Political activities and social responsibilities: do they
affect a firm’s financial health?
Amsterdam, 18/08/2017
Name: Bart van Vliet (10423273)
MSc Business Administration – Strategy Track
ABS, Uva
Supervisor: Dr. Pushpika Vishwanathan
Statement of originality
This document is written by Student Bart van Vliet 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.
Acknowledgements:
This is my master‘s thesis, this project marks the end of my university career, but it is just as much a reflection of it. Though the process was not smooth and far from flawless, I can be
proud to present it knowing that I gave my best. Knowing that I made mistakes, but I did my
best to fix them and keep going. I could not have done it without the help of some people so I
would like to thank a couple of people for their help and support. First of all I want to thank
my professor Dr. Pushpika Vishwanathan for sparking my interest for this topic and providing
me with the tools and guidance to be able to do this research. I would also like to thank Lara,
Brian and Quincy for reviewing my work and support during the process, and offer the
necessary distraction. I want to thank my parents for their support and for helping me to keep
motivated through the process, I hope I made you proud. The final person I want to thank is
my sister, without her I could not have done this, I could always lean on her when I was lost,
and ask her for guidance to help me go through the process. She also did a tremendous job in
Contents
Abstract 1
1. Introduction: 2
2. Literature review: 7
Corporate Political Activity (CPA) 7
H1. An increase in CPA activities will result in an increase in the CFP of a firm. 14
Corporate Social Responsibility (CSR) 14
H2. An increase in CSR activities will result in an increase in the CFP of a firm 17 Corporate Political Activity (CPA) & Corporate Social Responsibility (CSR) 17 H3. CSR activities of a firm will have a positive effect on the CPA – CFP relationship 19
3. DATA and Method 19
Sample 20 Data collection 21 Independent variables 21 Dependent variable: 25 Control Variables 25 Model 26 Data Source 27 Statistical procedure: 29 Data 29 Analyses 31 4. Results: 32
Results of the regression 34
Discussion of results 38
Limitations of results 39
Possible reverse causality. 39
5. Discussion 40
6. Conclusion: 45
7. References: 47
Appendix 1: 2016 Presidential race 52
Appendix 3: Top Contributors 54
Appendix 4. SIC code Coding: 55
1 Abstract
This report adds to the growing academic interest on the synergy between corporate political
activity (CPA) and corporate social responsibility (CSR). This is done by providing empirical
data on the theory that CSR has a moderating effect on the relationship between CPA and
corporate financial performance (CFP). This is done by using a multiple regression on 456
U.S. fortune 500 companies in the period between 2012 and 2015. The firm‘s PAC
contribution and the lobbying expenses give insight on its CPA. For rating the CSR activity of
a firm this report uses the KLD score. CFP is measured using the accounting measuring tool
return on assets, return on investment, and return on investment. The results provide evidence
of the direct positive relationship between both CPA and CSR on CFP, but did not find prove
of a moderating effect of CSR on the CPA-CFP relationship. This result contributes empirical
data to the academic understanding of the relationship between CSR and CPA. It also points
out that this is a highly complex relationship which needs to be studies further. Also the report
contributes to the debate about the usefulness of CPA and CSR for firms. This report shows
that both activities are beneficial to large U.S. companies, and recommence to do more
research to find synergies between the two, and study the relationship in different areas,
2 1. Introduction:
2017 is already breaking records in the U.S. with the appointment of Donald Trump as
president of the United State of America. It was the most expensive election campaign till this
date; with over 560 million dollars raised by Hilary and over 330 million by Trump by these
candidates alone. In total a 1.5 Billion U.S. Dollars was raised for campaigning by all
candidates of the election, with an additional 618 million dollar by PAC supporting them.
(www.opensecrets.org) (appendix 1). 2017 was also the first time a business man without any
political experience won the election. It raises the question how big the gap still is between
politics and business? With this election it has become even clearer that the gap between
business and politics is smaller than most think. Having former business men performing a
political function is not at all uncommon, and it is not unusual that former politicians continue
their career in a high level function in business. Business and politics are constantly
interacting with each other and they both try to influence how the other party behaves. Policy
makers set the law and boundaries for the companies and in return companies try to
manipulate the policy maker to form these policies in their favor. From a business perspective
it is understandable that they want to influence the lawmakers to their benefit and lobbying is
therefore not uncommon. Over the last four years there is about an average amount of 3.1
billion U.S. Dollars spent on lobbying by numerous companies in the U.S.
(www.opensecrets.org) (appendix 2). Companies also invest heavily in campaigning of the
candidates, for example both Microsoft and Google invested over 800.000 Dollars each in Obama‘s re-election campaign in 2012 (www.opensecrets.org) (appendix 3). But why do firms invest such large amount of resources into candidate contributions, lobbying, Political
3
(CPA)? What is the reasoning behind these investments? What determines if, and how much a
company is willing to invest in CPA? While examining the literature on CPA; the main driver
to engage in CPA is to be of political importance (Bonardi & Keim, 2005; Hillman, Keim, &
Schuler, 2004). Companies‘ foremost goal is to use these influences to increase their Corporate Financial Performance (CFP) by manipulating laws in their favor, as stated by
Schuler, Rehbein & Cramer (2002). For example Philips, who developed energy saving light
bulbs as part of its sustainability strategy, to replace traditional incandescent light bulbs.
When sales volumes of the new energy saving light bulbs remained unsatisfied, Philips
lobbied the European legal structure to phase out the sales of traditional light bulbs, arguing
for energy saving and the reduction of carbon dioxide emissions (Wynia, 2009). When the
law became more strict they enjoyed being ahead of the curve, while making life harder for
their competition. This resulted in a competitive advantage over their rivals. CPA can thus
lead to an advantage what could lead to increased CFP. But firms have to invest large
amounts of resources into CPA and the results are not always what they hope for. Firms can
have CPA that opposes each other. For example, in 2008 the U.S. chamber of Commerce
spent more than 60 million dollars against climate change legislation, and Southern
companies, a company dealing with high polluting power generators spent an estimated 14
million dollar to fight regulation (Delmas et al. 2016). At the same time, green companies
lobby for more regulations and more strict environmental laws. More strict laws do not only
improve the environment, but they can also be in favor of companies who are already ahead
of the curve in these areas and can thus benefit from their more environmental friendly
technology (Fremeth & Richter, 2011; Reinhardt, 1999; Vogel, 1995). Pacific Gas and
4
battle climate change at the federal level in 2008 (Delmas et al. 2016). The lobbying for both
more and less strict laws results in large amounts of resources put into CPA that cancel out
each other and in the end results in a lot of lost resources.
