• No results found

Corporate political activity and corporate financial performance : the moderating effects of industry regulations and corporate governance

N/A
N/A
Protected

Academic year: 2021

Share "Corporate political activity and corporate financial performance : the moderating effects of industry regulations and corporate governance"

Copied!
44
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Corporate Political Activity and Corporate Financial Performance:

The Moderating Effects of Industry Regulations and Corporate

Governance

Thesis MSc Business Administration - Strategy Track Marc Vogels

11411015

Thesis Supervisor: Pushpika Vishwanathan June 23rd, 2017

(2)

Statement of originality

This document is written by student Marc Vogels 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.

(3)

Table of contents

Abstract………..3

Introduction………....4

Literature review………....7

Firm performance through market and nonmarket strategies………....7

Corporate political activity………....8

Corporate political activity and financial performance………...9

Agency theory and corporate political activity………...10

Industry regulations and corporate political activity………...13

CEO corporate governance and corporate political activity………...16

Outcome-based incentives and corporate political activity……….... 18

Behavior-based incentives and corporate political activity……….19

Method……….21

Results………..24

Discussion………....32

Major findings………..32

Contributions of this study………...35

Limitations and future research………....36

Conclusion………....38

(4)

Abstract

This research explores the relationship between Corporate Political Activity (CPA) and Corporate Financial Performance (CFP) in the United States, based on the integration of agency theory and the theory of market and nonmarket strategies Furthermore, this research postulates a potential moderating effect of industry regulations on the CPA-CFP relationship. In addition, this study hypothesizes a moderation effect of CEO outcome-based incentives and CEO behavior-based incentives on the CPA-CFP relationship, in the context of nonregulated industries. The study finds no direct relation between CPA and CFP outcomes. Furthermore, no moderating effect is found for industry regulations on the CPA-CFP relationship. In addition, no moderating effect is found for either CEO outcome-based incentives or CEO behavior-based incentives on the CPA-CFP relationship. These findings are incongruent with the study’s predictions and suggest that there is no connection between corporate political activity and corporate financial performance.

Keywords: Corporate Political Activity; Corporate Financial Performance; Industry

(5)

Introduction

In the United States, lobbying expenditures doubled from $1.45 billion in 1998 to $3.22 billion in 2015 (Center for Responsive Politics, 2017), and campaign contributions to the Democratic Party more than quadrupled from $129 million in 1998 to $616 million in 2016 (Federal Election Committee, 2017). It is indisputable that corporations spend huge amounts of money to gain influence in the political atmosphere, and corporate efforts have increased significantly over the last years (Lord, 2000).

Corporate Political Activity (CPA), which is defined as ‘corporate attempts to shape government policy in ways favorable to the firm’ (Hillman et al, 2004, p.838), has been researched in fields varying from strategic management and economics to sociology and political science (Hillman et al, 2004). However, (empirical) knowledge of the effects of CPA has been inconclusive thus far (Lord, 2000) and difficult to measure due to the complexity of the (political) environment (Shaffer, 1995). A variety of theories has been applied in the literature regarding the CPA-CFP relationship, (Mellahi et al., 2016). The most commonly used theories include the agency theory, institutional theory, resource based view and resource dependency theory (Mellahi et al., 2016) and different theories predict different relationship outcomes between CPA-CFP (Mellahi et al., 2016; Hadani & Schuler, 2013). During the last few years, a shift has been noticed from the use of a single theory towards use of multiple theories to investigate the relationship (Mellahi et al., 2016).

Research outcomes investigating the direct effects of CPA on firm performance have been mixed so far (Mellahi et al., 2016; Hadani & Schuler, 2013; Lux et al., 2011)Reviews and research agendas have stressed the importance of and need for more research towards firm performance outcomes of CPA (Hillman et al, 2004). Previous studies have found the relationship between CPA and organizational performance to be mainly positive (Mellahi et al., 2016). However, there has also been noticeable amount of mixed, contingent,

(6)

insignificant and even negative relations found in the period 2010-2014 (Mellahi et al., 2016; Hadani & Schuler, 2013).Therefore, the effects of CPA on CFP remain relevant for further investigation (Mellahi et al., 2016).

Besides the research on direct effects of CPA, there has also been research devoted to exploring the conditions under which CPA is most effective (Mellahi et al., 2016). Industry regulations has been an important factor in determining under which circumstances the CPA-CFP relationship will take place (Sun et al., 2016; Hadani & Schuler, 2013; Mellahi et al., 2016). Thus, industry regulations can be an important factor in determining CPA strategies (Masters & Keim, 1985). Keim & Zeithaml (1986) found that the amount of CPA differs among industries and can potentially explain a firm’s rationale for investing in political activities. Firms in regulated industries have a tendency to engage more in CPA than firms in non-regulated industries (Hillman & Hitt, 1999), and there is a potential positive moderating effect of industry regulations on CPA and CFP (e.g. Masters & Keim, 1985; Holburn & Vanden Bergh, 2008; Hill et al., 2013; Mellahi et al., 2016; Hadani & Schuler, 2013; Sun et al., 2016).

Besides industry regulations, another potential factor that determines the effectiveness of CPA is a firm’s corporate governance structure. While there is evidence that corporate governance leads to financial performance (Coates, 2012) and reduces agency problems (Eisenhardt, 1989), research is limited about how corporate governance mechanisms can influence CPA investments to lead to financial performance outcomes (Hillman et al, 2004). The current study is focused on CEO governance structures as the CEO plays a major role within a firm concerning strategic decision-making (Eisenhardt & Bourgeois III, 1988; Calori et al., 1994), (political) investments, risk-taking (Coles et al., 2006) and external ties with political actors (You & Du, 2011). As agency problems occur more in nonregulated

(7)

industries (Agrawal & Knoeber, 1996), the focus of the effectiveness of governance structures on the CPA-CFP relationship will be in the context of nonregulated industries.

As there is mixed evidence concerning the relation between CPA investments and CFP outcomes, this research focuses on the conditions under which CPA may prove to increase firm performance. The main research question of this thesis is as follows:

Under which circumstances do the corporate political activities of a firm in the United States affect its corporate financial performance?

This research contributes to the literature through the following: First, it elaborates on the literature concerning the relationship between corporate political activity (CPA) and corporate financial performance (CFP). While there has been a great amount of research on the CPA-CFP relationship, there is no clear outcome of the effects of CPA on CFP yet (Mellahi et al., 2016). Therefore, both a positive relationship between CPA and CFP is postulated, as well as an alternative hypothesis that assumes a negative relationship. Secondly, the research builds further on previous research (e.g. Hadani & Schuler, 2013) by using the agency theory as theoretical benchmark. The theory of market and nonmarket strategies is applied and integrated with agency theory in order to performance effects of nonmarket strategy (Mellahi et al., 2016). Literature on the bridging mechanisms between agency theory and nonmarket strategies has been limited thus far (Mellahi et al., 2016). Market and nonmarket strategies theory and agency theory are complementary in their explanation of firm performance. The market and nonmarket strategy theory is centered on rent creation through engagement in the market and nonmarket environment, while agency theory is focused on rent appropriation by management (agents) (Sun et al., 2016). The current study combines the two theories to understand the circumstances in which CPA

(8)

investments may prove to benefit the firm. Thirdly, the study sets out to confirm the previous research findings concerning industry regulations as a positive moderation effect for the CPA-CFP relationship (Hadani & Schuler, 2013). Fourthly, within the context of nonregulated industries, the study explores whether and how corporate governance mechanisms can reduce agency problems in CPA investments (Agrawal & Knoeber, 1996)

This thesis proceeds as follows: the following section will review the literature concerning the relationship between CPA and CFP through multiple theoretical approaches. Thereafter, the data and research method will be described and the results will be presented. Subsequently, the results will be discussed. The research will conclude with limitations and future research suggestions.

