The Signaling Role of Politician Stock Ownership: Effects on Lobbying Intensity

Abstract

Adopting a signaling theory perspective, we argue that politician stock ownership sends signals of positive predispositions to firms, thereby alleviating some necessity for firms to emphasize lobbying expenditures to influence political action. Using data on congressional stock ownership, we find support for our arguments. We find that as the proportion of Congress owning stock in a firm increases, the firm decreases the intensity of lobbying. Furthermore, we find that the signals associated with stock-holding politicians with greater ability to affect the legislative agenda (i.e., affiliation with the majority party) relates to lobbying intensity. Our findings add to the literature on lobbying while also offering implications for practice and avenues for future research.

Keywords

lobbying signaling theory corporate political activity firm performance

Introduction

Investments increasingly put lawmakers in the position of voting or advocating on matters that could affect their personal wealth. (O’Harrow, Kindy, & Keating, 2009)

A growing body of research focuses on corporate political activity (CPA) (Lux, Crook, & Woehr, 2011; Oliver & Holzinger, 2008), defined as firm investments in political activities in an attempt to influence government officials to act in ways favorable to the firm (Getz, 2001; Hillman, 2003). CPA is so “well documented,” “pervasive,” and “ubiquitous” worldwide (Hillman, Keim, & Schuler, 2004: 837) that satirists argue politicians “should be compelled to wear uniforms like NASCAR drivers, so we can identify their corporate sponsors” (Baum, 2009). While comparing politicians to NASCAR drivers may be satirical, the analogy seems to have merit. In NASCAR sponsorships, firms provide resources to drivers in exchange for advertising and endorsements to influence customers. Similarly, a market exists where firms provide contributions to politicians’ election campaigns and engage in lobbying (the two primary forms of CPA; Lux et al., 2011) to sway the officials to act in ways that benefit the firm (Bonardi, Hillman, & Keim, 2005; Hillman & Hitt, 1999; Hillman & Keim, 1995; Shaffer, 1995).

This view of politics where “government favors can be bought” on the market (Shaffer, 1995: 495) suggests that firms are demanders of public policy while politicians and other governmental officials are suppliers (Hillman & Hitt, 1999; Keim & Zeithaml, 1986; Schuler, Rehbein, & Cramer, 2002). Yet if CPA operates similar to a market, then it is subject to the same informational asymmetries as other markets. For example, firms engage in CPA to influence politicians to act in a manner that is beneficial to the firm. However, the intensity of CPA in which firms must engage to influence politicians likely depends upon the firms’ knowledge about politicians’ probable actions. That is, if a firm is aware that a large number of politicians are positively predisposed towards the firm, the intensity in which the firm engages in CPA is likely to be lower as the firm will have less need to influence those politicians to act on the firm’s behalf. Therefore, in this article, we focus our attention on theorizing how politician stock ownership acts as a signal in the CPA market. Specifically, we argue that because politicians are both investors in firms as well as suppliers of policies that will affect those firms, the politicians’ stock ownership sends signals to firms that the politicians are positively predisposed to consider the firm when adopting policy positions, allowing firms to adjust their CPA accordingly.

We view politician stock ownership as a signal of positive predisposition from the vantage point of a firm for a number of reasons. First, politicians can utilize various means, such as stalling/pushing legislation, that influence firms directly (e.g., contracts, subsidies) as well as the environments (e.g., regulation, concentration) in which those firms operate (Gale & Buchholz, 1987; Hillman & Hitt, 1999), making politicians a particularly important stakeholder group for firms (Clarkson, 1995; Freeman, 1984). Second, politician stock ownership is easily identifiable. Politicians are required to make public disclosure of their stock portfolios, creating an observable signal as to the politicians’ likely predispositions toward the firms in which they own stock. Evidence suggests that information about stockholders is of utmost importance to firms, particularly powerful stockholders who have the ability to influence the firm (O’Sullivan, 2007). Lastly, and perhaps paramount, politician stockholding is growing and ties lawmakers’ personal wealth and self-interests to the fortunes of the firms in which they may either use legislative actions (e.g., advance or stall legislation) affecting those firms or act on inside information for financial gain (O’Harrow, Kindy, & Keating, 2009). Moreover, evidence suggests that politicians exhibit self-interested wealth-seeking behaviors in their positions, such as by voting differently when they have a direct interest in firms that will be affected by certain legislation and by utilizing inside information for personal gain (Schweizer, 2011; Tahoun & van Lent, 2011; Ziobrowski, Cheng, Boyd, & Ziobrowski, 2004).

Cumulatively, the aforementioned literature suggests that some politicians may utilize their positions either via legislative maneuvers or leveraging inside information to pursue financial gain (e.g., Buchanan & Tullock, 1965; Cohen, Diether, & Malloy, 2012; Downs, 1957; Keim & Zeithaml, 1986) and lends credence to the fact that firms are likely to view stock ownership as signaling positive politician predisposition toward them. Consistent with this view, we adopt a signaling theory framework to develop theory about how firms adjust their lobbying intensity in response to the signals of politician stock ownership. We focus explicitly on lobbying for two reasons. First, as Lux et al.’s (2011: 241) meta-analyses suggests, “lobbying appears to be the primary means of CPA with contributions [to political campaigns] acting more as a complementary activity.” Indeed, with respect to financial expenditures, evidence suggests that firms spend roughly 20 times more on lobbying than on campaign contributions (Hill, Kelly, Lockhart, & Ness, 2013; Milyo, Primo, & Groseclose, 2000). Focusing on lobbying, then, allows us to generate insight on the aspect of CPA that dominates firms’ political expenditures to better understand how available expenditures are allocated. Specifically, we focus on lobbying intensity, or the amount of lobbying expenditures relative to available resources, because from a strategic perspective, the decision to allocate $10 million towards lobbying for a $30 billion firm does not indicate the same emphasis placed on the political arena as the same $10 million lobbying expenditure from a $1 billion firm. Thus, focusing on the intensity of expenditures provides a better representation of lobbying intensity and strategic decision making than does a simple aggregate measure like the amount of lobbying, a view consistent with other strategic phenomena such as return on assets rather than pure returns and research and development (R&D) intensity rather than simply R&D expenditure (cf. Carpenter, 2000; Geletkanycz & Hambrick, 1997).

Second, focusing on lobbying answers calls for theory differentiating between the two primary means in which firms engage in attempts to influence politicians: campaign contributions and lobbying (cf. Bonardi, Holburn, & Bergh, 2006; Hillman et al., 2004; Lux et al., 2011; Vanden Bergh & Holburn, 2007). We expand upon Tahoun’s (2014) analyses of the relationship between politician stock investment and firms’ campaign contributions to articulate important differences between campaign contributions and lobbying that have implications for the relationship between politician stock investment and firms’ lobbying. Specifically, we first articulate how campaign contributions and lobbying differ on (a) the mechanisms that encourage politician actions (i.e., financial versus informational); (b) the nature of the firm’s investment (i.e., an investment in a quid pro quo financial exchange between a single firm and a single politician versus an investment in information dissemination for the purposes of influencing multiple politicians simultaneously); and (c) how politicians benefit from, and will respond to, changes in the CPA investment (i.e., politicians benefit from continued campaign contributions whereas politicians may not desire or even benefit from additional information through lobbying).

Based upon the differences between campaign contributions and lobbying, we then develop theoretical arguments that, as with campaign contributions, firms will not haphazardly invest (cf. Tahoun, 2014; Tripathi, 2000) in lobbying because lobbying does not “buy” any single politician (Shaffer, 1995). Since firms will not invest in lobbying haphazardly and because politicians generally will not view decreases in lobbying as a violation of a quid quo pro relationship as they do with firm’s campaign contributions, we contend that firms will not respond to signals that politicians are positively predisposed to them by increased lobbying intensity as they do with campaign contributions (Tahoun, 2014). Instead, because firms know that stockholding politicians are positively predisposed to them, they no longer have to lobby as intensively but, rather, need to do only a minimum amount to communicate needs (cf. Hillman & Hitt, 1999) and, thus, will decrease the intensity of their lobbying investment. Further, we develop arguments about the role played by the ability of signalers to fulfill the needs of the firm in eliciting a signal response.

Our study of how politicians’ stock ownership affects lobbying contributes to both theory and practice. With respect to theory, our study makes three principal contributions. First, we extend a growing line of research on antecedents to CPA and prior insights into CPA consisting of a market between firms and politicians (Hillman & Hitt, 1999; Hillman & Keim, 1995; Keim & Zeithaml, 1986; Schuler et al., 2002). Specifically, we answer calls to both focus on portions of the CPA market receiving less attention (politicians’ role), as well as to adopt the perspective of the firm to address factors that may determine a firm’s need to utilize CPA (Bonardi et al., 2005; Lux et al., 2011). Further, our development of arguments that differentiate lobbying and campaign contributions answers calls for theory distinguishing between the two primary means of CPA (cf. Bonardi et al., 2006; Hillman et al., 2004; Lux et al., 2011; Vanden Bergh & Holburn, 2007). Our findings not only highlight the reciprocal nature of exchange in the CPA market by addressing the signaling role that politicians play in influencing how firms engage in CPA (specifically lobbying) but also articulate important differences in campaign contributions and lobbying that can spawn future research.

