Department of Economics and Business
On the compatibility of economic prosperity and environmental sustainability:
The effect of welfare state regimes and firm internationalization on the relationship between green innovation and firm financial performance
Department: Economics and Business
Master: MSc Business Administration – International Business Thesis Supervisor: Dr. Mashiho Mihalache
Second Reader: Dr. Ilir Haxhi
Author: Lisa Charlotte Gockel Student Number: 11787082
Date: 21/01/2022
Statement of Originality
This document is written by Lisa Charlotte Gockel, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no other sources other than those mentioned in the text and its references have been used creating it.
The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Considering the debate on the compatibility of environmental benefits and profitability, this study scrutinizes the relationship between green innovation and firm financial performance.
Drawing on the resource-based, stakeholder, and institutional-based theories, the relationship is argued to be positive and moderated by country- and firm-level characteristics. On the country-level, the moderating effect of different welfare state regimes is investigated, and on the firm-level, the analysis explores the moderating role of firm internationalization. Testing the model, a panel study analyzes 60 firms from six different countries between the years 2014 and 2018. The results show that the relationship between green innovation and firm financial performance is not positive by itself, but that the moderators play crucial roles. Specifically, the effect of welfare state regimes sets the stage for future research, as it adds to the understanding of diversified national contexts, scrutinizing state intervention. Overall, the model furthers the knowledge on both country- and firm-level moderators of the relationship between green innovation and firm financial performance.
Keywords: green innovation, firm financial performance, welfare state regimes, firm internationalization
Table of Contents
1 Introduction ... 1
2 Literature Review ... 5
2.1 Green Innovation ... 5
2.2 Firm Financial Performance ... 7
2.3 National Context ... 8
2.3.1 Welfare State Regimes ... 10
2.4 Firm Strategy ... 12
2.4.1 Firm Internationalization ... 13
3 Theoretical Framework ... 16
3.1 Green Innovation and Firm Financial Performance ... 16
3.2 Welfare State Regimes ... 18
3.3 Firm Internationalization ... 22
3.4 Conceptual Model ... 24
4 Methodology ... 25
4.1 Sample and Data Collection ... 25
4.2 Variables ... 27
4.2.1 Independent Variable ... 27
4.2.2 Dependent Variable ... 28
4.2.3 Moderating Variables ... 29
4.2.3.1 Welfare State Regimes ... 29
4.2.3.2 Firm Internationalization ... 29
4.2.4 Control Variables ... 30
4.2.4.1 Firm-level ... 30
4.2.4.2 Industry-level ... 30
4.2.4.3 Country-level ... 31
4.3 Statistical Methods ... 31
5 Results ... 34
5.1 Descriptive Statistics ... 34
5.2 Test Statistics and Correlation ... 35
5.3 Fixed Effects Regression Analysis ... 40
6 Discussion ... 44
6.1 Findings ... 44
6.2 Academic Relevance ... 48
6.3 Practical Implications ... 50
6.4 Limitations and Future Research ... 51
7 Conclusion ... 53
8 Bibliography ... 55
List of Figures Figure 1: Conceptual Model……….24
Figure 2: Moderating Effects of the Welfare State Regimes on GI-ROA……….43
Figure 3: Moderating Effect of Firm Internationalization on GI-ROA……….43
List of Tables Table 1: Definition of Variables………...33
Table 2: Descriptive Statistics……….34
Table 3: Correlation Matrix……….………39
Table 4: Fixed Effects Regression Analysis on Firm Financial Performance………39
List of Abbreviations
Corporate Social Responsibility (CSR)
European Patent Office (EPO)
Firm Financial Performance (FFP)
Green Innovation (GI)
Operating Margin (OM)
Return on Assets (ROA)
Return on Equity (ROE)
Return on Sales (ROS)
United States Dollar (USD)
Variance of Inflation Factors (VIF)
1 1 Introduction
“We are the first generation to feel the effect of climate change and the last generation who can do something about it” - Barack Obama, Former US President
Society is facing one of the greatest challenges in history, containing the rise in temperature at about 1.5 degrees, in order to sustain life on earth (IPCC, 2019). Extreme weather events disrupt ecological systems and losses in biodiversity are witnessed around the whole world as the consequences of human made climate change (Gough & Meadowcroft, 2011;
IPCC, 2021). As a result, researchers, scientists, and politicians are trying to find answers to the central question; how to build a sustainable society? However, this is not a recent concern.
The Brundtland report of 1987 was one of the first attempts to formulate internationally, what has been debated ever since: how to generate economic growth that fulfills the needs of today’s world without compromising the life of future generations (UN, 1987).
Hereby, the business scholarship is positioned at the forefront of the discussion. This scholarship is able to address the puzzle that is pressuring industrialized and emerging economies – finding ways to reconcile profitability and the environment (Iman, 2019;
Wohlgezogen et al., 2020; Xue et al., 2019). One of the key topics relevant to this debate, concerns the role that green innovation (GI) can play for firms and society as a whole (Cancino et al., 2018; Oduro et al., 2021). Incorporating technological progress and environmental benefits, green innovation results in products and processes that contribute to sustainable growth (Oltra & Saint Jean, 2009; Zhang et al., 2019). Therefore, GI is seen as a central tool in addressing both, societal and corporate needs, as it enhances sustainability while simultaneously ensuring economic progress (Xue et al., 2019).
Considering its importance, researchers investigate external and internal drivers that impact the establishment of green innovation (Gonzalez, 2009; Hojnik & Ruzzier, 2016).
External drivers concern the effects of institutional stringency (Beise & Rennings, 2005;
2 Horbach et al., 2012), market competition and industry structure (Banerjee et al., 2003;
Brunnermeier & Cohen, 2003), as well as stakeholder pressure (Li & Wang, 2021; Sarkis et al., 2010). Internal drivers include managerial environmental concerns (Agan et al., 2013; Song et al., 2020), corporate governance characteristics (He & Jiang, 2019; Xu & Bai, 2019), and firm capabilities (Albort-Morant et al., 2018). Nevertheless, albeit responding to pressures and incentives, green innovation is said to most likely be implemented when contributing to firm financial performance (FFP) (Hojnik & Ruzzier, 2016; OECD/Eurostat, 2005). Following Stucki (2019), companies only invest in green technology when it is profitable to the firm.
However, when scrutinizing the relationship between GI and FFP, findings diverge.
Studies find a positive (Ambec & Lanoie, 2008; Ar, 2012; Huong et al., 2021), negative (Antonioli et al., 2016; Zaho, 2008), as well as u-shaped or insignificant relationship (Busch &
Hoffmann, 2011; Trumpp & Guenther, 2017). Consequently, different theories are used hypothesizing the link, ranging from stakeholder, to institutional- and resource-based views (Huang & Li, 2017; Shu et al., 2020). Furthermore, whereas several studies shed light on the moderating effects of managerial concerns, firm capabilities, and institutional pressures (Ar, 2012; Chen et al., 2015; Yao et al., 2019), open questions remain regarding the contingencies of the relationship. As researchers are navigating within a context of divergent findings, various relevant theories, and unresolved inquiries, they request greater scrutiny and enhanced investigation (Huong et al., 2021; Khan & Johl, 2019; Xue et al., 2019).