So does CPA really improve CFP? The empirical evidence of the relationship between
CPA and CFP is indeed still indecisive till today. Some studies found a positive relationship
between CPA and CFP (Cooper et al., 2010; Claessens et al., 2008). Other studies state that
investing in CPA is a zero sum game; benefits of CPA are cancelled out by its costs resulting
in a minor total profit or loss. (Godwin & Seldom, 2002; Hadini & Schuler, 2013; Lee &
Baik, 2010), or CPA even has a negative effect for CFP (Igan, Mishra, & Tressel, 2009;
Aggarwal, Meschke, & Wang, 2012; Coates, 2010) the investment does not pay out and
results in a net loss for the company. The differences in the academic world result in an
unclear vision of what CPA means for a firm‘s financial performance. Because of this, a large amount of resources is possible being wasted because of the ineffective use of CPA.
Therefore a deeper understanding of what enables the CPA – CFP relationship and what is
influencing this relationship is needed to help to get a better understanding of what CPA can
do for a firm.
In recent years a new trend has developed in the CPA sphere. Corporate Social
Responsibility (CSR) might be used to influence the effectiveness of CPA. Both CPA and
CSR can be used by firms as nonmarket strategies. A nonmarket strategy refers to a firm‘s
concerted pattern of actions to improve its performance by managing the institutional or
societal context of economic competition (Mellahi et al., 2016). Both CPA and CSR are
aimed at managing a corporation‘s spheres of influence, and little is known about the relationship between them. Both are aimed at the long term profitability of the firm. Den
5
Hond et al. (2004) noted that the CSR activity of firms are sometimes arguable political.
Firms operate politically in the sense that they interact with governmental decision makers
while aligning their CSR activities in the same way. As the previous example of Phillips
shows, aligning the CSR activity and CPA can result in complementary effect between the
two. Firms can invest their CSR activities in areas that they also influence via CPA, creating a
playing field they can have an advantage on above their competitors. Aside from
technological advantages den Hond et al (2004) mentions several ways in which CSR might
be used as a tool by companies to increase the effectiveness of CPA. CSR can be used to build
up a reputation (Brammer and Pavelin, 2006; Pfau et al., 2008), which could mean their CPA
efforts are taken more seriously. CSR might also complement the CPA and vice versa by
acquiring access to resources which can be helpful in the other activity (Rehbein & Schuler,
2013). Though there are conceptual and theoretical papers, the academic world has yet to
have clear empirical data on this relatively new theory. To address this gap, this report will
study the relationship between CPA and CFP and see if CSR does indeed moderate the effect
of this relationship with empirical data. Arguing that firms that are able to combine their CSR
and CPA effort to increase the effectiveness of their CPA effort to create a better CFP. To do
so the following research question will need to be answered:
Does CSR activity of a firm moderate the relationship between CPA and CFP?
The article will study the underlining effect CSR can have on the relationship between
CPA and CFP. This is done via a database research of the US Fortune 500 firms from 2012
till 2016. The financial results of the firms, funding in CPA, and CSR activity will be studied
6
First the relationship between CPA and CFP will be examined, followed by the
addition of CSR to the model to test for a moderating effect. For the CPA data the
organization Opensecret.org will be used. OpenSecrets.org is the US's premier website
tracking the influence of money on U.S. politics, and how that money affects policy and
citizens' lives. The lobbying expense, PAC contribution CEO contribution will be used to
measure the CPA activity of the firm. The CFP data will be retrieved from the Compustat
Database, to compare the financial data of the company the accounting financial performance
measurements ROE, ROA and ROI will be used to measure the financial performance. And
for the CSR data the RiskMetrics Group‘s KLD STATS will be used to compare the firms‘ CSR activity with each other. The KLD score will act as a scale to compare how much effort
is put into CSR by each firm.
This report is structured into six parts. Chapter 2 will provide an overview of the
existing literature on the topic of CPA, the different channels of CPA are discussed, the
intended results, the relationship towards CFP, CSR, and the existing literature linking CSR
with CPA. The chapter is concluded with an overview of the hypotheses that will be tested.
Chapter 3 describes the research methodology, the research strategy, and the data collection is
explained. Chapter 4 will include the data analyses and results. Both are described and
explained followed by Chapter 5, here the results and the limitations will be discussed.
Chapter 6 will include the conclusion of this report. The final chapter, Chapter 7 will conclude
7
2. Literature review:
In this chapter the theoretical background of Corporate Political Activity (CPA) and
Corporate Social Responsibility (CSR) will be discussed. First CPA is defined and the
strategic and theoretical thought process behind CPA is laid out. This is followed by the
empirical findings on the relationship between CPA and Corporate Financial Performance
(CFP). Next CSR is introduced; CSR is defined and the link between CSR and CFP is
introduced. Finally, it is explained how CSR would affect the relationship between CPA and
CPF. Based on this background the hypotheses are constructed.
Corporate Political Activity (CPA)
Corporate Political Activity (CPA) is defined as the corporate attempts to shape government
policy in ways favorable to the firm (Lawton et al., 2012). A firm does this by doing activities
like contributing to political establishments and figures, enlisting lobbyists, and organizing
events. The firm tries to gain influence in the govern policy in ways that are beneficial to the
firm. CPA is not a recent phenomenon, CPA can be traced back as far as the elections of
1904, were the republicans were handed large sums of money by large insurance companies
in an alleged attempt to gain legislation that would limit the ability of policyholders to sue
managers for breach of fiduciary duty (Campaign legal center, 2010 in Hadini & Schuler,
2013).
CPA has developed through the years, and firms can engage in CPA in several ways
nowadays. They can contribute financial resources to a political campaign in an attempt to get
influence, or start lobbying for certain laws and regulations that benefit the company. A firm
could also contribute to Political Action Committees (PACs), these are political committees
8
(opensecret.org). Money spent on these activities can be categorized as ‗hard money‘. These
are political donations that are regulated by law through federal election commissions. Firms can also use ‗soft money‘, this is money donated to political parties in ways that are
unregulated and are not officially registered.This study will use registered donations, and will
use the PAC contributions, and lobbying expenditure as a measurement of CPA. This data is
available on Opensecret.org. Via these activities, the firm tries to gain influence on the govern
policy, but what are the desired results of this influence? To better understand CPA the
reasoning behind CPA needs to be examined.
CPA is part of a firm nonmarket strategy. Non market strategy is considered to be the
firm‘s effort to control or influence the institutional or societal context of economic
competition (Boddewyn, 2003). By doing this, firms want to use their influence to obtain and
maintain a competitive advantage (North, 1990 in Lux, 2001). The general believe why firms
engage in nonmarket strategies including CPA, is to increase the firm performance (Mitchell
et al., 1997). Contributing resources to CPA is seen as an investment. Investing into CPA
generates political benefits that are considered to outweigh the costs (Baron, 1995; Mitchell et
al., 1997). Firms commit to CPA if it is perceived to generate better returns than investments
in other activities (Lux et al., 2001). To confirm if this is indeed true, the effect of CPA needs
to be observed.
What is the effect of CPA on corporate performance?
