Corporate Lobbying and Firm Performance

• Date: 01 April 2015
• Author: David Parsley,Ya‐Wen Yang,Hui Chen
• DOI: http://doi.org/10.1111/jbfa.12109
• Published date: 01 April 2015

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 42(3) & (4), 444–481, April/May 2015, 0306-686X

doi: 10.1111/jbfa.12109

Corporate Lobbying and Firm

Performance

HUI CHEN,DAVID PARSLEY AND YA-WEN YANG∗

Abstract: Corporate lobbying activities are designed to influence legislators, regulators and

courts, presumably to encourage favorable policies and/or outcomes. In dollar terms, corporate

lobbying expenditures are typically one or even two orders of magnitude larger than spending

by Political Action Committees (PAC), and, unlike PAC donations, lobbying amounts are direct

corporate expenditures. We use data made available by the Lobbying Disclosure Act of 1995

to examine this more pervasive form of corporate political activity. We find that, on average,

lobbying is positively related to accounting and market measures of financial performance.

These results are robust across a number of empirical specifications. We also report market

performance evidence using a portfolio approach. We find that portfolios of firms with the

highest lobbying intensities significantly outperform their benchmarks in the three years

following portfolio formation.

Keywords: corporate lobbying, accounting performance, market returns, portfolio, Citizens

United

1. INTRODUCTION

Lobbying refers to the political activities that special interests, including corporations,

are engaged in to influence legislators at various levels of the government. It has

a long history in the United States and is protected by the Constitution as a basic

right of ‘freedom of speech’. At the federal level, lobbying is defined as ‘any

communication made on behalf of a client to members of Congress, congressional

staffers, the President, White House staff and high-level employees of nearly 200

agencies, regarding the formulation, modification, or adoption of legislation’ (The

Center for Public Integrity).1It is regulated by the Lobbying Disclosure Act of 1995.

∗The first author is from the Department of Business Administration, University of Zurich. The second

author is from the Owen Graduate School of Management, Vanderbilt University.The third author is from

the School of Business, Wake Forest University. The authors wouldlike to thank Paul Chaney, Mara Faccio,

Alexei Ovtchinnikov, Steve Rock, Jacob Sagi, Phil Shane, Shang-Jin Wei, and seminar participants at the

University of Colorado, Colorado State University, Nanjing University Business School, Melbourne Business

School, Peking University,the University of New South Wales, the University of Queensland Business School,

Vanderbilt University,and the 2014 Financial Management Association annual meetings, for comments. We

also thank the Center for Responsive Politics (CRP) for providing the data and many discussions concerning

the methodology, and Christoph Schenzler for assistance in matching the lobbying data to the financial

information provided in the CRSP and Compustat databases. Parsley acknowledges financial support from

The Financial Markets Research Center at Vanderbilt University. (Paper received November 2013, revised

version accepted March 2014)

Address for correspondence: Hui Chen, Department of Business Administration, University of Zurich,

Plattenstrasse 14, Zurich, 8032 Switzerland. e-mail: hui.chen@business.uzh.ch

1 Reported by the Center for Public Integrity.

C

2015 John Wiley & Sons Ltd 444

CORPORATE LOBBYING AND FIRM PERFORMANCE 445

Lobbying can provide information to the government from the various interest

groups and individuals who are likely to be affected by current and pending govern-

mental actions and legislation. Purely as communication, such information can help

the government make more informed decisions. On the other hand, lobbying is often

characterized as an instrument of these same interests to co-opt the process of policy

formation and direct benefits toward lobbying interests. For example, companies

involved in lawsuits, controversies, or scandals such as Enron, WorldCom, Phillip

Morris, and Halliburton, spent heavily on lobbying.

The impact of corporate political activity, including lobbying, is a hotly debated

issue in the United States, especially since the 2010 Supreme Court Citizens United

ruling, which relaxed constraints on corporate (and other groups’) spending on

elections. Early studies of corporate political activity (e.g., Snyder, 1992) focused on

whether political contributions affect legislative voting outcomes. Snyder concludes

that ‘despite years of research by political scientists and economists, the extent to

which money actually buys political influence on a regular basis remains a mystery’.

More recently, studies have considered outcomes from the perspective of the firm,

rather than political outcomes, in areas such as tax rates (Richter et al., 2009),

regulatory oversight (Bonardi et al., 2006; and Lux et al., 2011), earmarks2(de

Figueiredo and Silverman, 2006), tariffs (Mayda et al., 2010), and government

contracts (Goldman et al., 2013).3

The goal of this study is to extend the analysis to overall corporate financial

performance, and to focus the analysis on corporate lobbying spending as opposed

to a more typical focus on contributions by Political Action Committees (PACs).