Second, our study adds insights into research on nonmarket strategies or firms’ efforts to influence the environments in which they operate (e.g., Holburn & Vanden Bergh, 2013). In particular, prior research notes that CPA operates in a quid pro quo fashion where firms trade CPA for favors (Hillman et al., 2004; Keim & Zeithaml, 1986; Shaffer, 1995). Our research finds that there are instances where firms may have a lesser need to influence their environments via CPA owing to an alignment of interests between firms and policy makers that arise when policy makers also have financial interests in the firms over which they simultaneously have legislative authority. We also extend nonmarket and CPA research by answering calls to bring new theoretical vantage points to investigate nonmarket strategies (Hillman et al., 2004; Lux et al., 2011). Along these lines, a third contribution of our study is to develop theoretical arguments about the signals sent by politician stock ownership and the manner in which this signaling affects the market for public policy. In this sense, we extend signaling theory to the context of CPA, and specifically lobbying, and illuminate the role played by signaler ability to fulfill the needs of the receiver. We conclude the article with a discussion of possible public policy implications as well as future research that can advance upon the theoretical insights we offer.

Theory and Hypotheses

Signaling Theory in a Political Arena

Firms engage in CPA in an attempt to shape public policy in ways favorable to them (Hillman, 2003; Shaffer, 1995). Signaling theory is a useful lens to explain firms’ CPA behavior because the theory applies to behaviors of a party when in a position of information asymmetry (Connelly, Certo, Ireland, & Reutzel, 2011), as is the case for firms in the political arena since firms are not perfectly aware of the intentions of politicians on issues that affect firms, and it is illegal to enter into formal agreements with politicians (Kroszner & Stratmann, 1998; Milyo et al., 2000; Tahoun, 2014). Specifically, signaling theory stems from the premise that since individuals and organizations base decisions upon information, differences in information between the two parties (i.e., information asymmetry) means that a party could make a more informed and potentially better decision if given access to the asymmetric information held by the other party (Stiglitz, 2002). Given that firms are not fully aware of the intentions of politicians, signaling theory suggests that firms will attempt to reduce information asymmetry by searching for signals regarding politicians’ intentions so that firms can engage in more informed decision making regarding their CPA towards the politicians (Spence, 1973). In addition to the contention that the exchange of information affects decisions, signaling theory further posits that for a receiver to respond to a signal, the signal must be both observable—or noticed by the receiver—and credible—or viewed by the receiver as providing relevant information about the quality or intent of the signaler (Stiglitz, 2002). Thus, our view is that firms are receivers that search for observable and credible signals from politicians, who are signalers. Firms then may use the information in the signal in deciding how to act towards the political arena. Importantly, whether politicians signal intentionally or unintentionally is not material to whether information is sent through signals—rather, receivers will search for and receive both intentional and unintentional signals that provide information about signalers (Janney & Folta, 2003; Stiglitz, 2002).

An observable, credible signal of politicians’ intent towards firms is politicians’ stockholding. The number of politicians owning stock is growing (O’Harrow et al., 2009). As O’Harrow and colleagues (2009) note, in the United States House of Representatives, “the number of lawmakers trading stock jumped from 91 in 2001 to 259 today … [including] 68 lawmakers who, as of the beginning of 2008, individually owned more than $100,000 in stock”; such investments “increasingly put lawmakers in the position of voting or advocating on matters that could affect their personal wealth.” Credibility of this as a signal to a firm is afforded because owning stock aligns the economic interests of the firm with those of the stockholding politicians since when the firm does well (poorly) the politician’s personal wealth improves (decreases). Given the interest alignment between firms and stockholding politicians, it is perhaps not surprising that “legislators who have a direct interest in firms often vote quite differently than other, uninterested legislators on legislation that impacts the firms in question” (Cohen et al., 2012: 24). Indeed, extant work has focused on politicians’ use of their positions for self-interest to advance or stall legislation that affects their personal wealth or because they have inside information that they can use to their financial benefit (e.g., Buchanan & Tullock, 1965; Downs, 1957; Keim & Zeithaml, 1986; Schweizer, 2011; Tahoun & van Lent, 2011; Ziobrowski et al., 2004). In either case, if politicians use their positions to support their personal wealth or use inside information to gain financial payoffs, owning stock in a focal firm, and the inherent alignment of the politicians’ personal wealth with the firm’s performance, provides a credible signal about the intentions of stockholding politicians. This line of argument is consistent with others who argue that retaining equity sends signals about individuals’ interest in, or knowledge of, the future performance of their firms (e.g., Certo, 2003; Connelly, Hoskisson, Tihanyi, & Certo, 2010).

Stock ownership ties politicians’ wealth to firms over which they legislate, raising the opportunity for possible self-interested action. Some scholars find evidence of this possible action. For instance, Ziobrowski and colleagues (2004) find that portfolios that mimic U.S. senators’ stock purchases beat the market by 85 basis points per month, while a portfolio holding stocks sold by senators lags the market by 12 basis points per month. Similarly, Tahoun and van Lent (2011) discuss how U.S. congresspersons’ stock ownership affects legislation and oversight in a manner consistent with the politicians’ wealth interests. Specifically, they find a relationship between stock ownership of members of the U.S. House of Representatives and the probability of a financial firm receiving governmental funding though the Troubled Asset Relief Program. Moreover, Tahoun (2014) finds evidence of a quid pro quo relationship between U.S. congresspersons’ stock ownership and political action committee (PAC) contributions.

If politician stock ownership remained private, the ability of firms to observe the information would be difficult, limiting the degree to which stock ownership could provide an interpretable signal to firms. However, politicians in many countries are legally required to disclose these investments publicly. Thus, firms may easily obtain information on which politicians are stockholders and are therefore likely to have a positive predisposition to the firm. Further, O’Sullivan (2007) and the National Investor Relations Institute (NIRI) note that firms spend heavily to obtain information about their stockholders. Indeed, a 2013 study by the NIRI shows that 88% of investor relations officers regard investor identification as extremely or very important, while more than half suggest they are “very familiar (know company stock ownership at a granular level) with their investor base.” The heavy spending on obtaining information on stockholders coupled with the public availability of information on politicians’ stockholdings suggests that firms are very likely to identify politicians who are stockholders.

Because firms engage in CPA in an effort to persuade politicians to take positions that are beneficial to them (Getz, 2001; Hillman et al., 2004; Lux et al., 2011; Oliver & Holzinger, 2008), we expect that signals indicating that politicians will take positions that benefit firms will elicit changes in firms’ CPA. Importantly, however, is that politicians do not necessarily need to engage in self-interested actions, but the ownership of stock in the firm does provide a signal that they may do so from a signaling perspective for firms. That said, we expect that there will be differences in how firms respond to stock investment between the the two primary types of CPA: campaign contributions and lobbying. Specifically, we argue there are three important differences between campaign contributions and lobbying that will affect how firms respond: how the type of CPA induces politician actions, the nature of the firm-politician relationship created by the type of CPA, and how politicians benefit from the type of CPA.

With campaign contributions, as Tahoun (2014: 90) notes, because “the main objective of the politicians in establishing a relation with firms is to obtain their contributions and support,” such contributions create a direct financial inducement for politicians to act as the only way to secure such contributions is to act in the interests of the donating firms. As such, contributions create a direct, one firm to one politician relationship often portrayed “as the functional equivalent of bribes” (Milyo et al., 2000: 75) where firms donate money and politicians, in an effort to continue the flow of contributions into their campaign coffers, return favors to the donating firm such as “favorable legislation and tax exemptions, having preferential access to finance, getting government contracts, or receiving help dealing with regulatory agencies” (Tahoun, 2014: 89). Tahoun empirically demonstrates the presence of such a noncontractible (i.e., cannot be an explicit contract because such agreements are bribes and are typically illegal; Kroszner & Stratmann, 1998; Milyo et al., 2000) quid pro quo relationship between politicians and firms such that there is a positive relationship between firm contributions and politician stock investments: politicians tend to invest in firms after firms donate to their campaigns and firms tend to discontinue contributions to politicians after politicians divest ownership of firms’ stock. Tripathi (2000: 69) notes a similar quid pro quo “transaction” or “spot market” such that firms “resolve free rider problems” by discontinuing contributions to politicians who either did not provide benefits in return to the donating firms or who no longer possess the ability to do so. In sum, with campaign contributions, a donating firm creates a financial inducement for any politician receiving a contribution to act in the interest of the firm such that a direct firm-politician quid pro quo relationship is created in which the firm receives favors and the politician receives contributions.

With lobbying, firms do not use financial inducement as is the case with campaign contributions but rather utilize information to influence politicians’ actions (Coen, 1997; Nownes, 2006). The informational inducement works similar to advertising, with firms broadly targeting a multitude of politicians simultaneously through means such as advocacy campaigns, providing information on impacts of extant and/or potential legislation/regulation, and distributing policy papers. As lobbying broadly targets multiple politicians, this form of CPA does not create a direct link from a firm to any particular politician. Further, while politicians may benefit from information, there is not a quid pro quo relationship expectation from politicians that a firm will maintain lobbying and politicians can still make decisions without such information. Thus, unlike campaign contributions in which a decrease in contributions from a firm to a politician will be viewed as a violation of the inherent quid pro quo contract (Tahoun, 2014), a decrease in lobbying not only does not violate a one-to-one, quid pro quo contract, but may also be unlikely to be viewed adversely by politicians.

