Modelling Financial Returns and Volatility Across Environmental Industry Sectors

In recent decades, the momentum of global environmental protection has culminated in the Kyoto Agreement of 1998, placing the limelight on “green” issues. This paper argues that the protection of environmental systems involves a fragile balance between the costs of environment preservation and the profit motivations of industrialists. In particular, one of the issues that needs to be addressed is the risk pressures environmental industries face in financial markets, where the higher the risk, the more pressure industries are under to exploit natural resources. Therefore, in order to devise effective environmentally-friendly yet economically viable policies, it is crucial to analyse the risks encountered by environmental industries in financial markets. The success of the autoregressive conditional heteroskedasticity (ARCH) or generalised ARCH (GARCH) models in explaining the stylised facts of financial asset returns has led to its widespread use in the empirical finance literature. By modelling the time-variation in conditional variances or volatility, the univariate ARCH model by Engle (1982) and the GARCH model by Bollerslev (1986) are able to capture the stylized features of the persistence of volatility, volatility clusters and kurtosis, while extensions of the GARCH model such as the asymmetric GARCH (GJR) model by Glosten, Jagannathan and Runkle (1993) can accommodate the additional stylized fact that positive and negative shocks have asymmetric effects, whereby a negative shock has a greater impact on volatility than a positive shock. This paper models the time-varying conditional variances of the returns on a variety of environmental industry sectors using the univariate ARMA(1,1)-GARCH(1,1) and the ARMA(1,1)-GJR (1,1) models. Our dataset consists of daily returns on seven Australian environmental industry sectors including Gold Mining, Other Mining, Mining Finance, Oil & Gas, Farming & Fishing, Forestry and Paper over their respective time periods. The findings of this paper suggest that the risks faced by environmental industries in financial markets are generally well-explained by the ARMA(1,1)-GARCH(1,1); the ARMA(1,1)GJR(1,1), on the other hand, received much less support due to the lack of asymmetric effects. The log-moment and second moment conditions were also satisfied empirically, implying that moments exist and the QMLE are both consistent and asymptotically normal. Therefore, inferences of the ARMA(1,1)-GARCH(1,1) estimates can be used to aid in formulating new “green” and economically viable environmental policies.