1 s2.0 S0921800922000891 Main

Ecological Economics 197 (2022) 107427
Contents lists available at ScienceDirect
Ecological Economics
journal homepage: www.elsevier.com/locate/ecolecon
The environmental and financial performance of green energy investments:

European evidence
Maria Céu Cortez a, *, Nuno Andrade b, Florinda Silva a
a
NIPE, University of Minho, Portugal
b
University of Minho, Portugal
A R T I C L E I N F O A B S T R A C T
JEL classification: This paper investigates the environmental and financial performance of investments in energy firms. For this
G11 purpose, we form portfolios of green energy European stocks compared to their non-green counterparts from
M14 January 2008 to November 2020 and assess their environmental and financial performance. Within firms with
Keywords: environmental ratings, those that are green perform better in environmental terms than their non-green coun
Environmental performance terparts, although the difference has narrowed in recent years. Regarding financial performance, our results show
Financial performance
that, in general, the green energy portfolio outperforms the market. Furthermore, within the energy sector, the
Green energy
Green finance
green energy portfolio performs better than its non-green counterpart. Moreover, we find that the out
Sustainable investments performance of the green portfolio is mainly due to a performance improvement in most recent years. Overall,
our results show that over this period investments in green energy firms perform at least as well as their non-
green energy counterparts.
1. Introduction the key role that financial markets play in transitioning to low carbon
economy (Scholtens, 2017).3 Financial institutions have also become
The influence of human activity on the climatic system and the more sensitive to climate-related issues, as reflected in the increasing
impact of climate on humanity are one of mankind’s main challenges.1 number of institutional investors committing to the UN Principles for
The main recommendations to deal with the adverse effects of global Responsible Investing, which encourage the integration of Environ
warming involve not only the implementation of energy efficiency and mental, Social and Governance (ESG) criteria and, particularly, climate
conservation practices but also replacing fossil fuels with low-carbon change issues into the investment process.4 This commitment of insti
and cleaner sources of energy (Ripple, Wolf, Newsome, Barnard, and tutional investors is not only a response to increasing public scrutiny
Moomaw, 2020). As such, there is an increased societal and political regarding the carbon footprint of their portfolios (Boermans and
awareness that transitioning from carbon-intensive fossil fuels to Galema, 2019) but also a risk management mechanism to address the
renewable and clean energy sources represents an important step to financial implications of climate risks (Krueger, Sautner, and Starks,
achieve the goals of reducing greenhouse gas emissions and converge to 2020).
a sustainable economy. In this regard, several governments and inter From the demand side, there is evidence that investors have been
governmental entities have been bringing together efforts to implement favoring green investments (e.g., Ammann, Bauer, Fischer, and Müller,
energy policies that promote investments in renewable and green en 2019; Hartzmark and Sussman, 2019) and, namely, those with a low
ergy.2 Following this trend, there has been an increasing recognition of carbon footprint (Ceccarelli, Ramelli, and Wagner, 2021; Humphrey and
* Corresponding author at: Universidade do Minho, School of Economics and Management, Gualtar 4710-057, Braga, Portugal.
E-mail address: [email protected] (M.C. Cortez).
1
In a paper signed by over 13,700 scientists from over 156 countries, Ripple, Wolf, Newsome, Barnard, and Moomaw (2020) declare that the world is facing a
climate emergency and warn that climate change is accelerating faster than expected, posing a major threat to global environmental sustainability.
2
The 2015 Paris Agreement on climate change and the Sustainable Development Goals, sponsored by the United Nations, are examples of major initiatives of the
international community to mitigate climate change by committing to carbon emission reduction targets.
3
To this end, for instance, the European Union has established an ambitious agenda on sustainable finance and set an action plan to connect finance with
sustainability.
4
https://www.unpri.org/climate-change.
https://doi.org/10.1016/j.ecolecon.2022.107427
Received 19 July 2021; Received in revised form 7 March 2022; Accepted 24 March 2022
Available online 27 April 2022
0921-8009/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
M.C. Cortez et al. Ecological Economics 197 (2022) 107427
Li, 2021). The intensifying divestment campaigns since 2011, which the recent decade, with increasing societal pressure and efforts of the
have put pressure on investors to move out of fossil fuel firms (Hal international community to address climate change, accompanied by
coussis and Lowenberg, 2019), are also a manifestation of investment significant advances in the renewable energy technology. From the de
preferences towards low-carbon investments. The popularity of envi mand side, financial markets are witnessing a shift in investors tastes
ronmentally friendly energy alternatives is, therefore, on the rise and the towards green assets. Motivated by these developments, we might
green energy sector shows an enormous growth potential (Ng and expect to observe changes in the performance of green energy in
Zheng, 2018). As a matter of fact, although more than 80% of the total vestments in more recent years. In this context, we further contribute by
global primary energy demand is still satisfied by coal, oil, and gas, analyzing the evolution of the performance of green investments over
renewable energy is the fastest-growing source of energy globally time.
(IRENA- International Renewable Energy Agency, 2020). Moreover, The remainder of the paper is organized as follows. Section 2 dis
despite the substantial decline in global energy demand following the cusses the theoretical arguments and empirical evidence regarding the
global coronavirus pandemic that unfolded in December 2019, leading financial impact of investing with environmental criteria, with a
many countries to implement lockdown and confinement measures, particular emphasis on green versus non-green energy investments.
renewable energy has so far been the source of energy most resilient to Sections 3 and 4 present the methods and data used in this research. The
the Covid-19 crisis (IEA- International Energy Agency, 2021). analysis and discussion of the empirical results are presented in section
Motivated by the growth of green investing, we specifically focus on 5, and section 6 concludes.
the financial impact of investing in green firms versus non-green firms in
the energy sector. This question is particularly relevant for energy in 2. Literature review
vestors, considering that this sector is at a crossroad, transitioning from
technologies that are heavily based on fossil fuels towards new and One of the arguments in the debate relating corporate environmental
greener ones, posing a higher uncertainty for investors’ capital alloca performance to financial performance is that firms that increase their
tions to different energy asset classes (Dragomirescu-Gaina, Galariotis, environmental performance must bear the costs of doing so at the
and Philippas, 2021). Companies operating in the energy sector, by expense of a diminished financial performance (Walley and Whitehead,
nature, have high environmental exposure (Gonenc and Scholtens, 1994). In contrast, another argument suggests that increasing environ
2017) and thus face an intense scrutiny regarding their environmental mental performance can result in a competitive advantage and lower
impact. Even companies that are involved in more controversial or costs for the firms (Freeman and Evan, 1990; Porter and van der Linde,
environmentally-sensitive sources of energy seek to incorporate envi 1995; Ambec and Lanoie, 2008). A vast literature explores the rela
ronmental practices in their strategy as a way to seek legitimation and tionship between corporate environmental and financial performance,
minimization of environmental risks and reputational damages (Kolk and despite the mixed evidence, several review studies show that com
and Levy, 2001; Du and Vieira, 2012; Agudelo, Johannsdottir, and panies with higher environmental performance tend to have higher
Davidsdottir, 2020).5 In this context, for investors wishing to go green financial performance (Molina-Azorín, Claver-Cortés, López-Gamero,
and support companies that proactively contribute to climate change and Tarí, 2009; Albertini, 2013; Dixon-Fowler, Slater, Johnson, Ell
mitigation, it is not only important to consider the extent to which strand, and Romi, 2013; Endrikat, Guenther, and Hoppe, 2014;
companies are engaged in environmental practices, but also to distin Guenther and Hoppe, 2014; Galama and Scholtens, 2021).
guish companies that supply fossil fuel energy from those that are Considering the important role of the energy sector in the economy, a
committed to alternative and sustainable sources of energy. set of studies has addressed the link between corporate environmental
So far, the financial implications of investing in these two different and financial performance in energy firms. One can argue that the
segments are still unclear. Several studies address the effects of investing higher costs associated to the adoption of new and disruptive technol
in clean or alternative energy firms by evaluating the performance of ogies may lower corporate profitability (Anderloni and Tanda, 2017).
actively managed renewable energy mutual funds (Reboredo, Quintela, Nevertheless, the progress in renewable energy technology has trans
and Otero, 2017; Martí-Ballester, 2019a, 2019b), ETFs (Alexopoulos, lated into lower costs and higher competitiveness of these sources of
2018; Miralles-Quirós and Miralles-Quirós, 2019) and market indices power over time (Popp, Hascic, and Medhi, 2011). Furthermore, it is
(Bohl, Kaufmann, and Stephan, 2013; Rezec and Scholtens, 2017). An important to consider that investment in green technologies may lead to
alternative methodological approach, which overcomes some of the competitive advantages as a result of a more efficient use of resources
limitations of evaluating actual portfolios (e.g., the effects of fund (Porter and van der Linde, 1995; Ambec and Lanoie, 2008). At the
managers’ skills in performance), relies on forming synthetic mutually empirical level, several event studies investigate the informational ef
exclusive portfolios formed on specific environmental characteristics ficiency of stock prices in the industry by analyzing the effects of envi
and evaluating their financial performance. Although this approach is ronmental incidents on stock prices of oil firms (e.g., Patten and Nance,
well established in the broader literature on the performance of socially/ 1998; Capelle-Blancard and Laguna, 2010).6 Yet, although event studies
environmentally screened portfolios, studies that apply it to assess the can determine a causal relationship in the days that follow an environ
performance of green energy-related stocks are scarce. An exception is mental event, it is difficult to evaluate its long-term impact (Ruggiero
Ng and Zheng (2018), who form portfolios of US green and non-green and Lehkonen, 2017). An alternative approach to explore the relation
energy stocks and show that investors do not bear a cost from holding ship between environmental performance and financial performance
portfolios of green energy firms compared to portfolios of non-green
firms. Regarding Europe, this issue is unexplored. This research con
tributes to fill this gap by forming portfolios of European green and non-
6
green energy stocks and assessing their environmental and financial These studies analyze the stock market reaction to catastrophes such as oil
performance, with the research questions lying on: how do green energy spills (e.g., the Exxon Valdez oil spill) and, in general, document negative
stock portfolios perform compared to non-green portfolios and how does abnormal returns of the companies in the period immediately following the
portfolio financial performance relate to its environmental perfor event. Remarkably, Patten and Nance (1998) observe a positive effect in other
companies in the industry following the 1989 Exxon Valdez oil spill, as there
mance? The energy sector has experienced important transformations in
was a significant price increase in petroleum products in the aftermath of the
accident. Furthermore, there is also evidence that the inclusion of energy firms
in sustainability indices has a negative short-term impact in its market value,
5
For instance, as argued by Hang, Geyer-Klingeberg, Rathgeber, and Stöckl possibly reflecting investors’ perception that commitment to sustainability
(2018), firms in dirty industries can more effectively deal with environmental strategies imply additional costs to the firms (Schmutz, Tehrani, Fulton, and
pollution. Rathgeber, 2020).
2
involves regression analysis. Within this approach, Sueyoshi and Goto on renewable energy ranges from neutral (Bohl, Kaufmann, and Ste
(2009), Clarkson, Li, Richardson, and Vasvari (2011) and Ruggiero and phan, 2013) to negative (Rezec and Scholtens, 2017) compared to their
Lehkonen (2017) investigate how environmental performance impacts conventional counterparts. Regarding ETFs, while Alexopoulos (2018)
financial performance of companies in the energy sector, but the results shows that investing in ETFs of clean energy companies is not the best
are mixed.7 Finally, within the IPO literature, Anderloni and Tanda strategy compared to investing in an aggregate portfolio that includes
(2017) investigate the underpricing and long-term performance of IPOs both clean and conventional energy ETFs, Miralles-Quirós and Miralles-
of European green energy companies from 2000 to 2014 and find that Quirós (2019) provide evidence that alternative energy ETFs outperform
the financial performance of green energy companies does not differ other energy ETFs. However, assessing the relationship between envi
from that of traditional energy ones. ronmental and financial performance by evaluating the performance of
So far, the financial implications of investing in green energy firms mutual funds, ETFs or market indices in the energy sector has several
are unclear. A contribution to the debate on the financial performance of limitations. First, actively managed mutual fund and ETF returns are
green versus non-green energy investments stems from Merton’s (1987) affected by fund management skills (Kempf and Osthoff, 2007; Ruggiero
incomplete information model and the concept of market segmentation and Lehkonen, 2017) and management fees (Auer, 2016), making it
