Evidence on the stationarity of electricity consumption in India

This research reevaluates the stationarity of sectoral electricity consumption in a panel of 18 Indian states. We use a novel panel unit root test with an allowance for structural breaks and cross-sectional dependence. The usual univariate unit root test, which accounts for structural breaks but not cross-dependence, rejects the null hypothesis of a unit root in all states under investigation. When we incorporate structural breaks and cross-sectional dependence, the null hypothesis's rejections significantly decrease. Electricity consumption appears nonstationary in many states across all sectors, suggesting that policy changes can have long-term effects.