The Moderating Effect of Firms’ External Political Environment: A Study of
the CSR – Financial Performance Relationship.
Master Thesis
Student:
Grant Kargol / Student No 11374705
MSc. Business Administration, Strategy track
University of Amsterdam, Faculty of Economics and Business
Supervisor: Dr. Pushpika Vishwanathan
University of Amsterdam, Amsterdam Business School
Date:
23
rdJune 2017
Statement of originality
This document is written by student Grant Kargol who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of contents
Abstract ... 3
Introduction ... 4
Theory and hypotheses ... 8
Corporate social responsibility and stakeholder theory ... 8
Corporate social responsibility and financial performance ... 11
Corporate social responsibility and politics ... 14
The moderating effect of location and external political environment ... 17
Methodology ... 20 Sampling strategy ... 20 Dependent variable ... 21 Independent variables ... 21 Moderating variable ... 22 Control variables ... 24 Statistical model ... 25 Results ... 26
Descriptive statistics and correlation analysis ... 26
Regression analysis ... 29
Discussion ... 35
Major findings ... 35
Contributions of the study ... 37
Limitations and future research ... 38
Abstract
This study examines the effects of a firm’s external political environment on corporate social responsibility activity and subsequent financial performance. I posit that the amount which a firm will benefit financially from corporate social responsibility activities will be moderated by the external political environment for which the firm is headquartered. Further, I suspect the political views of stakeholders will be reflected as a corresponding level of firm CSR. This study utilizes data of the 500 largest publicly traded U.S. based firms, as well as public voting and election records. I find that the more Democrat-leaning a firm’s home state is the higher the firm’s CSR score will be. Additionally, I find that corporate social responsibility has a positive relationship with accounting-based financial measures. These results are in large part consistent with my predictions, and suggest that external political environment has practical implications for managers, and furthermore, is relevant for stakeholder theorists.
Key words: Corporate social responsibility (CSR); external political environment (EPE); corporate financial performance (CFP); stakeholder theory; Democrat; Republican
Introduction
Socially responsible activity has for decades been used to explain variations in financial performance of firms (Barnett & Salomon, 2006; Barnett, 2007; Cochran & Wood, 1984; Davis, 1973; Graves & Waddock, 1994; Hillman & Keim, 2001; Margolis et al., 2007; McGuire et al., 1988; Orlitzky et al., 2003; Ruf et al., 2001; Tang et al., 2012; Waddock & Graves, 1997; Wang et al., 2016; Wood & Jones, 1995). However, given the immense amount of research scholars still lack a thorough understanding of what the effects of CSR are on the financial performance of the firm, since previous studies have produced inconsistent empirical results (Margolis et al., 2007; Orlitzky et al., 2003; Waddock & Graves, 1997; Wood & Jones, 1995). Further, when a large body of research on the topic is considered in sum, 192 studies over 35 years, we see that the overall effect is significant but small with r=.132 (Margolis et al., 2007). These studies generally indicate a positive relationship, if any at all, with just 2% producing significant negative results (Margolis et al., 2007; Orlitzky et al., 2003). Meanwhile, the relationship is strongest when financial performance predicts CSR, the reciprocal to most research, which is focused on determining the effect on financial performance, i.e., reverse causality (Allouche & Laroche, 2005; Margolis et al., 2007; Orlitzky et al., 2003). Yet, after 35 years of research, there is still difficulty defining the construct (Barnett, 2007; Clarkson, 1995; McWilliams, Siegel & Wright, 2006; Wood, 1991; Wood & Jones, 1995); the definition continues to evolve (Carroll, 1999), and fundamental questions about the relationship still remain unanswered (Margolis et al., 2007). Nonetheless, researchers are still trying to make ‘a business case for CSR’ (Carroll & Shabana, 2010).
In order to further develop our understanding of the relationship between social performance and financial performance, it is critical to consider dimensions of CSR not yet isolated to improve conceptualization of the theoretical connection to CFP. An overlooked dimension with regard to CSR is the headquarter location of the firm, which has many
implications for the firm that can be directly observed including, access to markets, competition, workforce, and external political environment. Meanwhile, headquarters location has garnered attention in other studies, which have identified that firms located in the same geographic area are favored by local investors and experience a strong co-movement in stock returns (Coval & Moskowitz, 1999; Gao et al., 2011; Pirinsky & Wang, 2006). Diving deeper, the external political environment encompasses the political views of many stakeholders, including the firm’s employees, suppliers, shareholders, customers, and regulators (Di Giuli & Kostovetsky, 2014). Additionally, the external political environment has been shown to significantly influence the CSR rating of companies, that is, firms located in Democrat-leaning states score high on CSR, and firms in Republican states tend to score low (Rubin, 2008). As a result, firms who are subject to the same relative headquarter location, and thus external political environment, can be expected to have corresponding levels of CSR as well as similar financial performance.
Therefore, I intend to examine the moderating effect of the external political environment for the firm headquarter location on the relationship between corporate social responsibility and financial performance. Researchers contend that “the role of values…and political values in particular in shaping investments has been underexplored” (Hong and Kostovetsky, 2010, p. 1). Since a firm’s location has been shown to influence CSR level (Rubin, 2008) as well as influence financial performance (Burris, 1987; Coval & Moskowitz, 1999; Pirinsky & Wang, 2006), it is reasonable to suggest that a firm’s location may have a moderating effect between CSR and CFP. In their article, Di Giuli and Kostovetsky (2014) show that “on average, a firm headquartered in a state that gave 60% of the vote to Democrats has 0.11 standard deviations higher KLD score than a firm in a state that gave 50% of the vote to Democrats” (p. 169). Further results from Di Giuli and Kostovetsky (2014) “indicate that it is the preferences toward CSR of stakeholders living in the state rather than the laws and
regulations of the sate that explain why firms in Democratic states have higher KLD scores” (p. 169). While it is also the preferences of local stakeholders (investors) that explain why there is local co-movement of equities (the covariance structure of underlying security returns), known as the geographic segmentation of domestic capital markets (Pirinsky and Wang, 2006). Demonstratively, firms with lower KLD scores perform better after Republican election presidential victories, a possible explanation of which is that firms which are socially responsible choose to locate their firm in more socially responsible states (Di Giuli and Kostovetsky, 2014).
The purpose of this research is to provide a deeper understanding of the causal mechanisms of the positive relationship between corporate social responsibility and financial performance, while taking into account the moderating effect of a firm’s headquarter location and hence external political environment. Consequently, this paper intends to answer the following research question:
What is the moderating effect of firms’ headquarters location on the relationship between corporate social responsibility and financial performance?