The foremost goal of CPA is to increase the Corporate Performance (Schuler, Rehbein
& Cramer 2002). The assumption is that engaging in CPA allows the firm to use their
influence to gain a competitive advantage by altering laws and rules to change the market
9
be done by a variety of ways looking at case studies done in previous research: One way is to
defend the domestic market against foreign competition. In a study on the steel industry by
Schuller (1996) it was observed that firms committed resources into CPA to influence new
import tariffs. Especially firms that had foreign competition engaged in CPA to influence the
import tax to protect the domestic market. This way the steel from foreign competitors had a
more difficult time in competing with the prizes of the US based firms. An interesting factor
seems to be that market share was found to be one of the more important for the levels of
political involvement. Schuller (1996) reasons that the CPA benefits the entire domestic
market, and thus the larger companies benefit more from CPA and consequently are more
likely to invest in CPA. If this is true in the steel market, it could be that this is also true in
other markets. If so, any effect of CPA on CFP would be more severe on the biggest players
in the market. The effect therefor should be most visible for the biggest companies in the US
who are dominant players in their market. A second way that CPA can be used is observed by
De Figueiredo & Silverma (2006). They did a study on the estimated returns to lobbying by
universities. Universities tried to obtain benefits by lobbying for earmarks. CPA was believed
to lead to receiving more earmarks from the government. They concluded that a University
that was represented the House Appropriation Committee or the Senate Appropriations
Committee was more successful in lobbying for earmarks then universities who were not
represented in these committees. This means that universities that are being represented in the
government are more likely to receive earmarks. CPA can thus result in a financial benefit for
universities. Another interesting conclusion of De Figueiredo & Silverman (2006) is that
interest groups that are closely aligned with the policymaker had to spend less resources to
10
the government reacts different on the request of a university than of a firm. Another example
of results of CPA is studied by Bonardi et al. (2006). Bonardi et al. (2006) studied non market
strategies in the U.S. electric utility sector. By using non-market strategies, firms try to
improve their financial performance by influencing the rate increase they received. The profit
levels of the utilities are regulated under a financial rate of return regime by state agencies. The rate of return was determined by a ―rate review‖. Utility firms are allowed to file for a rate review whenever they want to. If a firm filed for such a review, the firm tried to influence
state regulatory agent who determined the rate by the use of non-market strategies. By doing
this, the firm influenced the rate the customer has to pay. The firm who could justify a higher
rate of return would also increase their financial performance.
In short, by shaping the public policy environment to favor its business activities or to
mitigate the effects of harmful policies (Epstein, 1969 in Lux et al. 2001); and reinforce the
social relationship between business executives and public officials that maintain favorable
business conditions (Clawson, et al., 1998). This way the firm can increase its Corporate
Financial Performance (CFP). According to these studies, it is clear to say that CPA is a
diverse tool and can be used to achieve a large range of results. The desired results are diverse
but mostly serve the purpose of obtaining a financial benefit. Either by eliminating
competition, receiving subsidy, increase allowed rate of returns, in the end they are used as an
investment to increase the CFP. However CPA is a costly investment. The benefits of CPA
can pay dividends for the company, but the desired results are not always reached. The
empirical data about CPA and CFP do not clearly state that the CPA - CFP relationship is
positive. It has not convincingly supported the view that CPA has the desired financial results
11 CPA and CFP; a positive relationship
Various studies suggest that a positive relation exists. Cooper et al. (2010) studied the
firm level contribution of 1261 firms in the US during the political campaigns from 1979 and
2004, and found a positive relationship between CPA activity and CFP. The effect between
stock returns and corporate political contributions was strongest for firms that support a
greater number of candidates that hold office in the same state that the firms is based. Also the
effect was found to be stronger if firms had contributed to House candidates, or contributed to
democrats. The study is done over a larger period of time, but the most recent results are from
2004, more than decade ago. Since then the economy and the political landscape is changed.
Companies are increasingly global and the recent economic crisis may have altered the way
politics and business react towards each other. Beside the increase in stock returns for
companies who invest in CPA, Claessens et al. (2008) also found a positive relationship
between CPA and operating performance when studying the relationship between CPA and
CFP in Brazil. When studying Brazilian firms between the 1998 and 2002 election, they
found that firms that provided contribution to federal deputies experienced higher stock
returns than the firms who did not. They suggest that these contributions help the firm to
shape policy on a firm specific basic. This could imply that the relationship is not limited to
the U.S. alone. Kim (2008) studied the determinants of lobbying expenditures and campaign
contribution, and found that lobbying has a positive effect on the firm‘s equity returns relative to the market average, and to a lesser degree their peers in the same industry. This is
interesting because this implies that the effectiveness of lobbying allows for rivals in that
industry to also enjoy the benefits of CPA. It could also mean that the peers in the same
12
Schuller (1996) and his research of the steel market. Because of the CPA effort of one of the
players in that industry, the increase in import tax was beneficial to all the players in the
domestic market. The Steel industry might have an increased CFP, but the free riders benefit
who did not invested in CPA benefit without investing. The relationship between CPA and
CFP therefore might differ between the industry and an individual firm. Though these studies
show a positive relationship between CPA and CFP, some empirical studies have also found
contrary results.
CPA and CFP; a neutral and negative relation?
Investing in CPA is not without costs, and some of the financial benefits are
eliminated by the cost of CPA. Or their CPA effort is nullified by their rivals‘ efforts
(Stratmann, 2002). Therefore some studies concluded that the relationship between CPA and
CFP is a zero sum game. When Stratmann (2002) studied if PAC contributions had influence
on the legislators in the financial service legislation, he also showed that competition between
political activities implies that their contributions are partially offset by each other. This
makes their effort less effective. Instead of working cooperatively, they cancel out each
other‘s CPA effort, resulting in an arms race with other rivals who are politically active. This increase in costs of the political activities can result in an unclear or insignificant outcome of
these efforts. These zero sum games have also been reported in various studies, including
airline legislation (Godwin & Seldom, 2002 in Hadini & Schuler, 2013), and antidumping
(Lee and Baik, 2010).
Other studies though found a negative relationship. Aggarwal, Meschke, and Wang,
(2012) did a study of over 1800 firms who did, or did not, contribute money to CPA between
13
is associated with worse returns than not donating at all when looking at the stock prices of
the firms. They claim that resources contributed to political party are done by the managers to
benefit themselves. Donations would be part of a bigger agency problem in the firm, and
would be bad for the company and shareholders. CEO‘s would use the CPA efforts of the
company to benefit themselves, building relationships within the government to benefit their
career. The negative results between CPA and CFP was similar to the findings of Coates
(2010),who found a strongly negative relationship between lobbying and firm value when
studying the S&P500 in the period between 1998 and 2004. They concluded that political
activity is not in the interest of the shareholder when studying the observable political activity
like PAC contribution and lobbying expense. Coates (2010) added to this by suggesting that
unobserved political activity like soft money donations, would harm the shareholders even
more. Following this logic, CPA would be part of an agency problem in the firm. Managers
use CPA to benefit themselves, and would not be to increase the financial performance of the
company. Though this theory was proven by their study, it is contradicted by the study of
Marthur et al. (2013). They studied the lobbying behavior of sample firms for a period
between 1998 and 2003. They state that their evidence suggest that corporate lobbying is not
agency driven and may, in fact, create value for the shareholders als found in previous
research discussed before.
