The impact of the Board diversity on the firm
performance of power generation companies in the
European Union:
Navigating through the stormy waters of EU Energy
Transition
Master Thesis
MSc International Business & Management
Faculty of Economics and Business
University of Groningen
Matej Ribansky
S3299988
m.ribansky@student.rug.nl
Supervisor: Dr. M.H.F. Ridder de van der Schueren
Co-assessor: Dr. Mariko J. Klasing
Date: 25
thof February 2018
2 ABSTRACT
European Union is at the forefront of becoming a global leader in an energy transition from fossil fuels to near-zero carbon emission energy system (European Commision, 2010). To achieve this ambitious goal, renewable energy industry is a vital component in this complex matrix. In the past, a significant amount of research has been done in the area of renewable energy on subjects such as technology, innovation and social acceptance, but the Boards of Directors and their role in the energy transition have been often overlooked. Board diversity is often regarded as vital part of corporate governance. To successfully navigate within the energy industry, companies need to be able to gain a competitive advantage to overcome the fierce competition facing them from each side. This paper aims to develop a framework on how to analyze the diversity and previous experience of BoD, and their impact on the multinational enterprise performance in power generation industry (electric utilities). This research does so by utilizing age, nationality and gender diversity characteristics in combination with previous energy industry experience/ functional experience using a deductive approach and conducting a quantitative data analysis. Deductive approach builds on the pre-existing theories such as upper echelon theory and resource-based view. BoDs are analyzed based on their age diversity, nationality diversity, gender diversity and functional experience to describe their impact on firm performance measured by ROA, ROE and Tobin´s Q. All in all, the novel contribution of the paper is to explore whether the BoD diversity combined with previous experience of the BoD in the energy industry positively impacts the firm performance of electric utilities in the EU 28 region. As of now, there is almost no research focusing on BoD diversity and EU utility companies, which are considered to be a vital part of the EU energy transition towards carbon-free future, mainly due to the electrification push. Results of this study do not provide a conclusive evidence of the impact that BoD diversity and previous experience have on the firm performance of the EU power generation companies. Therefore, recommendations for future research were outlined in order to better navigate researchers in next editions of research on Board diversity in the energy industry.
3 TABLE OF CONTENTS ABSTRACT ... 2 TABLE OF CONTENTS ... 3 LIST OF ABBREVIATIONS ... 5 LIST OF FIGURES ... 6 LIST OF TABLES ... 6 1. INTRODUCTION ... 7
1.1. Energy Industry Transition in the European Union ... 7
1.2. Previous Research ... 8
1.3. Previous Know-How and Functional Experience ... 8
1.4. Upper Echelon Theory ... 9
1.5. Resource-based View ... 9
2. LITERATURE REVIEW ... 10
2.1. Board of Directors Diversity in the EU Energy Industry ... 11
2.2. Functional Background ... 15
3. METHODOLOGY ... 17
3.1. Research Approach ... 17
3.2. Data Collection ... 17
3.2.1. Company Sampling – Rules and Rationales ... 18
3.3. Variables... 20
3.3.1. Dependent Variables: Firm Performance – ROA, ROE and Tobin’ s Q ... 20
3.3.2. Independent Variables: Management Characteristics ... 21
3.3.3. Independent Variables: Functional Experience ... 22
3.3.4. Control Variables ... 23
3.4. Data Analysis ... 24
4. FINDINGS AND INTERPRETATION... 26
4.1. Descriptive Statistics ... 26
4.2. Basic Setup and Dependent Variables ... 28
4.3. Pearson Correlation ... 29
4.3.1. Nationality and Financial Performance ... 29
4.3.2. Age and Financial Performance... 29
4.3.3. Gender and Financial Performance ... 30
4.3.4. Experience and Financial Performance ... 30
4.4. Linear Regression ... 30
4.4.1. Nationality Diversity and Firm Performance ... 31
4.4.2. Age level and Firm Performance ... 31
4
4.4.4. Previous Experience in the Energy Industry and Firm Performance ... 32
4.4.5. Control Variables ... 33
5. CONCLUDING REMARKS ... 34
5.1. Conclusion and Discussion ... 34
5.2. Sound Management Abilities in times of Energy Transition ... 36
5.3. Limitations and Guidance for future exploration in the realm of energy research ... 36
REFERENCES ... 38
5 LIST OF ABBREVIATIONS
Below can be found the alphabetical list of the abbreviations utilized in this study:
AS Danish: Aktieselskab
English: Stock-based corporation BoD Board of Directors
C-Suite Highest level executives with title ‘Chief” CV Control Variable
COP 21 Conference of the Parties – Paris Climate Conference 2015 DV Dependent Variable
EU European Union
EU 28 28 European Union member countries
EC European Commission
FEB Faculty of Economics and Business
HQ Headquarters
IEA International Energy Agency IV Independent Variable
GHG Greenhouse Gas MLN Million
MNE Multinational Enterprise
NAICS North American Industry Classification System
OECD Organization for Economic Development and Cooperation PLC Public Limited Company
RBV Resource-based view ROA Return on Assets ROE Return on Equity
S.A. Portuguese: Sociedade Anónima English: Public Limited Company SPSS Statistical Software
TMT Top Management Team
6 LIST OF FIGURES
Figure 1: Top three renewable energy producers on country level in the EU – the map of the EU 28 region covered in this thesis………...10 Figure 2: Theoretical framework used in this study………..16
LIST OF TABLES
7
1. INTRODUCTION
1.1. Energy Industry Transition in the European Union
“Fossil fuels have lost”, explained Eddie O´ Connor, Chief Executive of Ireland´s Mainstream Renewable Power with regards to the recent surge in renewable energy investments (Clark, 2017a). “This clean energy disruption has just started and what is striking is how much of a financial impact it is already having on some companies”, noted Per Lekander, a portfolio manager in Lansdowne Partners, a UK-based Hedge Fund (Clark, 2017a). And this trend is expected to continue. The European Union has been vocally advocating the need to reduce the greenhouse gas emissions by at least 40 percent before the year 2030 and by 80 to 95 percent before 2050 (Roelofsen, de Pee, and Speelman, 2016). On the one hand, decision-makers in the European Parliament are aware that the well-being of the EU citizens, firms and the economy as a whole is dependent on safe, secure, sustainable and affordable energy policy (European Commission, 2012). On the other hand, it is known that emissions derived from energy generation account for up to 80% of the EU´s greenhouse gas emissions (European Commision, 2010).
