MSc Finance
Master Thesis: 20 EC
Do politically connected boards benefit firms?
Evidence from the United States
Name: Xiao Yao
Student Number: S3065901
Email: Y.Xiao.8@student.rug.nl
Supervisor: Dr. Swarnodeep Homroy
1
Abstract
This paper investigates the impact of connections between corporates and politics on the companies’ financial as well as social and governance performance in the United States from 2003 to 2019. The political connections are defined as the nomination directorship of a company, where the nominated board of directors was a former US politician. We collect 129 nomination announcements from the BoardEX and connect them to 82 companies from the S&P 500. We find that politically connected firms exhibit positive abnormal returns from the first day after the announcements were made. Therefore we conclude that politically connected boards add value to firms. Additionally, we find there is no significant difference between politically connected firms and unconnected firms in the performance of financial transparency, bribery and corruption as well as fair competition but corporate governance, which suggests that politically connected boards drive companies in the United States to have a better corporate governance performance.
Keywords: S&P 500; Political connections; Board of Directors; Firm Value; Corporate Social Performance; Corporate Governance.
JEL Codes: G14, G30, G34, G38
1
1 Acknowledgements: I would like to sincerely appreciate my supervisor Dr. Swarnodeep Homroy,
2
1. Introduction
3
performance of politically connected firms through investigating CSR indicators: financial transparency, bribery and corruption, corporate governance as well as fair competition. To achieve these objectives, the boundary of political connections should be confined. In this paper, the political connection is defined as : the nomination of directorship within a company, where the nominated board of director was a former US politician, more specifically, a former member of US Congress (a member of the Republicans, Democrats or any other party in the US Congress) in a company. The boards play an essential role in a company in terms of its efficiency and performance; this paper, therefore, analyzes the connection between historical politicians and board of directors based on the growing belief that the board of directors is considered the highest legal authority within a company and the linkage of externality (Petrovic, 2008, Bird, Buchanan and Rogers, 2004, Green and Homroy, 2018).
This paper mainly investigates 82 companies from S&P 500, which have a historical politician as a board director from 2003 to 2019 in the United States, and estimates the changes of financial as well as social and governance performance to clarify whether politically connected boards benefit the connected firms. The results show that there are significantly positive abnormal returns on the first and the third day after the nomination was announced, which implies politically connected boards add value to firms, and therefore the financial benefits of political connections are beyond the associated costs. Additionally, we find market value negatively influences such abnormal returns, which means that larger the firm is, less firm value increases, as a result of political connections. Last but not least, it is surprising that companies with politically connected boards have comparable performance as ones without politically connected boards (we refer to them unconnected firms in the later part), in terms of bribery and corruption training, financial transparency as well as fair competition; they, however, significantly outperform than unconnected firms in corporate governance in the years 2011 and 2016. Such results strengthen the hypothesis that political connections drive firms to burden more responsibilities. This paper contributes to the existing literature by extending the investigation into the impact of political connections on companies' financial as well as social and governance performance at the same time. Moreover, it supplements research of political connections in developed and sound legal countries.
4
next two parts describe the methodology and data selection, and in the fifth section, the results will be presented and interpreted. The last part contains the conclusions and discussions.
2. Literature Review and Hypotheses Development 2.1 Theoretical Background
2.1.1 Social Exchange Theory
The social exchange theory is defined as mutual rewarding transactions between two entities, which include a range of patterns: resource, reward, reinforcement, outcome, opportunity, payoff, etc. (Emerson, 1976). The transaction is two-way, which means when an entity acquires one thing, he/she needs to provide another thing in exchange. Political behaviors involve transactional activities (Curry and Wadf, 1968). The offerings from companies that have connections with politicians materialize in direct (donating for political campaigns, working for government’ interests) and indirect ways (devote to rebuild after natural disasters, creating employments beyond obligations). In return, the political connections provide connected companies with advantageous business resources, opportunities, and preferential outcomes.
2.1.2 Definition of Corporate Political Connections
Existing literature defines: a company is politically connected if one of its top officers (chief executive officer, the board of directors or large shareholders) was a politician (head of the state, government minister or member of the Congress), (Faccio, 2006, Faccio Masulis and McConnell, 2006). They call these direct connections. Relatively, indirect connections are referred as the situations: 1) when the top officers are the relatives or friends of the politicians; 2) when a company seeks to establish a sort of relationship with the top officers, such as campaign contributions and corporate lobbying (Faccio, et al., 2006). Corporate lobbying is any attempt to persuade a politician, the government, or an official group for the benefit of special interests (Hill, Kelly and Van Ness, 2013). 2.1.3 Motives of Appointing Politically Connected Boards
5
2005). Of course, there are alternative ways to have political connections, such as campaign contributions and lobbying. In this paper, we argue that the appointment of political boards is superior to the alternative ways for two reasons. First, appointing former politicians as board members can be a means to increase lobbying capacity, but which does not have to be lobbying only. Politically connected boards could provide extra. Second, compared to campaign contributions and lobbying, hiring former politicians is more flexible and efficient because it does not have to persuade the legislators and governors as the alternative ways do. Instead, with knowledge about the procedures and regulations, ex-politicians can advise companies to adjust their strategies and investments so that they can fit the regulations and laws that are supposed to implement. Third, compared to hiring full-time lobbyists, appointing ex-politician as the board of directors is more likely to bind the interests of ex-politicians to a company, thus mitigating the agency problem. However, no matter it is the corporate lobbying or the corporate political connections, the ultimate goal is to gain political influences and insurance so that benefit interests group.