Literature review

The following section discusses the main contributions from the existing literature on Corporate Political Activity (CPA) and Corporate Financial Performance (CFP). The review starts by discussing CPA as a form of nonmarket strategy and assesses positive outcomes of CPA in terms of firm performance. Subsequently, agency theory is used to discuss a potential negative outcome of the CPA-CFP relationship. Thereafter, industry effects and regulations are considered as a potential moderating effect. Finally, the corporate governance structures behavior-based incentives and outcomes-based incentives within the context of nonregulated industries are discussed and are linked to the CPA-CFP relationship.

Firm Performance through Market and Nonmarket strategies

Firm performance relies on both market and nonmarket strategies (Holburn & Vanden Bergh, 2014; Shaffer et al, 2000; Baron, 1995). Market and nonmarket strategies have a similar structure, in the sense that both strategies act in a certain environment to increase

(9)

economic performance and add value to a firm’s core competencies (Shaffer et al., 2000). The market environment influences interactions between the firm and other parties through the market (Baron, 1995), while the nonmarket environment intermediates through stakeholders, government, the media and other (public) institutions (Baron, 1995). Nonmarket strategies of a firm are focused on social and political actors, while market strategies focus on interactions within the ‘traditional’ market (Shaffer et al., 2000) The traditional market is more concerned with the stakeholders such as buyers, suppliers and competitors that have an influence on the products in the market environment (Shaffer et al., 2000), and entails a market-like exchange (e.g. buying, selling). Nonmarket strategies shape the firm’s market environment in a less direct manner through issues, institutions, interests and information (Baron, 1995). Firms develop nonmarket strategies to engage in the political market and, among other things, achieve favorable legislative, executive and policy outcomes (Bonardi et al, 2006). By interacting with the political market, in turn, a firm can shape the market environment and improve performance in the market environment (Baron, 1995; Holburn & Vanden Bergh, 2014). CPA can be both complementary as a substitute for market strategies (Hillman et al., 2004). Research in nonmarket strategy has been present for more than four decades and has demonstrated that successful nonmarket strategies are relevant for firm survival, organizational performance and potential sustainable competitive advantage (Mellahi et al., 2016).

Corporate Political Activity

A component of nonmarket strategy is a firm’s engagement in Corporate Political Activity (CPA) (Mellahi et al., 2016; Lux et al., 2011). The literature around CPA is diversified in several scholarly disciplines and fields of research (Shaffer, 1995). In strategic management, CPA studies are focused on government regulations in industries and the

(10)

corporate endeavors on gaining a competitive advantage through controlling the political agenda (Shaffer, 1995; Schuler et al., 2002).

Two frequently researched channels of CPA are Political Action Committee (PAC) campaign contributions to political parties and lobbying activities (Hillman et al., 2004; Lord, 2000; Hill et al., 2013; Lux et al., 2011). The channels operate in different manners (Hill et al., 2013). In the case of PAC contributions, firms are not allowed by law to directly contribute to a political party campaign. Instead, firms can set up Political Action Committees, in which employees and their relatives can make contributions. In turn, the PAC can support political parties in their elections by donating funds (Hill et al., 2013). On the contrary, lobbying expenditures are financed by corporations directly (Hill et al., 2013). Lobbying is defined as ‘a set of activities by which firms, or their representatives, attempt to establish communication channels with government representatives or their staffs.’ (McWilliams et al., 2002, p.710). Firms often use both channels simultaneously to gain favorable policy outcomes (Hill et al., 2013; Schuler et al., 2002). PAC contributions can be seen as an ex ante strategy to gain political influence, while lobbying activities are focused on policymakers after they have been enlisted/elected (Hill et al., 2013). Hill et al. (2013) found that firms invest more in lobbying activities than PAC contributions.

Corporate political activity and financial performance

There are various motivations for corporations to engage in CPA (Schuler et al., 2002; Oliver & Holzinger, 2008; Lux et al., 2011; Hadani & Schuler, 2013). One of the main determinants of a firm’s decision to take political action is to increase firm value (Oliver & Holzinger, 2008). The engagement in CPA is often perceived as an investment decision, in which resources are allocated to gain higher revenue in return (Lux et al., 2011). As a component of nonmarket strategy, CPA enables firms to interact with the market

(11)

environment through the (nonmarket) political environment. For example, Claessens et al., (2008) found that firms that provided contributions to federal deputies enjoyed higher stock returns during the elections of 1998 and 2002. Besides that, Cooper et al., (2010) discovered that contributions to U.S. political campaigns in the period of 1979 to 2004 were positively associated with future returns. This is referred to as value creation (Oliver & Holzinger, 2008). Besides value creation, another motivation for firms to take strategic action in the political arena is to protect their current position in the market (Oliver & Holzinger, 2008). This is referred to as value maintenance (Oliver & Holzinger, 2008), and efforts range from protecting their firm from (foreign) competition (Schuler, 1996, in Oliver & Holzinger, 2008; McWilliams et al., 2002), to mitigating transaction costs in the political market (Bonardi et al 2006). CPA can have direct and indirect effects on the financial performance of a firm (Hadani & Schuler, 2013). Profits can be attained in a direct manner via CPA through permits to operate or creating barriers for competition in the market (Hadani & Schuler, 2013). In an indirect manner a firm can boost its performance through influencing political policies to a beneficial situation for the firm (Hadani & Schuler, 2013; Hillman et al., 1999). As previously stated, these forms of CPA are grouped under nonmarket strategies and have proven to increase firm performance in the past (Mellahi et al., 2016; Hillman et al., 1999; Shaffer et al., 2000). Through nonmarket activities a firm is competitive in the market environment (Baron, 1995). Based on previous research and the theory of market and nonmarket strategies, the following hypothesis can be built:

Hypothesis 1a: Corporate political investments will be positively associated with corporate financial performance.

Agency theory and corporate political activity

An alternative hypothesis concerning the CPA-CFP relationship is formulated based on agency theory, which is a contrasting approach to understanding corporate political

(12)

activity. Agency theory is ‘directed at the ubiquitous agency relationship, in which one party (the principal) delegates work to another (the agent), who performs that work.’ (Eisenhardt, 1989, p.58). The key problem addressed in agency theory is that managers (the agent) may act opportunistic, at the cost of the firm’s owners (the principal) and other stakeholders (Jensen and Meckling, 1976, in Wang et al., 2008). When the agent has different interests and information from the principal, agency problems arise (Eisenhardt, 1989). Subsequently, various agency costs emerge in the relationships between the principal and agent, damaging the economic welfare of the involved parties, unless the opportunistic managerial behavior can be limited (Eisenhardt, 1989; Wang et al., 2008).