In the recently proposed ´Winter Package´ or ´Clean Energy for all Europeans´ plan, the European Commission sets out two main goals of European Union´s Energy Union project (Buchan and Keay, 2016). First, achieving decarbonization by adapting market and regulatory policy structures to match with the decarbonized energy market scheme of the future (Buchan and Keay, 2016). Doing so, by mainly focusing on decentralization of the electrical grid, more renewable energy power and prosumers (IEA, 2014). Second, it aims to focus more on an energy cooperation at EU-wide level or Europeanisation, by incorporating concepts such as cross-border capacity and renewables payments or so-called price zones (Buchan and Keay, 2016). Therefore, it is clear that renewables are a key component towards achieving the decarbonized and Europeanized energy system in the EU (IEA, 2014). Furthermore, a clear increase in share of the renewables on the EU energy mix is visible, rising from 1995 (5%) to 2014 (13%) as a share of total primary energy consumption (European Commission, 2016a). In addition, between 2008 and 2015, the renewables accounted for the majority (78%) of the new generating capacity added in the EU (Foley et al., 2015).
8 the long-term (European Commission, 2012). Therefore, additional research in the field of energy industry is not only necessary for the future, but it can be reasonably explained on the basis of a need to make EU energy mix more sustainable, affordable and secure, in order to enable safe and sound development of the EU in the upcoming years (European Commision, 2010).
1.2. Previous Research
Analyzing the situation from a theoretical approach, energy research has focused on a ´fight´ between so-called incumbents or dominant utilities mainly focused on fossil fuels and new entrants or renewables energy companies in the energy industry (Hockerts and Wuestenhagen, 2010). Also, researchers have often focused on social acceptance of the renewables Van der Horst (2007); Wuestenhagen, Wolsink and Buerer (2007) and renewable energy innovation together with market penetration, i.e. Johnstone, Haščič and Popp (2010); Lacerda and Van Den Bergh (2014). All in all, enormous amount of research has been done one various topics related to energy business strategies, consumer behavior and marketing, investments and finance, design of energy markets, engineering and others topics related to energy such as environmental policies1.
However, to date, a significant research gap has been left with regards to the importance of board of directors to energy companies and management studies in general. Superior BoDs are needed to successfully navigate firms within the ´stormy waters´ of evermore competitive and complicated EU energy industry, but particularly during the energy transition from fossil fuels towards renewable energy. Such as in the case of Danish company Dong Energy, now known as Ørsted, which changed its name after a BoD decision as a sign of departure from grey energy (Clark, 2017b).
1.3. Previous Know-How and Functional Experience
One of the mail goals of this paper is to depict the relationship between hiring professionals with experience within the field of energy industry to the Board of Directors of a company present in the energy industry in the EU and their impact on the firm performance. This paper presumes that managers familiar with the sector provide a competitive advantage to the firm, by bringing in their industry expertise. This in turn, should drive up the market capitalization and the profitability of the firm. This paper deems previous expertise as extremely important, namely, since the energy industry is being constantly confronted by changing subsidy policies,
1 See website of Energy.gov US research portal (https://energy.gov/eere/education/research-topics) and Centre for Energy
9 regulatory frameworks, and other complexities (Jaffe, Newell, and Stavins, 2005). These complexities are exemplified with what Barney, Wright, and Ketchen (2001); Wernerfelt (1995) describe as implicit industry knowledge, which may be hard to grasp for a manager not familiar with all the circumstances present within the industry. This paper relies on two below-mentioned theories, which can be directly linked to the researched area.
1.4. Upper Echelon Theory
First introduced in 1984 by Hambrick & Mason, the upper echelon theory postulates that it is not only a single individual within the TMT that undertakes the crucial decisions, but it rather is the whole TMT that holds the reins over the company and its managerial decisions (Carpenter, Geletkancz, and Sanders, 2004). Furthermore, upper echelon theory states that strategic decision making is not so much based on technical endeavours, as it is based on the executives perception of their firm´s strategic position via personalized lenses, which are formed by their previous experiences, personality, beliefs and character (Hambrick, 2007; Hambrick and Mason, 1984).
1.5. Resource-based View
Resource-based view postulates that the firm´s competitive edge is a result of combination of firm-specific resources which are not replaceable, valuable, rare and defines them as resources which cannot be easily duplicated – simply the VRIN framework (Barney, 1991). Moreover, it divides them into tangible and intangible assets, and suggests that competitive advantage comes from the ´bundle of resources´ that the firm owns and controls (Wernerfelt, 1995). Between these categories are included management skills such as specific knowledge, organizational processes, routines, and many more (Wernerfelt, 1995). Shortly, these human capital resources, as part of the RBV, should enable the firm to strive thanks to the sustained competitive advantage (Barney et al., 2001). Based on the abovementioned theories, this paper strives to answer following question:
10
2. LITERATURE REVIEW
Strong presence of renewable energy in EU primary energy consumption is critical in order to achieve the EU energy policy goals: Decarbonization and Europeanisation (Buchan and Keay, 2016). Policy changes, including the Renewable Energy Directive II currently under construction, must therefore be correctly interpreted by the BoDs in order to propel their companies through the 2020 to 2030 period safely. Focusing purely on the national or regional electricity markets within the EU 28 region would be a “fallacy”, as the interconnectedness, stability of the grid and cross-border electricity flows between different countries are already partially overseen by European Network of Transmission System Operators for Electricity (ENTSO-E), an EU body. Furthermore, combining this with other changes of the EU energy market legislation, the market for electricity in the EU has become much more competitive for power generating companies as well as electricity providers all around the EU 28 area (Erbach, 2016). Therefore, aligning the firm’s goals with EU energy policy goals, while keeping the company competitive in a global environment requires sound corporate governance abilities which is to a large extent the responsibility of Board of Directors.
11 2.1. Board of Directors Diversity in the EU Energy Industry
As executive teams exert a huge influence over the company´s steps in the future, a significant amount of research has already been conducted within this field, ranging from top management team international diversity to the top management team demographic characteristics such as gender, age and many more (Cannella, Park, Lee and Jr, 2014; Chemmanur and Paeglis, 2005; Hambrick and Mason, 1984). According to the upper echelon theory, decision-making in big multinational companies is usually done on a basis of consensus between top management team members, rather than a single executive (Hambrick, 2007). The same is normally the case for the Board of Directors as well. There undoubtedly exists a fine line between TMT (as defined in the upper echelon theory) and BoD concepts, however, they are closely-related when it comes to the concepts of the team dynamics and organizational performance concepts (Zahra & Pearce, 1989). Their composition is often based on comparable principles and should (desirably) lead to a higher creativity, innovation and quality decision-making that together improve firm performance.