Summarized, due to these mixed results there can be concluded that the relationship
between CPA and CFP is complex still incomplete and not fully understood. The relationship
has been proven positive, negative, and not existing.. To better understand this relationship,
first the CPA-CFP needs to be examined if such a relationship exists.Therefore based on the
14
testing the following hypotheses:
H1. An increase in CPA activities will result in an increase in the CFP of a firm.
Corporate Social Responsibility (CSR)
When firms engage in corporate political activity (CPA), firms are aware that their
reputation and trustworthiness are important external actors in the political area. Having a
good public reputation and be known to be good to the public could possibly complement
your political efforts (den Hond et al., 2014). Managing these external actors is important to
the firm‘s non-market strategies. To gain legitimacy, firms can participate in corporate social responsibility (CSR). By engaging in CSR the firm can improve their image and boost their
reputation. This synergy is discussed by den Hond et al. (2014), they argue that an alignment
of the CSR and CPA effort could boost the firm‘s reputation. Or CSR can act as a ‗lighting rot‘ to distract the attention from CPA to CSR, which has a better reputation (Vogel, 1996 in Hond et al. 2014). In contradiction to this theory, Cooper et al. (2010) shortly discussed the
possibility of charity spending as a sign of an agency problem. They argue that managers
simply engage in wasteful spending for their own good. Managers could be choosing charity
that are important to them, and spent firm resources on charities that are not in the interest of
the company. Charities are part of the CSR umbrella, and could be a sign of management
spending money in non-market strategies like CPA or CSR to benefit themselves. To better
understand the relationship between CSR and CPA, A clear definition and understanding of
CSR is needed.
As Frynas & Stephens (2015, p 484) said: ―A key challenge for building CSR theory
15
defined In this paper CSR is viewed as an umbrella term that contains the policies, processes,
and practices that a firm puts in place to attend to the social demands. It can be defined as all
activities that go beyond the requirements of the law (Carroll, 1979). Or as McWilliams and
Siegel (2001) specify: CSR is an investment in actions that advance a social cause, for
example by adding social features or characteristics to products, or by modifying production
processes to signify that the firm is seeking to advance a social objective. Because of the
nature of CSR it is has a very broad definition, and can be interpreted in various ways, it also
has a number of alternative terms that are used in business and sometimes the academic world
with (almost) similar meaning. Terms like corporate social performance (CSP), corporate
responsibility (CR), Corporate Citizenship (CC) Corporate Sustainability (CS). This paper
will include try to stay away from using these terms interchangeable, by using the definition
as described above as boundaries for the topic, and use the term CSR unless there is clearly a
difference.
The topic of CSR is already studied at large, and findings in this area have a wide
range (Agiomos & Glavas, 2012). The relationship that is relevant for most businesses is the
relation CSR can have on CFP (Agiomos & Glavas, 2012, p 943). If engaging in CSR can
directly or indirectly lead to an increase in CFP, it could become an attractive tool to create a
win-win situation for business and society. Numerous studies have studied the impact of CSR
for a firm‘s performance, and studies have found a variety of ways of how CSR could help a
firm increase its firm performance. CSR has found to have a positive relationship for
employees by increasing job satisfaction (Brammer et al.; 2007), or increase the employee
organizational commitment and organizational performance (Turker, 2009). Employees tend
16
resources to CSR. This perception could lead to firms being able to hold on to valuable
employees more easily, and find it easier to keep their employees happy. CSR also has an
effect on customers. CSR has also been found to increase the customer‘s valuation of the
company‘s brand. Sen & Bhattacharya (2001) found that CSR had a positive effect on customer‘s company evaluation. Customers conceived brands as more valuable when they knew the firm was committed to CSR activities, but Sen & Battacharya (2001) did not find an
increase in purchase intention of the customer. In a more recent study this was contradicted by
the findings of Huang, Yen, Lui and Huang (2014). This study also found a positive link
between customer‘s brand valuation and CSR, but also found an increase in purchase
intention. This could results in an increase of sales and thus firm financial performances. This
was confirmed in the study of Lou and Bhattacharya (2006) who found that customer
satisfaction was an important mediator in the link between CSR and CFP. Though studies
have found positive relationships between CSR and CFP, there is also a group of studies that
found a neutral or negative result. In the Meta analysis of Peloza (2009) that studied 167
studies related to CSR-CFP, it was found that 27% had mixed results, 14% a negative result,
and 59% had positive results. Other Meta analyses like Margolis & Walsh (2003) concluding
123 studies also found mostly positive relationship, which was confirmed in a more recent
study in Margolis, Elfenbrein & Walsh (2007).
Because CSR is shown to have a possible positive effect on CFP, it is important to
include this in the model when examining the possible synergy between CSR and CPA. To
confirm if such the relationship between CSR and CFP is present the following hypotheses
17 H2. An increase in CSR activities will result in an increase in the CFP of a firm
Corporate Political Activity (CPA) & Corporate Social Responsibility (CSR)
Looking at the relationship between CPA and CSR, the two were treated as separate
till only recent. A growing amount of literature is devoted to explore the impact of aligning
CPA and CSR on the performance of the firm (Liedong et al., 2015). They studied the
mediating effect of trust in policy influence, and argue that CSR and CPA should be aligned
for the successful influence of salient government policy. CSR is argued to be able to protect
the firm‘s reputation (Minor & Morgan, 2011), argued by Hond et al. (2014) it increases firm visibility, and it can grants legitimacy to the firm (De Roeck & Delobbe, 2012). Liedong et al.
(2015) therefor argues that through the benevolence effect of CSR, the politicians will see
firms as less selfish and hence tolerate influence. It also reduces the resistance of the public
and customers. This means reduced resistance towards the corporate political influence of the
firm (Fooks et al., 2013). And it also allows politicians to be acceptant to the firm's influence
(Schuler, 2008). Hond et al. (2014) therefore suggest that CPA and CSR can complement
each other. They argue that CSR efforts can help a firm‘s CPA activity via three ways. CSR
can contribute to getting access to valuable resources. The increased visibility of the firms due
to the CSR activity can establish direct contact in politics that can be used to gain access to
political and legislative decision making. A good reputation can help lower the barriers to
political entry(Wang & Qian, 2011 in Hond 2014), which can be established by engaging in
CSR activities.
Hond et al. 2014 also talks about the efficacy factor that CSR could have on CPA.