2.1.4 Financial Benefits
Former politicians announced as a member of boards is a positive signal to stakeholders and the market, because politicians are believed to have better access to various resources that benefit the companies than ordinary people. A former politician could benefit the companies where they are the board members with lighter taxation, and play a mediating role in corporate tax avoidance (Desai and Dharmapala, 2009). Wu, Wu, Zhou and Wu (2012) find that private firms with connected managers have lower effective tax rates (ETRs) than private firms without such managers do. Tax savings have a direct effect on increasing the after-tax equity of the firms. Therefore, good financial performance can be observed. Next to that, former politicians could bring investors and outside funds by the feat of personal influence and credibility. Having a former politician as a board director, firms are more likely to be granted credit by banks and other financial institutions and thus initiating projects with lower costs and achieving higher net margins than unconnected firms otherwise could (Faccio et al., 2006). Duchin and Sosyura (2012) argue that firms that have connections to politicians are more likely to receive government investment funds. According to the efficient market hypothesis (Malkiel and Fama, 1970), the benefits brought by former politicians will be reflected in positive abnormal returns of politically connected firms’ stock price. Given this, the first hypothesis of this paper is:
6
2.1.5 CSR Benefits
In most countries, politicians are elected and are therefore considered to be outstanding by the majority. Based on this, companies with politically connected boards are expected to have high social performance. Because politicians are considered to be experts in governance. Generally, high social performances are the outcomes of good governance (Ferrell, Liang and Renneboog, 2016). Moreover, as politicians have power and advantages over the general public, in return, they are required to take more social responsibilities and are required to disclose more information due to the supervision by the people according to the social exchange theory (Emerson, 1976). Based on similar intuition, politically connected firms should burden more social responsibilities rather than lessen those responsibilities. Consequently, they should do better in social and governance performances than ordinary firms do. Huang and Zhao (2016) find that companies with political connections significantly outperform those without political connections in society-oriented responsibility and customer-oriented responsibility. A study by Chatterji, Levine Toffel (2009) points out that the widely used ratings KLD fairly summaries companies’ social and environmental performance with good transparency. Based on this, the actual social and governance performance can be fairly reflected in the ratings. Therefore, given those, the second hypothesis of this paper is:
𝐻2: the rating (score) of social and corporate governance of politically connected firms is higher than that of unconnected firms in the United States.
2.1.6 Potential Bias of Event Study
In this section, we discuss potential sources of biased results when applying the event study. Partial anticipation issues appear when the participants in the financial market anticipate the nomination announcements. The estimates, in this case, are biased downwards. Schipper and Thompson (1983) show partial anticipation issues can be tackled by various sampling techniques.
7
reflecting a firm’s self-selection to choose the event. Self-selection is another channel of endogeneity.
Omitted variable bias
In an estimation model, there is a relevant variable for explaining the dependent variable left out and result in biased estimates and misleading inferences on the remaining parameters (Brooks, 2014). For example, in this study, the earnings of a subsidiary of a politically connected firm may influence the CARs of its holding company, but this variable is not included in the model, and then the estimates of the remaining parameters from OLS are biased and inconsistent.
Self-selection bias
Self-selection refers to situations when a rule rather than a random sampling is used to sample the underlying population that is objective of interest (Heckman, 2010). In this paper, the companies can choose when they would like to appoint the politicians. If the companies only hire ex-politician when they are in financial distress, then the effect of political connections on companies, in this case, maybe biased.
Reverse casualty bias Reverse casualty
In a model, we assume that independent variables explain the dependent variable, not other way around, however, this assumption does not hold sometimes. In this paper, we assume that politically connected boards lead to positive CARs. However, it also can be ex-politician tend to accept the appointment from companies with positive CARs.
2.1.7 Relevant Controversy
8
Similarly, political connections could also have negative influences on companies, in the fields of corporate social responsibility. Former politician boards can take advantage of their networks with policymakers or their influence on the government to hinder policies and regulations which are disadvantageous to them. Politically connected firms of less financial transparency are more able to reduce regulatory enforcement incidents against corporate fraud than unconnected firms are (Wu, Johan and Rui, 2016). Furthermore, being connected to politicians, corporates have more incentives to adopt political strategies, such as corruption and bribery due to the discretion over misuse of public power (Doh, Rodriguez and Uhlenbruck, 2003, Faccio, 2006, Zhou and Peng, 2012). To survive and win in intensive competitions, some companies tend to acquire vital qualifications and resources via backdoors especially when they are close to power. Moreover, because of their politically connected boards, these companies are advantageous in political lobbying so that they can block competitors by lobbying the government to formulate discriminatory policies (Yu and Yu, 2011). Last but not least, as believed by most people, politicians are experts at dealing with the agency problem but sometimes they may be overconfident about their experiences on governance issues and thus underestimating situations of corporate governance that might be different. Also, politically connected boards are more likely to neglect corporate governance because they believe, in the worst case, the political connections can act as last insurance for their interests.
2.2 Empirical Findings
Gu, Hasan and Zhu (2019) find that political connections have a significant impact on firms' daily operations at a micro-level. But how does this have impact firms? Qu and Harris (2019) further show that generally strong political links are positively associated with the likelihood of firm survival. Staying in business does not necessarily mean enhancing a firm’s value. According to Barney, Wright and Ketchen (2001), the growth of firms can be achieved only when firms obtain sustainable competitive advantages. It is unclear whether political ties provide such advantages to firms. A study by Hillman, ZardKoohi and Bierman (1999）investigates 31 US companies with political strategies incorporated into operations, and show such linkage positively correlates to firm value, and political strategies contribute to the firm-specific benefits. Additionally, Oliver and Holzinger (2008) find value-seeking is a determinant of firms' decisions on their engagement in political strategies.
9
political donations in the industries and find firms experience a positive stock price return after the winning election of Republicans in 2000, as a result of donating more to Republicans. Besides that, Green and Homroy (2018) find that firms connected to members of parliament exhibit positive abnormal returns during the change in parliamentary regulations in the United Kingdom. Next to that, Tian and Estrin (2008) prove that when the size of government shareholding on companies is sufficiently large, it positively affects corporate value compared to the situations where it is private. Similarly, Fisman (2001) investigates companies connected to the Suharto family in Indonesia and the result shows these companies lose value after the announcements of the health deterioration of President Suharto. Faccio and Parsley (2009) also show that the value of companies that were located in a politician’s hometown decreases when the politician’s unexpected death was announced.
Substantial studies investigate how lobbying impacts on companies' performance. Richter, Samphantharak, and Timmons (2008) find that for lobbying firms, a 1% increase in lobbying expenditure will lower the ETRs around one percentage point on average. Hill et al., (2013) show that connections via lobbying yield significant benefits to firms. Additionally, Chen, Parsley and Yang (2015) prove that lobbying firms have a better performance on income and free cash flows. These studies are conducted based on the direct monetary reward accruing to firms. However, in this paper, we use stock reactions as the proxy for the changes in the value of political connections. Our paper also differentiates study of Borisov, Goldman and Gupta (2015) by exploring politically connected boards, which less likely involves in unethical activities, and, under such condition, investigating the social and governance performance of those firms.