and is related to the effects of an increasing demand on stock prices. A difficult to disentangle the performance associated to the social/green
substantial number of investors seeking stocks in non-controversial characteristics of the stocks. Another limitation of these studies,
sectors would lead to a relative overpricing of green energy stocks including those evaluating the performance of energy indices, is that it is
compared to their non-green peers (Boermans and Galema, 2019), difficult to establish whether a premium exists, because holdings of
penalizing the performance of green investments. On the other hand, green funds and conventional funds are not mutually exclusive (Derwall,
another argument points out that investors holding fossil fuel companies Guenster, Bauer, and Koedijk, 2005). Finally, there is evidence sug
are substantially exposed to environmental and regulatory risks (Bat gesting that funds that self-designate themselves as socially or envi
tiston, Mandel, Monasterolo, Schütze, and Visentin, 2017; Trinks, ronmentally responsible may not fully comply with the ESG standards
Scholtens, Mulder, and Dam, 2018; Boermans and Galema, 2019; Bolton they claim to adhere to (Wimmer, 2013; Auer, 2016), so the financial
and Kacperczyk, 2020). In particular, investments in fossil fuels are at performance of these funds may not necessarily reflect their environ
risk of losing value due to the possibility of these reserves becoming mental features.
stranded assets as a result of policies that enforce reduction of emissions To overcome these limitations, an alternative approach to evaluate
(Griffin, Jaffe, Lont, and Dominguez-Faus, 2015; Andersson, Bolton, and the effects of investing in green versus non-green companies consists in
Samama, 2016). Furthermore, they are also subject to technological risk forming mutually exclusive portfolios based on different environmental
from innovations in alternative and more affordable sources of energy characteristics and investigating the portfolios’ return differences.
(Bolton and Kacperczyk, 2020). The price pressure on green stocks and Although this synthetic portfolio approach represents a large body of the
the returns associated to climate risks and other related ESG risks is also broader literature on the effects of investing with social/environmental
discussed by Pástor, Stambaugh, and Taylor (2021a, 2021b), who criteria,9 with exception of Ng and Zheng (2018), it has hardly been
develop an equilibrium model in which investor’s tastes for green assets applied in the context of the energy sector. Ng and Zheng (2018) form
explain their lower expected returns compared to brown assets, in the portfolios of US green and non-green energy stocks from 1990 to 2013
spirit of Fama and French (2007). However, it is possible for green assets and document that clean energy stocks perform at least as well as its
to benefit from higher returns if investors’ tastes towards green unex matching non-green portfolio and the benchmark. As a contribution to
pectedly increases, while the market portfolio adjusts to this shift. the literature, we explore this issue in the European market by forming
Another theoretical contribution to the discussion on investor tastes and portfolios of green versus non-green energy companies and assessing
price effects of sustainable investing is provided by Pedersen, Fitzgib their environmental and financial performance.
bons, and Pomorski (2021), who derive the ESG-efficient frontier in
which the investors’ asset allocation problem is explained as a trade-off 3. Methods
between the risk-adjusted return and the level of ESG. The authors as
sume the existence of heterogeneous investors regarding ESG prefer As mentioned previously, we follow the synthetic portfolio approach
ences and show that investors with ESG preferences can outperform to assess the impact of investing in green firms. As such, within the
other investors if the ESG factors contain security-relevant information European market, we form a value-weighted portfolio of green energy
that has not yet been incorporated by the market. companies and a value-weighted portfolio of non-green energy com
Empirically, several studies address the effects of investing in stocks panies. We also form a difference portfolio that reflects a strategy of
of renewable energy companies by evaluating the performance of assuming a long position in the green portfolio and a short position in
actively managed renewable energy funds (Reboredo, Quintela, and the non-green portfolio. We further form portfolios comprised of green
Otero, 2017; Martí-Ballester, 2019a, 2019b),8 market indices (Rezec and and non-green stocks with environmental ratings. We should expect
Scholtens, 2017; Bohl, Kaufmann, and Stephan, 2013) and ETFs (Alex higher environmental ratings of the green portfolio compared to that of
opoulos, 2018; Miralles-Quirós and Miralles-Quirós, 2019). Reboredo, the non-green portfolio, and thus focusing only on environmentally
Quintela, and Otero (2017) and Martí-Ballester (2019a, 2019b) show rated firms may provide additional insights concerning the relationship
that the performance of renewable energy funds is either similar or between environmental and financial performance.
worse than that of their black or conventional peers or the market. Assuming an investor’s perspective, we evaluate the financial per
Likewise, the evidence on the performance of market indices screened formance of portfolios using risk-adjusted returns based on the Carhart
(1997) and the Fama and French (2015) models, which include factors
that have been shown in the asset pricing literature to explain the cross-
7
section of equity returns of a wide variety of portfolios. The Carhart
The mixed findings are consistent with the evidence on the broader debate
(1997) four-factor model relates portfolio excess returns to the market,
on the relationship between corporate social/environmental performance and
size, and value factors, as in Fama and French (1993), further adding the
financial performance. Explanations for these inconsistent results include issues
related to the use of different measurement and theoretical approaches
momentum factor. Besides being recognized as a standard model in the
(Guenther and Hoppe, 2014). performance evaluation literature, previous findings suggest that the
8
These studies are related to a broader stream of the literature that focuses
on the performance of environmental or green mutual funds compared to other
9
funds (Climent and Soriano, 2011; Muñoz, Vargas, and Marco, 2014; Ji, Zhang, Some examples are Derwall, Guenster, Bauer, and Koedijk (2005), Kempf
Mirza, Umar, and Rizvi, 2021). and Osthoff (2007), Statman and Glushkov (2009), and Auer (2016).
3
size and value factors play a significant role in explaining the perfor green and non-green energy firms to be included in the dataset, we used
mance of green and/or fossil fuel portfolios (Ibikunle and Steffen, 2017; the Refinitiv Eikon database together with the altenergystocks.com
Ng and Zheng, 2018; Trinks, Scholtens, Mulder, and Dam, 2018). The website.12 Starting with Refinitiv Eikon, we selected the Energy sector
four-factor model is expressed as follows: and downloaded all the constituents from the “Energy - Fossil Fuel” and
“Renewable Energy” industries for European markets. The former in
rp,t = αp + β1 rm,t + β2 SMBt + β3 HMLt + β4 MOM t + εp,t (1)
cludes firms engaged in coal, exploration and production of oil and gas,
and related equipment and services, which we classify as non-green,
where rp, t is the excess return of portfolio p in period t; αp is the portfolio
whereas the latter includes firms engaged in the production of renew
performance measure, rm, t is the market excess return in period t, SMBt,
able fuels and renewable energy equipment and services, which we
HMLt, and MOMt represent the size, value and momentum factors in
classify as green. This resulted in an initial dataset of 387 non-green and
period t and εp, t is the error term. The size factor is defined as the dif
73 green firms, including surviving and non-surviving firms. The
ference between the return on a portfolio of small-cap stocks and a
altenergystocks.com website was then used to complement the green
portfolio of large cap stocks; the value factor is the difference between
portfolio (selecting only European firms and excluding trust funds and
the return on a portfolio of high book-to-market stocks and a portfolio of
ETFs). We checked all the equity lists available on the website for
low book-to-market stocks; and the momentum factor is measured by
additional green firms. This search increased the number of green firms
the difference between the return on a portfolio of stocks with high prior
to 125 (52 more). We further checked the industry classification as well
returns and a portfolio of low prior return stocks.
as the currency of the stocks and excluded those that although related to
Motivated by the proliferation of factors that have been proposed in
the energy sector were classified as financials and those quoted in
the asset pricing theory literature10 and the discussion around the model
USD.13 We then collected the monthly market value and return index
that provides the ‘best’ fit to a cross-section of equity returns, we also
data of the stocks over the period January 2008 to November 2020.14 As
make use of a more recent contribution - the Fama and French (2015)
for some firms the required financial data were not available, we ended
five-factor model, which adds two additional sources of systematic risk
up with the final dataset of 113 green energy and 278 non-green firms.
to the three-factor model - the profitability and investment risk factors.
The firms in the green portfolio are mainly located in five countries -
Although the implementation of this model is not so generalized as the
Germany, France, Sweden, United Kingdom and Italy, with the main
four-factor model, the inclusion of the profitability and investment
industries being Renewable Energy Equipment, Alternative Fuels,
factors has been shown to be relevant to mitigate the performance bias
Electrical Components and Alternative Electricity. Non-green firms are
associated to the low volatility anomaly (Jordan and Riley, 2015). The
mostly located in the United Kingdom, Norway, Russia, Romania,
use of the five-factor model in our research is further motivated by
France and Poland, with their main industries relating to Oil, Oil
Plantinga and Scholtens (2021), who show that fossil fuel stocks are
Equipment, Oil Refining, Coal, Offshore Drill & Services and Marine
more exposed to size, profitability and investment factors than non-fossil
Transportation.
fuel stocks.11 The five-factor model is expressed as:
From this dataset we formed two value-weighted portfolios: the
rp,t = αp + β1 rm,t + β2 SMBt + β3 HMLt + β4 RMW t + β5 CMAt + εp,t (2) green and the non-green energy portfolios. Excess returns are computed
using the 1-month Euribor rate as the risk-free rate. Table 1 presents the
where RMWt and CMAt represent the profitability and investment fac descriptive statistics of monthly excess returns and some characteristics
tors in period t. The profitability factor is defined as the difference in of the green and non-green portfolios, and the difference portfolio. The
return between a portfolio of highly profitable firm and low profitable green portfolio presents a higher mean return and a lower standard
firms. The investment factor is the difference in return between a port deviation when compared to the non-green portfolio. The difference
folio of firms with a low level of investment activity including divesting portfolio presents a positive mean return, although it is not statistically
firms and a portfolio of firms with a high level of investment activity. significant. Based on the Jarque-Bera test, we reject the hypothesis of a
We apply these models to the green and non-green portfolios as well normal distribution of the times series of monthly excess return for both
as to a difference portfolio, which is formed by subtracting the green the green and non-green portfolios but not for the difference portfolio.15
energy portfolio returns from the non-green energy portfolio returns. By In terms of firms’ age and market value, the two portfolios present
evaluating the difference portfolio, we are able to analyze the differ similar characteristics, with an average age of 16 years and an average
ences in performance and risk between the two investment approaches market value of around 3232 million euros for the firms in the green
and test its statistical significance. portfolio and an average age of 15.3 years and an average market value
of 3429 million euros for the firms in the non-green portfolios.
4. Data
Green energy stocks are defined as firms that produce energy directly
from environmentally friendly sources (e.g., solar, wind, biodiesel, etc.)
and firms closely related to energy efficiency (e.g., electricity storage, 12
The altenergystocks.com website, also used by Ng and Zheng (2018),
smart grid, clean transportation). In order to identify the European
identifies green energy stocks by listing firms that produce energy from alter
native or environmentally friendly sources as well as those that are involved in
the development of technologies for the production of green energy (Ng and
Zheng, 2018).
13
As our objective is to analyze European firms, we excluded those quoted in
10
In what Cochrane (2011) calls the ‘zoo of factors’. USD. Most of the excluded firms do not have information on the domiciled
11
Plantinga and Scholtens (2021) dropped the value (HML) factor after per country or, despite appearing as domiciled in a European country, they are
forming a factor redundancy test, in line with Fama and French (2015), who offshore firms.
14
conclude that HML is a redundant factor as its average return is fully captured The beginning of the sample period is constrained by the coverage of
by its exposures to the other risk factors and especially to the profitability and Refinitiv ESG ratings, which started in 2002. The coverage was somewhat
investment factors. Fama and French (2015) argue that in applications where limited in the early years, but it gradually increased over time. By starting in
the purpose is to estimate abnormal returns a four-factor model without HML 2008, we avoid the portfolio diversification restrictions associated to a very
performs as well as the five-factor model, but the five-factor model should be limited number of rated firms.
15
the choice if we are also interested in portfolio style tilts towards size, value/ As a robustness test, we also apply conditional models of performance
growth, profitability, and investment premiums. In our research, as we are also evaluation, that somewhat deal with the non-normality problem, as demon
interested in analyzing the factors coefficients we maintain the HML factor. strated by Adcock, Cortez, Armada, and Silva (2012).
4
Table 1 Environment pillar score

