Variance risk-premia in CO2 markets

This paper proposes a new methodology to measure the volatility of CO2 assets computed as the difference between model-free implied volatility (from option prices) and model-free realized volatility (from high-frequency intraday data), coined as ‘variance risk-premia’ (Carr and Wu, 2009; Bollerslev et al., 2009; Trolle and Schwartz, 2010), during 2008–2011. We find that variance risk-premia are equal to a daily sample average of 0.79 for European Union Allowances and 0.18 for Certified Emissions Reductions. In the spirit of the CAPM, we show that the beta can only explain a small portion, and that macro risk factors specific to CO2 markets and energy volatilities can improve this result. Hence, there exists a systematic variance risk factor in CO2 markets that asks for a highly risk premium. Further analysis shows that variance risk-premia are time-varying, and can be used as strong predictors for forecasting CO2 returns.