Manhattan Institute for Policy Research.
Subscribe   Subscribe   MI on Facebook Find us on Twitter Find us on Instagram      


Legal Policy Report

No. 15 June 2012

Corporate Political Spending: Why the New Critics Are Wrong

Robert J. Shapiro and Douglas Dowson


Center for Legal Policy.



Supreme Court clamping down on unions' political fundraising, Alana Semuels, Los Angeles Times, 06-22-12
Why Don't Investors Get To Opt Out Of Corporate Political Contributions?, Dan Fisher,, 06-22-12
Political Spending Pays, Editorial, Wall Street Journal, 06-19-12

James Copland interviews Robert Shapiro on his new paper, "Corporate Political Spending: Why the New Critics Are Wrong."

Table of Contents:
Executive Summary
About the Authors
Background: Theories of Why Corporations Participate in Political Activity
The Economic Value of Corporate Campaign Contributions
The Economic Value of Corporate Lobbying
The Economic Value of Corporate Political Connections
The New Revisionists: Claims that Corporate Political Activities Harm Companies and their Shareholders

Executive Summary

Since the Supreme Court’s 2010 Citizens United decision held that corporate political expenditures are free speech under the First Amendment, various groups and individuals have advocated imposing new limits on corporate political activity. These efforts include calls on shareholders to demand that corporations refrain from involvement in the political process. Such demands have been buttressed by an emergent academic literature which, in contrast to what had been an established perspective, has questioned whether corporate financial contributions and even lobbying are actually in the interest of corporate shareholders. This paper reviews this new literature, contrasts it with previous work on this subject, and determines that the new studies ultimately fail to establish that corporate political activity adversely affects shareholder returns.

Corporate political activities take a variety of forms, including direct campaign contributions, joining and supporting trade associations, lobbying, the hiring of former public officials, advertising to move public opinion, and grassroots advocacy promotions. Lobbying has long been the dominant form for political participation by corporations and other interests: In the 2010 election cycle, for example, firms and other interests spent $6.8 billion on lobbying, compared with PAC expenditures of $1.3 billion.

The dominant academic view for the last 20 years is that companies undertake political activity to secure advantages for themselves, based on a combination of opportunity and necessity. Their incentives to do so are clear, given that modern governments influence national economies in ways that affect the sales and returns of particular industries and companies.

There is a robust academic literature, both theoretical and empirical, on campaign contributions to candidates, especially those provided through corporate political action committees (PACs). The most common explanation for these PAC contributions is that they help corporations and other interests secure greater access to legislators and other public officials. Empirical studies also have shown that corporate PAC activities are positively related to a corporation’s size, concentration, level of regulation, and sales to the government. This research clearly suggests that cost-benefit considerations influence corporate decisions to form and use PACs.

The academic research on lobbying has stressed the role that corporate lobbying plays in providing information to legislators and other public officials, or, in one variation, providing political intelligence and connections. Further, various recent studies have shown that lobbying generates positive economic returns. A 2009 study published in the American Journal of Political Science, for example, found that for the average firm lobbying Congress, across industries and various measures of financial performance, each additional $1 spent on lobbying was associated with $6-to-$20 in new tax benefits. Similarly, a 2010 working paper by Hui Chen and two co-authors found that $1 spent on lobbying was associated with an additional $24–to-$44 in corporate income. The same study further found that those firms which lobbied most intensively—defined as lobbying expenditures as a share of assets, sales, and market capitalization—outperformed their benchmarks by 5.5 percent to 6.7 percent per year, for the three years following their intense lobbying.

Other studies have demonstrated the value of corporate political connections, measured through PAC contributions, and found additional, positive contributions from corporate political activity. These studies include, for example, event studies examining the effects on the value of politically-active firms of the sudden death of Senator Henry “Scoop” Jackson (D-WA) in 1983 and the surprise defection of Senator James Jeffords (R-VT) from the Republican Party in 2001. Further, a 2010 study published in The Journal of Finance suggests that the economic benefits of a corporation’s political connections are also evident over the long term. The authors reported that a one-standard-deviation increase in the number of candidates supported by a firm over the previous five years was associated with excess abnormal returns of 2.6 percent.

Extensive analysis and evidence, then, support the view that corporate participation in the political process yields generally positive returns for firms and their shareholders.

Three recent studies have challenged this broad consensus, arguing that, on balance, political activity by corporations harms shareholder value. The authors of these studies claim that firms that engage in political activities will perform worse than their peers because they waste firm resources to advance their executives’ personal interests and because they generally are more likely to have “poor corporate governance:”

  • A recently published study by Rajesh Aggarwal from the University of Minnesota and two colleagues focuses exclusively on corporate soft-money contributions to political parties and donations to 527 organizations unregulated by campaign finance laws, and finds that the corporations most likely to make these contributions underperform their peers.
  • The other two studies, from John Coates of Harvard Law School, argue that in many cases, shareholders are harmed by all forms of corporate political activity, including lobbying and PAC contributions.
    • In a 2010 article, Coates observes that S&P 500 firms with poor corporate governance as measured by an index of corporate governance indicators are more likely to be politically active. This study then purports to show that their corporate political activity harms their shareholders, using a series of regressions to allegedly test for a direct relationship between shareholder value and corporate political activity.
    • In a subsequent 2012 study, Coates upgraded his methodology from that employed in his 2010 study and accordingly modified his earlier claims and conclusions. He continues to argue that much corporate political activity is not shareholder-oriented and that the net effect of corporate political activity is negative for shareholders of companies that are neither highly regulated nor highly dependent on sales to government.

A close examination of the three new studies shows that their reasoning and findings do not actually challenge, much less refute, the academic consensus that corporate political activity benefits shareholders or, at a minimum, does not harm them:

  • The Aggarwal et al. Study. Although the authors infer that their soft-money and 527 contributions explain the underperformance of companies engaged in such activity, their results support, at most, an inference that those companies’ underperformance may be related to factors that also influence their decisions to contribute soft money and 527 funds.
  • The 2010 Coates Study. Coates purports to show that the inverse correlation between PAC activity and his preferred corporate governance index shows that such activity is a function of poor corporate governance. But his data show that the correlation reverses when one looks at PAC-contribution levels: as his own measure of corporate governance improves, the average level of PAC contributions increases. Further, Coates’s regressions purporting to show a negative relationship between firm performance and corporate PAC activity and lobbying are poorly designed and subject to selection and omitted-variable bias. Notably, Coates does not even include as variables in his regressions many of the firm- and industry-level characteristics associated with corporate political activity, mistakes that would be sufficient basis for any study’s rejection by any peer-reviewed journal.
  • The 2012 Coates Study. In Coates’ 2012 study, after addressing some of the methodological problems in his 2010 study, he finds a positive relationship between political activity and firm value for regulated industries, which by his definition encompass roughly one-third of GDP (including alcohol, tobacco, aircraft, pharmaceuticals, utilities, telecommunications, transportation, banking, and insurance). Further, when he controls for relevant variables such as firm size and industry, his correlations for unregulated industries largely disappear. The only correlation that remains statistically significant is a negative relationship between the decision to lobby and firm value for unregulated firms, and the coefficient for that relationship is so low that the actual effect could very well be zero.

Summary of Findings

The relationship between political activity and firm performance or shareholder value is varied and complex, but the body of research in this area has established several important findings:

  1. Firms employ a variety of strategies to influence the political process in ways that may, or should, improve their performance and benefit their shareholders.
  2. Corporate spending decisions on campaign contributions and lobbying efforts are generally made in a rational and strategic manner.
  3. This political spending does not appear to systematically affect congressional voting, but it does regularly influence policymaking.
  4. Corporate political activity appears to have a generally positive effect on firm value, as reflected in excess market returns.
  5. The precise mechanisms that produce these positive effects remain unclear.

About the Authors

Robert J. Shapiro is chairman of Sonecon, LLC, a private firm that advises U.S. and foreign businesses, governments, and nonprofit organizations. Shapiro and Sonecon have advised, among others, President Bill Clinton, Vice President Albert Gore, Jr., British prime ministers Tony Blair and Gordon Brown, and U.S. Senators Hillary Clinton and Barack Obama. Sonecon also has advised private firms including AT&T, Amgen, Gilead Sciences, Google, MCI, Nordstjernan of Sweden, and Fujitsu of Japan; and nonprofit organizations including the U.S. Chamber of Commerce and the American Petroleum Institute. Shapiro is an adviser to the International Monetary Fund, a senior fellow of the Georgetown University McDonough School of Business, chair of the NDN Globalization Initiative, and chair of the U.S. Climate Task Force. From 1997 to 2001, he was Under Secretary of Commerce for Economic Affairs; in that post, he oversaw the Census Bureau and the Bureau of Economic Analysis and directed economic policy for the U.S. Commerce Department. Prior to that, Shapiro was cofounder and vice president of the Progressive Policy Institute. He also served as principal economic advisor to Bill Clinton in his 1991–92 presidential campaign and as a senior economic advisor to Albert Gore, Jr. and John Kerry in their presidential efforts. In 2008, he advised the campaign and transition of Barack Obama. Prior to cofounding PPI, Shapiro was legislative director and economic counsel to Senator Daniel Patrick Moynihan.He has been a fellow of Harvard University, the Brookings Institution, and the National Bureau of Economic Research. He holds a Ph.D. and an M.A. from Harvard University, an M.Sc. from the London School of Economics and Political Science, and an A.B. from the University of Chicago.

Douglas Dowson is senior economic analyst at Sonecon, LLC. Prior to joining Sonecon, he was research assistant at the Peterson Institute for International Economics, where his work focused on macroeconomic and international financial issues including global imbalances, exchange rate policy, and international trade and investment. Mr. Dowson holds a B.A. in economics and political science from the University of California at Berkeley and a master’s of international affairs in international economic policy from Columbia University’s School of International and Public Affairs. He has served as a Rosenthal Fellow in the U.S. Treasury’s Office of International Affairs and participated in the European Union Visitors Program in Brussels.

I. Introduction

The impact and significance of corporations engaging actively in the political process have been matters of debate and controversy throughout much of American history. The legitimacy of political activity by noncommercial entities of every type is rarely questioned, whether they are universities; foundations and museums; religious, minority, and civic groups; or simple assemblages of citizens with strongly held views about issues of the day. But when corporations similarly contribute to PACs, lobby, and otherwise engage in activities meant to affect an election or a debate in Congress, some observers see grave dangers. In response, Congress has applied various limits at various times to the campaign activities of corporations. Unsurprisingly, concerns in some quarters about corporate political activities intensified in the wake of the Citizens United decision by the Supreme Court.

Since that decision, some advocates of imposing new limits on corporate political activity have called on shareholders to demand that corporations withdraw from politics. Some of these efforts draw on new academic analyses that claim to show that corporate political contributions and lobbying actually damage the shareholders. These new analyses are avowedly contrarian, as they contradict the findings of scores of scholarly studies that, over several decades, have found that the political activities of corporations generally serve the interests of their shareholders. This study evaluates these contradictory findings. We have carefully reviewed both the new analyses and the corpus of previous theoretical and empirical work in this area. We find that the reasoning and methodologies of new analyses are flawed, while the vast bulk of other work in this area has been generally sound. Whatever the social costs or merits of corporate political activity, and however it might be regulated under the Constitution, we find that such activities generally benefit shareholders.