10
However, some studies have contrary conclusions on the impact of political connections on firm financial performances. Due to the association with distortions in investment efficiency, Duchin and Sosyura (2012) indicate that politically connected firms are more likely to underperform than unconnected firms. Faccio (2006) also proves politically tied firms compared with untied peers exhibit a significantly worse financial performance when the governments bailed out financially distressed firms. Additionally, some research finds that in terms of performance on sales returns and profit growth, non-connected companies demonstrated better performance compared with connected companies (Fan, Wong and Zhang 2007, Wang and Wu 2008).
Companies treat the dedication to pro-social activities as strategies for exchanging preferential treatments. Sánchez (2000) finds that firms in El Salvador donate to charitable foundations as an exchange for political legitimacy. Sim (2005) provides a piece of compelling evidence indicating that major firms in the US regularly use corporate social responsibility (CSR) as a tool to modify or influence the regulatory framework in their favor. Politicians build a positive image and reputation and gain political support from devoting to pro-social activities. Liston-Heyes and Ceton (2007) prove that CSR engagement assists the politician in the attainment of his (her) policy goals, and so reinforces support for him (her). So naturally, politicians tend to take CSR as a strategy for gaining support from stakeholders and their interests when they are appointed as a board of directors. Due to this strategic overlap, the effects work. As a result, the closer connections to the politician the companies are, the more and better responsibilities the companies take (Yi and Xu 2014).
11
connections are more likely to exhibit poor corporate governance practices. Such a negative impact exists as the management of companies slacks off corporate governance, and the political connection on the other hand act as a backup for their slackness. Faccio et al. (2006) further support this point by proving politically connected firms are significantly more likely to be bailed out and supported by governments than the similar unconnected firms when they are in a crisis. However, such a negative relationship is not always the case. In China, it is found that private companies have politically connected managers outperform similar companies without connected managers (Wu et al., 2012).
Financial transparency and the probability of being involved in financial fraud are influenced by a company's background. Political connections could act as a shelter to some extent when companies intend to conceal or cheat in their financial data. Chen, Ding and Kim (2010) investigate companies from 17 jurisdictions during the period 1997 and 2001 and illustrate high-level political connections made it harder to predict companies’ earnings. Such a phenomenon is more prevalent in countries with weak legal systems. Bliss and Gul (2012) investigate politically connected and unconnected firms in Malaysia and find that politically connected firms have a significantly higher likelihood to be audited by big audit firms.
12
political network that entrepreneurs develop and the extent of bribery that they have been engaged in.
3. Methodology
3.1Event Study Methodology
This paper investigates the effect of the nomination between former politicians and board directors in companies. Therefore, the event study is applied simply because it is designed to detect the immediate market response to an event announcement, and it can be suggestive for the long-term effects of the event (MacKinlay, 1997, Ait-Sahalia, Andritzky, Jobst and Nowak, 2012). In our study, if such nomination indeed adds value to companies, there should be a positive cumulative abnormal return (CAR) detected within the event window and vice versa. The event window is composed of three business days before and after the announcement date [-3, +3]. To obtain CARs, the abnormal returns (ARs) is first calculated based on the market model which is given as:
𝑅𝑖,𝑡 = 𝛼𝑖 + 𝛽𝑖𝑅𝑚,𝑡+ 𝜀𝑖,𝑡 (1) E(𝜀𝑖,𝑡) = 0 Var(𝜀𝑖,𝑡) = 𝜎𝑖,𝑡2
Where 𝑅𝑖,𝑡 is the return of stock i at date t, 𝑅𝑚,𝑡 is the market premium at date t, 𝛼𝑖 is the constant, 𝛽𝑖 is the coefficient and 𝜀𝑖,𝑡 is the error term.
The AR then can be obtained by:
𝐴𝑅𝑖,𝑡 = 𝑅𝑖,𝑡− 𝐸[𝑅𝑖,𝑡] (2) Where 𝑅𝑖,𝑡 is the return of a stock i at date t, 𝐸[𝑅𝑖,𝑡] is the expected return of a stock i at date t.
To have the expected return, we use the daily returns within the estimation window which is from days -120 to days -30 before the event window to estimate the parameters 𝛼𝑖 and 𝛽𝑖 based on the market model. Then the expected returns can be obtained via:
13
𝐶𝐴𝑅𝑖,𝜏1−𝜏2 = ∑𝜏2 𝐴𝑅𝑖,𝑡
𝑡=𝜏1 (4)
3.2 Cross-Sectional Analysis
Followed by, we explore an array of potential factors that may influence the CARs of politically connected firms by running a regression analysis. Three criteria are taken into account in factors selection: 1) whether it is a systematic factor; 2) whether it is a firm-specific factor; 3) whether it is a factor implied by the data. After referring the articles of Goldman et al., (2009) and Ait-Sahalia, et al., (2012)2 , we select three factors: macroeconomic environment (proxy with the global financial crisis in 2008), the position of a political party and market value. The regression model is presented below. In this model, the first two independent variables are dummy variables, the third one is a numerical variable.
𝐶𝐴𝑅𝑖,𝜏1−𝜏2 = 𝛼0+ 𝛽1𝐷1 + 𝛽2𝐷2 + 𝛽3𝑀𝑉𝑖 + 𝜀𝑖 (5) Where:
𝐶𝐴𝑅𝑖,𝜏1−𝜏2 is the cumulative abnormal return from 𝜏1 to 𝜏2 for stock i;
𝐷1 = 0 if the nomination is announced during the Global Financial Crisis period, July 2007 to December 2008;
𝐷1 = 1 otherwise.
𝐷2 = 0 if the nominated former politician is not a member of the party in power when the announcement was made;
𝐷2 = 1 if the nominated former politician is a member of the party in power when the announcement was made.
𝑀𝑉𝑖 is the market value i. 𝜀𝑖 is the error term.
