Empirical modelling of environmental risks

Compared to the banking sector, the insurance industry has shown more interest in incorporating environmental risk assessment to their financial analysis since it has experienced the negative impact of natural events in its financial balances. With the enforcement of environmental regulation credit and investment banks are paying more attention to the risks and opportunities of environment-related aspects. In this doctoral thesis the return distribution of several environmental sensitive securities with different environmental performance of the chemical, oil and alternative energy sectors together with the Dow Jones Sustainability Index have been analyzed in order to assess the financial impact of environment-induced risks. Four aspects have been focused: the modelling of the statistical characteristics of environmental sensitive assets based on stable distributions, the dynamic of environmental risks by means of ARMA-GARCH models, the systematic or unsystematic character of environmental risks and, at the portfolio level, the risk-return structure of environmentally sound funds. Results of this research suggest that the risk profile of polluting firms is extreme leptokurtic in comparison to firms with a higher environmental performance. This characteristic may have important implications for risk management since the risk of securities that exhibit extreme kurtosis is more difficult to diversify. For environmental asset management, screening procedures based on stable distributions show a more realistic risk-return structure and therefore lead to better financial results.