In order to investigate this question, I analyze metrics from the largest firms in the United States. A composite CSR score is determined based upon data from the widely used KLD dataset (Graves & Waddock; 1994; Hillman & Keim, 2001; Johnson & Greening, 1999; Margolis et al., 2007; Ruf et al., 2001; Waddock & Graves, 1997). External political environment score is computed based upon prior research from Di Giuli and Kostovetsky (2014). To determine financial performance, I follow two distinct methods, each following prior research. To test each, a proxy is used; Tobins Q is used for market performance, while return on equity (ROE) is used for accounting performance. Either of these ratios lend unique
insights. By testing market performance, sentiment of investors is reflected, and thus elements such as a scandal can have an effect. Further, market performance represents value created for shareholders. While the accounting measures indicate how the firm is performing fundamentally.
The research conducted and analyzed in this paper makes several contributions to the existing literature. First, it addresses the under researched subject of external political environment as it relates to social responsibility and financial performance. By doing so, practical implications for firms across the United States in terms of CSR and financial performance will be realized. Thus, generating insights as to how a firm should incorporate CSR based upon the state in which they are headquartered. Theoretically, this research extends the reach of stakeholder theory. By rooting the research of external political environment in stakeholder theory it is ensured the findings will yield meaningful results; thus, providing a point for continued discussion and research going forth. Additionally, it adds to previous studies, which have exhaustively attempted to prove the relationship between corporate social responsibility and financial performance by adding a further layer of analysis. This study goes on to distinguish between accounting and market-based financial measures on a theoretical level rather than an operational level. The result is a more clearly constructed and defined model, which has implications that are more clearly distinguishable. Finally, this study presents a potential causal mechanism for the ‘location bias’ of investors.
Theory and hypotheses
The following section discusses the main insights of the relevant literature on corporate social responsibility (CSR) as it is related to stakeholder theory, financial performance (CFP), political preference, and external political environment (EPE), while presenting the hypotheses of this research. First, I introduce the nature of corporate social responsibility and its roots in stakeholder theory, as it is relevant to this research topic. Next, I discuss the corporate social responsibility – financial performance relationship and possible causal mechanisms, at which point hypothesis 1 is introduced. Then, I introduce the concept of external political environment and present hypothesis 2. Continuing, related literature on the how political preference and location influence financial performance lead to the development of hypothesis 3. Subsequently, the moderating effects external political environment has on the CSR – CFP relationship is discussed, and hypothesis 4 presented. Finally, I outline several related variables that may have an impact on the moderating effect.
Corporate social responsibility and stakeholder theory
Strategy literature has paid much attention to how firms allocate their resources and increasingly, factors such as social issues have impacted how the resources are distributed (Prahalad & Hamel, 1994; Waddock & Graves, 1997). Over time, social issues have gained relevance due to increased expectations of stakeholders (Freeman, 1984), and the increasing popularity of stakeholder theory (Clarkson, 1995; Freeman, 1984; Hillman & Keim, 1995; Jones, 1995; Ruf et al., 2001; Wood & Jones, 1995). With increased attention from stakeholders focused on social issues, firms tend to spend more money on programs for things like environmental management, community welfare, and employee benefits (CSR) (McGuire et al., 1988). Advocates believe that in turn, engaging in CSR attracts investors, adds to
critics of stakeholder theory, namely Milton Friedman (1970), among others, argue against this practice stating that managers only responsibility is to increase shareholder’s wealth. Raising the question of what responsibility managers of a firm have to their shareholders, and who else, if anyone at all.
To whom else the firm is responsible to is at the core of the debate, which produced stakeholder management theory, and as a result directly effects how resources are allocated to CSR activities. One side of the argument, as mentioned prior, is that of Milton Friedman (1970), reasoning that the primary responsibility of the firm is ‘to maximize the value created for the shareholders of the firm’ (Barnett & Salomon, 2012; Carroll, 1999; Rupp et al., 2010). This rationale however, has little utility for investing in socially responsible activities, and thus, CSR changed the approach to who the firm was responsible (Rupp et al., 2010). Rather, stakeholder theory as its name entails has the belief that firms are responsible for all stakeholders, (parties who are affected by the organization), not just shareholders (Carroll, 1991; Freeman, 1984). Stakeholder theory proposes that firms who partake in stakeholder management should perform better than those that do not (Harrison et al., 2010). The view that stakeholder management provides financial benefits is fundamental for CSR, as it is the basis for why firms engage in CSR activities, provides justification for CSR expenditures (Barnett, 2007; Wood & Jones, 1995), and thus provides a theoretical foundation to investigate CSR activities.
Stakeholder theory lends itself as an outline to whom the firm is responsible, thus answering managers long awaited question, how to allocate resources (Prahalad and Hamel, 1994). Even with stakeholder theory though managers still must make judgments on what CSR activities to allocate resources to. Interestingly, research indicates that the values and opinion of managers plays a large role in determining how much and the type of CSR they will pursue,
as they do not have a clear direction (Hong & Kostovetsky, 2012; McWilliams & Siegel, 2001; Waddock & Graves, 1997).
Further, stakeholder management theory is not only a determinant of what CSR activities a firm will engage in, but is also is a crucial factor in implementing them (Noble, 1999). Often, stakeholders take a passive role in CSR as they are outsiders to the firm, and are the one who is being affected by the firm’s activities. When discussing strategy implementation however, stakeholders are in an active operational role in the form of employees (Lev et al., 2010). As employees, stakeholders are an important part of implementing an integrated CSR strategy (Alagaraja and Shuck, 2015; Glavas, 2016; Greening & Turban, 2000; Lev et al., 2010). A major factor for an organization to successfully implement an integrated CSR strategy is alignment within the firm (Alagaraja and Shuck, 2015; Glavas, 2016). To preform its best, the firm must have shared goals from top down amongst all employees, so that the strategic vision is uniform throughout the firm (Alagaraja and Shuck, 2015). Employees’ contributions to daily operations give them a main role in strategy implementation, and ultimately make a significant impact on the success of the firm (Glavas, 2016; Greening & Turban, 2000). This is true for all strategies a firm implements, but is especially important for CSR activities as stakeholders play an additional role, and can serve as catalysts in CSR implementation (Hemingway, 2005). One reason employees can further enhance CSR implementation is because as the organization implements CSR activities, the employees are motivated as part of a bigger goal (Deci & Ryan, 1985; Glavas, 2016), due to the alignment between their personal values and the values and goals of the organization (Aguinis & Glavas, 2013). That is, employees will recognize the firm as having the same values as they personally have, resulting in them feeling that they can ‘bring their whole selves to work’ and ultimately lead to them performing better (Aguinis & Glavas, 2013; Kahn, 1990). Thus, when employee (stakeholder) values are congruent with those of the firm, is when firms will perform best (Alagaraja and
Shuck, 2015; Aguinis & Glavas, 2013; Glavas, 2016; Greening & Turban, 2000; Lev et al., 2010).
