European Union Emission Trading Scheme: The effect of the European Union Allowances on the stock market performance and the performance of individual firms in Europe.
Paul Hemmingson Student ID:11730641 Supervisor: A. Kaakeh
Bachelor Thesis University of Amsterdam
Faculty of Economics and Business – Finance Amsterdam, June 30, 2020
Abstract: This paper investigates the market effects of the European Union Emission
Trading Scheme (EU ETS) during its third phase. A multifactor market model has been used to measure the effect the European Union Allowances (EUA) price has on the performance of STOXX 600 Utility Index, as well as the effect on corporate value of firms within the energy and aircraft sector. The results suggest that there is an insignificant positive correlation between the EUA and the STOXX EUROPE 600 Utility Index. Furthermore, the results indicate that the EUA has an insignificant positive effect on firms within the aircraft sector, whereas it is insignificant negative for firms within the energy sector. The effect of the EUA price changes on corporate value is sector and firm specific. The conclusion that has been reached is that the effect resulted from the sector specific EUA allocation policy.
Statement of Originality
This document is written by Student Paul Hemmingson who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
1. Introduction 05 2. Literature Review 10 3. Hypothesis 13 4. Data 15 4.1 Energy Sector 16 4.2 Aircraft Sector 17 5. Methodology 18 5.1 Energy Sector 20 5.2 Aircraft Sector 20 6. Results 21
6.1 Correlation between the Euro Stoxx and EUA 21
6.2 Energy Sector 23
6.3 Aircraft Sector 25
6.4 Comparison between Sectors 27
7. Discussion 28
8. Conclusion 30
9. Bibliography 32
1.Introduction
Sustainability receives a high degree of attention in the literature. Often it is
associated with the environment, but sustainability takes on an important role in corporations as well. When in the 1970s the first environmental regulations were enforced, debates
occurred on how they affect businesses (Makridou, Doumpos, & Galariotis, 2016).
According to Makridou et al. (2016), the regulations can have negative effects on corporate value, but on the other can also increase the economic growth of a firm. Nowadays,
stakeholders (shareholders) are, at an increasing rate interested in investing in companies that have well defined sustainability goals. These goals do not only benefit the planet, but
environmental performance simultaneously improves financial performance (Albertini, 2013). According to Busch and Hoffmann (2007) companies are trying to follow the Kyoto Protocol guideline and reduce their effect on climate change and are trying to minimize their ecological footprint. A great example of that Busch and Hoffman (2007) idea is the
Norwegian Pension fund. One of the largest pension funds in the world which manages around 1.1 trillion USD$1 sold their shares of RWE but also Sasol, Glencore, AGL Energy and Angelo America because they use and produce coal2.
Since the industrial revolution which started in the late 1700s, carbon dioxide (CO2) emissions are constantly increasing, leading to environmental changes affecting everyone already today and in the long-term. Therefore, it was necessary to implicate changes to improve the situation. The Kyoto protocol was adopted during the Framework Convention on Climate Change in Kyoto 1997 and was the first milestone to include binding commitments for developed countries to limit and reduce their emissions3. To ratify the protocol at least 55 countries that account for at least 55% of the total CO2 emissions in 1990 had to sign it, which took until 20054. Each country had to commit itself to reduce its emissions such as Carbon dioxide (CO2), methane (CH4), and laughing gas (N2O). There are three common ways to decrease air pollution (1) command-and-control (2) pollution tax (3) emissions trading according to Stuhlmacher, Patnaik, Streletskiy, and Taylor (2019). The European
1 https://www.institutionalinvestor.com/article/b1kjglcy66gpz1/Norway-s-Massive-Pension-Fund-Returns-19-9-Percent
2 https://www.faz.net/aktuell/finanzen/norwegischer-staatsfonds-veraeussert-rwe-anteile-16767746.html 3 https://www.bmu.de/en/topics/climate-energy/climate/international-climate-policy/kyoto-protocol/ 4 https://www.bmu.de/en/topics/climate-energy/climate/international-climate-policy/kyoto-protocol/
Union choose the emission trading approach to reduce its emission, a reason could be that it is most economically efficient and gives the most flexibility (Stuhlmacher et al., 2019).
By directive the idea of the European Union Emissions Trading Scheme (EU ETS) was established in 2003, to start its first phase in 2005 (Zetterberg, Nilssom, Åhman, Kumlin, & Birgersdotter, 2014). It was the first of its kind and is the largest emission trading scheme, that follows a ‘cap and trade’ approach. Giving out a limited number of allowances the EU can steer and reduce its overall emissions each year, but now the EUAs can also be valued, making it the medium of exchange within the EU ETS (Makridou et al., 2019). By putting a price on the emission, the scheme has a direct effect on the investment and production decisions of the included firms (Löschel, Lutz, & Managi, 2019).
According to Stuhlmacher et al. (2019) an advantage of the cap-and-trade is the increased flexibility regarding compliance, on the other hand, this could lead to regional clustering of pollution if it is cheaper to buy allowances than to reduce the emissions. The EU ETS includes all 28 EU member states as well as Island, Liechtenstein, and Norway, who joined in 2008 during phase two. Two phases have been completed, since the start in 2005. The scheme is now in its third phase, which will end in 2020.
Each phase of the EU ETS is divided into three steps, first, an annual cap on the overall emissions is set, second, the total allowances within the set cap are allocated and auctioned off to the emitters, and finally, each installation chooses how much to emit
(Oestreich, & Tsiakas, 2015). The allowances are called European Union Allowances (EUA). Participants either get them for free or they have to buy them. Each owned allowance gives the individual firm the right to emit one tonne of CO2. At the end of the year, each company has to have enough EUAs to cover its emissions, if they don’t, they will face significant fines. On the other hand, however, if a company has more EUAs than needed, by reducing its total emission, it can keep the allowance to cover future needs or trade them to another company5. All large energy-intensive firms are obliged to take part in the EU ETS, however, there are some sectors in which only firms over a certain size had to participate (Makridou et al., 2019).
Phase one started in 2005 and was set up as a trial phase. It lasted until the end of 2007 and only covered CO2 emissions from two different industrial sectors, power generators and the energy-intensive industries6. At first almost all of the allowances were given out for
5 https://ec.europa.eu/clima/policies/ets_en
free, in case a company did not have enough allowances to cover all its emissions at the end of the year, a fine of 40 Euros per tonne of CO2 had to be payed7. The EU ETS was a decentralized trading scheme. Each country had to set their cap level, allocate the permits, monitor, report and verify their emissions (Kruger, Oates, & Pizer, 2007). Each member state had to publish their National Allocation Plan (NAP) in 2004, which had to be in line with the directive and had to be approved by the European Commission (Zetterberg et al., 2014). However, since there wasn’t a centralized cap set by the EU, non-state and market actors started to influence the regulatory mechanisms, through lobbing for example (Stuhlmacher et al., 2019).
The limited time to submit the NAP, increased the hurdle to obtaining the right data for the allocation of the allowances. This data problem existed in all member states and regardless of how the data was collected in the past. Countries such as Sweden and Germany who had a lot of energy and environmental regulations in the past faced similar problems as countries that had less strict rules before such as Spain or Italy (Buchner, Carraro, &
Ellerman, 2011). Some countries had collected data that was covered by the EU ETS, but discrepancies between earlier data and the newly submitted data differed by up to 20% (Buchner et al., 2011). Denmark was the only exception, they had developed a CO2 trading scheme before, which covered similar emissions as the EU ETS. The data problem can be traced back to, too little legal authority to collect the relevant data, as well as the tight
deadline to submit the NAP (Buchner et al., 2011). Due to the time pressure governments had to rely on voluntary submission, while trying to receive the necessary legal authority. The missing data enabled some member states to choose benchmarks on emissions from 1998, in the case of Sweden, Denmark, and the UK (Buchner et al.,2011). Therefore, during phase one reliable emission data wasn’t present. The cap had to be estimated, leading to an oversupply of allowances, dropping the price to zero. However, phase one was still successful because a price for carbon and its trade was established and most importantly the infrastructure to monitor, report, and verify the emission was set up. During phase one the scheme limited 11,000 heavy energy-using installations (plans) for firms within the energy sector and other carbon intensive industry such as coke ovens, iron and steel, paper, refining, roasting and sintering, brick and ceramics, cement and lime and glass production (Stuhlmacher et al., 2019). It was thereby covering 45% of Europe’s total CO2 emission and about 30 % of its greenhouse gas emissions (Betz, Rogge, & Schleich, 2011).