3.3 Social and Governance Performance Analysis
Next, this paper investigates whether politically connected boards would cause a better result of the social and governance performance of the connected firms. To address this question, it is necessary to set a benchmark (unconnected firms). Utilizing comparing social and governance scores of politically connected firms with unconnected firms, the
2
14
improvement in social and governance performance of firms with politically connected boards can be detected. There are two different types of data of social and governance performance, dummy and numeric. A transformation is required to synthesize those two types of data. The method of transformation is through a scoring system. For a dummy variable, if it is a “Y” in the database, then it is given a “1”, otherwise given a “0” instead. As for the numerical variable, Corporate Governance needs to be transformed as well to unify the mensuration of these four indicators. The transformation for a numerical variable is that if the politically connected firm has a higher score than the unconnected firm, then the politically connected firm is given a "1", and unconnected firms is given a "0"; otherwise the politically connected firm has a “0”, and the unconnected firm has a “1”. To obtain the total score of these four indicators, we sum the score of individual indicators with equal weight. This method consequently makes the synthesis score range from 0 to 4.3
4. Data
There are three technical reasons that we base our research in the United States and the S&P 500. First, the United States is the most representative developed country with good social transparency and a sound legal system. Second, although the number of politically connected firms in the United Kingdom is much greater than of those in the United States, the US is still more advantageous than the UK when it comes to the number of politically firms with available data, according to Faccio (2006). Third, compared to NASDAQ-100 and Dow Jones Industrial Average, S&P 500 covers more firms and industries, and the literature we refer to also focuses on S&P 500 companies, which makes the comparison more indicative than the literature with otherwise index.
4.1 Data Availability
We collect board and director announcements from BoardEx, which covers information from January of 2000 to March of 2019. The historical members of Republicans, Democrats and other parties in the US Congress are available on the official US government websites (https://www.govtrack.us/congress/members/all). The former US
politicians are traced back to 1930 until 2010. Financials of companies from S&P 500 are obtained from DataStream, covering a period from July of 2003 to March of 2019. The social performance data (Financial Transparency, Bribery and Corruption Training, Corporate Governance, Fair Competition, and ESG total score) are also obtained from
3
15
DataStream in consecutive 7 years starting from 2010 to 2016 (the latest information available).
4.2 Data Selection
As mentioned earlier, this paper focuses on politically connected boards because of the crucial roles they play within a company. Additionally, compared to countries, such as Italy, France and China, where government ownership is less restrained, state-owned companies are relatively rare in the United States and therefore it is technically hard to obtain relevant data (Shleifer, 1998). Last but not least, some political connections forms, such as corporate lobbying and political donation, are more likely to involve unethical activities between firms and policymakers (Borisov, Goldman and Gupta, 2015), and lead to bad corporate governance (Aggarwal, Meschke, and Wang, 2008). However, one objective of this paper is to investigate how political connections per se impact on corporate social and governance performance, therefore, any form of political connection which might provide directive results on corporate social and governance performance should not be taken into account. The politically connected firms in this paper are obtained by matching the intersection of the names of the politician with the names of board directors of the companies that made nomination announcements. With the help of programming, initially, 79 intersected names were found (79 former US Congress members), which connect to 91 companies with 144 nomination announcements. Among these names, we find some politicians had more than one nomination within one company at different years or had nominations at other different companies. To keep our sample as large as possible, we do not exclude such cases because the nominations were announced at different times. Based on the assumption of event study, different event dates should be counted as different observations. Because event samples may not always random, there might be residuals correlation over time for a specific industry. This would provide biased and downwards standard errors. A good way to cancel out this possibility could be clustering standard errors at the industry level. There is no formal test to tell at which level the standard errors should be clustered, however, the general rule is always clustered at the highest level, and there should be somehow even spread between observations in each cluster.
16
difficult, and the criteria to which degree of relationship should be taken into account is hard to unify. Second, the effect of political connections on firms' value might be explained by more possibilities when involved in indirect connections. The effect can be determined by how the close relationship between the former politician and the nominated boards. When they were just ordinary friends, such political connections may not add value to firms at all. Therefore, on one hand, such form of political connections probably fails to reflect the real value of political connections itself; on the other hand, it is also econometrically difficult to detect the pure effect of political connections.
The analyses of the effect of nomination announcements on firm value are based on the event study. When applying the event study, the overlap problems should be taken into account. After checking the disclosed news of the initial 91 companies, we exclude 9 companies with stock price affecting events within one month before the nomination announcements, including announcements: M&A, restructuring, share repurchase, and major personnel and strategy changes. Those screening criteria finally limit our sample to 82 politically connected firms, 68 nominees and 129 nomination events. We further break down these 68 nominated boards and 129 announcements. All the 68 historical politicians are either Democrat (39 people) or Republican (29 people). Among them, 44 people were from House of Representatives, 23 were the members of the Senate and 1 was a former US president (shown in table 1). Next to that, we also notice that the number of nominations varies over the years, but overall, there are more nominees from the Democrat than the Republican every year except the year 2015, 2016 and 2017. It is also implied from figure 1 that, in some years (2008-2012, 2014, 2018) with a weak macroeconomy4 and major political events5, the number of political nominations is greater than that of other years, which suggests that companies tend to reduce damage by political connections when there are exogenous uncertainties. In a similar vein, as shown in table 2, there are more nominations announcements in the second and third year of the presidential term than that in the first and fourth years. This phenomenon is also economically reasonable because there are many uncertainties in the first year (policy direction and personnel appointment) and fourth year (potential party alternation) of a presidential term, therefore companies may avoid any political connection in these years due to the political uncertainties.
4
From2008 to 2012, the United States suffered Subprime Mortgage Crisis and Economic Recession.
17
Table 1— Summary of Data.
Variables Number
Nomination Announcement 129
Companies 82
Former US Congress Member 68
Representative 44
Senator 23
Former US President 1
Democrat 39
Republican 29
Table 2 — Timing of Nomination.
Nomination Announcements Total Nominations Democrat 76 Republican 53 Total 129 Classification
Year in Presidential Term first year second year third year fourth year 16 26 19 15 9 23 12 9 25 49 31 24
This table shows the break-down of the number of nomination announcements of politically connected boards in a specific classification. The first classification is based on the year of the presidential term in which the nomination announcement was made. First-year includes years: 2004, 2008, 2013, and 2017; second-year covers years: 2005, 2009, 2014, and 2018; third-year contains years 2006, 2010, and 2015; fourth-year refers to years: 2007, 2011, 2012, and 2016. Since the elections in 2012 and 2016 took place in November, the end of the year, therefore, the first presidential term in these cases starts in the next year, which is 2013 and 2017.