This conclusion is consistent with one of the basic tenets of American political culture and the social science that describes it: individuals, voluntary organizations, corporations, and other institutions that contribute money and time to political campaigns, petition their representatives, and find other ways of conveying their views do so because they expect to benefit in some way or because they are expressing deeply held views or notions. A large body of research and analysis in political science, economics, and sociology supports this view. Evaluating the effectiveness or success of these efforts, however, is more complex. For example, corporate PAC contributions aimed at electing particular candidates inevitably disappoint those who back the losers. But even losing campaigns can sometimes influence policy outcomes. Similarly, many policy processes provide opportunities for partial successes, as when lobbying efforts to oppose a new law or regulation succeed in tempering its provisions. In this sense, the political environment in which most organizations and individuals act provides a wide range of opportunities to influence policy outcomes in ways that advance their interests.

This view of private political activity as broadly rational has been studied, tested, and analyzed with particular intensity with regard to corporations. The interest of economists, political scientists, and other academics in the political activities of corporations is unsurprising. To begin, the concerns of a corporation or an industry about a particular election, law, or regulation often can be stipulated with precision, and data on corporate spending for elections and lobbying efforts are publicly available. In addition, there is a long tradition of scholarly inquiry into the political influence of various groups, especially those such as corporations with substantial resources. As we will show, the broad consensus of this research conducted over several decades is that corporations engage in politics because it generally serves their own economic interests and therefore those of their shareholders.

Some popular accounts characterize these efforts by business as sinister and corrupt, with powerful companies using their resources to “buy” votes on major pieces of legislation or regulation.[1] In this view, campaign contributions and other forms of financial support are exchanged for favors in a quid pro quo that benefits influential politicians and moneyed interests at the expense of the public.[2] One new variant of this stark view adds a new feature, holding that corporations spend money on campaign contributions and lobbying not to advance or satisfy their own interests, but rather the personal interests of their executives.[3] According to this view, corporate political activity is a symptom of poor corporate governance or of executives run wild. Moreover, by expending resources on contributions and lobbying that are not designed to serve the corporation’s broader interests, these political activities actually damage a company’s bottom line. This new view, therefore, rejects the academic consensus in holding that most companies and industries that are politically active produce lower returns for their shareholders.

This new view is fully articulated in three recent academic articles. The most far-reaching claims are contained in two studies conducted by John Coates of Harvard Law School,[4] while Rajesh Aggarwal of the University of Minnesota and his colleagues Felix Meschke and Tracy Wang offer a less sweeping analysis.[5] All three studies draw on a long line of work focused on what economists call the “principal-agent” or “agency” problem. Adolf Berle and Gardiner Means initiated this inquiry in the 1930s with a seminal analysis of how and why the interests and decisions of corporate managers sometimes diverge from the interests of the shareholders, who are the company’s principals.[6] Following the recent collapse and near-failures of financial institutions whose managers were handsomely rewarded for decisions that ultimately crippled their companies, few Americans would doubt the salience of agency problems.

This paper reassesses the economic benefits and costs of corporate political activity by critically surveying the literature and carefully analyzing the new work by Coates and Aggarwal et al. The relationship between business and government, we find, is complex and multifaceted. Firms employ a variety of approaches to influence the political process, with varying degrees of success. However, we find that the two generations of scholars across several disciplines who have carefully examined corporate political activity are broadly correct: most firms, like most individuals, behave rationally and strategically in their spending decisions on campaigns and lobbying, devoting resources in ways that, they have reason to expect, will benefit the corporations themselves and their shareholders. While this spending does not necessarily translate into the desired policy outcomes, corporate political activities do influence policymaking. Furthermore, numerous studies demonstrate that lobbying and other corporate efforts to engage with policymakers can have positive effects on firm value and shareholder returns. Moreover, there is virtually no credible evidence showing that corporate political activity harms firms and their shareholders. A close reading of the three studies that claim to establish such harm shows that their reasoning, methodology, and conclusions are flawed in fundamental ways.

This assessment is broadly confirmed by a forthcoming, magisterial study of interest-group influence in the United States by three of the nation's leading political scientists: Kay Schlozman, the J. Joseph Moakley Endowed Professor at Boston College; Sidney Verba, University Professor at Harvard University; and Henry Brady, dean of the Goldman School of Public Policy at the University of California at Berkeley.[7] The study covers political activity by all kinds of interests, including issue-oriented groups, unions, and nonprofit institutions, as well as corporations. The authors reviewed all published statistical studies of such influence and concluded that “[s]ome of them find significant influence; others show no significant influence. However, there are none that demonstrate a significant negative impact of organized interest activity on policy."[8] They also reviewed all published case studies of the same question and found: “1) organized interests do not always win; 2) they often get their way; 3) and, win or lose, organizations are never worse off, and are usually better off, for having gotten involved than they would have been if they had not been at the table.” [9]

II. Background

Theories of Why Corporations Participate in Political Activity

In this section, we will review theories of corporate political participation and the history of federal laws governing corporate political activities, including campaign contributions and lobbying. The view that companies undertake political activity to secure advantages for themselves is based on a combination of opportunity and necessity. Modern governments influence national economies in major ways that affect the sales and returns of particular industries and companies. While the United States has a smaller public sector than other large, advanced economies, its federal spending alone still accounts for one-fifth to one-quarter of annual GDP. This spending directly supports demand for the goods and services purchased either by government or by particular groups that receive income transfers from government. This federal spending, in turn, supports the sales of the companies and industries that produce those particular goods and services.

The government revenues that finance this spending are drawn from individual and business taxpayers, and the distribution of their contributions depends on a complex system of tax rates, deductions, credits, and exemptions. On the business side, those rates, deductions, credits, and exemptions, in turn, are based on particular characteristics that favor or disfavor some companies and industries, compared with others. One analysis by the Stern School of Business at New York University, for example, found that the effective corporate tax rate averages about 15.5 percent across all industries but ranges from 2.5 percent for biotechnology companies to 34.4 percent for retail automobile companies. [10] Counting only profitable companies, the average tax rate is 28.2 percent for all industries and ranges from 4.6 percent for private equity firms to 40.9 percent for telecom utility companies and 44.2 percent for security brokerages. Securing favorable tax treatment can clearly have large effects on the final bottom line of a company or an industry.

Similarly, countless large regulatory decisions have large economic effects for particular companies and industries. These decisions range from the application of a tariff or quota on certain imported goods, the sale or award of spectrum, and the application of reserve requirements to new financial products, to bailouts for certain companies deemed “too big to fail” and approval of certain alternative energy products for the renewable fuel standard program.

The powers wielded by lawmakers and other public officials to spend, tax, regulate, and otherwise constrain or stimulate certain economic activities create powerful incentives for firms and industries to compete for access to the political system, in order to influence the decision-making process in ways that could benefit their interests. [11] These corporate political activities assume a variety of forms, including direct campaign contributions, joining and supporting trade associations, lobbying, the hiring of former officials, advertising to move public opinion, and grassroots advocacy promotions.

The most commonly researched topic in this field is campaign contributions, and political scientists and economists have developed a number of theories to explain why firms and other interest groups form “political action committees,” or PACs, which then provide financial support to political campaigns.[12] The most common explanation is that campaign contributions help corporations and other interests secure greater access to legislators and other public officials.[13] This “access-seeking” theory focuses on the value to firms and other interests of being able to directly present their case and views to public officials and thereby increase the likelihood that those views will be considered in the decision-making process. By contrast, an alternative, “vote-buying” theory posits that interest groups exchange campaign contributions for specific decisions by legislators, on a quid pro quo basis.[14] Finally, a third, “influence elections” theory focuses on the potential impact of campaign contributions on election outcomes. According to this view, corporations, trade associations, and other interests contribute to candidates not to secure direct access or influence particular decisions but to help reelect incumbents or support challengers who already share their interests and policy preferences. These three theories of why corporations contribute to campaigns—to secure access, buy votes, or influence elections—are not mutually exclusive. Moreover, all three theories assume that corporations and other interests expect to benefit from their campaign contributions.

In contrast to the research on campaign contributions, which has focused on firms as a source of funds, the academic research on lobbying has stressed the role that firms play in providing information to legislators and other public officials. Members of Congress, it is generally assumed, are motivated primarily by the desire to win reelection, consistent with public choice theory,[15] and to advance policies that they believe are good for their district, state, or the country. Lobbyists can help them achieve both goals by providing information and expertise on specific matters pertinent to legislation and keeping them informed about how people in their districts or states feel about those matters. This view downplays the transactional quality often ascribed to interest-group politics, focusing instead on how corporations and other interests use information and expertise to persuade policymakers to adopt their views on policy matters. When lobbyists and legislators or executive-branch officials already agree on a matter and persuasion is unnecessary, lobbying can still serve a reinforcing role, especially when the members of Congress or other officials have limited time and resources to devote to each issue.[16] Other scholars have advanced a variation on this view, holding that the value that lobbyists provide is not information or policy expertise but political intelligence and connections.[17] Again, all these theories assume that the corporations or other interests that pay for lobbying expect to benefit from it.

Laws Governing Corporate Political Activity

“You don’t buy a United States Congressman with a contribution, of course, but you buy access and access is the name of the game.”—Rep. Mike Synar (D-OK)[18]

Laws Covering Campaign Contributions

Given the large economic incentives for corporations to contribute to campaigns and to finance lobbying activities, we should expect corporations to invest heavily in such activities. Federal law, however, has long placed significant constraints on the nature and extent of their political activities. For more than a century, Congress and the Supreme Court have tried to balance freedom of speech and the integrity of the democratic process in shaping the parameters of campaign finance law. The first major law governing campaign finance dates back to the presidency of Teddy Roosevelt. In the months following his election in 1904, it was revealed that Roosevelt had received some $2.2 million ($56.4 million in today’s dollars) in contributions from large corporations and wealthy individuals, including Standard Oil and J. P. Morgan. The resulting public outcry prompted Roosevelt to call for a ban on “all contributions by corporations to any political committee or for any political purpose," [19] and Congress did so in 1907 under the Tillman Act. After millions of Americans joined labor unions in the 1930s, Congress extended the ban to unions in 1943, and made that ban permanent in 1947 under the Taft-Hartley Act. Taft-Hartley also barred labor unions and corporations from making “independent expenditures”—spending not coordinated with a candidate or an affiliated committee—to influence the outcome of a federal election.[20] Despite President Truman’s protest that those provisions constituted “a dangerous intrusion on free speech,” Congress approved the bill by large margins.

The PAC was also invented in the same period. In 1943, the Congress of Industrial Organizations (CIO) was looking for a way to support President Franklin D. Roosevelt’s 1944 reelection campaign without violating the new law. To circumvent the ban on union contributions, the CIO created an independent entity to raise funds for political contributions and other activities from the CIO’s members, and named it a Political Action Committee.[21] Other unions as well as corporations and trade associations quickly created their own PACs, and the acronym became a generic term for all independent political fund-raising organizations.