- Descriptive statistics of monthly excess returns and stock characteristics of the
green and non-green portfolios. 80.00
Descriptive statistics Green Non-Green Difference 70.00
Mean (%) 0.914 0.526 0.388 60.00

Median (%) 1.531 0.896 0.740
50.00
Maximum (%) 18.809 28.784 14.999
Minimum (%) − 21.320 − 18.839 − 13.754 40.00
Standard Deviation (%) 5.478 6.317 4.867
30.00
Skewness − 0.688 0.217 − 0.097
Kurtosis 5.173 5.467 3.493 20.00
JB test 42.728 40.521 1.810
10.00
P-value 0.000 0.000 0.405
Monthly observations 155 155 155 0.00
Average firm age (years) 15.997 15.323 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Average firm market value (million €) 3231.538 3429.349
Green Non-Green
This table presents the descriptive statistics of monthly excess returns of the
green and non-green value-weighted portfolios and their difference over the
Fig. 1. - Evolution of the Environmental pillar scores for the green and non-
period January 2008 to November 2020. The green and non-green portfolios are
green portfolios.
composed of 113 green and 278 non-green firms, respectively. P-value is the
This figure shows the mean Environmental pillar score of the portfolios of green
probability of the Jarque-Bera normality test. The average firm age, at the end of
and non-green rated firms over the period 2008 to 2019. The green and non-
November 2020, and the average market value over the period under analysis
green portfolios comprise 50 and 77 rated firms, respectively. (For interpreta
are also reported.
tion of the references to colour in this figure legend, the reader is referred to the
web version of this article.)
After checking which green and non-green firms in the dataset are
rated by Refinitiv ESG,16 we further formed green and non-green port
folios comprised of only those firms that have environmental scores Table 2
available (50 and 77 firms, respectively). Fig. 1 plots the evolution over Mean Environmental pillar scores of the green and non-green portfolios.
time of the average Environmental pillar score of the two portfolios.17 Environmental scores Green Non-green Difference
We can see that the green portfolio’s annual average score is higher than
Environmental pillar score 65.012 52.609 12.403***
that of the non-green portfolio, with the average difference being sta Emissions score 64.660 64.030 0.630
tistically significant at the 1% level, as we can observe in Table 2. Resources Use score 68.739 61.466 7.272***
However, in the last years there is a decrease on the average of the Environmental Innovation score 60.169 19.347 40.822***
Environmental pillar score of the green portfolio and, consequently, a Number of firms 50 77
decrease on the difference between the green and non-green portfolios. This table shows the aggregate mean Environmental pillar score as well as its
Potential explanations for the narrowing of this difference may be the dimensions, the Emissions score, the Resources Use score, and the Environ
increasing regulatory and other stakeholder pressures for high polluting mental Innovation score of the green, non-green, and the difference portfolios
firms to implement environmentally friendly practices (Cadez, Czerny, over the period 2008 to 2019. The green and non-green portfolios are composed
and Letmathe, 2019), and also the increasing number of rated firms. of 50 and 77 rated firms, respectively. The significance of the difference port
Looking at the different dimensions of the Environmental pillar score, folios is based on a two-sample t-test for mean differences. ***, **, and * denote
statistical significance at the 1%, 5%, and 10% levels.
namely the Emissions, Resource Use, and Environmental Innovation
scores, we conclude that the major difference between the two portfolios
comes from the Environmental Innovation score, for which the green likely to be.
portfolio presents a clearly higher score compared to the non-green Table 3 presents the descriptive statistics for the rated portfolios and
portfolio. The emissions score is similar for the two portfolios and it is the difference portfolio. We observe that the rated green portfolio con
the only dimension for which the difference is not statistically signifi tinues to have a higher mean return and a slightly lower standard
cant. Although this may be surprising, it may reflect the more stringent
Table 3
environmental regulations for high polluting firms, especially those that
- Descriptive statistics of monthly excess returns and stock characteristics of the
are larger and more subject to public scrutiny, as rated firms are more green and non-green portfolios considering only the environmental rated firms.
Descriptive Statistics Green Non-Green Difference
16 Mean 0.721 0.279 0.442