In the light of these inconsistencies, the study sets out to investigate the following research question: How do welfare state regimes and firm internationalization moderate the relationship between green innovation and firm financial performance? This angle of research is critical to explore due to several reasons. Answering to the research gaps and practical relevance of the topic, the debate on GI and FFP is furthered in several ways.
The study makes three contributions. First, knowledge on country-level institutional differences influencing the GI-FFP relationship remains “scarce and fragmented” (Tamayo-
3 Orbegazo et al., 2017: 1349). Authors repeatedly demand to investigate country-level differences through other perspectives than regulatory stringency, to better understand the context that moderates the GI-FFP relationship (Przychodzen et al., 2020; Tamayo-Orbegazo et al., 2017). Thus, a fresh perspective is employed scrutinizing country-level variation through the lenses of welfare state regimes. Importantly, the welfare state perspective widens the framework of the institutional-based view. The regime theory uses insights from other fields of social science, scrutinizing different degrees and forms of state intervention as well as diversified national conditions of the welfare state (Arts & Gelissen, 2010; Aspalter, 2019).
Therefore, bridging different disciplines and synthesizing various theories represents as a promising avenue for further research (Farrukh et al., 2021).
Second, research concerning firm-level moderators, points towards potential effects of the firm’s international strategy, which is yet to be investigated in-depth (Bermúdez-Edo et al., 2017; Przychodzen & Przychodzen, 2015). The link between multinationality and performance is scrutinized, and questions are raised to what extent this connection moderates the GI-FFP relationship. Consequently, differences in the degree of firm internationalization are focused on, viewing the firm not only as being embedded within a larger institutional framework, but as carrying its own agency and interest (Aguilera & Jackson, 2003; Oliver, 1991). As a result, the study provides a somewhat holistic approach. On the one hand, it includes diversified national conditions of welfare state regimes, which shape the context of firms. On the other hand, the study attributes variety to the firm’s nature, investigating the differences in firm internationalization. As future research has been suggested to integrate both, the firm’s strategy as well as the national context it operates in, the study contributes to two research gaps (Przychodzen et al., 2020).
Third, the study allows for practical contributions. The analysis furthers the long- standing debate between profitability and sustainability, scrutinizing the national and strategic context of the relationship. Consequently, the research angle provides important insights into
4 the peculiarities of the GI-FFP relationship. This is crucial, as profitability is often seen as a prerequisite for firms to implement green innovation (Hojnik & Ruzzier, 2016; Stucki, 2019).
Therefore, the findings are important for managers and policymakers alike, as the former have to design the right strategy to address green innovation, and the latter the right form of state intervention to support the GI-FFP relationship.
To contribute in this three-fold manner, the study sets out to analyze firms’ green innovative behavior and their financial performance. Thereby, it is focused on firms from archetypical welfare state regimes, scrutinizing the United Kingdom, Ireland, Germany, France as well as Sweden and Denmark. Conducting a longitudinal analysis over five years, the GI- FFP relationship is investigated applying country- and firm-level moderators. Indeed, the study delivers crucial insights into the significant moderating role of welfare state regimes and firm internationalization, furthering the understanding of the GI-FFP relationship.
Consequently, the remaining sections are structured as follows. In the literature review, key constructs are elaborated on, synthesizing academic findings, and leading towards the central research question. Subsequently, the question is addressed in the theoretical framework, proposing several hypotheses. The next chapter discusses the methodology, outlining the data collection, variables, and statistical methods employed. Thereafter, the results are stated, adding an in-depth analysis in the discussion section, as well as an outlook concerning crucial implications, before diving into final conclusions.
5 2 Literature Review
The literature review starts with discussing the independent variable, green innovation.
Subsequently, the dependent variable, firm financial performance, as well as the country- and firm-level moderators are analyzed. The sections build upon one another, leading towards the key gap in the literature, which is then addressed by the research question.
2.1 Green Innovation
Green innovation is a term of growing academic importance (Oduro et al., 2021). As stated above, it concerns organizational attempts that enhance technological innovation while simultaneously benefiting the environment (Zhang et al., 2019). More comprehensively, the Organization for Economic Co-operation and Development defines green innovations as newly established or improved products, goods as well as services, processes, organizational structures, marketing methods, and institutional arrangements that contain relative environmental improvements (OECD, 2009). Thus, it potentially spans all firm activities and requires measurable environmental enhancements.
Nevertheless, most research focuses on the dimensions of product and process innovation (Putri & Soewarno, 2020; Tang et al., 2018). These concern green innovation regarding new or improved product development as well as the reduction of used resources or other negative externalities involved in the firm's internal production (Putri & Soewarno, 2020).
Few studies, such as the one by Khan and Johl (2019), include green service and organizational innovation. The former refers to the after sales operational activity that customers are confronted with, and the latter refers to non-operational activities such as reducing electricity consumption in non-production processes. Overall, whether implemented within the production, process, delivery or organizational structures, green innovation refers to technological progress that reduces environmental externalities.
6 However, the definition remains partly vague when situated next to similar terms (García-Granero et al., 2018; Hojnik & Ruzzier, 2016; Schiederig et al., 2012). While most researchers use green, eco, environmental, ecological, and sustainable innovation interchangeably, regarding the last term an important difference has to be made. Sustainable innovation includes the social, next to the environmental dimension, and is thus distinct to green innovation, which focuses on technological and environmental progress (Schiederig et al., 2012). Oduro et al. (2021) build upon this distinction, concluding that the other terms can indeed be regarded as similar typologies. Additionally, next to the attention devoted to the dimensions of green innovation, its causes and effects are further scrutinized.
The drivers of green innovation can be divided into macro-, meso-, and micro-level ones. First, the macro-level drivers concern institutional pressures as well as consumption behaviors of populations (Hojnik & Ruzzier, 2016; Li & Wang, 2021). For example, Aguilera- Caracuel and Ortiz-de-Mandojana (2013) employ the institutional-based view, dividing the pressures into environmental regulation stringency and the degree of environmental normative levels within a country. Both factors are found to increase the likelihood of a company to adopt green innovative actions. At the meso-level, the collaborative networks of firms and supply chain characteristics play an important role (Huang & Li, 2017; Melander & Pazirandeh, 2019).
Hereby, the stakeholder theory is used, finding external stakeholder and network pressures to positively impact green innovation (Hojnik & Ruzzier, 2016). Finally, micro-level drivers relate to corporate governance models, management characteristics, and firm capabilities (Albort- Morant et al., 2018; He & Jiang, 2019; Huong et al., 2021; Song et al., 2020). Thereby, the resource-based view allows for insights why firms deem the focus on GI as a viable strategy and under what circumstances their capabilities allow for it (Hojnik & Ruzzier, 2016).
Several authors make additional efforts to sort the drivers according to their importance.