A firm’s CSR activities are affected by more than just internal stakeholders though; external stakeholders also play a role in determining the CSR strategy a firm will pursue as well as the success of said strategy. Increasingly, shareholders are growing concerned about CSR and its affect on firm performance (Porter & Kramer, 2006). As a result, firms must appease the shareholders and initiate CSR. Another example of outside stakeholders affecting a firms CSR strategy are activists who may protest the company, or customers boycotting their products due to them being socially irresponsible. The increase in attention among the public to CSR has caused almost all companies to engage in at least some form of CSR (Sen & Bhattacharya, 2001).
Corporate social responsibility and financial performance
Research surrounding CSR is now investigating the relationship between CSR and CFP, producing results which are inconsistent (Barnett & Salomon, 2006; Barnett, 2007; Cochran & Wood, 1984; Davis, 1973; Graves & Waddock, 1994; Hillman & Keim, 2001; Margolis et al., 2007; McGuire et al., 1988; Orlitzky et al., 2003; Ruf et al., 2001; Tang et al., 2012; Waddock & Graves, 1997; Wang et al., 2016; Wood & Jones, 1995). Multiple meta-analysis based on a plethora of studies have been conducted in recent years. Orlitzky et al. (2003) for example, found that there is a positive correlation between social performance and financial performance based on 52 studies spanning two decades. In an even larger meta-analysis comprising 192 studies over 35 years, Margolis et al. (2007) concluded that there is a small but significant positive effect on CFP. Studies have also observed the relationship through different financial instruments as well, Jiao (2010) found that both Tobin’s Q and ROA both are positively correlated. While the Orlitzky et al. (2003) study found that there is a
stronger relationship when considering accounting based financial metrics rather than market based metrics. These results are consistent with many other studies findings on CSR and financial performance (Preston and O’Bannon, 1997; Hillman and Keim, 2001; Bebchuk et al., 2013), but still there is not unanimous consensus. Further, these studies use a multitude of measures for financial performance. As there are more financial metrics and ratios than can be named, they are generally grouped into two main categories, accounting-based and market-based. This is problematic because conceptually these measures are unique, as accounting measures are thought of as current or short term indicators of financial performance, while market measures are future or long term (Hoskisson et al., 1994). Also, market-based values are more reflective of investor sentiment and public reaction to CSR. For example, an environmental scandal may cause significant impact on a firm’s stock price, while it may not be reflected in annual financial statements. Furthermore, the same CSR activity could positively influence one and negatively influence the other. Such as could be the case if a firm elected to raise wages for employees. Enhancing the reputation of the firm could result in the stock price rising, but the increase in expenses will adversely effect profitability. Thus, in an attempt to add a further layer of analysis to existing studies and justify later hypotheses, the first hypotheses of this research are as follows:
H1a: Corporate social responsibility will positively influence firm accounting-based financial performance.
H1b: Corporate social responsibility will positively influence firm market-based financial performance.
Previously, I discussed stakeholder theory as it relates to CSR. Continuing the review, stakeholder theory again plays a central role as it is the main model in existing CSR-CFP
literature (Freemen, 1984; Donaldson and Preston, 1995; Mitchell et al., 1997). Those who support the theory believe that “the favorable social performance is a requirement for business legitimacy and that social and financial performance tend to be positively associated over the long term” (Preston and O’Bannon, 1997, p.420). Further, by paying attention to stakeholder interests, a firm can increase its reputation and financial performance (Preston & O’Bannon, 1997; Jensen, 2001). Similarly, Waddock (1997) argues that positive relationships with stakeholders will help firms gain better financial performance. To achieve this type of result in practice a firm could take action in many ways, for example, they could have a CSR policy on employee relations, thus boosting employee motivation, or could keep community initiatives high and benefit from good relations with local government.
Still though, there is a tradeoff between focusing on all stakeholders and the financial performance of the firm. Accordingly, Baron (2001) introduces the concept of “profit-maximizing CSR,” that is, a firm engages in CSR with the goal of “profit-maximizing profits. In this way, CSR is a tool for firms to enhance performance and competitive advantage while still being grounded in stakeholder management. Baron (2001) lists many ways that CSR can contribute to gains in firm value through stakeholders: One, consumers could value CSR and be willing to pay a premium for the products the firm provides. Two, CSR can help recruiting employees as well as motiving existing ones. Three, some investors known as green investors, may pay a premium for a firm which have a good CSR reputation. Four, firms can substitute some advertising and branding by having a good CSR reputation. Another article expressing a similar idea; McWilliams and Siegel (2001), argues that consumers seek out products that are produced in socially responsible ways; thus, firms can enhance financial performance through “consumer-oriented CSR”. This rationale is supported by other studies that show how CSR affects firm performance through stakeholders, including increasing demand by attracting
socially conscious consumers (Benabou & Tirole, 2010), and having good relationships with interest groups and the environment (Konar & Cohen, 2001).
Consensus among scholars who advocate for a stakeholder view is that firms must pursue the interests of both internal and external stakeholders to maximize their value (Freemen, 1984, Donaldson and Preston, 1995; Mitchell et al., 1997; Jensen, 2001; Preston & O’Bannon, 1997). Furthermore, a firm’s CSR performance is an extension of its reputation among internal and external stakeholders, and can contribute to additional financial returns (Porter & Kramer, 2006). Given that stakeholders and their perception of firm reputation are credited with impacting firm performance, it is surprising there is not more research done investigating which outside factors influence stakeholders and further influence firm performance as a result.
Corporate social responsibility and politics
Many studies have investigated the relationship between CSR and political activity (Schuler, 1996; Holburn & Van den Bergh 2013; Boddewyn & Brewer 1994; Hansen and Mitchel, 2000; Hillman and Keim 1995; Hillman and Hitt 1999). Naturally, social responsibility and politics are closely related, as government rules and regulations in large part determine the landscape which firms operate in, and therefore must develop a CSR strategy based upon. Further, CSR has become a politicized topic, as corporate responsibility issues often coincide with political issues such as environmental protection, product safety, and employee rights. Similarly, support for CSR varies across political parties (Di Giuli & Kostovetsky, 2011; Rubin, 2008). Thus, it can be said that ones “political affiliation is a natural measure of preference for social responsibility” (Di Giuli & Kostovetsky, 2011, p. 159). Research shows that the Democratic Party platform is more in line with the values and principles of CSR related issues such as environmental protection, sustainability,
anti-discrimination laws, employee wellbeing, and community support (Di Giuli & Kostovetsky, 2011; Rubin, 2008). Contrarily, the Republican party has values that are less in line with CSR, favoring decreased regulation on topics such as environmental protection and worker rights.