The ESG scores from Refinitiv provide information on firms’ ratings in
Median 1.269 0.646 0.899
terms of several dimensions of corporate social responsibility, namely on the
Maximum 18.606 13.833 15.790
Environmental, Social and Governance pillars. The overall rating of the Envi Minimum − 21.166 − 16.503 − 9.475
ronmental pillar results from the aggregation of the scores in its three cate Stand. Deviation 5.313 5.696 4.498
gories: Emissions score, Resource Use score and Environmental Innovation Skewness − 0.760 − 0.233 0.006
score. The emissions score “measures a company’s commitment and effective Kurtosis 5.350 3.428 3.471
ness towards reducing environmental emissions in its production and opera JB test 46.987 2.404 1.331
tional processes.” The resource use score “reflects a company’s performance P-Value 0.000 0.301 0.514
and capacity to reduce the use of materials, energy or water, and to find more Monthly observations 144 144 144
Average firm age (years) 22.224 21.079
eco-efficient solutions by improving supply chain management”. Finally, the
Average firm market value (million €) 6888.958 12,162.407
innovation score “reflects a company’s capacity to reduce the environmental
costs and burdens for its customers, thereby creating new market opportunities This table presents the descriptive statistics of monthly excess returns of the
through new environmental technologies and processes, or eco-designed green and non-green value-weighted portfolios, comprising only the environ
products.” (Refinitiv, 2021). mental rated firms, and their difference from January 2008 to December 2019.
17
The data on these scores were downloaded from Refinitiv Eikon Datastream The green and non-green portfolios are composed of 50 and 77 rated firms,
on 19th January 2021. Due to the fact that the majority of the companies did respectively. P-value is the probability of the Jarque-Bera normality test. The
not have the scores of 2020 available yet, we do not include this year in the average firm age, at the end of December 2019, and the average market value
analysis. over the period under analysis are also reported.
5
deviation compared to the rated non-green portfolio. The difference to-market firms) and with higher profitability. Plantinga and Scholtens
portfolio presents a positive mean return that is higher than in the case (2021) also document the exposure of the fossil fuel portfolio to the
of portfolios of all firms, but again it is not statistically significant. As for profitability factor. Thus, these results suggest that the profitability risk
the normality test, we only reject the hypothesis of the monthly excess factor is relevant, in particular for the evaluation of non-green
returns being normally distributed for the rated green portfolio. As ex portfolios.
pected, the average age and market value of the firms in the two rated It is worth noting that the explanatory power of the model depends
portfolios are clearly higher than those considering all the firms. How on the index used to measure the market factor. Although the Energy
ever, while the average age of the rated green and non-green firms index better explains the excess returns of the non-green portfolio, its
continues to be similar, the average market value of the rated non-green ability to explain the excess returns of the green portfolio is quite
firms is substantially higher, with a value of 12,162 million euros limited. Furthermore, when the Energy index is used, the market beta
compared to 6889 million euros. estimates of the green portfolio are quite low (0.358 and 0.325,
Regarding the risk factors, we use the excess returns on both a respectively, using the four- and five-factor model).19 As such, and
specialized and a general stock market index as alternative proxies for consistent with the literature, our analysis focuses mainly on the risk
the market factor. The specialized benchmark is proxied by the MSCI factor loadings estimates obtained with the general market index.
Europe Energy index. The return index data of this index was also
collected from Refinitiv Eikon Datastream. As the general market 5.2. Performance of green vs non-green portfolios of environmentally
benchmark, we use the European market return series from the Professor rated stocks
Kenneth French’s data library18 converted to euros. To compute the
excess returns of the market indices, we use the 1-month Euribor rate. In this section, we focus the analysis on the firms of the dataset that
The other risk factors (size, value, momentum, profitability and in are ESG rated and have information available for the Environmental
vestment factors) for the European market were also collected from pillar score. This analysis thus excludes firms that in either portfolio do
professor’s Kenneth French webpage. not have data regarding the Environmental pillar. The purpose of this
analysis is to ensure that we only compare firms that have a measure
5. Empirical results ment of environmental impact and observe whether removing these
“unrated” firms impacts the results shown in the previous section. For
5.1. Performance of green vs non-green energy portfolios both categories of green and non-green firms, we also analyze the dif
ference between a portfolio including only environmentally-rated firms
We analyze the performance of the green and non-green portfolios and a portfolio including all firms.
using the Carhart (1997) four-factor and the Fama and French (2015) Table 5 presents the regression estimates for the four- and five-
five-factor models, as described in the methods section. Table 4 presents factor models. In what concerns portfolio performance, both models
the regression estimates of the models using all the firms included in the provide similar results. In general, the alphas of the portfolios are
dataset. We perform the regressions using both the European market lower in value compared to those reported in Table 4 and there is less
returns from professor’s Kenneth French webpage (FF Market Factor) evidence that the green portfolio outperforms the market. However, as
and the MSCI Europe Energy index as the market factor. In order to before, within the energy sector, the alphas of the difference portfolio
assess the performance difference between the green and non-green are positive and statistically significant, so we can still conclude that
portfolios we also present the estimates of the difference portfolio, portfolios of green energy rated firms outperform their non-green rated
which consists of a strategy of a long position in the green portfolio and a counterparts.
short position in the non-green portfolio. Looking at the difference portfolios between the environmentally
As shown in Table 4, the green portfolio presents positive and sta rated firms and all firms, we observe that in terms of performance there
tistically significant alphas. This evidence of outperformance compared is no difference in the case of the green portfolios, but the non-green
to the market is robust to whatever market index is used as benchmark portfolio of rated firms underperforms its non-green counterpart
(the FF market factor or the MSCI Europe Energy index) and to the use of formed by all non-green firms. A possible explanation for this result
the four- or five-factor models and is consistent with the results of Ng might be the higher costs associated to the efforts of improving the
and Zheng (2018) for US green energy portfolios. The non-green port environmental performance of large and more publicly scrutinized non-
folio also tends to present positive alphas, but they are only statistically green rated firms. As to the coefficients of the risk factors, regardless of
significant when the Energy index is used as the market factor. The al the stock market proxy used as the market factor or the multi-factor
phas of the green portfolio are always higher than those of the non-green model used, we find that the green portfolio including only environ
portfolio, and the difference between both is statistically significant at mentally rated firms is less exposed to the size factor than the green
the 5% level when the Energy index is used. We thus conclude that the portfolio including all the firms, which suggests that rated firms are
green energy portfolio outperforms the general market and, considering bigger firms.
the energy sector, it performs better than the non-green portfolio. This
finding is consistent with a portion of the return of the European green 5.3. Performance of green vs non-green energy portfolios over time
energy sector not being fully captured by the risk factors.
Regarding the risk factors estimates, the results show that the co There is evidence in the literature documenting the time-varying
efficients are sensitive to the performance evaluation model used. nature of green portfolio performance over the years (e.g., Climent
Focusing on the results based on the general market index, the estimates and Soriano, 2011; Ibikunle and Steffen, 2017; Ng and Zheng, 2018).
of the Carhart (1997) four-factor model show that both the green and Yet, this issue is largely unexplored in the European energy sector,
non-green portfolio returns are mainly explained by the market risk especially considering the transformations that it has undergone in the
factor. In the case of the Fama and French (2015) five-factor model,
there is evidence that the value and profitability factors are also useful to
explain the returns of the non-green portfolio. This suggests that the 19
Although not reported, we also ran the regressions with clean/alternative
non-green portfolio comprises firms that are more mature (higher book- energy indices as the market factor. The results show that even though clean/
alternative energy indices are better able to explain the returns of the green
portfolio, they are clearly less adequate to explain the returns of the non-green
18
http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.ht portfolio. The general market index does a better job at explaining the returns
ml. of both portfolios. These results are available upon request.
6
Table 4
Performance of green and non-green portfolios – Carhart (1997) four-factor and Fama and French (2015) five-factor models.
Four-factor model αp β1Mkt β2SMB β3HML β4MOM Adj. R2
FF Market Factor
Green 0.558** 1.060*** -0.053 0.117 -0.038 83.4%
(0.002) (0.056) (0.106) (0.095) (0.056)
NonGreen 0.206 1.017*** 0.060 0.268 -0.015 61.3%
(0.003) (0.083) (0.191) (0.170) (0.121)
Difference 0.352 0.043 -0.113 -0.151 -0.023 -2.0%
(0.004) (0.101) (0.241) (0.216) (0.144)
MSCI Europe Energy

Green 1.215*** 0.358*** 0.060 0.275 -0.341*** 42.0%
(0.004) (0.077) (0.301) (0.196) (0.130)
NonGreen 0.459** 0.914*** 0.411*** 0.034 -0.108 88.7%
(0.002) (0.035) (0.130) (0.066) (0.073)
Difference 0.755** -0.556*** -0.351 0.242 -0.233* 38.5%
(0.003) (0.058) (0.218) (0.196) (0.119)
Five-factor model αp β1Mkt β2SMB β3HML β4RMW β5CMA Adj. R2
FF Market Factor
Green 0.671*** 1.028*** -0.145 0.095 -0.326 -0.301 84.0%
(0.002) (0.059) (0.099) (0.165) (0.282) (0.215)
NonGreen -0.037 0.993*** 0.122 0.723*** 0.909*** 0.024 63.1%
(0.003) (0.076) (0.191) (0.234) (0.276) (0.246)
Difference 0.707 0.035 -0.267 -0.628** -1.236*** -0.326 4.4%
(0.004) (0.107) (0.228) (0.301) (0.421) (0.347)
MSCI Europe Energy