In 2008 the second phase started which ended in 2012. A major change compared to phase one was that the total cap was cut by 6.5% at the beginning of phase two8, as well as the fee for not having enough allowance increased to 100 Euros per tonne (Stuhlmacher et al., 2019). During that phase three countries joined the scheme, Iceland, Liechtenstein, and Norway, as well as the inclusion of the aviation sector. Furthermore, the free allocation dropped down from 95% to 90% and therefore the share of allowance that had to be auctioned increased by 5% (Mo, Zhu, & Fan, 2012).
In phase two the cap reducing was based on actual emission data. During the world financial crisis in 2008 the total emission was less than expected. The total amount of exceeded the expected amount, leading to a surplus of allowances and credits and simultaneously a decrease the EUA price9.Credits are divided into two categories Clean Development Mechanism (CDM) and Joint implementation (JI), both represent a removed or reduced tonne of CO2 from the atmosphere outside of the company itself, with a similar effect as the EUAs10.
In 2013 the third phase started, 40% of the allowances were auctioned11 and the aircraft sector was included in the scheme. Since the NAP12 was too complex the EU decided to have an overall cap, single EU cap, starting in phase three (Ellerman, & Patnaik,2019), with around 2 billion allowances issued, decreasing the total amount by 1,74% each year13. In the current period, the EU aimed to auction of 57% of the EUA to companies by the end of 2020, the rest will be allocated free of charge. The manufacturing industry receives around 30% of its allowances for free, the aircraft industry mostly receives all of their allowances for free during this phase, whereas the power generators do not receive any certificates for free since 201314.
Until now the drivers of the EU ETS are still not fully understood, generally, the price fluctuations are explained by the unstable supply and demand, that advanced as a result of the financial crisis in 2009 (Flachsland, Pahle, Burtraw, Edenhofer, Elkerbout, & Fischer, 2020).
8 https://ec.europa.eu/clima/policies/ets/pre2013_en 9 https://ec.europa.eu/clima/policies/ets/pre2013_en 10 https://ec.europa.eu/clima/policies/ets/credits_en 11 https://ec.europa.eu/clima/policies/ets/auctioning_en 12 https://ec.europa.eu/clima/policies/ets_en 13 https://ec.europa.eu/clima/policies/ets/cap_en 14 https://ec.europa.eu/clima/policies/ets/allowances_en
Fig. 1. The price of EUA futures from 2010 until 2019.
Löschel et al. (2019) found that the scheme could lead to a competitive disadvantage for firms that are part of the EU ETS compared to ones outside the EU, especially from the manufacturing sector which sells their goods and services on a global scale.
Mo et al. (2012) found that the difference between phase one and phase two is the fact that the EUAs became scare, thereby effecting corporate value. Veith, Werner, and
Zimmermann (2009) concluded in their research that in addition to the energy sector, other sectors will need to be studied to understand if the results are generalisable. Therefore, this paper is going to analyse how different sectors are effect by the EU ETS during phase three, by answering the following research question:
Is there a positive (negative) correlation between the STOXX EUROPE 600 Utility Index return and the increases (decrease) of the EUA price, with a special focus on how the free allocation of EUAs affect firms within the energy- and aircraft sector?
To analyse the effects of the EUAs on the firms from the energy and aircraft sector a multifactor regression model will be used. This research paper will contribute to the existing
literature by being one of the first to analyse the effect the EUA price has on the energy sector in comparison to the aircraft sector during phase three of the EU ETS.
The remainder of this thesis is organised into 7 further sections. The first section will analyse the current literature on the relationship between carbon emissions and stock market performance. The four hypotheses will be formulated and discussed in section 2, section 3 will introduce the analysed data and variables. Section 4 will outline the methodology, which will be followed by section 5, which will presents the results of this thesis. Section 6 will discuss the results and section 7 will conclude.
2. Literature Review
The following section will discuss the current literature that analysed the impact the EU ETS had on the European stock markets and different sectors.
There has been a lot of research on the EU ETS that analyses it from different perspectives (Hoffmann, 2007). Hoffmann (2007) found that the research has a strong focus on expected effects rather than conducting empirical research to understand the actual effect, which he explained with the time of the research. Friedrich, Mauer, Pahle, and Tietjen (2020) argued that there is still too little understanding of the drives of the EUA prices.
Generally, Hoffmann (2007) categorised the research done on EU ETS into four different groups, namely, how the scheme affects the incentive to invest into technology innovation. The second category analyses the effect on the operating and investment
decisions, that changed due to the increasing cost for different technologies. The third set of studies that analyses the attitude of companies towards the scheme. Finally, there is a set of studies that analyses the effect on current and future performance of the scheme, and how well the goal, the reduction of greenhouse emissions, is achieved. An accompanying topic in the discussion on the attitude of the companies is the way on how to decrease emissions, either by investing in new technologies or by buying extra allowances, with either way resulting in additional costs (Montgomery, 1972).
A lot of research has been conducted that tries to explain the relationship between CO2 emissions and the financial performance of companies, however, it is still very limited (Tian, Akimov, Roca, & Wong, 2016; Anger & Oberndorfer, 2007). Mo et al. (2012) found that there are a lot of different views within the existing literature when it comes to the effect of EU ETS on stock performance. They concluded that there is a wider spectrum of thoughts,
Bode (2006) who associates it with increasing opportunity costs and Oberndorfer, Rennings, and Sahin (2006) who found that the effect is less significant.
One of the first econometric analysis of how EU ETS affects the stock market was done by Oberndorfer (2009). Their research focused on the first phase of the EU ETS from 2005 until 2007, during that phase certificates were given out for free, in which they analysed the impact of EUAs on 12 European companies within the energy sector. Their research focused on country and time-specific effects on the stock returns, which was analysed with a panel approach. Furthermore, the generalized autoregressive conditional heteroskedasticity (GARCH) model was used as well as the least squares sequence (OLS) regression. They found that the EUAs and the stock returns of the energy companies are positively related, when the EUA increases in value the value of the individual firm increases and vice versa. Moreover, they found that the correlation is time and country specific such as that firms from Spain show a negative relationship whereas firms from Germany and the UK show a positive relationship. In theory, however, according to Busch and Hoffmann (2007), there should be a negative correlation between EUA and the stock prices due to additional volatility to the cash flows and therefore the increased chance of making a loss.
Veith et al. (2009) also covered the first phase of the EU ETS, they included 22 European power companies, that generated two third of the electricity within the EU. They used a multifactor model to find that the stock returns and the rising of EUA prices are positively correlated, however, it is less significant when it comes to companies with a high proportion of fossil electricity generation. On the other hand, Jong, Couwenberg, and Woerdman (2014) studied 2167 firms via an event study. They analysed the impact the introduction of the allowances had on the share price during the first phase in 2006. They found that there is an asymmetric association between stock prices and carbon certificates (EUA). Furthermore, they also found that having more allowances compared to how much CO2 they emit, has a positive effect on their competitiveness, which is in line with the results of Makridou et al. (2019).