Agency theory has been used to examine the link between non-market strategies and firm performance (Mellahi et al., 2016) and potentially explains rationales for managers’ investments in political donations (Aggarwal et al., 2012; Hadani & Schuler, 2013). Agency theory predicts that managerial engagements in non-market activities are pursued because of private benefits and are at the expense of the shareholders (Mellahi et al., 2016). Besides that, shareholders are more likely to encounter agency problems in the CPA context due to the limited legal disclosure of CPA investments (Coates, 2012). Previous research has found that CPA investments are associated with higher measures of managerial agency costs and lower shareholder power (Coates, 2012). Therefore, agency theory predicts a negative relationship between CPA and CFP (Mellahi et al., 2016; Hadani & Schuler, 2013; Aggarwal et al., 2012).

Among other potential reasons, there are several reasons based on agency theory that postulate a negative relationship between CPA and CFP (Hadani & Schuler, 2013). First, senior managers that invest in CPA have higher risk-taking in decision-making (Hadani & Schuler, 2013). The firms that were engaged in political activities took higher risks and were associated with a decreased performance (Igan et al., 2009, in Hadani & Schuler, 2013). This

(13)

is an example of a moral hazard contracting problem, as these firms left the reduced financial performance consequences of the risk-taking to the government for financial assistance (Eisenhardt, 1989; Igan et al., 2009, in Hadani & Schuler, 2013).

Secondly, managers invest in CPA through poor decision-making (Hadani & Schuler, 2013). The CPA investments were done with deficient evaluation of the potential profit, instead of investing in more profitable (non)market strategies (Hadani & Schuler, 2013). This can refer to the bounded rationality (Eisenhardt, 1989) of managers, as they have to make decisions in complex environments and do not have full information on the effects of the investments.

Thirdly, there are monitoring problems (Jensen, 1993) for shareholders in assessing CPA investments in comparison to other (non)market strategies (Hadani & Schuler, 2013; Coates, 2012). While there is a legal obligation for firms to report on their CPA investments to government institutions, the firms do not have to present these numbers on their financial reports (Hadani & Schuler, 2013). This creates information asymmetries (Eisenhardt, 1989) between the managers that are engaged in CPA and the shareholders, therefore creating agency costs (Eisenhardt, 1989; Hadani & Schuler, 2013; Yu & Yu, 2011).

Fourthly, senior managers’ personal interests and needs may be the driving force behind CPA investments (Hadani & Schuler, 2013; Coates, 2012; Aggarwal et al., 2012; Mathur et al., 2013). Instead of being strategic investments for profit maximization, these (non)market investments can be due to image reinforcement and needs for attention (Petrenko et al., 2015). Besides that, managers may invest in CPA based on their personal political preference, or because they want to pursue a political career (Coates, 2012). In turn, this relates to agency problems as the interests of the senior manager are diverging from the shareholders (Eisenhardt, 1989).

(14)

investing in CPA is managerial entrenchment (Mathur et al., 2013). Managers invest in political relations in order to make themselves more valuable and needed for the firm (Shleiffer & Vishny, 1989; Mathur et al., 2013). As political ties can be viewed as a type of manager-specific investment, it can lead to managerial entrenchment (You & Du, 2011). These investments are more aimed at the manager’s interest, which is remaining in a strong position in the firm. Firms with more entrenched management are more likely to engage in lobbying activities (Mathur et al., 2013). Previous research has found that entrenched boards create lower market value (Bebchuk & Cohen, 2005).

Sixthly, another reason besides entrenchment is the manager's decision horizon (Antia et al., 2010). Managers prefer to pursue short-term results at the expense of long-term firm performance. Decisions that may be optimal for a manager may be suboptimal for a firm (Antia et al., 2010). Related to manager-specific investments, short-term gains from investments in political ties may evaporate once a manager leaves the firm. Based on these six reasons, the following alternative hypothesis is stated:

Hypothesis 1b: Corporate political investments will be negatively associated with corporate financial performance.

Industry regulations and corporate political activity

Regulations concerns the ‘limitations on firm behaviors placed by governments’ (Hadani & Schuler, 2013, p.170) and is a factor in determining corporate political strategy (Sun et al., 2016; Hillman, 2005). It is in the interest of the firm to make sure that governmental policies are created in their favor in order to remain competitive in terms of (non)market strategies (Hadani & Schuler, 2013; Hillman et al., 1999). In these business-government interactions there are several forms of regulation, including antitrust regulations, economic regulations and social regulations (Hillman et al., 1999). These regulations ensure

(15)

competitiveness of the market and industry in terms of pricing, production, licensing, and also deal with issues such as environmental law and labor issues (Hillman et al., 1999). Over the years in the United States, there has been an increase in regulatory policies created by political decision-makers and legislators in the U.S. government (Hillman et al., 1999). With the increasing regulations, firms need to manage the increased uncertainty (Hillman et al., 1999) and remain competitive (Hadani & Schuler, 2013).

Based on the theory of market and nonmarket strategies, it is rational that firms dealing with increased regulations invest more in CPA (Baron, 1995). Regulation is a major driving force in the nonmarket environment of business (Hillman, 2005). Firms that are more government-controlled than market-controlled rely more on nonmarket strategies for firm performance than the market-controlled ones (Baron, 1995). As CPA is a form of nonmarket strategy, it can be argued that regulated firms will invest more in CPA to increase firm performance in the market environment (Baron, 1995).

There are several mechanisms how non-market activities help firms in regulated industries. First, firms in heavily regulated industries are more affected by regulatory policies and therefore need to invest more in having these policies serve their best interest (Sun et al., 2016). Among other reasons, firms lobby in order to gain favorable regulation rules (Yu & Yu, 2011) and gain political intelligence to better react to incoming policy changes (Yu & Yu, 2011). When regulatory policies have a higher impact on regulated firms, these firms are more likely to engage in CPA to improve their performance (Sun et al., 2016; Coates, 2012; Hadani & Schuler 2013). Holburn & Vanden Bergh (2014) used the context of the electricity sector (as regulated industry) and found that firms use campaign contributions to influence regulatory merger approvals. Besides that, Masters & Keim (1985) discovered that firms competing primarily in the (regulated) transport, energy and communications (TEC) industries have higher PAC activity than in other industries. In addition to PAC activity,

(16)

lobbying also varies between industries and is of greater importance to firms in industries that are regulated in a more extensive manner (Hill et al 2013).

Secondly, regulated firms can engage more directly in targeting political actors to gain favorable policies (Hadani & Schuler, 2013; Holburn & Vanden Bergh, 2008). This allows to build up more steady interactions between the firms and the policy makers and to ensure long-term relationships (Hadani & Schuler, 2013). As the government is important in limiting and controlling firm behaviors, in turn this can potentially reduce agency problems (Hadani & Schuler, 2013). Previous research has found that firms in regulated industries gain higher performance with more CPA investments than firms that invest less (Hadani & Schuler, 2013).