18
After having the list of politically connected firms, we collect the stock price of those companies and market index on a daily frequency. All calculations related to stock returns and market returns are on a daily basis. Unlike stock prices, we collect the firm characteristics (firm market value) on a yearly basis from 2004 to 2018. The market values of firms are dynamic and change over time. Figure 2 suggests that the market value changes consistently with the US macroeconomic environment. In recent years, as the US economy hits new highs, the market values of US firms also continuously grow, while during the period of the financial crisis, they reached their lowest points.
Figure 2— Firm Characteristics Over Time
As mentioned above, this study also investigates the social and governance performance of firms, in the following four indicators: Financial Transparency, Bribery and Corruption Training, Corporate Governance, and Fair Competition. We select these indicators based on two main reasons: 1) data from these indicators are relatively complete and deviated from company to company. It is non-sensible to compare the difference in indicators in which all companies have the same data; 2) extensive studies focused on these indicators have been done in developing countries, which provides enough empirical evidence for our study. All indicators are on a yearly basis because the corporate governance report is
0 5000 10000 15000 20000 25000 30000 35000 40000 45000 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Ma rket V al u e in ml n Year
19
published annually. Companies completely without ESG information are not selected, nor those companies which have missing information about their social and governance performance for consecutive 2 years. The number of companies does not change after adding this condition.
To these ends, we gather data of equal-weighted S&P 500 returns from WRDS on a daily basis. The equal-weighted S&P 500 returns are calculated based on the equal-weighted portfolio across the companies from the S&P 500. Because of the missing data in 2019, the valid events decrease to 127 when we conduct the robustness tests based on the data from WRDS. Next to that, we collect annual data of the Equal Weight Rating of FESG (we refer as Total ESG Score in the following parts) from DataStream. The range of Equal Weight Rating of ESG is from zero to one hundred.
4.3 Data Processing
Following the method used by Faccio et al. (2006), we match the politically connected firms with unconnected firms which have comparable market value with a 20% deviation to mitigate the difference in social and governance performance between these two kinds of firms raised by the difference in financial power. The assumption for this matching is that comparable companies, measured by market value, should have similar capacities and abilities to devote to socially responsible activities. In the end, there are 82 politically connected firms and 81 unconnected firms in our observations. Apparently, there is one unconnected firm that pairs two politically connected firms at the same time. This is because, for several extreme cases, the options available for politically connected firms that can be matched are very limited under the restriction of the 20% deviation in market value. From 2010 to 2016, the average market value of politically connected firms is 39866.19 million USD, compared to 39601.47 million USD of unconnected firms; the median of market value for connected firms is 18750.25 million USD while for unconnected firms, it is 18713.26 million USD.
20
two consecutive years, there are still some firms with missing data in the early years, for example in the year of 2010 or 2011. If we screen such companies out, our sample would be very limited. To overcome this dilemma, we assume that the score for the year without information is the same as the score of the next coming year. Based on this assumption, we fill some missing data in the original databases which shows as “NA”.
After transforming dummy variables into numerical variables via scoring system mentioned in the last part. The synthetic score is obtained. As seen from table 3, the politically connected firms have a higher average synthetic score (3.11) than that (2.95) of unconnected firms over 2010 to 2016, with the standard deviation 0.079 and 0.090 respectively. Moreover, the total ESG score of politically connected firms is also greater than that of unconnected firms over the same period. However, it is worth to be noticed that differences in those two scores between these two kinds of firms are narrow. We also find the politically connected firms have higher synthetic scores over unconnected firms on average in each year during the period of 2010 to 2016. To present the difference between politically connected firms and unconnected firms and show how they change over the years in a straightforward way, we make figures in the below:
Table 3—Descriptive Statistics of Politically Connected Firm and Unconnected Firm.
Variable N Mean Median Std. Dev. Min Max
Politically Connected Firms:
Market Value 82 39866.19 18750.25 51389.79 3082.94 196584.9
Synthetic Score 7 3.1080 3.1341 0.0795 2.9390 3.1707
Total ESG Score 7 79.7753 78.26293 4.7117 75.8471 88.3246
Unconnected Firms:
Market Value 81 39601.47 18713.26 51340.81 3079.34 216158.5
Synthetic Score 7 2.9512 3.000 0.0902 2.8171 3.0366
Total Score ESG 7 79.3593 77.39 4.6455 75.5212 87.6282
21
Figure 3—Average Synthetic Score of Politically Connected Firms and Unconnected Firms From 2010 to 2016.
In this graph, it shows how the social and corporate governance performance of politically connected firms and unconnected firms changes over time based on the synthetic score which composes of Financial Transparency, Bribery and Corruption Training, Corporate Governance and Fair Competition with equal weight.
2.75 2.8 2.85 2.9 2.95 3 3.05 3.1 3.15 3.2 2009 2010 2011 2012 2013 2014 2015 2016 2017
Average Synthetic Score
22
4.4 Data Description
Table 4— Description of Indicators of Social and Governance Performance.
Indicator Form Description
Financial Transparency Dummy Describes whether the company has a policy to improve financial transparency.
Fair Competition Dummy Describes whether the
company claims to have processes in place to improve fair competition.
Bribery and Corruption Training
Dummy Provides the information if the
company trains its employees in preventing corruption and bribery.
Corporate Governance Numeric Reflects a company's capacity and the extent to which the board of directors and executives work for the interests of their shareholders.
Total ESG Score Numeric Reflects the extent of
company performance in terms of economic, environmental, social, and corporate governance pillars in equally weighted balance. This table provides the descriptions of indicators used in Social and Governance Performance.
5. Result
5.1 Cumulative Abnormal Returns of Politically Connected Firms
23
Table 5 — T Statistics of Cumulative Abnormal Returns (CARs) With Different Intervals.
CAR Interval
N Mean Std. Err. T Statistics
(-1,+1) 129 0,0037 0,00246 1,512*
(0,+1) 129 0,0037 0,00214 1,736**
(0,+3) 129 0,0047 0,00278 1,684**
(+1,+3) 129 0,0043 0,00234 1,828**
*,**,*** denote 10%, 5%, 1% significance level respectively; Std. Err. stands for standard error mean. In this table, it presents the results of stock reaction (value-weighted CARs) to the nomination between US former politicians and the board of directors with different event intervals. The estimation window is from 120 to 30 days before the event window, and the event windows are chosen three days before and after the announcement date.
