Time series properties of the renewable energy diffusion process: Implications for energy policy design and assessment

Confronted by increasingly tight budgets and a broad range of alternative options, policy makers need empirical methods to evaluate the effectiveness of policies aimed at supporting the diffusion of renewable energy sources (RES). Rigorous empirical studies of renewable energy policy effectiveness have typically relied on panel data models to identify the most effective mechanisms. A common characteristic of some of these studies, which has important econometric implications, is that they assume that the contribution of RES to total electricity generation will be stationary around a mean. This paper reviews such assumptions and rigorously tests the time series properties of the contribution of RES in the energy mix for the presence of a unit root. To that end, we use both individual and panel unit root tests to determine whether the series exhibit non-stationary behavior at the country level as well as for the panel as a whole. The analysis, applied to a panel of 19 OECD countries over the period 1990–2012, provides strong evidence that the time series of the renewable share of electricity output are not stationary in 17 of the 19 countries examined. This finding has important implications for energy policy assessment and energy policy making, which are discussed in the paper.