Green 1.257*** 0.325*** -0.285 0.854*** -0.353 -1.701*** 53.9%
(0.003) (0.065) (0.214) (0.214) (0.378) (0.276)
NonGreen 0.464*** 0.888*** 0.248** 0.289** -0.070 -0.787*** 90.7%
(0.002) (0.032) (0.110) (0.120) (0.201) (0.117)
Difference 0.793** -0.563*** -0.534*** 0.565*** -0.283 -0.915*** 42.1%
(0.003) (0.053) (0.175) (0.184) (0.366) (0.250)
This table presents estimates of monthly abnormal returns (alpha), expressed in percentage, factor loadings, and the adjusted R2 obtained from regressing eq. (1) and
(2). Mkt corresponds to the excess returns of the market benchmark, proxied by the broad market returns of the European market available in the Professor Kenneth
French’s website converted to Euros or the returns of the MSCI Europe Energy index. The 1-month Euribor rate is used as the risk-free rate. SMB, HML, MOM, RMW and
CMA correspond to the risk factors for the European market available on the Professor Kenneth French’s website also converted to Euros. The portfolios are value-
weighted. The period of analysis is from January 2008 to November 2020. ***, **, and * indicate statistical significance at the 1%, 5% and 10% level, respec
tively. Standard errors, reported in parenthesis, are corrected for autocorrelation and heteroscedasticity using the Newey and West (1987) method.
last decade. This research addresses this issue by evaluating the per contraction period based on the number of oil rigs identified by Plan
formance of European green and non-green energy portfolios over time. tinga and Scholtens (2021).20
For this purpose, we divide the period under analysis into three sub-
periods of similar length: 2008–2011, 2012–2015, and 2016–2020. In 5.4. Robustness tests
addition, we also perform regressions considering rolling window pe
riods of four years starting with the period 2008–2011 and then drop We perform several additional tests to check the robustness of our
ping the oldest year and adding the following year. Portfolio findings to different portfolio formation schemes as well as to different
performance estimates using the general stock market index as the specifications of the performance evaluation models used.
market factor are reported in Table 6. As can be observed from the re As in related studies (e.g., Kempf and Osthoff, 2007; Statman and
sults of both models, the performance of the green portfolio tends to Glushkov, 2009), we analyze the sensitivity of the results to an alter
increase over time and it is on the most recent periods that it shows native portfolio weighting scheme by forming equally-weighted port
abnormal performance compared to the market. This improvement in folios. In general, with equally-weighted portfolios we find less evidence
performance is in line with the results of Ibikunle and Steffen (2017) of outperformance of the green portfolio either compared to the market
with respect to green mutual funds. The performance of the non-green or compared to the non-green portfolio.21 This finding suggests that
portfolio is more unstable. Comparing both portfolios, their perfor
mance is not statistically different except for the 2012–2015 period. In
this subperiod, the alphas from the five-factor model show that the non- 20
Although not reported, following Trinks, Scholtens, Mulder, and Dam
green portfolio underperforms the market and performs significantly (2018) we perform an additional analysis to explore how the results of the
worse than the green portfolio. These results are also consistent with subperiods relates to oil prices. In this additional test, we add the log returns of
Ibikunle and Steffen (2017), who find that green mutual funds outper oil prices (Europe Brent spot price) as an additional explanatory variable. The
form black mutual funds from 2012 to 2014. The explanation for this results show that the strong underperformance of the non-green portfolio
underperformance is the large oil price declines that occurred during compared to market and the green portfolio in the 2012–2015 period disap
mid-2014 until early 2016 that also coincides with the strong pears, confirming our explanation. These results are available upon request.
21
These results are available upon request.
7
Table 5
Performance of green and non-green portfolios including only environmentally rated firms.
FF Market Factor
RatedGreen 0.343* 1.090*** -0.069 0.245*** -0.044 86.6%
(0.002) (0.046) (0.094) (0.080) (0.048)
RatedNonGreen -0.174 1.005*** 0.068 0.161 0.065 54.6%
(0.003) (0.098) (0.217) (0.191) (0.108)
RatedGreen-RatedNonGreen 0.518 0.085 -0.137 0.084 -0.109 0.90%
(0.004) (0.095) (0.234) (0.224) (0.119)
RatedGreen-AllGreen -0.006 0.000 -0.030*** -0.006 -0.002 12.96%
(0.000) (0.003) (0.005) (0.005) (0.005)
RatedNonGreen-AllNonGreen -0.438*** 0.022 -0.010 0.024 0.019 2.9%
(0.000) (0.015) (0.037) (0.013) (0.022)
MSCI Europe Energy

RatedGreen 0.944*** 0.454*** -0.093 0.412** -0.377*** 49.2%
(0.003) (0.071) (0.289) (0.203) (0.115)
RatedNonGreen 0.029 0.962*** 0.333*** 0.119* -0.115** 88.1%
(0.002) (0.030) (0.127) (0.068) (0.049)
RatedGreen-RatedNonGreen 0.915*** -0.507*** -0.426** 0.293 -0.262** 34.1%
(0.003) (0.063) (0.206) (0.215) (0.113)
RatedGreen-AllGreen -0.007 0.001 -0.030*** -0.007 -0.002 13.1%
(0.000) (0.002) (0.005) (0.005) (0.005)
RatedNonGreen-AllNonGreen -0.432*** 0.018*** -0.006 0.024* 0.014 4.2%
(0.000) (0.005) (0.036) (0.014) (0.024)
FF Market Factor
RatedGreen 0.277 1.113*** -0.043 0.306* 0.081 0.056 86.5%
(0.002) (0.043) (0.099) (0.165) (0.282) (0.215)
RatedNonGreen -0.317 0.944*** 0.075 0.575** 0.727** -0.164 55.6%
(0.003) (0.078) (0.201) (0.256) (0.315) (0.295)
RatedGreen-RatedNonGreen 0.594 0.169* -0.118 -0.269 -0.646 0.220 1.6%
(0.004) (0.091) (0.237) (0.307) (0.424) (0.323)
RatedGreen-AllGreen -0.006 0.003 -0.029*** -0.013 -0.009 0.012 13.6%
(0.000) (0.004) (0.005) (0.009) (0.015) (0.013)
RatedNonGreen-AllNonGreen -0.429*** 0.017 -0.013 0.018 0.013 0.005 0.6%
(0.001) (0.013) (0.030) (0.048) (0.071) (0.061)
MSCI Europe Energy

RatedGreen 1.073*** 0.390*** -0.380* 0.842*** -0.377 -1.429*** 54.4%
(0.003) (0.068) (0.210) (0.215) (0.399) (0.322)
RatedNonGreen 0.134 0.925*** 0.169** 0.232** -0.236 -0.724*** 89.9%
(0.001) (0.026) (0.070) (0.095) (0.173) (0.138)
RatedGreen-RatedNonGreen 0.939*** -0.534*** -0.548*** 0.610*** -0.141 -0.705** 33.6%
(0.004) (0.066) (0.200) (0.227) (0.383) (0.285)
RatedGreen-AllGreen -0.004 0.002 -0.030*** -0.013 -0.0113 0.009 13.7%
(0.000) (0.002) (0.006) (0.009) (0.015) (0.010)
RatedNonGreen-AllNonGreen -0.421*** 0.017** -0.011 0.011 -0.005 -0.005 2.5%
(0.000) (0.006) (0.025) (0.049) (0.069) (0.071)
This table presents estimates of monthly abnormal returns (alpha), expressed in percentage, factor loadings, and the adjusted R2 obtained from regressing eqs. (1) and
(2). The green and non-green portfolios are composed of only those stocks that have the Environmental pillar score available. Mkt corresponds to the excess returns of
the market benchmark, proxied by the broad market returns of the European market available in the professor Kenneth French’s website converted to Euros or the
returns of the MSCI Europe Energy index. The 1-month Euribor rate is used as the risk-free rate. SMB, HML, MOM, RMW and CMA correspond to the risk factors for the
European market available on the Professor Kenneth French’s website also converted to Euros. The portfolios are value-weighted and monthly rebalanced. The period
of analysis is from January 2008 to December 2019. ***, **, and * indicate statistical significance at the 1%, 5% and 10% level, respectively. Standard errors, reported
in parenthesis, are corrected for autocorrelation and heteroscedasticity using the Newey and West (1987) method.
large green energy firms may be performing better than their small similar to the ones using the FF market factor.
green energy peers. In addition, we also analyze the performance of the Another important issue is that our analysis is based on uncondi
portfolios considering an alternative general stock market index for the tional models, which might lead to biased estimates of performance.
European market - the FTSE Europe index - and the results were very Thus, we also apply the four- and five-factor models in a conditional
8
Table 6
Performance of green and non-green portfolios over different subperiods.
Four-factor model 2008-2011 2012-2015 2016-2020
Green 0.134 0.225 1.239***

(0.003) (0.002) (0.003)
Non-Green 0.685 -0.632 0.826
(0.006) (0.006) (0.005)
Difference -0.551 0.8567 0.413
(0.005) (0.007) (0.006)
2008-2011 2009-2012 2010-2013 2011-2014 2012-2015 2013-2016 2014-2017 2015-2018 2016-2019 2017-2020
Green 0.134 -0.012 0.061 0.180 0.225 0.232 0.391 0.638** 0.789** 1.342***
(0.003) (0.002) (0.002) (0.002) (0.002) (0.004) (0.003) (0.003) (0.003) (0.004)
Non-Green 0.685 0.283 -0.057 -0.351 -0.632 1.075 0.964 1.397** 1.033* 0.215
(0.006) (0.005) (0.004) (0.005) (0.006) (0.009) (0.008) (0.007) (0.006) (0.005)
Difference -0.551 -0.295 0.118 0.531 0.857 -0.843 -0.572 -0.759 -0.243 1.127
(0.005) (0.004) (0.004) (0.005) (0.007) (0.008) (0.007) (0.008) (0.007) (0.007)
Five-factor model 2008-2011 2012-2015 2016-2020
Green 0.374 -0.117 1.242***