In light of the foregoing facts, we agree with Tahoun (2014: 107) that firms are incentivized to maintain campaign contributions with politicians who own their stock for fear of being viewed as “rengeging on their promises,” which “can lead to a breakdown in the politician-firm exchange market.” However, as the nature of lobbying does not create a quid pro quo relationship but instead targets a broad swath of polticians simultaneoulsy, if a firm receives signals that a number of politicians are likely to act on their behalf without needing to attempt to influence their decisions, the signal reduces the necessity of the firm to emphasize lobbying to obtain outcomes favorable to the firm. Rather than investing a wealth of resources to influence politicians, firms only need to engage in a minimal amount of lobbying to communicate needs; or, in other words, a firm can shift from an expensive relational information strategy to a less intensive, and cheaper, transactional information strategy (cf. Hillman & Hitt, 1999). From a signaling theory perspective, the signal of politician stock ownership helps to alleviate a portion of informational asymmetry in the CPA market, particularly as the percentage of politicians owning stock increases. Unlike campaign contributions, we expect the firm to respond to the signal that politicians are positively predisposed to the firm by engaging in less intensive lobbying, which is no longer necessary to influence politicians to act in the interests of the firm as, instead, the firm only needs to engage in a minimal level of lobbying to communicate needs. Interestingly, this argument could be viewed as complementary to Tahoun (2014). For example, if a firm and a politician create a quid pro quo relationship through campaign contributions and politician stock ownership as Tahoun finds, then there is even less need for firms to lobby these politicians, strengthening the signal because of the prior development of a quid pro quo relationship. Therefore, as more politicians signal their intent to the firm, the firm can have greater confidence that the legislative process will favor the firm. As a greater proportion of politicians own stock in the focal firm, the firm will have greater assurance that less emphasis on lobbying is necessary to influence the legislative process and thus respond by reducing the intensity of the firm’s lobbying.

Hypothesis 1: The proportion of politicians owning stock in a firm will negatively relate to firm lobbying intensity.

Signaler Ability to Fulfill Receiver Needs

In determining how to respond to a signal, receivers consider the “ability of the signaler to fulfill the needs or demands of an outsider observing the signal” (Connelly et al., 2011: 43). In the CPA context, firms may consider whether politicians are able to influence policy before reacting to signals received from them. That is, if politicians have little ability to influence the legislative agenda in a manner conducive to the firm, the firm is less likely to respond to the signal of those politicians’ stock ownership by reducing lobbying, such as is the case when firms quit donating to the campaigns of politicians who are no longer positioned to return benefits to the firms (cf. Tahoun, 2014; Tripathi, 2000). Consistent with this notion, we argue that a firm is more likely to respond to signals of politician stock ownership when politicians have the ability to influence the legislative agenda. In nontotalitarian governments, two mechanisms enhance politicians ability to influence legislation: (a) occupying ranking positions in legislative subgroups; and (b) membership in the political party that holds the majority of the legislative positions, or what we refer to as majority party (versus minority party, which holds fewer legislative positions).

First, with respect to ranking positions, legislative bodies typically divide duties into suborganizations (e.g., committees) that allow members to develop specialized knowledge of the area pertinent to the committee. Each committee appoints politicians to ranking positions to facilitate committee operation, which we refer to as “ranking members.” Ranking members “have more power than their colleagues to raise questions, hold hearings, and push through targeted legislation that in some cases governs the industries in which they have investments” (O’Harrow & Keating, 2010). Similarly, legislative bodies are typically divided into chambers (e.g., the Senate and the House of Representatives in the U.S. Congress; the House of Commons and House of Lords in the United Kingdom, or U.K., Parliament), each of which appoints individuals to ranking positions who, like ranking members in committees, possess additional power to control the legislative process. Indeed, politicians holding ranking positions are “in a better position of influencing the performance of their investments or at least appearing to have that ability” (O’Harrow & Keating, 2010). Thus, the ability of politicians in ranking positions to influence the legislative process suggests that when firms receive signals of intent sent by these individuals, they are more likely to respond to the signal than they would to signals from politicians with lesser ability to affect the legislative process in ways beneficial to the firm. As such, we expect that when a firm has a larger percentage of politicians in ranking positions holding stock in the firm, the firm will view the signal as indicating a higher likelihood that public policy will be shaped in ways beneficial to the firm, decreasing the firm’s need to engage in lobbying to influence the legislative process.

While we do not believe that firms will respond to this need by cutting contributions, which might be viewed as violating quid pro quo and lead politicians who lost contributions to use their power to hurt the firm (Tahoun, 2014), we do believe that firms will have a lower need to engage in lobbying to influence decisions. In particular, if influential politicians, such as those in ranking positions hold stock, the firm will not need to lobby as broadly or as intensely to gain influence but, rather, can scale back their lobbying overall and invest only enough to communicate needs to those with influence. Therefore, we argue that a greater proportion of politicians as stock owners who also hold ranking positions will have a negative relationship with firm lobbying intensity.

Hypothesis 2: The proportion of politicians owning stock in a firm who occupy a ranking position will negatively relate to firm lobbying intensity.

A second mechanism that enhances a politicians’ ability to influence the legislative agenda is membership in the majority party. Nontotalitarian political systems typically operate with a limited-party system consisting of individuals representing different ideas (e.g., the Republican and Democratic Parties in the U.S.; the Conservative and Labour Parties in the U.K.). At any point in time a single party may constitute the majority in the legislative body with respect to the ratio of legislative positions held and hence controls the legislative agenda. In the limited-party system, the majority party has substantial control over the legislative agenda, providing politicians in the majority party with greater ability to influence the legislation being pursued. The majority party controls the legislative agenda for the legislative body and as such is able to bring up (or delay) legislation as they deem fit. For example, the passage of both the Deficit Reduction Act of 2005 and the Patient Protection and Affordable Care Act in 2010 were passed through both chambers of the U.S. Congress and signed into law without a single favorable vote from politicians representing the minority party.

With the increased ability of the majority party to control the legislative process, firms are more likely to respond to signals of politician stock ownership stemming from politicians in the majority party. Thus, as a greater number of politicians in the majority party own stock in the firm, the necessity for firms to engage in CPA is likely to decrease. As with our previous hypothesis, we do not expect firms to respond to signals by reducing contributions for fear of retribution (Tahoun, 2014), but we do believe that they will respond by adjusting their lobbying accordingly. In particular, the firm will not need to lobby as intensely but, instead, can shift lobbying both away from the minority party and away from costly relational information strategies towards a narrowed, less expensive transaction lobbying approach such that the overall intensity of the firms lobbying efforts will be lower. As such, we hypothesize the following:

Hypothesis 3: The proportion of politicians in the majority party owning stock in a firm will negatively relate to firm lobbying intensity.

Methods

Sample

The study sample contains S&P 500 firms in 2005 where CPA and firm data were available from 2005 through 2010. Data for CPA and politician stock ownership were obtained from the Center for Responsive Politics and were linked to firm data from COMPUSTAT. After removing firms with missing data, the final sample included 2,369 firm-year observations from a total of 420 firms, averaging 5.6 observations per firm. To establish temporal precedence, the dependent variable was measured in the year following the independent and control variables.

Dependent Variables

Lobbying targets politicians broadly (Nownes, 2006), and as such, our measure of lobbying is at the level of aggregate politicians. Since we are primarily interested in the strategic emphasis that firms place on lobbying relative to available resources rather than simple aggregate expenditures, we measure lobbying intensity as the firm’s total lobbying expenditures per $1 million of market value in the year. The resulting measure gives us our cumulative measure of the intensity of lobbying in which each firm engages which is in line with the measurement of other important strategic variables such as R&D intensity and capital intensity as well as return on assets (ROA) or return on equity (ROE) that are scaled by appropriate denominators.

Independent Variables

The variable politician ownership was measured as the percentage of U.S. congresspersons that held stock in a firm each year. While U.S. congresspersons are required to report the value of any security held (e.g., stocks, bonds, mutual funds), we utilize only stocks held directly in a single company. The measure ranking member ownership was calculated as the percentage of stock owning politicians that held the positions of chairperson, vice chairperson, or ranking member in a committee or held a ranking position in the full U.S. Congress (e.g., Speaker, majority/minority leader). To create the measures of majority party politician ownership, we first determined the controlling political party of both chambers of the U.S. Congress (the House of Representatives and Senate) and the number of members in both the majority and minority parties. We then calculate the number of U.S. congresspersons affiliated with the majority party holding stock in the focal firm. Thus, majority party ownership was calculated as the number of U.S. congresspersons affiliated with the majority party who owned stock in a focal firm divided by the total number of U.S. congresspersons affiliated with the majority party.