Thirdly, firms in heavily regulated industries use CPA to reduce and avoid regulatory costs and legal penalties (Sun et al., 2016). For example, Yu & Yu (2011) examined the relation between lobbying activities and fraud detection and discovered that firms that lobby have lower hazard rates of detection for fraud, longer evasion of fraud detection, and are less likely to be detected by regulators (Yu & Yu, 2011). Industries vary in terms of managerial fraud (Zahra et al., 2005). Firms in industries that are more regulated (e.g. petroleum refining, transportation, financial services and health care) have a long history with fraud and are more tolerant of managerial fraud (Zahra et al., 2005). A rationale for firms in regulated industries is to lobby in order to reduce the costs and penalties of fraud. Besides fraud, Correia (2014) found that long-term contributions and lobbying can lead to lower penalties and costs imposed by the Securities and Exchange Commission (SEC). These political investments spent on the SEC directly are found to be more effective than other expenditures to reduce enforcement costs (Correia, 2014). As regulatory policies affect firms more in regulated industries (Sun et al., 2016) and are more present in these industries, regulatory compliance costs are likely to be higher in regulated industries than in non-regulated industries. As a

(17)

consequence, CPA investments by a firm in a regulated industry may prove to be more effective in terms of compliance cost-reduction.

As firms in regulated industries are more affected by regulatory policies (Sun et al., 2016), can target political decision makers more directly (Hadani & Schuler, 2013; Holburn & Vanden Bergh, 2008), and reduce regulatory compliance costs (Correia, 2014), the following hypothesis can be stated:

Hypothesis 2: industry regulations have a positive moderation effect on the relationship between corporate political activity and corporate financial performance.

CEO corporate governance and corporate political activity

As discussed in the section of the alternative hypothesis concerning CPA-CFP, agency problems might occur when CPA is used as a nonmarket strategy. Governance structures can minimize the agency costs associated with the previously discussed agency problems (Eisenhardt, 1989). Concerning CPA investments, the current study focuses on chief executive officer (CEO) governance structures in assessing the effect of CPA investments on CFP based on several reasons.

First, CEOs play a huge role in determining strategy and the making of investments (Calori et al., 1994), and have the power to make these decisions (Daily & Johnson, 1997). A CEO has to deal with uncertainty and ambiguity in its decision-making, and has a selective perception of the environment and the matters that are relevant to making those decisions (Calori et al., 1994). The important role of the CEO and bounded rationality (Eisenhardt, 1989) in decision-making are relevant to CPA investments as the CEO can determine the scope of the political investments.

Secondly, CEOs determine the political agenda of the firm and have their own political agenda (Eisenhardt & Bourgeois III, 1988). Related to decision-making, CEOs can

(18)

make decisions with uncertain outcomes through the exercise of power and coalition forming (Eisenhardt & Bourgeois III, 1988). If a CEO has a political preference, for example, it can steer the strategic decision process of the firm into investing in a certain political party. In turn, this may not benefit firm performance.

Thirdly, CEOs are involved in maintaining external ties with political actors (You & Du, 2011). As part of their function, CEOs maintain close contact with policy makers and can gain competitive advantages through the creation of political ties (You & Du, 2011). While this could improve the firm’s performance, it also could enable the CEO to strengthen its own position in the firm. As political ties can be viewed as a type of manager-specific investment, this can lead to managerial entrenchment (You & Du, 2011).

Fourthly, the CEO is important when it comes to risk-taking in investments (Coles et al., 2006). A CEO can be risk-averse (Mehran, 1995), or may be taking too much risk at the expense of the firm (Hadani & Schuler, 2013). Investments in political activity can be quite risky as their outcomes are not entirely clear (Hadani & Schuler, 2013).

These reasons show the important role of the CEO in the context of CPA investments. Therefore it is important to discuss how corporate governance creates optimal conditions to ensure that CPA investments benefit the firm.

In order to make sure that the CEO acts in the best interest of the firm, there is need for effective governance incentives, such as behavioral and outcome-based contracts (Eisenhardt, 1989). Besides handling in the best interest of the firm, these contracts share the risk between the CEO and shareholders (Eisenhardt, 1989). According to Core et. al (1999), firms with weaker governance structures have greater agency and, in turn, their firm performance is worse. Vice versa, shareholder-friendly corporate governance structures are associated with positive firm value (Coates, 2012). This study focuses on outcome-based incentives (CEO link to shareholder returns ) and behavior-based incentives (CEO duality).

(19)

Outcome-based incentives and corporate political activity

A form of an outcome-based contract is linking the chief executive officer’s (CEO) compensation to shareholder returns (Eisenhardt, 1989). In this governance structure, the CEO’s payment is dependent on the total profit of the firm. This pay structure will ensure that the CEO’s interests are aligned with the shareholder’s interests, as these forms of compensation reward executives when they increase firm performance (Carpenter & Sanders, 2002). If the CEO’s own wealth is tied to the returns made by the firm, its decision-making is more focused on making more value-maximizing decisions (Coles et al., 2006; Mehran, 1995).

Outcome-based incentives are likely to influence CPA investments, as it may lead CEOs to make more effective CPA investments that will increase firm performance. This is because the problem of diverging interests will be (partially) resolved. Personal benefits of the CEO include image reinforcement, needs for attention (Petrenko et al., 2015), managerial entrenchment (Mathur et al., 2013), self-aggrandizement (Hadani & Schuler, 2013), or the pursuit of a political career (Coates, 2012). Furthermore, CEOs can have their own political agenda and steer the agenda of the board towards investing in certain political parties that may not benefit the firm’s performance (Coles et al., 2006). The firm, on the other hand, benefits more from value-maximization. When the CEO’s compensation is linked to the firm’s performance, the CEO will be more likely to invest in order to increase this performance (Carpenter & Sanders, 2002; Hadani & Schuler, 2013).

The current study hypothesizes that the moderating effects of CEO outcome-based contracting will be more salient in nonregulated firms. As agency problems occur more in nonregulated industries (Agrawal & Knoeber, 1996) governance structures will be more effective in reducing agency costs in nonregulated industries. Accordingly, the following hypothesis is stated:

(20)

Hypothesis 3: In nonregulated industries, corporate governance outcome-based incentives have a positive moderation effect on the relationship between corporate political activity and corporate financial performance.

Behavior-based incentives and corporate political activity

Besides the outcome-based incentives, there are also behavior-based incentives (Eisenhardt, 1989). For example, the board of directors serves as an internal control system at the top of a firm (Jensen, 1993) and as an information system for shareholders to monitor potential opportunistic behavior of top management (Eisenhardt, 1989; Hillman & Dalziel, 2003). Besides that, the board can hire, fire and reward the CEO and provide strategic advice (Jensen, 1993). The board as monitoring system is important as it can reduce the potential agency costs when management serves its own interests at cost of the firm (Hillman & Dalziel, 2003). In turn, agency cost reduction leads to firm performance (Hillman & Dalziel, 2003).

The configuration of the board of directors affects how well it can function (Lipton & Lorsch 1992). A type of structure is the independent leadership structure (Rechner & Dalton, 1991), in which the CEO and chairperson of the board are two different persons. Previous research has shown that firms with an independent leadership structure outperform those with a dual leadership structure (Rechner & Dalton, 1991). A separation of the CEO and the chairperson increases the amount of control the board has on top management through an independent chairperson (Fama & Jensen, 1983).