They argue that CSR firms have a stronger and more diverse set of relationships with
18
relationships they can obtain information from these partners to establish a superior issue
position. They can use the information to gain a more increase debt knowledge and more clear
picture of what the community thinks, compared to companies who only rely on their own
personal. This information is very important for politicians, and can be used to trade to boost
politician influence. It is also possible to use this relationship to establish a coalition with
similair minded parties. By working together, and pressure politicians from multiple angles, it
allow for an increase in influence the company and his coalition can exercise on the political
issue. Crafting coalitions with non-business partners through CSR increases the number of
players that think alike, helps to channel their efforts and thus increases their influence in the
political sphere.
Another possible benefit is a cost reduction. The cost of CPA activities may also go
down if a firm is active in CSR. Hond et al (2014) argues that a firm with a good reputation is
less required to do financial donations, but politicians are more likely to consider their opinion
than a firm what has a bad reputation. A study by (Wang & Qian, 2011 in Hond 2014),
studied Chinese firms have found that firms that were a positive philanthropy performance
gained political resources because politicians want to be associated with firms that are well
received. Hond et al. 2014 also suggest a second way that CSR reduces the cost of CPA.
Because of the increased reputation of the firm, the amount of demonstrating compliance may
be reduced. The goodwill and reputation the firm has built due to their CSR activities may
allow firms to perform in a less regulated market and allow the firm to self-regulate without
interference of the government. Reed (2009) studied the relationship of relational
characteristics to explain the incremental variance in predicting the frequency of monitoring
19
(2009) then reasons that banks who are less regulated can switch their attentions elsewhere,
improving their performance. Besides the government, NGO may also be less critical of firms
who have a good reputation then firm.
Following the reasoning of the above literature, CSR can have an influence on the
successfulness of CPA activities via numerous ways. This results in a more effective use of
resources for CPA and thus to an increased corporate financial performance. But this
relationship has been mostly conceptual. There is a significant lack of empirical data available
on this relationship. Therefor this theory needs empirical data to validate its credibility. If the
relationship between CPA, CSR and CFP finds empirical support, it would benefit the
academic understanding of the CPA – CFP relationship and add to the understanding of CSR
as a tool in business. To test if just a relationship exist this report will test the following
hypotheses:
H3. CSR activities of a firm will have a positive effect on the CPA – CFP relationship 3. DATA and Method
This chapter presents the data and methodology of this report. Paragraph one will
describe and explain the sample used in this report. In the second paragraph the variables are
explained. The third paragraph will explain how the variables are measured, and discuss why
these variables are chosen. The fourth paragraph will elaborate on the source of the data, and
how the data is collected. The chapter will be concluded with paragraph five; here the model
20 Sample
The sample population is based on companies in the US listed in the Fortune 500 over
the period of 2012 to 2015. The sample exists of publicly listed firms that were listed on the
Fortune 500 of 2012 or 2014. This resulted in starting pool of 554 companies. Companies
were filtered to exclude companies without the required information over the period between
2012 and 2015.This resulted in a final data pool of 456 companies. This report limits itself to
US companies only because of the availability of the data.
There are limitations to this data set. First the use of only U.S. based companies; the
results will only be valuable to test the relationship in the U.S. The U.S. political system and
their view on CPA and CSR can be different than the views elsewhere in the world. The use
of Fortune 500 companies ensures the availability of data, but limits this report to only focus
on large influential organizations. These companies are not limited to the US market only, and
can also have interest in different parts of the world. This can influence the CFP by investing
in CPA activity in the rest of the world, which are now not taken into account. CPA can also
have different outcomes in the world because of the differences in governmental strength and
political system. (Windsor, D. 2007). During the period of the data, the political landscape has
changed in the US. This database does not include switches in power in the political spheres that may have an impact on the outcome of CPA. The CPA data is only based on financial resources spend via regulated channels. With the possible contributions of soft money and non-currency contribution, therefor it is possible the true value of contributions is much higher.
21 Data collection
Variables
This section will explain the variables that are used to answer the research question.
To test the hypotheses, the following independent variables will be used: Corporate political
activity (CPA), corporate social responsibility (CSR) and the following dependable corporate
financial performance (CFP).
Independent variables CPA
To measure the amount of effort a company puts into CPA this report follows Lux et
al. (2004) by measuring the hard money contribution. A firm can contribute financial
resources to their CPA by donating soft money, or hard money. Soft money is an unregulated
donation and has no limit on the amount someone can contribute. This makes soft money hard
to trace. Large sums of money are contributed without disclosure and are therefore unreliable
as a variable for CPA. The other method is hard money contributions. To measure this, this
report uses the amount of political action committee (PAC) contributions and lobbying
expense as variables to measure the financial resources to CPA. PAC contributions are
defined by Opensecret.org as political committees that control most of the corporate money.
These committees pool contributions from individuals and distribute them to candidates,
political parties and other PAC‘s. These PAC‘s are controlled by companies, trade
association, unions, issue groups and politicians. Contributions towards these PAC‘s are a
way for firms to boost the PAC and use these PACs to push their political agenda. All
22
(FEC). This information is made available through the center for responsive politics via the
website Opensecret.org. Because of the full disclosure of the information and the
completeness of the data, it is a reliable source to see financial contributions by firms to CPA.
The second ways of hard money contributions are lobbying expenses. Lobbying expense are
the expenses made on professional advocates that lobby members of congress and
government officials on the issues their clients care about. A higher amount of expenses
would indicate a higher amount of CPA. This data is also made available by on the
Opensecret.org website.
CSR
CSR will be measured using the Kinder, Lydenberg & Domini (KLD) score of the
company. The KLD score has been widely used to examine CSR (Harjoto et al., 2011;
Margolis and Walsh) and is one of the most inclusive ways to measure the corporate social
and environmental rating in the US. Therefor the KLD score is chosen as the measurement for
CSR. The KLD score consist of 13 social performance criteria ranging from community,
corporate governance, diversity, employee relations, environment, human rights, product, and
participation in controversial business activities. The KLD score consist of positive and
negative performance indicators concerning environmental, social and governance
performance. The KLD score of the period between 2012 and 2015 will be included into the
database. A firm with a higher average KLD score in this period is regarded to be more
invested into CSR.