Control Variables

We include several control variables for firm and political influences on the signaling relationship and/or the ability of the firm to engage in CPA, and specifically lobbying. First, firm size was measured as the logarithm of firm sales and was included because it has been shown to be a major driver of CPA (Lux et al., 2011). Because the performance of the firm may impact the level of CPA in which the firm is engaged, firm performance was also included and measured as ROA. Although our results are also robust to Tobin’s Q, we omit the latter from the tables for parsimony (Hadani & Schuler, 2013). Measures of the firm’s debt to equity ratio and capital intensity (capital expenditures divided by firm sales) were also included as indicators of firm strategic alternatives (Henderson & Fredrickson, 2001). We also include the measure of prior lobbying intensity because the level of lobbying in which a firm engages in a certain year is likely influenced by prior lobbying investments. Additionally, we include government contracts as a control variable to indicate the dependence each firm may have on government authority. This measure is calculated as the logarithm of the value of serviceable contracts in each year using data from usaspending.gov. Although politicians have various means with which to influence firms (e.g., enacting/not enacting legislation can affect regulated firms or unregulated firms), the ability to do so may vary by the degree to which an industry is regulated. Thus, we control for regulated industry, as a dichotomous variable for firms in more heavily regulated industries following the typology developed by Grier, Munger, and Roberts (1994) and utilized more recently by Hadani and Schuler (2013), and also include industry dummies in our analyses using two-digit standard industrial classification (SIC) codes. Further, we tested an interaction of the regulation dummy with politician ownership and find a nonsignificant interaction effect, indicating that our results are not moderated by industry regulation. We also control the amount of campaign contributions, measured as the average amount of PAC contributions per U.S. congressional candidate in a year. Because campaign contributions and politician stock ownership have been shown to be codetermined (e.g., Tahoun, 2014), we regress politician stock ownership on campaign contributions and then include the residuals of this regression as a control variable. In robustness tests, we omit campaign contributions, utilize alternative measures (including an instrumental variables measurement) and employ seemingly unrelated regressions and find results consistent with those we report here.

At the political level, we include controls for politicians’ investment values by political party. We did so for two reasons. First, a firm may consider the relative value of an investment to a politician in determining whether stock ownership aligns interests. Second, to the degree that firms hold preconceived notions about the policy leanings of an individual based upon party, party affiliation may affect the perceptions of interest alignment and signals being sent. Thus, we included a measure of the value of stock ownership relative to politicians’ net wealth calculated along political party lines. Politicians are required to provide good faith estimates of the range of value for each stock owned, total assets, and total liabilities, although some individuals report exact values. To determine the value of stock ownership, we used the exact value of investments where provided; otherwise we utilized the average value of the estimated range for each politicians’ investment. Politicians’ net wealth was determined by taking the maximum reported value of all assets owned in the ranges provided minus the minimum value of all liabilities in the ranges. We used the maximum value of assets and the minimum value of liabilities to provide a net worth that is the maximum possible, providing us with the most conservative estimate of the value of each investment relative to net wealth of politicians. Thus, Democrat value owned and Republican value owned were created as the value of politicians’ stock investment divided by member net wealth, averaged by party. Also, because stock may be sold relatively frequently, we included a control for the level of stability within politicians’ portfolios. Specifically, we created the measure ownership stability as the number of sales transactions (within the entire portfolio of politician owners) divided by the total number of assets held by a politician owner, averaged by firm. Similarly, we included a measure of the diversity of the type of investor who holds stock in the focal firm. The measure owner investment diversity was included as the coefficient of variation of the total number of investments owned by each politician owner.

Finally, to account for the possibility that firms in which politicians own stock systematically differ from those in which no stock is held, we estimated a first-stage model to create a treatment control (cf. Barrios, 2004; Guo & Fraser, 2010). To do so, we created a dummy variable with a value of 1 if at least one member of the U.S. Congress held stock in the firm and 0 otherwise. We then identified several instrumental variables that predict ownership and estimate a first-stage probit model (results appear in the appendix). Using the first-stage probit results, we calculated the inverse Mills ratio and used this value as a control variable in second-stage lobbying models.

Analysis

To ensure robust testing of our hypotheses, we employed two different analytical techniques, one on what we consider to be the full sample and a second on a subsample of firms engaging in lobbying. We use two samples because it may be argued that firms that either do not or have not engaged in lobbying are not in the market and thus should not be considered to react to signals from politicians. Thus, we utilized the subsample of firms that previously engaged in lobbying as an additional robustness test because our theory builds upon the premise of such investments taking place within a political exchange market. Within the full sample, the dependent variable is left-censored at zero because not all firms engage in lobbying in a given year (680 of our 2,369 firm-year observations had zero lobbying expenditures), and so we employed a Tobit specification to account for censoring (Greene, 2000). The time period that we utilize in hypothesis testing (i.e., 2005 to 2010) coincided with fluctuation in the S&P 500 index. As such events may give rise to contemporaneous correlation that may bias estimates, we employ time dummies, adding a binary indicator variable equal to 1 in a focal year and 0 otherwise, to account for events within a unique time period such as a yearly fluctuation in the S&P index (Beck & Katz, 1995; Certo & Semadeni, 2006). Within the full sample, we employed a random effects model as Tobit models for panel analysis are not defined for fixed effects (Greene, 2000; Souder, Simsek, & Johnson, 2012). Within the subsample of firms previously and currently engaged in lobbying, analysis was performed using the Arellano-Bond method (Arellano & Bond, 1991; Arellano & Bover, 1995), a variant of the standard fixed effects model. This analysis is appropriate because in the subsample the censoring of the dependent variable is not a concern. Arellano-Bond tests were performed for autocorrelation and Hansen tests for the validity of the instrumentation strategy. In all models, the autocorrelation and Hansen tests do not reject the null (p > .05), indicating that the Arellano-Bond estimator is asymptotically consistent and that the instruments are exogenous.

Results

Table 1 reports the means, standard deviations, and correlations for study variables, while Tables 2 and 3 provide hypotheses testing results. We first show our base model that includes only control variables (Table 2, Model 1; Table 3, Model 5). Hypothesis 1 posits that politician ownership will negatively relate to firm lobbying intensity. The results provide support for Hypothesis 1 with a negative coefficient (p < .01) on politician ownership in both the full sample (Table 2, Model 2) and subsample (Table 3, Model 6). Supporting our theory, politician ownership negatively relates to firm lobbying intensity, suggesting that such ownership signals a beneficial position regarding the firm and minimizing the need to emphasize lobbying.

Table 1

Descriptive Statistics and Correlation Matrix

		Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1	Lobbying intensity^a	85.07	165.2
2	Prior lobbying intensity^a	84.16	162.2	.55
3	Politician ownership^b	1.40	2.50	−.02	−.03
4	Ranking member ownership^b	9.60	16.90	.01	.06	.15
5	Majority party ownership^b	1.20	2.50	−.04	−.04	.96	.15
6	Government contracts^c	10.15	8.24	.14	.14	.21	.04	.20
7	Campaign contributions	1,581.87	1,826.91	.26	.30	.33	.11	.29	.27
8	Firm performance^b	6.78	7.81	−.14	−.17	.09	.00	.09	−.05	−.12
9	Inverse Mills ratio (λ)	0.47	0.35	−.11	−.11	−.53	−.26	−.46	−.21	−.39	.00
10	Democrat value owned^b	0.09	0.41	.01	.01	.03	.01	.05	−.01	.02	.00	−.03
11	Republican value owned^b	0.17	1.86	.03	.03	.01	−.01	.01	−.02	.04	−.01	−.04	.00
12	Ownership stability	0.68	1.44	.07	.11	.05	.29	.05	−.01	.08	.04	−.16	−.01	.05
13	Capital intensity	0.08	0.15	.02	.02	−.03	−.03	−.04	−.17	.00	−.06	.04	.02	.00	.00
14	Firm size^c	9.02	1.19	.15	.16	.58	.12	.53	.26	.44	−.08	−.61	.01	.00	.07	−.16
15	Owner investment diversity	0.46	0.44	.08	.10	.51	.32	.44	.15	.36	.02	−.62	.12	.00	.21	.00	.44
16	Debt to equity	0.72	9.18	.07	.05	−.04	.00	−.04	.02	.00	.02	.03	.00	.00	.03	.00	−.02	−.02
17	Regulated industry	0.34	0.47	.05	.07	−.06	.08	−.06	.01	.16	−.20	−.02	.04	−.02	−.01	.37	−.03	.03	.01

Note: Correlations over .04 are significant at the .05 level.

Multiplied by 1 million for scaling.

Percentage.

Logarithm.

Table 2

Tobit Analyses of Politician Stock Ownership on Firm Lobbying Intensity (Full Sample)

	Model 1	Model 2	Model 3	Model 4
Constant	−192.75*	−276.19***	−190.27*	−252.58**
	(78.92)	(82.27)	(79.09)	(80.58)
Inverse Mills ratio (λ)	−51.68**	−56.87***	−53.07**	−55.57***
	(16.20)	(16.20)	(16.26)	(16.18)
Prior lobbying intensity	0.29***	0.29***	0.29***	0.30***
	(0.05)	(0.05)	(0.05)	(0.05)
Campaign contributions	0.02***	0.02***	0.02***	0.02***
	(0.00)	(0.00)	(0.00)	(0.00)
Government contracts	0.77	0.91	0.76	0.86
	(0.72)	(0.71)	(0.72)	(0.71)
Regulated industry	8.73	0.35	9.63	2.415
	(29.31)	(29.03)	(29.37)	(28.93)
Democrat value owned	7.33	8.59	7.07	9.07
	(9.86)	(9.82)	(9.87)	(9.81)
Republican value owned	−0.51	−0.47	−0.55	−0.45
	(1.77)	(1.76)	(1.77)	(1.77)
Ownership stability	0.29	0.01	0.88	0.15
	(2.48)	(2.48)	(2.56)	(2.48)
Owner investment diversity	18.86	25.24*	20.47	21.21
	(11.45)	(11.57)	(11.58)	(11.44)
Capital intensity	34.67	53.35	34.43	48.47
	(44.63)	(44.51)	(44.68)	(44.34)
Firm size	21.07***	32.97***	20.97***	30.10***
	(6.17)	(7.15)	(6.18)	(6.86)
Firm performance	−1.44**	−1.23*	−1.44**	−1.30*
	(0.54)	(0.54)	(0.54)	(0.54)
Debt to equity	0.68	0.65	0.68	0.66
	(0.36)	(0.36)	(0.36)	(0.36)
Predictors
Politician ownership		−89.31**
		(26.92)
Ranking member ownership			−2.28
			(2.40)
Majority party ownership				−70.66**
				(23.79)
Wald χ²	701.71***	730.15***	702.02***	727.95***
Change in χ²		11.01***	0.90	8.82**

Note: n = 2,369 firm-year observations; standard errors in parentheses; industry and time dummies included.

p < .05.

p < .01.

