Print
E-MailRui Albuquerque is Associate Professor of Finance at Boston College Carroll School of Management. This post is based on a recent paper authored by Professor Albuquerque; Zicheng Lei, Lecturer in Finance and Accounting at the University of Surrey; Jörg Rocholl, President and Professor at the European School of Management and Technology; and Chendi Zhang, Associate Professor of Finance at Warwick Business School. Related research from the Program on Corporate Governance includes Shining Light on Corporate Political Spending and Corporate Political Speech: Who Decides?, both by Lucian Bebchuk and Robert Jackson (discussed on the Forum here and here).
There is increasing evidence that state public pension funds preferentially direct their investments towards local corporations, creating a bias which cannot be justified by subsequent returns. In our paper, Institutional Investors and Corporate Political Activism, we investigate the political activism of firms and how it is influenced by the presence of state public pension fund ownership. The paper shows that state-level political connections appear to be an important mechanism of political activism by corporations with state public pension fund ownership.
Adam Smith (1776) postulates utmost care when dealing with political demands by capital-owners, as their self-interest may significantly deviate from public interest. Capital-owners today are not restricted to businesses such as merchants and master manufacturers. Rather, states have amassed significant amounts of capital and control of business through state pension funds. This raises the important question of whether the actions taken by states as capital-owners have to be considered with the same care as those taken by businesses.
The U.S. Supreme Court’s landmark decision on Citizens United v. Federal Election Commission in January 2010 provides a unique opportunity to investigate the important question of a potential agency conflict in U.S. public pension funds. The decision asserts for the first time that corporations benefit from First Amendment protection regarding freedom of speech in the form of independent political expenditures. The ruling generated significant controversy and resulted in a seven-fold increase in independent expenditures to federal elections. President Barack Obama (2010) voiced the opinion of many regarding Citizens United in his State of the Union Address:
“Last week, the Supreme Court reversed a century of law that I believe will open the floodgates for special interests—including foreign corporations—to spend without limit in our elections. Well I don’t think American elections should be bankrolled by America’s most powerful interests …”
The Council of Institutional Investors (CII) and the Center for Political Accountability (CPA) urged S&P 500 companies in a letter to adopt rules to disclose all corporate political contributions and called on boards to review and approve such contributions (CPA-CII, 2010).
Citizens United represents the most dramatic change in corporate campaign financing since the Taft-Hartley Act of 1947 that prohibited corporations from making any expenditure in connection to federal elections. It thus provides a unique experiment to study how corporations with different ownership structures adjust their inputs to political activism. Corporations are not new to political activism and have used political connections, lobbying, and contributions by executives and Political Action Committees (PACs). We revisit and broaden Adam Smith’s concern by noting that public pension funds are agencies of state governments that could pressure the companies they are invested in to pursue political agendas outside the scope of public corporations, see e.g. Romano (1993), Mitchell and Hsin (1997). This creates a potential conflict of interest between public pension funds and other shareholders (see e.g. Woidtke, 2002, and Coronado, Engen, and Knight, 2003) and raises the important question of whether the market response to Citizens United depends on having institutional investors who may be engaged in political activism themselves.
Using a sample of 1,722 firm-year observations, we find that the average three-day return on the announcement of Citizens United amounts to 0.92%. In the cross-section, firms with more political connections exhibit lower three-day abnormal stock returns than firms with less political connections. This negative effect is concentrated on firms with high institutional ownership, whereas we find a positive market reaction for the firms with no institutional ownership. A one-standard-deviation increase in the number of political connections leads to a 1.20% lower three-day abnormal return for firms with high institutional ownership than for firms with zero institutional ownership, a relative loss of $83 million in market capitalization. This result is consistent with a general inability of high institutional ownership firms with established political connections to adjust to the presence of a new input to political activism.
We further investigate the result above. We show that it is the institutional owners without business ties to the corporation that drive the negative market reaction, suggesting that an arm’s length relationship may be more effective in imposing constraints on management. We further divide institutions without business ties to the corporation into investment companies, public pension funds, and private pension funds. Consistent with the potential of an agency conflict for states as owners of companies, we find that our main results are concentrated on public pension funds.
To shed more light on this finding, we explore the fact that twenty-three states had bans on independent political expenditures by corporations on state elections prior to Citizens United, besides the ban on all states on independent political expenditures on federal elections. State bans had been ruled constitutional by the U.S. Supreme Court in 1990 in Austin v. Michigan Chamber of Commerce. The decision in Citizens United overruled Austin and gives rise to a cross-sectional difference that allows the identification of the effect of Citizens United on corporate decisions based on company headquarter state. Corporations headquartered in ban states serve as the treatment group, while corporations in no-ban states form the control group (see also Spencer and Wood, 2014).
Firms in ban states, i.e. the treatment group, establish on average less state-level political connections after Citizens United than firms in no-ban states, i.e. the control group. This effect depends on the level of institutional ownership. Citizens United has a negative net impact on state-level political connections for low institutional-ownership firms, consistent with an ability to adjust to the presence of the new input. In contrast, high-institutional-ownership firms do not significantly change or even mildly increase state-level political connections after Citizens United. We again divide domestic institutions into several groups and find that the results are again driven by public pension funds. This evidence suggests that public pension funds put constraints on firms to not use the new avenue of political activism created by Citizens United.
The full paper is available here.