Thereafter, some agreement appears that regulatory pressures act as the most important reason for adopting GI (Hojnik & Ruzzier, 2016; Karimi Takalo et al., 2021). In general, the findings
7 on green innovation drivers can be argued to be complementary, creating a diverse framework.
Differently, controversies emerge when turning towards the effects of green innovation.
Broadly seen, these concern the firm’s performance, which is discussed beneath.
2.2 Firm Financial Performance
Depending on the context, firm performance entails the dimensions of financial, social, environmental, human-resource, technological as well as innovative and managerial firm performance (DeNisi & Smith, 2014; Goyal et al., 2013; Prado et al., 2020; Shu et al., 2020). In relation to GI, operational, environmental, and financial firm performance are most commonly investigated (Ambec & Lanoie, 2008; Ar, 2012; Xue et al., 2019). Operational performance refers to the outcomes of a firm’s processes, as for example the costs and speed of production life cycles (Xue et al., 2019). Environmental performance describes the innovation’s mechanism and impact regarding the environmental consequences of the firm. Finally, financial performance concerns the profitability of the firm and the generated returns due to green innovation (Oduro et al., 2020). Studies investigating the effects of GI on operational and environmental performance tend to yield positive results, with the greatest controversy found regarding firm financial performance (Antonioli et al., 2016; Huong et al., 2021; Trumpp &
Guenther, 2017). Considering the lacking consensus and the importance to combine green innovation and profitability, this study focuses on firm financial performance.
Setting the stage for the relationship between GI and FFP, conflicting findings and various theories emerge. Arguing for a positive relationship, the following theories are employed. Most often, the ‘Porter Hypothesis’ and resource-based views are used. The theory states that green innovation leads to quality improvements and reduced resource consumption.
In turn, GI yields differentiation and cost leadership benefits which generate profitability (Hart, 1995; Huong et al., 2021; Porter & Linde, 1995). Second, from the theory of the first-mover
8 advantage follows that green innovative firms are able to develop new markets, resulting in the preemption of competitors (Chen et al., 2006; Hart & Dowell, 2011). Additionally, also a stakeholder view is employed, arguing that GI improves the relations within the firm’s network, leading to increased quality and firm value (Ambec & Lanoie, 2008; Huang & Li, 2017; Huong et al., 2021). Finally, it is argued that GI enhances the reputation of firms, therefore increasing financial returns (Huang & Li, 2017; Shu et al., 2020).
Contrary, in favor of a negative relationship it is argued that GI increases the overall costs of a firm, while decreasing the product quality (Madaleno et al., 2020; Murphy &
Gouldson, 2000). Furthermore, when serving different goals, such as environmental ones, firms are said to conflict with shareholders’ interests, hampering FFP (Przychodzen et al., 2020;
Zaho, 2008). It is for example argued that shareholders regard an overconcentration on green innovation as significantly risk increasing (Przychodzen et al., 2020). Furthermore, the negative relationship has also been attributed to the nature of some green innovations, which do not improve output as such, but reduce public bads (Antonioli et al., 2016). Taking these views and non-significant empirical findings into account (Busch & Hoffmann, 2011), greater consensus needs to be established. Consequently, the analysis turns towards the context of the relationship.
Scrutinizing country- as well as firm-level characteristics, the next sections address possible moderating effects on the relationship between GI and FFP.
2.3 National Context
Within the literature on green innovation, certain national conditions prevail as the focus of investigation. The development degree (Lopes Santos et al., 2019; Przychodzen &
Przychodzen, 2015) and the intensity of institutional pressures (Aguilera-Caracuel & Ortiz-de- Mandojana, 2013; Shu et al., 2020; Yao et al., 2019) are common country-level moderators of the GI-FFP relationship.
9 To exemplify, Aguilera-Caracuel and Ortiz-de-Mandojana (2013) investigate institutional contingencies, comparing firm performance of green innovative and non-green innovative firms. The authors include the moderating effect of stringent environmental regulations and environmental normative levels, concluding on mixed results (Aguilera- Caracuel & Ortiz-de-Mandojana, 2013). The normative levels have an insignificant and the stringent regulations a significant negative effect, as firms are hampered in their efforts to differentiate themselves and to gain profits. This diverges from what the ‘Porter-Hypothesis’
predicts, namely that stringent regulations not only drive GI, but that they enhance competitiveness and thus firm profitability (Porter & Linde, 1995). Aguilera-Caracuel and Ortiz-de-Mandojana (2013) conclude that a missing aspect in the research is the long-term scope, as the positive effects of GI on FFP might best materialize over time. Furthermore, they suggest other forms of government involvement to have a positive moderating effect, such as voluntary agreements, joint development efforts, subsidies, and taxes. Finally, the authors recommend considering different institutional settings, which allow for greater stakeholder involvement and potentially impact the relationship positively (Aguilera-Caracuel & Ortiz-de- Mandojana, 2013).
These aspects are also stated within other concluding reflections. For example, Huang and Li (2017) emphasize the longitudinal effect and the collaborative type of stakeholder relations that might be vital for the relationship. Considering these suggestions, it appears relevant to focus on national characteristics that account for the specifics of the GI-FFP relationship. However, as national contexts in the form of institutional conditions remain focused on regulation pressures, further research is suggested to include diversified country- level moderators (Przychodzen et al., 2020; Tamayo-Orbegazo et al., 2017).
To broaden the perspective of institutional contexts, insights from other disciplines are considered. The typology of welfare state regimes is used in economics, political science, sociology, and other social sciences. It has however also been connected to business and
10 management research before (De Corte & Verschuere, 2014; Den Dulk et al., 2012). Moreover, it relates to the entrepreneurial and innovative structures of a country (Cooke, 2011) as well as to environmental policies in general (Marquart-Pyatt et al., 2019). Welfare state research allows to investigate country-level differences from a distinct perspective than the strictness of regulations, namely by taking the varying degrees and forms of state intervention into account.
Thus, it appears to contain promising theoretical insights and possibly new empirical findings.
Employing a fresh take on national contexts, this study investigates the moderating effect of welfare state regimes on the relationship between green innovation and firm financial performance.
2.3.1 Welfare State Regimes
Welfare states as such are often attributed to the development of the modern nation state in industrialized countries (Pierson & Leimgruber, 2010). Due to the intensity of novel policy innovations, its ‘birth’ period is broadly associated with the years from 1875 to 1914. The theoretical construct of welfare state regimes may well be traced back to the highly influential work by Esping-Andersen on the three political economies of the welfare state (Esping- Andersen, 1990). While theoretical efforts have been made before by Titmuss (1974), the regimes were the first three-fold welfare state typology, which remains the basis of contemporary welfare state research (Aspalter, 2019). Classifying the regimes, Esping- Andersen investigated the different types of relationships between markets (property) and the state (democracy). Employing a system- and institutional-based view, he traced cross-country similarities, establishing different clusters of institutional configurations.