(0.005) (0.002) (0.003)
Non-Green 0.324 -1.320*** 0.835*
(0.007) (0.005) (0.004)
Difference 0.050 1.203** 0.407
(0.007) (0.006) (0.006)
2008-2011 2009-2012 2010-2013 2011-2014 2012-2015 2013-2016 2014-2017 2015-2018 2016-2019 2017-2020
Green 0.374 -0.089 -0.135 -0.326 -0.117 -0.069 0.198 0.561** 0.776*** 1.202***
(0.005) (0.002) (0.003) (0.003) (0.002) (0.002) (0.002) (0.002) (0.002) (0.003)
Non-Green 0.324 0.127 -0.267 -0.580 -1.320*** -0.357 0.042 0.834 0.958* 0.002
(0.007) (0.006) (0.005) (0.005) (0.005) (0.007) (0.005) (0.006) (0.005) (0.005)
Difference 0.050 -0.217 0.132 0.254 1.203** 0.288 0.156 -0.272 -0.182 1.200
(0.007) (0.005) (0.004) (0.006) (0.006) (0.007) (0.005) (0.008) (0.006) (0.007)
This table presents estimates of monthly abnormal returns (alpha), expressed in percentage, obtained from regressing eqs. (1) and (2) considering different subperiods.
The portfolios are value-weighted. ***, **, and * indicate statistical significance at the 1%, 5% and 10% level, respectively. Standard errors of the abnormal returns
estimates, reported in parenthesis, are corrected for autocorrelation and heteroscedasticity using the Newey and West (1987) method.
setting that allows for alphas and betas to vary linearly over time as a dividend yield. In the case of the non-green portfolio, the variable term
function of a vector of conditioning information (Ferson and Schadt, spread appears as statistically significant with a negative sign, meaning
1996; Christopherson, Ferson, and Glassman, 1998).22 As conditioning that performance decreases when the term spread increases. Moreover,
information, we consider two public information variables: the term when we consider the five-factor model, the coefficient of this variable
spread and the dividend yield. These variables were also used in other
previous studies (e.g., Leite and Cortez, 2018; Leite, Cortez, Silva, and
Adcock, 2018). The term spread is computed as the difference between Table 7
the yield of a long-term bond (proxied by the 10-year Germany bench Performance of green and non-green portfolios using conditional models.
mark bond) and a short-term interest rate (represented by the 3-month Four-factor model α0p TS DY Adj. R2
Euribor). The dividend yield is based on the STOXX Europe 600 index.23 Green 0.557** -0.087 -0.268 83.6%
This data was collected from Refinitiv Eikon Datastream. (0.002) (0.005) (0.008)
Table 7 presents the regression estimates for the conditional four- NonGreen 0.216 -1.471** -1.690 69.4%
and five-factor models using the FF market factor as the proxy for the (0.003) (0.007) (0.011)
Difference 0.340 1.385 1.422 18.7%
stock market. We report the conditional performance estimates and the
(0.005) (0.011) (0.016)
explanatory power of the models. We observe that the performance es
Five-factor model TS DY Adj. R2
timates of the green and non-green portfolios are similar to the ones α0p
reported in Table 4. Furthermore, the performance of the green portfolio Green 0.696*** 0.534 -0.100 84.1%
does not seem to change with the information variables term spread or (0.003) (0.006) (0.009)
NonGreen -0.198 -2.219*** -1.196 67.8%
(0.004) (0.007) (0.011)
Difference 0.894* 2.754*** 1.096 17.7%
22 (0.005) (0.010) (0.016)
The conditional four-factor and five-factor models become:rp, t = α0p +
Ap′ zt− 1 + β0prm, t + β0p′ (zt− 1rm, t) + β1pSMBt + β1p′ (zt− 1SMBt) + β2pHMLt + This table presents estimates of the average conditional alphas, the coefficients
β2p′ (zt− 1HMLt) + β3pMOMt + β3p′ (zt− 1MOMt) + εp, t andrp, t = α0p + A′ pzt− 1 + of the time-varying alphas, and the adjusted R2 obtained from the conditional
β0prm, t + β0p′ (zt− 1rm, t) + β1pSMBt + β1p′ (zt− 1SMBt) + β2pHMLt + β2p′ (zt− 1HMLt) versions of the four- and five factor models. The excess returns of the European
+ β3pRMWt + β3p′ (zt− 1RMWt) + β4pCMAt + β4p′ (zt− 1CMAt) + εp, t where zt− 1 = market available in the professor Kenneth French’s website converted to Euros is
Zt− 1 − E(Z) represents the vector of deviations of Zt− 1 from their unconditional used as the proxy for the market factor. The conditioning information variables
average values, β0p, β1p, β2p, β3p are the average betas, β0p′ , β1p′ , β2p′ , β3p′ are the used are the term spread (TS), computed as the difference between the yield of a
vectors that measure sensitivity of conditional betas to the information vari long-term bond and a short-term interest rate (proxied by the 10 year Germany
ables Zt− 1, A′ p is a vector that measures the response of the conditional alpha to benchmark bond and the 3-month Euribor, respectively), and the dividend yield
the information variables, and α0p is the average conditional alpha. (DY) of the STOXX Europe 600 Index. The portfolios are value-weighted. The
23
To mitigate the possibility of spurious regressions, we followed the sug period of analysis is from January 2008 to November 2020. ***, **, and *
gestion of Ferson, Sarkissian, and Simin (2003) and stochastically detrended indicate statistical significance at the 1%, 5% and 10% level, respectively.
these series by subtracting the 12-month moving average. Also, these variables Standard errors are corrected for autocorrelation and heteroscedasticity using
were used in the mean-zero form. the Newey and West (1987) method.
9
for the difference portfolio is also statistically significant at the 1% level. periods and are supportive of the argument that green investments are
As the term spread tends to be higher in recessions, this may indicate more resilient to unstable periods in the oil industry.
that the non-green portfolio shows a lower performance in recessions
and it performs significantly worse than the green portfolio in these 6. Conclusions
market states. Although not reported, the evidence regarding the risk
factors that appear as relevant to explain the portfolios’ excess returns is The growing concerns on the effects of climate change have
similar to that reported in Table 4. Furthermore, there is also evidence prompted worldwide efforts to transition from carbon-intensive fossil
that the exposure to some of the risk factors changes with the infor fuels towards renewable and cleaner sources of energy. As a conse
mation variables. In general, the explanatory power of the conditional quence, investments in green energy are on the rise. Yet, the effects of
models is higher than those of the unconditional models. investing in portfolios of energy versus non-green energy stocks are still
Finally, we further explore how different oil market conditions unclear. This paper contributes to fill this gap by forming portfolios of
impact the performance of the green and non-green energy portfolios by European green versus non-green energy stocks and evaluating their
identifying periods of contraction and expansion in the oil industry performance over the period January 2008 to November 2020.
based on the trailing 12-month growth rate of the number of global oil Focusing on the subset of green and non-green energy firms that are
rigs, as in Plantinga and Scholtens (2021). Months with negative (pos rated, we observe that the environmental performance of the green ones
itive) growth correspond to contraction (expansion) periods. is higher than that of their non-green counterparts, although the dif
Table 8 reports the results of the regressions of the four- and five- ference narrows in recent years. Consistent with the compliance costs
factor models using the FF market factor for contractions and expan argument of carbon-intensive firms (Clarkson, Li, Pinnuck, and
sion periods. As we can observe, the outperformance of the green Richardson, 2015), the stricter regulatory requirements and the
portfolio relative to the market and the non-green portfolio is more increasing public scrutiny seem to be forcing high polluting energy firms
evident in contraction periods. Furthermore, in contraction periods the to implement environmentally responsible practices and processes
non-green portfolio is more exposed to the profitability and investment aiming to mitigate the harmful effects of their activity on climate
factors than the green portfolio. In particular, the results on the in change.
vestment factor are consistent with green (non-green) firms having more In terms of financial performance, our results show that the green
aggressive (conservative) investment strategies when the oil industry is energy portfolio presents abnormal returns when compared to the
experiencing turmoil. In all, these results are consistent with the market, whatever the asset pricing model used or the stock market proxy
improvement of the performance of green portfolios in the most recent for the market factor. The results obtained when using the industry
Table 8
Performance of green and non-green portfolios in periods of expansion and contraction in the oil industry.
Expansion periods (85 monthly observations)

Green 0.540** 1.066*** 0.064 0.378*** -0.119 85.1%
(0.002) (0.061) (0.144) (0.108) (0.091)
NonGreen -0.203 1.176*** -0.068 0.172 0.361** 64.7%
(0.005) (0.086) (0.208) (0.232) (0.159)
Difference 0.743 -0.110 0.133 0.206 -0.480***
(0.006) (0.090) (0.256) (0.254) (0.177)
Contraction periods (70 monthly observations)
Green 1.084*** 1.010*** -0.252** -0.009 -0.037 83.5%
(0.003) (0.077) (0.098) (0.093) (0.063)
NonGreen 0.350 0.826*** 0.065 0.210 -0.317 62.6%
(0.007) (0.156) (0.318) (0.200) (0.198)
Difference 0.735 0.184 -0.317 -0.219 0.281 4.5%
(0.008) (0.203) (0.364) (0.252) (0.243)
Expansion periods (85 monthly observations)

