Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Analyses of residential electricity demand primarily rely on expenditure or aggregate data. However, newer sources of data, such as that from meter readings, are becoming available. In most circumstances, these newer sources cannot be matched to detailed information about the household. In this research, we make use of South Africa’s Domestic Electrical Load Study, one of the only sources available in a developing country that includes both meter and household level data. Due to some gaps in the meter readings, we focus our attention on average peak electricity consumption, estimating the income elasticity with respect to morning, evening, and the average across both morning and evening peaks. Although we find differences in income responsiveness in the morning, relative to the afternoon, and across quantiles of electricity consumption, these differences tend not to be statistically significant. We do, however, find heterogeneities in those elasticities that can be correlated with, in particular, appliance ownership, suggesting that the ownership of appliances makes electricity more of a necessity, or at least makes the services derived from electricity more necessary for the household.