Do companies with a low ESG rating
experience different abnormal
returns than companies with a high
ESG rating around announcements of
environmental policy?
Findings on the impact of the Paris Agreement
announcement on the returns of low- and high ESG
rated companies based in Europe vs China
Name: Yannik Petrus ‘t Hart
Student number: 11669977
Institute: University of Amsterdam
Program: Economics and Business Economics – Finance Specialization
Date: 28 June 2020
Statement of originality
This document is written by Yannik ‘t Hart who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is 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.
Index
Statement of originality 2
1.Introduction 4
2.Literature review 5
3.Research Question and Hypothesis 9
4.Methodology 11
5.Results 15
6.Discussion 20
7.Conclusion 21
4 1. Introduction
This paper studies the effect of environmental policy announcements on high ESG rated companies. It looks at companies based in China and companies based in Europe. This is done specifically by looking at the abnormal returns, if any, of those companies around the announcement date of the Paris Agreement. The Paris Agreement is the successor of the Kyoto Protocol. It was announced on the 12th of December
2015.
Since the beginning of this century, people have become increasingly aware of the climate change problem. Investors more and more look to invest in a philanthropic way where less attention is payed to making a profit and more on investing in companies who adhere to the values of the investors (Schwarts and Saiia, 2012). Companies who take into account their pollution and who treat their employees well are called socially responsible. The ESG rating has been developed to help quantify how well companies are doing in these aspects. ESG stands for Environmental, Social and corporate Governance.
The objective of this research is to discover if environmental policy announcements affect the stock returns
of both low- and high ESG rated companies, and if that effect differs between companies based in China or Europe. The outcome of this research is relevant, because a high ESG rating becomes increasingly important for investors nowadays. The outcome is also interesting for (institutional) investors who have part of their wealth invested in high ESG rated companies. It will help understand how environmental policies may be affecting the performance of their portfolio around the announcement date of these policies. Furthermore, the outcome may help clarify how sensitive companies in Europe and China are to the announcement of environmental policies. This might be of interest for investors, so they can adjust their portfolio to get their desired amount of risk and volatility.
This research uses the method of event studies. To establish if the announcement was really a surprise for the market, I look at the effect the announcement had on the returns of highly polluting companies. After having looked at these companies, I look at low ESG rated companies and high ESG rated companies to see if they experienced different abnormal returns. This is done for companies based in both Europe and China.
5 2. Literature review
As said before, the environmental policy on which this paper focusses on is the Paris Agreement. In an article about climate change policies, Seo (2017) states the different features agreed on in the Paris Agreement. The agreement was announced on the 12th
of December 2015 with as its main goal to keep temperature increase below 1.5˚C above pre-industrial levels. The agreements made in it are not legally binding, therefore it is called an agreement rather than a treaty. The agreement went into effect approximately a year after the announcement date. The reason for this is that at least 55 countries, who together should produce at least 55% of the greenhouse gas emission, had to accept the agreement.
In line with the Paris Agreement, all countries should set their own specific goals which are called the Nationally Determined Contributions (NDCs). Every five years these goals should be updated in a progressive way, this ensures that countries stay ambitious. The system works in such a way that countries can cooperate, so the emission reductions can be transferred from one to another. As of February 2020, all countries who are a member to
the UNFCCC have signed the agreement. The only high emitting countries whom are not yet party to the contract are Turkey and Iran. However, on 1 June, 2017 president Donald Trump of the United States announced that his country will withdraw from the Paris Agreement. Since a notice of withdrawal from the agreement is only possible 3 years after the agreement went into effect, and then it will still take a year before a country can withdraw, the United States can officially leave the agreement on November 4, 2020 the earliest (Urpelainen and Van de Graaf, 2017).
The strict long-term goal of the Paris Agreement is to keep the temperature increase to below 1.5˚C above pre-industrial levels, while the lenient version is to keep the temperature increase below 2˚C above pre-industrial levels. The agreement, however, does not clearly state what the pre-industrial levels are. Scientists use different periods for this. In a paper written by Schleussner et al. (2016), the average of the period 1850-1900 is used. To reach this long-term goal, there are three different targets to which countries must aim for. They have to reduce their CO2-emissions by 20%, they have to increase their energy efficiency by 20%
6 and they have to increase the market
share of renewable energy to at least 20%. To achieve those goals, prices on carbon emissions are needed. Those prices can for example be in the form of taxes or a trading scheme (Bassi et al., 2015). This may increase the costs of low ESG rated companies relatively more than the costs of high ESG rated companies. The increasing costs of low ESG rated companies can cause lower profits and therefore lower future values. This might cause a drop of those stock prices, which in turn should be reflected by negative abnormal returns. Furthermore, this might be a reason for investors to switch their money into high ESG rated companies, which can cause positive abnormal returns. Batten et al. (2016) find such significantly positive abnormal returns for companies from the renewable energy sector in an event study performed on the announcement of the Paris Agreement. They also find negative abnormal returns for oil and gas companies, however, those returns were not significant.
ESG consist of three components, Environmental, Social and corporate Governance. When talking about environmental, the focus is on how efficient companies use their resources. Think about water usage and waste
production etc. Under the heading of social, the treatment of clients and workers is covered. How well run the company is and how diversified the boards are, is looked at under corporate governance.