Based on the three dimensions of de-commodification, social stratification and state- market interaction, Esping-Andersen (1990) established the liberal, conservative/corporate, and social democratic welfare state regimes. The liberal ones, exemplified by the United Kingdom, United States, and Ireland are characterized by low levels of state intervention and laissez-faire
11 of market structures in social security provision, resulting in higher levels of commodification and social stratification. Germany and France, as archetypical conservative/corporate welfare states, illustrate higher levels of state involvement, following a contribution-based social security system that reinforces social stratification. As examples for the social democratic welfare states, most often Sweden, Norway and Denmark are named, illustrating a highly interventionist state and universalist security structures (Arts & Gelissen, 2010; Esping- Andersen, 1990).
However, the three-tier typology is not the only way to classify welfare state regimes.
The critiques towards Esping-Andersen’s typology were numerous, proposing four, five or even more welfare state regimes instead (Arts & Gelissen, 2010; Danforth, 2014). Furthermore, the typology is focused on industrialized nations, with several authors attempting to design frameworks that incorporate emerging markets (Cerami & Wague, 2013; Sumarto, 2017).
Nevertheless, in the light of this research, the relevance and usefulness of the classification by Esping-Andersen is two-fold. First, the focus is not on comparing developed versus developing country institutions, but on country-level similarities as well as differences between developed nations. Second, even when modifying the typology, most contemporary research including welfare state regimes returns to the original typology by Esping-Andersen, as its explanatory power remains valid (Aspalter, 2019; Van Kersbergen & Vis, 2014). Furthermore, it is important to note that the regime structures impact the state, economy, and society in countless ways, and are not limited to social security provision. Moreover, they set the stage for a country's institutional framework and stakeholder relations (Arts & Gelissen, 2002; Esping- Andersen, 1990; Pierson & Leimgruber, 2010). Consequently, the theoretical construct has been applied to relevant topics.
For example, Marquart-Pyatt et al. (2019) investigate climate change views, energy, and eco-policy preferences as well as intended actions on the country-level, using evidence from the European Social Survey. Strikingly, they find great similarities across welfare state regimes
12 as predictors for the preferences and intended actions: social democratic regimes report the highest percentage of participants favoring eco-policies, followed by conservative/corporate and lastly liberal regimes. Therefore, the authors point towards further investigating the effects of political, economic, and material contexts of welfare state regimes on environmental improvements. Furthermore, albeit naming the regimes somewhat differently, Cooke (2011) conducted several case analyses of countries clustered in similar groups, displaying a common tendency in promoting eco-innovation on the institutional level.
In general, welfare state regimes have been connected to the varieties of capitalism by Hall and Soskice (2001), challenging the view that liberal types contain a better structure for radical market innovation than coordinated ones (Akkermans et al., 2009). Akkermans et al.
(2009) conclude that while Hall and Soskice’s two-fold differentiation enables relevant insights, different frameworks allowing for greater variety are needed. Therefore, the three-fold typology on welfare state regimes might provide additional theoretical insights. Moreover, the classification is yet to apply to the relationship between green innovation and firm financial performance. Insights into state and market relations, innovation climate, and stakeholder interactions could enlarge the debate on how the national context impacts the relationship. Thus, welfare state regimes are a crucial moderator to investigate regarding the GI-FFP relationship. Consequently, as this study aims to include both, a country- as well as firm-level perspective, the subsequent section discusses the strategic agency of firms.
2.4 Firm Strategy
Since this study tries to approach potential moderators of the GI-FFP relationship from an integrative perspective, firms are not only considered as being part of an institutional framework but also as agents that formulate strategies (Aguilera & Jackson, 2003; Oliver, 1991). Thus, additional attention is paid to firm-level moderators. Hereby, numerous studies are relevant, with most of them concerning the nature of management and corporate governance (Huong et
13 al., 2021; Song et al., 2020) or capability and resource commitment (Li, 2014; Shu et al., 2020).
It has for example been established that environmental concern and absorptive capacities positively moderate the relationship between GI and FFP (Ar, 2012; Xue et al., 2019). However, avenues for further research remain.
Conducting a comprehensive review of articles hypothesizing positive, negative, or neutral relationships between green innovation and firm financial performance, Przychodzen and Przychodzen (2015) conclude that additional insights are needed regarding the moderating role of firm characteristics. Hereby, it is pointed towards possible effects of the firm’s international strategy (Bermúdez-Edo et al., 2017; Przychodzen & Przychodzen, 2015). This perspective is relevant to further investigate due to two reasons, one being empirical and one theoretical in nature. First, research that analyzed the impact of internationalization on the performance link yielded mixed results, as discussed beneath (Bermúdez-Edo et al., 2017;
Przychodzen & Przychodzen, 2015). Second, when taking the academic literature that concerns the general link between multinationality and performance into account, different theories emerge (Chang & Rhee, 2011; Hennart, 2007; Vermeulen & Barkema, 2002). As the topic carries potential explanatory power for the GI-FFP relationship, further investigation is needed.
2.4.1 Firm Internationalization
To address firm internationalization, a wide range of possible approaches appears. A firm’s international strategy is broadly defined as the strategy that captures the relationship a firm has with foreign countries (Głodowska et al., 2019). Thereby, researchers concentrate on one or more dimensions of the international strategy, the international speed, scope, or scale (Głodowska et al., 2019). The following section’s focus lies on firm internationalization in relation to green innovation. Additionally, some attention is paid to the relationship between multinationality and performance in general.
14 First, firm internationalization has been investigated as a driver of green innovation.
Researching export intensity and green innovation, a study by Choi and Yi (2018) provides crucial insights. The authors explore the effect of export intensity on environmental process and product innovation, consequently focusing on export behavior as an antecedent. As a result, export intensity positively impacts environmental process innovation (Choi & Yi, 2018), stimulating the question concerning the moderating effect of export behavior on the profitability link. Indeed, Przychodzen and Przychodzen (2015) suggest a positive moderating effect of the firm’s exporter-status: these firms are often larger and possibly more valuable in general. In fact, as research has repeatedly shown, exporters tend to be more profitable than non-exporters (Bernard & Jensen, 1999). However, it remains questionable whether profitability causes export behavior, or export behavior causes profitability (Bernard & Jensen, 1999; Hennart, 2007).
Second, investigating the moderating effect of the firm’s international scope on GI and FFP, Bermúdez-Edo et al. (2017) focus on foreign direct investment as the measurement and conclude on mixed findings. They differentiate between the potential effects of knowledge and capability sourcing as well as the exploitation of innovations in diversified geographic scopes.
Doing so, the authors build upon the well-known theoretical contribution by Bartlett and Ghoshal (1989), who categorized international business strategies by employing the two- dimensional framework of global integration and local responsiveness. Using panel data from the information and communication technology industry, the authors present inconsistent results (Bermúdez-Edo et al., 2017). Whereas the geographical scope exploiting the innovations positively affects the relationship, the geographical scope of knowledge sourcing has negative impacts.