Some researchers move to describe human resources of a firm as the single most importance origin of firm´s competitive edge in the market (Lo and Fu, 2016). For Barney (1991), human resources represent a strategic resource with a potential to deliver competitive advantage to the firm. In this thesis the role of Board of Directors, a governing body elected by the company’s shareholders, will be the main concept discussed with regards to the human resources as a source of strategic advantage for the company. BoD normally consists of various elected executives and not all board members are typically elected at the same time. The size of the board can shrink or increase depending on the shareholders’ decision at the annual general shareholder meetings. Board of Directors has usually an odd number of members in order to be able to reach a majority in terms of decision-making (Bengston et al., 2004).
12 Nationality Diversity and its importance to the company recently became a hot topic in the team diversity studies. Increase in BoD nationality diversity was often used a public relations effort to demonstrate an absence of discrimination, but it is unclear what the impact is on the performance (Erhardt, Werbel, & Shrader, 2003). Various studies postulate that MNEs try to cope with the increasing complexity of their businesses by increasing the number of the nationalities on their executive teams (Nielsen & Nielsen, 2013). Nationality diversity also has an important value for companies, as it should lead to implementation of a global strategy in a local market based on the differing views of each national present on the team (Lauring, 2013). Often, the research results are inconclusive as to the impact of nationality diversity on the MNE performance in a multinational environment (Nielsen & Nielsen, 2013; Milliken & Martins, 1996). Formal and informal institutions have impact on the way the individuals deal with uncertainty and take appropriate actions (Hambrick, 2007). In other words, directors with different nationalities should in theory come up with various differing proposals on which strategic path is best for the company. The pitfall of having too much nationality diversity in the BoD could be, that due to various differences in culture, institutional norms and knowledge basis, the decision-making process could become inefficient or rather slow due to increased team conflict on the Board of Directors which is in line with argument of (Stahl, Maznevski, Voigt, & Jonsen, 2010). However, some authors note that negative consequences of nationality diversity on team processes and performance tend to diminish over time (Earley & Mosakowski, 2000).
Therefore, since the research does not provide clear results and is often somewhat incomplete or rather, in some aspects, under-researched, this thesis would try to contribute to this area, and further explore the impact of nationality diversity of BoDs on firm performance of power generation utilities in the EU. The nationality diversity is also very interesting characteristic to look at, since many some of the power generation companies (utilities) are subject to national regulations and are partially state-owned such as in the case of EDF, which clearly influences the decision-making path of these companies (Chassany & Keohane, 2017). Moreover, the utilities business is considered to be a part of national security in a broader sense, and therefore the states might be prone to have their own nationals sitting on Board of Directors of these strategic companies. Nationality Diversity requirements could be specified in the company bylaws as percentage of all directors appointed to the Board of Directors, but is not required in any EU country by the law. In general, there are almost no differences in this area in the EU 28 region.
13 ethnic diversity is relevant in management teams. First, they point at the trend of employing competent people from different cultures (ethnicities) instead of less qualified people from the same to the corporate boards. Second, they believe that different ethnicities and genders on the BoDs lead to better corporate governance, which in turn leads to higher profitability. Last but not least, this mix of talent creates specific intangible knowledge which can be considered valuable, rare and hard to imitate characteristic, when analyzed from the point of resource-based view (Wernerfelt, 1995). Therefore, the proposition:
H1: Higher nationality diversity of the BoD positively influences the strategic vision and direction of the company in power generation industry, leading to a higher firm performance.
Age Diversity is compatible with the aim of this thesis. BoD age may influence the decision taken, strategic vision and also the degree of creativity in the decision-making process (Kimberly & Evanisko, 1981; Yoon, Kim, & Song, 2016). Jackson & Bantel (1989) and Nielsen (2010) research stream uses a core argument, that BoD and TMT diversity such as age, educational background and other characteristic are important as they inspire variety in terms of knowledge and information sharing, leading to new creative ideas (Milliken & Martins, 1996). Moreover, McIntyre, Murphy, & Mitchell (2007) discovered that age diversity within BoD may lead to an improved firm performance. Furthermore, their study also brings forward another interesting proposition, where having a BoD with different ages can lead to increased social network. That in turn represents an advantage for the company. Higher level of experience also usually goes hand in hand with the age. Maximum or minimum director age levels are normally not required by law to be within a specified range, but upper bounds are usually specified (the maximum age) in the company bylaws.
Next, financial literature shows that older BoD members are usually less risk-prone when it comes to insecure outcomes of projects with high risks, in which case the long project pay-off periods are avoided (Antia, Pantzalis, & Park, 2010). Contrary to this, younger BoD members might be more open to more risky projects. Risky projects, in the case of utilities or power generating companies might be branded as disruptive energy generation technologies, such as the Power-to-Gas2 and others. Hence, the optimal mix between age diversity leads to the second hypothesis:
14 H2: Age diversity of the BoD positively influences strategic choices and organizational creativity in power generation industry, leading to an improved firm performance.
Gender Diversity is used as independent variable to uncover potential impact on the firm performance. Heterogenous groups are better in responding to dynamic market changes according to some researchers. Erhardt et al. (2003) generally find positive link between gender diversity (more women directors) on BoD and firm performance. Trying to replicate their approach, but when looking at the whole EU 28 region could happen to produce ambiguous results for some countries. An interesting reference in this particular case can be made to Germany, where the Boards of Directors of the biggest companies must have 30 percent of seats allocated to women based on a law passed in 2015 (Smale & Cain Miller, 2015). Germany is a unique case in the EU28 requiring 30% females on the BoD3. It is the only such country effectively enforcing this rule in the EU as of the time of writing this thesis. Companies from Germany will, nonetheless, be also analyzed in this thesis. Gender diversity is one of the key team diversity measures and therefore was taken into account as independent variable. Milliken & Martins (1996) argue that age and gender differences should lead to creation of various visions and opinions on the BoD which might lead to conflicts. Therefore, as noted before, BoD gender equality has been chosen as an independent variable, due to differences in the men/women ratio in BoDs based on the different regional groupings in the EU. As additional example of the differences, Nordic countries tend to have more senior female executives than Eastern European countries.
Very recent studies have focused on analyzing impact of gender diversity in TMTs on firm performance by taking a closer look at the effects of having female chief executives in the TMT (Jeong & Harrison, 2017). Jeong & Harrison (2017) found that higher percentage of female executives in TMTs positively impacts company performance in long-term, however it tends to adversely impact the share value in short-term. This thesis aims to analyze the effect of having more males on Board of Directors and its impact on firm performance in line with the abovementioned literature.
3 The list of the countries delivering on gender equality goals seems to be growing and makes the research interesting.
15 H3: The higher the percentage of male directors on the Board of Directors in the EU 28 power generating companies, the worse companies’ financial performance.