The next major piece of campaign finance legislation came in 1972, with the Federal Election Campaign Act. The new law enhanced disclosure requirements for contributions to federal campaigns; but since corporations could not make those contributions, they were unaffected. In the wake of the Watergate scandal, however, Congress in 1974 set new limits on contributions from PACs as well as individuals: PACs could give no more than $10,000 to any one candidate in an election cycle, and individuals were limited to $2,000 per candidate, per cycle. While this new law was intended to reduce the political impact of major donors, the $2,000 limit on individual contributions and the law’s new clarity about PACs inadvertently spurred the explosive growth of PACs sponsored by corporations and other interests: between 1974 and 1978, the number of PACs more than tripled, growing from 516 to 1,828, and total PAC donations similarly increased, from $12.5 million to $35.1 million. [22]

Over the next several decades, Congress and the courts continued to modify and reinterpret the campaign finance laws. In 1975, U.S. Senator James Buckley and a coalition of interests challenged the 1974 act as a violation of freedom of speech. In the landmark 1976 case Buckley v. Valeo, the Supreme Court agreed in part, striking down limits on overall campaign expenditures and on expenditures by outside groups. In the years following the Buckley ruling, corporations, unions, and other interests found new ways to direct funds to political candidates. For example, while the Federal Election Campaign Act limited direct campaign contributions to federal candidates, it did not regulate money spent by political parties on campaign-related activities such as get-out-the-vote efforts and generic advertising. This omission allowed interests and individuals to direct “soft money” to federal campaigns without exceeding contribution limits. According to Federal Election Commission (FEC) data, over the 1992 to 2002 election cycles, soft-money contributions increased more than fivefold, from $88 million to $458 million. Many of these funds were directed to so-called issue ads, which many considered to be another loophole in the existing campaign finance law. Buckley v. Valeo had held that for ads to be covered by campaign finance laws, they must expressly advocate the election or defeat of an identified candidate, using terms such as “vote for,” “elect,” and “defeat.” Individuals, corporations, and other entities could therefore spend unlimited funds on ads supporting federal candidates, so long as they avoided those particular terms. In every other respect, these issue ads were often indistinguishable from regular campaign ads but were not subject to contribution limits and disclosure requirements.

In 2002, Congress attempted to close some of these campaign finance loopholes by passing the Bipartisan Campaign Reform Act (BCRA), commonly known as McCain-Feingold. In its most important provision, as it relates to this analysis, McCain-Feingold banned soft-money as well as corporate spending on issue ads intended to influence elections, referred to as “electioneering communications.”

Finally, in 2011, the Supreme Court eliminated most restrictions on corporate political spending. In Citizens United v. Federal Election Commission, the Court struck down the 62-year-old provision from Taft-Hartley barring corporations and unions from using their general treasury funds to finance campaign ads or other “independent expenditures” advocating for or against a candidate in a federal election. The justices held that such “express advocacy” funded by corporations is protected by the First Amendment, provided that it is not done in coordination with a candidate. This ruling has been followed by sharp increases in campaign contributions to “Super PACs” by very wealthy individuals, some of whom control private companies. However, data suggest that public companies account for less than 1 percent of Super PAC spending.

Laws Governing Lobbying Activities by Corporations

Lobbyists—individuals paid usually to influence legislation or regulation—have been a feature of American political life since the earliest days of the Republic. Yet Congress did not regulate their activities in a serious way until after World War II. To be sure, in 1876, the House of Representatives passed a resolution directing lobbyists to register with the Clerk of the House; and 43 years later, in 1919, Congress prohibited the use of federal funds to influence government policy under the Anti-Lobbying Act.[23] However, it was not until 1946 that Congress passed legislation requiring that all lobbyists not only register with the Clerk of the House and the Secretary of the Senate, but also file quarterly financial reports documenting their clients, their lobbying income and expenses, and the legislation they are paid to influence.

Even so, the 1946 act was routinely circumvented; and 37 years later, a Justice Department official in the administration of George H. W. Bush described the act to a Senate committee as “ineffective, inadequate and unenforceable.”[24] For example, the act’s requirements applied only to lobbyists whose “principal purpose” was to influence legislation in Congress, exempting lobbying efforts directed at the White House, cabinet departments, and regulatory agencies. There also was a large exemption for the in-house lobbyists of corporations or other organizations, who today account for some 40 percent of all registered lobbyists.[25] In-house lobbyists were required to register but not to disclose their activities. In addition, the Secretary of the Senate and the Clerk of the House had no enforcement authority, while the Justice Department emphasized voluntary compliance and rarely prosecuted anyone for failing to comply. In the 1954 case United States v. Harriss, the Supreme Court further limited the scope of the act, ruling that the law applied only to lobbyists who “directly communicate” with members of Congress on “pending legislation.” This interpretation effectively exempted from regulation all phone calls with members, grassroots lobbying efforts, and meetings with congressional staff, with whom lobbyists spend an estimated 98 percent of their time while on Capitol Hill. [26] In 1995, Congress strengthened regulation of lobbyists by passing the Lobbying Disclosure Act. This law extended disclosure requirements to in-house lobbyists at corporations, trade associations, and advocacy groups, and broadened the definition of lobbying to cover lobbying of executive-branch officials and staff, congressional staff, and lobbying for federal contracts.[27] However, critics contend that compliance remains weak because of relaxed oversight and enforcement by the Justice Department.[28]

The question remains: Do corporations benefit from expending money and other resources on campaign contributions and lobbying? A recent Harvard Business School case notes that “unlike other investments, the return on engaging in the political process is difficult, if not impossible, to calculate.”[29] In the following sections, we review in detail the literature on the impact of campaign contributions, lobbying, and the use of political connections, focusing on studies that measure the effects of such corporate political activities on firm performance. This analysis will help establish whether such activities generally help or harm corporations and their shareholders.

III. The Economic Value of Corporate Campaign Contributions

“It is ludicrously naive to contend that PAC money never influences congressmen’s decisions, but it is irredeemably cynical to believe that PACs always, or even usually, push the voting buttons in Congress.”—Larry Sabato, 1985[30]

Until the Citizens United decision, virtually all campaign spending by corporations and trade associations flowed through PACs established by those corporations and associations. The Federal Election Commission reports that from 1974 to 2011, the number of PACs registered with the FEC grew nearly eightfold, from 608 to 4,657[31] (Figure 1). In 2011, about one-third of those 4,657 PACs were corporate PACs; roughly one-third were established by trade associations, labor unions, and other politically active interests; and the remaining one-third were PACs defined by a policy or ideological position.[32] During the 2010 election cycle, corporate PACs spent over $305 million, including over $165 million in direct campaign contributions.

Figure 1: Growth of PACs and their Campaign Contributions

These extensive PAC activities are not distributed evenly across the economy: only about 10 percent of publicly traded firms, for example, have PACs.[33] Early studies found that resources and economic incentives determine which firms form PACs. A 1985 study published in The Journal of Politics was one of the first to explore this question, testing the impact of firm size, level of government regulation, and industry concentration on PAC activity.[34] The authors hypothesized, first, that larger firms with more employees have an advantage in raising PAC money because they have more executives, administrative personnel, and shareholders to bear the costs. They also reasoned that firms most affected by government policy, including those in industries that are heavily regulated or heavily dependent on government purchases, would be more likely to form PACs. Finally, the authors argued that firms in industries dominated by a relatively small number of large firms would be more likely to create PACs. This hypothesis was based on collective action theory, which holds that concentrated industries provide fewer inducements to free-ride and therefore are more likely to organize politically.[35] Based on the PAC contributions of 1,152 Fortune-ranked companies during the 1981–82 election cycle, the authors found, as expected, that the number of employees, the level of government regulation or purchases, and industry concentration all correlated positively with corporate PAC activity.

These findings were confirmed by a 1994 study published in the American Political Science Review.[36] Using data covering 124 industries over the 1978–86 election cycles, the researchers found that corporate PAC contributions are significantly higher in industries with greater sales, industries that are regulated, industries that are highly concentrated, and industries that generate a significant portion of their sales from government. For example, over this period, regulated industries on average contributed $167,000 more than unregulated industries (1988 dollars). The study also found that an additional $1 million in government sales was associated with an additional $132,000 in annual corporate PAC contributions to campaigns.

This research clearly suggests that cost-benefit considerations influence corporate decisions to form and use PACs. This view that firms behave rationally and strategically in their corporate political contributions is reinforced by evidence on the spending patterns of corporate PACs. Researchers in this area have found that corporate PACs tend to contribute to incumbents who are more likely to be reelected, candidates who support policies that benefit them, candidates in close races who most need campaign contributions, and candidates expected to have the most influence in Congress.[37] For example, corporate PACs contribute overwhelmingly to congressional incumbents, who also win the vast majority of their races. Corporate PACs also tend to favor Republican or conservative candidates whose voting records reflect support for business interests, in contrast to union PACs, which contribute overwhelmingly to Democratic candidates whose voting records are pro-labor. The finding that corporate PACs are more likely to contribute to candidates in close races is believed to reflect a view that those candidates will value the contributions more, and therefore will be more attentive and accountable to the contributing corporation. In addition, the expected value of the donation—its expected impact on the outcome—may be higher when the outcome is expected to be close.[38] Finally, numerous studies have shown that corporate PACs are more likely to contribute to party leaders, committee chairs, members with great seniority, and members of powerful committees such as the House Ways and Means, Financial Services, and Energy and Commerce committees.[39]

These various findings suggest that corporate PACs strategically target their congressional campaign contributions, generally directing them to those members who they expect will be most likely to pursue and deliver on legislative provisions consistent with the corporation’s interests. A 1992 study by Harvard economist James Snyder found an even broader range of strategic considerations in play. He found that contributions by corporate PACs, along with those by other nonideological PACs with narrowly defined economic interests, exhibit the characteristics of long-term investments: they persist over time, manifest a preference for younger legislators with long careers ahead of them, and favor those House members thought to be most likely to become senators (i.e., members from small states and from states with elderly senators).[40] Unsurprisingly, PACs also contribute less to members who have announced plans to retire or run for nonfederal office.[41]

The observation that PAC contributions reflect strategic or direct economic considerations has led some observers to conclude that the contributions directly influence congressional voting behavior. Yet, as one economic expert in this area wrote recently, “campaign donations can be ‘rational’ even when they do not alter how an individual politician votes.”[42] As an empirical matter, it is virtually impossible to know how a member of Congress would have voted on a particular bill in the absence of receiving campaign contributions from corporate PACs. Statistical analysis of the relationship between contributions and roll-call voting is problematic. Even in cases in which analysts can identify a correlation between PAC contributions and voting behavior, it may be impossible to determine its direction of the effect, if any: Did the contributions influence voting behavior, or did expected voting behavior influence contributions—or did a third factor, such as ideology, independently influence both PAC contributions and members’ voting behavior?[43]

Evidence of a systematic relationship between corporate PAC contributions and congressional voting behavior is mixed. In 2003, political scientists from MIT reviewed 36 published studies and concluded that roughly three-quarters of the time, there was no evidence of a statistically significant effect of those contributions on legislation.[44] Similarly, University of Chicago economist Steven Levitt concludes in 1998 that “[w]hile numerous papers document a correlation between PAC contributions and a politician’s voting record, those papers that most carefully attempt to identify a causal effect of PAC contributions typically fail to uncover evidence that PACs influence legislative roll-call voting patterns.”[45]

In some instances, case studies can establish that PAC contributions have influenced policy outcomes, especially when the policies in question involve relatively concentrated benefits and broadly diffused costs. A case in point is a well-known series of House votes on farm subsidies in 1985. A 1991 study found that eight out of ten of these votes were significantly influenced by PAC contributions.[46] Further, the researchers concluded that campaign contributions were decisive in at least one crucial vote, when recipients of $212,000 in contributions from 13 PACs representing sugarcane and beet growers blocked an amendment to reduce the support price of sugar from 18 cents to 15 cents per pound. This price-support program cost taxpayers and consumers, through higher prices, an estimated $3.3 billion per year.[47]

Other case studies of corporate PAC contributions and policy outcomes have failed to find any effects. This is unsurprising; across the policy arena, corporate interests sometimes win and sometimes do not. However, there are no case studies showing that corporations that provide PAC contributions were worse off for having been involved.