In the context of CPA, the behavior-based contract (CEO-chairman separation) can benefit the firm performance in several ways. Firstly, it reduces the amount of hierarchical power of the CEO (Daily & Johnson, 1997). Instead of the CEO, the chairman is able to determine the agenda of board meetings, lead the meetings and decide the information that is

(21)

received by the board members (Daily & Johnson, 1997). In terms of decision-making and risk-taking in CPA, the chairman is able to control the CEO’s activities concerning (non)market investments. Consequently, the CEO is less able to invest based on political preference, personal needs, or other personal benefits at the cost of firm performance.

Secondly, CEO-chairman separation reduces monitoring problems (Jensen, 1993; Daily & Dalton, 1997). As a representative of the shareholders, the chairperson can demand the CEO to provide more information concerning CPA investments. While it is not obligatory to present CPA investments in the financial reports, the chairman can put pressure on the CEO and top management to provide this information. This way of monitoring can influence the CEO to invest less in CPA that does not improve firm performance, because the board can find out about these investments and disciplinary measures may follow.

Thirdly, related to corporate political activities, independent leadership structure in politically connected firms is perceived as less risky than a dual leadership structure (Bliss & Gul, 2012). Consequently, firms with an independent leadership structure need to pay lower audit fees because of less external auditing, and are charged with lower interest rates by lenders (Bliss & Gul, 2012). Summarizing, besides internal monitoring and controlling, an independent leadership structure in a politically active firm may also prove to be more beneficial from an external view.

As with the outcome-based incentives, this research expects the moderating effects of behavior-based incentives to also be more prevalent in the context of nonregulated industries (Agrawal & Knoeber, 1996). Related to behavioral-based incentives, monitoring is less in nonregulated industries (Demsetz & Lehn 1985). The monitoring and controlling through behavioral-based incentives will ensure that political strategies are aimed to benefit the firm. Therefore, the following hypothesis is stated:

(22)

incentives have a positive moderation effect on the relationship between corporate political activity and corporate financial performance.

Method

Description of initial steps in the analysis

In the current study, 593 firms listed in the Fortune 500 between 2012 and 2015 are examined. The website of the Center for Responsive Politics (2017) provides the lobbying expenditures and amounts of PAC campaign contributions to political parties of the aforementioned firms, as was done in previous research (e.g. Hadani & Schuler, 2013). First, a frequency check is performed to detect errors in the dataset. Thereafter, missing values are excluded listwise. From the original 593 firms, a sample of 528 firms is used that include complete Corporate Political Activity (CPA) data. After excluding missing values listwise for the financial performance and control variables as well, a final sample of 405 firms is included.

Variable description

Dependent Variables

Return on assets: Corporate Financial Performance is measured through an accounting-based

measure, namely return on assets (ROA) (Hillman, 2005). This measure is derived from the Compustat database. Return on assets is widely used to measure firm performance (Carpenter & Sanders, 2002) as it reveals how efficient a firm is in making use of its assets. A four-year average is taken between 2012 and 2015 to deal with potential abnormalities in a specific year’s performance.

(23)

Independent Variables

Lobbying expenditures: Corporate Political Activity is measured through Lobbying

expenditures by firms (e.g. Hill et al., 2013; Hadani & Schuler, 2013). Data for lobbying expenditures is derived from the Center for Responsive Politics (2017) and its website opensecrets.org.

Political Action Committee (PAC) contributions: A different way of measuring Corporate

Political Activity is through PAC contributions (e.g. Hill et al., 2013; Hadani & Schuler, 2013). The data for PAC contributions is also derived from the Center for Responsive Politics (2017) and its website opensecrets.org.

Moderator variables

Industry regulations: A dummy measure of industry regulations is created to differentiate

between regulated and nonregulated industries the research, as was done by Coates (2012). Among others, regulated industries include telecommunications, banks, pharmaceuticals and utilities (Coates, 2012). A firm was coded ‘1’ when it belongs to a regulated industry and ‘0’ otherwise (e.g Coates, 2012; Sun et al., 2016). The firms’ SIC codes are derived from Compustat and the regulated industries list is based on Coates’ (2012) classification.

CEO compensation link to shareholder returns: To test the outcome-based governance

incentives, the variable ‘CEO compensation link to shareholder returns’ is used. For this variable, a dummy variable is created. This dichotomous variable differs between whether a CEO is linked to the shareholder returns. The firm is coded ‘1’ if it has the compensation linked and a ‘0’ if it has not. The data for CEO compensation link to shareholder returns is derived from Datastream - Asset4.

CEO-chairman separation: To test the behavior-based governance incentives, the variable

(24)

was ‘1’ if it has a separate CEO and chairman and ‘0’ if it has a dual leadership structure (Daily & Johnson, 1997; Aggarwal et al, 2012). The data for CEO-chairman separation is also taken from Datastream - Asset4.

Control Variables

Firm size: As firm size is affiliated with firm performance and governance (Dalton et al.,

1998, in Hillman, 2005), firm size is included as a control variable. Firm size is measured as the logged number of employees of a firm. Measurements for firm size are derived from Compustat.

Board size: A well-known corporate governance structure is the board size of a firm, which

can affect firm performance (Core et al., 1999; Lipton & Lorsch, 1992). Board size is used as a control variable to exclude potential performance differences caused by board size differences, as has been done in previous research (Hadani & Schuler, 2013). Measurements for board size are derived from Compustat.

Leverage: Leverage is the ratio of debt to assets, and has been previously used in researches

concerning CPA (see Coates, 2012) as a control variable. Measurements for leverage are derived from Compustat.

(25)

Results

The mean, standard deviations and correlations are found in Table 1. Several significant correlations were found, with Lobbying expenditures and PAC contributions having the strongest correlation (r=0.741, p < 0.01). No significant correlation was found between lobbying expenditures and return on assets. Besides that, there was a non-significant correlation between PAC contributions and return on assets.

Table 1. Means, Standard Deviations, Correlations

Variables M SD 1 2 3 4 5 6 7 8 9 1. Lobbying expenditures 2067023.19 3118969.42 - 2. PAC contributions 247818.07 415872.99 .741** - 3. Firm size 1.47 .51 .341** .341** - 4. Leverage 5.49 31.66 .045 0.85 .030 - 5. Board size 11.16 1.80 .275** .260** .232** .024 - 6. Industry regulations 0.32 .47 .245** .215** -.127* -.023 .268** - 7. CEO compensation link to shareholder returns 0.64 .48 .129** .139** -.066 .052 .172** .079 - 8. CEO chairman separation 0.62 .48 .075 .106* .110* .040 .090 -.057 .062 - 9. Return on assets 0.05 .07 .064 .020 .218** -.015 .007 -.056 -.009 .065 -

** Correlation is significant at the p < 0.01 level (2-tailed) * Correlation is significant at the p < 0.05 level (2-tailed) Listwise N=405

(26)

Direct effect of corporate political activity on corporate financial performance

Hierarchical multiple regression (table 2) was performed to investigate the potential effect of lobbying expenditures and PAC contributions on return on assets, after controlling for firm size, leverage and board size. In the first model of the hierarchical multiple regression, the predictor firm size, leverage and board size were entered. This model proved statistically significant (F (3, 407) = 7.05, p < .001) and explained 4.9% of variance in return on assets. After entry of lobbying expenditures and PAC contributions in Step 2, the model showed no significance in explaining variance. In the final model only one out of five variables was statistically significant, which was the firm size (β = .24, p < .001). In other words, if a firm’s size increases for one, the return on assets will increase for 0.24. For the rest, there is no effect on return on assets if lobbying expenditures or PAC contributions increases for one.