While behaving in a corporate socially responsible way is now more important than ever, in the past people were hugely influenced by the view of Milton Friedman who argued that the cost of behaving in such a way would outweigh the benefits it entails (Freeman and Liedtka, 1991). Friedman agreed with Adam Smith’s view of capitalism, he was influenced by the multiple failing socialist regimes during his lifetime. He argued that companies behaving in a corporate socially responsible way have too much restrictions on them, therefore they cannot pursue their goal to maximize firm value. Maximizing firm value would lead to more wealth for the people, whom in their turn could invest that in a socially responsible system. However, Dowell et al. (2000) find that firm value is linked to a companies’ environmental policy. The companies with a stricter environmental policy have higher Tobin’s Q values, and thus relatively higher market values. This induces that investors prefer putting their money in environmental friendly
7 companies. A reason for this could be
that being environmental friendly leads to a lower perceived level of risk (Feldman et al, 1995). When announced in a policy that rules are going to be more strict, environmental friendly companies (high ESG), may have an advantage compared to low ESG companies which might give them positive abnormal returns.
Multiple research has been conducted to the behavior of investors. Evidence is found that there are investors who care about other things than money (Derwall et al, 2011; Beal and Goyen, 1998; Statman, 2005; Hong and Kacperczyk, 2009). They are called values-driven investors. Instead of solely investing in companies on the basis of their profitability, they also take into account their social and personal values. This is called Socially Responsible Investing (SRI) and it emerged in the 18th
century (Derwall et al, 2011). Derwall et al (2011) come up with two different hypothesis, the shunned stock hypothesis and the errors-in-expectation hypothesis. The errors-in-expectation hypothesis says SRI stocks have higher returns, because the investors are slow in recognizing the positive impact of those stocks and thus the stocks will be underpriced. The shunned stock
hypothesis states that because investors might hold stocks for other reasons than solely making a profit, they create an excess demand for responsible stocks and a shortage demand for irresponsible stocks. When an announcement as big as the Paris Agreement is made, investors might themselves become more aware of the irresponsible stocks they possess and might switch and masse to the more responsible stocks, like high ESG rated stocks.
According to Dowell et al. (2000), investors expect companies to incur extra costs due to the environmental cleanup after the policy announcements. Dowell at al. (2002) state that these costs tend to be lower for high ESG rated companies than for low ESG rated companies. The reason for this is that the high ESG rated companies have less cleaning up to do. This might be a reason for them to invest in high ESG rated companies rather than low ESG rated companies. This switching to high ESG rated companies might create positive abnormal returns for those stocks. In previous research from Shane and Spicer (1983), Hamilton (1995), Klassen and McLaughglin (1996), and Karpoff et al. (2005) is shown that stock prices do change significantly after news about environmental issues, like new policies
8 or research which addresses
environmental problems. However, positive abnormal returns after positive news are smaller than negative abnormal returns after negative news.
When investigating the effect on European based companies versus Chinese based companies, it is important to take into account past actions on climate matters. Both the EU (10 percent) and China (30 percent) are in the top three largest emitters (Schreurs, 2016). Schreurs (2016) also addresses that Europe has always had the most stringent rules and restrictions on emissions while the Chinese government was very unwilling to implement such rules in the past. This might imply that European companies already did a lot to become more environmental friendly, while there is lots more to do for Chinese companies. However, during the Paris Agreement, China announced they are now willing to reduce their emissions and strengthen their climate action commitments wherever possible. In a research to the probability of achieving the emission targets set in the Paris Agreement for the top ten emitters, Dong et al. (2017) find that next to India and Russia, China is very likely to achieve their targets.
This paper makes use of event studies. This method has also been applied by Ramiah et al. (2013) and Ramiah et al. (2014). For an event study to work, the event which is studied must be a surprise. Then, according to the efficient market hypothesis, the new released information should immediately be priced in. From behavioral finance, we know that there are a lot of investor that suffer from the conservatism bias. This means they will not react immediately to the new information. Therefore the event study should have a time period of more than what is expected by the efficient market hypothesis.
Previous research done on the effect of (environmental) policy announcements on the prices of sin stocks ran into different problems. A possible reason why they could not find any abnormal returns after an environmental policy announcement, is that institutional investors like pension funds invest relatively less in socially irresponsible stocks (Hong and Kacperczyk, 2009). Institutional investors account for around 80% of market capitalization. So after an announcement like the Paris Agreement, they do not switch from ‘bad’ to ‘good’ stock since they already did not own the
9 bad stocks. Therefore, no significant
negative abnormal returns were observed. Only (some of) the individual investors may switch, and since they account only for about 20% of the market cap, that might not be enough to create significant abnormal returns. Ramiah et al. (2014) conclude that a possible explanations for the failure to see negative abnormal returns at highly polluting companies, is that those companies might be able to pass the extra cost on to their consumers. In another research, Ramiah et al. (2013) find that companies often bring out a statement of a possible change in their policy shortly after the policy announcement. This statement might influence the effect the policy announcement had in the first place. All these things are also challenges for this research. Furthermore, an extra obstacle may be that the Paris Agreement was drafted during the 21st conference of the
parties of the United Nations Framework Convention on Climate Change (UNFCCC). This convention is known worldwide and therefore people might already expect that an environmental policy would be announced during this convention. Hence, the announcement may not be a surprise and the abnormal returns will be less strong.
3. Research Question and Hypothesis
I start by looking if the announcement was even a surprise at all by looking at the effect it had on the abnormal returns of high polluting companies. As stated in the literature review, to achieve the goals of the Paris Agreement, prices on carbon emission are needed. Those can come in the form of taxes or a trading scheme (Bassi et al., 2015). This may increase costs for the high polluting companies and therefore decrease profitability. I expect investors might react on this by selling the stocks of these companies which translates in the first hypothesis tested.