The nomination between a former politician and board director is value-enhancing to a company because the market believes the political halo brought by former politicians could help a firm to acquire kinds of beneficial business resources, and these benefits are more than the associated costs. Moreover, a politician is supposed to be a superb administrator or governor who is good at making vital and strategic decisions that would benefit a company’s long-term value. Having such a board of directors, the company is more likely to succeed. Overall, the nomination between a former politician and a board of director is a positive signal to the market and a company, and such signal is reflected in its stock price. 5.2 Cross-Sectional Analysis
24
observations, which also can result in insignificant results. Furthermore, the results indicate that the consistency of party position between the former politicians and the party in power does not affect the CARs either. This seems to be inconsistent with the finding by Goldman et al., (2009) that companies connected to the Republican Party increase value, while those connected to the Democratic Party decrease value when the Republicans won the election in 2000. However, the results are answers to two different questions. The study by Goldman et al., (2009) focuses on a specific election year. Normally, in the election year, a political party would spend many resources on political propaganda, therefore the effects of political influence on stock price would be more significant than that in any other non-election years. The nomination announcements in our observations are not limited to election years, therefore the position of a political party may not affect the CARs.
Table 6 — Cross-Sectional Analysis of Cumulative Abnormal Returns (CARs).
Independent Variable (Value-weighted) Dependent CAR(-1,+1) Variable
CAR(0,+1) CAR(0,+3) CAR(+1,+3)
Financial Crisis Coef. 0.0064 0.0069 0.0102 0.0079 t-statistics (0.66) (0.90) (0.90) (0.87) Political Party Coef. -0.0018 -0.0059 -0.0034 -0.0015 t-statistics (-0.35) (-1.38) (-0.60) (-0.33) Market Value
Coef. -4.56e-08 -5.11e-08 -6.10e-08 -4.22e-08
t-statistics (-1.46) (-1.89*) (-1.70*) (-1.28) Constant Coef. t-statistics 0.0012 (0.12) 0.0032 (0.40) 0.0003 (0.03) 0.0001 (0.01) 𝑅2 0.0152 0.0241 0.0262 0.0267 N 129 129 129 129
25
5.3 Social and Governance Comparison Between Politically Connected Firms and Unconnected Firms
Table 7 shows the results of differences in synthetic scores between politically connected firms and unconnected firms. As presented, only in the year of 2011, the synthetic score of politically connected firms is significantly higher than that of unconnected firms, at a 5% level. This difference implies that, overall, politically connected firms have a better social performance than unconnected firms, in terms of financial transparency, bribery and corruption, corporate governance as well as fair competition (the same conclusion can be made in 2016 if it is at a 12% significance level). While in other years, differences between these two kinds of firms are insignificant.
Table 7— T Statistics of the Difference in the Social and Governance Performance Between Politically Connected Firms and Non-connected Firms Based on Synthetic Score between 2010 and 2016.
Year N Mean Std. Err. T
Statistics 2010 82 0.1098 0.1753 0.626 2011 82 0.3537 0.1721 2.055** 2012 82 0.0976 0.1634 0.597 2013 82 0.1342 0.1550 0.866 2014 82 0.0610 0.1631 0.374 2015 82 0.1342 0.1550 0.866 2016 82 0.2073 0.1635 1.268
*,**,*** denote 10%, 5%, 1% significance level respectively; Std. Err. stands for standard error mean. In this table, it shows the difference in social and governance performance of politically connected firms and unconnected firms based on the synthetic score which compose of Financial Transparency, Bribery and Corruption Training, Corporate Governance and Fair Competition.
26
Table 8—T Statistics of the Difference in individual indicator of the Social and Governance Performance Between Politically Connected Firms and Unconnected Firms between 2010 and 2016.
2010 2011 2012 2013 2014 2015 2016 F. T. Mean t statistics -0.0244 (-0.630) 0.0122 (0.376) 0.0244 (0.815) 0.0244 (0.815) 0.0244 (0.815) 0.0244 (0.815) 0.0244 (0.815) B. C. Mean t statistics 0.0488 (0.705) 0.0732 (1.180) -0.0122 (-0.207) -0.0244 (-0.406) -0.0244 (-0.406) -0.0244 (-0.406) -0.0244 (-0.406) C. G. Mean t statistics 0.0244 (0.220) 0.2195 (2.025**) 0.0732 (0.660) 0.1220 (1.106) 0.0488 (0.440) 0.1220 (1.106) 0.1951 (1.791**) F. C Mean t statistics 0.0610 (1.092) 0.0488 (0.893) 0.0122 (0.228) 0.0122 (0.228) 0.0122 (0.228) 0.0122 (0.228) 0.0122 (0.228) N 82 82 82 82 82 82 82
*,**,*** denote 10%, 5%, 1% significance level respectively; F. T., B. C., C. G. and F. C. stand for the indicator: Financial Transparency, Bribery and Corruption Training, Corporate Governance and Fair Competition respectively. In this table, it shows the difference in social and governance performance of politically connected firms and non-connected firms based on the individual score of Financial Transparency, Bribery and Corruption Training, Corporate Governance and Fair Competition respectively.
27
regardless of whether the firms are politically connected or not, they perform no difference in many oriental-setups indicators because of the similar strategies every company takes. Furthermore, stock exchange commissions may have requirements for listed companies on their social and governance performance. For example, in 2008, the Shanghai Stock Exchange issued an announcement that requires all listed companies to improve their social responsibility. To be listed for public trading, firms need to strengthen their social responsibility so that they could meet the requirements by stock exchange commission. Consequently, social and governance performance is insignificantly different among the listed firms due to standardization. As for the reason why the difference only appears in two years out of seven. A possibility is that the effect caused by the former politician boards in the social and governance performance of companies takes time to work and cannot be caught by data immediately. As a result, it may take four to five years to be reflected in data.
5.4 Robustness Tests and Further Tests
In this section, robustness tests and further analyses are performed. The results are presented below.
5.4.1 Equal-weight Returns
28
Table 9 — T Statistics of Cumulative Abnormal Returns (CARs) With Different Intervals.
CAR
Interval N Mean Std. Err. T Statistics
(0,+1) 127 0,0028 0,00190 1.450*
(0,+3) 127 0,0038 0,00267 1.420*
(+1,+3) 127 0,0039 0,00280 1.407*
*,**,*** denote 10%, 5%, 1% significance level respectively; Std. Err. stands for standard error mean. In this table, it presents the results of stock reaction (equal-weighted CARs) to the nomination between US former politicians and the board of directors with different event intervals. The estimation window is from 120 to 30 days prior to the event window, and the event windows are chosen three days before and after the announcement date.
