Determinants and Effects of Corporate Lobbying

• DOI: http://doi.org/10.1111/fima.12032
• Published date: 01 December 2013
• Author: Robert A. Ness,G. Brandon Lockhart,Matthew D. Hill,G. Wayne Kelly
• Date: 01 December 2013

Determinants and Effects of Corporate

Lobbying

Matthew D. Hill, G. Wayne Kelly, G. Brandon Lockhart,

and Robert A. Van Ness∗

We examine the determinants and value effects of corporate lobbying, controlling for corporate

political action committee (PAC)campaign contributions. We find evidence that firms with greater

potential payoffs from favorable policy and regulations lobby most actively, and that managers

often utilize both lobbying and campaign contribution channels to influence the political cli-

mate affecting the firm. We also find that shareholders value the lobbying activities pursued by

management on their behalf, particularly if the firm does not have a PAC that contributed to

an election campaign. The results are robust to a number of tests designed to mitigate potential

omitted-variable and self-selection bias.

The question of whether political influence matters for business has been debated extensively

in the economics and political science literature. The potential payoffs of political influence

are often very large. Hence, it is not surprising that firms will seek to influence political and

regulatory outcomes (Stigler, 1971; Grossman and Helpman, 1994), typically either through

direct campaign contributions or indirectly through lobbying. However, it is less obvious which

channel will dominate the other for firms with different types of characteristics.

The two channels work in different ways. Firms are not allowed to make direct contributions

to political campaigns from the corporate treasury. Instead, they can form political action com-

mittees (PACs) to which firm directors, employees, and their families can contribute and can,

in turn, support candidates for elections up to a maximum of $5,000 per candidate per elec-

tion. In contrast, lobbying expenditures are not limited and can be funded from the corporate

treasury.1

Recent research has typically focused on the first channel. Cooper, Gulen, and Ovtchinnikov

(2010) for example, analyze corporate PAC contributions to election candidates and find that the

number of candidates supported by a corporate PAC is positively linked with the firm’s future

earnings and stock returns. We extend this line of research by analyzing the lobbying activities

Wethank Ken Frenchfor providing portfolio breakpoints and returns data via his website at Dartmouth University,Raghu

Rau (editor), two anonymous referees,and Adam Yore (discussant, 2010 FMA Annual Meeting), for their comments. This

researchwas conducted while Lockhart was an Assistant Professor of Finance at the University of Nebraska-Lincoln.All

errors remain sole responsibilityof the authors.

∗Matthew D. Hill is an Assistant Professor of Finance and the J. Ed Turner Chair of Real Estate at the University

of Mississippi in University, MS. G. Wayne Kelly is an Associate Professor of Finance at the University of Southern

Mississippi in Hattiesburg, MS. G. Brandon Lockhart is an Assistant Professor of Finance at Clemson University in

Clemson, SC. Robert A. Van Ness is a Professor of Finance and Tom B. Scott Chair of Financial Institutions at the

University of Mississippi in University, MS.

1In a third channel, investigated by Faccio (2006), firms can also seek to influence policy when a large shareholder,

officer, or board member of a firm is a politician. However, US senate rules against potential conflicts of interest limit

the usefulness of studying this channel in our US context.

Financial Management •Winter 2013 •pages 931 - 957

932 Financial Management rWinter 2013

of US public firms. Specifically, we ask two research questions. First, which firm characteristics

predict a firm’s level of lobbying activity and, second, do shareholders capitalize this political

activity into the firm’s share price?

We collect lobbying expenditure and corporate PAC campaign contribution data for Center

for Research in Security Prices (CRSP)-Compustat firms from 1998 to 2011. We find evidence

that firms use both channels of influence to engage politically. For example, we find that 31.0%

(17.0%) of sample firms lobby (contribute), and that 14.9% pursue both strategies at some

point during the sample period. We also find evidence that firms allocate significantly greater

resources to lobbying activities relative to (highly regulated) PAC contributions (Milyo, Primo,

Groseclose, 2000). For the average firm in our sample, conditional on lobbying, expenditures

are $1.01 million, while conditioning on contributing, corporate PAC contributions are on the

order of $0.120 million. In a multivariate setting, we find that lobbying is positively related to

size, investment opportunities, and industry concentration, while negatively related to cash flow,

financial leverage, the number of elected politicians likely representing the firm’s interests in

Washington,and the distance in miles between the f irm’s physical headquarters and the respective

state capital building.

We estimate regressions based on three valuation models to learn whether shareholders value

the lobbying activities pursued by management over our sample period, controlling for their

attempts to influence the political process directly via corporate PAC campaign contributions.

We find strong evidence that shareholders value the lobbying activities pursued by management

even after controlling for PAC contributions. The interaction between lobbying and campaign

contributions is often negative in our value estimations, suggesting that shareholders place less

value on the lobbying activitiesof f irms that directly contributed to politicians prior to an election.

The magnitudes of our value estimates indicate greater value to corporate PAC contributions

relative to lobbying,consistent with an incremental value of lobbying.2Our conclusions regarding

the value implications of corporate lobbying survive severalrobustness tests for omitted variables

and selection.

Our research is related to that of Cooper et al. (2010), Faccio (2006), Faccio and Parsley

(2009), Fisman (2001), and Goldman, Rocholl, and So (2009), who examine the value of po-

litical connections. To our knowledge, we are the first to analyze the value of lobbying activ-

ities controlling for the corporate PAC campaign contributions that Cooper et al. (2010) show

are valued by shareholders.3Our results are particularly interesting given the implications of

the US Supreme Court decision on Citizens United vs. Federal Election Commission, and be-

cause previous research has connected lobbying to managerial-shareholder agency problems

(Hochberg, Sapienza, and Vissing-Jorgensen, 2009; Yu and Yu, 2011). The Citizens United

2It is true that lobbying and contributions occur throughout the calendar year. However, lobbying a particular politician

occurs after their election, by definition. Shareholders likely know whetherthe f irm’sPACcontributed to campaigns prior

to the lobbying activity.This timing predicts a negative sign on the interaction of lobbying and contributions. Lobbying

will be valued most by shareholders of firms not contributing, and the incremental value of lobbying will be lower for

firms that contribute.

3Kim (2008) finds mixed results when examining the market value impact of lobby expenditures. Inferences from the

study are limited by the focus on S&P 500 firms in election years. Further, the implications are weakened by potential

econometric issues such as omitted variables bias due to a parsimonious model, as well as sample selection bias. Kim

(2008) notes that the model represents a “preliminary step” in understanding the returns to lobbying. Chen, Parsley,

and Yang (2012) study issues related to firm performance and lobbying. In a working paper, Borisov, Goldman, and

Gupta (2012) use the Jack Abramoff scandal as a natural experiment to estimate the effect of corrupt lobbying on market

performance.