***

p < .001.

Table 3

Generalized Method of Moments (GMM) Analyses of Politician Stock Ownership on Firm Lobbying Intensity (Subsample)

	Model 5	Model 6	Model 7	Model 8
Constant	−49.11	−241.93	−40.43	−196.88
	(93.33)	(129.61)	(138.92)	(109.16)
Inverse Mills ratio (λ)	−8.57	−14.39	−10.68	−14.61
	(14.58)	(15.02)	(23.91)	(14.89)
Prior lobbying intensity	0.47***	0.44***	0.47***	0.45***
	(0.07)	(0.07)	(0.07)	(0.07)
Campaign contributions	0.01*	0.01	0.01	0.01
	(0.00)	(0.00)	(0.01)	(0.003)
Government contracts	1.09	1.26	1.29	1.48
	(1.23)	(1.22)	(1.21)	(1.19)
Regulated industry	8.82	−3.32	12.85	−5.60
	(26.90)	(23.50)	(29.45)	(23.38)
Democrat value owned	−1.95	−0.12	−3.15	0.02
	(6.33)	(6.99)	(10.37)	(7.11)
Republican value owned	−0.35	−0.18	−0.38	−0.21
	(0.82)	(0.92)	(1.28)	(0.89)
Ownership stability	−0.53	−0.80	0.54	−0.50
	(1.49)	(1.51)	(2.86)	(1.47)
Owner investment diversity	−21.82*	−13.00	−19.00	−19.39
	(10.59)	(10.47)	(23.35)	(10.72)
Capital intensity	13.38	50.72	14.12	58.82
	(40.61)	(46.81)	(55.53)	(46.46)
Firm size	11.50	34.94*	10.79	29.51*
	(9.80)	(14.57)	(13.36)	(12.09)
Firm performance	−1.87*	−1.47	−1.93	−1.52*
	(0.86)	(0.76)	(1.01)	(0.76)
Debt to equity	0.61	0.50	0.60	0.53
	(0.70)	(0.69)	(0.80)	(0.69)
Predictors
Politician ownership		−139.10**
		(49.28)
Ranking member ownership			−3.99
			(5.36)
Majority party ownership				−99.66**
				(35.09)
Wald χ²	759.33***	592.46***	2355.09***	630.50***
Change in χ²		7.97**	0.55	8.07**
AR(2)	1.54	1.53	1.55	1.53
Hansen	357.64	356.09	360.80	357.34

Note: n = 1,689 firm-year observations; standard errors in parentheses; industry and time dummies included.

p < .05.

p < .01.

***

p < .001.

Hypothesis 2 posits that the percentage of politician owners holding ranking member positions will be negatively related to lobbying intensity. Results are not supportive, as the coefficient for ranking member ownership is negative but not significant in either the full sample (Table 2, Model 3) or subsample (Table 3, Model 7). Hypothesis 3 suggests that the proportion of the majority party that owns stock in the focal firm will negatively relate to firm lobbying intensity. Coefficients for majority party ownership are negative and significant (p < .01) in both full sample (Table 2, Model 4) and subsample (Table 3, Model 8), lending support to Hypothesis 3. In totality, the results from both the full sample and subsample support the theoretical arguments that politician stock ownership provides signals to firms which relate to reduced lobbying intensity (Hypothesis 1) and that signals from majority party ownership affects firm response (Hypothesis 3) although the argument for ranking member signals alone (Hypothesis 2) is not supported.

In a given year in our sample, 1.4% of the politicians in U.S. Congress hold stock in our sample firms, while the average firm that lobbies expends just under $2 million on this form of CPA ($1,995,147). This means that in a given firm-year in our sample, about 7 of the 535 members of the U.S. Congress own stock in any particular S&P 500 firm. S&P 500 firms are quite large (average market value in our sample of $23.453 billion); thus, the average lobbying intensity is a fraction of market value. Practically speaking, our results indicate that as politician ownership increases by one standard deviation, or 2.5% (i.e., from 1.4% to 3.9%), lobbying intensity decreases 20% on average (i.e., from .00008507 to .00006806), which represents a decrease in roughly $400,000 in lobbying expenditures for the average firm in our sample. Similarly, an increase in majority party ownership from 1.2% to 3.7% (i.e., the mean plus one standard deviation) corresponds to a roughly 15% decrease in lobbying intensity, which amounts to a reduction of $300,000 in lobbying expenditures for the average firm.

As noted in our theoretical exposition, we assume that stock ownership places lawmakers in a position to either act in ways that affect their personal wealth or to use inside information as a means to do so; in either case, the interests of firms and the politicians owning their stock are aligned because when the politician acts in a way that benefits their personal wealth, the fortunes of the firm also benefit. Given that politicians and firms interests are aligned when politicians own a focal firm’s stock, we argue that politician stockholding acts as a signal to the focal firm to then decrease efforts to sway politicians to act in ways beneficial to the firm. Although a wealth of prior research supports the contention that politicians gain personal financial benefits from either the advancing/stalling legislation or inside information (e.g., Buchanan & Tullock, 1965; Cohen et al., 2012; Downs, 1957; Keim & Zeithaml, 1986; Schweizer, 2011; Tahoun, 2014; Tahoun & van Lent, 2011; Ziobrowski et al., 2004), in Table 4 we test the assumption of interest alignment in the sample utilized in our study and in the context of stockholding. Specifically, in the models labeled Positive Outcome Assumption in Table 4, using politician ownership and all other control variables in time t and two measures of firm performance (ROA and approximate Tobin’s Q; Richard, Devinney, Yip, & Johnson, 2009) in time period t + 1 as the dependent variables, we find a significant and positive effect (p < .001), offering support for our assumption that politician stock ownership has a positive relationship with firm subsequent performance outcomes.

Table 4

Results of Robustness Testing

	Positive Outcome Assumption		Reverse Causality Model
	ROA	Tobin’s Q	Reverse Causality Model
Constant	21.70***	3.21***	−0.10**
	(3.54)	(0.62)	(0.03)
Total shareholder returns			0.00**
			(0.00)
Stock price	0.01	−0.00	0.00
	(0.01)	(0.00)	(0.00)
Inverse Mills ratio (λ)			−0.04***
			(0.01)
Campaign contributions	−0.00	−0.00	0.00
	(0.00)	(0.00)	(0.00)
Regulated industry	−2.38*	−0.35**	−0.00
	(1.17)	(0.13)	(0.01)
Democrat value owned	0.36	−0.02	0.01
	(0.27)	(0.03)	(0.00)
Republican value owned	−0.01	−0.01*	0.00
	(0.02)	(0.01)	(0.00)
Ownership stability	0.09	0.01	0.00
	(0.08)	(0.01)	(0.00)
Owner investment diversity	1.00	0.22***	0.01
	(0.53)	(0.05)	(0.01)
Capital investment	−2.74	−0.02	0.02
	(1.97)	(0.21)	(0.02)
Firm size	−1.95***	−0.31***	0.02***
	(0.40)	(0.07)	(0.00)
Firm performance	0.29***	0.63***	0.00
	(0.05)	(0.04)	(0.00)
Debt to equity	−0.03*	−0.00	−0.00
	(0.01)	(0.00)	(0.00)
Politician ownership	5.50***	0.78***	0.91***
	(1.55)	(0.19)	(0.01)
Lobbying intensity	−0.00	0.00	−0.00
	(0.00)	(0.00)	(0.00)
Wald χ²	176.54***	1,284.40***	21,175.81***

Note: Standard errors in parentheses.

p < .05.

p < .01.

***

p < .001.