CPA
The total amount of financial contributions by a firm is calculated by summing the
23
contributions made by the firm. This is based on the method used in Hadani & Schuler
(2013). The dataset concerns the period between 2012 and 2015. This is to ensure to eliminate
the possibility of one time contribution and decrease the influence of outliers in the data. This
data is available on Opensecrets.org. To search for the lobbying expense, the Opensecret.org‘s
search engine was used. To find the relevant firms, the name of the firm, or sections of the
name was used in an attempt to find the firm. The firms that came up in the search results
were all verified to make sure the firm is the exact firm that was searched for. If the firm was
indeed the correct one, the lobbying expenditures for those relevant years were copied to the
dataset for the relevant year. If there was no lobbying reported for a particular year, the
assumption was made that there was no lobbying by that firm in that period, and a zero was
inserted into the dataset. The PAC contributions were obtained in a similar manner. The
search option in the PAC Contribution section of Opensecret.org was used to search for the
firm. If the firm was found in the search results, the PAC contribution was inserted into the
database for the election cycle of 2012 and 2014. If the firm did not show up in the search
results, or there was no PAC contribution reported for the relevant years, a zero was inserted
into the relevant cell. The final CPA variable was then constructed by adding the PAC
contribution and Lobbying expense together. This variable was labeled as CPA Contribution.
CSR
To work with the KLD score this report follows Graves & Waddock (1994). This is
received as the standard protocol for measuring Corporate Social Performance (CSP) by
Mattingly (2017), Because of the high similarity between CSP and CSR, this protocol is also
followed to measure CSR. The KLD score uses a binaurally code to measure a firm. The firm
24
are categorized in six sections. By using the weighting scheme by Capelle-Blancard & Petit
(2014) (Table 1), it is possible to add weighted positive indicators and distract the negative
indicators to generate a KLD score. To weight the scores, the firms are divided into sectors
given by Capelle-Blancard & Petit (2014) to align the KLD score and to make sure the effect
of CSR is measured accurately. The weighed score allows the companies to be compared with
each other and give a representative KLD score.
Table 1 Weighted schemes by sector
Enviroment Social Governance
ENV COM HUM EMP+DiV PRO CGOV
(%) (%) (%) (%) (%) (%) Panal A: Concerns Banks 7 17 10 32 10 24 Basic Resources 34 18 10 15 6 16 Chemicals 35 13 4 10 27 11 Consumer G&S 12 12 5 32 26 13 Industrial Goods 18 11 4 46 9 13 Technology 14 8 11 38 9 21 All Sectors 20 13 8 29 14 16 Panal B Strengts Banks 24 36 4 15 13 9 Basic Resources 31 32 3 11 9 13 Chemicals 37 23 3 8 16 14 Consumer G&S 32 25 3 14 18 8 Industrial Goods 53 15 1 8 11 12 Technology 41 20 3 8 15 12 All Sectors 36 24 3 11 15 11
Source. Capelle-Blancard & Petit (2014)
Note. Data: Covalence-EthicalQuote. Sample period: 2002-2010. NGO = nongovernmental organization; ENV = environment; COM = community; HUM = human rights; EMP + DIV = employee relations + diversity (women/minorities); PRO = product; CGOV = corporate governance.
25 Dependent variable:
CFP
To measure CFP, this report follows the article by Lux et al. (2011). Lux et al. (2011)
based their measurements on studies done by Bonardi et al., (2006); De Figueiredo &
Silverman, (2006); and Lux, (2008). Firm performance is measured by using accounting
based outcomes. In this article the return on equity (ROE) return on assets (ROA), and return
on investment (ROI) is used. A firm would make an investment only if that investment yields
higher returns than an alternative investment opportunity. If CPA does not give a higher
return than other investments, the firm that would have invested in other investments has a
higher return. Therefor a firm who invest in CPA should have a higher return on equity (ROE)
return on assets (ROA), and return on investment (ROI).
Control Variables
Because the size and the industry have been suggested to be a factor in other studies
(Tsoutsoura, 2004; Hillman, 2004; Udauasankar, 2008), and the industry sector has influence
on lobbying behavior as well (Schuler et al, 2002; Mathur et al., 2013); each of these
characteristics is used as a control variable. Also the agency problem with concern to CPA as
described in Aggarwal, Meschke, and Wang, (2012) is taken into account as a control
variable. First to address the size of a firm; Size is an important control variable because the
size of the organization influence the motivation to commit to CSR principle (Udayasankar,
2008), and influence the amount of resources available for CPA (Hillman, 2004 Hansen and
Mitchell, 2000; Brasher and Lowery, 2006). Size will be measured based on the study by
Tsoutsoura (2004) and will be done by using a logarithm of the assets of the firm and a
26
Schuler et al. (2002) and Mathur et al. (2013) emphasize the importance of industry
economics and regulatory factors that influence a firm‘s lobbying behavior. To control for any
differences between industries, the firms are categorized into sectors. This is done following
Capelle-Blancard & Petit (2014). Firms are grouped into six industries: Banks (80 firms);
Basic Resources including Oil & Gas (50 firms); Chemicals including Health Care (38 firms);
Consumer Goods and Services including Personal & Household Goods, Food & Beverages,
and Retail (135 firms); Industrial Goods including Automobiles & Parts (129 firms); and
Technology (60 firms). Firms that did not fit into these categorize were given the all sectors
label (31 firms). To categories the firms the 4 digit SIC code was used.
To control for the agency problem, this report follows Aggarwal, Meschke, and
Wang, (2012) by taking the CEO contribution as a variable to test for possible own interest in
CPA activity. This report uses CEO contribution is a sign of own interest in CPA activity, and
control for the possible agency problem complementing CPA and CSR in the form of charity
expenses as suggested by Cooper et al. (2010). CEO contributions are seen as a sign of
personal investment from CPA and a possible source for personal gain. A higher CEO
contribution would have a negative effect on the CFP.
Model
To visualize the three hypotheses Model A will be used. H1 states that an increase in
CPA will have a positive effect on CFP and results in an increase in CFP. An increase in the
amount of lobbying expenses and political action committee (PAC) contributions will result
in a higher CFP. H2 states that an increase in CSR will result in an increase in CFP. Firms
with a higher KLD score will have a higher CFP. H3 states that a high CSR will positively
27
positively moderate the relationship between CPA and CFP.
CFP is measure in three different ways, ROA, ROE and ROI. The model will be tested
with the ROA, ROE and ROI separately. The model has 4control variables. The first is CEO
contribution; CEO contribution could have a negative effect on CFP due to the agency
problem. The model controls for the effect of firm size by adding firm size based on number
of employees and amount of assets to the model. The model is both tested for all firms at
once, and separated into industry sectors to see if the hypotheses are influenced by the
differences in industry.
