem
in the context of appliance efficiency standards by revisiting our previous analysis, utilizing
data released over the last ten years and identifying additional estimates of durable goods
price elasticities in the literature.
Reviewing the literature, we find the following ranges of market-level price elasticities:
-0.14 to -0.42 for appliances; -0.30 to -1.28 for automobiles; -0.47 to -2.55 for other durable
goods. Brand price elasticities are substantially higher for these product groups, with most
estimates -2.0 or more elastic. Using market-level shipments, sales value, and efficiency level
data for 1989-2009, we run various iterations of a log-log regression model, arriving at a
recommended range of short run appliance price elasticity between -0.4 and -0.5, with a
default value of -0.45.