Lastly, to attempt to rule out alternative explanations, we conduct an additional robustness test. Our hypotheses contend that a firm will adjust lobbying intensity in response to signals of intent to act in ways beneficial to the firm stemming from politicians holding stock in a firm. To the degree that politicians are concerned with suffering repercussions if their stockholdings give rise to appearances of impropriety, it is possible that the inverse of our arguments is true such that as a firm engages in more lobbying to influence politicians, politicians would decrease their stockholdings. If this were true, the negative relationships that we observe in our hypotheses tests may arise from reverse causality. We note four reasons why we are skeptical of reverse causality in our context. First, as Bonardi, Holbrun, and Bergh (2006: 1210) argue, “most voters remain rationally ignorant” about the activities of politicians “because of the costs of becoming fully informed,” which in turn provides “some scope to trade policies that deviate from the ‘public interest’” (Aranson, 1990) without fearing electoral repercussions. Additionally, voters typically side with the incumbent (Abramowitz, 1991; Cox & Katz, 1996; Gelman & King, 1990), suggesting that politicians may not fear electoral repercussions. Second, because of the length and complexity of legislation, small changes to legislation may provide large benefits to firms, yet not garner public attention. Similarly, legislative maneuvers such as stalling of legislation may positively affect a politician’s portfolio but go unnoticed (e.g., Schweizer, 2011). Third, politicians may act in a manner that is beneficial to both their stockholdings and public interest. For instance, a politician owning stock in a U.S. automotive company and also representing a district surrounding Detroit may take legislative actions that benefit automotive companies as well as the member’s personal wealth and that of constituents (cf. Cohen et al., 2012; Kindy & O’Harrow, 2010). Finally, and perhaps most importantly, lobbying is not tied directly to any single politician but, instead, disseminates information targeting multiple politicians, limiting the degree to which lobbying activities give rise to appearances of impropriety for any single politician. While we are skeptical that politicians decrease their stockholdings in a firm in response to lobbying, we nonetheless conducted analyses to investigate this possibility. Specifically, we directly tested reverse causality by using lobbying in time t as the independent variable and politician stock ownership in time t + 1 as the dependent variable. The results of this analysis are shown in the column labeled Reverse Causality Model in Table 4. While the overall model is significant (p < .001), suggesting that we are predicting a significant amount of variance in politician stockholding, we find a nonsignificant coefficient on lobbying intensity in the Reverse Causality Model, providing no evidence of reverse causality, as lobbying intensity is not a significant predictor of politician stock ownership. As a robustness test, we regressed just lobbying intensity on politician ownership and tried models either excluding extant or including alternative control variables (as well as alternative measures) to assess whether our results were distorted by including/excluding control variables (cf. Carlson & Wu, 2011; Spector & Brannick, 2011); and in each case, we find a nonsignificant effect of lobbying intensity on politician stockholding, offering additional support for our conclusion that the relationship is not reverse causal in nature. While we tried various models without finding a significant relationship for lobbying intensity predicting stockholding, which we cumulatively view as supporting our contention, it is nonetheless possible that alternative predictors were not identified, and future research may wish to address whether lobbying plays a role in determining stock ownership.

Discussion

This study sought to understand the signaling effect of politician stock ownership on firm lobbying intensity. Specifically, we argued that because politicians are investors in firms as well as suppliers of policy that affects those firms, their ownership in a firm will send signals that the politician will likely act in ways that will benefit the firm. As such, firms can utilize these signals to base subsequent lobbying decisions. We argue that key differences between campaign contributions and lobbying with respect to the mechanisms that encourage politician action, the nature of the form of CPA investment, and politicians’ likely response to changes in CPA investment have important implications for firms’ response to signals from politicians stockholding. Specifically, we build upon Tahoun’s (2014) work arguing that because relationships between politician stockholding and firms’ campaign contributions create a noncontractible quid pro quo relationship with each single politician who invests in the firm, firms will increase their campaign contributions in response to politician stockholding to maintain their half of the exchange relationship (i.e., where firms provide financial contributions to induce politicians to act on their behalf). In contrast, because lobbying uses informational rather than financial means and targets politicians broadly rather than creating a one-to-one quid pro quo relationship, there is not an expectation of continued engagement from a single politician as there is with campaign contributions. As such, signals of positive predispositions by politicians lower the necessity of the firm to engage in lobbying. Our findings support this assertion, indicating that as a greater proportion of politicians own stock in a firm, the lower the lobbying intensity of the firm’s CPA investments. Moreover, we find that a single standard deviation increase in politician stockholding corresponds to a 20% decrease in firm lobbying intensity. The practical implications of these findings, in light of the fact that politicians are increasingly investing in firms, is that firms are both experiencing performance benefits and, to the degree that lobbying is indicative of larger actions, taking note and responding to politician stockholding by reducing the intensity of the lobbying. Broadly, there have been a number of calls by politicians and practitioners alike to limit the influence of firms by limiting or prohibiting campaign contributions. Yet our findings suggest that if firms influencing politicians via donations is under question, so too should be politician stockholdings. As such, our findings build on prior research demonstrating that politicians are subject to self-interested biases associated with stock ownership such that as their personal interests are tied to a firm, they will likely support policies that are beneficial to the firm (e.g., Barley, 2007; Buchanan & Tullock, 1965; Cohen et al., 2012; Downs, 1957; Keim & Zeithaml, 1986; Tahoun & van Lent, 2011; Ziobrowski et al., 2004).

Beyond the general signals provided by politician stock ownership, we also sought to understand if firms responded differently to signals based upon the ability of the signaling politicians to influence legislation. We theorized that politicians owning stock who are also ranking members or who are affiliated with the majority party have greater ability to influence the political process, and thus that firms will respond to these signals by limiting the intensity of lobbying. Our findings partially offer support for these arguments. Specifically, while we do not find support for our arguments that a greater portion of politicians owning stock and holding ranking positions will negatively relate to the intensity of lobbying in which the firm engages, we do find support for our argument that the proportion of politicians in the majority party who own stock in a focal firm negatively relates to firm lobbying intensity. These results suggest that the ability of the signaler (here, politicians) to fulfill the actions associated with the signal affects how firms respond to the signal to the degree that the signals are from majority parties.

This study offers multiple contributions to theory and practice. From a theoretical perspective, a first contribution is that while prior research notes that there is a market for CPA between firms and politicians within a market (Hillman & Hitt, 1999; Hillman & Keim, 1995; Keim & Zeithaml, 1986; Schuler et al., 2002), this line of inquiry has paid little attention to the role that politicians play (Hillman et al., 2004; although Tahoun, 2014, stands as a notable exception) or to reasons that may affect the degree to which firms need to utilize the CPA market (Bonardi et al., 2005; Lux et al., 2011). Our findings address both issues.

With respect to the role that politicians play, prior research addresses that there is a market for public policy such that firms are demanders of public policies (who, as economic entities, desire the policies to be in their favor) and politicians are suppliers with the ability to meet demanders needs (Hillman & Hitt, 1999; Keim & Zeithaml, 1986; Schuler et al., 2002). While a related literature suggests that politicians may also be economically motivated because they use their political positions for gain (e.g., Buchanan & Tullock, 1965; Cohen et al., 2012; Downs, 1957), minimal attention has been given to the role of policy suppliers’ economic motivations in the policy market. We respond to calls to identify reasons that drive the degree to which firms, as demanders, utilize the CPA market by considering the economic motivations of politicians on the CPA market. Our findings suggest that demanders (i.e., firms) act upon suppliers’ (i.e., politicians) perceived economic motivations. Cumulatively, our findings imply that the CPA market consists of more than firms providing resources to politicians to “buy” favorable policies (Shaffer, 1995) but rather, that demanders do not wantonly purchase CPA and, instead, consider factors that may affect suppliers desire to provide the policies demanded, such as their economic motivations for doing so, in determining how to act on the CPA market.

Second, we extend prior research that views the CPA market as functioning on a quid pro quo basis where “firms give something valued by public officials in exchange for getting favorable policies” (Getz, 2001: 318) by indicating that much like a market, factors that affect the willingness of a demander (here, firms) to spend on a suppliers product (here, favorable actions) are at play. As we articulate above, there are important differences between campaign contributions and lobbying. As with campaign contributions, our results suggest that firms seek to strategically invest (cf. Tahoun, 2014; Tripathi, 2000), but in building upon and extending Tahoun’s (2014) campaign contribution work, our research suggests how these differences result in strategic investments in lobbying that are opposite of those in campaign contributions. Identifying these underlying differences contributes to understanding about how lobbying and campaign contributions function differently, answering calls for theorizing about how the two CPA types differ (cf. Bonardi et al., 2006; Hillman et al., 2004; Lux et al., 2011; Vanden Bergh & Holburn, 2007) while also posing important implications for the CPA market. Specifically, our findings suggest that investigations of the CPA market in which firms attempt to sway politicians to act in their favor should consider the underlying mechanisms of how campaign contributions and lobbying function. Thus, future research should consider the implications of differences between campaign contributions and lobbying can add to our understanding of the CPA market.

On a broader level, our use of a signaling lens to investigate CPA, specifically lobbying, contributes to the literature by answering calls to apply theory to nonmarket strategy research (Hillman et al., 2004; Lux et al., 2011). Specifically, we apply signaling theory to help explain antecedents of firms’ nonmarket strategies in a political arena. In doing so, we answer calls to gain a deeper understanding of lobbying because of the significance of lobbying in comparison with campaign contributions, as firms expend more resources on lobbying than campaign contributions. Relatedly, by articulating the role of politicians’ ability to fulfill the demands of the firm in eliciting a response to the signal, we add nuance to our understanding of the role of differences between politicians that may offer avenues for future nonmarket research.

