During the last few decades, oil prices have experienced both sharp increases and, more recently, equally sharp decreases. A number of researchers (e.g. Gately 1993a, 1993b; Hogan 1993; Shealy 1990) have examined the demand responses to these price movements and concluded that price-induced changes in demand have not been symmetric. More specifically, this asymmetry is said to have taken the form of larger demand changes for price increases than for price decreases. To explain this phenomenon, it is typically argued that the oil price rises of the 1970s and early 1980s set in motion behavioural changes that had long-lasting effects, so that when price decreases occurred later in the 1980s, relatively small changes in demand were observed. Shealy (1990), among others, points to the capital stock as playing an important role in this process, but does not explicitly incorporate infornmation on the stock of capital in the model he estimates. Studies that have examined this issue are typically based on as many as three broad levels of aggregation. They tend to aggregate geographically, focusing on entire countries or even groups of countries. Many of these studies also aggregate across sectors of the economy, and thus fail to take into account the different patterns of energy utilization in different regions and/or sectors of the economy. Finally, by looking only at the demand for oil products or at total energy demand, these studies do not explicitly allow for inter-fuel substitution. In contrast, we examine residential energy demand in a single province (Ontario) and allow for inter-fuel substitution by specifying a system of fuel expenditure share equations, which is estimated using annual data for the period 1962-1989. Our focus on the residential sector is predicated on the