Hypothesis 1:
H0: The cumulative average abnormal returns of high polluting companies are not affected by the announcement of the Paris Agreement.
H1: The cumulative average abnormal returns of high polluting companies are significantly negative around the announcement of the Paris Agreement.
Following the view of Dowell et al. (2000) and Dowell et al. (2002) where low ESG rated companies incur higher extra costs due to the cleanup after environmental policy announcements, combined with the expectation that investors sell the shares of those
10 companies, we form hypothesis 2.
Thereafter, following the findings of Batten et al. (2016) where significantly positive abnormal returns for companies from the renewable energy sector are found, I check if this also applies to high ESG rated companies in hypothesis 3.
Hypothesis 2:
H0: There is no effect on the cumulative average abnormal returns of low ESG rated companies around the announcement of the Paris Agreement H1: The cumulative average abnormal returns of low ESG rated companies are significantly negative around the announcement of the Paris Agreement.
Hypothesis 3:
H0: There is no effect on the cumulative average abnormal returns of high ESG rated companies around the announcement of the Paris Agreement H1: The cumulative average abnormal returns of high ESG rated companies are significantly positive around the announcement of the Paris Agreement.
After separately checking if the cumulative average abnormal returns are negative for low ESG rated companies and positive for high ESG rated companies, I look if low ESG rated companies have significantly lower abnormal returns than high ESG rated
companies. In other words, if the effect of being a high ESG rated company is significantly positive. I expect this effect to be present following the ‘switch and masse’ theory explained in the literature review and the findings of Dowell et al. (2002).
Hypothesis 4:
H0: There exists no difference between the cumulative abnormal returns of low ESG rated companies and the cumulative abnormal returns of high ESG rated companies around the announcement of the Paris Agreement.
H1: The cumulative abnormal returns of high ESG rated companies are significantly larger than the cumulative abnormal returns of low ESG rated companies around the announcement of the Paris Agreement.
Consequently, I look at hypotheses comparing European based companies to Chinese based companies. Following Schreurs (2016) and Dong et al. (2016), I expect the abnormal returns for Chinese based companies to be larger than those of European based companies since European based companies already did a lot to become more environmental friendly. Hence, I assume the Paris Agreement has a bigger impact on the abnormal returns of Chinese based
11 companies. I test this with the following
hypothesis:
Hypothesis 5:
H0: There exists no difference between the cumulative abnormal returns produced by European based companies and the cumulative abnormal returns produced by Chinese based companies around the announcement of the Paris Agreement.
H1: The cumulative abnormal returns of Chinese based companies are significantly larger than the cumulative abnormal returns of European based companies around the Paris Agreement.
Lastly, I expect the difference in abnormal returns for low- and high ESG rated companies is higher for Chinese based companies than for European based companies. I expect this, because the cleanup that has to be done after environmental policy announcements will most likely be more costly for the low ESG rated companies. I expect the difference to be bigger for Chinese based companies since European companies already have more restrictions and rules than Chinese companies to comply to and therefore might have less cleaning up to do (Schreurs, 2016),. This will be tested in the following hypothesis:
Hypothesis 6:
H0: The difference in cumulative abnormal returns between low ESG rated companies and high ESG rated companies is similar for European based companies and Chinese based companies.
H1: The difference in cumulative abnormal returns between low ESG rated companies and high ESG rated companies is significantly higher for Chinese based companies than for European based companies.
4. Methodology
To determine the effect of the Paris Agreement on the abnormal returns of low- and high ESG rated companies, this paper conducts an event study on those companies around the announcement date of the Paris Agreement. As explained in the research question section, I do this for companies based in both China and Europe. I start by looking at high polluting companies from Europe to see if there exists a surprise effect at all. Thereafter I look at low and high ESG rated companies to see if the effect differs for them. Lastly, I look at the difference between European based companies and Chinese based companies. This paper assumes
12 that the efficient market hypothesis
holds, but with some delay. This is because people may be subject to conservatism bias and therefore react to new information with delay. So instead of looking at abnormal returns the immediate trading day after the announcement, this paper looks at the 5 trading days before and after the announcement.
This paper uses the measure of Carbon Intensity as a criteria to select high polluting companies. Since there is almost no data available for companies in China for this variable, this is only done for the companies based in Europe. I use a sample of 22 companies. It is important for event studies that the event is a surprise. Hence, I check this by looking at the abnormal returns of the high polluting companies first. An announcement for a more stringent environmental policy should affect high polluting companies the most and cause negative abnormal returns. To rank companies in terms of ESG, I rank the companies based on their Environment Score. Both for China and Europe, the paper looks at the top 30 companies from the ranking for both high and low ESG ratings. Not all data was available for those companies. Hence, from those top 30’s, this paper looks at 25 low ESG
rated companies and 21 high ESG rated companies from Europe. For both low- and high ESG rated companies based in China, 29 companies are studied. The data for all the stocks is retrieved from the WRDS – Sustainalytics database.