A final, related theoretical contribution is extending signaling theory, which our research does in several ways. Beyond extending the theory to the context of nonmarket strategies, and specifically CPA in the form of lobbying, we also answer calls to explore the role that signaler intent plays in affecting receiver response and expand research adopting the vantage point of the receiver (Connelly et al., 2011). Our study also adds to the paucity of research addressing signals that may be sent unintentionally (Janney & Folta, 2003; Stiglitz, 2002). Collectively, our development of a signaling theory perspective within the exchange view of CPA, and specifically lobbying, provides new insights into the signals that politicians (or other government officials) may provide firms. Stock ownership signals may not be knowingly provided by politicians. Still, because politicians are subject to both the same self-interested biases as other agents (Buchanan & Tullock, 1965; Downs, 1957) and to a legal obligation to divulge investments, firms may consider stock ownership as a signal even when politicians have no knowledge or expectation of this possibility. Our findings that firms are likely to respond to this signal is particularly relevant because over the last several years, stock ownership by politicians has increased substantially (O’Harrow et al., 2009), meriting closer investigation to understand the degree to which it influences firms. Moreover, by bringing the possible influence of politician stock ownership to light, we provide an avenue for future research to consider how firms may adapt their strategies in response to the interests of governmental officials, particularly their stock ownership. That is, if firms are adjusting CPA in response to politician stock ownership, whether in the form of campaign contributions as in Tahoun’s (2014) work or in lobbying intensity as in this research, it stands to reason that the signal that politicians provide to firms may help explain strategic outcomes other than CPA. Thus, future research may be apt to investigate other ways in which firms respond to politician stockholding, such as strategy adjustments in anticipation of favorable actions by politicians on their behalf.

Beyond the contributions of our research to the academic literature, our findings have practical ramifications for public policy. Our findings suggest that the investments of politicians provide signals to firms that affect firm decisions. Government officials may want to consider the signals their portfolios send to firms and whether such signals are beneficial. That is, policy makers may want to revisit rules regarding whether stockholdings and the requirement that they publicly disclose such purchases are appropriate given the potential conflicts that may arise. For example, laws in some countries require that judges recuse themselves from cases with potential to affect their personal finances or the finances of those close to them and that certain public officials place investments in blind trusts to remain impervious to influences associated with the investments. In the U.S., for example, the Stop Trading on Congressional Knowledge Act of 2012 prohibits U.S. congresspersons from using information obtained in their positions for personal benefit (such as with stock trading). However, critics of the act note that while the legislation “is a good first step in deterring these abusive practices,” it does not “go far enough to protect the American people from members of Congress who chose to act with self-interest over public good” (Weissmann, 2012; see also Clark & Embree, 2014). Difficulties associated with identifying, and proving, that politicians engaged in trading on private information as well as both loopholes in the specific act and difficulties in legislating such actions, generally, may still enable politicians to profit off of their position (cf. Clark & Embree, 2014; Cohen et al., 2012; Schweizer, 2011). Given difficulties in identifying, proving, and legislating politicians’ trading on private information and in light of our findings, requirements such as prohibiting investments and/or placing investments in blind trust may be worthy of consideration for members of U.S. Congress as well as for politicians in other settings where stockholding is permitted. Similarly, firms may want to consider whether they are leveraging the information provided by the composition of their stockholders. We find that firms adjust their emphasis on lobbying based upon having politicians as stockholders. Yet other stockholders may also be providing signals, and firms may want to consider what these signals mean and whether they are actionable, including extending analyses to regional government officials (e.g., Holburn & Vanden Bergh, 2013).

Relatedly, our test of the assumption that firms can expect to benefit from stockholding politicians and the finding that our Positive Outcome Assumption is empirically supported offers further evidence that politicians may not be solely serving the public good but also pursuing their self-interest in the form of financial gain associated with their stock portfolios. Moreover, while we add to a wealth of research with this finding of self-interested politicians, a new caveat from our research is that firms may consider this in strategic decisions. That is, our findings suggest that firms know that politicians are positively predisposed to the firm when they hold stock and, thus, will invest more strategically by reducing lobbying intensity. As such, not only do our results suggest that there may be social costs from politician stockholding to the degree that politicians pursue self-interest rather than public interest stemming from their investments, but, since firms are considering this possibility, our results bring to light questions as to the dealings between firms and the politicians that directly influence firms (e.g., contracts, subsidies) and their environments (e.g., regulations, concentration). From a practical standpoint, then, firms may wish to either “chum the water,” so to speak, by doing things to entice politicians to invest in their firm or, in contrast, work to level the playing field by working to prohibit such favoritism by firms. Certainly, the benefits (costs) incurred by firms from either “chumming the water” or working to level the playing field may differ based upon a number of contingencies, and some firms (like some politicians) may not like limits being placed on politicians’ stockholdings. More broadly, however, there are a number of societal implications from firms’ influencing government officials. For example, Barley (2007) notes many ways in which firms influence government is detrimental to public good, such as policy benefitting firms (or a subset of firms) at the expense of people or the environment and co-optation of the oversight process. We add to discussions about potential ills of firms influencing politicians and, specifically, suggest societal stakeholders should be cognizant of the implications politician stockholding has for both the competitive environments of firms and public good.

As with any study, our findings must be considered concurrently with possible limitations. In particular, this study builds upon the assumption that politician ownership is an observable and credible signal of intent to act in ways beneficial towards firms. While this view is consistent with empirical evidence (e.g., Buchanan & Tullock, 1965; Cohen et al., 2012; Downs, 1957; Getz, 2001; Tahoun & van Lent, 2011; Ziobrowski et al., 2004) and research noting that owning stock signals intent to take action to increase the value of that stock (e.g., Certo, 2003; Connelly et al., 2010), it may be that some politicians are more resistant to self-serving biases than others or that actions that appear self-interested are actually in the best interest of constituents. In turn, when considering legislative actions, stock ownership may have a differential impact on certain politicians, and future research would benefit from investigating the prior behavior of politicians to ascertain whether certain members hold a bias towards their stock ownership and the firms in which they invest as well as the signals emitted. Moreover, as prior work argues, there are other possible reasons that lobbyists target certain politicians beyond a rational economic view that we adopt in this study. Specifically, in this study we adopted a bounded rational view of information asymmetry, signaling, and market exchange to gain further understanding of firm impacts and lobbying. Yet as Hall and Deardorff (2006) argue, lobbying can occur because there is a match of similar policy agendas between lobbyists and politicians. Thus, although lobbying occurs in a market, there are multiple reasons that firms might lobby specific politicians, beyond the signal of stock ownership. Therefore, future research might not only view the signaling exchange view but also integrate alternative motivations, and particularly the matches between firms and politicians, when investigating lobbying by firms. Hence, while our study investigates avenues through which politicians may have a differential ability to impact legislation, future research could also add to our understanding in this realm. Such inquiries, despite obstacles to measurement and numerous complexities with respect to the nuances of lobbying, would not only provide additional insight into signaling in the political arena but could also add to our understanding of CPA. What is more, financial expenditures on lobbying and campaign contributions are not the only ways firms can influence politicians—for example, having current or former politicians as employees or board members are others (e.g., Hadani & Schuler, 2013; Hillman, Zardkoohi, & Bierman, 1999)—and future research could incorporate such relationships into lobbying to better understand the firm-politician interface. Finally, although lobbying and politician stockholding occur in countries across the globe (Choi, Jia, & Lu, 2014; Djankov, Porta, Lopez-de-Silanes, & Shleifer, 2009; Hillman et al., 2004), our investigation occurs within the U.S. context. We expect that the arguments and findings of our research will apply in other countries where lobbying and stock ownership by legislators occur (cf. Coen, 1997; Hillman et al., 1999). Nonetheless, idiosyncratic factors to any single context (here, the U.S.) may affect the relationships of interest in this research and, as such, future investigations can address the degree to which contextual factors are at work.

In conclusion, this study investigates the signaling properties of politician stock ownership in the market exchange view of CPA. To our knowledge, this study is the first to apply this signaling approach to provide evidence that politician stock ownership impacts the intensity of lobbying in which firms engage, building upon Tahoun’s (2014) work investigating quid pro quo relationships that campaign contributions create. Further, the ability of the politician owning stock (through majority party allegiance) to enhance the signal provides additional insight into the signaling process in the market exchange approach. This study brings to light the necessity to look at both sides of the market exchange (i.e., firms and politicians) when considering CPA as well as a need to discern how the two types of CPA, campaign contributions and lobbying, function.

Footnotes

Appendix

First Stage Regression Results

	Treatment Model
Constant	−0.09
	(0.05)
Firm size _{t − 1}	0.02***
	(0.00)
Politician sales stability _{t − 1}	−0.24***
	(0.05)
Politician purchase stability _{t − 1}	0.84***
	(0.08)
Owner average wealth _{t − 1}	0.01***
	(0.00)
Firm performance _{t − 1}	−0.30
	(0.35)
Election cycle year	−0.22***
	(0.06)
χ²	929.62***

Note: Standard errors in parentheses.

***

p < .001.

Acknowledgements

This article was accepted under the editorship of Patrick M. Wright.

References

Abramowitz

A. I.

1991. Incumbency, campaign spending, and the decline of competition in US House elections. Journal of Politics, 53(1): 34-56.

Aranson

P. H.

1990. Theories of economic regulation: From clarity to confusion. Journal of Law and Politics, 6: 247-286.

Arellano

Bond

1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58: 277-297.

Arellano

Bover

1995. Another look at the instrumental variable estimation of error-components models. Journal of Econometrics, 68(1): 29-51.

Barley

S. R.

2007. Corporations, democracy and the public good. Journal of Management Inquiry, 16(3): 201-215.

Barrios

2004. Generalized sample selection bias correction under RUM. Economics Letters, 85: 129-132.

Baum

2009, March 26. Congress should wear t-shirts with endorsements. Bloomberg.