In sum, the regression model is as followed:
ROA =αi + β1 CPA contribution + β2 KLD Score + β3 CPA contribution * KLD Score+ β4
CEO Contribution + β5 Number of Employees + β6 Amount of Assets + µi
ROE =αi + β1 CPA contribution + β2 KLD Score + β3 CPA contribution * KLD Score+ β4
CEO Contribution + β5 Number of Employees + β6 Amount of Assets + µi
ROI = αi + β1 CPA contribution + β2 KLD Score + β3 CPA contribution * KLD Score+ β4
CEO Contribution + β5 Number of Employees + β6 Amount of Assets + µi Data Source (-) Firm Size (+/-) MODEL A H3 (+) CEO contribution Industry sector H2 (+) Independent variable
CPA
(CPA contribution) Dependable variableCFP
(ROE ROA ROI)
Independent variable
CSR
(KLD Score)28
CPA
For the CPA data, the Opensecret.org site is used. This website is used by the Center
for Responsive Politics to publish data on CPA related activities. It is a non-partisan,
independent and non-profit premier research group that tracks money in the U.S. politics. It is
seen as the most comprehensive resource for federal campaign contribution, lobbying data
and analysis available for the U.S. The data is openly available and manually searched for in
this report. The data for the political activities and contribution were retrieved from
Opensecrets.org. The data obtained from this source includes ―hard money‖ contribution. To get a clear view of the CPA activity of the firm, a measure of 5 year of data was included. The
included data is from the election cycle of 2012 and 2014 for the PAC contribution. The
lobbying expenses were also gathered from opensecrets.org for the period of 2012 till 2015.
The CPA data was retrieved manually, and assembled into a database. The data was retrieved
in March 2017.
CSR
The CSR data is gathered by using the RiskMetrics Group‘s KLD STATS. The KLD score is chosen because of its reliability, validity and completeness in measuring the broad
definition of CSR. The criteria for the KLD score are publicly available and constantly
revisited. The ratings are based on a combination of quantitative and qualitative information,
and including some interpretation and judgment. These judgments however are in many
respects not different from the interpretation that one can underlie on a financial statement,
29
CFP
The financial data needed is taken from the Compustat Database. From this database
the ROE, ROI, ROA is used, and the Firm size is measured by taking the number of
employees, or amount of assets.
The CSR and CFP data are both retrieved from their respected database. The CSR and
CFP data was modified and merged with the CPA database to transpire as one database that
will be used for this report.
Statistical procedure: Data
CSR
To deal with the missing data from the CSR data, the data was transformed to increase
the validity and reliability of the KLD score. First, the firms who had no data available on all
variables were excluded from the database. This decreased the number of firms from 554 to
513. Next only variables that had data available for more than 80% of the firms in all years
were included in the database. This is to make sure the firms were compared more equally and
to make sure a firm did not receive a high or low score based on data available. For example,
a firm that has no data available on variables that are considered a ‗concern‘, would not have
been taken into account for the KLD score, and thus would receive no penalty on their KLD
score because of this. This leads to a higher KLD score then the firm should have acquired.
This could case firms that have limited data available to receive inadequate scores. After this
was done, it decreased the number of variables that were compared each year from 67 to 28.
30
score were categorized into six categories; Environmental (ENV); Community (COM),
Employee (EMP); Product (PRO); Corporate Government (CGOV); and Human Rights
(HUM). This was done to add the right weight to each variable to get the KLD score. After
this was done the binaural score of 0‘s and 1‘s of the strengths were added together and the concerns were retracted based on the weighted schedule of Capelle-Blancard & Petit (2014)
(Table 1) (Appendix 5). To do this, all firms were categorized into the following seven
categories: Banks (80 firms); Basic Resources including Oil & Gas (50 firms); Chemicals
including Health Care (38 firms); Consumer Goods and Services including Personal &
Household Goods, Food & Beverages, and Retail (135 firms); Industrial Goods including
Automobiles & Parts (129 firms); and Technology (60 firms). Firms that did not fit into these
categories were given the all sectors label (31 firms). The KLD score is then calculated by
adding the weighted strengths and concerns for each firm based on their industry, and the
category of the variable.
CPA
First, the database had to be checked on reliability because it was manually obtained
from the Opensecret.org website. To do this, the database was checked by pairs of 2 students to control each other‘s manually collected data. Each person checked 10 companies who the other person collected, and checked the data with the use of a inter rater reliability score. The
data had been measured to be above 99% similarity. Any differences in the found data were
discussed and resolved. Next the lobbying expenses, PAC contributions, CEO contribution
and were taken from the respectable period. For lobbying expenses the expenses of the period
2012 till 2015 was added together and divided by the number of years to get to an average
31
same is done for PAC contributions and CEO contribution for the election cycle of 2012 and
2014. Firms who did not have data available were left blank.
CFP
The variables ROA, ROE, ROI to measure the CFP, were given by The Compustat
Database. To increase the reliability of the data, the average data of multiple years was used
to measure the variable. This meant the ROE from 2012 till 2015 were averaged to one ROE.
This is done in the same way for ROI, ROA, and the control variables. The data for all the
firms on all CFP variables were complete, and needed no extra transformation. The data of all
three databases were then merged into one database. The firms that were removed in the
process for one of the variables were also excluded from the final database, resulting in a data
pool consisting of 487 firms.
Analyses
The data checked for outliers. This was done by standardizing the scores of the
variables to a Score by SPSS. The norm for outliers was set to mark all cases that had a
Z-Score > 3,29 and a Z-Z-Score <-3,29. This is done because of the use of secondary databases.
There is no way to verify if these scores outside this range are reliable. Therefor these firms
were removed to increase the reliability of the database. When the outliers were removed the
database the number of firms was reduces from 487 to 456.
Next the data was checked if it was normally distributed. The CEO contribution
variable was found to have a significantly high skewness and kurtosis, the data was therefore
transformed; this was done by taking the 10 logarithm of the variable to obtain a maximum
skewness of 1 and a kurtosis below 3; meeting the requirements of a normal distribution. A 1
32
results in a ―0‖.
Next the moderating variable was created needed to test hypotheses 3. This was done
by computing the variable of the KLD Score and CPA contribution to one variable by
standardizing them and multiplying them together to create the variable CPA*CSR. The
hypotheses were tested by doing a linear regression analysis using SPSS. Two models were
used to do this. The first model containing the KLD score and CPA Contribution with control
variable to test hypothesis 1 and 2, the second model had the moderator added to test
hypothesis 3. To control for industry sector both models were used again for with the data
split based on their individual industry sector.
4. Results:
In this chapter the results of the analysis will be discussed. The chapter will start with
an overview of the descriptive data of the database. Thereafter the results of the analyses are
displayed. These results will then be explained and discussed. Finalizing this chapter is a
discussion of the limitations of the data.
A total of 456 companies remain in the sample after companies missing either
corporate financial performance (CFP), corporate social responsibility (CSR) or corporate
political activity (CPA) data were eliminated. Table 2 gives on overview on the average
scores for each of the variables tested in the model including CSR, CPA, CFP and control
variable. This is given per industry sector and for the overall database. Looking at the CPA
contribution, the average company contributes 1,89 million U.S. dollars in CPA each year.
The chemicals sector has a relative high CPA contribution with an average of 3,22 million
U.S. Dollars invested in CPA per firm each year. Industries in the basic resources sector has
33
score of the sample was -0,02 with the banking sector scoring highest with 0,86 on the KLD
score, and basic resources scoring lowest with -0,70 on average.