Third, the mixed findings are in line with further research arguing that multinationality is not naturally connected to increased performance (Chang & Rhee, 2011; Hennart, 2007;
Vermeulen & Barkema, 2002). Hennart (2007) employs the transaction cost theory, illustrating that theoretically, no automatic positive relationship is given. Instead, arguments are made for
15 a rapid expansion to bring the greatest gains (Chang & Rhee, 2011), as well as for a coordinated and balanced internationalization (Vermeulen & Barkema, 2002), or one that is best evaluated in different phases (Lu & Beamish, 2004). Thus, the arguments diverge and the context a firm operates in plays a vital role. Consequently, further theoretical and empirical investigation is required. Specifically, Przychodzen et al. (2020) call to examine both, firm internationalization as well as the impact of various national characteristics, as two moderators for the GI-FFP relationship.
The objective of this study is to answer to these gaps in the literature. Having discussed the academic state of the art regarding green innovation and firm financial performance, the study is designed to further the understanding of the relationship and its possible moderators. Therefore, the following research question aims at synthesizing a greater understanding of country-level variation, as well as firm-level differences:
How do welfare state regimes and firm internationalization moderate the relationship between green innovation and firm financial performance?
16 3 Theoretical Framework
3.1 Green Innovation and Firm Financial Performance
Scrutinizing the link between green innovation and firm financial performance, several theories suggest a positive relationship. Hereby, the resource-based and stakeholder view, as well as the theory of the first-mover advantage allow for important insights.
First, the resource-based view explains how the competitive advantage of a firm is gained through its linkage with the natural environment (Hart, 1995; Hart & Dowell, 2011; Lee
& Min, 2015; Marín-Vinuesa et al., 2020). The causal mechanisms concern the relationship of the firm’s resources and capabilities with its performance. Within the production process, firms prevent pollution through minimizing the generated emissions and waste disposal. The overall costs are reduced accordingly, positively contributing to firm performance (Hart, 1995; Hart &
Dowell, 2011; Przychodzen & Przychodzen, 2015). Furthermore, green innovation in product design entails following higher production standards, leading to quality improvements (Huong et al., 2021; Tang et al. 2018). Consequently, the firm’s competitive advantage is gained through cost leadership and product differentiation (Ar, 2012; Chen & Liu, 2018; Porter &
Linde, 1995). Moreover, as the competitive advantage does not materialize overnight, the GI- FFP relationship is argued to yield greatest positive results over a longer period (Hart & Ahuja, 1996; Huang & Li, 2017; Xue et al., 2019). Thus, GI is seen as a resource that generates a competitive advantage and ensures the firm’s success over the long run, enabling financial sustainability (Hart, 1995; Hart & Dowell, 2011; Huong et al., 2021).
Second, predicting a positive relationship between GI and FFP, the theory of the first- mover advantage provides additional insights. Hereby, green innovation pioneers are able to establish a strong consumer preference (Porter & van der Linde, 1995; Przychodzen et al., 2020). Associating the firm with environmental leadership, contributes to a green image that sticks with the company and constitutes an advantage compared to its competitors (Ar, 2012;
17 Przychodzen et al., 2020). Thus, the early association with green innovation, as well as the innovative product design and process result in the preemption of competitors (Aguilera- Caracuel & Ortiz-de-Mandojana, 2013; Hart, 1995; Hart & Dowell, 2011). Furthermore, being a green pioneer allows a firm to sell its products and services for higher prices and to develop new markets. Consequently, the firm generates greater revenues and increases its financial performance (Chen et al., 2006; Przychodzen et al., 2020).
Third, based on the stakeholder theory, green innovation is rewarded by actors inside and outside of the firm, leading to greater firm value (Huang & Li, 2017; Huong et al., 2021).
To ensure lasting financial performance, it is vital for firms to balance the needs of different stakeholders. Hereby, green innovation helps to acquire and keep the inclusive balance (Huong et al., 2021; Weng et al., 2015). Focusing on reciprocity mechanisms between stakeholders, it is demonstrated that integrating ecological objectives into the firm’s network and supply chain increases their overall trust and engagement, leading to greater firm value (Huang & Li, 2017).
Specifically, findings show that the ecological objectives of green innovation significantly improve employee commitment and employment rate, increasing firm financial performance (Aldieri, et al., 2019; Muhammed et al., 2015).
Finally, adding to the stakeholder view, green innovation is also rewarded by actors outside of the firm, enhancing the firm’s perceived reputation on the market (Eidat et al., 2008;
Huang & Li, 2017; Shu et al., 2020). As a result, the corporation yields greater customer loyalty which increases financial performance (Shu et al., 2020; Zhu et al., 2014). Considering the environmentally beneficial products and services, the customer retention rate expands significantly, and the firm is additionally able to sell the products at a greater price premium (Kurdi et al., 2020; Shu et al., 2020). Thus, the revenue advantage holds even besides the argument of being a first-mover.
18 In conclusion, green innovation generates a strengthened competitive position, first- mover advantage as well as improved stakeholder relations and customer loyalty which increase firm value and profitability. Consequently, the following hypothesis is proposed:
H1: Green innovation positively affects firm financial performance.
3.2 Welfare State Regimes
Dividing industrialized countries along their welfare state dimensions, the liberal, conservative/corporate, and social democratic types emerge. Employing the institutional and stakeholder theory, the structures of the liberal regime appear to negatively moderate the GI- FFP relationship. Differently, the conservative/corporate and social democratic types moderate the relationship positively.
First, according to the institutional-based theory, firms have to comply with regulatory as well as normative rules in order to attain legitimacy and secure firm survival (Aguilera- Caracuel & Ortiz-de-Mandojana, 2013). Consequently, conforming and proactively adapting to institutional pressures that push for environmental standards positively affects the relationship between GI and FFP, as the firm gains in legitimacy (Shu et al., 2020). However, stringent regulations are also suggested to hamper the link, as they are coercive in nature and potentially limit the firm’s agency (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013). Research grants, investment support, innovation, and ecological incentives are instead proposed to have a positive effect (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Shu et al., 2020). Combining the peculiarities of the GI-FFP relationship, the perspective of welfare state regimes and the institutional-based theory deliver crucial explanations.
Specifically, green innovation most likely generates profitability over a longer period (Huang & Li, 2017; Xue et al., 2019). Looking into policy recommendations to promote the
19 link between GI and FFP, it is suggested that due to the long-time horizon, businesses have to be provided with ongoing support and predictability (Del Río et al., 2010). It appears necessary, to limit the uncertainty for investors as well as the risk incorporated in establishing new innovations that prosper over the long run (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013).
Thus, extended state intervention beyond mere regulations positively moderates the relationship between GI and FFP (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Shu et al., 2020). The interventions provide certainty, support, and a long-term provision that strengthen the trust of investors, consequently increasing the possibility for firms to gain legitimacy.
Importantly, the degree of state intervention and support is reflected in the different forms of state-market interaction in the welfare state regimes (Arts & Gelissen, 2010; Aspalter, 2019; Esping-Andersen, 1990). As the liberal type illustrates low levels of intervention, it is argued to negatively moderate the GI-FFP relationship. Differently, the conservative/corporate type displays a medium and the social democratic type a high degree of intervention and support. As a result, both regime types moderate the relationship positively, with the social democratic regime yielding a greater positive effect.