The impact politics has on a firm can be explained in part with stakeholder theory. A firm’s major stakeholders are the customers, employees, suppliers, and shareholders to name a few. Of these stakeholders, all tend to cluster around the firm headquarters, i.e., live near the headquarters (Porter, 2000; Coval & Moskowitz, 1999). Stakeholders cluster around the firm headquarters for many reasons, employees must live close to work, suppliers are nearby for logistical reasons, and shareholders tend to invest with firms they are more familiar with (Coval & Moskowitz, 1999). Therefore, a significant portion of stakeholders live near the firm’s headquarters, and in turn comprise one factor of the firm’s external political environment.
The external political environment is defined by Di Giuli and Kostovetsky (2014) as “the political views of the firm’s employees, suppliers, shareholders, customers, and regulators” (p. 160). Additionally, pressure from stakeholders’ influences firms to engage in CSR (Sen & Bhattacharya, 2001). Hence, as a result of the political views of their stakeholders, firms may enact CSR differently than they otherwise would. It is important to note that the scope of the environment could be of many magnitudes; state level, country or regional level all could have impact on the firm. However, for this study we will focus within the United States, and how the political environment within a state effects firms. By investigating at state level, the study will be narrow enough to analyze the impact of political partisanship, and varying degrees of ideology while still having meaningful results.
Consider a firm in a Democratic-leaning state. As a result of the company being in a Democratic state, the stakeholders which cluster around the firm, will be proportionately Democratic. Therefore, the employees who work at the headquarters will be more in favor of CSR activities. This could result in more attention being given to CSR due to vocal employees
who want their company to be reflective of them. Also, operationalization of CSR initiatives may be better as a result of the alignment of values between the employees and their work. Customers, who also tend to cluster near the headquarters, may encourage a firm to engage in CSR activities. The firm’s product may be sought out or supported because of the positive reputation you have for CSR. In contrast, customers can also negatively respond to a firm’s CSR through means such as boycotting products that may be made in socially irresponsible ways, or protesting the firm completely. Additionally, regulators and local governments are also stakeholders. In a Democrat-leaning state regulators are more likely to enforce strict regulations on CSR activities; thus, forcing firms to adapt or be faced with penalties. Conversely, lawmakers who are Democrats may create tax incentives for firms who perform exceptionally with regard to CSR, or allow things like environmental expenditures to be tax deductible.
Contrarily, firms in heavily Republican states will experience much less of a push towards CSR. Employees are more likely to be skeptical of environmental issues. Customers are less likely to be concerned with how a product was made, or if the company uses renewable energy. Regulators are in favor of decreased regulation, while local governments generally don’t support issues such as a higher minimum wage or paid paternity leave.
As was illustrated, it would behoove firms to have CSR levels that are representative of their external political environment. As a result, it is expected that the pressure from stakeholders will have influenced firms’ CSR levels; thus, given the existing literature and the reasoning presented here the next hypothesis is as follows:
H2: There will be a positive relationship between the proportion of the external political environment that is Democrat and CSR level.
The moderating effect of location and external political environment
Finance literature observes that there is preference among investors to invest in firms which are headquartered locally, and further, that the stock prices of firms geographically close exhibit co-movement (Coval & Moskowitz, 1999; French & Poterba, 1991; Gao et al., 2011; Huberman, 2001; Pirinsky & Wang, 2006; Tesar & Werner, 1995). Pirinsky & Wang (2006) argue that the trading patterns of local residents are the cause for the geographic co-movement in stock prices between firms, as they find fundamentals do not have an effect. That is, fundamental financial metrics such as earnings or profit for companies in the same geographic area do not cause the co-movement in stock prices, rather, ‘location bias’ does. Explanations for the local bias of investors is limited and uncertain though, pointing to things such as information asymmetry (Coval & Moskowitz, 1999) and word of mouth (Brown et al., 2004; Pirinsky & Wang, 2006). Political views could be attributed to location bias as there is a reasonable geographic component to political ideology, as well as, additional finance literature that indicates there is a relationship between political views and CFP (Hong & Kostovetsky 2012). Their research shows that mutual fund managers who are Democrats invest more in firms that are socially responsible than Republican mutual fund managers. Stating that because Democrats care about CSR and comprise such a large proportion of investors, that they will have an important effect on stock prices (Hong & Kostovetsky, 2012).
Given that political views correspond to preference of CSR (Hong & Kostovetsky, 2012), and that it is common practice to divide political views geographically, then it is reasonable to say that sentiment toward CSR also has some level of a geographic component. It was discussed that stakeholders cluster, and are more likely to reside nearby the firm’s headquarters (Porter, 2000; Coval & Moskowitz, 1999); thus, it was established that the stakeholders are representative of the firm’s external political environment. This is significant because stakeholders in many ways impact the performance of a company, both internally
through employees and externally through suppliers, shareholders, and the local community (Aguinis & Glavas, 2013; Barnett, 2007; Baron, 2001; Waddock, 2007). The effects on the firm are two fold, impacting both accounting and market based financial measures. On one hand there are things such as tax deductions or regulatory penalties. These directly impact the bottom line for firms, and are thus reflected in the accounting statements and accounting measures. On the other hand, customer and investor sentiment can directly impact the stock price of the firm. Additionally, productivity gains and improvements in operationalization of strategy can come from alignment between firm and employees’ values (Aguinis & Glavas, 2013).
The two channels through which external political environment influences financial performance are useful to demonstrate how the moderation effect takes place. In order to show this effect, we will consider two scenarios. In the first, ‘Firm X’ is going to be switching to clean energy and investing in a new fleet of electric vehicles. Firm X could be located in either a Democratic or Republican state. In the Democrat-leaning state Firm X will be allowed to deduct the cost of the vehicles from their state income taxes, whereas in a Republican state Firm X would not benefit from such a law. In the second scenario let’s consider ‘Firm Y’. ‘Firm Y’ has recently been caught exceeding toxic emissions levels and now must deal with repercussions. In the Democrat-leaning state Firm Y is a top story on local news. Customers are boycotting their products because of their personal views regarding the environment, and investors sell the stock due to panic and uncertainty. While in the Republican state not much notice is given to the issue, as the local community doesn’t have concern for environmental issues, and investors feel its business as usual.