2.2. Functional Background
Previous Energy Industry Experience or Functional Background is used as the last independent variable, and therefore we also satisfy criteria with regards to strategic dynamism, which is considered to be very important by some researchers (Naranjo-Gil, Hartmann, & Maas, 2008). Various opinions have surfaced on the topic of BoD functional background. Holland (1976) posits that previous experience influences the behavior of a leader, in this thesis, the senior manager. On the one hand, some authors pose that functional background differences between the leading employees can translate into increased productiveness and creativity of the BoD (Bunderson and Sutcliffe, 2002; Milliken and Martins, 1996). On the other hand, according to Cannella, Park, & Lee (2008) when a team of senior executives has a common background and certain overlaps between their job competencies exist, they will be able to work better as a team. In the middle of the spectrum, Jackson and Bantel (1989) and (Cannella et al., 2014) found that there has not been done enough research in this area, and therefore the multiplicity could possibly lead to adverse performance. This shows that there is a clear hesitance in the literature.
Obviously, common sense dictates that having all directors on BoD with the same background will probably at the end of the day cause more harm than good. This may lead to a creation of narrow-minded thinking (Raskas and Brick, 1992) or decision biases (Burke and Steensma, 1998). This research does not suggest that functional background variety has a negative effect on the BoD structure. It only points a finger at a potential research gap, which should be investigated when it comes to functional experience in energy-related industry. According to some authors, the higher the senior executive team expertise is, the better the quality of the projects undertaken, and hence the return on assets (ROA) or net present value of these projects is higher. This in turn, together with other performance factors drives up the value of the company (Chemmanur and Paeglis, 2005). Therefore, the fourth hypothesis:
16 Chemmanur and Paeglis (2005) support this assumption partially, as they looked at both the pre-initial public offering phase, but also the phase after listing on a stock exchange or long-term stock performance. It has to be noted that this research focuses only on companies that are publicly listed on one of the European stock exchanges. This thesis interprets the abovementioned assumptions in the following manner:
Figure 2: Theoretical framework used in this study
Board of Directors Diversity (Age, Gender & Nationality)
Firm Performance (ROA, ROE & Tobin’s Q)
17
3. METHODOLOGY
The aim of this research is to explore the research gap left by analyzing the impact of Board diversity by looking at age diversity, nationality diversity, gender diversity and the previous energy sector experience (independent variables) on the firm performance represented by ROA, ROE and Tobin’s Q (dependent variables). In order to make sure that the research can be done properly, appropriate methodological approach has to be chosen – data collection and data interpretation must have a system (Saunders, Lewis and Thornhill, 2008). Saunders et al. (2008) outline various possibilities, such as using multi-method approach however, this research will stick to the mono-method.
3.1. Research Approach
This research uses deductive approach: it develops hypotheses based on relevant pre-existing theories and then proceeds to test them (Silverman, 2013). Deductive approach is deemed to be well suited for the positivist approach (Saunders et al., 2008; Snieder and Larner, 2012). Using positivist approach seems to be proper in this case, given that the data used for the research are given and cannot be influenced by the researcher (Remenyi, Williams, Money, and Swartz, 1998). With regards to obtaining valid results on the research question, quantitative research will be conducted. After conducting a quantitative data analysis, the result should lead to a statistically significant outcome based on using relevant statistical methods. For that we need a compatible and consistent data set, based on choosing the right sample size and other characteristics so that it can be quantitatively measured (Saunders et al., 2008).
3.2. Data Collection
18 First, in order to conduct a sound cross-sectional analysis this thesis will solely focus on year 2016 where the data was gathered for 76 publicly traded electric utilities or power generating companies from 17 countries (see Appendix 1). Pre-research done also shows that data in 2016 is available for more publicly traded electric utilities than in any of the previous years (2013 - 2015) and, most importantly, the data available for the below mentioned variables this thesis will be using is only sufficient in terms of quality and representative in year 2016. This thesis does not choose a panel review analysis, as the pre-research has shown that some companies were either not public during the whole period between 2013 and 2016 or data was missing and that would have resulted into a smaller and less representative sample. Furthermore, the data on management characteristics for panel review would have required additional proprietary data sources, as BoardEx and Orbis did not provide data on all companies dating back to 2013 and the data could not have been searched back via publicly available sources such as annual reports. That is due to the various reporting standards in different EU countries. Moving forward, each of the 76 companies included is listed on one of the European Stock Exchanges.
Second, the 76 companies were chosen mainly based on the availability of the data and due to the fact that these companies have, as mandated by law in each country, relatively high level of disclosure regarding their management and company activities compared to private companies.
3.2.1. Company Sampling – Rules and Rationales
19 power generation industry. This classification is different from OECD and EC where companies with more than 250 employees are considered to be large companies. Next, very large companies were also incorporated in this sample. In the electric utilities industry, only large and very large firms have a meaningful impact on the electricity market and wholesale electricity prices. Smaller operators are, simply put, price-takers and there would have been a significant number of outliers in terms of their financial performance depending on the power generation mode chosen by these small producers. Therefore, small and medium size companies were not chosen. In general, this approach is not fully in line with Miletkov, Poulsen, and Wintoki (2016) where they look at the relative size of the company by dividing the market capitalization by gross domestic product. However, this research only looks at the publicly traded companies on the EU level, and therefore their approach is distant and partially irrelevant for this research.
20 3.3. Variables
3.3.1. Dependent Variables: Firm Performance – ROA, ROE and Tobin’ s Q
Naturally, as the research goal is to monitor the company performance by looking at ROA – Return on Assets, we chose Net Income and Total Assets as criteria. ROA is commonly used financial ratio used to capture both short- and long-term aspects of the firm performance (Jeong & Harrison, 2017; Nielsen & Nielsen, 2013). It enables us to observe firm performance changes in time and compare them. Furthermore, ROA is less susceptible to so-called financial engineering than ratios such as return on equity. However, utilizing both will provide even better picture. Further, as noted earlier by ticking Book value of equity and Market value of equity boxes in Orbis, we acquired data that enable us to calculate the ROE – Return on Equity and Tobin’s Q. Tobin’s Q is the last ratio used to measure the firm performance in this thesis. “Tobin’s Q or average Tobin’s Q is constructed by using financial markets data and is often measured by the market-to-book ratio, expressed as the ratio between the market value of equity and the book value of equity” (Pietrovito, 2016). Therefore, by incorporating the Tobin’s Q the element of the company market value will be also captured. Chemmanur and Paeglis (2005) also use more variables to control for other external factors influencing the market capitalization, therefore ROE and Tobin’s Q is also used on top of ROA to obtain proper results.