Second, the stakeholder theory provides additional insights discussing the moderating effect of welfare state regimes. In general, cooperative and well-established stakeholder relations generate mutual trust and engagement within the firm’s network. Subsequently, the relationship between green innovation and profitability is strengthened, as overall firm performance improves (Huang & Li, 2017; Tamayo-Orbegazo et al., 2017). Regarding welfare state regimes, several mechanisms are relevant that moderate the described relationship. These concern the provision of social capital between external and internal stakeholders, employee commitment, processes of knowledge transfer between different sectors, and the perceived market reputation in welfare states.
When promoting a participative approach between stakeholders, states are able to generate a suitable interaction within and between the firm’s network and supply chains
20 (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Del Río et al., 2010). Indeed, strong welfare states are associated with providing increased social capital between stakeholders, fostering trust and mutual relations (Kääriäinen & Lehtonen, 2006). The generated social capital and trust within the firm’s network strengthen the relationship between GI and FFP, as the stakeholders’
interaction improves (Huang & Li, 2017).
Furthermore, strong welfare states are said to positively affect the relationship between the perceived societal usefulness of the work, which is potentially provided by green innovation and greater job satisfaction (Kjeldsen & Andersen, 2013). Subsequently, the increased job satisfaction leads to enhanced employee commitment and thus firm performance (Muhammed et al., 2015). Therefore, countries with greater state intervention and support positively moderate the relationship between the societal usefulness of green innovative work and employee engagement, increasing firm value.
Additionally, governmental intervention plays a vital role in promoting knowledge transfer between the different stakeholders that is relevant for a successful performance of the firm. Knowledge transfer as such, not only generates trust within the network, but also increases profitability, as ideas, new innovations, and quality standards are diffused (Palacios-Marques et al., 2013; Reed, 2010). Importantly, increased state support improves the coordination and transfer of green innovative knowledge between private and public sectors, positively moderating the relationship between GI and FFP (Martínez-Ros & Kunapatarawong, 2019).
Finally, as established in the section above, green innovation enhances the firms’
reputation on the market, increasing customer loyalty and financial performance (Eidat et al., 2008; Huang & Li, 2017; Kurdi et al., 2020; Shu et al., 2020). The positive relationship between GI and FFP, established due to the increased market evaluation, is further moderated by the welfare state regimes. Indeed, findings demonstrate different degrees of eco-policy preferences and favorable attitudes towards eco-innovation, when comparing social democratic, corporate/conservative, and liberal welfare states (Cooke, 2011; Marquart-Pyatt et al., 2019).
21 The level of favorable preference and attitude ascends with welfare state strength and intervention (Marquart-Pyatt et al., 2019). Consequently, it is argued that strong welfare states displaying greater eco-preference positively moderate the relationship between GI and FFP, as they strengthen the market evaluation of green innovative firms.
Therefore, several arguments can be made. For an institutional context to positively impact the relationship, state intervention and the provision of predictability are needed on all stages of the innovation process (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Del Río et al., 2010). Additionally, the interaction and attitudes of various stakeholders are vital to guarantee a successful link between GI and FFP (Tamayo-Orbegazo et al., 2017). Both aspects relate to the nature of the different welfare state regimes. Clustering these on a continuum of governmental support, welfare state strength, and the promotion of stakeholder relationships, they are grouped in ascending order: liberal, conservative/corporate, and social democratic welfare state regimes (Arts & Gelissen, 2010; Kääriäinen & Lehtonen, 2006; Kjeldsen &
Andersen, 2013; Spicker, 2012). Consequently, two hypotheses are proposed:
H2a: Liberal welfare state regimes have a negative moderating effect on the relationship between green innovation and firm financial performance.
H2b: Conservative/corporate and social democratic welfare state regimes have a positive moderating effect on the relationship between green innovation and firm financial performance, with the effect of social democratic welfare state regimes being stronger.
22 3.3 Firm Internationalization
Finally, the theoretical framework turns towards the proposed firm-level moderator. As established in the literature review, one cannot assume a natural positive link between multinationality and firm performance (Hennart, 2007). Moreover, various elements impact the relationship, rendering it context dependent (Verbeke et al., 2009). These concern the variety of strategic motivations in international expansion, foreign environmental characteristics, and importantly, the capability portfolio. It is argued according to the resource- and institutional- based theory, as well as with insights from the organizational learning perspective that within the specific GI-FFP context, the level of internationalization has positive moderating impacts.
First, employing the resource-based view, research repeatedly emphasizes the importance of superior capabilities within the multi-national context and competition (Chang
& Rhee, 2001; Lu & Beamish, 2004). This view can be synthesized with the argument that GI leads to cost leadership and differentiation benefits, and thus improves the firm’s competitive advantage and value (Ar, 2012; Chen & Liu, 2018; Porter & Linde, 1995). Thereafter, GI acts as the superior capability within the international strategy. Internationalization, as for example reached through exports or foreign operations, positively moderates the relationship between GI and FFP, as it allows to reap the benefits of the innovative capabilities in an increased market scale (Bermúdez-Edo et al., 2017; Chen et al., 2006; Przychodzen et al., 2020). Consequently, firm internationalization accelerates the exploitation of the differentiation and cost leadership advantages, increasing financial performance (Porter & Linde, 1995).
Second, including firm internationalization as a moderator enables the study to employ the institutional-based view once more, in order to account for the effects of international regulations. Indeed, Przychodzen and Przychodzen (2015) hypothesize that having an exporter- status moderates the relationship between GI and FFP, as the firm has to navigate various regulatory frameworks. As a result, the firm becomes part of the global business environment
23 (Tashman et al., 2019). Thus, it is exposed to the scrutiny of numerous global legitimacy actors, public and private, that monitor the firm’s workings (Bermúdez-Edo et al., 2017; Marano &
Kostova, 2016; Tashman et al., 2019). Consequently, the extended scrutiny pushes the firm to improve its green innovative qualities to gain in legitimacy, thereby strengthening its reputation and firm performance.
Finally, insights from the organizational learning perspective and resource-based view suggest that being exposed to the international context not only heightens the level of scrutiny for the firm, but also its possibilities to improve and develop (Hsu & Pereira, 2008; Tashman et al., 2019). As the firm is introduced to different practices, the possibilities to explore alternative production processes and to source knowledge are enlarged (Bermúdez-Edo et al., 2017). Consequently, the intensified exposure to knowledge and the possibility to test different approaches lead to improved quality standards in green innovative products and processes, which enhance firm value and financial performance.
In sum, internationalization leads to capability exploitation benefits and strengthens the quality of green innovation through navigating various regulatory contexts and sourcing different forms of knowledge. Thus, the GI-FFP relationship is positively moderated.
Consequently, the last hypothesis is stated:
H3: Firm internationalization has a positive moderating effect on the relationship between green innovation and firm financial performance.