Table 2. Hierarchical multiple regression

R R2 R2 Change B SE β t Step 1 .22 .05* Firm size .03 .01 .23* 4.58 Leverage -.00 .00 -.02 -.45 Board size -.00 .00 -.05 -.99 Step 2 .23 .05 .00 Firm size .03 .01 .24* 4.51 Leverage -.00 .00 -.02 -.35 Board size -.00 .00 -.05 -.88 Lobbying expenditures .00 .00 .07 .94 PAC contributions -.00 .00 -.10 -1.35

* Correlation is significant at the p < 0.001 level (2-tailed) Listwise N=405

(27)

Industry regulations moderating effect on CPA-CFP relationship

A simple moderation analysis (table 3)was performed via the Hayes (2013) Process macro for SPSS, model 1, to investigate the potential moderating effect of industry regulations on the CPA - CFP relationship. A standardized lobbying expenditures was used as independent variable, a standardized firm size, leverage and board size as control variables, and return on assets as dependent variable. Industry regulations was used as a dummy variable. The interaction effect between standardized lobbying expenditures and industry regulations was found insignificant, meaning that no moderation effect is taking place.

Table 3. Simple Moderation effects of industry regulations on lobbying expenditures - return on assets

Coefficient SE t p constant .049 .00 10.09 <.0001 ZLobbying Expenditures(X) c1 .00 .00 .26 .796 Industry Regulations (M) c2 -.00 .01 -.38 .701 int_1 (XM) c3 -.00 .00 -.22 .826 ZFirmsize .02 .00 3.25 .001 ZLeverage -.00 .00 -.49 .621 ZBoardsize -.00 .00 -1.02 .308 R2 =.050 p <0.001 F(6,404)=3.982

(28)

Another simple moderation (table 4) was performed through the A.F. Hayes (2013) Process add-on to investigate the potential moderating effect of industry regulations on the CPA - CFP relationship. This time, a standardized PAC contributions was used as independent variable, a standardized firm size, leverage and board size as control variables, and return on assets as dependent variable. Industry regulations was again used as a dummy variable. The interaction effect between standardized PAC contributions and industry regulations was again found insignificant, meaning that there is no moderation effect is taking place either.

Table 4. Simple Moderation effects of industry regulations on PAC contributions - return on assets

Coefficient SE t p constant .049 .00 9.93 <.0001 ZPAC contributions(X) c1 -.00 .00 -.74 .459 Industry Regulations (M) c2 -.00 .01 -.13 .893 int_1 (XM) c3 -.00 .00 -.15 .882 ZFirmsize .02 .00 3.50 .001 ZLeverage -.00 .00 -.46 .648 ZBoardsize -.00 .00 -.86 .390 R2 =.052 p <0.01 F(6,404)=3.292

(29)

CEO compensation moderating effect on CPA-CFP in nonregulated industries

Several simple moderation analyses (table 5) were performed via the Hayes (2013) Process macro for SPSS, model 1, to investigate the potential moderating effect of CEO compensation link to shareholder returns on the CPA - CFP relationship in the context of nonregulated industries. Standardized lobbying expenditures in regulated industries and in non-regulated were used as independent variables. Standardized firm size, leverage and board size were again used as control variables, and return on assets was used as dependent variable. CEO compensation link to shareholder returns was used as a dummy variable. The interaction effect between standardized lobbying expenditures and CEO compensation link to shareholder returns was found insignificant both in regulated as in the non-regulated context, meaning that no moderation effect is taking place.

Table 5. Simple Moderation effects of CEO compensation on lobbying expenditures - return on assets

Regulated Non regulated Coefficient SE t p Coefficient SE t p constant .040 .01 4.96 .00 .048 .01 7.89 .00 ZLobbying expenditures (X) c1 .00 .00 .74 .46 -.01 .01 -1.17 .24 CEO compensation (M) c2 .01 .01 .98 .33 .00 .01 .05 .96 int_1 (XM) c3 .01 .01 1.27 .21 .01 .01 1.24 .22 ZFirmsize .00 .01 .19 .85 .02 .01 3.46 .00 ZLeverage -.00 .01 -.14 .88 -.00 .00 -.36 .72 ZBoardsize -.01 .00 -1.93 .06 -.00 .00 -.16 .87 R2 =.060 p > 0.05 R2 =.070 p <0.05 F(6,123) =1.877 F(6,268)= 2.433

(30)

Besides using lobbying expenditures as independent variables, two simple moderation analyses (table 6) were performed with PAC contributions as independent variables in regulated and in non-regulated industries. Standardized firm size, leverage and board size were again used as control variables, and return on assets as dependent variable. CEO compensation link to shareholder returns was used as a dummy variable. The interaction effect between standardized PAC contributions and CEO compensation link to shareholder returns was found insignificant both in regulated as in the non-regulated context, meaning that no moderation effect is taking place.

Table 6. Simple Moderation effects of CEO compensation on PAC contributions - return on assets

Regulated Non regulated Coefficient SE t p Coefficient SE t p constant .04 .01 5.01 .00 .05 .01 7.96 .00 ZPAC contributions (X) c1 .00 .00 .02 .98 -.01 .01 -2.00 .05 CEO compensation (M) c2 .01 .01 1.05 .30 .00 .01 .23 .82 int_1 (XM) c3 .00 .00 .71 .48 .01 .01 1.58 .12 ZFirmsize .00 .01 .87 .38 .02 .01 3.54 .00 ZLeverage -.00 .02 -.18 .86 -.00 .00 -.34 .74 ZBoardsize -.01 .00 -1.75 .08 -.00 .00 -.02 .99 R2 =.038 p > 0.05 R2 =.073 p <0.05 F(6,123) =.823 F(6,268)= 2.733

(31)

CEO-chairman separation moderating effect on CPA-CFP in nonregulated industries To investigate the potential moderating effect of CEO-chairman separation on the CPA - CFP relationship in the context of nonregulated industries, two simple moderation (table 7) were performed. Standardized lobbying expenditures in regulated industries and in non-regulated were used as independent variables. Standardized firm size, leverage and board size were again used as control variables, and return on assets as dependent variable. CEO-chairman separation was used as a dummy variable. The interaction effect between standardized lobbying expenditures and CEO-chairman separation was found insignificant both in regulated as in the non-regulated context, meaning that no moderation effect is taking place.