Specifically, we attempt to estimate causal effects of corporate political activity on

firms’ subsequent financial performance. By studying all firms for which we can obtain

financial data, including those that are not politically active, we take advantage of the

data’s panel structure, thereby allowing us to control for all time invariant firm-specific

omitted variables when comparing firms that lobby with those that do not. Nichols

(2007) and Angrist and Pischke (2009) advocate this as one way to address the bias

due to unobserved confounders, common in studies relying on observational data. As

a second identification strategy we use a matching technique to study the stock market

performance of portfolios of lobbying and matched firms.

As noted, most studies of corporate political activity examine spending by corporate

affiliated PACs. Milyo et al. (2000) explain the ‘inordinate attention to PAC contribu-

tions’ by noting that ‘data on contributions are readily available and PACs are easily

linked to their corporate or industry sponsors’. Brasher and Lowery (2006) caution

that such research is too ‘narrow’, and suggest that empirical findings concerning PAC

behavior may not be generalizable; their study then argues that corporate lobbying

is the more appropriate focus. The key distinction, according to Ansolabehere et al.

(2003), is that since PAC dollars are actually contributions by individuals, and not

corporations, PAC spending should not even be considered ‘corporate’. In contrast,

2 Congressional earmarks in the US refer to designated federal funds that are spent on specified projects.

They differ from the regular congressional budget appropriation that grants each federal agency a certain

amount of money for a broad purpose. Earmarks are funds that must be directed to a specific project by an

agency.

3 Other recent studies of corporate political activity include: Agrawal and Knoeber (2001), de Figueiredo

and Tiller (2006), Fisman (2001), Ansolabehere et al. (2004), Hillman et al. (2004), Fan et al. (2007), Kim

(2008), Alexander et al. (2009), Goldman et al. (2009), Igan et al. (2009), Cooper et al. (2010), Yu and Yu

(2011), and Hill et al. (2013).

C

2015 John Wiley & Sons Ltd

446 CHEN, PARSLEY AND YANG

lobbying is a corporate expense – presumably made with a view toward the firm’s

bottom line. Given these critiques, this study focuses instead on the impact of

corporate lobbying on subsequent corporate financial performance.

Additional motivation for this study is provided by recent research into the financial

implications of corporate connections with politicians. Generally, these studies con-

clude that connected firms receive specific benefits from political connections, and

importantly, that the value of these connections is priced in financial markets.4Recent

studies find, for example, that political connections lead to better access to finance,

and/or lower taxation, government bailouts, higher market returns, more government

contracts, and greater market share, e.g., Faccio (2006), Faccio et al. (2006), Leuz and

Oberholzer-Gee (2006), Claessens et al. (2008) and Goldman et al. (2013).

Econometrically, we rely on the assumption that the impact of time-dependent

omitted variables is minimal. This suggests that given the severity of the global

financial crisis, the likelihood that firms made unique, one-off spending decisions,

cannot be discounted. For this reason we primarily focus on lobbying during the

pre-crisis years 1998–2005.5We hand match lobbying data to financial data from

Compustat, and to stock market returns and security pricing data from the Center

for Research in Security Prices (CRSP) for as many firms as have data available.

We begin with the approach taken by researchers studying the value of corporate

Research and Development (R&D) expenditures (e.g., Sougiannis, 1994; and Lev and

Sougiannis, 1996). We examine three accounting measures of performance released

in firms’ financial statements: income before extraordinary items, net income, and

operating cash flows. The evidence we present points to a robust, positive relationship

between corporate lobbying expenditures and firm financial performance, though

the results are weaker when focusing on cash flows from operations. In robustness

exercises, we attempt to provide a more nuanced view by considering non-linearity in

the relationship, and principal agent interpretations via sub-sample regressions (e.g.,

limiting to large lobbying spenders only, and to firms with weak investor protection

only).

In Section 3(ii), we abandon the panel approach in favor of a portfolio approach

focusing on market returns. Specifically, we compare excess returns of portfolios

of matched lobbying and non-lobbying firms. We follow the approach used by

Chan et al. (2001) who study the stock market valuation of R&D expenditures; hence

both of our empirical approaches derive from studies of R&D. One important benefit

of this portfolio-based approach is that it mitigates concerns about reverse causality

and endogeneity, since the focus of the analysis is on future (i.e., one-, two-, and

three-years ahead) excess market returns. Forecasts of future market performance

at these horizons are highly unreliable, thus making endogeneity bias and reverse

causality arguments less credible. We find that lobbying is positively correlated

with future excess returns. In particular, firms with the highest lobbying intensities

significantly outperform their benchmarks. Our results also suggest that most lobbying

expenditures are not associated with abnormal returns, and that simply spending the

most on lobbying does not necessarily lead to better stock market returns. However,

4 These studies do not conclude that firms with political connections necessarily enjoy superior financial

performance (e.g., Fan et al., 2007).

5 Following the suggestion of a referee, we do examine one important out-of-sample event. Specifically, in

Section 3(iii) we ask whether lobbying activity or market performance of firms engaged in lobbying activities

changed relative to other firms in the time period following the US Supreme Court Citizens United decision.

C

2015 John Wiley & Sons Ltd