Figure 2: Sources of Congressional Campaign Financing,
PACs and All Other, 1980–2010

There are many reasons for these mixed conclusions. First, PAC contributions represent only a modest share—roughly a quarter—of total campaign spending;[48] and, as noted earlier, corporate PACs account for only one-third of total PAC contributions. During the 2010 election cycle, for example, congressional candidates reported raising a total of $1.8 billion, including $1.1 billion for House races and $740 million for Senate races. Of the $1.8 billion in total campaign contributions, $400 million, or 22 percent, came from PACs (Figure 2). Taking a longer view, PAC contributions represented 20–30 percent of all congressional campaign contributions from 1980 to 2010. Contributions from individuals continue to dominate congressional campaign fund-raising, representing some 55 percent of all contributions. To be sure, some individual contributions, including many contributions by corporate executives, are tied to the interests of the corporation and its industry. Other individual contributions, such as those of some tort lawyers, may reflect the interests of an industry composed largely of unincorporated businesses.

Furthermore, many factors influence the voting decisions of members of Congress, including the preferences of their party caucuses and leaders, the views of their constituents, and their own ideologies and personal beliefs. In a recent interview, Congressman Barney Frank, former chairman of the House Financial Services Committee and consequently a recipient of substantial PAC contributions, acknowledged that contributions influence congressional decision making in Congress. However, he noted: “If the voters have a position, the voters will kick money’s rear end any time. I’ve never met a politician … who, choosing between a significant opinion in his or her district and a number of campaign contributors, doesn’t go with the district.”[49] In addition, influence may occur without leaving evidence of it, and what appears to be evidence of influence may be misleading. In the first case, for example, Congress may enact legislation seen as adverse to the interests of corporate PAC contributors, but its provisions may have been subtly relaxed in ways not apparent to popular commentators or academic statisticians. Alternatively, laws that advance the interests of corporate PAC contributors might have been enacted in any case.

The view of many political scientists is that PAC contributions influence policy outcomes not by buying votes but by securing greater access for the contributors to government officials. Such access gives corporations and trade associations the opportunity to present their case and thereby influence legislators’ opinions on relevant policy issues. Interest groups intensively engaged in lobbying tend to make campaign contributions in ways consistent with this “access-seeking” theory.[50] In fact, researchers have correlated contributions and access. For example, a 1986 analysis using data on how congressmen spent their time in a typical workweek in 1977 found that a $294,000 contribution earned interest groups an average of an hour with a congressperson, compared with 35 minutes for contributors who gave $115,000.[51]

Table 1: Predicted Access to Members of Congress,
for Corresponding PAC Contributions, 1977 [52]
Average Time of Access
60 minutes
47 minutes
35 minutes
25 minutes

The general consensus drawn from the literature, then, is that corporations use PACs to make strategic, long-term investments in the political process, in the hopes of influencing policy outcomes in ways that help them and their shareholders. Corporate PAC contributions have the potential to directly influence certain legislation, and votes in Congress and corporate PAC contributors to members may often be aligned. Nevertheless, the evidence does not point to a systemic or determinative relationship between corporate PAC campaign contributions and votes by members on legislation that directly affects the interests of their corporate contributors. While media sometimes portray corporate PAC contributions as a form of bribery used to manage the political process, these contributions would be better characterized as a kind of “entrance fee” for firms interested in making their case to legislators and perhaps reinforcing the impact of other lobbying.[53]

IV. The Economic Value of Corporate Lobbying

“Turning to business lobbyists to draft legislation makes sense, according to DeLay, because ‘they have the expertise.’ ”—Washington Post, March 12, 1995

Lobbyists have been a feature of Washington life since at least the latter half of the nineteenth century. In 1913, Woodrow Wilson famously said, “This town is swarming with lobbyists, so you can’t throw bricks in any direction without hitting one.”[54] In recent years, lobbying has grown especially rapidly: from 2001 to 2008, spending on lobbying increased more than 10.5 percent annually, or about 7.5 percent per year in real terms. For purposes of comparison, corporate spending on gross domestic investment increased just 1.2 percent annually, in real terms, over the same years. By 2011, there were 12,654 actively registered lobbyists in Washington, or more than 23 lobbyists for each member of Congress.[55] Lobbying has long been the dominant form for political participation by corporations and other interests: in the 2010 election cycle, firms and other interests spent $6.8 billion on lobbying, compared with PAC expenditures of $1.3 billion (Figure 3).

Figure 3: PAC and Lobbying Expenditures,
Election Cycles of 2000–2010

While lobbyists often contribute to the campaigns of the same members of Congress they lobby, their principal role is to provide information and expertise to legislators, their staffs, and executive-branch officials and their staffs. At times, members of Congress even rely on lobbyists to help draft legislation.[56] In evaluating the impact of lobbying, our first question is: Who hires lobbyists? Last year, a team of economists led by William Kerr of the Harvard Business School published a study to answer that question.[57] To begin, over the period of 1998 to 2006, only about 10 percent of publicly traded firms lobbied in any given year.[58] Moreover, the firms that hire lobbyists are generally large, reporting average annual sales 3.8 times those of firms that do not lobby and total workforces 3.3 times the size of firms that do not lobby. In addition, firms that lobby do so persistently: the likelihood that a firm that lobbied last year will also lobby this year is 92 percent. Finally, there is a fairly strong correlation between lobbying expenditures by companies and campaign contributions.[59]

The larger question is whether firms lobby because it produces net economic benefits. Throughout the 1980s, research in the area of corporate political activities focused mainly on the relationship between campaign contributions and congressional voting behavior. In that period, most academics assumed that most lobbying activity was dictated by the policy preferences of legislators and their constituents. In 1990, however, John R. Wright at the University of Iowa published the first study to combine data on campaign contributions and lobbying activities, in order to better understand the relationship between the overall political activities of interests and voting at a congressional committee level.[60] His analysis was a case study of how lobbying influenced votes in the House Ways and Means Committee and the Agriculture Committee on two controversial bills in 1985.[61] The analysis drew on a survey of organizations that had lobbied those committees in the period leading up to the two votes. Using two different regression approaches[62] and controlling for political party, ideology, and constituency preferences, Wright found that organized lobbying efforts affected the voting in the committees to a statistically significant degree. In 1998, Wright and a colleague published a related study assessing the impact of lobbying efforts on the voting on Supreme Court nominations in the Senate.[63] Controlling again for party, ideology, and constituency, the authors found that here, too, lobbying had a statistically significant effect on senators’ votes on the confirmations of Robert Bork, David Souter, and Clarence Thomas. The results further suggested that if the lobbying for the Thomas nomination had been reduced by 10 percent, Justice Thomas would not have been approved; similarly, if lobbying against the Bork nomination had been reduced by 25–50 percent, Judge Bork would have been confirmed. While the study involved lobbying by primarily noncorporate organizations, it demonstrates the general capacity of modern lobbying to provide information and expertise that ultimately affects congressional voting on even very high-profile matters.

The Economic Returns to Lobbying

“There’s a reason that corporations invest in lobbying Congress: they see a return, and it’s good for their bottom line”—Mary Boyle, Common Cause [64]

While numerous early studies provided empirical evidence that voting by members of Congress can be influenced by lobbying, they did not estimate the returns or other benefits for those financing the lobbying. One reason is that data that might be used to estimate those returns were generally unavailable until the later 1990s, when the Lobbying Disclosure Act of 1995 mandated the collection and publication of official data on lobbying expenditures. Even so, measuring the returns to lobbying has remained difficult. To begin, firms that lobby are often different from firms that do not lobby—for example, in size, dependence on government purchases, industry regulation, and degree of industry concentration. Attempts to simply compare the economic performance of firms that lobby with those that do not, therefore, are distorted by “selection bias.” For example, if large companies earn lower returns than small companies, and large companies are more likely to lobby, a finding that companies that lobby earn lower returns may have nothing to do with their lobbying. (As we will see, this problem affects the Coates analysis.) In addition, it is often difficult to measure the value of government policies to individual companies that lobbied for or against them. Consider, for example, the passage of NAFTA in 1994 or the repeal of Glass-Steagall in 1999. In such cases and many others, economic analysis can establish the direction of the effect for particular industries, but quantifying the impact on the returns of particular companies is more problematic. Finally, it is often impossible to isolate the impact of lobbying from the effects of other corporate political activities, including campaign contributions, public advocacy, or grassroots organizing.

The first analysis to successfully address these methodological challenges focused on the returns of lobbying by universities.[65] The lobbying activities of universities are a particularly apt subject for this analysis because they are directed almost entirely at securing earmark funding, which can be directly measured. In addition, universities are not allowed to form PACs, mobilize grassroots campaigns, or engage in a number of other political activities, so their success or failure in securing earmarks can be directly attributed to their lobbying. The analysis found that universities located in the districts of members of the House and Senate appropriations committees (HAC and SAC) benefited substantially from their lobbying efforts: for universities with HAC representation, a 10 percent increase in lobbying produced a 2.8 percent increase in earmarks. Similarly, for those with SAC representation, a 10 percent increase in lobbying produced a 3.5 percent increase in earmarks. These results did not reflect merely the members’ propensity to channel funds to their own districts: universities with committee representation that spent less on lobbying received fewer earmarks, suggesting an independent positive relationship between lobbying and earmarks. A 10 percent increase in lobbying by universities without appropriations committee representation yielded a 1.5 percent increase in earmarked funding, although this result was not statistically significant.

In dollar terms, the authors estimated that each $1 in lobbying by universities with SAC representation returned $36 in earmarks, and each $1 in lobbying by those with HAC representation returned $25 return in earmarks.[66] Furthermore, the return on an additional dollar of lobbying by universities with SAC representation—their marginal return on lobbying—was $5.24. The marginal return in additional earmarks for an additional dollar spent on lobbying was $4.52 for universities with HAC representation and $1.56 for universities without such committee representation; but neither result was statistically significant.
Figure 4: Corporate Returns Relative to Benchmark,
Based on Intensity of Lobbying[69]

At a minimum, the results suggested that universities with representation on the Senate Appropriations Committee underinvested in earmark lobbying in this period and therefore “left money on the table.”

Another study of the economic impact of lobbying with a broader and more corporate focus appeared in the American Journal of Political Science in 2009.[67] Its authors sought to estimate the tax benefits associated with corporate lobbying. Using a sample of more than 6,200 firms, they found that every additional 1 percent in lobbying expenditures was associated with a 0.5–1.6 percentage-point reduction in effective tax rates the following year. For the average firm lobbying Congress, across industries and various measures of financial performance, this meant that an additional $1 for lobbying was associated with $6–$20 in new tax benefits.

Several recent studies have sought to estimate the impact of lobbying by publicly traded firms on their shareholders’ market returns. These analyses suggest that the market returns to lobbying are significant. One 2010 study found a positive correlation between lobbying expenditures by corporations and financial performance as defined by income before extraordinary items.[68] The authors reported that $1 spent on lobbying was associated with an additional $24–$44 in corporate income, depending on the model’s specifications. They further found that firms that lobbied most intensively—defined as lobbying expenditures as a share of assets, sales, and market cap—outperformed their benchmark by 5.5–6.7 percent per year, for the three years after their intense lobbying (Figure 4). Moreover, these results were derived using rigorous analytic methodologies.

