Political activities and social responsibilities : do they affect a firm’s financial health?

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

CFP

(ROE ROA ROI)

Independent variable

CSR

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