The European Union Emission Trading Scheme (EU ETS), currently in its fourth phase (2021-2030), is a multinational cap-and-trade system, which was established in 2005 under the Kyoto Protocol to reduce greenhouse gas emissions effectively. The EU ETS “provides an institutional framework for market forces to determine the price of carbon emissions in energy-intensive industrial sectors of Europe”, bolstering hopes that it will be the forerunner to a truly global carbon market (European Commission, 2021).

Broadly, emissions trading seeks to achieve a given emissions target by standardizing marginal abatement costs across firms because profit-maximizing firms will try to reduce emissions if it is cheaper to do so than acquiring allowances on the market (Hintermann, 2016; Montgomery, 1972; Tietenberg, 1985). Thus, the “efficient allowance price is equal to the cost of reducing emissions to one unit below the emissions cap, which is generally referred to as the market’s marginal abatement cost” (Hintermann, 2016). Efficient abatement decisions rely on informative allowance prices that truly reflect the cost of lowering emissions to realize the cap and act as financial instruments that support firms’ profit-maximizing behavior. The role of allowance prices in signaling whether a carbon market can be viable has spurred significant literature on the price composition of EU ETS allowances. However, we still cannot confidently conclude if the allowance price is “right” in terms of reflecting “abatement costs, or whether there is a price wedge caused by uncertainty, transaction costs, and/or price manipulation” (Hintermann, 2016).

Recent years have also seen an exponential growth in activity on microblogging platforms such as Twitter, which has in turn spurred a wave of information mining on these platforms to determine how people are thinking or feeling about a certain topic or product. In this age of rapid information dissemination, short-term sentiment has the potential to not only illustrate the interplay of economic and political influences that are prevalent in the market but also plays a critical part in the short-term performance of financial instruments. Research already suggests that social media excitement is impactful at the microeconomic level, especially in the financial markets (Mao, 2011; Gilbert, 2010; Sprenger, 2010; Zhang, 2009). Rao (2014) shed light on how high frequency traders track the volume of memes in microblogging forums as a proxy for public sentiment while making short term investment decisions. Similarly, Souza (2015) found a statistically significant relationship between volume and Twitter sentiment, and stock returns and volatility, concluding that social media is an important source of information that determines the financial mechanisms of the retail sector, more so than other mainstream news sources.

Thus, this study seeks to build on this prior research about the EU ETS and Twitter sentiment in order to broaden our understanding of the factors that may influence allowance pricing. More specifically, this paper will investigate whether there is a relationship between EUA futures pricing and Twitter sentiment and if that relationship can be employed to forecast EUA futures movements in the short-term. By investigating the relationship between EUA futures and Twitter sentiment, this paper contributes to the existing literature in a couple of ways. Firstly, it expands the limited research on the effect of emissions sentiment in an emission market and will illustrate whether negative and positive aspects of public mood hold the power to track movements of EUA futures in the short-term. It essentially seeks to uncover a potential information asymmetry, which in turn will have implications for traders, policymakers, financial researchers, and financial regulators.

Keywords: carbon permits, sentiment analysis, Twitter, ARIMAX, forecast, emissions market



References

1. European Commission. (2021). Questions and Answers – Emissions Trading — Putting a Price on Carbon. https://ec.europa.eu/commission/presscorner/detail/en/qanda_21_3542

2. Gilbert, E and Karahalios, K. (2010). Widespread worry and the stock market,” Artificial Intelligence, pp. 58–65

3. Hintermann, B., Peterson, S., and Rickels, W. (2016). Price and Market Behavior in Phase II of the EU ETS: A Review of the Literature. Review of environmental economics and policy, The University of Chicago Press, 10, 108–128. https://doi.org/10.1093/reep/rev015.

4. Mao, H., Counts, S., Bollen, J. (2011). Predicting Financial Markets: Comparing Survey, News, Twitter and Search Engine Data,” arXiv.org, Quantitative Finance Papers 1112.1051 https://doi.org/10.48550/arXiv.1112.1051

5. Montgomery, W. D. (1972). Markets in Licenses and Efficient Pollution Control Programs. Journal of Economic Theory 5: 395–418.

6. Souza, T. T. P., Kolchyna, O., Treleaven, P. C., and Aste, T. (2015). Twitter Sentiment Analysis Applied to Finance: A Case Study in the Retail Industry. https://doi.org/10.48550/arXiv.1507.00784

7. Sprenger, T. O., Tumasjan, A., Sandner, P. G., and Welpe, I. M. (2014). Tweets and Trades: the Information Content of Stock Microblogs. European Financial Management: the Journal of the European Financial Management Association, Oxford: Blackwell Publishing Ltd, 20, 926–957. https://doi.org/10.1111/j.1468-036X.2013.12007.x.

8. Tietenberg, T. H. (1985). Emissions trading. An Exercise in Reforming Pollution Policy. Washington, DC: Resources for the future.

9. Zhang, X., Fuehres, H., and Gloor, P. A. (2011), “Predicting Stock Market Indicators Through Twitter ‘I hope it is not as bad as I fear,’” Procedia - Social and Behavioral Sciences, 26, 55–62. https://doi.org/https://doi.org/10.1016/j.sbspro.2011.10.562