Another 2010 study produced similar results.[70] This analysis estimated that firms that lobbied Congress outperformed those that did not by about 2 percent per year, controlling for financial performance. The study also found that among firms that lobbied, those that lobbied more aggressively achieved better returns: a one-standard-deviation increase in lobbying expenditures was associated with a 1.2 percent increase in excess returns—that is, even higher returns relative to non-lobbying firms than those achieved by other companies that lobbied. While some studies that compare lobbying and non-lobbying firms suffer from sampling bias, these two studies controlled for such bias by modeling both the decision to lobby and the effects of lobbying expenditures on market returns.[71]

Figure 5: Market Performance of Firms That Lobby Intensively, Versus the S&P 500

The generally positive returns from lobbying have not escaped the market’s attention. In 2009, Strategas Research Partners, a New York–based research firm that advises institutional investors, developed the “K Street Index,” a portfolio of firms that lobby the federal government most intensively, defined as lobbying expenditures relative to total assets.[72] Strategas updated its index in 2010, and it has since outperformed the S&P 500 by an average of more than 3 percent per year (Figure 5).[73]

In sum, lobbying can influence congressional voting behavior, which, in turn, can produce tangible benefits for firms and industries that lobby. The benefits of lobbying can take many forms, including lower taxes or higher federal spending. For publicly traded firms, the benefits of lobbying have been reported in stronger financial performance and stronger stock-market returns. Moreover, firms that lobby intensively tend to outperform their benchmarks to larger degrees. While the estimated benefits of lobbying vary from study to study, and lobbying in specific cases may not produce benefits, the literature contains no instance in which lobbying is associated with lower returns for firms and their shareholders.

V. The Economic Value of Corporate Political Connections

Much as the impact of corporate campaign contributions is often entangled with lobbying by the same corporations, the apparent benefits from lobbying may also reflect a firm’s political connections, sometimes created and enhanced by lobbying. These distinctions are subtle but genuine. The economic value attributed to lobbying may reflect not only the knowledge, expertise, and persuasiveness that lobbyists provide, but also the political connections generated by the lobbying. Those connections, in turn, can lead to increased access, valuable information about policy developments, and reduced uncertainty.[74]

Much of the research on lobbying acknowledges the importance of political connections. One of the studies covered in the previous section notes, for example, that “lobbying expenditures proxy firms’ political connections.”[75] Another study also reviewed above suggests that “lobbying could be correlated with an unobserved variable, e.g., government connections, which may be the real source of value to the firm.”[76] This theme runs through much of the research on both lobbying and campaign spending.

In recent years, the academic literature on political connections has grown considerably, reflecting a growing consensus that political connections have economic effects. The analytic challenge has been to find a rigorous way to define and measure political connections. Some of the proxy measures for these connections explored in recent studies include, in addition to campaign contributions and lobbying expenditures, the number of politically connected individuals serving on a corporation’s board and the number of board members or executives who served in prominent positions in Washington.

One of the first studies to examine the economic value of political connections appeared in the American Journal of Political Science in 1990. Professor Brian Roberts of the University of Texas sought to measure the impact of relationships with senior members of Congress on corporations. To do so, he conducted an “event study” to measure the effect of the sudden death of Senator Henry “Scoop” Jackson (D-WA) in 1983 on the value of firms.[77] As the ranking Democrat on the Senate Armed Services Committee, Senator Jackson had developed close connections with several publicly traded firms. Moreover, his death was unexpected and thus served as an exogenous shock to the perceived or expected value of those firms. The study found that firms located in Jackson’s home state of Washington and firms that had made PAC contributions to Jackson experienced negative abnormal returns on the first trading day after the news of the senator’s death. For example, 11 Washington State firms that had made PAC contributions to Jackson in the preceding two years experienced abnormal losses of 2.5 percent, and 24 firms located in other states that had made PAC contributions experienced abnormal losses of 1.1 percent. Both results were statistically significant.

A similar study published in 2006 focused on the impact of the surprise defection of Senator James Jeffords (R-VT) from the Republican Party in 2001, a move that shifted control of the Senate to the Democrats.[78] This analysis also used an event-study approach, measuring the impact of that significant and unexpected political event on the share prices of firms connected to each of the political parties. Using data on soft-money contributions by large public companies during the 2000 election cycle, the author found that the surprise shift in power and committee leadership from Republicans to Democrats did affect the market performance of politically connected firms—which the author called the “Jeffords effect.” The analysis found that each $100,000 donated to the Republican Party in the 2000 election cycle was associated with a negative abnormal stock return of 0.33 percent in the week of Senator Jeffords’s announcement. Similarly, each $100,000 donated to the Democratic Party was associated with a positive abnormal stock return of 0.17 percent (Figure 6). In dollar terms, every $1 that a firm donated to the Republican Party was associated with a loss of $2,313 in market value when Jeffords defected; and every $1 that a firm had donated to the Democratic Party was associated with a $2,219 gain in market value.[79] Other event studies have found that close elections and the nominations of politically connected individuals to corporate boards also cause abnormal stock returns in politically engaged firms.[80]

Figure 6: The Impact of the Jeffords Defection on the
Market Value of Politically Connected Firms[81]

By demonstrating that stock prices respond to events that affect the perceived value of a company’s political connections, these studies suggest that corporate political activities that enhance a company’s political connections benefit its shareholders. Moreover, a 2010 study published in The Journal of Finance suggests that the economic benefits of a corporation’s political connections are also evident over the long term.[82] Its authors reasoned that the number of politicians supported by a firm’s PAC money was a good proxy for the extent of a firm’s connections to politicians. Using data on more than 1,200 publicly traded firms between 1979 and 2004, their study found that firms that made PAC campaign contributions to more political candidates experienced higher subsequent market returns. Specifically, the authors reported that a one-standard-deviation increase in the number of candidates supported by a firm over the previous five years was associated with excess abnormal returns of 2.6 percent.[83]

While the studies noted above provide sound empirical evidence of a relationship between a firm’s political connections and its market performance, they do not explain precisely how such connections actually create firm value. One recent study, however, documents at least one channel through which firms can profit from their political connections—namely, through government contracts. Indiana University economist Eitan Goldman and two colleagues designed an event study around the elections of 1994 and 2000, when power shifted from the Democratic Party to the Republican Party in Congress and the White House, respectively.[84] They reported that following the 1994 midterm elections, when Republicans gained control of both houses of Congress, the 81 S&P 500 firms in their sample with board members who previously had held political positions as Republicans or in Republican administrations experienced an increase in the value of their government procurement contracts. Similarly, the 39 S&P 500 firms with comparable connections to the Democratic Party experienced a decline in their procurement contracts.

The authors estimated that the average value for a firm of having connections to the Republicans in 1994 was nearly $120 million per year in additional government contracts from 1995 to 1998. The study also found that following the 2000 election, when control of the White House shifted to Republicans, 55 S&P 500 firms with strong Republican connections and 39 S&P 500 firms with strong Democratic connections experienced similar shifts in government contracts. The average value for a firm of being connected to the Republicans in 2000, by the authors’ estimate, was about $45 million per year in government contracts from 2001 to 2004.

On balance, the literature shows that firms benefit from establishing ties with government officials. However, measuring the value of those benefits is challenging because the value of a corporation’s political connections is often difficult to distinguish from the value of other political activities. To address these methodological issues, some researchers have relied on event studies that measure the stock-market response to unexpected political events for politically connected firms. These studies find that such connections do affect share prices; and over the long term, politically connected firms can outperform their peers by 2–3 percent per year. Finally, across the literature, we have not found any cases in which researchers have identified a negative relationship between the political connections of a corporation or an industry and its shareholder returns.

VI. The New Revisionists: Claims that Corporate Political Activities Harm Companies and their Shareholders

Extensive analysis and evidence, then, support the view that corporate participation in the political process yields generally positive returns to firms and their shareholders. Three recent studies have challenged this consensus, arguing that, on balance, political activity by corporations harms shareholder value. The basic hypothesis is that the corporate executives responsible for these activities, especially for political contributions, act in their own interest, not those of their shareholders. By this view, corporate political activity is a symptom of what economists call an “agency” or “agent-principal” problem. From this, the authors of these studies infer that firms that engage in political activities will perform worse than their peers because they waste firm resources to advance their executives’ interest and because they generally are more likely to have poor corporate governance. A close examination of these studies shows that their reasoning and findings do not actually challenge, much less refute, the academic consensus that corporate political activity benefits shareholders or, at a minimum, does not harm them.

A recently published study by Rajesh Aggarwal from the University of Minnesota and two colleagues sets aside the corporate political activities that receive the most attention in the literature, including PAC contributions, lobbying, and political connections.[85] Instead, these authors focus exclusively on soft-money contributions to political parties to support operations such as get-out-the-vote campaigns and issue-focused ads, and donations to 527 organizations unregulated by campaign finance laws.[86] These contributions can come close to explicit support for candidates, but they remain technically independent of their campaigns. The study’s results suggest that the corporations most likely to make these contributions underperform their peers, from which the authors infer that their soft-money and 527 contributions were one reason for this underperformance. As we will see, these results support, at most, an inference that those companies’ underperformance may be related to other factors that also influence their decisions to contribute soft money and 527 funds. The results do not support the conclusion that even these limited corporate political activities harm shareholders.

The other two studies, from John Coates of Harvard Law School, argue that, in many cases, shareholders are harmed by all forms of corporate political activity, including lobbying and PAC contributions.[87] This provocative argument has received considerable attention from academics and media.[88] Like Aggarwal et al., Coates relies on the proposition that corporate political activity reflects the personal interests of executives, even to the point of harming their shareholders, and he characterizes this activity as a consequence of poor corporate governance. As we will see, Coates’s methodology is flawed. In addition, he fails to show that his results did not reflect other factors that may have independently influenced both the political activity and the governance issues in those corporations. This explanation would lead to conclusions much more consistent with the past generation of scholarly studies in this area.

Aggarwal: Do Soft-Money and 527 Contributions by Corporations Harm their Shareholders?

As noted, Aggarwal and his coauthors examined two types of unregulated campaign spending drawn directly from corporate funds: soft-money contributions to political parties and donations to 527 organizations. Their data covered 1991 to 2004. Soft-money corporate contributions were banned in 2002, but many corporations continued to make donations from corporate funds to unregulated 527 organizations, with the stipulation that the 527s would not “coordinate” with the campaigns of candidates. The authors report that for each additional $10,000 spent on soft-money or 527 contributions, a contributing firm underperformed its benchmarks by 0.74 percent, or a “negative excess return of 0.074 percent.” They calculate that this was equivalent to a loss in market value of $1.33 million per $10,000 contribution for the median donating firm. That result is equivalent to a remarkable $133 decline in market value for every $1 contributed.

Previous studies of corporate political giving generally focused on “hard-money” contributions to candidates by corporate PACs, drawn from the company’s employees, shareholders, or their families and subject to FEC regulatory limits. Aggarwal et al. focus on corporate political giving that falls outside FEC regulation. Such corporate donations, they hold, reflect the political preferences of the firm’s managers. Further, they claim that this “perquisites consumption” by corporate executives indicates “agency problems” in those firms, which, in turn, have negative effects on the corporation’s market performance and value. If this connection is correct, it would imply, at most, that agency issues, not political activities, are responsible for these firms’ underperformance.

