Exploring Renewable Energy Adoption in the OECD: An Observational Panel Data Analysis of the Energy Mix Dynamics

Abstract

This thesis aims to develop understanding of the determinants of the energy transition by employing panel data analysis of 37 OECD countries over the time period 1990 to 2020. To control for unobserved heterogeneity, we utilize a fixed effects model in analysis of this dataset. To provide a robust analysis, we correct for identified violations of the OLS assumptions. Furthermore, we provide a critical review of how a difference-in-differences study design would be useful in our context, and how it would be infeasible with our data given the underlying assumptions. We challenge the existing literature by applying the renewable energy share of the total energy supply as our dependent variable. Unlike in the existing literature, which tend to deploy renewable consumption in isolation from the total energy mix, we do not find a link between GDP per capita and the energy mix development. As a step to achieve stationary variables, we perform growth transformation of all continuous, non-dummy variables. We find the share of renewable energy to have a significant and positive relationship to EU Emissions Trading System membership, Brent Crude oil price, population, and renewable energy per capita. The EU Emissions Trading System sees the greatest magnitude in our model, and aligns with presented theory about international climate agreements. The Brent Crude oil price’s positive and significant relationship with the dependent variable also suggest that the substitution theory holds true for the energy markets in our sample. A negative and significant relationship is found between the dependent variable and the total energy supply, as well as industrial activity levels. No significance is found in the volatility of the Brent Crude oil price.