21 where more than 95% of the data was obtained and checked via Bloomberg Terminal, the renowned financial intelligence platform.
For the few missing pieces of information, where Bloomberg has omitted the data, FactSet offered us a clear overview of the financial performance ratios. The missing information was simply downloaded to excel per each company by using excel add-in and clicking on “download button: Price, Financials & Ratios History” in the FactSet interface after choosing the correct company name in the search menu and choosing option “Snapshot” on the menu bar. The check was conducted for companies available both in Bloomberg and FactSet databases to ensure consistency in the reporting and calculating methods of both platforms.
3.3.2. Independent Variables: Management Characteristics
Age Diversity, Gender Diversity and Nationality Diversity were used as independent variables. The data compilation for each of the respective variables is described below, as many of the steps were done simultaneously.
As very briefly introduced above in the subsection ‘3.2. Data Collection’, in order to ensure that we have the right data on Board of Directors characteristics, the box in Orbis platform called “Directors & Managers” had to be opened and option Board of Directors was ticked. Orbis allows us to check for resignations by stating exact start and end dates, which in the case of my thesis were 01/01/2016 until 31/12/2016. Resignations are typically also available in annual reports, financial press and on company websites. Following platforms enabled me to gather almost full information on the executives: BoardEx, FactSet and Orbis. These platforms have been supplemented by using Reuters.com intelligence service and in case of Polish companies by using the company intelligence website Infoveriti.pl. Webpage announcement of respective companies and their annual reports were used to check and clarify any ambiguous information. Following explains in detail all the steps taken to gather the data:
22 confirm the affiliation with the company. In general, this thesis finds that there are significant differences in reporting standards between various EU member states when it comes the data on company management. Continuing, it must be noted, that FactSet and Orbis were used at all times in conjunction to gather data on management characteristics and therefore the cross-check was conducted for each of the companies. Nonetheless, where available, company websites were checked to make sure that data on the executive ages was accurate. In this way, the Director Age and Gender characteristics were obtained, and furthermore Nationality attributes per each director were also gathered and are used in this thesis to measure Nationality Diversity.
3.3.3. Independent Variables: Functional Experience
23 Finally, given the numerous crosschecks, the data obtained represents more than 90% of director characteristics in the overall sample. What is more, this thesis discounted all of the data back one year to 2016 as a benchmark year for the independent variables measured. Tenure and age were both accounted for in time as t-1, where t represents number of years as of November 2017. This is due to the fact, that the data in FactSet starting from age to tenure was obtained first in November and December 2017.
3.3.4. Control Variables
Firm age and Board size are standard variables used to control for any confounding effects as will be described in the later sections. Control Variables will be put into simple linear regression models to uncover any potential confounding effects on other predictors (independent variables). Should the effects become more positive or negative by adding the control variable (confounder), we know that CV is acting as either suppressor of reinforcer of the effect. In that case we would be talking about spurious relationship, whereby one of the IVs is thought to have a causal relation ship with the DV, but in reality, this is wrongly inferred by the researcher. Firm age and BoD are used as control variables, as previous research shows that older companies can deal better with unexpected circumstances (Jackling & Johl, 2009). There are various reasons why board size is a viable control variable in case of this research. We can directly relate the board size to the group interaction on BoD and the results of this cooperation, such as the strategic decisions or outcomes, for the availability of the resources on a top management team depends from how many executives we have on the team using the argument of (Hambrick & D’Aveni, 1992). Others argue that cognitive resources of a team that is larger could lead to improved dynamic in terms of decision-making within the group, emergence of innovative ideas and overall efficiency (Webber & Donahue, 2001). Some authors also find both negative and positive relationship between Tobin’s Q and management team size, which is one of the performance measures incorporated in this thesis, and the number of members on the board Jackling & Johl (2009), which points at further viability of the assumption to utilize its potential as a control variable.
24 thesis also controlled for tenure of at least 1 year and directors who were with the BoD for least than one year were taken out of the sample.
3.4. Data Analysis
The data will be analyzed by using the IBM SPSS 25 statistical software which is provided by the University of Groningen in the library premises or is offered with a discount to University of Groningen scholars. The Basic Linear Regression in SPSS will be used to conduct regression in this study and bivariate correlations between the variables will be conducted for each and every variable used in the study.
First, simple linear regression analysis will be conducted on the relationship between the BoD age and the firm performance variables. Age diversity will be measured as average age of all BoD members as defined in the earlier section of the thesis. Based on the average age in the BoDs, standard deviation from the mean will be regressed with the financial performance measures.
Second, analysis on the relationship between the BoD Nationality diversity and the firm performance variables will be examined in this thesis. Nationality diversity will be measured based on the HQ country of the company (home country), where all nationals with different nationality will be taken as foreign directors. Total number of BoD members will be divided with the number of foreign executives. This will lead to an overview of how big the percentage of foreigners on the BoD is. On top of this ratio, this thesis uses also more qualitative and representative measure to account for nationality diversity, called Blau Index (Blau, 1977). Blau Index is calculated by using the following formula: 𝐵 = [1 − ∑(𝑝𝑖)2], where the p is the
percentage of members in the ith group (in this research the nationality). Blau Index is a
common tool between researchers and serves as a measure in the field of demographic research. This will allow us to account for number of different nationalities on the BoD and exemplify the true impact of nationality diversity (various fractions). Blau Index is renowned and known as an ideal and frequently used measure of diversity which includes variations within a group (Miller & Triana, 2009; Nielsen & Nielsen, 2013). In this research the nationality diversity is measured in such a way that if board nationality diversity increases, the Blau Index increases as well4. The higher the Blau Index, the higher is the diversity we observe. Blau Index is very similar to the Simpson’s index that is used to measure species diversity and also shares familiar traits with Hirschman-Herfindahl Index (HHI). All in all, this study aims to capture the full picture when it comes to the nationality diversity (see Appendix 7).
25 Third, Gender diversity computed as percentage of male directors on the BoD and the firm performance indicators will be also regressed against the ROA, ROE and Tobin´s Q.
Fourth and the last, Functional Experience as defined below will be also regressed with financial performance indicators. Previous experience in the energy-related industries (according to the aforementioned NAICS classification codes) will be used as confirmation of previous energy sector experience. By creating three categories: 0-2 years, 2-4 years and 4 to more years of experience, this study captures the entry, moderate and professional experience level in the energy industry. Dividing the number of the directors in each of the categories with the board size, we end up with the percentage amount of directors per each category on BoD in the respective company. Since the experience level is stored in the data of a categorical nature, by omitting one of the categories when measuring the impact on ROA, ROE and Tobin´s by conducting linear regression, dummy variable is being used.