This example gives a simplified illustration of how the external political environment could influence the degree to which CSR impacts financial performance. It is important to note that this example distinguished between events that effect accounting and market performance,
however, they most certainly are interdependent. In the first example, the external political environment determined a tax benefit. Where a firm with the same level of CSR would be better off financially in a Democrat-leaning state. While in example two, the opposite was true. In this example the external political environment moderated how a scandal was perceived by stakeholders, and thus, in turn how it affected the stock price. Following this rationale leads to hypothesis 3 which is as follows:
H3: The external political environment of a firm will have a moderating effect on the CSR-CFP relationship
Methodology
This section provides a detailed summary of the research design and how the research was conducted. In this section the research design and subsequent variables are explained. First, the data collection method and sample are discussed. Then, in detail, the dependent, independent, moderating, and control variables are reviewed. Finally, the statistical methods used to analyze the data are outlined.
Sampling strategy
This study utilizes data gathered from Wharton CRSP – Compustat Merged database. The CRSP – Compustat database provides the financial and industry data from the year 2014. The dataset was collected for all firms in the Fortune 500 list, an annual report of the 500 largest U.S. based firms by total revenue size. Of these records, several firms were acquired or merged during the year; thus, they were removed from the sample. Additionally, non-profit firms are also removed as there are not comparable to the rest of the sample. CSR data is gathered from the Kinder Lydenberg and Domini social audit (KLD) database following other studies (Choi & Wang, 2009; Garcia-Castro et al., 2010; Griffin & Mahon, 1997; Godfrey et al., 2009; Hillman & Keim, 2001; Johnson & Greening, 1999; Jo & Harjoto, 2011; Nelling & Webb, 2009; Ruf et al., 2001; Turban & Greening, 1996; Waddock & Graves, 1997; Waddock & Graves, 2000). Political data is retrieved from the Federal Election Commission, the government association tasked with reporting election results. The year 2014 was selected because it was the most recent year that complete financial data and KLD data were available for. The final sample consists 319 firms and 17226 firm level observations during the 2014 year.
Dependent variable
Financial Performance. The dependent variable in this study is the firm’s financial performance measured by both accounting-based and market-based metrics. The proxy used for accounting based financial performance is return on equity (ROE), and is consistent with previous studies (Bowman & Haire, 1975; Cowen et al., 1987; Griffin & Mahon, 1997; Johnson & Greening, 1999; Mahoney & Roberts, 2004; Roberts, 1992; Ruf et al., 2001; Waddock & Graves, 1997). To measure market based performance I utilize Tobins Quotient (Tobins Q). The quotient is a ratio of the market value of a firm’s assets and their replacement value, and is a commonly used metric to gauge market financial performance across firms (Choi & Wang, 2009; Dowell, Hart, & Yeung, 2000; Garcia-Castro, et al., 2010; Jo & Harjoto, 2011; King & Lenox, 2001; Konar & Cohen, 2001; Surroca et al., 2010). Further, I use the logarithm of Tobins Q as it is more normally distributed (Hirsch and Seaks, 1993), and is commonly used among scholars (Chung and Pruitt, 1994; Lang and Stulz, 1994; Youn et al., 2015). Some data points were not available for Tobins Q as a result, thirty-eight firms were removed from the data set for this reason. Additionally, due to systematic accounting and operational differences amongst financial firms, all financial firms (SIC code 600-680) were removed from the data set.
Independent variables
CSR level. The Kinder, Lydenberg, and Domini database has become one of the most widely used measurements of corporate social responsibility available today, and was chosen because all companies in the Fortune 500 are rated. The KLD dataset is comprised of 67 dimensions in total, for my research only 36 dimensions (9 strengths and 27 concerns) had data available for the entire sample of companies. The database records the variables with values as 1 or 0 indicating “yes” or “no” if the company has either strength or concern. Following
common practice, the total number of strengths and concerns are each calculated (Di Giuli & Kostovetsky, 2014). Concerns are multiplied by negative one as they are a negative indicator. Finally, a total CSR score is calculated, by subtracting the total concerns from the total strengths (Di Giuli & Kostovetsky, 2014). For all of these scores, a higher number represents better corporate social responsibility. From the sample the mean concern score was 1.38 while the average strength score was 1.83 and the average total score was .47. Finally, the CSR scores are standardized so that the mean is zero and the standard deviation is one thus simplifying the interpretation of regression coefficients (Di Giuli & Kostovetsky, 2014).
Moderating variable
External political environment. Each firm in the sample has their home state recorded from the Fortune website. An issue that arose is that of incorporation vs. headquarter. Many firms are incorporated (i.e. registered, listed, etc.) in a state which they have no physical presence. To illustrate this point, the most extreme case is Delaware, which has only 900,000 residents, currently has over 1 million companies incorporated there. Yet there is often little more than a piece of paper tying a firm to the state. External political environment is being researched due to its composition of firm stakeholders; thus, most important is the physical location of the firm headquarters, not incorporation state. As a result, only the physical headquarters’ location provided by Fortune is recorded. In order to assign each firm a score, I compute a composite external political environment (EPE) score for each state and use it as a proxy. Following the work of Di Giuli and Kostovetsky (2014) I compute an EPE score based upon the following dimensions: state level results of the most recent presidential election, the partisan makeup of the states congress, the political party of the governor, and the political party of the state legislature. Each dimension is calculated as a % Democrat, so a score of 1 would represent being 100% Democrat (liberal) environment, and a score of 0 would represent
being 0% democrat; thus, a Republican (conservative) environment. In this study only two political parties are identified, Democrat and Republican as these are the only major political parties in the U.S. Further, Independent party votes makes up less than 2% of total presidential vote, and only 2 of 535 congressmen from the 113th congress. All the data is recorded for the 2014 year, since presidential elections only occur every 4 years the most recent election data available, at the time of this study, was 2012. The first dimension, president vote D%, is the percent of the vote received by the Democratic candidate in the 2012 presidential election. Second, Congress delegation, is composed of both the members of the senate and house of representatives (congressmen). Each U.S. state has 2 senators and a varying number of congressmen. The proportion of senators that are democrat is multiplied by .5 then added to the proportion of congressmen that are democrat multiplied by .5. The final dimension is State Government, which is composed of the governor and the state legislature. If the governor is (not) a Democrat, then a value of one (zero) is recorded and multiplied by .5 then added to the proportion of the state legislature that is Democrat controlled multiplied by .5. The average state in the U.S. receives a score of .46, with a range from .90 (Hawaii) to .09 (Wyoming). Not all states are home to a Fortune 500 firm however; thus, the range of scores that are included in the sample are from .11 (Oklahoma) to .88 (Rhode Island). The mean EPE score for a firm in our sample however is .52, meaning that the average firm in our sample is in an environment that leans slightly Democrat-leaning. This is consistent with the results of Di Giuli and Kostovetsky (2014). Finally, I normalize the EPE score to have a mean of zero and standard deviation of one. Di Giuli and Kostovetsky (2014) also discuss the possibility that presidential election results alone will accurately predict CSR scores. They posit that presidential election results may more purely show the preference toward CSR among stakeholders rather than composition of law makers or regulators. Thus, in addition to the external political environment
score which was calculated, presidential election will also be used as another operationalization of the external political environment variable.