Mo et al. (2012) focused on the first and second phases of the EU ETS and used a modified multifactor model, to adjust for thin trading especially during the first phase, to analyse 12 European electricity firms. They discovered that there is indeed a relationship between stock prices and the EUA, in phase one the EUA increased the value of the
individual company, whereas in phase two the EUA decreased the value. The change in the relationship from phase one to phase two can be explained by two factors according to Mo et al. (2012). One, the EUA allocation plan was adjusted and two, the regulations also got more
rigorous. A similar effect was found by Oestreich and Tsiakas (2015) after they conducted an empirical study on the stock returns of 65 German companies from 2003 until 2012,
including the first two phases of the EU ETS. Firms that received free CO2 certificates outperformed the ones that did not receive free allowances, resulting in a higher stock price.
Makridou et al. (2019) and Dutta, Bouri, and Noor (2018) are one of the first to include phase three in their research. Makridou et al. (2019) used a multilevel cross-classified model to analyse the profitability of 3,952 firms from the major energy intensive sectors (takes up to 80% of total energy consumption), who are part of the EU ETS from 2006 to 2014. They found that the amount of emissions and the amount of allowances has a positive effect on the profitability of the firm. Dutta et al. (2018) focused their research on the
analysis of firms producing clean energy from the clean energy European Renewable Energy TR Index (ERIX) and how they are affected by the EUA prices. A bivariate
vector-autoregressive (VAR) GARCH model was used for their analyses including data from phase two (2009) to phase three (2017). They found that there is a insignificant positively
correlation between the allowances and the stock price. However, this correlation is country- or region-specific, similar as Oberndorfer (2009).
Other literature is suggesting that the correlation is dependent on the sector of the firm. Bushnell, Chong, and Mansur (2013) did an event study and analysed 552 stock from the STOXX EUROPE 600 Utility Index (Euro Stoxx) in April 2006. They found that when prices for EUAs dropped, firms within the carbon and electricity industries faced a significant decrease in stock price, concluding that the degree of correlation between EUA and ETS depends on the industry. This outcome is in line with the research of Tian et al. (2016) who noted that carbon intensive firms are negatively affected by the EUA price development, and vice versa for firms that are not carbon intensive.
There has been a lot of research made on the EU ETS effects on how individual companies’ stock performs, with a strong focus on the energy sector. The results in previous literature are mostly similar. The most common findings are that there is an effect which depends on certain criteria. In previous research different samples as well as different time frames were studied, making it difficult to come to an overall conclusion. There is very little research on how different sectors are affected by EUA prices (Veith et al., 2009). Against this background, this phenomenon needs to be further empirically investigated.
In this section, the hypotheses will be introduced. Previous research has shown that the EU ETS impacts the financial markets and consequently has economic effects on the firms involved (Oberndorfer, 2009). Oberndorfer (2009) stated, that firms that are fully grandfathered – firms that are getting the allowances for free based on historical values - will make a profit. By the end of phase three more than half of the allowances will be auctioned off (EU link) of the overall firms less will be grandfathered and therefore they will have to buy additional EUAs. The increasing overall costs will decrease the net profit and decrease the stock value of the corporation. Finally, Bushnell et al. (2013) found that the more CO2 a firm emits the worst its stocks performs.
Hypothesis 1: The movement of the EU Emissions Allowances prices and the performance of the STOXX EUROPE 600 Utility Index is negatively correlated.
Compared to phase one, an increasing part of the allowances are auctioned off in phase three. In 2013, 40 % of the EUAs were auctioned off, between 2013 and 2020 the amount was estimated to increase to 57%. Benz and Trück (2008) found that the change in the EUA price has a direct effect on the value of the allowance held by a firm. Moreover, the price for each allowance directly influences the held cash flow, therefore directly affecting the overall value of the firm. This effect found is country and time specific due to country specific NAP (Oberndorfer,2009)
Since previous research is mainly based on phases one and two, during which the EU ETS was decentralized, the found results could be based on the country specific NAP. This should not be the case because only phase three is included, in which an overall single, EU-wide cap was established. Finally, as more sectors are included in the scheme and as fewer allowances are given out for free, the more trading is occurring. This increased trading behaviour exposes the firm value to price changes in the allowance’s prices (Mo et al., 2012). Busch and Hoffmann (2007) conclude that in theory, there should be a negative correlation between EUA and the stock prices due to additional volatility to the cash flows and therefore the increased chance of making a loss.
Hypothesis 2: The relationship between the EUA price and the stock performance of companies within the energy sector is negatively correlated.
Oberndorfer (2009) states that firm profits are depended on the intensity of the use of carbon and consequently the emission of carbon dioxide, and thereby directly effecting the firm’s value and its stock price. Further, he stated that under full grandfathering, firms within the energy sector would make profit. However, since they are not grandfathered anymore in phase three the opposite could be assumed. Oberndorfer (2009) concluded that in theory the stock price and the EUA should be negatively correlated because of the increase in volatility to the cash flow, which could lead to a loss. With the start of phase three in 2013, the
allowances have to be fully purchased by the energy sector15 , thereby directly affecting their cash flows. This assumption is the opposite of what Bushnell et al. (2013) assumed since, during that phase, allowances were given out for free. On the other hand, the increase of the EUA price will be reflected in the electricity prices. Consequently, if the price for an
allowance increases, the extra cost can be pushed on to the final customer (Oberndorfer, 2009). Finally, increased participation in the allowance trading towards the end of phase two increased the exposure of the corporate value to EUA price changes (Mo et al., 2012).
Hypothesis 3: The relationship between the EUA price and the stock performance of companies within the aircraft sector is positively correlated.
The aircraft sector still receives more than 80% of the EUAs for free during the third phase. Therefore, it is a similar setting as when Oestreich and Tsiakas (2015) conducted their first analysis of the energy sector during phase one. The profit of a firm depends on its carbon intensity. It, therefore, simultaneously effects its stock value. As the EUA prices increase, the future cash flows of the firm increases as well. Oestreich at al. (2015) found that firms, who received free emission allowance performed better than firms who did not receive them. The free allocation provides another source of revenue for the aircraft sector, which gives them an advantage over the energy sector (Bushnell et al., 2013). On the other hand, even though the aircraft sector receives most of their allowances for free, they might need more allowances and therefore have to engage in trading, whereby they expose themselves towards the EUA price changes (Mo et al., 2012). Finally, as most previous research focused on the first two phases and found that the free allocation has a positive impact on the revenue Oestreich et al. (2015), a similar impact on the aircraft sector can be assumed.
Hypothesis 4: An increase in the EUA price positively affects the aircraft sector, whereas the energy sector is negatively affected by it.
Bushnell et al. (2013) and Oberndorfer (2009) analysed German companies during phase one and two and found that the ones that receive free allowances outperformed those that did not. Since almost all allowances within the aircraft sector are given out for free, they should outperform the firms within the energy sector. Because firms within the energy sector have to buy all of their emission allowances since 2013. As previously mentioned, these price increases directly affect the future cash flows of a firm and therefore their stock value.
4.Data
In this section the data will be introduced that was necessary to conduct the research. The analysis covers the third period of the European Union Emission Trading Scheme, with sufficient data from January 2, 2013 until April 30, 2020. I excluded phase one and two because extensive research has been made already on these two phases and research one phase three is limited (Meleo, Nava, & Pozzi, 2016). Furthermore, as Mo et al. (2012) found, there has not been enough reliable emissions data in the first phase, which led to price for EUAs to drop down close to zero, which they adjusted for with lead and lag terms. In
addition, in phase two the prices for EUAS dropped down to zero again because of the world-wide financial and economic crisis in 200816 (Dutta et al., 2018). Makridou et al. (2019) and Dutta et al. (2018) analysed phase three of the EU ETS, this paper will extend this research and thereby contribute to existing literature.
