A new methodological perspective on the impact of energy consumption on economic growth: time series evidence based on the Fourier approximation for solar energy in the USA

Abstract

From the empirical energy literature, it is observed that studies focusing on the energy-economic growth nexus ignore the possible existence of gradual breaks as they employ methods without or with sharp structural breaks. Therefore, one can argue that they may yield biased and inefficient output in the presence of gradual breaks. The goal of this paper is to investigate the impact of solar energy consumption on GDP utilizing quarterly data over the period 1984-2018 for the USA. For this purpose, the paper performs a unit root test and a cointegration test that are based on the Fourier approximation to take gradual breaks into account. The paper also performs the dynamic ordinary least squares estimator to estimate long-run parameters. The findings document that there exists cointegration in the empirical model and that GDP is positively associated with solar energy consumption. Some implications based on the empirical findings are presented in the paper.