Table 7. Simple Moderation effects of CEO-chairman separation on lobbying expenditures - return on assets

Regulated Non regulated Coefficient SE t p Coefficient SE t p constant .04 .01 6.26 .00 .04 .01 6.12 .00 ZLobbying expenditures (X) c1 .01 .01 1.39 .17 -.00 .01 -.58 .56 CEO-chairman separation (M) c2 .00 .01 .37 .71 .01 .01 .95 .34 int_1 (XM) c3 -.01 .01 -.80 .42 .00 .01 .52 .60 ZFirmsize .00 .01 .06 .95 .02 .01 3.20 .00 ZLeverage -.00 .02 -.16 .88 -.00 .00 -.41 .68 ZBoardsize -.01 .00 -1.71 .09 -.00 .00 -.41 .68 R2 =.052 p > 0.05 R2 =.071 p <0.05 F(6,123) =1.539 F(6,268)= 2.716

(32)

Besides using lobbying expenditures, several simple moderation analyses (table 8) were performed to investigate the potential moderating effect of CEO-chairman separation on the CPA - CFP relationship. Standardized PAC contributions were used as independent variables in regulated and in non-regulated industries. Standardized firm size, leverage and board size were again used as control variables, and return on assets was used as dependent variable. CEO-chairman separation was used as a dummy variable. The interaction effect between standardized PAC contributions and CEO-chairman separation was found insignificant both in regulated as in the non-regulated context, meaning that no moderation effect is taking place.

Table 8. Simple Moderation effects of CEO-chairman separation on PAC contributions - return on assets

Regulated Non regulated Coefficient SE t p Coefficient SE t p constant .04 .01 6.44 .00 .04 .01 6.14 .00 ZPAC contributions (X) c1 .01 .01 .56 .64 -.02 .01 -1.35 .18 CEO-chairman separation (M) c2 .00 .01 .47 .64 .01 .01 .97 .33 int_1 (XM) c3 -.00 .01 -.34 .74 .02 .01 1.24 .21 ZFirmsize .00 .01 .54 .59 .02 .01 3.36 .00 ZLeverage -.00 .02 -.23 .82 -.00 .00 -.40 .69 ZBoardsize -.01 .00 -1.52 .13 -.00 .00 -.14 .89 R2 =.029 p > 0.05 R2 =.080 p <0.05 F(6,123) =.739 F(6,268)= 2.770

(33)

Discussion

The aim of this study was to investigate under which circumstances the corporate political activities of a firm in the United States affect its corporate financial performance. Overall, the results of the data analysis proved to be incongruent with the stated hypotheses. First, the major findings and contributions will be reviewed in the discussion section. Thereafter, the limitations and recommendations future research will be discussed. Finally, the discussion will end with concluding remarks.

Major findings

Earlier studies between 2000-2014 have found the relationship between CPA and organizational performance to be predominantly positive (Mellahi et al., 2016). However, a considerable amount of mixed, contingent, insignificant and even negative relations has been found in the period 2010-2014 (Mellahi et al., 2016; Hadani & Schuler, 2013). In the current study, no relationship was found between corporate political activity and corporate financial performance, contrasting previous research findings and rejecting the first hypotheses concerning either a positive or negative relationship.

The finding that CPA is not related to CFP has important implications for the discussion regarding the effectiveness of CPA as a nonmarket strategy. Two alternative theories can be stated to explain the non-significant relation between CPA and CFP. First of is the behavioral theory of the firm, which depicts that firms face uncertainties in evaluating political investments and that the ambiguity in measuring benefits and costs of CPA will not lead to higher performance (Hadani & Schuler, 2013). Another approach is the political marketplace theory, which predicts that firms compete with one another to gain access to legislators and political decision makers. As there are many firms competing over limited

(34)

resources (i.e. favorable legislations and policies), the performance outcomes are unlikely to be positive (Hadani & Schuler, 2013).

However, in the period between 2000-2014, out of 51 CPA studies, only one previous CPA study found an insignificant relationship (Mellahi et al., 2016) between nonmarket strategy and organizational performance. Given the amount of research that has found a significant relationship between CPA and CFP, it is not in line with the expectations that a firm that engages in CPA has neither a positive nor a negative effect on their financial performance (e.g. Hillman et al., 2004; Mellahi et al., 2016). Therefore, the results of the current study seem counterintuitive.

A possible explanation for the insignificant relationship between firm performance and CPA is that CPA can affect firm performance, but with a time lag (Hadani & Schuler, 2013; Shaffer et al., 2000). Like many nonmarket activities, building relationships with regulators and political decisionmakers takes time (Hadani & Schuler, 2013). It takes time to impact on the policy maker’s legislation through lobbying and PAC contributions, so this may not instantaneously affect financial performance. Large donations over several electoral cycles, combined with adequate information on legislations may lead to increased performance (Hadani & Schuler, 2013).

Besides that, a difference between the current study and other research is the time range of the dataset. Among other variables, data for lobbying expenditures and PAC contributions was in previous research taken over a different time period. For example, Hadani & Schuler (2013) used data ranging from 1998-2008, while the current study has a range of 2012-2015. This difference might impair the comparability of the researches and outcomes of the CPA-CFP relationship. As the political environment and legislations might have changed, this could affect the effectiveness of corporate political investments on financial performance.

(35)

Another difference between this study and other studies is the way financial performance was measured. This research used the accounting-based measure ‘return on assets’ as firm performance indicator, while other researches used a market-based measure such as Tobin’s q (e.g.Coates, 2012), or market-to-book (Faccio, 2010). For example, Faccio (2010) found that politically active companies have higher leverage and higher market shares, but underperform based on an accounting-based measure (Faccio, 2010). However, tests have been reperformed through a market-based (tobin’s q) measure, which also yielded non-significant results.

Furthermore, in contrast to previous studies (e.g. Hadani & Schuler, 2013), no moderating effect was found for industry regulations on the CPA-CFP relationship. A possible important difference between this research and previous research is the classification of the industry regulations variable. The industry regulations variable in the current study is based on Coates’ (2012) definition of which industries are regulated and which are not. In contrast, Hadani & Schuler (2013) relied on the classifications offered by Grier et al. (1994), which, in turn, are based on Pittman’s (1977) own judgement. These differences in classification could have led to different firms being classified as operating in regulated industries and vice versa. As discussed before, time-based differences between this study and previous research could affect the outcomes of the CPA-CFP relationship. In terms of industry regulations, it could mean that regulatory policies within industries have changed and that the degree of regulatory power over firms has changed over the years. This might affect the strength of ‘regulated industries’ on firms. If, for example, regulations have weakened over the period of 2012-2015, there would be less of a difference between regulated and nonregulated industries. In turn, this would lead to less of a moderating effect of industry regulations.

(36)

Besides the absence of both a direct effect of CPA on CFP and a moderating effect of industry regulations, no moderating effect was found for the CEO outcome-based incentives. A possible explanation for this finding is that linking a CEO’s compensation to shareholder returns will expose the CEO to more risk and might therefore refrain him/her from investing in high-risk CPA projects (Coles et al., 2006). In turn, this may reduce the profits that a firm could have potentially made investing in the CPA projects. The finding that there is no difference between regulated and nonregulated industries could mean that regulatory power in regulated industries might have decreased, leading to less of a difference between regulated and nonregulated firms.

Next to the outcome-based incentives, no moderation effect was found for the behavioral-based incentives. As the separation of CEO and chairman positions may increase board monitoring and controlling of the CEO, there are also disadvantages to the independent leadership structure (Boyd, 1995). These disadvantages include the sacrifice of leadership and decision speed (Boyd, 1995), which could in turn reduce the decision-making process and (effectiveness of) CPA investments done by CEOs. This may explain the non-significant moderating effect found in the CPA-CFP relationship.

Contributions of this study

Firstly, this study elaborated on the existing literature concerning the effect of corporate political activity (CPA) on corporate financial performance (CFP). In contrast to most previous research, this study found no evidence for the ability of corporate political activity to affect corporate financial performance.