Daily returns are needed for the event study. A CAPM model is created from the daily returns in the estimation window. To implement the CAPM model, a benchmark is needed to estimate the market return. This paper uses the daily returns from the STOXX Europe 600 index as benchmark for the European companies. For the Chinese companies, the Daily WRDS index for China is used. The yields of 10 year government bonds are used as risk-free rates. The model looks as follows:
𝑅𝑖𝑡 − 𝑅𝑓𝑡 = 𝛽0 + 𝛽1(𝑅𝑚𝑡 − 𝑅𝑓𝑡) + 𝜀𝑖𝑡 (1) where Rit is return of the specific stock i, at time t. Rft is the risk-free rate at time t. Rmt is the return of the benchmark market at time t. β0 is the intercept of the regression and β1 measures the extent to how much the return of a specific stock i will change when the market return changes. The estimation window consists of 30 days. That model is used to obtain normal returns:
13 Those estimated normal returns are then
compared to the actual returns of the stocks during the event window. The event window consists of the 5 trading days before (t1) and 5 trading days after (t2) the announcement. The difference between them will be seen as the abnormal return:
𝐴𝑅𝑖𝑡 = 𝑁𝑅𝑖𝑡 − 𝑅𝑖𝑡 (3)
All abnormal returns combined per stock, gives the cumulative abnormal returns::
𝐶𝐴𝑅𝑖 = ∑ 𝐴𝑅𝑖𝑡
𝑡2
𝑡1
(4) Consequently, for all five groups (high polluting, low ESG Europe, high ESG Europe, low ESG China and high ESG China), we compute the cumulative average abnormal returns:
𝐶𝐴𝐴𝑅 =1
𝑛∑𝐶𝐴𝑅𝑖
𝑛
𝑖=1
(5)
where n is the number of companies investigated in that specific group.
The abnormal returns can be significantly negative, not significantly different from zero or significantly positive. When they are significantly negative, the interpretation is that investors expect that the firm’s future value will decrease and this may drive away investors and therefore decrease
the price of that stock. I assume the effect of the announcement is that investors sell their shares of that specific company, because they may think it will become less profitable caused by the new policy announced. This effect is expected for the group of high polluting companies and for all the low ESG rated companies. When the abnormal returns are insignificant (negative or positive), the reason could be that the policy announcement is not perceived as a surprise to investors. The announcements made were already expected by the investors and therefore already priced in. Other explanations could be that companies can pass on the extra costs caused by the announced agreements to their consumers and therefore will not become less profitable or that the effect is simply not affecting companies significantly. The last scenario is that the abnormal returns will be significantly positive. These effects are expected for the high ESG rated companies.
To get a better overview of the what happens in the data, I construct table 1-5 in which the cumulative abnormal returns per company are shown. Only in table 6 are the t-test statistics reported which are used to reject or not reject hypotheses.
14 Expressed in statistical terms, the
following hypothesis are tested:
Hypothesis 1:
For this hypothesis I perform a one sided t-test on the cumulative average abnormal returns of the high polluting companies:
𝑡 = √𝑁𝐶𝐴𝐴𝑅 𝑆𝑐𝑎𝑎𝑟
(6) where n is the number of companies investigated in that group, Scaar is the standard deviation of the cumulative abnormal returns from the companies in that specific group. Since there are 22 companies in this sample, the degrees of freedom is 22-1=21. The critical t-value to be significant at a 5% level is therefore -1.721.
H0: 𝐶𝐴𝐴𝑅 = 0 H1: 𝐶𝐴𝐴𝑅 < 0
The following hypotheses are separately tested for all companies, European based companies and Chinese based companies.
Hypothesis 2:
For this hypothesis I perform a one sided t-test on the cumulative average abnormal returns of the low ESG rated companies using formula 6. Since there are 25 European and 29 Chinese companies in this sample. The degrees of
freedom are therefore, respectively, 24 and 28. The critical t-values to be significant at a 5% level are therefore, respectively, -1.711 and -1.701.
H0: 𝐶𝐴𝐴𝑅 = 0 H1: 𝐶𝐴𝐴𝑅 < 0
Hypothesis 3:
For this hypothesis I perform a one sided t-test on the cumulative average abnormal returns of the high ESG rated companies using formula 6. Since there are 21 European and 29 Chinese companies in this sample, the degrees of freedom are, respectively, 20 and 28. The critical t-values to be significant at a 5% level are therefore, respectively, 1.725 and 1.721.
H0: 𝐶𝐴𝐴𝑅 = 0 H1: 𝐶𝐴𝐴𝑅 < 0
Hypothesis 4:
For this hypothesis I made a dummy variable HighESG which is equal to 1 if the company is a high ESG rated company, and 0 otherwise. I look at the following regression formula:
𝐶𝐴𝑅𝑖𝑡 = 𝛽0 + 𝛽1 ∗ 𝐻𝑖𝑔ℎ𝐸𝑆𝐺 (7)
The β0 coefficient represents the abnormal returns for low ESG companies. By checking if the β1 coefficient is significantly positive we can check if being a high ESG rated company causes significantly larger abnormal
15 returns.
H0: β1 = 0 H1: β1 > 0
The following hypothesis is separately tested for all companies, low ESG rated companies and high ESG rated companies.
Hypothesis 5:
For this hypothesis I made a dummy variable China which is equal to 1 if the company is based in China, and 0 otherwise. I look at the following regression formula:
𝐶𝐴𝑅𝑖𝑡 = 𝛽0 + 𝛽1 ∗ 𝐶ℎ𝑖𝑛𝑎 (8)
The β0 coefficient represents the abnormal returns for European based companies, by checking if the β1 coefficient is significantly positive we can check if being a Chinese company causes significantly larger abnormal returns. H0: β1 = 0
H1: β1 > 0
Hypothesis 6:
For this hypothesis we will compare the β1 coefficient from formula 7 applied to European based companies, to that same β1 coefficient from the formula applied to the Chinese based companies. We will do this with the Z-test, provided by Clogg et al. (1995):
𝑍 = 𝛽1𝑐 − 𝛽1𝑒
√(𝑆𝐸(𝛽1𝑒)2+ 𝑆𝐸(𝛽1𝑐)2)
(9)
β1c is the beta coefficient from the formula applied to the Chinese based companies and β1e is the beta coefficient from the formula applied to the European based companies.
