A Note on Price Asymmetry as Induced Technical Change

Abstract

This note evaluates whether fixed time effects (yearly dummy variables) are a better representation than separate price-decomposition terms for induced technical change in energy and oil demand. Fixed time effects are a proxy for all omitted variables that change similarly over time for all countries. Many of these omitted variables have little relevance to technical change. Empirically, statistical tests applied to previous studies reject an important premise of the fixed-time-effect model that energy or oil demand responds symmetrically to price increases and decreases. Moreover, when price-decomposition techniques allow for price-asymmetric responses, the estimated income elasticities are not dramaticalxly different from their fixed-time-effect counterparts, as it is sometimes alleged. There are also practical reasons for choosing models that allow for asymmetric responses to price, especially when evaluating the long-run implications of a number of important energy and environmental issues.