The Informational Efficiency of the European Carbon Market

Abstract

This thesis examines the informational efficiency of the European carbon market based on the European Union Emissions Trading Scheme (EU ETS). The issue is approached from three different perspectives. I explore whether the volatility embedded in carbon options is a rational forecast of subsequently realized volatility. Then, I investigate if, and to what extent, new information about the structural and institutional set-up of the market impacts the carbon price dynamics. Lastly, I examine whether the European carbon market is relevant for the firm valuations of covered companies. First, perhaps because the market is new and derivatives’ trading on emission allowances has only started recently, carbon options have not yet been extensively studied. By using data on options traded on the European Climate Exchange, this thesis examines an aspect of market efficiency which has been previously overlooked. Market efficiency suggests that, conditional upon the accuracy of the option pricing model, implied volatility should be an unbiased and efficient forecast of future realized volatility (Campbell et al., 1997). Black (1976) implied volatility and implied volatility estimates directly surveyed from market participants are used in this thesis to study the information content of carbon options. Implied volatility is found to be highly informative and directionally accurate in forecasting future volatility. There is no evidence, however, that volatility embedded in carbon options is an unbiased and efficient forecast of future realized volatility. Instead, historical volatility-based forecasts are shown to contain incremental information to implied volatility, particularly for short-term forecasts. In addition, this thesis finds no evidence that directly surveyed implied volatility estimates perform better as a forecast of future volatility relative to Black’s (1976) estimates. Second, the market sensitivity to announcements about the organizational and institutional set-up of the EU ETS is re-examined. Despite their importance for the carbon price formation, demand-side announcements and announcements about the post-2012 framework have not yet been researched. By examining a very comprehensive and updated dataset of announcements, this thesis adds to the earlier works of Miclaus et al. (2008), Mansanet-Bataller and Pardo (2009) and Lepone et al. (2011). Market participants are found to rationally incorporate new information about the institutional and regulatory framework of the emissions trading scheme into the carbon price dynamics. However, they seem to be unable to accurately assess the implications of inter-temporal banking and borrowing on pricing futures contracts with different maturities. The impact of macroeconomic conditions on the market responsiveness is investigated by splitting the dataset into subsamples according to two alternative methods: 1) a simple split into pre-crisis and full-crisis time periods, and 2) according to a Bai-Perron structural break test. Evidence is found that in the context of economic slowdown and known allowances oversupply, the relationship between the carbon price and its fundamentals (institutional announcements, energy prices and extreme weather) breaks down. These findings are consistent with the arguments in Hintermann (2010), Keppler and Mansanet-Bataller (2010) and Koop and Tole (2011) that carbon price drivers change in response to the differing context of the individual trading periods. Third, the role of carbon performance in firm valuation is understudied. Since companies were not obliged to disclose their carbon emissions prior to the launch of the EU ETS, there exists little empirical evidence of the effect of carbon performance on market value. Earlier studies of the European carbon market have only focused on the impact of ETS compliance on the profitability and competitiveness of covered companies (e.g. Anger and Oberndorfer, 2008). There is also little research on how the newly available emissions data has altered the carbon performance of companies. This thesis addresses these gaps in the literature by examining the stock price reactions of British and German firms on the day of verified emissions release under the EU ETS over the period 2006 – 2011. An event study is conducted using a Seemingly Unrelated Regressions model to deal with the event clustering present in the dataset. Limited evidence is found that investors use information about the carbon performance of companies in their valuations. The information contained in the carbon emissions reports is shown to be somewhat more important for companies with high carbon-intensive operations. This thesis finds no conclusive evidence that the cap-and-trade programme has been able to provide regulated companies with enough incentives to de-carbonize their operations. The market does not punish companies which continue to emit carbon at increasing rates or reward companies which improve their carbon performance. In brief, the results of the thesis suggest that the market is not fully efficient yet. Inefficiently priced carbon options may allow for arbitrage trades in the market. The inability of investors to incorporate rules on inter-temporal banking and borrowing of allowances across the different trading periods leads to significant price reactions when there should be none. A recessionary economic environment and a known oversupply of emission allowances have led to a disconnect between the carbon price and its fundamental drivers. And, lastly, the signal embedded in the carbon price is not strong enough to invoke investor action and turn carbon performance into a standard component of investment analysis.