Secondly, the research built further on previous research (e.g. Hadani & Schuler, 2013) by integrating agency theory with the theory of market and nonmarket strategies in order to understand performance effects of nonmarket strategy (Mellahi et al., 2016).

(37)

Thirdly, the results of this research contrast previous research concerning a moderating effect of industry regulations on the CPA-CFP relationship. No evidence was found for such an effect.

Fourthly, corporate governance theory was extended in the context of CPA by exploring if well-governed firms in nonregulated industries perform better. Again, no such effect was found.

Limitations and future research

As stated earlier, in the period between 2000-2014, out of 51 CPA studies, only one previous CPA study found an insignificant relationship between nonmarket strategy and organizational performance (Mellahi et al., 2016). The insignificant findings of this research contrast the overall evidence of the CPA-CFP relationship. This study has a number of potential limitations that could affect the research outcomes. Also, recommendations for future research are suggested.

First of, like other studies of CPA, this research is challenged by obligatory disclosure of CPA of firms. Even public firms only have to disclose political activities such as lobbying expenditures above a certain amount (Coates, 2012). This means that, potentially, important figures on e.g. lobbying expenditures are made without public disclosure. In turn, these figures are not taken into account in the CPA investments that could affect firm performance. Besides that, if a firm hires another firm (such as a law or public relations firm) to engage in political activity without the legal obligation to label the activities as ‘lobbying’, the investments will remain unknown for the public and the current study (Coates, 2012). Currently, lobbying laws are predominantly unenforced (Coates, 2012).

Secondly, related to the limited availability of data, the current study is constrained to a certain amount of years (2012-2015) for variables such as lobbying expenditures. A longitudinal study is necessary to determine whether the effects of Corporate Political

(38)

Activity on Corporate Financial Performance could be significant over a longer period of time. In contrast to previous research (e.g. Hadani & Schuler, 2013), fewer independent variables were tested to research the potential relationship between CPA and CFP. A suggestion for future research is to include variables such as contributions to ‘527 groups‘ (Hadani & Schuler, 2013).

Thirdly, it may be difficult to find a direct effect of CPA on firm performance (Hillman et al., 2004; Shaffer, 1995). As a firm engages in many market and nonmarket activities simultaneously in extremely complex environments, finding a direct relation is challenging (Shaffer, 1995). A suggestion for future research is to include more political tactics (e.g. constituency building) and more control variables (e.g. firm diversification) and to extend the literature concerning indirect effects of CPA on firm performance (Hillman et al., 2004). Also, more qualitative research concerning the nature of political relations between a firm and legislators may prove to find new insights concerning the CPA-CFP relationship (Hadani & Schuler, 2013).

Fourthly, another important limitation is the U.S. centric view, which might impair the external validity of the current study. As this research focuses only on firms in the United States, the business and political activity can be expected to differ from other institutional contexts (Hillman, 2005). For example, a Chinese context with a complete different political regime and heavier government influence can yield different results in political activities (Sun et al., 2016).

Fifthly, an important limitation of this research concerning the CEO variables is that there has been no criterion made concerning CEO succession. CEO succession is important as CEO change can lead to power shifts and might influence a firm’s processes and power structure (Daily & Johnson, 1997). Consequently, different investment decisions in political activity might be made or differences in political ties might exist between the old and new

(39)

CEO. A suggestion for future research is to include only firms that experience no CEO succession during the entire period of the dataset.

Conclusion

First, this research explored the relationship between Corporate Political Activity and Corporate Financial Performance. The empirical testing found no significant results for the CPA-CFP relationship. Furthermore, analyses were performed to test a potential moderating effect of industry regulations on the CPA-CFP relationship. The moderating effect of industry regulations for both the lobbying expenditures and PAC contributions on return on assets turned out to be non-significant. These results contrast previous research (e.g Hadani & Schuler, 2013) on the CPA-CFP relationship. Finally, moderation tests were performed to explore potential moderating effects of CEO outcome-based incentives and CEO behavior-based incentives on the CPA-CFP relationship. Again, no significant results were found, contradicting the formed hypotheses. Future research might extend on the CPA-CFP relationship discussion by testing different political tactics, integrating different theories and focusing on potential indirect effects of CPA on CFP.

(40)

Reference list

Aggarwal, R. K., Meschke, F., & Wang, T. Y. (2012). Corporate political donations: Investment or agency. Business and Politics, 14(1), 3.

Agrawal, A., & Knoeber, C. R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of financial and quantitative analysis,

31(03), 377-397.

Baron, D. P. (1995). Integrated strategy: Market and nonmarket components. California

management review, 37(2), 47-65.

Bebchuk, L. A., & Cohen, A. (2005). The costs of entrenched boards. Journal of Financial

Economics, 78(2), 409-433.

Bliss, M. A., & Gul, F. A. (2012). Political connection and cost of debt: Some Malaysian evidence. Journal of Banking & Finance, 36(5), 1520-1527.

Bonardi, J. P., Holburn, G. L., & Bergh, R. G. V. (2006). Nonmarket strategy performance: Evidence from US electric utilities. Academy of Management Journal, 49(6), 1209-1228. Boyd, B. K. (1995). CEO duality and firm performance: A contingency model. Strategic

Management Journal, 16(4), 301-312.

Calori, R., Johnson, G., & Sarnin, P. (1994). CEOs' cognitive maps and the scope of the organization. Strategic Management Journal, 15(6), 437-457.

Carpenter, M. A., & Sanders, W. M. (2002). Top management team compensation: The missing link between CEO pay and firm performance?. Strategic management journal, 23(4), 367-375.

Center for Responsive Politics (2017, 31 March). Retrieved from www.opensecrets.org Claessens, S., Feijen, E., & Laeven, L. (2008). Political connections and preferential access to finance: The role of campaign contributions. Journal of financial economics, 88(3), 554-580. Coates, J. C., (2012). Corporate politics, governance, and value before and after Citizens United. Journal of Empirical Legal Studies, 9(4), 657-696.

Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerial incentives and risk-taking.

Referenties

GERELATEERDE DOCUMENTEN

Coupled with the asymmetry of the mean horizontal velocity profiles in figure 6 , the resulting viscous boundary layer becomes thinner at the cold wall, and a larger C f ,loc

Given that current operational strategies rely heavily on the onsite skills and experiences of operators, who have implicitly learnt from their experience in previous projects, it

Schematic representation of the fabrication of micron-scale surface chemical gradients of the alkyne- functionalized thiol-sensitive probe 14 via electrochemically promoted CuAAC on

Maar daardoor weten ze vaak niet goed wat de software doet, kunnen deze niet wijzigen en ook niet voorspel- len hoe de software samenwerkt met andere auto-software. Laten we

13 H2a: The cultural variable power distance negatively influences the positive relationship between corporate social responsibility and corporate financial performance

The regression is estimated using ordinary least squares with fixed effects including the control variables size and risk (Altman Z-score when using ROA and MTB, volatility of

Lastly, it should be noted that in this paper I used the result from content analysis as the comprehensive evaluation index to measure the performance of corporate social

The conse- quence is two-fold: (i) a suitable selection method is required to select those components for which de- veloping prognostic methods is useful, and (ii) an approach