H0: β1c = β1e H1: β1c > β1e
5. Results
For hypothesis 1-3, this paper studies the cumulative average abnormal returns (CAAR) of the companies divided into 5 groups: High polluting (Europe), Low ESG Europe, High ESG Europe, Low ESG China and High ESG China. The cumulative average abnormal returns are constructed by taking the average of the cumulative abnormal returns (CAR) of the companies in that specific group. Those cumulative abnormal returns are shown in tables 1-5. For an effect to be considered as significant, I use the 5% level as a boundary. Companies highlighted in bold green produced a significantly positive cumulative abnormal return and companies with a significantly negative cumulative abnormal return are highlighted in italic
16 red. These tables are meant to clarify
the data and not to reject hypotheses. For the high polluting companies, I expect the cumulative abnormal returns to be significantly negative. As shown in Table 1, there are 13 companies (59%) whose cumulative abnormal returns react significantly. From those 13 companies there are 9 companies (41%) who had a significant negative effect as expected by this paper.
Table 1
High polluting (Europe)
COMPANY NAME CAR
ENDESA SA -0.010
ANGLO AMERICAN PLC -0.162***
RIO TINTO GROUP -0.079***
ELECTROCOMPONENTS PLC -0.067***
ALCATEL-LUCENT -0.095***
LOGITECH INTERNATIONAL
SA -0.036**
NATIONAL EXPRESS GROUP
PLC 0.003 ARM HOLDINGS PLC -0.056*** POSTNL NV -0.006 ANTOFAGASTA PLC -0.079*** STAGECOACH GROUP PLC -0.185*** NATIONAL GRID -0.004 JENOPTIK AG 0.022* AURUBIS AG -0.204*** ACEA SPA -0.023* KONTRON AG 0.040** GLENCORE PLC 0.147*** OESTERREICH POST AG 0.006 AMG ADVANCED METALLURGICAL 0.019 TNT EXPRESS NV -0.027* POLYMETAL INTL PLC -0.011 ROYAL MAIL HOLDINGS -0.019
***p<0.01, **p<0.05, *p<0.1
For the European based low ESG rated companies, I also expect the
cumulative abnormal returns to be significantly negative. As shown in Table 2, there are 18 companies (72%) whose cumulative abnormal returns react significantly. From those 18 companies there are only 9 companies (36%) who had a significant negative effect as expected by this paper. The expected effect seems already to be less present for the low ESG rated companies in Europe than for the high polluting companies.
Table 2
Low ESG Europe
COMPANY NAME CAR
SAMSONITE INTERNATIONAL SA -0.027**
IKB DEUTSCHE INDUSTRIEBANK 0.005
INVESTOR AB -0.027**
OBERBANK AG -0.004
ACKERMANS & VAN HAAREN
NV/SA 0.031**
BANCA POPOLARE DI SONDRIO -0.030**
MBANK SA 0.091***
BB BIOTECH AG 0.088***
RIT CAPITAL PARTNERS PLC 0.034**
MP EVANS GROUP PLC -0.085***
BANCA CARIGE SPA GEN &
IMPER -0.099***
SPAREBANK 1 SR BANK -0.060***
INTERMEDIATE CAPITAL GROUP 0.014
JUMBO SA 0.033**
LPP SA -0.140***
AZIMUT HOLDING SPA 0.026**
POWSZECHNA KASA
OSZCZEDNOSCI 0.089***
AMERICAN SHIPPING CO ASA -0.145***
MELKER SCHORLING AB -0.003
VZ HOLDING AG -0.106***
SALVATORE FERRAGAMO SPA -0.004
GETIN NOBLE BANK SA 0.147***
ALIOR BANK SPOLKA AKCYJNA 0.028**
KENNEDY WILSON EUR REAL EST 0.008 CARD FACTORY PLC 0.015
17 For Chinese based low ESG rated
companies, I also expect the cumulative abnormal returns to be significantly negative. As shown in Table 3, there are 18 companies (62%) whose cumulative abnormal returns react significantly.
Table 3
Low ESG China
COMPANY NAME CAR
LIANHUA SUPERMARKET
HOLDINGS 0.027**
JIANGXI COPPER CO LTD 0.027**
SHANGHAI JIN JIANG INTL
INDL 0.026**
SHANGHAI JINJIANG INTL TRAVL 0.019* ANHUI EXPRESSWAY CO LTD -0.018 FOSHAN ELECTRICAL &
LIGHTING 0.008
HUADIAN ENERGY CO LTD 0.050***
SHANDONG CHENMING PAPER
HLDG -0.005
GUANGDONG PROVINCIAL
EXPRESS 0.013
ENN ENERGY HOLDINGS LTD 0.039***
SINOTRANS LTD -0.043***
CITIC SECURITIES CO LTD 0.002 WEIQIAO TEXTILE CO LTD -0.011
ZIJIN MINING GROUP CO LTD 0.049***
GREAT WALL MOTOR CO 0.455***
BANK OF CHINA LTD 0.030**
HAITONG SECURITIES CO LTD 0.004
CHINA YURUN FOOD GROUP
LTD 0.082***
CIMC ENRIC HOLDINGS LTD -0.071***
PARKSON RETAIL GROUP LTD 0.002
DALIAN PORT (PDA) CO LTD -0.059***
INTIME RETAIL (GROUP) CO
LTD 0.047***
SINOTRUK (HONG KONG) LTD -0.014
ZHONGSHENG GROUP HLDGS
LTD 0.149***
CHINA LESSO GROUP HLDGS LTD -0.078***
CHINA HONGQIAO GROUP LTD 0.065***
CHINA MACHINERY ENG CORP 0.012
XINYI SOLAR HOLDINGS 0.067***
HUISHANG BANK CORP LTD 0.079***
***p<0.01, **p<0.05, *p<0.1
However, from those 18 companies there are only 4 companies (14%) who had a
significant negative effect as expected by this paper, while 14 companies (48%) reacted significantly positive as opposed by the expectations from this paper. Hence, there seems to be no negative effect at all for Chinese based companies, in fact, there seems to be an opposite effect.