24 3.4 Conceptual Model
The conceptual model in Figure 1 summarizes the hypotheses. Overall, green innovation strengthens the firm’s competitive position, generates a first-mover advantage as well as improved stakeholder relations, and thus increases firm financial performance. The welfare state regimes moderate the GI-FFP relationship differently, according to their level of governmental support, welfare state strength, and the promotion of stakeholder relationships.
Finally, firm internationalization leads to capability exploitation benefits and strengthens the quality of green innovation, as the firm navigates various regulatory contexts and sources different forms of knowledge. Thus, firm internationalization has a positive moderating effect on the relationship between GI and FFP.
Figure 1: Conceptual Model
H1: Green Innovation (+) Firm Financial Performance H3:
Internationalization (+)
H2a & H2b: Welfare State Regimes Liberal (-)
Conservative/corporate (+) Social Democratic (++)
25 4 Methodology
The methodology is designed to investigate the research question and test the hypotheses quantitatively. This is in line with most other research investigating GI and FFP (Huang & Li, 2017; Oduro et al., 2021). Therefore, the sample and data collection are elaborated on, before discussing the variables. The chapter ends on a brief description of the statistical methods.
4.1 Sample and Data Collection
To select the right sample various steps are taken. First, as discussed in the literature review, the goal is to scrutinize differences between industrialized countries. Additionally, welfare state research has originally been concerned with European as well as North American countries (Aspalter, 2019; Esping-Andersen, 1990; Pierson & Leimgruber, 2010). Therefore, only arche- typical countries representing the three regime types are chosen to preempt diverted results (Arts & Gelissen, 2010; Esping-Andersen, 1990; Powell & Barrientos, 2004): for the social democratic type, Sweden and Denmark are selected. For the conservative/corporate one, Germany and France are the most common named examples. Finally, the United Kingdom and Ireland are crucial representatives for the liberal type. The United States is also grouped as an arche-typical liberal welfare state regime, however, due to its size and historical roots, the country is considered as an extreme case compared to European welfare states and is thus not selected (Pierson & Leimgruber, 2010). Whereas the study could focus on one representative per regime type, including two cases allows for more robust results.
Second, since specific countries are the key interest of this study, it is not focused on one single industry. Instead, all industries that can be grouped under material or industrial industries, as well as industries concerning personal health or technology are included, as further explained in the variables section (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013).
This is done to avoid biasing the results by selecting an industry that is particularly strong in
26 one, but not in another country. To guarantee that industry differences do not divert the analysis, they are controlled for using dummy variables (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Huang & Li, 2017).
Third, the firms are chosen by their performance, using the Financial Times 500 list (Bermúdez-Edo et al., 2017). Hereby, the Financial Times Europe 500 list is used due to its appropriateness of the research focus on European countries. The companies are ordered by their market capitalization, reflecting the stock value. All values were converted into United States Dollar (USD) to compare the firms correctly (Dullforce, 2015). To employ the list for this study, companies that are grouped as “banks” in Compustat Capital IQ were excluded from the ranking, due to the different variables used for FFP and in accordance with the industry classification by Aguilera-Caracuel and Ortiz-de-Mandojana (2013). Subsequently, for each regime type, the first twenty firms on the list are selected. For both, Germany, and France, the first ten firms listed in the Financial Times Europe 500 are analyzed. The same is done for Sweden and Denmark. In the case of the liberal regimes, the first ten companies from the United Kingdom are selected and all of Ireland’s listed firms (five), with the remaining five taken from national lists of best performing companies.
Fourth, the objective is to conduct a longitudinal analysis, choosing a timeframe of five years. Compared to cross-sectional analyses, the design allows to scrutinize the long-term effect that GI might have on FFP, generating more robust results that account for potential change (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Bono & McNamara, 2011; Huang &
Li, 2017). Furthermore, as GI is a topic of increasing significance, a recent timeframe is recommendable (Oduro et al., 2021). Therefore, the years from 2014 to 2018 are analyzed, representing a timeframe that is not yet distorted by the economic drawbacks in 2020, which resulted from the global pandemic. Consequently, the sample includes 60 firms which are traced over five years, resulting in 300 observations and selecting the needed data from several sources.
27 Information on GI is extracted from the European Patent Office (EPO), which is one of the three largest patent offices worldwide, including a classification scheme for green patents (Aragon-Correa et al., 2016; EPO, 2021). In accordance with other studies, the financial data is retrieved from Compustat Capital IQ (Aragon-Correa et al., 2016; Leyva-de la Hiz et al., 2021).
Moreover, missing data is selected from the Orbis database as well as from annual financial reports of the firms. Additionally, to secure comparability, all values are converted into USD.
Furthermore, for firm internationalization, the number of foreign countries the firm operates in is hand-picked for every year, analyzing official annual and strategic reports that are publicly accessible via the firms’ websites. Finally, information on the country-level control variable is retrieved from the World Bank database (Bermúdez-Edo et al., 2017).
4.2 Variables
4.2.1 Independent Variable
Depending on the research design, GI has been measured in various ways. Firstly, several studies employ own or adapted scales to measure different types of green innovation through surveys and questionnaires (Tang et al., 2018). However, this approach entails some limitations considering the scope of the research as well as the longitudinal design and is thus not suited for this study. Applying other quantitative measurements, they can be grouped into input measures (e.g., R&D, and other innovation expenditures), intermediate output measures (e.g., number of patents), direct output measures (e.g., number of innovations) and indirect impact measures (e.g., changes in eco-efficiency) (Bonturi & Pilat, 2009).
Moreover, there is an increasing number of studies that employ green patent data to operationalize GI, as this measurement contains several benefits (Aguilera-Caracuel & Ortiz- de-Mandojana, 2013; Przychodzen et al., 2020). Whereas green patents represent intermediate output data, not guaranteeing that they will always lead to the commercialization of the
28 innovations, the patents are clearly measurable, comparable between industries and extensively available, which are all crucial aspects for the cross-industry longitudinal design (Aragon- Correa et al., 2016; Leyva-de la Hiz et al., 2021; Przychodzen et al., 2020). The data is taken from the EPO, introducing a cooperative patent classification code for GI (Leyva-de la Hiz et al., 2021). Hereby, Y02 and Y04 classify all technologies and applications for the mitigation or adaptation against climate change as well as sustainable smart grids (Bermúdez-Edo et al., 2017; EPO, 2021). Only granted patents and the family representative of each innovation are selected for reliable results and to avoid double-counting on the European and national level (Aragon-Correa et al., 2016; Leyva-de la Hiz et al., 2021). Finally, to account for different degrees of R&D intensity as well as varying patent behavior across industries, the granted green patents are calculated as the percentage of all granted patents within a year (Przychodzen et al., 2020). The number of all annual granted patents by a firm is similarly selected from the EPO database.