Table 10 — Cross-Sectional Analysis of Cumulative Abnormal Returns (CARs). Independent Variable (Value-weighted) Dependent CAR(0,+1) Variable CAR(0,+3) CAR(+1,+3) Financial Crisis Coef. 0.0003 0.0021 0.0015 t-statistics (0.05) (0.19) (0.17) Political Party Coef. -0.0053 -0.0044 -0.0032 t-statistics (-1.38) (-0.80) (-0.56) Market Value
Coef. 5.60e-09 -5.87e-08 -7.93e-08
t-statistics (0.25) (-1.47) (-1.97*) Constant Coef. t-statistics 0.0051 (0.76) 0.0070 (0.62) 0.0080 (0.93) 𝑅2 0.0154 0.0190 0.0241 N 127 127 127
*,**,*** denote 10%, 5%, 1% significance level respectively; Coef. Stands for coefficient. In this table, it presents the results of impact of factors (independent variables) on equal-weighted CARs (dependent variables) in 3 different intervals where the CARs are significantly positive at 10% significance level [(CAR(0,+1), CAR(0,+3), CAR(+1,+3)] respectively.
5.4.2 Partial Anticipation Issues Check
29
30
Table 11 — The Effect of Politically Connected Boards on Firms’ Cumulative Abnormal Returns (CARs).
Classification Number of Observation Value Weighted
CAR (0, +1) Total Announcement 129 0.0037 (1.736**) Democrat (Ⅰ) Republican (Ⅱ) Difference (Ⅱ-Ⅰ) Panel A: 76 53 Party of Nominee 0.0016 (1.122) 0.0067 (1.252**) 0.0051 (1.173)
Panel B: Former Position in Congress Senate (Ⅲ) House of Representatives (Ⅳ) Difference (Ⅳ-Ⅲ) 42 86 0.0047 (1.201) 0.0034 (1.292*) -0.0013 (-0.282) Panel C: Industry Exposure Non-Exposure (Ⅴ) Exposure (Ⅵ) Difference (Ⅵ-Ⅴ) 71 58 0.0005 (0.743) 0.0077 (1.624**) 0.0072 (1.684**) Panel D: Relatedness Related (Ⅶ) Unrelated (Ⅷ) Difference (Ⅷ-Ⅶ) 54 75 0.0071 (1.350**) 0.0013 (1.002) 0.0058 (1.340*)
31
politicians and the board of directors in the event window (0, +1), which is calculated based on two the value-weighted. The estimation window is from 120 to 30 days prior to the event window, and the event windows are chosen three days before and after the announcement date. The announcements are classified into four sub-sections: panel A is based on the party of the nominee, panel B on the former position in the Congress of the nominated boards, panel C on whether the companies that made the nomination announcements come from the following industries: Financials, Energy, Utilities, and Industrials according to General Industry Classification (GIS), and panel D on whether the nominated former politician whose former political duties are related to the operating activities of the companies where they are appointed as the board of directors.
In panel C, companies are sorted into two sub-groups based on the general industry classification (GIC). Companies in some industries, such as Banking and Finance, Energy, Utilities and (Military) Industrials, have a higher likelihood of exposure to government dominant projects and contracts, therefore, they might benefit more than those from otherwise industries. The results in panel C prove that politically connected firms with industries exposure are more likely to gain positive CARs, and such positive CARs are significantly higher than politically connected firms without industries exposure. Similarly, companies with boards whose previous political duties are related to companies' operating activities are more likely to benefit from political connections. Panel D illustrates the results about the relatedness of former political duties of nominees and the sectors in which the companies operate. The significantly positive CARs of companies with related boards and significant differences between related and unrelated companies indicate that the political connections are more beneficial to companies with a board of directors who served in related sectors as the companies operate in.
5.4.3 Endogeneity test
32
Table 12— First-stage results of the 2SLS regression and OLS Financial
Crisis
Political Party Market Value
2SLS CAR(0, +1) OLS CAR(0, +1) Difference 0.0099 0.0069 0.0030 -0.0016 -0.0059 -0.0043 -2.93e-07 -5.11e-08 -0.0044 N 129 129 129
*,**,*** denote 10%, 5%, 1% significance level respectively. In this table, it presents the results of first-stage results of the 2SLS regression and OLS. The dependent variable is CAR(0, +1) based on value-weight. The independent variables are financial crisis, political party and market value.
5.5.4 Total ESG Score
The robustness check for social and governance performance analysis between politically connected firms and unconnected firms is conducted via comparing differences in total ESG score. The results (presented in table 13) show that none of the means is significant, which means that throughout from 2010 to 2016, the overall ESG performance of politically connected firms and that of unconnected firms is not different. The results remain unchanged compared to the results of the synthetic score.
Table 13— Robust analysis of the Difference in the Social and Governance Performance Between Politically Connected Firms and Non-connected Firms between 2010 and 2016.
N Mean Std. Err. T Statistics
2010 82 -0.1943 3.6536 -0.053 2011 82 0.8729 3.4147 0.256 2012 82 0.3290 3.6122 0.091 2013 82 1.397 3.5356 0.395 2014 82 -0.0552 2.8684 -0.192 2015 82 0.3623 2.2247 0.163 2016 82 0.6965 1.5831 0.440
*,**,*** denote 10%, 5%, 1% significance level respectively; Std. Err. stands for standard error mean. In this table, it shows the difference in social and governance performance of politically connected firms and non-connected firms based on the total ESG score.
33
Companies benefit from political connections, meanwhile, they need to pay for these connections directly or indirectly according to the social exchange theory (Emerson, 1976), which leaves controversies and questions about the net effect of political connections on companies.
34
2016). Given these, we can assert that politically connected firms in the United States do not use political connections as a tool to lessen the social responsibilities that they are supposed to take, and such connections help the firms to have a better performance in corporate governance.