26
4. FINDINGS AND INTERPRETATION
4.1. Descriptive Statistics
Table 1 captures the core elements of descriptive statistics which were used in this study. The study observes that even though the measures such as choosing very large companies that are publicly traded were used, the most sizeable company when measured by revenues Électricité de France (82$bn USD in 2016) tended to be much bigger than the firms ranking the lowest with revenues close to millions USD in high-single digits or low double digits. This thesis therefore returns with recommendation for future researchers with regards to EU energy industry in the section 5. Continuing, overall number of directors in study amounted to 562 directors for 76 companies, which translates to more than 7 directors per company on average. From the 76 companies used in this sample, we found the youngest firm with age of only 7 years, company founded in 2009 whereas the extreme on the other end of the spectrum was firm Eszak-Magyarorszagi i Aramszolgaltato founded in 1897 with age of 119 years of existence followed closely by RWE of Germany, founded in 1898.
Table 15
Descriptive statistics of key variables used
Type of Variable Variable Obs. Mean Min Max
Control STDEV Firm Age 76 33.55 7.00 119.00
STDEV Board size 76 1.56 0.20 6.80
Independent
STDEV Age 76 8.59 1.53 23.51
EXP 0 -2 y % 76 0.11 0.00 0.67
EXP 2 - 4 y % 76 0.27 0.00 1.00
EXP 4 - more y % 76 0.62 0.00 1.00
Blau Index Nationality 76 0.20 0.00 0.83
Male % of Directors 76 0.82 0.40 1.00
Dependent
ROA 16 76 0.57 -59.51 17.96
ROE 16 76 2.32 -132.89 86.99
Tobin 16 76 1.20 -6.74 15.45
In terms of board size, we found the most sizeable board of directors in the case of EDP – Energias De Portugal, S.A., a Portuguese company which also boasted the highest number of foreigners when taken as absolute number. The number of foreign directors on the BoD in this case could be explained by increased foreign ownership of the company, where more than 20% of the company’s voting shares are owned by a Chinese Investment Fund. The smallest in terms of Board size were mostly Polish power generating companies such as the PGE Polska Grupa
27 Energetyczna a.s. with only three directors on their BoD. Clearly, the argument of having an odd number of directors on BoD was supported in the abovementioned cases to ease and streamline the decision-making.
Furthermore, the Inter Rao Lietuva, a Lithuanian electricity producer has achieved the lowest average age of directors on its BoD, with the average age of only 41.25 years and a conservative standard deviation of age compared to mean of only 7.27 years (when compared to the analyzed sample). The highest age of directors was on the board of company Envitec Biogas AG of Germany, where the average age closely surpassed the mark of 66 years. With standard deviation of only 3.79 years, we can clearly see that all of the directors on the BoD were above the age of 60. This BoD composition could have also affected the lackluster performance of the company when measured by Tobin’s Q, where in all years the company had Tobin’s Q value of lower than 1, which means that the company was ‘undervalued’. This could be partially caused by the so-called risk-avoidance in longer-term projects mentioned earlier in the study (Antia et al., 2010).
When we take a deeper look into the financial performance indicators, we arrive at a fact that the EU 28 power generation sector delivered overall positive performance when measured by ROA and ROE in year 2016. As described by popular financial press, electric utilities in the EU perform rather as a kind of a bond instrument when it comes to their earnings potential than regular equity shares – in simpler words, they have a much clearer outlook for their earnings during each regulatory period compared to their counterparts in the US (Crooks, 2018). Regulatory period, depending on the national regulator ranges usually from 3 to 5 years, where the allowed Rate of Return6 is set based on mutual meetings between the company and the regulators. Based on this, company gains a sort of ‘guarantee’ when it comes to the stability of the income on their assets, and therefore their overall short- to mid-term performance. This could justify the findings of this thesis, where in general the positive performance was delivered, however the return was significantly lower compared to the sectors outperforming7 on the major stock indices for year 2016. Since Tobin’s Q takes into account market capitalization, this study has uncovered that on average the EU 28 power generation sector has overperformed, meaning that the value of Tobin’s Q vas above 1, and therefore the market value of the firm was higher than the book value of company’s assets. Taking a deeper look into the underlying numbers, we have also uncovered significant outliers in the sample of 76 companies when we talk about the financial performance. In case of ROAs the highest value
28 stood at 17.96% compared to the lowest of -59.51% which was a significant underperformer with the standard deviation of 42.69 compared to the average of 3.14. However, interestingly enough, in case of ROEs there even more significant deviation between the companies and their performance, where one company delivered a positive performance of 86.99%8, whereas the worst performer ‘achieved’ the ROE of -132.89%9 with standard deviation of 95.6 compared to the average of 9.3 One of the reasons for these significant deviations in the year 2016 is the ongoing energy transition, which brought significant financial strain on companies exposed to fossil fuels in the power production as introduced in the beginning sections of this thesis. Future researchers could adjust their research for these outliers in order to come up with more robust results. This study will further describe avenues for further research improvements in this area in section 5.
Continuing, the study has found that on average only 17% of all directors on BoDs of all 76 companies were females (see Appendix 8), with highest female percentage representation of 60%10. This goes in line with the general notion of just how male dominated are energy and technical industries in general at the present moment (Byrne, 2017; Lee & Inajima, 2017). However, given the broad span in this sample, this study merely points out at the result and confers no final conclusion in this area upon the reader. Looking at cultural diversity achieved, this study found on the sample of 76 companies analyzed that on average, only 13% directors from the overall 562 directors in this study were foreigners. Furthermore, out of these 13% or 73 directors, only 21 boasted various nationalities when sitting on BoD with another foreigner. In simpler words, this thesis finds that more often rather than not, foreigners on the BoD are from the same country.
4.2. Basic Setup and Dependent Variables
This study uses significance level (p) of 0.05 or 5% as a benchmark which is a standard. Furthermore, this thesis uses following scales to analyze correlations between the variables based on Cohen’s classification (Cohen, 1990): we utilize four different categories 0 to 0.1 stands for having no effect size or marginal 0.1 to 0.3 has small impact and for 0.3 to 0.5 this study uses term moderate effect. Lastly, 0.5 and higher stands for large impact.
In the Appendix 9, the correlations between the dependent variables ROA and ROE used in this thesis are shown. Obviously, when we look at the financial performance there is a relation between ROA and ROE where both use the Net Income as part of the formula for their
8 Lowest value (ROE 16): RWE AG. 9 Highest value (ROE 16): Centrica Plc.
29 calculation. ROA and ROE have a linear relationship, which can be interpreted as a positive linear relationship where a higher ROA would immediately lead to higher ROE. Given the large correlation and significance, this study will use only one of the variables (where deemed appropriate) together with Tobin’s Q in order to describe the relationship between independent variables and ROA, respectively ROE.
