The Volatility Structure Between Aluminum and Ethanol Prices, and the Stock Returns of Automotive Companies by Copulas

Abstract

Aluminum and ethanol usages have been accumulating over time under a fluctuated crude oil price and compulsory environmental regulations. The automotive sector is related to these inputs both upstream and downstream in order to meet fuel efficiency standard and lower GHG emission which this sector statistically consumed 64.5% of global oil and emit 16% of CO2 on road transportation in 2014 (IEA and OICA). Both commodities are claimed to be alternative to steel and crude oil, however also additive. Aluminum is substituted to steel to make lighter weight vehicles and save CO2 emissions unless some parts have been limited to steel and aluminum price is five times expensive. While E100 runs on very limited model, mostly it is mandated to minimum volume mixed into gasoline and use in flex-fuel vehicles. The weekly returns from November 22, 2010 until January 30, 2017 of aluminum, ethanol, steel, and 6 auto companies’ shares had shown to be skewed and excess kurtosis, and to exhibit an asymmetric or tail dependence structure. Therefore, this study is analysed using the copula-based GARCH model based on Wu, Chung, and Chang (2012) to fill the gap between energy - ethanol, metals - aluminum and steel, and its biggest-related industry – automotive companies, in term of the conditional volatility and dependence structures between commodities’ returns and equity returns. The results shows that the past volatility positively affect to current volatility of all returns, however, only weekly returns of aluminum, steel, General Motors, Ford, Volkswagen, and BMW are positively related to its own past residuals; others has no relationship with its own residuals. While the leverage effects are significant only for Ford and BMW returns, while others is insignificant which means negative returns have a greater effect on future volatility. The dependence structure between aluminum and 6 automotive companies’ returns is positive, while dependence structure of ethanol and 6 automotive companies’ returns can be grouped into 3 patterns, and dependence structure between steel and 6 automotive companies’ returns is negative, even though, dependence structure between steel and Honda returns is negative, but it become positive in curtain period. The estimation and analysis of the conditional volatility and dependence structure between aluminum, ethanol, steel, and 6 auto companies’ returns can provide useful information for investors and stakeholders that are concerned with the aluminum, ethanol, steel, and automotive markets.