Control variables
Firm size. Many studies have measured the impact firm size has on CSR (Darnall et al., 2009; Lepoutre & Heene, 2006). Large firms incur bias with regard to CSR as they are more visible, and thus, are subject to being over reported. Further, firm size can directly effect how well a firm can manage the cost and labor burden associated with CSR. Since size can be an important determinant of CSR we control for it, both in terms of employees and revenue. The logged value of the number of employees and revenues are used so that the values are normally distributed; thus, a percentage change in firm size will have a uniform effect (Audia & Greve, 2006).
Organizational slack. Slack resources can be defined as “potentially utilizable resources that can be diverted or redeployed for the achievement of organizational goals” (George, 2005, p.661). Researchers believe that slack resources may have a moderating effect on the CSR and CFP relationship (Hong & Anderson, 2011; Waddock & Graves, 1997). Slack resources address the issue of reverse causality, as increased financial performance resulting in slack resources allowing the firm the luxury to invest in CSR. Thus, a control is added for slack resources.
Industry. Similar to firm size, industry has a direct impact on the reputation of a firm’s CSR, and thus, lends to reporting bias especially when it comes to concern measures of CSR. For example, some industries are more harmful to the environment (oil and gas) as compared to those that are perceived to be more environmentally friendly (technology). Furthermore, industries are subject to specific regulations that address CSR issues such as pollution restrictions or worker compensation. Using standard industrial classification codes (SIC code)
I categorize the firms by industry. Of those, two were identified as being problematic, and having potential to influence the results. First, all financial firms (SIC code 6000-6799) were removed from the dataset. Due to systematic accounting and operational practices among financial firms their financial performance is incomparable to the rest of our sample and are therefore removed. Further, upon review I found the sample consists of 138 mining industry firms. The mining sector, as it is classified by SIC code (1000-1500), includes industries such as, oil drilling, gas exploration, and metal mining. For reasons such as over reporting and adverse regulatory effects, a dummy variable is added for firms which have a SIC code between 1000-1500. When considering the mining industries, negative relationship with CSR and large composition of the sample, it is logical to assume that industry effect is likely to influence the results. For that reason, I also control for the mining industry (SIC code 1000-1500). To do so, I added a dummy variable for SIC code 1000-1500, which can be seen in Table 1, 2, and 3 listed as Industry.
Statistical model
The following statistical analysis is performed in SPSS Statistics. Initially a frequency check is run to ensure no missing values. Then, both CSR score and EPE score are standardized so that each have a mean of zero and a standard deviation of one. Upon validating the data, I run an OLS regression. Then following Di Giuli and Kostovetsky (2014), I subsequently add control variables to examine if industry, firm size, or organizational slack influence the effect. Finally, using the PROCESS add-on, I run a moderation test for hypothesis 3.
Results
In the following section I will report the results of this research. First, I present the descriptive statistics of the data to provide insights of the sample as a whole. Then, a Bivariate correlation analysis is performed and significant correlations are reported. Subsequently, multiple OLS regressions are performed to test hypothesis 1 and 2. Finally, a moderation test is run for hypothesis 3.
Descriptive statistics and correlation analysis
The complete set of descriptive statistics and correlations I report can be found in Table 1. A bivariate (Pearson) correlation analysis was performed on all variables discussed in this research, and the coefficients with significance at p = 0.05 level or lower are presented; additionally, some of the descriptive statistics are reviewed. First, the correlations between the control variables and all other variables are discussed. Then, I continue on with the correlations between the moderating variables, and both dependent and independent variables. Concluding with the correlation between independent and dependent variables.
In this study four controls are used, two controlling for size (revenue and employees), one for industry effects, and another for organizational slack. The strongest correlation shown in this study is between the revenue and CSR concerns (r = 0.494). Revenue was correlated less strongly to CSR strengths (r = 0.259) and similarly to total CSR score (r = -0.237). The correlation to total becomes negative because concerns is a negative indicator, and is thus subtracted from concerns. Industry on the other hand is positively correlated to CSR score (r = 0.195) as well as strengths (r = 0.336) but shows no significance to concerns. With regard to employees there is correlation between strengths (r = 0.245) and concerns (r = 0.217) but not total score overall. Organizational slack contrary to other controls is negatively correlated to
Just Tobins Q is correlated to slack (r = 0.140) and employees (r = 0.138). The moderating variables are correlated to all controls but revenue. The external political environment is correlated to organizational slack (r = 0.171) and also employees (r = 0.168). While the presidential election results show slightly weaker correlations to employees (r = 0.151), the correlation is stronger for organizational slack (r = 0.188). Additionally, presidential election results are correlated to industry effects (r = 0.120).
The main moderating variable, external political environment, as well as a component dimension of the score, presidential election results is reported. I include presidential election results following the work of Di Giuli and Kostovetsky (2014) whose results indicate election results may be particularly predictive of CSR scores. The mean external political environment score for our sample is 0.521 on a scale with 1 being fully Democrat (liberal) and 0 being fully Republican (conservative). The standardized scores used for regression analysis are also included in Table 1, yet correlations are not effected. With regard to the dependent variable measures, Tobins Q shows significant correlation to the main moderating variable as well as its component dimension. With the correlation to the Q being slightly stronger for the external political environment as a whole (r = 0.204) than for the presidential election results (r = .185). The independent variable, CSR, is also correlated to EPE (r = 0.160). Subsequently, presidential election results are also correlated to CSR, but to a stronger degree (r = 0.212). Further both components of CSR, strengths (r = 0.135) and concerns (r = -0.117) are correlated to the presidential election results.
Finally, considering the correlation with the independent variable and dependent variable, only one of the dependent variable measures shows correlation. Return on equity is correlated with total CSR score (r = 0.140) and slightly less with CSR strengths (r = 0.121).