4.3. Pearson Correlation
Given that this thesis has already described the strongly positively correlated dependent variables ROA and ROE, we move forward to describe the correlation between the independent and dependent variables between each other.
4.3.1. Nationality and Financial Performance
First, this study inquired the relationship between nationality diversity and company performance by correlating both variables used for measuring the nationality diversity versus the ROA, ROE and Tobin’s Q. This thesis finds, that having more foreigners on board, and also more nationalities between foreigners on BoD leads to a marginally positive correlation, therefore higher performance when measured by Tobin’s Q. However, as shown in Appendix 10, all results were insignificant and therefore no conclusions can be derived from them. We could assume that the very small positive correlation (almost close to zero) could have been caused by potential conflicts on the Board of Directors caused due to the various perspectives taken on each of the topics discussed, and therefore led to issues with effective decision-making (Stahl et al., 2010).
Resulting correlations for ROA were similar both in terms of positive correlation and significance of the results. Connecting these results with already existing literature, relatively recent research papers on the topic pose that having more nationals in the long-run leads to increased firm performance (Carter et al., 2010). Nonetheless, the ambiguous nature of the results in general provides limited explanatory value and future research should employ more elaborate tools to confirm the above sketched relationships.
4.3.2. Age and Financial Performance
30 once again the result is highly insignificant. Therefore, this thesis finds no support for the hypothesis 2.
4.3.3. Gender and Financial Performance
Third, when we correlate percentage of males on BoD with the ROA, ROE and Tobin’s Q, we end up with having a marginally negative correlation between the variables. The correlations remain insignificant, they are higher than alpha value of 0.05 also in this case and therefore no fine explanation of the results can be provided (see Appendix 11).
4.3.4. Experience and Financial Performance
Fourth, ROA and the three experience levels utilized in the thesis achieve very low levels of correlation. They are almost non-correlated or close to zero as can be seen in Appendix 12. Furthermore, the variables are also non-significant. We can, however observe a strange trend with mostly very marginal and low impact in the group of directors with experience 0-2 years having slightly positive correlation with financial performance measured by ROA, whereas with experience 2-4 years, we see a marginally declining performance (i.e. negative correlation), followed by a low increase in firm performance in case of the most experienced group.
When describing the relationship between the Tobin’s Q and levels of experience, this thesis finds interestingly enough that there is a significant and small positive correlation of 0,199 between functional experience of 2-4 years with Tobin’s Q (see Appendix 12). This might be a result of positive investor sentiment towards executives with moderate experience rather than inexperienced or very experienced managers (with long tenure), who can become entrenched in their positions after a longer period of time (Tirole, 2006). Shleifer & Vishny (2016) and Tirole (2006) argue that managers have many ways in which they can undertake actions which hurt shareholders but enable them to make themselves indispensable for the company and secure their position. This thesis finds that some directors tend to have very long tenures on BoD11. With that said, future work in the area of energy research could focus on this and incorporate tenure as one of the independent or moderating variables.
4.4. Linear Regression
This thesis now moves to describe relationship between various independent (predicting) variables and the dependent (outcome) variables used in this thesis and pre-defined in the abovementioned sections. There exist three most common types of linear relationship: positive, negative or no linear association. The key metrics this thesis looks for in the SPSS are
11 For example, this thesis finds that Italian executive Tomaso Tommasi di Vignano, the CEO of Hera Group, sits on the
31 squared (variance explained), level of significance (Sig.) and unstandardized coefficients: constant (intercept) and the coefficient of each IV used (representing the slope or beta). Using simple linear regression in IBM SPSS 25, regression analysis per each combination of variables was conducted.
4.4.1. Nationality Diversity and Firm Performance
First, linear regression between Blau Index (nationality) and measures of firm performance was conducted. Blau Index was used as a sole measure of nationality diversity, as there is significant and highly positive correlation of 0.924** between Blau Index and the percentage of foreign directors on BoD (see Appendix 10). The variance explained by the Blau Index (nationality) out of the ROA 16 was only at 5% level. The relationship between nationality diversity and ROA 16 is highly insignificant at 0,564 (see Table 2). The trend seems to be towards a positive impact of nationality diversity measured by Blau Index on ROA 16 at 2.177, but the results are insignificant and provide no explanatory value. Regressing nationality diversity measure with ROE 16 and Tobin’s Q does not provide us with any different results and therefore we have a status quo when it comes to the interpretation of linear relationship between nationality diversity and firm performance.
Table 2: Blau Index (Nationality) and ROA 16 – Linear Regression Results12
Model Summary
Model R R Square Adjusted R Square Std. Error of the Estimate
1 ,067a ,005 -,009 8,926631290601694 Coefficientsa Model Unstandardized Coefficients Standardized Coefficients
t Sig. 95.0% Confidence Interval
for B
B Std. Error Beta Lower
Bound Upper Bound 1 (Constant) ,142 1,266 ,112 ,911 -2,381 2,664 Blau Index 2,177 3,755 ,067 ,580 ,564 -5,306 9,660
a. Dependent Variable: ROA 16
4.4.2. Age level and Firm Performance
Second, linear regression between standard deviation of age and measures of firm performance was conducted. The variance explained by the age diversity in ROA 16 was only at 4% level. The relationship between age diversity and ROA 16 is highly insignificant at 0,591.
32 There exists a small positive impact of age diversity on ROA 16 at 0.157, but the results are insignificant and therefore have little or no explanatory value (see Appendix 13 for model). Regressing age diversity measure with ROE 16 and Tobin’s Q does not provide us with any different results and therefore we end up with the same when it comes to the interpretation of linear relationship between age diversity and firm performance.
4.4.3. BoD Gender Mix and Firm Performance
Third, linear regression between percentage of males on BoD and measures of firm performance was conducted. The variance explained by the gender diversity in ROA 16 represented 16% (see Appendix 14 for model). The relationship between gender diversity and ROA 16 was insignificant at 0,273. There is a small negative impact of lower gender diversity (having masculine BoD) on ROA 16 at -0.071, but since the results are insignificant, they provide no explanatory value. Regressing the measure of gender diversity with ROE 16 and Tobin’s Q does not provide us with any different results. Therefore, we cannot approve neither deny the Hypothesis 3.
4.4.4. Previous Experience in the Energy Industry and Firm Performance
33 4.4.5. Control Variables
By conducting separate simple linear regressions between IVs and DVs and then incorporating CVs, this thesis finds no significant confounding effect from board size on any independent variable. Firm age also has no confounding effect on any variable, except for the effect of standard deviation of firm age on standard deviation of age (age diversity). Firm age acts as a suppressor and there exists a spurious relationship between standard deviation of age and the ROE 16 according to this finding (see Appendix 17 for model).