The Euro Stoxx is used in this analysis and was used to calculate the price return (Mo et al., 2012; Dutta et al.,2016). The Euro Stoxx was used as a dependent variable and was retrieved from the Database Thomson Reuters Eikon (TRE). The EUA future price was used, which was collected from TRE, which was used as the independent variable. Originally, the daily stock return was used, however, the data was incomplete, therefore, the monthly prices were used to calculate the changes of the EUA future price.
The S&P 500 and the MSCI Index were also retrieved from TRE. The index prices were used to calculate the market return and were then used as a benchmark as well as control variables to represent the economy. Furthermore, the iBoxx EUR Liquid Sovereigns
Diversified 25 Index17 (ETF) is included, it tracks the 25 most liquid Euro denominated government bonds within the eurozone, which was retrieved from TRE.
4.1 Energy Sector
In the second part of the analysis the effect of the EUAs prices on the stock price of each sample company within the energy sector will be analysed. In line with Mo et al. (2012) I used the same electricity firms as them, which are included in the Euro Stoxx. By using a similar data as well as similar model as Mo et al. (2012), I can compare my results to the prior findings (Schmale, 2014). The companies that were used A2A (Italy - IT), Drax Group (United Kingdom - UK), Électricité de France (France - FR), ENDESA (Spain - ES), ENEL (IT), Fortum (Finland - FI), IBERDROLA (ES), Public Power Corporation (Greece - GR), Red Eléctrica (ES), Scottish & Southern Energy (UK) and TERNA (IT). In addition, I added E. ON (Germany - GE) and RWE (GE) to represent the German market (Oberndorfer, 2009). International Power (UK) was taken out due to incomplete data, and its stock price was close to zero. The monthly stock prices for each company were gathered from TRE, to calculate the percentage change in stock prices.
The data for the EUA future price was retrieved from the European Energy Exchange (EEX) in Leipzig, which is free of charge for research purposes. The price was used as an independent variable (Oberndorfer, 2009). The Euro Stoxx is used in this analysis and was used to calculate the market return. The data was also retrieved from TRE. In this analysis the market return, the Euro Stoxx, is used as a control variable, which helps to reflect market movements such as financial crisis (Mo et al., 2012).
According to Oberndorfer (2009), the EEX is one of the biggest EUA marketplace next to Nord Pool, European Climate Exchange and Powernext. Since the prices for the allowances (EUAs) have developed in a similar way in all marketplaces (Oberndorfer, 2009) the choice of which one to use is not essential. The German market is the biggest electricity market in the EU, therefore, in accordance with previous literature, the German electricity prices, called Phelix, was used as a proxy for the price development (Oberndorfer, 2009). The monthly German electricity futures prices (Phelix) were retrieved from TRE and were used to calculate the price change. As mentioned by Mo et al. (2012), the electricity price effects the stock return of a company within the energy sector.
In conformance with Oberndorfer (2009), and Mo et al. (2012) the changes in oil prices were included because it is one of the main indicators for energy price development. Dutta et al. (2018) argued that the EUA prices are affected by the economic activity and energy fuel prices such as oil. The Europe Brent Spot Price (FOB) was used to calculate the percentage change and were retrieved TRE.
4.2 Aircraft Sector
For the aircraft sector, ten firms of the STOXX Europe Total Market Airlines Index (link), were used, since their primary source of revenue is air transportation. Enter Air
(Poland - PL) and Wizz Air (Hungary – HUN) were taken out due to incomplete data and the International Airline Group (IAG) was added. Therefore, the following nine aircraft
companies were used for this research, Ryanair (IE), International Airline Group (GB), Lufthansa (GE), EasyJet (GB), Aegean Airlines (GR), Air France-KLM (FR), SAS (Sweden - SE), Finnair (FI), Norwegian Air Shuttle (Norway - NO). Moreover, the same variables will be used as in the previous sector analysis. The Phelix variable was left out because it does not represent airline prices, nor does it affect the airline industry. The changes in oil prices were included because they have an effect on the performance of corporations within the aircraft sector. Unfortunately, there is no proxy such as the electricity prices (Phelix) for the energy sector that can be used for the aircraft industry. The airline industry was included in the EU ETS in 2012 and still receives most of the allowances for free, therefore the companies do not incur extra cost from buying allowances (Semin, Grevtsev, & Egoshina, 2018).
The same data was used to calculate the sector average to be able to compare the two sectors. A description of the variables and their descriptive statistics can be found in the appendix.
5. Methodology
This section examines the methodology that is used. The research is structured into three different sections. The correlation between the development of the EUAs prices and the STOXX EUROPE 600 Utility Index (Euro Stoxx) value, the relationship between the
development of the stock prices from companies within the energy sector and EUA prices as well as the stock value of firms within the aircraft sector and the EUA prices during the third phase of the scheme. Finally, the two sectors will be compared.
Within this paper, the multifactor market model, based on the Capital Asset Pricing Model (CAPM), was used (Mo et al., 2012). This model is commonly used in related empirical research and is able to include different common and fundamental factors that have an effect on the value of individual corporations (Mo et al., 2012).
For the correlation between the EUA prices and the Euro Stoxx return, first, the Pearson correlation test was conducted, second, a multifactor market model was used (Wang, Xie, Chen, Yang, & Yang, 2013). In the energy sector and aircraft sector analysis, the
multifactor market model was used to analyse the correlation between the sector and the development of the Euro Stoxx, which is in accordance with Mo et al. (2012). However, Mo’s et al. (2012) multifactor market model was adjusted, the lead and lag terms were left out because the trading increased significantly during phase three.
Fig. 2. The trading volume of the EUA futures during phase three (2013 – 2019) (data from the European climate exchange)
Furthermore, in addition to the firm’s Mo et al. (2012) included in their model, E.ON (Germany - GE) and RWE (GE) were added. They have been included because they take up a significant share in the electricity market and, the German electricity market is one of the biggest in Europe (Oberndorfer, 2009). Furthermore, apart from Italy, Spain, and the UK, Germany is one of the biggest CO2 emitters in Europe. These four countries combined take up 69% of the total allowances within the EU ETS market (Makridou et al., 2019).
First, the Pearson correlation coefficient (𝐶𝑖𝑗) was used to find the correlation between the variables, with a main focus on the two variables of interest Euro Stoxx and EUA (Koch, Fuss, Grosjean, & Edenhofer, 2014):
𝐶𝑖𝑗 = (𝑟𝑖𝑟𝑗) − (𝑟𝑖)(𝑟𝑗) 𝜎𝑖𝜎𝑗
in which 𝑟𝑖 and 𝑟𝑗 are the returns of the Euro Stoxx and EUA price respectively. The standard deviation ( 𝜎 ) is calculated as follows:
𝜎𝑖 = √(𝑟𝑖2) − (𝑟𝑖)2
After correcting for multicollinearity, a multifactor market model was used to analyse the relationship between the return of EUA and the return of the Euro Stoxx.