Regression analysis
Following this section Table 2, Table 3 and Table 4 present the complete results of the regressions and moderations tests performed in this research. Table 2 presents the results with return on equity as the dependent variable, while Table 3 presents the results with the log of Tobins Q as the dependent variable. Table 4 presents the results from hypothesis 2 where external political environment is the dependent variable and CSR is the independent variable. The empirical study starts by regressing corporate social responsibility and financial performance in order to confirm a causal link between the two, and further confirm the relationship among the sample. Hypothesis 1 states that corporate social responsibility will positively influence firm financial performance, with H1a making this statement with regard to accounting performance, and H1b with regard to market based performance. In Table 2 and 3 I utilize a hierarchical OLS regression to gauge the effect of CSR on financial performance, controlling for industry, organizational slack, and firm size. In Model 2 regressing CSR and ROE, the result is a coefficient of 0.165, significant at the 1 percent confidence level. Further, the R^2 of 0.039 and a model F of (F = 2.562); which is significant at the p < .05 level, validate the models effect. This result is generally consistent with prior studies that found a very small but significant relationship (Margolis et al., 2007). The coefficient means that an increase of one on CSR score will result in an 0.165 increase of ROE. Therefore, hypothesis 1a, that corporate social responsibility will positively influence firm accounting based financial performance, is supported. The same analysis was performed using log of Tobins Q as the dependent variable rather than ROE in Table 3. In contrast to ROE, Tobins Q shows coefficients that are significant for all controls aside from organizational slack. Yet the main relationship between CSR and Tobins Q shows no significance and further the coefficient is nearly zero (0.002). As a result, considering the coefficients in Model 5, there was no support found for hypothesis 1b based upon this sample.
Hypothesis 2a states that there will be a positive relationship between a firm’s external political environment and CSR level. As was mentioned previously, the effect of presidential election results is also investigated independently from external political environment; thus, Model 9 uses presidential election results as the dependent variable. Model 8 shows that there is a positive interaction between external political environment and corporate social responsibility. The coefficient for the interaction is 0.121 and is significant at the p < .01 level. Further, the model F (F = 10.355) is significant at the p < .001 level, thus, the R^2 value of 0.142 can be relied upon. This suggests that with our controls, a firm’s external political environment explains 14.2% of variation in observed CSR score. Therefore, there is significant evidence that hypothesis 2a is supported. Indicating that a 1-point increase in external political environment will result in a .121 rise in CSR score. Additionally, I test the hypothesis using presidential election results rather than external political environment. Similar, but stronger results are yielded. Presidential election results showed a coefficient for the interaction of 0.161 which was significant at the p < .001 level; greater significance than was shown for the relationship with external political environment. Further, the model F (F = 11.251) is significant at the p < .001 level, thus the R^2 of .152 is significant. Hence, hypothesis 2b is also supported. This result is consistent with Di Giuli and Kostovetsky’s (2014) suspicion that presidential election results will be more predictive of CSR than external political environment overall.
Finally, I test hypothesis 3 by performing moderation tests that can be seen as Models 3 and 6 in Tables 2 and 3. Model 3, testing the moderating effect of external political environment on the relationship between CSR and ROE, shows no significance. Therefore, this test did not yield results supporting hypothesis 3, but is consistent given Model 2. Moving on to Table 3, the moderating effect of external political environment on the relationship between CSR and Tobins Q is tested. In Model 6, we can see that there is no significance of a moderating
effect of external political environment the relationship between CSR and Tobins Q. What’s more, the coefficient for external political environment is .324 and is significant at the p < .01 level; while, the model f (F = 11.251) is significant at the p < .001 level. Yet still, these results do not support a case for moderation.
Discussion
The goal of this research was to empirically investigate how a firm’s external political environment could moderate the relationship between corporate social responsibility and financial performance. Further, I investigated the link between the external political environment of a firm and its CSR score. The results of the study were mixed, with some but not all of the hypotheses being supported. In this section I will cover the major findings of the research and possible implications. Then the theoretical contributions made by this study will be discussed. Finally, I address limitations of the study as well as suggestions for future research.
Major findings
Over the years there has been great debate on the business case for CSR, in particular if social responsibility generates improved financial performance (Barnett & Salomon, 2006; Barnett, 2007; Cochran & Wood, 1984; Davis, 1973; Graves & Waddock, 1994; Hillman & Keim, 2001; Margolis et al., 2007; McGuire et al., 1988; Orlitzky et al., 2003; Ruf et al., 2001; Tang et al., 2012; Waddock & Graves, 1997; Wang et al., 2016; Wood & Jones, 1995). Many studies have investigated this relationship and have produced mixed results (Margolis et al., 2007; Orlitzky et al., 2003). Summarizing the findings, Margolis performed a meta analysis of 192 studies and found that there is a small but significant relationship between social responsibility and financial performance. While this analysis utilized studies that used multiple measures for financial performance, the impact of CSR was not differentiated with respect to financial measures. In this study I compared the effect of CSR on both accounting performance (ROE) and market performance (Tobins Q). The research found consistent results with studies which state that CSR leads to better financial performance when using accounting based
financial metrics vs market based financial metrics (Margolis et al., 2007; Orlitzky et al., 2003). Furthermore, the research was also consistent with studies that found no relationship between CSR and market based financial performance (McGuire et al., 1988). More explicitly, the results demonstrate that firms which score higher on CSR by 1 will have an increase of return on equity by .165, but will not benefit from improved market performance. Thus confirming that financial performance cannot be generalized when trying to make a business case for CSR, as the underlying causal mechanisms influencing each are unknown and evidently inconsistent. Further, managers should be aware of the financial performance gains they can expect so that they are aligned with their specific objective.
There has also been great speculation on how individual’s political ideologies impact business (Hillman & Hitt, 1999; Hong & Kostovetsky, 2012; Rubin, 2008). Scholars tend to agree that Democratic views are more in line with the principles of corporate social responsibility than are Republican views who favor decreased regulation. This study aimed to investigate if this notion holds true, and can be empirically shown to have an effect on a firm’s level of social responsibility. What was found was consistent with this belief, and demonstrated that the more Democrat-leaning a firm’s external political environment, is the higher they will score on CSR. More precisely the political views of a firm’s stakeholders, as represented by the external political environment measurement and presidential election results, will correspond to the firms’ CSR score. Suggesting that the personal political views of stakeholders can manifest into a more socially responsible firm.
Further findings with regard to external political environment found that there is a link with financial performance in the form of Tobins Q. Interestingly, external political environment was the most correlated variable with either of the financial performance metrics. The correlation suggests that the more Democrat-leaning the state is in, which the company is headquartered, the higher their Tobins Q.
Finally, while external political environment is related to both CSR and financial performance, this study found that it does not moderate the CSR – financial performance relationship. Many explanations are available for why there is a lack of a moderating effect. For one, with the issue of reverse causality not resolved between CSR and CFP, then determining moderation becomes more difficult. The lack of moderation however, does not necessarily mean there is no relationship between CSR, financial performance and political environment. The issue of causality arises again as there are many different models of moderation and mediation, as well as, potential influential variables which have not been identified. Overall, the findings from this study provide useful insights as to how external political environment interacts with CSR and financial performance.