For the European based high ESG rated companies, I expect the cumulative abnormal returns to be significantly positive. As shown in Table 4, there are 11 companies (52%) whose cumulative abnormal returns react significantly.
Table 4
High ESG Europe
COMPANY NAME CAR
BT GROUP PLC -0.048***
KONINKLIJKE PHILIPS NV -0.042***
BNP PARIBAS -0.014
GREAT PORTLAND ESTATES
PLC -0.038** HAMMERSON PLC -0.006 STMICROELECTRONICS NV -0.041** TRAVIS PERKINS PLC -0.005 HOLMEN AB 0.041*** NOVARTIS AG 0.056*** ACCIONA SA -0.026* ICADE -0.052*** TECHNICOLOR SA 0.087*** ABB LTD -0.050***
NOKIAN RENKAAT OYJ -0.126***
SONOVA HOLDING AG -0.009 ELRINGKLINGER AG 0.028* GETINGE AB 0.007 GECINA 0.007 CASTELLUM AB -0.006 QINETIQ GROUP 0.042*** TELEFONICA DEUTSCHLAND 0.002 ***p<0.01, **p<0.05, *p<0.1
From those 11 companies there are only 4 companies (19%) who had a significant
18 positive effect as expected by this paper.
This means there were 7 companies (33%) who reacted significantly negative.Also for these companies there seems to be an opposite effect as for what was expected.
For the Chinese based high ESG rated companies, I expect the cumulative abnormal returns to be significantly positive as well. As shown in Table 5, there are 20 companies (69%) whose cumulative abnormal returns react significantly. From those 20 companies there are 14 companies (48%) whom had a significant positive effect as expected by this paper. The expected effect seems to be present for these companies.
In table 6, the results for hypothesis 1-3 are shown. The results confirm the expectations for hypothesis 1. I expected the cumulative average abnormal returns for the high polluting companies to be significantly smaller than zero. Following the p-value of 0.000, I can reject the null-hypothesis for hypothesis 1 in favor of the alternative hypothesis. This means I can expect that the policy announcement will be perceived as a surprise for investors and therefore is not already priced in.
When looking at the cumulative average abnormal returns for low ESG
rated companies based in Europe I observe the same effect, but this effect is insignificant. Moreover, for low ESG rated Chinese based companies the effect is highly insignificant since the p-value is equal to 1.000.
Table 5
High ESG China
COMPANY NAME CAR
SINOPEC SHANGHAI
PETROCHEM 0.134***
YANZHOU COAL MINING CO
LTD 0.111***
CHINA PETROLEUM & CHEM
CORP 0.017
SEMICONDUCTOR MFG INTL
CORP 0.057***
PING AN INSURANCE GROUP 0.008
ANGANG STEEL CO LTD 0.042***
CHINA SOUTHERN AIRLINES -0.107***
CHINA VANKE CO LTD 0.256***
TSINGTAO BREWERY CO LTD 0.005
TIANJIN CAP ENVMNTL PROTN -0.066***
SHENZHEN EXPRESSWAY CO
LTD 0.025**
BOE TECHNOLOGY GROUP CO
LTD 0.007
BENGANG STEEL PLATES CO
LTD 0.077***
SAIC MOTOR CORP LTD 0.011
ZTE CORP 0.041***
INNER MONGOLIA YITAI COAL
CO 0.039***
BAOSHAN IRON & STEEL CO
LTD 0.031**
BYD COMPANY LTD -0.034**
CHINA TELECOM CORP LTD -0.001 CHINA OILFIELD SERVICES LTD -0.010
INNER MONGOLIA YI LI IND CO 0.083*** CHINA SHENHUA ENERGY CO
LTD -0.036***
SHUI ON LAND LTD -0.050***
CHINA COAL ENERGY CO 0.002 INDUSTRIAL BANK CO LTD 0.016
CHINA MOLYBDENUM CO LTD 0.054***
CHINA ZHONGWANG HLDGS LTD -0.089*** CHINA CONCH VENTURE
HOLDINGS 0.043***
CGN POWER CO LTD 0.042***
19 Even a contrary effect may be expected
for this group. Based on this strong evidence, I cannot reject the null-hypothesis for null-hypothesis 2.
For the high ESG rated companies I observe a comparable effect. The companies based in Europe do not experience a significantly positive return. They experience a significantly negative return as opposed to what is expected. However, the Chinese based companies experience highly positive returns as expected. Therefore, we cannot reject the null-hypothesis from hypothesis 3 for the European based companies. For the Chinese based companies we can reject the null-hypothesis in favor of the alternative hypothesis.