Green 0.447* 1.109*** 0.052 0.336 -0.134 0.070 84.5%
(0.002) (0.058) (0.135) (0.283) (0.379) (0.253)
NonGreen 0.120 0.955*** -0.161 0.478* 0.468 -0.673** 63.3%
(0.005) (0.098) (0.248) (0.265) (0.353) (0.303)
Difference 0.328 0.155 0.212 -0.142 -0.602 0.743** 4.4%
(0.005) (0.121) (0.288) (0.350) (0.573) (0.373)
Contraction periods (70 monthly observations)
Green 1.158*** 0.953*** -0.345*** 0.131 -0.236 -0.563** 85.1%
(0.003) (0.083) (0.102) (0.221) (0.329) (0.238)
NonGreen -0.122*** 0.948*** 0.287 0.808** 1.230** 0.700** 64.8%
(0.005) (0.140) (0.272) (0.369) (0.490) (0.330)
Difference 1.280* 0.005 -0.631** -0.677 -1.467** -1.263*** 16.5%
(0.007) (0.183) (0.293) (0.485) (0.634) (0.425)
This table presents estimates of monthly abnormal returns (alpha), expressed in percentage, factor loadings, and the adjusted R2 obtained from regressing eq. (1) and
(2) for periods of expansion and contraction in the oil industry based on the trailing 12-month growth rate of the number of global oil rigs. The contraction periods we
identify are January 2009–January 2010, August 2012–November 2013, January 2015–December 2016 and July 2019–November 2020. The Mkt corresponds to the
excess returns of the market benchmark, proxied by the broad market returns of the European market available in the Professor Kenneth French’s website converted to
Euros or the returns of the MSCI Europe Energy index. The 1-month Euribor rate is used as the risk-free rate. SMB, HML, MOM, RMW and CMA correspond to the risk
factors for the European market available on the Professor Kenneth French’s website also converted to Euros. The portfolios are value-weighted. ***, **, and * indicate
statistical significance at the 1%, 5% and 10% level, respectively. Standard errors, reported in parenthesis, are corrected for autocorrelation and heteroscedasticity
using the Newey and West (1987) method.
10
index as benchmark indicate that within the energy subsegment of the Kölbel, and Rigobon, 2020), future research could explore the sensitivity
market, green energy portfolios outperform the non-green ones. Our of the results to the use of environmental ratings from different rating
findings for the European market are aligned with those of Ng and Zheng providers and, by doing so, the results could shed light on how ‘green’
(2018), who find that green energy firms perform at least as well as green energy investments really are.
matching non-green energy firms.
Forming portfolios that only include environmentally rated stocks Funding
yields similar results, although in this case there is less evidence of
outperformance of the green portfolio compared to general market. We This work was supported with National Funds of the FCT – Portu
also find no differences between the performance of environmentally guese Foundation for Science and Technology within the project UIDB/
rated firms and all firms in the case of the green portfolios. However, the 03182/2020.
non-green portfolio comprising only environmentally rated firms
underperforms the non-green portfolio of all firms, consistent with non-
Declaration of Competing Interest
green rated firms assuming additional compliance costs in search of
environmental legitimation and minimization of environmental risks.
The authors declare that they have no known competing financial
Moreover, the green portfolio restricted to environmentally rated firms
interests or personal relationships that could have appeared to influence
is less exposed to the size factor than the green portfolio including all
the work reported in this paper.
firms, suggesting that green rated firms are larger than their unrated
peers.
References
The analysis considering different subperiods shows that the per
formance of the green portfolio has improved over time and that the Adcock, C.J., Cortez, M.C., Armada, M.R., Silva, F., 2012. Time varying betas and the
outperformance compared to the market is concentrated in the most unconditional distribution of asset returns. Quant. Fin. 12 (6), 951–967.
recent years, characterized by increased turmoil in the oil industry. The Agudelo, M.A.L., Johannsdottir, L., Davidsdottir, B., 2020. Drivers that motivate energy
companies to be responsible. A systematic literature review of Corporate Social
improvement in performance over time is consistent with the results of
Responsibility in the energy sector. J. Clean. Prod. 247, 119094.
Climent and Soriano (2011) and Ibikunle and Steffen (2017) and sug Albertini, E., 2013. Does environmental management improve financial performance? A
gests that investments in green energy may be benefiting from the meta-analytical review. Organ. Environ. 26 (4), 431–457.
Alexopoulos, T.A., 2018. To trust or not to trust? A comparative study of conventional
transition to a low-carbon economy and from a higher resilience to
and clean energy exchange-traded funds. Energy Econ. 72, 97–107.
turmoil periods in the oil industry. Furthermore, the outperformance of Ambec, S., Lanoie, P., 2008. Does it pay to be green? A systematic overview. Acad.
the green portfolio relative to the market and the non-green portfolio is Manag. Perspect. 22 (4), 45–62.
more evident in contraction periods. In all, these results are consistent Ammann, M., Bauer, C., Fischer, S., Müller, P., 2019. The impact of the Morningstar
sustainability rating on mutual fund flows. Eur. Financ. Manag. 25 (3), 520–553.
with the improvement of the performance of green portfolios in the most Anderloni, L., Tanda, A., 2017. Green energy companies: stock performance and IPO
recent periods and are supportive of the argument that green in returns. Res. Int. Bus. Financ. 39 (Part A), 546–552.
vestments are more resilient to turmoil periods in the oil industry. Andersson, M., Bolton, P., Samama, F., 2016. Hedging climate risk. Financ. Anal. J. 72
(3), 13–32.
Overall, our results suggest that forming portfolios based on green Auer, B.R., 2016. Do socially responsible investment policies add or destroy European
energy screening may result in financial gains compared to the market. stock portfolio value? J. Bus. Ethics 135 (2), 381–397.
The results can be interpreted in light of the theoretical predictions of Battiston, S., Mandel, A., Monasterolo, I., Schütze, F., Visentin, G., 2017. A climate stress-
test of the financial system. Nat. Clim. Chang. 7 (4), 283–288.
Pástor, Stambaugh, and Taylor (2021a), according to which in equilib Berg, F., Kölbel, J., Rigobon, R., 2020. Aggregate confusion: the divergence of ESG
rium green assets have lower expected returns than brown assets, ratings. SSRN Working Paper. https://papers.ssrn.com/sol3/papers.cfm?abstrac
although they can generate higher realized returns in periods when in t_id=3438533.
Boermans, M.A., Galema, R., 2019. Are pension funds actively decarbonizing their
vestors’ tastes shift towards green products. According to our evidence,
portfolios? Ecol. Econ. 161, 50–60.
this shift seems to coincide with the post-Paris agreement period and is Bohl, M.T., Kaufmann, P., Stephan, P.M., 2013. From hero to zero: evidence of
consistent with increasing climate change concerns, as in Pástor, performance reversal and speculative bubbles in German renewable energy stocks.
Energy Econ. 37, 40–51.
Stambaugh, and Taylor (2021b).
Bolton, P., Kacperczyk, M.T., 2020. Carbon Premium around the World. SSRN Working
Our paper contributes to the literature on the impact of green Paper. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3550233.
investing in the energy sector. It also relates to a strand of the literature Cadez, S., Czerny, A., Letmathe, P., 2019. Stakeholder pressures and corporate climate
that addresses the financial consequences of fossil fuel divestment doc change mitigation strategies. Bus. Strateg. Environ. 28 (1), 1–14.
Capelle-Blancard, G., Laguna, M.A., 2010. How does the stock market respond to
umenting that moving out of fossil fuel investments does not impair chemical disasters? J. Environ. Econ. Manag. 59 (2), 192–205.
financial performance (Trinks, Scholtens, Mulder, and Dam, 2018; Carhart, M.M., 1997. On persistence in mutual fund performance. J. Financ. 52 (1),
Plantinga and Scholtens, 2021; Boermans and Galema, 2019). These 57–82.
Ceccarelli, M., Ramelli, S., Wagner, A.F., 2021. Low-carbon mutual funds. In: Swiss
findings have practical implications for investors, the financial industry, Finance Institute Research Paper, 19–13. https://papers.ssrn.com/sol3/papers.cfm?
and policymakers. For investors, the results contribute to a better un abstract_id=3353239.
derstanding of the financial effects of investing in environmentally Christopherson, J.A., Ferson, W.E., Glassman, D.A., 1998. Conditioning manager alphas
on economic information: another look at the persistence of performance. Rev.
friendly energy firms. In sum, investors playing an active role in tran Financ. Stud. 11 (1), 111–142.
sitioning from carbon-intensive fossil fuels to renewable and clean en Clarkson, P.M., Li, Y., Richardson, G.D., Vasvari, F.P., 2011. Does it really pay to be
ergy sources do not necessarily face a trade-off between environmental green? Determinants and consequences of proactive environmental strategies.
J. Account. Public Policy 30 (2), 122–144.
and financial performance. For the financial industry, our research
Clarkson, P.M., Li, Y., Pinnuck, M., Richardson, G.D., 2015. The valuation relevance of
contributes to inform increasingly engaged portfolio managers’ asset greenhouse gas emissions under the European Union carbon emissions trading
allocation decisions when directing investments towards the energy scheme. Eur. Account. Rev. 24 (3), 551–580.
Climent, F., Soriano, P., 2011. Green and good? The investment performance of US
sector. Finally, in the aftermath of the 21st Conference of the Parties
environmental mutual funds. J. Bus. Ethics 103 (2), 275–287.
(COP21), our results are particularly relevant to policymakers aiming to Cochrane, J.H., 2011. Presidential address: discount rates. J. Financ. 66 (4), 1047–1108.