Contributions of the study
This research set out to investigate the relationship between social responsibility and financial performance, consequently, empirically testing the validity of said relationship as well as the moderating effect of external political environment. In doing so, this study advances the field of CSR research by again validating the link between social responsibility and financial performance, while attempting to uncover the causal mechanisms. Further, it contributes specificity to the field, by showing that the relationship is stronger for accounting based financial measures, and found no evidence of a relationship with market based financial metrics. The theoretical implications of which warrant added consideration from scholars who investigate this area in the future.
Additionally, this study suggested that a firm’s external political environment may have a moderating effect on the relationship between CSR and financial performance. In doing so another possible moderating mechanism was investigated, in this case yielding a non finding.
As a result, it is shown that while correlated to both social responsibility and financial performance, external political environment has no moderation effect on our sample.
Finally, this study produced insights into the effects of external political environment on firm social responsibility, as well as financial performance. The results indicate that the external political environment will correspond to their relative level of CSR. Thus, demonstrating that organizations which are located in a state that leans Democrat will score higher on CSR, while an organization which is located in a state that leans Republican will score relatively lower on CSR. Further, it is shown that presidential election results are a strong indicator of the political environment, as it encompasses stakeholder political ideology. This provides empirical evidence that there is a business implication for the notion that Democrats favor CSR more than Republicans. Furthermore, while investigating external political environment, significant correlation was found between political environment and market financial performance. Hence, the more Democratic a state is, the better market performance firms in that state have. As this effect is largely unexplained, the causal mechanisms for such relationship remain unknown, and thus could be a potential departure point for future research. Further, this demonstrates a correlation between market performance and a geographically determined component. What’s more, is also a strikingly similar issue to finance literature, which found an unexplained relationship between location and market performance (Coval & Moskowitz, 1999; French & Poterba, 1991; Gao et al., 2011; Huberman, 2001; Pirinsky & Wang, 2006; Tesar & Werner, 1995). Hence, the results of this study could be of value to researchers who are more explicitly investigating ‘location bias’ of investors.
Limitations and future research
While this research was successful, there are several factors which were limiting and could be improved upon for future research. This study made use of the largely popular KLD
database which measures firms on a multitude of strength and concern dimensions. However, this study had limited data from KLD, and as a result had disproportionate number of concern measures as compared to strengths. This could have caused bias within our sample in many ways, by over punishing firms that have concerns, while also not rewarding firms who have strengths. Future studies should strive to have a complete dataset from KLD in order to most accurately asses CSR performance. Also, this study focused on exclusively large U.S. based firms, in the end only sampling 319 firms. Future researchers could expand the scope of firms sampled, for example, by utilizing larger indexes such as the Russell 3000 rather than the Fortune 500, as was done in this study. It was pointed out that large firms are subject to a reporting bias, and therefore, by including smaller firms may impact results (Darnall et al., 2009; Lepoutre & Heene, 2006). Furthermore, smaller firms are more likely to be impacted by location bias, and thus, external political environment, since stakeholders are more likely to be located nearby (Coval & Moskowitz, 1999)
Further, this study did not make use of time series data. It would be interesting to examine if there are time effects that influence the results. It is possible that market financial performance lags CSR, and as a result did not show significance in this study. Moreover, with the obscurity regarding the relationship between the variables, time series data could provide deeper insights. Along the same lines, this study did not test for reverse causality. It is well documented that the issue of reverse causality has plagued the CSR-CFP field (Allouche & Laroche, 2005; Margolis et al., 2007; Orlitzky et al., 2003). This study makes no exception, as this issue could very well impact the results. In addition to testing for reverse causality between CSR and CFP, it also becomes an issue between CSR and EPE. If this issue is resolved, then a more concrete understanding of the causal mechanisms will be established, thus, opening the door for further research. For example, many correlations between EPE, CSR and financial
performance were found. Thus, it’s not implausible to say that there may be undiscovered moderation or mediation occurring.
Moreover, this study set out with a focus on CSR and financial performance, research that is commonly described as ‘the business case for CSR’ (Carroll & Shabana, 2010). Meanwhile, evidence was found that social responsibility is in large part a function of political ideology. Further, it was shown without explanation that firms in Democrat-leaning states performed better as measured by Tobins Q. Therefore, it would be interesting for future research to shift the discussion from making ‘a business case for CSR’ to ‘a business case for Democrats’.
Conclusion
This study shows that firms which perform better with regard to CSR perform better on accounting-based financial measures. The study highlights the distinction between accounting and market-based financial performance and demonstrates the impact of CSR on each. The results confirm that CSR does not affect all financial performance metrics equally, as there is significance between return on equity, but not for Tobins Q. Further, the study shows that there is a relationship between political preference and preference toward CSR, so that being a Democrat is associated with being in favor of CSR. Additionally, it shows that the relationship has an operational impact on companies, as the external political environment of a firm corresponds to their CSR score. Furthermore, it illustrates why external political environment is crucial for firms to consider as it relates to their stakeholders, social responsibility, and financial performance. Future research can broaden the scope of the insights generated in this study by examining the external political environment under an array of different conditions and circumstances. Ideally, this research will garner the attention of other scholars and perpetuate future studies in this direction.
Reference List
Aguinis, H., & Glavas, A. (2013). Embedded versus peripheral corporate social responsibility: Psychological foundations. Industrial and Organizational Psychology, 6(4), 314-332.
Alagaraja, M., & Shuck, B. (2015). Exploring Organizational Alignment-Employee
Engagement Linkages and Impact on Individual Performance: A Conceptual Model. Human Resource Development Review, 14(1), 17–37.
https://doi.org/10.1177/1534484314549455
Allouche, J., & Laroche, P. (2005). A Meta-analytical investigation of the relationship between corporate social and financial performance. Review of Human Resource Management, (57), 18. https://doi.org/10.5465/AMR.1982.4285580
Audia, P. G., & Greve, H. R. (2006). Less likely to fail: Low performance, firm size, and factory expansion in the shipbuilding industry. Management science, 52(1), 83-94. Barnett, M. L. (2007). Stakeholder Influence Capacity and the Variability of Financial
Returns to Corporate Social Responsibility. The Academy of Management Review, 32(3), 794–816. https://doi.org/10.2307/20159336
Barnett, M. L., & Salomon, R. M. (2006). Beyond dichotomy: the curvilinear relationship between social responsibility and financial performance. Strategic Management Journal, 27(11), 1101–1122. https://doi.org/10.1002/smj.557
Baron, D. P. (2001). Private Politics, Corporate Social Responsibility, and Integrated Strategy. Journal of Economics & Management Strategy, 10(1), 7–45.
https://doi.org/10.1111/j.1430-9134.2001.00007.x
Bebchuk, L. A., Cohen, A., & Wang, C. C. (2013). Learning and the disappearing association between governance and returns. Journal of Financial Economics, 108(2), 323-348.