By running simple linear regression just for ROE 16 and firm age standard deviation, the results are close to the significance benchmark of p = 0.05, in this case p = 0.061 with high negative correlation of -0.556. Analyzing this relationship in detail, however, we find that the results are significantly distorted by the underperformance of RWE AG as mentioned earlier in the thesis. RWE AG has one of the highest firm ages and reported the lowest ROE 16 – it is an outlier. Therefore, when I omitted the RWE AG from the company sample, the relationship immediately changes and becomes highly insignificant, variance explained is at 1% and positive effect of only 0.058 is measurable (see Appendix 18 for model). Therefore, based on this empirical exercise we cannot perceive standard deviation of firm age as a confounding variable. Then argument of Jackling & Johl (2009) that older firms are better positioned to withstand tough business conditions in not confirmed.
34
5. CONCLUDING REMARKS
5.1. Conclusion and Discussion
In the previous editions of research in the arena of BoD diversity and its impact on firm performance, authors were, in numerous cases, able to come up with significant results to support or reject their hypotheses. However, as this thesis demonstrates below, there have been instances where researchers have not been able to do so. This study built its grounds on the of theories and arguments used by well-known researchers, in order to analyze the relationship between BoD diversity by looking at age dispersion, gender distribution on BoD, nationality diversity coupled with functional experience, and finally the financial performance variables ROA, ROE and Tobin’s. The empirical research was conducted on the sample of 76 publicly quoted companies in the EU 28 region based on the data for year 2016. To control for any undesired (confounding) effects in order to uncover any spurious correlations, this thesis utilized firm age and board size as control variables. Statistical analysis of the aforementioned variables was conducted by using basic descriptive statistics in excel, bivariate correlations and simple linear regression in IBM SPSS 25.
The end results were overall insignificant, with hardly interpretable effects. Taken together, the contradictory nature of the results and their insignificance coupled with low correlations showed that the results are hardly sound. Consequently, they must be treated with utmost vigilance in mind and should advisably not be relied upon. In the table below, the recap of the hypotheses formulated and subsequent findings of the researched are summarized:
Table 3
Initial assumptions based on the previous literature review and industry development
# Hypothesis Supported/Rejected
/Not Clear
H1
Higher nationality diversity of the BoD positively influences the strategic vision and direction of the company in power generation industry, leading to a higher firm performance.
NOT CLEAR
H2
Age diversity of the BoD positively influences strategic choices and organizational creativity in power generation industry, leading to an improved firm performance.
35 H3
The higher the percentage of male directors on the Board of Directors in the EU 28 power generating companies, the worse companies’ financial performance.
NOT CLEAR
H4
The BoDs lead to a higher firm performance when they have previous experience in the energy sector or energy-related industry.
NOT CLEAR
This study has hoped to discover similarities with previous studies by analyzing BoD diversity, only measured in a different geographical setting and industry, assuming that BoD leads to abundance of new ideas and rise in creativity which could have an impact on firm performance (Jackson & Bantel, 1989; Nielsen, 2010).
Argument of Nielsen & Nielsen (2013), where authors posit that firms might opt to have more nationalities on their BoD to reduce the complexities arising in their international business operations is not confirmed. However, the results of my thesis also have not supported the Milliken & Martins (1996) results where authors expected conflicts in very diverse teams to have on the firm performance. Carter et al. (2010) find similar tendencies in their research when it comes to obtaining insignificant effect size on correlation between the nationality diversity and firm performance.
Furthermore, no decisive effect was found between age diversity on the BoD with regards to the firm performance. Therefore, we cannot relate to the (McIntyre et al., 2007) and their finding where authors discovered positive relation in the increasing age diversity of board members and firm performance. Their study also used data only for one year. Therefore, statistical reliability of the measures incorporated in this thesis may have caused the ambiguous results and should be reconsidered in the future.
When it comes to the gender diversity, Jeong & Harrison (2017) argue in their paper that gender diversity does have an impact on the firm performance, however, this thesis could not provide clear results to be able to make a comparison. Therefore in case of hypothesis 3, the impact of gender diversity was not found to have significant effect on the firm performance, we end up with results that are in line with (Adams & Ferreira, 2009).
36 of the external effects). Therefore, this thesis cannot neither confirm or reject the argument of Hambrick (2007), where the author argues that decision-making is less about technical endeavours and more emphasis is directed on executives´ strategic perception of the firms´ standing. This thesis comes back with a set of recommendations for future researchers such as focusing on different geographical areas. As it is visible on the sample of 76 companies chosen which is a relatively small and diverse sample of companies13, this could be a reason for not getting clearer results, since previous research tended to focus on a more homogenous sample such as the banking industry in the United States of America (Bantel & Jackson, 1989).
5.2. Sound Management Abilities in times of Energy Transition
Even though this study did not find supportive results for the hypotheses formulated earlier, there is doubtlessly a need to develop key firm resources in the EU 28 power generation industry. Human capital, one of the three main firm resources as formulated by RBV, when effectively put into the work may gain the company a sustainable competitive advantage (Barney et al., 2001). It would be difficult to say whether the original theories focusing on TMT and BoD diversity can or cannot be applied in the EU power generation industry Exploratory case studies might constitute a more favourable way of conducting this research to confirm the viability of management studies in the energy industry.
5.3. Limitations and Guidance for future exploration in the realm of energy research In future, in case of any attempts to research energy industry in the EU 28 area setting, researchers should focus only on large and very large companies with joint company status rather than focus solely on publicly quoted companies. This approach will not only increase the sample size, but also make the sample more representative as the vast majority of the power generation sector in the EU 28 region will be incorporated. This sample, focusing only on publicly quoted enterprises, has omitted companies such as EPH14, at the time of writing of this thesis, the seventh biggest energy company in the EU 28 whose primary business is power generation (Thomas, 2017).
The reason for obtaining insignificant results when it comes to the functional experience in the energy industry can be a result of interference of national regulatory bodies15 in the energy industry, government interventions or wrongly chosen variable measuring methods. In case of the wrongly chosen variable measuring methods or sample, this thesis would recommend to
13 76 companies from 17 EU countries, see Appendix 1
14 EPH itself represents a very interesting study proposition, as the company belongs to the top consolidators of grey energy,
namely coal power in a time when most of the companies pursue an exit strategy from the coal power generation industry.
15 Regulatory bodies can have a huge influence over the economics of the companies’ projects by changing subsidy schemes,