Beck

Katz

J. N.

1995. What to do (and not to do) with time-series cross-section data. American Political Science, 89(3): 634-647.

Bonardi

J.-P.

Hillman

A. J.

Keim

G. D.

2005. The attractiveness of political markets: Implications for firm strategy. Academy of Management Review, 30(2): 397-413.

10.

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.

11.

Buchanan

J. M.

Tullock

1965. The calculus of consent: Logical foundations of constitutional democracy. Ann Arbor: University of Michigan Press.

12.

Carlson

2011. The illusion of statistical control: Control variable practice in management research. Organizational Research Methods, 15(3): 413-435.

13.

Carpenter

J. N.

2000. Does option compensation increase managerial risk appetite? Journal of Finance, 55(5): 2311-2331.

14.

Certo

S. T.

2003. Influencing initial public offering investors with prestige: Signaling with board structures. Academy of Management Review, 28(3): 432-446.

15.

Certo

S. T.

Semadeni

2006. Strategy research and panel data: Evidence and implications. Journal of Management, 32(3): 449-471.

16.

Choi

S.-J.

Jia

2014. The structure of political institutions and effectiveness of corporate political lobbying. Organization Science, 26(1): 158-179.

17.

Clark

Embree

2014. Faux transparency. In Ala’l

Vaught

(Eds.), Research handbook on transparency: 312-336. Northampton, MA: Elgar.

18.

Clarkson

M. E.

1995. A stakeholder framework for analyzing and evaluating corporate social performance. Academy of Management Review, 20(1): 92-117.

19.

Coen

1997. The evolution of the large firm as a political actor in the European Union. Journal of European Public Policy, 4(1): 91-108.

20.

Cohen

Diether

K. B.

Malloy

2012. Legislating stock prices (Harvard Business School Working Paper No. 13-010). Cambridge, MA: Harvard Business School.

21.

Connelly

B. L.

Certo

S. T.

Ireland

R. D.

Reutzel

C. R.

2011. Signaling theory: A review and assessment. Journal of Management, 37(1): 39-67.

22.

Connelly

B. L.

Hoskisson

R. E.

Tihanyi

Certo

S. T.

2010. Ownership as a form of corporate governance. Journal of Management Studies, 47(8): 1561-1589.

23.

Cox

G. W.

Katz

J. N.

1996. Why did the incumbency advantage in US House elections grow? American Journal of Political Science, 40: 478-497.

24.

Djankov

Porta

R. L.

Lopez-de-Silanes

Shleifer

2009. Disclosure by politicians. Cambridge, MA: National Bureau of Economic Research.

25.

Downs

1957. An economic theory of democracy. New York: Harper.

26.

Freeman

R. E.

1984. Srategic management: A stakeholder approach. Marshall, MA/Pitman.

27.

Gale

Buchholz

1987. The political pursuit of competitive advantage: What business can gain from government. In Marcus

Kaufman

Beam

(Eds.), Business strategy and public policy: 231-252. New York: Quorum.

28.

Geletkanycz

Hambrick

1997. The external ties of top executives: Implications for strategic choice and performance. Administrative Science Quarterly, 42(4): 654-681.

29.

Gelman

King

1990. Estimating incumbency advantage without bias. American Journal of Political Science, 34: 1142-1164.

30.

Getz

2001. Public affairs and political strategy. Journal of Public Affairs, 1: 305-329.

31.

Greene

2000. Econometric analysis (4th ed.). Upper Saddle River, NJ: Prentice Hall.

32.

Grier

K. B.

Munger

M. C.

Roberts

B. E.

1994. The determinants of industry political activity, 1978–1986. American Political Science Review, 88(4): 911-926.

33.

Guo

Fraser

M. W.

2010. Propensity score analysis: Statistical methods and applications. Thousand Oaks, CA: SAGE Publications.

34.

Hadani

Schuler

D. A.

2013. In search of El Dorado: The elusive financial returns on corporate political investments. Strategic Management Journal, 34(2): 165-181.

35.

Hall

R. L.

Deardorff

A. V.

2006. Lobbying as legislative subsidy. American Political Science Review, 100(1): 69-84.

36.

Henderson

Fredrickson

2001. Top management team coordination needs and the CEO pay gap: A competitive test of economic and behavioral views. Academy of Management Journal, 44(1): 96-117.

37.

Hill

M. D.

Kelly

G. W.

Lockhart

G. B.

Ness

R. A.

2013. Determinants and effects of corporate lobbying. Financial Management, 42(4): 931-957.

38.

Hillman

2003. Determinants of political strategies in US multinationals. Business & Society, 42: 455-484.

39.

Hillman

Hitt

M. A.

1999. Corporate political strategy formulation: A model of approach, participation, and strategy decisions. Academy of Management Review, 24: 825-842.

40.

Hillman

Keim

1995. International variation in the business-government interface: Institutional and organizational considerations. Academy of Management Review, 20(1): 193-214.

41.

Hillman

Keim

G. D.

Schuler

2004. Corporate political activity: A review and research agenda. Journal of Management, 30(6): 837-857.

42.

Hillman

A. J.

Zardkoohi

Bierman

1999. Corporate political strategies and firm performance: Indications of firm specific benefits from personal service in the US government. Strategic Management Journal, 20(1): 67-81.

43.

Holburn

G. L. F.

Vanden Bergh

R. G.

2013. Integrated market and nonmarket strategies: Political campaign contributions around merger and acquisition events in the energy sector. Strategic Management Journal, 35: 450-460.

44.

Janney

J. J.

Folta

T. B.

2003. Signaling through private equity placements and its impact on the valuation of biotechnology firms. Journal of Business Venturing, 18(3): 361-380.

45.

Keim

G. D.

Zeithaml

C. P.

1986. Corporate political strategy and legislative decision making: A review and contingency approach. Academy of Management Review, 11: 828-843.

46.

Kindy

O’Harrow

R. E.

Jr.

2010, March 28. Dingells and GM illustrate limits of congressional conflict-of-interest rules. Washington Post.

47.

Kroszner

R. S.

Stratmann

1998. Interest-group competition and the organization of Congress: Theory and evidence from financial services’ political action committees. American Economic Review, 88: 1163-1187.

48.

Lux

Crook

T. R.

Woehr

D. J.

2011. Mixing business with politics: A meta-analysis of the antecedents and outcomes of corporate political activity. Journal of Management, 37(1): 223-247.

49.

Milyo

Primo

Groseclose

2000. Corporate PAC campaign contributions in perspective. Business and Politics, 2(1): 75-88.

50.

National Investor Relations Institute. 2013. The annual report. Alexandria, VA: Author.

51.

Nownes

A. J.

2006. Total lobbying: What lobbyists want (and how they try to get it). New York: Cambridge University Press.

52.

O’Harrow

Keating

2010, June 14. When public duties and private investments intersect. The Washington Post.

53.

O’Harrow

Kindy

Keating

2009, November 23. Rise in stock ownership among lawmakers brings ethics concerns. The Washington Post.

54.

Oliver

Holzinger

2008. The effectiveness of strategic political management: A dynamic capabilities framework. Academy of Management Review, 33(2): 496-520.

55.

O’Sullivan

2007. Who owns your stock? Retrieved from CFO.com

56.

Richard

P. J.

Devinney

T. M.

Yip

G. S.

Johnson

2009. Measuring organizational performance: Towards methodological best practice. Journal of Management, 35: 718-804.

57.

Schuler

D. A.

Rehbein

Cramer

R. D.

2002. Pursuing strategic advantage through political means: A multivariate approach. Academy of Management Journal, 45(4): 659-672.

58.

Schweizer

2011. Throw them all out. New York: Houghton Mifflin Harcourt Trade.

59.

Shaffer

1995. Firm-level responses to government regulation: Theoretical and research approaches. Journal of Management, 21(3): 495-514.

60.

Souder

Simsek

Johnson

S. G.

2012. The differing effects of agent and founder CEOs on the firm’s market expansion. Strategic Management Journal, 33(1): 23-41.

61.

Spector

Brannick

2011. Methodological urban legends: The misuse of statistical control variables. Organizational Research Methods, 14(2): 287-305.

62.

Spence

1973. Job market signaling. Quarterly Journal of Economics, 87(3): 355-374.

63.

Stiglitz

J. E.

2002. Information and the change in the paradigm in economics. American Economic Review, 92(3): 460-501.

64.

Tahoun

2014. The role of stock ownership by US members of Congress on the market for political favors. Journal of Financial Economics, 111(1): 86-110.

65.

Tahoun

van Lent

2011. Personal wealth interest of politicians and government intervention in the economy. Unpublished manuscript.

66.

Tripathi

2000. PAC contributions and defense contracting. Business and Politics, 2(1): 53-73.

67.

Vanden Bergh

R. G.

Holburn

G. L. F.

2007. Targeting corporate political strategy: Theory and evidence from the US accounting industry. Business and Politics, 9(2): 1-31.

68.

Weissmann

2012. On the passage of the STOCK Act. Retrieved from http://ericweissmann.com/_blog/

69.

Ziobrowski

A. J.

Cheng

Boyd

J. W.

Ziobrowski

B. J.

2004. Abnormal returns from the common stock investments of the US Senate. Journal of Financial and Quantitative Analysis, 39(4): 661-676.