Table 6
Cumulative average abnormal returns
CAAR St.Err. t-value p-value High Polluting Low ESG EU Low ESG CH High ESG EU High ESG CH -0.083 -0.012 0.095 -0.019 0.071 0.019 0.015 0.021 0.010 0.016 -4.257 -0.823 4.511 -1.850 4.512 0.000 *** 0.209 1.000 0.960 0.000 *** *** p<0.01, ** p<0.05, * p<0.1
All results combined, I observe a different effect than expected. Instead of different effects between the low- and high ESG rated companies, I observe a
different effect depending on the place the company is based in.
As next check, I regress the cumulative abnormal returns on the dummy variable HighESG, which is equal to 1 if a company has a high ESG rating, and 0 otherwise. The results are shown in Table 7. With this regression, I observe if being a high ESG rated company causes significantly higher cumulative abnormal returns. None of the groups (all companies, European companies, Chinese companies) experience significantly higher cumulative abnormal returns as expected. Therefore, we cannot reject the null-hypothesis from hypothesis 4.
Table 7
CARi = β0 + β1*HighESG
(1) (2) (3) VARIABLES ALL EUROPE CHINA HighESG -0.00512 -0.00433 -0.00849 (0.0147) (0.0176) (0.0218) Constant 0.0154 -0.00477 0.0328* (0.0117) (0.0145) (0.0175) Observations 104 46 58
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
For hypothesis 5, I checked if cumulative abnormal returns for Chinese based companies are higher than for European based companies. This is done by regressing those cumulative abnormal returns on a dummy variable equal to 1
20 for Chinese companies, and 0 otherwise.
As shown in table 8, this effect is present for all groups (all companies, low ESG rated companies and high ESG rated companies). However, the effect is only significant at the 10% level for the low ESG rated companies. Therefore, we can reject the null-hypothesis in favor of the alternative hypothesis for hypothesis 5 for all companies together or for the group of high ESG rated companies. For the group of low ESG rated companies alone, the effect is not significant enough.
Table 8
CARit = β0 + β1*HighESG
(1) (2) (3)
VARIABLES ALL LOW ESG HIGH ESG China 0.0353*** 0.0376* 0.0335***
(0.0141) (0.0227) (0.0164) Constant -0.00675 -0.00477 -0.00910 (0.00899) (0.0144) (0.0100) Observations 104 54 50
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
To observe if the difference in cumulative abnormal returns between low ESG rated companies and high ESG rated companies is similar for European based companies and Chinese based companies, I applied the Z-test. The p-value for this test is equal to 0.4404, this is highly insignificant and therefore we cannot reject the null-hypothesis from hypothesis 6.
6. Discussion
This research has found different results than what was expected. Unlike a separation in abnormal returns for the low vs. high ESG rated companies, I observe a separation between European based companies and Chinese based companies. European based companies tend to have negative abnormal returns, while Chinese based companies tend to have positive abnormal returns. As mentioned in the literature review, there might be multiple reasons why abnormal returns did not arise after the announcement of the Paris Agreement. One of them is that companies can pass on their extra costs to their consumers. However, I do observe significant cumulative average abnormal returns for the high polluting companies and for the high ESG rated companies from China. A possible explanation for this may be that this research selected the companies based on their environment score, while the research question focuses on the overall ESG rating and not just the environmental component. This might explain why the companies selected for the high polluting group experienced significant effects as expected in contrast to the companies in the other groups.
Moreover, the cumulative average abnormal returns of the low ESG rated
21 companies from China and the high ESG
rated companies from Europe are also significant, but in the opposite direction as expected. Another reason stated in the literature review is that the institutional investors do possess very little ‘bad’ stocks, so they do not go through the ‘switch and masse’ process. The weight of the trading volume that the individual investors produce would therefore not be enough to produce significant abnormal returns. The last reason stated is that investors already expected the policy and according to the efficient market hypothesis therefore already priced that information in. However, this seems to be rejected by the results from testing hypothesis 1.
Further research must look into the difference between stock returns for Chinese companies and from those of European companies around the time of the 21st conference of the parties of the
UNFCCC in which the Paris Agreement was announced. It must be checked in more detail if any China specific or Europe specific events occurred which might have been of influence on the returns of those stocks. If so, this effect must be incorporated in the models to exclude it from the possible effect the announcement of the Paris Agreement might have had. This paper uses an
estimation window of only 30 days. The abnormal returns are therefore very vulnerable to such specific events and this might thus be a drawback for this paper. Furthermore, it might be of importance to either select companies based on their overall ESG rating instead of only the environmental component or specify the research to the specific components.
7. Conclusion
The objective of this paper is to discover if environmental policy announcements affect the stock returns of both low- and high ESG rated companies, and if that effect differs between companies based in China or Europe. In this paper, a specific study is done to the effect of the announcement of the Paris Agreement. In this environmental policy, agreements are made to keep temperature increase below 1.5˚C above pre-industrial levels. To achieve this, companies must produce in a more environmental friendly way. For this reason, companies might have to adjust their production system. This might lead to higher costs and hence lower profits. This may cause investors to sell the stocks whom lose their profitability the most and buy the stocks
22 whom are affected the least. To check for
this effect, this paper studied the cumulative average abnormal returns of low- and high ESG rated companies from both Europe and China. Consequently, I regressed the cumulative abnormal returns on two different dummy variables, HighESG and China. The results of the first regression did not confirm the expectations, therefore I do not conclude that being a high ESG rated company caused higher abnormal returns. However, Chinese companies experienced higher cumulative abnormal returns.