𝑹𝒎,𝒕 = 𝜶 + 𝜷𝒊𝒎𝑹𝒊𝒎,𝒕+ 𝜷𝒊𝒐𝑹𝒔,𝒕+ 𝜷𝒊𝒊𝑹𝒊𝒊,𝒕+ 𝜷𝒊𝒄𝑹𝒄,𝒕+ 𝜺𝒊,𝒕
where 𝑹𝒎,𝒕 is the price change of the Euro Stoxx at day t, α is constant. 𝑹𝒊𝒎,𝒕 is the price change of the MSCI index on day t. 𝑹𝒔,𝒕 stands for the price change of S&P 500 index on day t. 𝑹𝒊𝒊,𝒕 is the price change of ETF index invested into government bonds. 𝑹𝒄,𝒕
represents the percentage change of the EUA price on day t. Moreover, 𝜺𝒊,𝒕 is a disturbance term with Ε (𝜺𝒊,𝒕) = 𝟎, 𝑉𝐴𝑅(𝜺𝒊,𝒕) = 𝝈𝟐. The betas (𝜷) respectively stand for the effect of
the market return (𝜷𝒊𝒎) on the index, the effect of the change of the oil prices (𝜷𝒊𝒐) on the index, the effect of the ETF price change on the index (𝜷𝒊𝒊), and the effect of EUA price change (𝜷𝒊𝒄) on the index. The betas were estimated by using the OLS regression and the robustness standard error wereused to obtain unbiased standard errors to account for heteroscedasticity.
5.1 Energy Sector
To analyse the impact on the chosen companies within the energy sector, a multifactor market model was used, in accordance with Mo et al. (2012). However, they included lead and lag terms to adjust for low trading volume during the first and second phase, which was left out in this analysis since the trading volume increased significantly during phase three (Fig.2.).
𝑹𝒊,𝒕= 𝜶 + 𝜷𝒊𝒎𝑹𝒎,𝒕+ 𝜷𝒊𝒆𝑹𝒆,𝒕+ 𝜷𝒊𝒐𝑹𝒐,𝒕+ 𝜷𝒊𝒄𝑹𝒄,𝒕+ 𝜺𝒊,𝒕
where 𝑹𝒊,𝒕 is the price change of the company within energy sector at day t and 𝜶 is constant. 𝑹𝒎,𝒕 is the price change of the Euro Stoxx. 𝑹𝒆,𝒕 is respectively the change of the electricity price on day t. 𝑹𝒐,𝒕 stands for the change in oil price on day t. 𝑹𝒄,𝒕 is a placeholder for the change in EUA price on day t. Moreover, 𝜺𝒊,𝒕 is a disturbance term with Ε (𝜺𝒊,𝒕) = 𝟎, 𝑉𝐴𝑅(𝜺𝒊,𝒕) = 𝝈𝟐. The betas (𝜷) respectively stand for the effect of the market return (𝜷𝒊𝒎) on the stock return, the effect of the change in electricity price (𝜷𝒊𝒆) on the stock return, the effect of the change of the oil prices (𝜷𝒊𝒐) on the stock return and the effect of EUA price (𝜷𝒊𝒄) change on stock return. The betas were estimated by making us of the OLS regression and the robustness standard error were used to obtain unbiased standard errors to account for heteroscedasticity.
5.2 Aircraft Sector
To analyze the aircraft sector the same adjusted multifactor market model was applied.
𝑹𝒊,𝒕= 𝜶 + 𝜷𝒊𝒎𝑹𝒎,𝒕+ 𝜷𝒊𝒐𝑹𝒐,𝒕+ 𝜷𝒊𝒄𝑹𝒄,𝒕+ 𝜺𝒊,𝒕
where 𝑹𝒊,𝒕 is the price change of the company within the aircraft sector at day t and 𝜶
is constant. 𝑹𝒎,𝒕 is the price change of the market Euro Stoxx. 𝑹𝒐,𝒕 stands for the change in oil price on day t. 𝑹𝒄,𝒕 is a placeholder for the change in EUA price on day t. Moreover, 𝜺𝒊,𝒕 is
a disturbance term with Ε (𝜺𝒊,𝒕) = 𝟎, 𝑉𝐴𝑅(𝜺𝒊,𝒕) = 𝝈𝟐. The betas (𝜷) respectively stand for the effect of the market return (𝜷𝒊𝒎) on the stock return, the effect of the change of the oil prices (𝜷𝒊𝒐) on the stock return and the effect of EUA price (𝜷𝒊𝒄) change on stock return. The betas were estimated by making use of the OLS regression and the robustness standard error were used to obtain unbiased standard errors.
In both models the sign before the ß indicates how the variable is correlated with the stock price performance each company, namely if 𝜷 is positive it means that the variable is positively correlated to the stock price of the individual company. A positive EUA 𝜷 would result in an increase in the stock value and vice versa.
In the last part of the analyses the two sectors are compared. The stock prices were used to calculate the returns for each company. Then the average for each sector was calculated to be able to compare them, which results in a multifactor model as before.
This empirical study consists of one Pearson correlation, eight regression of the EUA price and its effect on the Euro Stoxx performance as well as of 84(12 firms * 7 years) individual regression for the energy sector as well as 63 (9 firms * 7 years) individual regression for the aircraft sector. There will be two regressions to compare the effect of the EUA price on each sector.
6. Results
The following section is discussing the results of the research. The results are split in four different sections, correlation between the Euro Stoxx and the EUA prices, the
relationship between the energy sector and EUA prices, the aircraft sector and the EUA prices and finally a comparison between the two.
6.1 Correlation between the Euro Stoxx and EUA
The test statistics of the Pearson’s correlation coefficient is presented in table 1. The descriptive statistics can be found in the appendix (A.1.
Table 1.
Correlation between the variables (n = 88)
Variables 1 2 3 4 5
2. EUA Return .2073 -
3. MSCI Index Return .5319*** .1572 -
4. S&P 500 Index Return .4356*** .1296 .9785*** -
5. ETF Index Return .5666*** -.0174 .2532* .2335* - *p < .05. **p < 0.01. ***p < .001.
There is a weak positive non-significant correlation of around 0.21 between Euro Stoxx and the EUA price. The results indicate that rising economy and rising share prices are stimulating the prices for EUAs, however, insignificant. After testing for multiclonality it comes apparent that MSCI and the S&P 500 variables are correlated and not independent. The model was adjusted by taking out the S&P 500 variable, to increase the validity of the model. The MSCI variable had a higher 𝑅2 value (0.5055) compared to S&P variable (0.4493), explaining a higher percentage of the variance and thereby increasing the overall value of the model (A.7).The robustness multifactor model was then used to estimate the correlation between the EUA and the Euro Stoxx. The average estimated betas are presented in table 2.
Table 2.
Mean and median estimate for 𝜷𝒎, 𝜷𝒊𝒊, 𝜷𝒊𝒄 from 2013 until 2019.
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒊𝒊(𝑬𝑻𝑭 𝒃𝒆𝒕𝒂) 𝜷𝒊𝒄(𝑬𝑼𝑨 𝒃𝒆𝒕𝒂)
Mean .3072 1.4274 .0323
Median .3355 1.2771 .0117
From the table 2 it can be seen that there is a weak positive, non-significant
correlation between the performance of the Euro Stoxx and the EUA price. When the EUA variable 𝑅2 is taken out the only loses around 7 percent of the explained variance. The mean 𝜷𝒊𝒄 of 0.0323 indicates that if the Euro Stoxx increases by 10% the value of the EUA price
will increases by 0.32% on average. Therefore, it can be said that the Euro Stoxx and the EUA prices are insignificant correlated. It is interesting that the 𝜷𝒊𝒊 is more significant than
the 𝜷𝒎. An increase of 10% would indicate an increase of 14.2% of the ETF index. The 𝜷𝒎, however, increases by only 3.1% if the Euro Stoxx increases by 10%.
The estimated market model for each sample company within the energy sector from 2013 until 2019 can be found in the Table A.4. The aggregate results of my data from 2013 until 2019 are presented in table 4. The results are compared to the research of Mo et al. (2012), done during the first phase of the EU ETS from 2006 until 2009. It has to be taken into account, that two new firms were added, one was taken out in my data set, as well as the difference in the time periods. Consequently, the comparability is limited.
Table 3.