It may be reasonable to expect that, compared with PAC contributions to candidates, soft-money donations to parties and contributions to 527 organizations are less likely to help firms gain access to officeholders or develop connections to those officeholders, which, in turn, can benefit shareholders. However, there is no reason to expect, as Aggarwal et al. claim, that such soft-money donations and contributions to 527 organizations would actually harm shareholders and especially that “donating to either winners or losers is associated with worse returns than not donating at all.”[89] Moreover, while the study excludes corporate PAC contributions and lobbying, the authors assert that “our results are quite similar if we also include individual and PAC donations” [90] without reporting those results or providing any evidence for this claim.

Other, more direct explanations for the market underperformance found by the authors are apparent when they report that the 11.4 percent of public companies that made soft-money or 527 contributions tended to be firms with below-average R&D and investment spending. They grant that the low investment and R&D spending of these firms may reflect poor management, the impact of which could overwhelm the positive effects arising from their political activity. Similarly, they report that better corporate governance was associated with smaller donations. Even accepting the authors’ definitions of good and bad corporate governance, this finding may merely suggest that better corporate governance was associated with allocating more firm resources to R&D and other investments. Alternatively, companies with low investment and R&D spending may simply be firms with lower expected profits and low returns on investment, without reference to the quality of their governance. For such firms in industries with strong competition from imports, it might well be a sound business decision to contribute to a party or politicians skeptical of the trade agreements reached in the 1990s that intensified that competition. Aggarwal et al. did not explore these possibilities but simply attributed the results to agency factors. In the end, Aggarwal and his colleagues do not establish in any definitive or reliable way that these political donations led to the lower returns for shareholders, much less that each $1 donated caused a $133 decline in a firm’s market value.

The authors’ statistical approach, while generally rigorous, is weakened by technical issues and subject to biases. For example, they measure the returns by the donating companies over the following year, compared with an event-study approach, in which the analysis of market performance is tied much more closely to the timing of the donations. They recognize that such “long term stock returns are noisier than event horizon returns” and that many other factors may influence the returns for a company that made soft-money or 527 donations in the previous year.[91] They also concede that standard regression techniques cannot address that much “noise.” They defend their design by noting that the donations are not disclosed to shareholders or the market and “are therefore likely to be capitalized into stock prices only slowly over time.”[92] However, that explanation only suggests that the price declines reflect more fundamental factors unrelated to the donations and, as noted earlier, plausibly unrelated to any governance issues. At a technical level, their results cannot establish any firm connection between the donations and the next year’s returns.[93]

In the end, Aggarwal and his associates virtually concede that they have not proved their case. The study yields two main observations. First, they find a positive correlation between certain features that they consider evidence of poor governance—large boards of directors, CEOs who also serve as chairmen of their boards of directors, and low ownership by institutional investors—and soft-money contributions and donations to 527 entities. Second, they observe that firms that made such contributions and donations underperform the market. Those observations, however, do not tell us how political donations influence firm performance. How much of this underperformance can be attributed to the elements of weak corporate governance that they use, independent of any political activity? Do firms that already perform poorly choose to allocate more corporate funds to political donations? If that is the case, their poor performance may be wholly unrelated to governance issues or political donations. Finally, how much of this poor performance, if any, can be attributed to its political donations? The authors cannot answer these questions. Further, they fail to identify any mechanism by which unregulated corporate campaign spending could damage firm performance. By contrast, numerous studies reviewed here demonstrate how corporate political activities translate into higher shareholder returns, whether through earmarks, lower taxes, or government contracts. The alternative theory that these unregulated corporate donations cause market underperformance remains vague and unintuitive.

Coates: Corporate Political Activity More Generally Harms Shareholders

The two Coates studies have attracted more attention than the Aggarwal et al. analysis because Coates’s focus and claims are broader. Where Aggarwal and his colleagues confine their case to soft-money and 527 donations, Coates argues in one study that firms that contribute to PACs and lobby harm their shareholders by doing so; he argues in a second study that the shareholders of at least some firms that contribute to PACs and lobby are thereby harmed. Much like Aggarwal et al., Coates’s models focus more on links between weak shareholder rights and agency costs that can impair a firm’s value than on the impact of corporate political activities. Like Aggarwal et al., he interprets his results as evidence that corporate political activity is a symptom of agency problems.

In both articles, Coates acknowledges that numerous studies show that corporate political activities have paid off for companies and industries through favorable trade provisions, earmarked spending, lower taxes, and eased regulation. Coates also grants that many scholars have found positive correlations between corporate PAC contributions and abnormally high returns, and between corporate lobbying expenses and higher corporate returns. Nevertheless, he argues that politically engaged firms tend to have poor governance, their political activities reflect the personal interests of their executives, and those political activities harm shareholders.

In his second article, Coates offers a theory of how this occurs: managers engaged in political activity become distracted, diluting their companies’ strategic focus and affecting corporate decisions regarding large projects. He presents a series of correlations that he claims demonstrate, if not prove, this thesis. For example, he focuses on companies with executives who use corporate jets for personal use, and uses that subset as a proxy for firms likely to act contrary to their shareholders’ interest. He then finds a positive correlation between such corporate jet use and corporate political activity, which, he says, “suggests that corporate political activity may also not be in shareholder interests.”[94] However, his assumption that the personal use of corporate jets is a sound proxy for poor governance is unproved. He offers no data on what share of corporate jet use by these companies was for personal use, or what share of corporate spending or profits such corporate jet use represents. Coates also does not explore whether such corporate jet use is an efficient use of corporate resources for some companies. Furthermore, he singles out CEOs who later were appointed to or nominated for public positions, and finds them more likely than other CEOs to have worked for firms that engaged in lobbying. That seems plausible; but he uses this finding to claim that the prospect of future political appointment shaped the political activity of the companies that those CEOs headed, without offering any evidence for this conclusion.

Figure 7: The Relationships Between PAC Activity
and Corporate Governance

Coates’s analysis has other inconsistencies and technical problems. In his 2010 article, he argues that S&P 500 firms with poor corporate governance, measured by an index of corporate governance indicators, are more likely to be politically active.[95] To support this position, Coates observes that 90 percent of the firms scoring highest on this index, which indicated poor governance, contributed to PACs, compared with only 63 percent of firms that scored lowest on the index, indicating sound governance.[96] Those same data show that the correlation reverses when one looks at PAC-contribution levels: as corporate governance improves from 4 to 3, 3 to 2, 2 to 1, and 1 to zero, the average level of PAC contributions increases (Figure 7). This finding, which Coates does not discuss, contradicts his basic thesis.

Coates’s second attempt in the 2010 study to show that corporate political activity harms shareholders involves a set of regressions to test for a direct relationship between shareholder value and corporate political activity. He reports that these regressions show that firms that lobbied or made PAC contributions between 1998 and 2004 had a lower ratio of market value to book value—“relative Q”—than firms that did not expend resources on lobbying or PACs. If this finding were true, it could suggest that firms with comparable physical assets (their book value) have lower market value if they also undertake political activity than those that do not. In practice, Coates’s test is poorly designed. Studies trying to measure the influence of corporate political activity on firm value typically look at financial variables, such as income, or at market variables, such as benchmark-adjusted returns.[97] Instead, Coates uses his relative Q to measure a firm’s value, for which there is little basis in the literature. Coates’s relative Q depends on Tobin’s Q, which measures the ratio between the market value and replacement value of a firm’s physical assets. In the literature, Tobin’s Q is sometimes used as a control variable but not as the primary variable of interest. Further, a firm’s book value—the market value of its physical assets—is an accounting concept. The relation between a firm’s book value and market value may largely reflect the value of its intangible assets, such as intellectual property and brands, which vary substantially by industry and firm size. Coates’s study ignores all these issues.

In addition, Coates’s regressions fail to control adequately for certain features of firms that are politically active, introducing self-selection bias into his sample. It is universally recognized in the literature that firms that lobby and form PACs are different from those that do not do so. Politically active firms, as noted earlier, are much larger: the average market cap of PAC-contributing firms in 2004 was larger than that of 92 percent of companies listed on the New York Stock Exchange.[98] Similarly, firms in industries that are heavily regulated or that depend on government contracts are more likely to be politically active. Without controlling for these differences, an observed correlation between firm performance and political activity may merely reflect a correlation between performance and some factor, such as size or industry, that makes it more likely that the firm will be politically active. Most rigorous studies in this area address this issue by using what is called a “two-stage Heckman selection model.”[99] Coates neglects to do so. He does not even include as variables in his regressions many of the firm- and industry-level characteristics associated with corporate political activity. This failure to adequately control for self-selection bias and omitted-variable bias would be sufficient basis for any study’s rejection by any peer-reviewed journal. However, it may explain why Coates’s 2010 results so thoroughly contradict the rest of the literature on the subject.

In his 2012 study, Coates addresses some of the methodological errors and problems from his 2010 study and accordingly modifies some of his 2010 claims and conclusions. He acknowledges that “corporate politics could fit into a good corporate strategy” and that some corporate political activity is “shareholder-oriented.”[100] He concedes that the political activities of regulated industries and industries with large sales to government—for example, financial services, telecommunications, and defense—enhance the returns of their shareholders. However, he continues to argue that much corporate political activity is not shareholder-oriented and that the net effect of corporate political activity is negative for shareholders of companies that are neither highly regulated nor highly dependent on sales to government. His results do not support those conclusions.

Using a sample covering the S&P 500 from 1998 to 2004, plus 2008 to 2010, Coates finds a positive relationship between political activity (PAC contributions and lobbying expenses) and firm value (measured by industry-adjusted Tobin’s Q) for regulated industries: high firm value correlates with greater PAC contributions and lobbying expenditures. [101] Coates’s definition of regulated industries encompasses roughly one-third of GDP and includes alcohol, tobacco, aircraft, pharmaceuticals, utilities, telecommunications, transportation, banking, and insurance. For other, unregulated industries, however, he claims to find a negative relationship, suggesting that their political activities harm their shareholders.

These last findings do not hold up to scrutiny. Coates improves on his 2010 study by controlling for relevant variables such as firm size and industry. But when he applies those controls, his correlations largely disappear. The only correlation that remains statistically significant is a negative relationship between the decision to lobby and firm value for unregulated firms. Even in that case, the coefficient’s 95 percent confidence interval reaches a value of –0.01, which means that the actual effect could very well be zero. In layman’s terms, while Coates finds that firms in unregulated industries that lobby have market-to-book ratios 7 percent lower than unregulated firms that do not lobby, at a 95 percent level of statistical confidence, the difference in firm value between companies that lobby and those that do not may be only 1 percent. In addition, his results showing a positive relationship between political activity and firm value for firms in regulated industries are much stronger, when the appropriate control variables are included.

Coates’s 2012 study has other problems as well. Several variables could have introduced bias into his models. There is also a possibility of reverse causation, which he tacitly acknowledges by noting that “it seems likely that politics and shareholder value influence each other.” For example, firms that already were performing poorly may turn to political activity in an effort to increase their value, only to find that it is not enough, given whatever factors produced their poor performance. In addition, in contrast to most of the literature in this area, both Coates studies consider only the decision to contribute or lobby while ignoring the level and intensity of the associated expenditures. This surprising and unexplained omission from Coates’s models render his remaining conclusions, at best, highly problematic. In the end, neither of Coates’s studies establishes that the political activities of corporations harm their shareholders.