Improvements for the research on this topic are given in the discussion section. The most important, but also the most difficult one being the inclusion of exogenous effects on the abnormal returns of the companies studied in this paper. Altogether, the effect of the announcement of the Paris Agreement seems not to be different for low- and high ESG rated companies. Nevertheless, the announcement seems to have had a different effect on Chinese based companies than on European based companies.
References
BASSI, S., BOYD, R., BUCKLE, S., FENNELL, P., MAC DOWELL, N. MAKUCH, Z., STAFFELL, I. (2015). Bridging the gap:
improving the economic and policy framework for carbon capture and storage in the European Union. Policy Brief, June 2015. Centre for Climate Change Economics and Policy and Gratham Institute.
BATTEN, S. (2018). Climate Change and the Macro-Economy: A Critical Review. SSRN
Electronic Journal,
29-30.https://doi.org/10.2139 /ssrn.3104554
BATTEN, S., SOWERBUTTS, R., & TANAKA, M. (2016). Let’s Talk About the Weather: The Impact of Climate Change on Central Banks. SSRN Electronic Journal, 6–23.
https://doi.org/10.2139/ssrn.27 83753
BEAL, D., GOYEN, M. (1998). Putting your money where your mouth is, a profile of ethical investors. Financial Services Review 7, 129-142.
23 CLOGG, C. C., PETKOVA, E., & HARITOU, A.
(1995). Statistical Methods for
Comparing Regression
Coefficients Between
Models. American Journal of
Sociology, 100(5), 1261–1293. https://doi.org/10.1086/230638 DERWALL, J., KOEDIJK, K., & TER
HORST, J. (2011). A tale of values-driven and profit-seeking social investors. Journal of Banking &
Finance, 35(8), 2137–2147. https://doi.org/10.1016/j.jbankfi n.2011.01.009
DONG, C., DONG, X., JIANG, Q., DONG, K., & LIU, G. (2018). What is the probability of achieving the carbon dioxide emission targets of the Paris Agreement? Evidence from the top ten emitters. Science
of The Total Environment, 622– 623, 1294–1303. https://doi.org/10.1016/j.scitote nv.2017.12.093
DOWELL, G., HART, S., & YEUNG, B. (2000). Do Corporate Global Environmental Standards Create
or Destroy Market
Value? Management
Science, 46(8), 1059–1074. https://doi.org/10.1287/mnsc.4 6.8.1059.12030
FELDMAN, S., P. SOYKA, P. AMEER. (1996). Does Improving a Firm’s
Environmental Management System and Environmental Performance Result in a Higher Stock Price? ICF Kaiser, Washington.
FREEMAN, R. E., & LIEDTKA, J. (1991). Corporate social responsibility: A critical approach. Business Horizons, 34(4), 92–98. https://doi.org/10.1016/0007-6813(91)90012-k HAMILTON, J. T. (1995). Pollution as News: Media and Stock Market Reactions to the Toxics Release Inventory Data. Journal of Environmental Economics and Management, 28(1), 98–113. https://doi.org/10.1006/jeem.19 95.1007
HONG, H., & KACPERCZYK, M. (2009). The price of sin: The effects of social norms on markets. Journal of Financial Economics, 93(1), 15– 36. https://doi.org/10.1016/j.jfineco .2008.09.001
KARPOFF, J.H., LOTT JR, J.E., WEHRLY, E.W. (2005). The
24 reputational penalties for
environmental violations: empirical evidence. Journal of Law and Economics, 48, 653-675.
KLASSEN, R. D., & MCLAUGHLIN, C. P. (1996). The Impact of Environmental Management on Firm Performance. Management
Science, 42(8), 1199–1214. https://doi.org/10.1287/mnsc.4 2.8.1199
RAMIAH, V., MARTIN, B., & MOOSA, I. (2013). How does the stock market react to the announcement of green policies? Journal of Banking &
Finance, 37(5), 1747–1758. https://doi.org/10.1016/j.jbankfi n.2013.01.012
RAMIAH, V., PICHELLI, J., & MOOSA, I. (2014). Environmental regulation, the Obama effect and the stock market: some empirical results. Applied Economics, 47(7), 725–738.
https://doi.org/10.1080/000368 46.2014.980572
SCHLEUSSNER, C.-F., ROGELJ, J., SCHAEFFER, M., LISSNER, T., LICKER, R., FISCHER, E. M., HARE,
W. (2016). Nature Climate
Change, 6(9), 827–835. https://doi.org/10.1038/nclimat e3096
SCHREURS, M. A. (2016). The Paris Climate Agreement and the Three Largest Emitters: China, the United States, and the European
Union. Politics and
Governance, 4(3), 219. https://doi.org/10.17645/pag.v4 i3.666
SCHWARTZ, M. S., & SAIIA, D. (2012). Should Firms Go “Beyond Profits”? Milton Friedman versus Broad CSR1. Business and Society
Review, 117(1), 1–31. https://doi.org/10.1111/j.1467-8594.2011.00397.x
SEO, S. N. (2017). Beyond the Paris Agreement: Climate change policy negotiations and future directions. Regional Science Policy
& Practice, 9(2), 121–140. https://doi.org/10.1111/rsp3.12 090 SHANE, P.B., SPICER, B.H. (1983). Market response to environmental information produced outside the firm. The
25 STATMAN, M. (2005). The Religions
of Social Responsibility. The
Journal of Investing, 14(3),
14–21.
https://doi.org/10.3905/joi.200 5.580542
URPELAINEN, J., & VAN DE GRAAF, T. (2017). United States non-cooperation and the Paris agreement. Climate Policy, 18(7), 839–851.
https://doi.org/10.1080/146930 62.2017.1406843