Mo et al. (2012) mean and median estimates for 𝜷𝒎, 𝜷𝒆, 𝜷𝒐, 𝜷𝒄 (2006 - 2009).
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒆(𝒆𝒍𝒆𝒄𝒕𝒓𝒊𝒄𝒊𝒕𝒚 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean 0.9005 0.0680 0.0420 -0.0140
Median 0.8702 0.0577 0.0160 -0.0021
Table 4.
Mean and median estimate for 𝜷𝒎, 𝜷𝒆, 𝜷𝒐, 𝜷𝒄 for 84 company – year observation for 12 firms (2013 – 2019)
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒆(𝒆𝒍𝒆𝒄𝒕𝒓𝒊𝒄𝒊𝒕𝒚 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean 0.9205 0.1091 -0.0262 -0.0218
Median 0.9157 0.0418 -0.0289 -0.0107
Small differences can be found from Mo et al. (2012) and my study. The difference can be explained by the development of the scheme over time as well as the different combination of firms. We can see that the market beta (𝜷𝒎) increased by around 0.01 and
still represents the most significant factor within the model. The electricity beta (𝜷𝒆)
increased by 0.04, increasing the effect on the stock price. The EUA beta (𝜷𝑪) decreased by around 0.006, increasing the effect the EUA price has on the individual firm. A 10% increase in the EUA price leads to a 0.2% decrease in the stock value of firms within the energy sector. A more significant change can be seen in the oil beta (𝜷𝒐) , that decreased by around 0.06 and is negative now. This means that an increase of 10% of the oil price will lead to an average decrease of 0.26% in the value of a firm within the energy sector. It can be said that the effect of the EUA on corporate value did not change much from phase one to phase three.
Figure (3) visualizes the evolution of the EUA beta (𝜷𝑪)for each firm over time. The previous table presented the mean and median, therefore the average of all companies within the energy sector. Each company is differently affected by the EUA price movements, because of the amount of the allowances needed.
Fig.3. The development of 𝜷𝒄 from 2013 to 2019 for each energy company.
From this graph it can be seen that the beta for most companies have a similar development over time. However, there are still differences between companies. These differences are very significant in 2019. Firms such as the public company (GR), Red Eléctrica (ES) and Terna (IT) in 2019 have a large negative beta compared to the rest. Given a negative beta results in a decreasing stock value, when the EUA prices increases. In 2019 most 𝜷𝒄 were negative, the betas ranged from almost negative 4 to 0.16.
Table 4 presents a summary of the development of the estimated carbon price beta for each year.
Table 5.
Development of mean and median 𝜷𝒄 from 2013 until 2019.
2013 2014 2015 2016 2017 2018 2019
Mean -.0283 0.0006 0.0452 0.0114 0.0114 0.0999 -0.3577 Median -0.038 0.0039 0.0884 0,0073 0.0073 0.0088 -0.0158
The mean EUA beta is mostly stable, except at the beginning in 2013 and at the end in 2019. I would not have expected that the 𝜷𝒄 is mostly positive and only decreases in 2019, effecting the overall effect. The results could be explained by the increase of allowances that had to be bought to cope with economic growth and the increased EUA price (Fig.1)
6.3 Aircraft Sector
The estimated market model for each sample company within the aircraft sector from 2013 until 2019 can be found in the Table A.5. The aggregate results of my data from 2013 until 2019 for firms within the aircraft sector are presented in table 6. The results are compared to the previous findings for firms in the energy sector (table 5.).
Table 6.
Mean and median estimate for 𝜷𝒎, 𝜷𝒆, 𝜷𝒐, 𝜷𝒄 for 78 company – year observation for 13 firms (2013 – 2019)
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒆(𝒆𝒍𝒆𝒄𝒕𝒓𝒊𝒄𝒊𝒕𝒚 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean 0.9205 0.1091 -0.0262 -0.0218
Median 0.9157 0.0418 -0.0289 -0.0107
Table 7.
Mean and median estimate for 𝜷𝒎, 𝜷𝒐, 𝜷𝒄 for 54 company-year observation for 9 firms (2013 – 2019)
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean .2332 -0.3217 0.0671
Median .2555 -0.3063 0.0505
The aircraft sector market beta is comparably smaller than the one from the energy sector. If the market return increases by 10% on average the value of the stock of companies within the aircraft sector increases by 2.3%. Further, on average, if the EUA prices increase by 10 % the stock price of firms from the aircraft sector increases by 0.6% on average, compared to the beta of the energy sector, where a 10 % increase lead to a decrease of 0.2%. Finally, it can be seen that the increase of the oil price of 10% lead to a decrease of 3.2% decrease of the stock value on average. Compared to the energy sector, where an increase of 10% leads to a decrease of around 0.3%. It can be said, that the aircraft sector is more effected by oil price changes then the energy sector. Similar to the energy sector, the betas are representing the average, the effect is firm specific and changes over time. The
development of the firm specific EUA beta within the aircraft sector are summarized in figure 4.
Fig.4. The development of 𝜷𝒄 from 2013 to 2019 for each aircraft company.
It can be seen that there is a decrease in the EUA beta in 2015 for all companies within the aircraft sector. Noticeable is that apart from the dip in 2015 all company betas are positive throughout the time period from 2013 until 2019. Further, in 2019 all betas are positive, which could be explained by the increasing price of the EUA towards 2019 (Fig 1). The price increase makes the allowances more valuable, by selling EUAs these firms can earn extra (positive) cash flow. Further, there are less outliers compared to the energy sector.
In table 8 the average development of the EUA beta is presented for the aircraft sector.
Table 8.
Development of 𝜷𝒄 from 2013 until 2019.
2013 2014 2015 2016 2017 2018 2019
Mean -0.0418 0.0775 -0.3425 0.2917 0.2059 0.0894 0.1960 Median -0.0393 0.0601 -0.3365 0.2619 0.1917 0.1368 0.2070
Looking at the results in table 8, it can be said that there is slight fluctuation going from negative to positive within the first three years. This effect could be explained by the reason that the aircraft sector was included into the EU ETS in 2012 and didn’t participate in the trial phase from 2005 until 2008. However, as of 2016 the mean beta estimate shows a common positive trend.
6.4 Comparison between Sectors
The market model was estimated for each year taking the average of each sector (Table A.6.). The aggregate results of both the energy sector as well as the aircraft sector can be found in table 8.
Table 9.
Mean and Median 𝛽, estimates for energy sector from 2013 until 2019.
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒆(𝒆𝒍𝒆𝒄𝒕𝒓𝒊𝒄𝒊𝒕𝒚 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean 0.9913 0.1091 -0.0199 -0.0206
Median 1.0245 0.1487 -0.0382 0.0061
Table 10.
Mean and Median 𝛽, estimates for aircraft sector from 2013 until 2019.
𝜷𝒎(𝒎𝒂𝒓𝒌𝒆𝒕 𝒃𝒆𝒕𝒂) 𝜷𝒐(𝒐𝒊𝒍 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂) 𝜷𝑪(𝑬𝑼𝑨 𝒑𝒓𝒊𝒄𝒆 𝒃𝒆𝒕𝒂)
Mean 0.3038 0.0000 0.0585
Median 0.5286 0.0000 0.0911
The mean for the market beta is in both cases positive, however for the energy sector it is almost equal to 1, therefore it is positively correlated. The correlation to the EUA price is in line with the previous analysis. The energy sector is slightly negative correlated to the EUA prices, whereas the aircraft sector has a slight positive correlation. The results suggest that the value for firms within the aircraft sector, on average, increases when EUA prices increases. The opposite is true for firms within the energy sector.
The development of the EUA betas for each year, for each sector can be found in table 11.
Table 11.