Table 3. Correlations matrix for key variables
CPA CSR ROE ROA ROI Assets
CSR -,057 ROE ,050 ,070 ROA ,059 ,145*** ,276*** ROI ,090 ,123*** ,209*** ,822*** Assets ,583*** -,051 -,041 -,147** -0,099** Employees ,292*** ,097** ,022 ,196** ,156*** ,264***
***. Correlation is significant at the 0.01 level (2-tailed).
**. Correlation is significant at the 0.05 level (2-tailed). *. Correlation is significant at the 0.1 level (2-tailed).
The correlations between the key variables are summarized in table 3. ROA and ROI
both have a positive relationship with CSR (r=0,145 & r = 0,123). This echoes the findings in
previous research in the CSR-CFP relationship (Margolis & Walsh, 2003; Margolis,
Elfenbrein & Walsh, 2007). CPA is positively linked to the sized of the firm, both in number
of employees and amount of assets, implying that larger companies invest more resources into
CPA than smaller companies. Reinforcing the findings of Hillman (2004), Hansen and
1
Table 2. Descriptive statistics
Independent variable Dependable variable Control variable
N CPA Contribution KLD ROE ROI ROA
Firm Size (assets) Firm Size (Employees) CEO contribution
Industry Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean Mean Mean
(in Millions) (Log10) (Log10) (Log10)
Consumer G&S 125 1,05 2,16 0,18 1,11 0,22 0,38 0,12 0,12 0,07 0,05 3,94 1,54 3,25 Industrial Goods 126 2,00 2,81 0,74 1,11 0,17 0,47 0,10 0,11 0,05 0,05 4,11 1,51 3,51 Banks 65 2,20 2,09 0,76 0,82 0,14 0,11 0,07 0,57 0,03 0,03 4,77 1,24 4,15 Basic Resources 41 1,48 1,92 -0,70 1,37 0,05 0,20 0,06 0,06 0,04 0,04 4,18 1,06 4,25 Chemicals 37 3,22 3,59 0,20 1,34 0,19 0,15 0,11 0,08 0,07 0,05 4,35 1,43 4,60 Technology 54 2,42 2,68 -0,30 1,07 0,16 0,07 0,53 0,08 0,03 0,03 4,45 1,15 4,31 Others 8 2,98 4,47 -0,80 1,00 0,05 0,14 0,07 0,05 0,04 0,03 4,33 1,93 4,04 Total Sample 456 1,89 2,64 -0,02 1,14 0,17 0,40 0,09 0,10 0,52 0,05 4,22 1,40 3,79
34 Results of the regression
Table 4, table 5 and table 6 present the panel data analysis. Model 1 includes the
Independent variable CPA, CSR, and control variables CEO contribution, and firm size. Model 2 includes the Moderator CPA*CSR variable as well. Table 5 represents the results when
ROA is used as a dependable variable for CFP, Table 6 when ROE is used, and for table 7
ROI.
Table 4 Hypotheses Test results Independent variable ROA
ROA Model 1 Model 2 Adjusted R2 0,104 0,103 Beta SE Beta SE CPA 0,183** ,000 0,182** ,000 CSR 0,119** ,002 0,110* ,002 CEO -,011 ,001 -,013 ,001 Assets -0,300** ,005 -0,301** ,005 Employees 0,210** ,005 0,211** ,005 CPA*CSR ,033 ,003
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Taking a look at table 4, model 1 result in a model with an adjusted R2 of 0,104.The
model shows a positive relationship between ROA in regards to CPA, CSR, and firm size
measured in number of employees. CPA is positively related to ROA with a Beta of 0,183
and a p<0,01. CSR is positively related to ROA with a Beta of 0,119 and a p<0,01. Firm size
is both positively and negatively related to ROA, depending on how it is measured. The
amount of assets a firm has is negatively related to the ROA (Beta = -0,300, p<0,01) and
number of employees is positively related to ROA (Beta = 0,210, P<0,01). The amount of
CEO contribution has a negative effect on CFP, but seems to have no significant relationship
35
Model 2 tests if the CSR has a moderating effect on the CPA – CFP relationship. The
model has an adjusted R2 of 0,103 and is thus a weaker model than model 1. Model 2 shows
no significant difference in the results, the CPA*CSR moderator variable has been found to
be of no significant value in this model. Based on these results, CPA and CSR both are
positively related to the CFP when measured in ROA in both models. Therefor H1 and H2 are
thus supported. CPA has a positive relationship with the CFP of a firm and CSR has a positive
relationship with CFP. Because of the lack of support for a significant relationship between
the moderating variable CPA&CSR and ROA, H3 is not supported. There has been found no
prove of a moderating effect of CSR on the relationship between CPA and CFP.
Table 5 Hypotheses Test results Independent variable ROE ROE Model 1 Model 2 Adjusted R2 0,005 0,002 Beta SE Beta SE CPA 0,116* ,000 0,117* ,000 CSR ,070 ,017 ,074 ,017 CEO -,015 ,012 -,014 ,012 Assets -,104 ,044 -,103 ,044 Employees ,009 ,042 ,009 ,042 CPA*CSR -,013 ,024
*. Correlation is significant at the 0.05 level (2-tailed).
Looking at table 5, model 1 result in a model with an adjusted R2 of 0,005, and is thus
a weaker model then the ROA model. Model 1 result in a positive relationship between ROE
in regards to CPA. CPA is positively related to ROE with a Beta of 0,116 and a p<0.05. CSR
is positively related to ROE, but is not significant. Firm size is both positively and negatively
related to ROE, depending on how it is measured. The amount of assets a firm has is
36
relationships are not significant. The amount of CEO contribution has a negative effect on
ROE, but also has no significant relationship to the ROE of the firm.
Model 2 has resulted in no significant difference in the results, the CPA*CSR
moderator variable has been found to be of no significant value in this model. Based on these
results, CPA is positively related to the CFP when measured in ROE. Therefor H1 is
supported. CPA has a positive relationship with the CFP of a firm. Because of the lack of
support for a significant relationship between CSR and ROE, H2 is not supported. The
relationship between the moderating variable CPA&CSR and ROE is also not significant and
thus H3 is not supported. There has been found no prove of a moderating effect of CSR on the
relationship between CPA and CFP.
Table 6 Hypotheses Test results Independent variable ROI ROI Model 1 Model 2 Adjusted R2 0,071 0,068 Beta SE Beta SE CPA 0,204** ,000 0,204** ,000 CSR 0,106* ,004 0,104* ,004 CEO -,039 ,003 -,040 ,003 Assets -0,241** ,011 -0,241** ,011 Employees 0,150** ,010 0,150** ,010 CPA*CSR ,007 ,006
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Table 6 represents the models when ROI is used as a measurement for CFP. Model 1 has an
adjusted R2 of 0.071. Model 1 shows a positive relationship between ROI in regards to CPA,
CSR, and firm size measured in number of employees. The CPA variable is positively related