VII. Conclusion

The relationship between political activity and firm performance or shareholder value is varied and complex. Case studies show that companies that make PAC contributions, conduct lobbying campaigns, and establish political connections sometimes win and sometimes do not. However, there are no case studies showing that industries or firms are worse off than they would have been, had they never become involved at all.[102] Similarly, some multivariate statistical studies find that such political activities have significant positive effects on the firms carrying them out, while other studies find no significant or discernible impact.[103] The Aggarwal et al. study may suggest that corporate activity directed to 527s and other unregulated, party-level entities is less likely to produce positive effects than PAC contributions, lobbying efforts, and political connections focused on particular officeholders and policy decisions, although no other studies confirm this result. Moreover, until the Aggarwal et al. and Coates studies, there have been no statistical analyses that claimed to find significant adverse effects arising from corporate political activity. We find that the Aggarwal et al. and Coates studies ultimately fail to establish such adverse effects.

The body of research in this area has established several important findings. First, firms employ a variety of strategies to influence the political process in ways that may, or should, improve their performance and benefit their shareholders. Second, corporate spending decisions on campaign contributions and lobbying efforts are generally made in a rational and strategic manner. Third, this political spending does not appear to systematically affect congressional voting, but it does regularly influence policymaking. Fourth, corporate political activity appears to have a generally positive effect on firm value, as reflected in excess market returns. Finally, the precise mechanisms that produce these positive effects remain unclear.


Abramson, Jill, and Timothy Noah. “In GOP-Controlled Congress, Lobbyists Remain as Powerful as Ever—and Perhaps More Visible.” Wall Street Journal, April 20, 1995.

Aggarwal, Rajesh K., Felix Meschke, and Tracy Yue Wang. “Corporate Political Contributions: Investment or Agency?.” Business and Politics 14, no. 1 (April 2012): [p. span].

Alston, Chuck. “Byrd’s Provision Still Leaves More Loophole than Law.” Congressional Quarterly Weekly Report 47, no. 43 (1989): [p. span].

Andres, Gary J. “Business Involvement in Campaign Finance: Factors Influencing the Decision to Form a Corporate PAC.” PS: Political Science & Politics 18, no. 2 (spring 1985): 213–20.

Ansolabehere, Stephen, James M. Snyder Jr., and Micky Tripathi. “Are PAC Contributions and Lobbying Linked? New Evidence from the 1995 Lobby Disclosure Act.” Business and Politics 4, no. 2 (2002): 131–55.

Ansolabehere, Stephen, John M. de Figueiredo, and James M. Snyder Jr. “Why Is There So Little Money in U.S. Politics?.” The Journal of Economic Perspectives 17, no. 1 (winter 2003): 105–30.

Austen-Smith, David. “Campaign Contributions and Access.” American Political Science Review 89, no. 3 (September 1995): 566–81.

———, and John R. Wright. “Competitive Lobbying for a Legislator’s Vote.” Social Choice and Welfare (1992): 229–57.

Bebchuk, Lucian A., Alma Cohen, and Allen Ferrell. “What Matters in Corporate Governance?.” Review of Financial Studies 22, no. 2 (2009): 783–827.

Berle, Adolf A., and Gardiner C. Means. The Modern Corporation and Private Property. [City]: Transaction Publishers, 1932.

Bertrand, Marianne, Matilde Bombardini, and Francesco Trebbi. “Is It Whom You Know or What You Know? An Empirical Assessment of the Lobbying Process.” NBER Working Paper No. 16765, February 2011.

Bhagat, Sanjai, Brian Bolton, and Roberta Romano. “The Promise and Peril of Corporate Governance Indices.” European Corporate Governance Institute, Law Working Paper No. 89/2007, October 2007.

Bloxham, Eleanor. “Boards Take Heat for Political Spending.” Fortune, October 18, 2010.

Bronars, Stephen G., and John R. Lott, Jr. “Do Campaign Donations Alter How a Politician Votes? Or, Do Donors Support Candidates Who Value the Same Things That They Do?.” Journal of Law and Economics 40, no. 2 (October 1997): 317–50.

Byrd, Robert C., and Wendy Wolff. The Senate, 1789–1989: Addresses on the History of the United States Senate. Washington, D.C.: Senate Historical Office, 1991.

Caldeira, Gregory A., and John R. Wright. “Lobbying for Justice: Organized Interests, Supreme Court Nominations, and the United States Senate.” American Journal of Political Science 42, no. 2 (April 1998): 499–523.

Center for Responsive Politics. 2012.

Chen, Hui, David Parsley, and Ya-Wen Yang. “Corporate Lobbying and Financial Performance.” Working paper, April 2010.

Clymer, Adam. “PAC Money’s Role in Congress Raises Suspicions.” New York Times, January 19, 1983.

Coates, John C. “Corporate Governance and Corporate Political Activity: What Effect Will Citizens United Have on Shareholder Wealth?.” Harvard Law and Economics Discussion Paper No. 684, 2010.

———. “Corporate Politics, Governance, and Value Before and After Citizens United.” Harvard Law School, Working Paper Series, 2012.

Cooper, Michael J., Huseyin Gulen, and Alexei V. Ovtchinnikov. “Corporate Political Contributions and Stock Returns.” The Journal of Finance 65, no. 2 (April 2010): 687–724.

Damodaran Online. Data set. de Figueiredo, John M., and Brian S. Silverman. “Academic Earmarks and the Returns to Lobbying.” Journal of Law and Economics 49, no. 2 (October 2006): 597–625.

Drutman, Lee J. “The Business of America Is Lobbying: The Expansion of Corporate Political Activity and the Future of American Pluralism.” Ph.D. thesis, University of California at Berkeley, 2010.

Elstein, Aaron. “Lobbying D.C. Really Does Work.” Crain’s New York Business, July 19, 2010.

Federal Election Commission. “PAC Count—1974 to Present.” 2012.

Fitzgerald, Alison. “Tobacco-Maker Lorillard Among Companies Whose Lobby Pays Off for Investors.” Bloomberg News, June 2, 2011.

Goldman, Eitan, Jörg Rocholl, and Jongil So. “Do Politically Connected Boards Affect Firm Value?.” Review of Financial Studies 22, no. 6 (2009): 2331–60.

Grenzke, Janet M. “PACs and the Congressional Supermarket: The Currency Is Complex.” American Journal of Political Science 33, no. 1 (February 1989): 1–24.

Grier, Kevin B., and Michael C. Munger. “Committee Assignments, Constituent Preferences, and Campaign Contributions.” Economic Inquiry 29, no. 1 (January 1991): 24–43.

———. “Comparing Interest Group PAC Contributions to House and Senate Incumbents, 1980–1986.” The Journal of Politics 55, no. 3 (August 1993): 615–43.

———, and Brian E. Roberts. “The Determinants of Industry Political Activity, 1978–1986.” American Political Science Review 88, no. 4 (December 1994): 911–26.

Grossman, Gene M., and Elhanan Helpman. “Protection for Sale.” American Economic Review 84, no. 4 (September 1994): 833–50.

Hall, Richard L., and Alan V. Deardorff. “Lobbying as Legislative Subsidy.” American Political Science Review 100, no. 1 (February 2006): 69–84.

Heckman, James. “Sample Selection Bias as a Specification Error.” Econometrica 47 (1979): 153–61.

Herndon, James F. “Access, Record, and Competition as Influences on Interest Group Contributions to Congressional Campaigns.” The Journal of Politics 44, no. 4 (November 1982): 996–1019.

Hill, Matthew D., G. Wayne Kelly, G. Brandon Lockhart, and Robert A. Van Ness. “Determinants and Effects of Corporate Lobbying.” Working paper, 2010.

Hillman, Amy J., Asghar Zardkoohi, and Leonard Bierman. “Corporate Political Strategies and Firm Performance: Indications of Firm-Specific Benefits from Personal Service in the U.S. Government.” Strategic Management Journal 20, no. 1 (January 1999): 67–81.

Jayachandran, Seema. “The Jeffords Effect.” Journal of Law and Economics 49, no. 2 (October 2006): 397–425

Kerr, William R., William F. Lincoln, and Prachi Mishra. “The Dynamics of Firm Lobbying.” NBER Working Paper No. 17577, November 2011.

Kim, Jin-Hyuk. “Corporate Lobbying Revisited.” Business and Politics 10, no. 2 (2008).

Koenig, Robert L. “Loopholes ... ‘Information Purveyors,’ Others Ease Through Gaps in U.S. Lobbying Law.” St. Louis Post-Dispatch, January 29, 1992.

Kranish, Michael, and Susan Milligan. “Oversight of Lobbyists Found Lax; Compliance with Rules Can Reveal Little.” Boston Globe, January 16, 2006.

Krueger, Anne O. “The Political Economy of the Rent-Seeking Society.” American Economic Review 64, no. 3 (June 1974): 291–330.

Kwak, James. “Citizens United v. FEC Turns 2—and It’s Still Wrong.” The Atlantic, January 20, 2012.

Lambert, Richard, and Donald Larcker. “Golden Parachutes, Executive Decision-Making and Shareholder Wealth.” Journal of Accounting and Economics 7 (1985): 179–203.

Langbein, Laura I. “Money and Access: Some Empirical Evidence.” The Journal of Politics 48, no. 4 (November 1986): 1052–62.

Levitt, Steven D. “Are PACs Trying to Influence Politicians or Voters?.” Economics and Politics 10, no. 1 (March 1998): 19–35.

Masters, Marick F., and Gerald D. Keim. “Determinants of PAC Participation Among Large Corporations.” The Journal of Politics 47 (1985): 1158–73.

Milyo, Jeffrey, David Primo, and Timothy Groseclose. “Corporate PAC Campaign Contributions in Perspective.” Business and Politics 2, no. 1 (2000): 75–88.

New York Times staff. “80,000 Lobbyists? Probably Not, but Maybe….” New York Times, May 12, 1993.

Oberholzer-Gee, Felix, Libby Cantrill, and Patricia Wu. “Note on Lobbying.” Harvard Business School Note 707-471, January 30, 2007.

Olson, Mancur. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, Mass.: Harvard University Press, 1965.

Poole, Keith T., and Thomas Romer. “Patterns of Political Action Committee Contributions to the 1980 Campaigns for the United States House of Representatives.” Public Choice 47, no. 1 (January 1985): 63–111.

———, and Howard Rosenthal. “The Revealed Preferences of Political Action Committees.” American Economic Review 77, no. 2 (May 1987): 298–302.

Richter, Brian Kelleher, Krislert Samphantharak, and Jeffrey F. Timmons. “Lobbying and Taxes.” American Journal of Political Science 53, no. 4 (October 2009): 893–909.

Rivlin, Gary. “The Billion-Dollar Bank Heist: How the Financial Industry Is Buying Off Washington—and Killing Reform.” Newsweek, July 18, 2011.

Romer, Thomas, and James M. Snyder, Jr. “An Empirical Investigation of the Dynamics of PAC Contributions.” American Journal of Political Science 38, no. 3 (August 1994): 745–69.

Roosevelt, Theodore. “State of the Union Address.” December 5, 1905.

Sabato, Larry. PAC Power: Inside the World of Political Action Committees. New York: Norton, 1985.

Schlozman, Kay Lehman, Sidney Verba, and Henry E. Brady. The Unheavenly Chorus: Unequal Political Voice and the Broken Promise of American Democracy. Princeton, N.J.: Princeton University Press, 2012.