2013 2014 2015 2016 2017 2018 2019 Energy Sector 𝜷𝒄 -0.0305 -0.0002 0.0487 0.0122 0.0829 0.1076 -0.3852
Aircraft Sector 𝜷𝒄 -0.0464 0.0861 -0.3805 0.3241 0.2288 0.0911 0.1063
Comparing the two it can be seen that both betas have a similar development. However, the aircraft beta is more extreme in most cases. The energy beta moves around zero, except in year 2019. Interestingly is that the beta for the energy sector is mostly positive and then decreases a lot in 2019 to almost negative 0.4, which could be explained by an increase in economic growth. If there is economic growth, there is increased production and more electricity is needed, which means that more certificates have to be bought by the energy firms18.
The reason why the aircraft beta decreased that much during 2015 could be explained by incidents that happened during that year, not only within Europe but worldwide, such as the Germanwings crash and the crash of Metrojet Airbus over Egypt. All events that happen within the aircraft industry directly effects the customer perception on the safety and
therefore leads to a decrease of passengers (Chen-Wei, Veng Kheang, & Yai, 2015).
7. Discussion
My results suggest that there is, against my proposed theory an insignificant positive correlation between the price change of the EUA and the value of the Euro Stoxx. In this respect, an increase in the index increases the value of the allowances. An explanation could be that when the economy is booming, the overall production increases. More allowances are required by corporations increasing the overall demand and thereby the price per EUA. On the other hand, if the economy is in a recession the demand for allowances decrease therefore prices decrease, affecting each sector (Stuhlmacher et al., 2019).
The effect is sector and therefore firm specific. As predicted, firms within the energy sector are negatively correlated to EUA price changes, as the prices for the allowances increase the (stock) value decreases. The opposite is true for firms within the airline sector, which could be an effect of the almost free allocation of allowances. This insignificant effect was also found by Dutta et al. (2018) for clean energy firms. The similarity in the result could be explained by the fact, that being able to sell the allowances when not needed has a similar
effect as getting the allowances for free. Moreover, when comparing the two sectors the aircraft sector profits from increasing EUA prices and the energy sector does not.
Oberndorfer (2009) analysed the energy sector during a different phase (one) and came to a similar conclusion, namely that there is a slightly positive correlation. Moreover, a lot of previous research has found a country specific effect such as Mo et al. (2012) or Dutta et al. (2018), however, the described effect was not present in this research due to the change from a decentralized cap-and-trade system to a single EU-wide cap19. Finally, the results made apparent that the effect depends on the EUA allocation policy, which is in line with findings from Mo et al. (2012).
This research is limited due to time and monetary reasons. The sample selection was based on previous research, however, more companies for each sector could have been used. One company from the energy sector had incomplete data and was taken out. A similar problem was found in the aircraft sector. Two firms had stock prices from 2015 onwards and were taken out, decreasing the sample to nine firms.
Furthermore, the daily stock price data was incomplete, therefore the monthly data had to be used. The variables that have been used to analyse the energy sector were based on previous research, therefore no new variables were introduced that might affect the overall outcome. Bushnell et al. (2013) found that the firm performance is affected by the type of power generation of the individual firm, firm specific variables were left out. The variables used for the aircraft sector were based on the energy sector variables. There is a high chance, that firms within the aircraft sector also have firm specific variables that effect their stock value. Further, there was no proxy included that represented airplane ticket price, as there was for the energy sector. This proxy is very important, when taken into account the statement from Bushnell et al. (2013), that costs are passed on to the final customer.
Generally, it can be assumed that there are firm specific influences that need to be accounted for (Oberndorfer, 2009).
It can be concluded that the EUAs has an impact on participating firms and their investment decisions (Stuhlmacher et al., 2019). Firms will have to take into consideration, how they are going to be dealing with the EU ETS, namely if they invest in new technologies or if they buy extra allowances. The less CO2 is emitted the more the firm will profit from the trading scheme (Mo et al., 2012). Furthermore, the empirical findings could be
important for investors as well as policymakers (Dutta et al., 2018). According to Dutta et al.
(2018) investor base their decision on big indices such as the Euro Stoxx, which is affected by the EUA price movements.
It would be interesting to conduct further research on how the other sectors are
affected by the EU ETS and to compare them to previous findings. Furthermore, firms within Europe should be studied and how they might have a competitive disadvantage over the ones outside the EU, especially regarding the manufacturing sector that sells goods and services on a global scale (Löschel et al., 2019). Finally, it will be interesting to further investigate how the effect will develop over phases four. Phase four will start in 2021 and will last until 2030. The annual reduction will increase to 2.2% to be able to achieve the targets of the Paris agreement by 203020. To achieve this target Europe has to cut 43% of its total emissions compared to 200521. This goal will result in less allowances and therefore an increase in the price, therefore this new setting will have to be studied further.
8. Conclusion
This paper empirically tested the effect of the European Trading Scheme on the value of the Euro Stoxx, as well as on firms within the energy and aircraft sector during phase three of the EU ETS. This research contributes to current literature research by extending previous studies that focused on phase one and two of the EU ETS, done by Oberndorfer (2009) and Mo et al. (2012). This was done by choosing an European index (Euro Stoxx) and two different sectors within the index that are part of the EU ETS.A multiple market model was used to analyse the effect, in line with previous empirical research.
The main results of this paper suggest that the European Union allowances have an effect corporate value. The effect mainly depends on the allowance’s allocation, namely whether the firm has to buy the allowances or gets them freely allocated. It suggests that firms within the aircraft sector, who get most of their allowances for free, are better off than the firms within the energy sector. The research question: “Is there a positive (negative) correlation between the STOXX EUROPE 600 Utility Index return and the increases (decrease) of the EUA price, with a special focus on how the free allocation of EUAs affect firms within the energy- and aircraft sector?” can be answered. The results suggest, that firms who are part of the aircraft sector, thereby receive most of the allowances for free, are better
20 https://ec.europa.eu/clima/policies/ets_en
off than firms who are part of the energy sector and have to buy them. Further, there is, even though it’s insignificant, a positive correlation between the Euro Stoxx and the EUA. It becomes apparent that firms from different sectors are also affected by the movements of the EUA prices.
To come to a conclusion, the European Union Trading Scheme does affect financial (stock) markets as a whole and therefore has an effect on the individual firm, which is depended on the EUA allocation policy.
It is, therefore, important for corporation who are part of the EU ETS, to take further actions to decrease their total amount of emitted greenhouse gases, to assure economic growth. Environmental regulation can negatively affect a corporation, but if done right, lead to economic growth (Markidou et al.,2016).
9. Bibliography
Albertini, E. (2013). Does Environmental Management Improve Financial Performance? A Meta-Analytical Review. Organization and Environment, 26(4), 431–457.