4.2.2 Dependent Variable
For FFP as the dependent variable, the return on assets (ROA) measurement is used most in relation to GI, describing the extent to which the firm generates returns from its assets (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Putri & Soewarno, 2020). Conducting a comprehensive analysis, authors employ additional measures, such as operating margin (OM), return on employed capital (ROEC), return on sales (ROS), or return on equity (ROE) (Fujii, et al., 2013; Lee et al., 2016; Przychodzen et al., 2020). To ensure robust results, the analysis uses three measurements: ROA, ROE and ROS. The data is retrieved from the Compustat Capital IQ database, Orbis as well as financial firm reports. Moreover, a t+1 value is employed for the measurements, to account for the lagged influence of GI on FFP (Aragon-Correa et al., 2016;
Leyva-de la Hiz et al., 2021). Consequently, whereas the data on GI is selected for 2014-2018, the financial data is retrieved for the years of 2015-2019. Thereby, ROA is calculated dividing
29 net income by total assets, ROE is defined as net income divided by shareholders’ equity, and ROS is calculated dividing net income by total revenue (Aguilera-Caracuel & Ortiz-de- Mandojana, 2013; Lopes Santos et al., 2019; Przychodzen & Przychodzen, 2015).
4.2.3 Moderating Variables 4.2.3.1 Welfare State Regimes
The first moderator concerns the welfare state regimes, by employing dummy variables (Kim
& Zurlo, 2009; Kjeldsen & Andersen, 2013). As discussed above, two representative countries are analyzed for each type (Arts & Gelissen, 2010; Esping-Andersen, 1990; Powell &
Barrientos, 2004). Therefore, all firms from the United Kingdom and Ireland are grouped as
“1”, all firms from Germany and France as “2” and all firms from Sweden and Denmark as “3”.
Subsequently, dummy variables are introduced to test hypothesis 2a and 2b. Thereby, liberal welfare state regimes are taken as the base group, having one dummy variable representing conservative/corporate welfare state regimes and one representing social democratic states.
4.2.3.2 Firm Internationalization
The second moderator reflects the firm’s level of internationalization. Firm internationalization is measured by the number of countries the firm is operating in. This approach is evaluated to be superior compared to non-equity modes, such as exporting behavior, as it represents an increased international involvement (Brouthers & Hennart, 2007; Tse & Pan, 2000).
Furthermore, the number of operating countries allows for greater insights than looking at the number of foreign subsidiaries (Głodowska et al., 2019). In case of the latter, a firm with many subsidiaries in only one single foreign country would score similarly as a firm with internationally dispersed subsidiaries. Therefore, the annual and strategic reports of every firm for each year are analyzed to extract the number of countries the firm operates in.
30 4.2.4 Control Variables
To account for other aspects impacting FFP, five control variables are included. On the firm- level, studies highlight the importance to control for firm age, size and prior financial performance (Huong et al., 2021). Furthermore, with the study not focusing on one industry alone, a control variable for industry sectors is added (Aguilera-Caracuel & Ortiz-de- Mandojana, 2013; Huang & Li, 2017). Finally, on the country-level, the GDP growth is controlled for (Bermúdez-Edo et al., 2017; Jakobsson et al., 2018).
4.2.4.1 Firm-level
As firm size is argued to positively influence FFP, associating larger firms with higher levels of resources, the potential effect is controlled for (Chen & Liu, 2018; Tang et al., 2018).
Therefore, firm size is measured by the logarithm of total assets (Przychodzen et al., 2020;
Trumpp & Guenther, 2017). Furthermore, firm age is measured counting the years since the firm’s incorporation (Huang & Li, 2017; Tang et a., 2018). Older firms have been associated with greater levels of experience and learning capacity, thereby being more profitable (Aragon- Correa et al., 2016; Leyva-de la Hiz et al., 2021; Tang et al., 2018). Finally, analyzing the impact of GI on FFP, it is vital to control for prior financial performance. Thus, the ROA average from three previous years (2011, 2012, 2013) is calculated (Aguilera-Caracuel & Ortiz- de-Mandojana, 2013). The data needed for all firm-level control variables is retrieved from the Compustat Capital IQ as well as from the Orbis database.
4.2.4.2 Industry-level
The incentive for firms to conduct green innovation is said to vary with different industries (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Aragon-Correa et al., 2016; Leyva-de la Hiz et al., 2021). Therefore, the industries are categorized according to four sectors: material,
31 industrial, personal health, and technology (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013).
The sectors are defined by the following characteristics: material industries include mining, metal, and chemical functions; industrial industries concern activities related to automobiles, electrical equipment, aerospace, and machinery; personal health industries include everything regarding pharmaceuticals, food, personal products, and health care; and finally, technological industries describe all activities associated with semiconductors, software, and technology hardware (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013). Each firm is placed under one category, and subsequently, dummy variables are introduced to control for the industry type (Aguilera-Caracuel & Ortiz-de-Mandojana, 2013; Huong et al., 2021).
4.2.4.3 Country-level
To account for other country-level factors that could potentially divert the impact of the welfare state regime moderator, the differences in the national market size and growth are controlled for (Jakobsson et al., 2018). Thus, data from the World Bank is extracted, measuring the annual GDP growth (Bermúdez-Edo et al., 2017; Jakobsson et al., 2018). Thereby, the GDP is displayed in trillion USD, which is subsequently transformed with the natural logarithm (Białkowski et al., 2008; Jakobsson et al., 2018; The World Bank, 2021).
4.3 Statistical Methods
The STATA 17 software is employed for the statistical analysis. As the study follows a longitudinal design, the method is required to account for repeated measures and within-subject correlation (Aragon-Correa et al., 2016; Leyva-de la Hiz et al., 2021). In accordance with other studies, a Hausman Test was conducted first, subsequently endorsing fixed effects (X2 = 29.21, p = 0.0000) in the panel linear regression model (Aragon-Correa et al., 2016; Leyva-de la Hiz et al., 2021). Additionally, the fixed effects of the industries and years are controlled for
32 introducing dummy variables (Kohler & Kreuter, 2009). Moreover, to ensure that serial correlation within a firm does not divert the model across time, robust standard errors are utilized (Aragon-Correa et al., 2016; Baltagi, 2021). Finally, analyzing the effect of GI on FFP and including the moderators, the following equation is tested:
ROAint = a + b1 * GIint + b2 * WSRint + b3 * FIint + b4 * GIint * WSRint + b5 * GIint * FIint + b6
* Cint + bn * In + bt * It + eint
For robustness checks, two additional equations are employed:
ROEint = a + b1 * GIint + b2 * WSRint + b3 * FIint + b4 * GIint * WSRint + b5 * GIint * FIint + b6
* Cint + bn * In + bt * It + eint
ROSint = a + b1 * GIint + b2 * WSRint + b3 * FIint + b4 * GIint * WSRint + b5 * GIint * FIint + b6
* Cint + bn * In + bt * It + eint
The variables are defined in Table 1. The subscripts refer to a firm i that is observed in the data in industry n and year t. Additionally, a is the intercept and b represents the estimated coefficients. The firm- and country-level control variables are included in C. Moreover, In and It refer to the industry and year fixed effects, and e represents the error term.