Snyder, Jr., James M. “Long-Term Investing in Politicians; Or, Give Early, Give Often.” Journal of Law and Economics 35, no. 1 (April 1992): 15–43.

Sorauf, Frank J. “Political Action Committees.” In Campaign Finance Reform: A Sourcebook, ed. Anthony Corrado. Washington, D.C.: Brookings Institution Press, 1997.

State Journal-Register (Springfield, Ill.) staff. “U.S. Sugar Subsidy an Unsweetened Deal.” State Journal-Register, September 20, 1985.

Stigler, George J. “The Theory of Economic Regulation.” Bell Journal of Economics and Management Science 2, no. 1 (spring 1971): 3–21.

Stratmann, Thomas. “What Do Campaign Contributions Buy? Deciphering Causal Effects of Money and Votes.” Southern Economic Journal 57, no. 3 (January 1991): 606–20.

Teitelman, Robert. “Corporate Political Activities and Governance.” The Deal Magazine, October 4, 2010.

Terhune, Chad, and Keith Epstein. “The Health Insurers Have Already Won: How UnitedHealth and Rival Carriers, Maneuvering Behind the Scenes in Washington, Shaped Health-Care Reform for Their Own Benefit.” BusinessWeek, August 17, 2009.

The Economist staff. “Money and Politics: Ask What Your Country Can Do for You.” The Economist, October 1, 2011.

This American Life staff. “Take the Money and Run for Office.” This American Life, March 30, 2012.

Wall Street Journal staff. “Political Contribution Groups Face Test in House.” Wall Street Journal, October 17, 1979.

Weiser, Carl. “Enforcement of Law Almost Non-Existent.” Gannett News Service, November 16, 1999.

Weisskopf, Michael, and David Maraniss. “Forging an Alliance for Deregulation; Rep. DeLay Makes Companies Full Partners in the Movement.” Washington Post, March 12, 1995.

Wright, John R. “Contributions, Lobbying, and Committee Voting in the U.S. House of Representatives.” American Political Science Review 84, no. 2 (June 1990): 417–38.

Yahoo! Finance. “Historical Prices,” S&P 500 returns (dividends included).^GSPC&a=00&b=1&c=2005&d=11&e=31&f=2005&g=m.


  1. Rivlin (2011).
  2. Terhune and Epstein (2009).
  3. Bloxham (2010).
  4. Coates (2010); and idem (2012).
  5. Aggarwal et al. (2012).
  6. Berle and Means (1932). Using similar reasoning, Adam Smith in The Wealth of Nations expressed skepticism of publicly held enterprises managed by executives who were not their actual owners.
  7. Schlozman et al. (2012).
  8. Ibid., p. 296.
  9. Ibid., p. 294.
  10. Damodaran Online, data set.
  11. See Stigler (1971) and Krueger (1974).
  12. Herndon (1982).
  13. Austen-Smith (1995).
  14. Grossman and Helpman (1994).
  15. In political science, public choice analysis applies economic models and the paradigm of self-interest to understand the decisions of political officials and other political phenomena. The early, major works of public choice theory include Anthony Downs, An Economic Theory of Democracy (1957); James M. Buchanan and Gordon Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (1962); and Mancur Olson, The Logic of Collective Action (1965). Social Choice and Individual Values (1951) by Kenneth Arrow, Nobel Laureate in economics, is often cited for its influence on the early formulation of public choice theory.
  16. Hall and Deardorff (2006).
  17. Bertrand et al. (2011) and Drutman (2010).
  18. Quoted in Clymer (1983).
  19. Roosevelt (1905).
  20. Specifically, the Taft-Hartley Act made it unlawful “for any corporation whatever, or any labor organization to make a contribution or expenditure in connection with any [federal] election.”
  21. See Sorauf (1997).
  22. Wall Street Journal staff (1979).
  23. Byrd and Wolff (1991).
  24. Alston (1989), p. 2845-2846.
  25. Bertrand et al. (2011).
  26. Koenig (1992).
  27. A requirement that grassroots lobbying be covered by the law was dropped in response to opposition from House Republican Speaker Newt Gingrich and his Republican colleagues in the Senate.
  28. See Weiser (1999); and Kranish and Milligan (2006).
  29. Oberholzer-Gee et al. (2007).
  30. Sabato (1985), p. 126.
  31. Federal Election Commission (2012).
  32. Ibid.
  33. Cooper et al. (2010).
  34. Masters and Keim (1985); see also Andres (1985).
  35. Olson (1965). Collective action theory analyzes “free-rider” situations in which a number of individuals (including corporations) could benefit from taking certain actions, but the associated costs of doing so discourage any one of them from addressing the situation alone. The rational choice is to undertake this effort as a collective action with shared costs. In industries that are highly concentrated, however, the large players may have sufficient individual incentives to take action, even if their competitors could free-ride on those efforts.
  36. Grier et al. (1994).
  37. Poole and Romer (1985), Poole et al. (1987), Grier and Munger (1991), Grier and Munger (1993), and Romer and Snyder (1994).
  38. Corporate PACs may also believe that contributions to close races will be highly valued by party leaders.
  39. Ansolabehere et al. (2002).
  40. Snyder (1992).
  41. Bronars and Lott (1997).
  42. Ibid., p. 346-347.
  43. Grenzke (1989), Bronars and Lott (1997), and Ansolabehere et al. (2003).
  44. Ansolabehere et al. (2003).
  45. Levitt (1998), p. 20.
  46. Stratmann (1991).
  47. State Journal-Register staff (1985).
  48. This point is made by, among others, Milyo et al. (2000) and Ansolabehere et al. (2003).
  49. Quoted in This American Life staff (2012).
  50. Ansolabehere et al. (2002).
  51. Langbein (1986). Amounts have been adjusted to 2011 dollars.
  52. Ibid.
  53. Milyo et al. (2000).
  54. Quoted in Byrd and Wolff (1991),
  55. Center for Responsive Politics (2012).
  56. See, e.g., Abramson and Noah (1995); and Weisskopf and Maraniss (1995).
  57. Kerr et al. (2011).
  58. Richter et al. (2009) and Cooper et al. (2010) also report that only about 10 percent of firms lobby.
  59. Ansolabehere et al. (2002).
  60. Wright (1990).
  61. The Ways and Means vote involved paying for the renewal and expansion of an environmental “superfund” to clean up hazardous waste dumps. The Agricultural Committee vote involved setting price-support loan rates and direct subsidy payments for farmers to be included in the farm bill.
  62. Ordinary least squares, or OLS; and two-stage least squares, or 2SLS.
  63. Caldeira and Wright (1998).
  64. Quoted in Fitzgerald (2011).
  65. de Figueiredo and Silverman (2006).
  66. Each $1 in lobbying by universities without House or Senate Appropriations Committee representation produced $8–$12 in earmarks; but again, the results were not statistically significant.
  67. Richter et al. (2009).
  68. Chen et al. (2010). The study controlled for both industry and firm characteristics, avoiding selection bias.
  69. Ibid.
  70. Hill et al. (2010).
  71. As noted earlier, firms that lobby tend to differ from firms that do not lobby in a number of ways, including size, industry, regulatory burden, and industry concentration. In order to properly estimate the relationship between lobbying and firm performance, it is important to first control for these characteristics. These studies address this potential selection bias by first modeling for the decision to lobby, and then including those results in a second model to test the effects of lobbying on firm performance.
  72. Elstein (2010) and The Economist staff (2011).
  73. We tested the Strategas 2009 K-Street Index against the S&P 500 in 2010–12: we found that it outperformed the S&P 500 by an average of 3.2 percent per year. Strategas. Investment & Sector Strategy, May 2010,
  74. Hillman et al. (1999).
  75. Hill et al. (2010).
  76. Chen et al. (2010).
  77. Senator Jackson was sometimes referred to as “the Senator from Boeing” because of his strong support for the Washington State–based aerospace and defense contractor.
  78. Jayachandran (2006).
  79. This positive result was not statistically significant at the 5 percent level.
  80. Goldman et al. (2009).
  81. Jayachandran (2006).
  82. Cooper et al. (2010).
  83. The authors did not offer a strong hypothesis for why the value of political connections is not immediately priced into stock values, suggesting only that “investors may lack knowledge concerning the benefits to firms from participating in the political process” (ibid., p. 718).
  84. Goldman et al. (2009).
  85. Aggarwal et al. (2012).
  86. McCain-Feingold barred such soft-money donations by corporations (and other entities such as unions) in 2002, until the Supreme Court declared the provision unconstitutional in the Citizens United case in 2010.
  87. Coates (2010).
  88. See, e.g., Bloxham (2010), Kwak (2012), and Teitelman (2010).
  89. Aggarwal (2012), p. 4. They also do not consider the special circumstances surrounding certain elections that could have an impact on contributors, such as the unexpected Republican takeover of the House of Representatives in 1994 or the equally unexpected shift of Senate control to the Democrats in 2003, when Senator Jim Jeffords left the GOP.
  90. Ibid., p. 8.
  91. Ibid., p. 6.
  92. Ibid., p. 5.
  93. Ibid. also report that the “negative effects” of soft-money and 527 donations were unaffected by whether the industry of the firm was regulated. Similarly, they report that such donations by companies in industries with large fractions of sales coming from government were associated with neither negative nor positive effects on returns. Every other study that examines political activity by regulated industries and by firms that derive much of their sales from government, including Coates’s second study, conclude that such activity benefits shareholders. Aggarwal and his coauthors do not explain why soft-money and 527 contributions should have such different effects or, alternatively, whether such anomalous findings might reflect problems in their research design.
  94. Coates (2012), p. 18.
  95. Furthermore, the corporate governance index used by Coates here includes problematic elements. For example, the index counts the use of golden parachutes as an indicator of poor corporate governance. Yet, other scholars have found that the use of golden parachutes can produce positive price effects for shareholders (Lambert and Larcker [1985]). A recent study of the E-index of corporate governance notes that it does not demonstrate that “good governance can improve shareholder returns” but only that the evidence is “suggestive” that the governance provisions affect performance (Bhagat et al. [2007]).
  96. The E-index has six elements: staggered boards: limits to shareholder bylaw amendments; poison pills; golden parachutes; requiring supermajorities for mergers; and requiring supermajorities for charter amendments. See Bebchuk et al. (2009).
  97. E.g., to test for a relationship between PAC contributions and firm performance, Cooper et al. (2010) use a firm’s returns adjusted for an industry benchmark to measure its performance, and confirm their results using returns in excess of the Treasury bill rate and changes in firm earnings. Similarly, to explore the relationship between lobbying expenditures and firm performance, Chen et al. (2010) use income before extraordinary items to measure performance, and confirm their results using a portfolio analysis of benchmark-adjusted returns. Even Aggarwal et al. (2012) use benchmark-adjusted returns to measure firm performance.
  98. Cooper et al. (2010).
  99. Heckman (1979). The first stage models the decision to become politically active using the individual firm and industry characteristics that have been shown to be related to the decision to lobby or form a PAC, including sales, number of employees, and industry dummies that account for such things as government regulation and industry concentration. The resulting inverse Mills ratio (IMR), which represents the predicted likelihood that a firm becomes politically active, can be used as an independent variable in the second model, which estimates the actual correlation between lobbying expenditures or PAC contributions and firm performance.
  100. Coates (2012), p. 26.
  101. Coates does not explain why he omits 2005–07 but suggests that he did not have a full data set for those years. Schlozman et al. (2012).
  102. Schlozman, (2012).
  103. Ibid.