Betz, R., Rogge, K., & Schleich, J. (2006). EU emissions trading: An early analysis of national allocation plans for 2008–2012. Climate Policy, 6(4), 361–394.
https://doi.org/10.1080/14693062.2006.9685608
Bode, S. (2006). Multi-period emissions trading in the electricity sector-winners and losers. Energy Policy, 34(6), 680–691. https://doi.org/10.1016/j.enpol.2004.06.017
Buchner, B. K., Carraro, C., & Ellerman, A. D. (2011). The Allocation of European Union Allowances: Lessons, Unifying Themes and General Principles. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.929109
Busch, T., & Hoffmann, V. H. (2007). Emerging carbon constraints for corporate risk management. Ecological Economics, 62(3–4), 518–528.
https://doi.org/10.1016/j.ecolecon.2006.05.022
Bushnell, J. B., Chong, H., & Mansur, E. T. (2013). Profiting from Regulation: Evidence from the European Carbon Profiting from Regulation: Evidence from the European Carbon Market Market. https://doi.org/10.1257/pol.5.4.78
Chen-Wei, L., Veng Kheang, P., Yai, T. (2015). The Effects of Aviation Accidents on Public Perception toward an Airline.
https://www.researchgate.net/publication/294495442_The_Effects_of_Aviation_Ac
Dutta, A., Bouri, E., & Noor, M. H. (2018). Return and volatility linkages between CO2 emission and clean energy stock prices. Energy, 164, 803–810.
https://doi.org/10.1016/j.energy.2018.09.055
Flachsland, C., Pahle, M., Burtraw, D., Edenhofer, O., Elkerbout, M., Fischer, C., Tietjen, O., & Zetterberg, L. (2020). How to avoid history repeating itself: the case for an EU Emissions Trading System (EU ETS) price floor revisited. Climate Policy, 20(1), 133–142. https://doi.org/10.1080/14693062.2019.1682494
Friedrich, M., Mauer, E.-M., Pahle, M., Tietjen, O. (2020). A Service of zbw
Leibniz-Informationszentrum Wirtschaft Leibniz Information Centre for Economics Standard-Nutzungsbedingungen: From fundamentals to financial assets: the evolution of understanding price formation in the EU ETS *. www.econstor.eu
Hoffmann, V. H. (2007). EU ETS and Investment Decisions:. The Case of the German Electricity Industry. European Management Journal, 25(6), 464–474.
https://doi.org/10.1016/j.emj.2007.07.008
Jong, T., Couwenberg, O., & Woerdman, E. (2014). Does EU emissions trading bite? An event study. Energy Policy, 69, 510–519. https://doi.org/10.1016/j.enpol.2014.03.007
Koch, N., Fuss, S., & Grosjean, G. (2014). Causes of the EU ETS price drop: recession, CDM, renewable policies or a bit of everything?-New evidence.
Kruger, J., Oates, W. E., & Pizer, W. A. (2007). Decentralization in the EU Emissions Trading Scheme and Lessons for Global Policy.
https://doi.org/10.1093/reep/rem009
Löschel, A., Lutz, B. J., & Managi, S. (2019). The impacts of the EU ETS on efficiency and economic performance – An empirical analyses for German manufacturing firms. Resource and Energy Economics, 56, 71–95.
https://doi.org/10.1016/j.reseneeco.2018.03.001
Makridou, G., Doumpos, M., & Galariotis, E. (2019). The financial performance of firms participating in the EU emissions trading scheme. Energy Policy, 129, 250–259.
https://doi.org/10.1016/j.enpol.2019.02.026
Marcel Oestreich, A., Tsiakas, I. (2015). WP 15-18 CARBON EMISSIONS AND STOCK RETURNS: EVIDENCE FROM THE EU EMISSIONS TRADING SCHEME Carbon Emissions and Stock Returns: Evidence from the EU Emissions Trading Scheme.
www.rcfea.org-secretary@rcfea.org
Meleo, L., Nava, C. R., & Pozzi, C. (2016). Aviation and the costs of the European Emission Trading Scheme: The case of Italy. Energy Policy, 88, 138–147.
https://doi.org/10.1016/j.enpol.2015.10.008
Mo, J. L., Zhu, L., & Fan, Y. (2012). The impact of the EU ETS on the corporate value of European electricity corporations. Energy, 45(1), 3–11.
Montgomery, W. D. (1972). Markets in licenses and efficient pollution control programs. Journal of Economic Theory, 5(3), 395–418.
https://doi.org/10.1016/0022-0531(72)90049-X
Oberndorfer, U. (2009). EU Emission Allowances and the stock market: Evidence from the electricity industry. Ecological Economics, 68(4), 1116–1126.
https://doi.org/10.1016/j.ecolecon.2008.07.026
Oberndorfer, U., Rennings, K., & Sahin, B. (2006). The Impacts of the European Emissions Trading Scheme on Competitiveness and Employment in Europe-a Literature Review A report commissioned by World Wide Fund for Nature (WWF) Structure of Report.
Schmale, J.T. (2014). The impact of EU carbon price changes on the stock performance of European electricity firms. Honor Thesis. University of North Carlina.
SEMIN, A., GREVTSEV, N., & EGOSHINA, O. (2018). Systems Approach in Modeling Production and Consumption of Peat Products. Journal of Environmental
Management and Tourism, 8(5), 957–967.
https://doi.org/10.14505//jemt.v8.5(21).01
Stuhlmacher, M., Patnaik, S., Streletskiy, D., & Taylor, K. (2019). Cap-and-trade and emissions clustering: A spatial-temporal analysis of the European Union Emissions Trading Scheme. Journal of Environmental Management, 249, 109352.
Tian, Y., Akimov, A., Roca, E., & Wong, V. (2016). Does the carbon market help or hurt the stock price of electricity companies? Further evidence from the European context. Journal of Cleaner Production, 112, 1619–1626.
https://doi.org/10.1016/j.jclepro.2015.07.028
Veith, S., Werner, J. R., & Zimmermann, J. (2009). Capital market response to emission rights returns: Evidence from the European power sector. Energy Economics, 31(4), 605–613. https://doi.org/10.1016/j.eneco.2009.01.004
Wang, G. J., Xie, C., Chen, S., Yang, J. J., & Yang, M. Y. (2013). Random matrix theory analysis of cross-correlations in the US stock market: Evidence from Pearson’s correlation coefficient and detrended cross-correlation coefficient. Physica A: Statistical Mechanics and Its Applications, 392(17), 3715–3730.
https://doi.org/10.1016/j.physa.2013.04.027
Zetterberg, L., Nilsson, K., Åhman, M., Kumlin, A.-S., & Birgersdotter, L. (2014). Analysis of national allocation plans for the EU ETS. www.ivl.se,
Table A.1. Definitions
𝑹𝒊 The return for each individual company ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒎 The return of the Euro Stoxx ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒊𝒎 The return of the MSCI Index ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒊𝒊 The return of iBoxx EUR Liquid Sovereigns Diversified 25 Index (ETF) ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒆 The change in the Phelix Future price ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒐 The change in Europe Brent Spot price (FOB) ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
𝑹𝒄 𝑹𝒔
The change in EUA price ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1)
The change in the S&P 500 price ln ( 𝑃𝑟𝑖𝑐𝑒𝑡
𝑃𝑟𝑖𝑐𝑒𝑡−1) Table A.2.
Descriptive Statistics
Mean Std. Dev. Min Max
𝑹𝒎 0.0031 0.0402 -0.1657 0.0789 𝑹𝒊𝒎 0.0077 0.0381 -0.1350 0.1155 𝑹𝒊𝒊 0.0028 0.0109 -0.0224 0.0243 𝑹𝒆 0.0014 0.0534 -0.1135 0.1672 𝑹𝒐 -0.0085 0.1006 -0.5501 0.2073 𝑹𝒄 0.0190 0.1461 -0.4284 0.4219 Table A.3
The estimation results of the 𝜷𝒎, 𝜷𝒊𝒊, 𝜷𝒊𝒄 (7 individual regressions).
Year 𝜷𝒎 P value 𝜷𝒊𝒊 P value 𝜷𝒊𝒄 P value
STOXX 600 Utility Index 2013 - 0,080188 0.72 2.680601 0.000 0.0252422 0.43 2014 0,5873434 0.239 - 0.0960001 0.95 - 0.0034861 0.976 2015 0,626628 0.001 0.5100409 0.509 0.2649586 0.222 2016 0,1553103 0.362 2.044251 0.002 0.1521647 0.000 2017 0,1826443 0.701 2.574707 0.015 -0.0476372 0.656 2018 0,4977302 0.12 3.29591 0.002 0.0233873 0.536 2019 0,4884358 0.018 0.41004 0.525 -0.1561303 0.008
