Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Earlier literature often overlooked monthly data's detailed insights into electricity price changes' effects on demand and the impact of economic shocks like recessions on electricity consumption. We conducted static and dynamic panel analyses of state-level data since 2001, using various techniques to address endogeneity and serial correlation. Our analysis incorporates diverse techniques such as linear regressions, Two-Stage Least Squares, Driscoll-Kraay standard errors, and quantile regression for static models, while dynamic models employ the two-step method, forward-orthogonal deviations, deviations from within-group means-GMM method, and the method of moments quantile regression to address potential endogeneity, serial correlation, and cross-sectional dependence. After conducting over a thousand regressions, our major findings are as follows; first, we find that our average monthly long-run (short-run) elasticity of −0.84 (−0.68) is about five (eight) times higher than our average annual elasticity of −0.15 (−0.09). Second, our recession variable is statistically significant for most of our monthly specifications, indicating that residential electricity consumption decreases during economic downturns. Third, using a between estimator yields similar static elasticity results between the data frequencies. Our findings underscore the importance of selecting both methodological approaches and data types, while illustrating the intricate relationship between electricity consumption, pricing elasticity, and economic fluctuations.