promote the best allocation of capital resources while contributing to a Derwall, J., Guenster, N., Bauer, R., Koedijk, K., 2005. The eco-efficiency premium
more sustainable economy. puzzle. Financ. Anal. J. 61 (2), 51–63.
Dixon-Fowler, H.R., Slater, D.J., Johnson, J.L., Ellstrand, A.E., Romi, A.M., 2013. Beyond
We note that our research, as other studies, relies on a static industry “does it pay to be green?” a meta-analysis of moderators of the CEP–CFP
definition of green and non-green energy firms, thereby not capturing relationship. J. Bus. Ethics 112 (2), 353–366.
the firm dynamics over time. Furthermore, it is limited by the use of the Dragomirescu-Gaina, C., Galariotis, E., Philippas, D., 2021. Chasing the ‘green
bandwagon’ in times of uncertainty. Energy Policy 151, 112190.
environmental ratings of Refinitiv ESG and by its coverage. Considering Du, S., Vieira, E.T., 2012. Striving for legitimacy through corporate social responsibility:
the divergence of ESG ratings across different rating providers (Berg, insights from oil companies. J. Bus. Ethics 110 (4), 413–427.
11
Endrikat, J., Guenther, E., Hoppe, H., 2014. Making sense of conflicting empirical Martí-Ballester, C.P., 2019a. The role of mutual funds in the sustainable energy sector.
findings: a meta-analytic review of the relationship between corporate Bus. Strateg. Environ. 28 (6), 1107–1120.
environmental and financial performance. Eur. Manag. J. 32 (5), 735–751. Martí-Ballester, C.P., 2019b. Do European renewable energy mutual funds foster the
Fama, E.F., French, K.R., 1993. Common risk factors in the returns on stocks and bonds. transition to a low-carbon economy? Renew. Energy 143, 1299–1309.
J. Financ. Econ. 33 (1), 3–56. Merton, R.C., 1987. A simple model of capital market equilibrium with incomplete
Fama, E.F., French, K.R., 2007. Disagreement, tastes, and asset prices. J. Financ. Econ. 83 information. J. Financ. 42, 483–510.
(3), 667–689. Miralles-Quirós, J.L., Miralles-Quirós, M.M., 2019. Are alternative energies a real
Fama, E.F., French, K.R., 2015. A five-factor asset pricing model. J. Financ. Econ. 116 alternative for investors? Energy Econ. 78, 535–545.
(1), 1–22. Molina-Azorín, J.F., Claver-Cortés, E., López-Gamero, M.D., Tarí, J.J., 2009. Green
Ferson, W.E., Schadt, R.W., 1996. Measuring fund strategy and performance in changing management and financial performance: a literature review. Manag. Decis. 47 (7),
economic conditions. J. Financ. 51 (2), 425–461. 1080–1100.
Ferson, W., Sarkissian, S., Simin, T., 2003. Is stock return predictability spurious? Muñoz, F., Vargas, M., Marco, I., 2014. Environmental mutual funds: financial
J. Invest. Manag. 1 (3), 1–10. performance and managerial abilities. J. Bus. Ethics 124 (4), 551–569.
Freeman, R., Evan, W.M., 1990. Corporate governance: a stakeholder interpretation. Newey, W., West, K., 1987. A simple, positive-definite, heteroskedascity and
J. Behav. Econ. 19 (4), 337–359. autocorrelation consistent covariance matrix. Econometrica 55, 703–708.
Galama, J.T., Scholtens, B., 2021. A meta-analysis of the relationship between Ng, A., Zheng, D., 2018. Let’s agree to disagree! On payoffs and green tastes in green
companies’ greenhouse gas emissions and financial performance. Environ. Res. Lett. energy investments. Energy Econ. 69, 155–169.
16 (4), 043006. Pástor, Ľ., Stambaugh, R.F., Taylor, L.A., 2021a. Sustainable investing in equilibrium.
Gonenc, H., Scholtens, B., 2017. Environmental and financial performance of fossil fuel J. Financ. Econ. 142 (2), 550–571.
firms: a closer inspection of their interaction. Ecol. Econ. 132, 307–328. Pástor, Ľ., Stambaugh, R.F., Taylor, L.A., 2021b. Dissecting green returns. In: NBER
Griffin, P.A., Jaffe, A.M., Lont, D.H., Dominguez-Faus, R., 2015. Science and the stock Working Paper, 28940. https://www.nber.org/papers/w28940.
market: Investors’ recognition of unburnable carbon. Energy Econ. 52, 1–12. Patten, D.M., Nance, J.R., 1998. Regulatory cost effects in a good news environment: the
Guenther, E.M., Hoppe, H., 2014. Merging limited perspectives: a synopsis of intra-industry reaction to the Alaskan oil spill. J. Account. Public Policy 17 (4–5),
measurement approaches and theories of the relationship between corporate 409–429.
environmental and financial performance. J. Ind. Ecol. 18 (5), 689–707. Pedersen, L.H., Fitzgibbons, S., Pomorski, L., 2021. Responsible investing: the ESG-
Halcoussis, D., Lowenberg, A.D., 2019. The effects of the fossil fuel divestment campaign efficient frontier. J. Financ. Econ. 142 (2), 572–597.
on stock returns. North Am. J. Econ. and Finance 47, 669–674. Plantinga, A., Scholtens, B., 2021. The financial impact of fossil fuel divestment. Clim.
Hang, M., Geyer-Klingeberg, J., Rathgeber, A., Stöckl, S., 2018. Economic development Pol. 21 (1), 107–119.
matters: a meta-regression analysis on the relation between environmental Popp, D., Hascic, I., Medhi, N., 2011. Technology and the diffusion of renewable energy.
management and financial performance. J. Ind. Ecol. 22 (4), 720–744. Energy Econ. 33 (4), 648–662.
Hartzmark, S.M., Sussman, A.B., 2019. Do investors value sustainability? A natural Porter, M.E., van der Linde, C., 1995. Green and competitive: ending the stalemate. Harv.
experiment examining ranking and fund flows. J. Financ. 74 (6), 2789–2837. Bus. Rev. 73, 120–133.
Humphrey, J.E., Li, Y., 2021. Who goes green: reducing mutual fund emissions and its Reboredo, J.C., Quintela, M., Otero, L.A., 2017. Do investors pay a premium for going
consequences. J. Bank. Financ. 126, 106098. green? Evidence from alternative energy mutual funds. Renew. Sust. Energ. Rev. 73,
Ibikunle, G., Steffen, T., 2017. European green mutual fund performance: a comparative 512–520.
analysis with their conventional and black peers. J. Bus. Ethics 145 (2), 337–355. Refinitiv, 2021. Environmental, social and governance (ESG) scores from Refinitiv. https
IEA- International Energy Agency, 2021, March 2. Global Energy Review 2020: CO2 ://www.refinitiv.com/content/dam/marketing/en_us/documents/methodology/re
emissions in 2020. https://www.iea.org/articles/global-energy-review-co2-emission finitiv-esg-scores-methodology.pdf.
s-in-2020. Rezec, M., Scholtens, B., 2017. Financing energy transformation: the role of renewable
IRENA- International Renewable Energy Agency, Global Renewables Outlook: Energy energy equity indices. Int. J. Green Energ. 14 (4), 368–378.
transformation 2050, 2020. In: https://irena.org/publications/2020/Apr/Global- Re Ripple, W.J., Wolf, C., Newsome, T.M., Barnard, P., Moomaw, W.R., 2020. World
newables-Outlook-2020. scientists’ warning of a climate emergency. BioScience 70 (1), 8–12.
Ji, X., Zhang, Y., Mirza, N., Umar, M., Rizvi, S.K.A., 2021. The impact of carbon Ruggiero, S., Lehkonen, H., 2017. Renewable energy growth and the financial
neutrality on the investment performance: evidence from the equity mutual funds in performance of electric utilities: a panel data study. J. Clean. Prod. 142, 3676–3688.
BRICS. J. Environ. Manag. 297, 113228. Schmutz, B., Tehrani, M., Fulton, L., Rathgeber, A.W., 2020. Dow Jones sustainability
Jordan, B.D., Riley, T.B., 2015. Volatility and mutual fund manager skill. J. Financ. Econ. indices, do they make a difference? The US and the European Union companies.
118 (2), 289–298. Sustainability 12 (17), 6785.
Kempf, A., Osthoff, P., 2007. The effect of socially responsible investing on portfolio Scholtens, B., 2017. Why finance should care about ecology. Trends Ecol. Evol. 32 (7),
performance. Eur. Financ. Manag. 13 (5), 908–922. 500–505.
Kolk, A., Levy, D., 2001. Winds of change: corporate strategy, climate change and oil Statman, M., Glushkov, D., 2009. The wages of social responsibility. Financ. Anal. J. 65
multinationals. Eur. Manag. J. 19 (5), 501–509. (4), 33–46.
Krueger, P., Sautner, Z., Starks, L.T., 2020. The importance of climate risks for Sueyoshi, T., Goto, M., 2009. Can environmental investment and expenditure enhance
institutional investors. Rev. Financ. Stud. 33 (3), 1067–1111. financial performance of US electric utility firms under the clean air act amendment
Leite, P., Cortez, M.C., 2018. The performance of European SRI funds investing in bonds of 1990? Energy Policy 37 (11), 4819–4826.
and their comparison to conventional funds. Invest. Anal. J. 47 (1), 65–79. Trinks, A., Scholtens, B., Mulder, M., Dam, L., 2018. Fossil fuel divestment and portfolio
Leite, C., Cortez, M.C., Silva, F., Adcock, C., 2018. The performance of socially performance. Ecol. Econ. 146, 740–748.
responsible equity mutual funds: evidence from Sweden. Business Ethics: A Eur. Rev. Walley, N., Whitehead, B., 1994. It’s not easy being green. Harv. Bus. Rev. 72, 46–51.
27 (2), 108–126. Wimmer, M., 2013. ESG-persistence in socially responsible mutual funds. J. Manag.
Sustain. 3 (1), 9–15.
12

1 s2.0 S0921800922000891 Main

Uploaded by

Copyright:

Available Formats

1 s2.0 S0921800922000891 Main

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 s2.0 S0921800922000891 Main

Uploaded by

Copyright:

Available Formats

Ecological Economics 197 (2022) 107427

Contents lists available at ScienceDirect

The environmental and financial performance of green energy investments:

Table 1 Environment pillar score

Descriptive statistics Green Non-Green Difference 70.00

Mean (%) 0.914 0.526 0.388 60.00

16 Mean 0.721 0.279 0.442

MSCI Europe Energy

Five-factor model αp β1Mkt β2SMB β3HML β4RMW β5CMA Adj. R2

MSCI Europe Energy

MSCI Europe Energy

Five-factor model αp β1Mkt β2SMB β3HML β4RMW β5CMA Adj. R2

MSCI Europe Energy

Green 0.134 0.225 1.239***

Five-factor model 2008-2011 2012-2015 2016-2020

Green 0.374 -0.117 1.242***

Expansion periods (85 monthly observations)

Five-factor model αp β1Mkt β2SMB β3HML β4RMW β5CMA Adj. R2

Expansion periods (85 monthly observations)

You might also like