This paper supplements the previous literature on the topic of political connections. Compared to the extensive studies which focus on developing countries, we shift our horizon to a developed country, the United States. Moreover, our study combines political connections with financial as well as social and governance performance of connected firms, which is relatively new to the existing studies. Next to that, the findings from this study have implications on corporate strategies for companies in the US. More importantly, the insufficiency and flaw of our study could point out the possibilities and directions for future researches. One insufficiency of this paper which cannot be ignored is that the methodology we use is based on the assumption that there is no causality between financial performance as well as social and governance performance. But in fact, it is not the case. The previous study suggests that good financial performance can be a result of good social and governance performance (Preston and O’Bannan, 1997). The methodology applied in this paper fails to distinguish whether the increasing value to the politically connected firms is a consequence of political connections or if it is just because of the improved performance of corporate governance. The second flaw is the assumption about the missing data of ESG. We fill the missing data with the information from next year only if there is no particular news or announcement disclosed that their performances in those two years were different. This assumption will not be held and there will be data bias in our study if the information was concealed or missed. Last, as we mentioned before, we attribute the insignificant difference in social performance between companies from the S&P 500 to the requirements by the stock exchange commission. The suggestions for future researches on this topic are: 1) private companies that are not subjective to the stock exchange commission requirements can be selected in the data; 2) it is optimal to develop and apply a model that can mitigate the causality problem between financial performance and social performance.
7. Reference
35
Aggarwal, R., Meschke, F., Wang, T., 2012. Corporate political donations: investment or agency? Business and Politics 14, 1-38.
Barney, J., Wright, M., Ketchen, D. J., 2001. The resource-based view of the firm: ten years after 1991. Journal of Management 27, 625-641.
Bénabou, R., Tirole, J., 2010. Individual and corporate social responsibility. Economica 77, 1-19.
Bird, A., Buchanan, R., Rogers, P., 2004. The seven habits of an effective board European Business Journal 16, 128-32.
Bliss, M. A., Gul, F. A., 2012. Political connection and cost of debt: some Malaysian evidence. Journal of Banking & Finance 36, 1520-1527.
Borisov, A., Goldman, E., Gupta, N., (2016). The corporate value of (corrupt) lobbying. The Review of Financial Studies 29, 1039–1071.
Brooks, C. 2014. Introductory econometrics for finance. Cambridge university press.
Charoensukmongkol, P., 2016. The interconnections between bribery, political network, the government supports, and their consequences on export performance of small and medium enterprises in Thailand. Journal of International Entrepreneurship 14, 259-276.
Chatterji, A. K., Levine, D. I. and Toffel, M. W., 2009. How well do social ratings actually measure corporate social responsibility?. Journal of Economics & Management Strategy 18, 125-169.
Chen, C. J. P., Ding, Y., Kim, C. F., 2010. High-level politically connected firms, corruption, and analyst forecast accuracy around the world. Journal of International Business Studies 41, 1505-1524.
Chen, H. , Parsley, D. and Yang, Y., 2015. Corporate lobbying and firm performance. Journal of Business Finance & Accounting, 42, 444-481.
Desai, M. A., Dharmapala, D., 2009. Corporate tax avoidance and firm value. Review of Economics and Statistics 91, 537-546.
Dinç, I. S., 2005. Politicians and banks: political influences on government-owned banks in emerging markets. Journal of Financial Economics 77, 453-479.
Doh, J. P., Rodriguez, P., Uhlenbruck, K., 2003. Coping with corruption in foreign markets. Academy of Management Perspectives 17, 114-127.
36
Emerson, R. M., 1976. Social exchange theory. Annual Review of Sociology 2, 335-362.
Faccio, M., 2006. Politically connected firms. American Economic Review 96, 369-386. Faccio, M., Masulis, R. W., McConnell, J. J., 2006. Political connections and corporate bailouts. The Journal of Finance 61, 2597-2635.
Faccio, M., Parsley, D. C., 2006. Sudden deaths: taking stock of political connections. CEPR Discussion Paper No. 5460. Available at SSRN: https://ssrn.com/abstract=899378. Faccio, M., Parsley, D. C., 2009. Taking stock of geographic ties. Journal of Financial and Quantitative 4, 683-718.
Fama, E. F., French, K. R., 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics 33, 3-56.
Fan, J. P. H., Wong, T. J., Zhang, T., 2007. Politically connected CEOs, corporate governance, and Post-IPO performance of China’s newly partially privatized firms. Journal of financial economics 84, 330-357.
Ferrell, A., Liang, H., Renneboog, L., 2016. Socially responsible firms. Journal of Financial Economics 122, 585-606.
Fisman, R., 2001. Estimating the value of political connections. American Economic Review 91, 1095-1102.
Fisman, R., Wang, Y., 2015. The mortality cost of political connections. The Review of Economic Studies 82, 1346-1382.
Goldman, E., Rocholl, J., So, J., 2009. Do politically connected boards affect firm values? The Review of Financial Studies 22, 2331-2360.
Green, C. P., Homroy, S., 2018. Bringing connections on board: the value of political influence. Available at SSRN: https://ssrn.com/abstract=3279041.
Gu, X., Hasan, I., Zhu, Y., 2019. Political influence and financial flexibility: evidence from China. Journal of Banking and Finance 99, 142-156.
Guenther, D. A., Matsunaga, S. R., Williams B. M., 2017. Is tax avoidance related to firm risk? The Accounting Review 92, 115-136.
Hellman, J. S., Jones, G., Kaufmann, D., 2003. Seize the state, seize the day: state capture and influence in transition economies. Journal of Comparative Economics 31, 751-773.
37
Hill, M. D., Kelly, G. W., Lockhart, G. B, Ness, R. A., 2013. Determinants and effects of corporate lobbying. Financial Management, 42, 931-957.
Hillman, A. J., 2005. Politicians on the board of directors: do connections affect the bottom line? Journal of Management, 31, 464–481.
Hillman, A. J., Zardkoohi, A., Bierman, L., 1999. Corporate political strategies and firm performance: indications of firm-specific benefits from personal service in the U.S. government. Strategic Management 20, 61-87.
Huang, H., Zhao, Z., 2016. The influence of political connection on corporate social responsibility—evidence from listed private companies in China. International Journal of Corporate Social Responsibility 1, 1-19.
Johnson, S., Kaufmann, D., McMillan, J., 2000. Why do firms hide? Bribes and unofficial activity after communism. Journal of Public Economics 76, 495-520.
Kitzmueller, M., Shimshack, J., 2012. Economic perspectives on corporate social responsibility. Journal of Economic Literature 50, 51-84.
Kothari, S. P., Warner, Jerold B, 2007. Chapter 1 - econometrics of event. Handbooks in Finance, Handbook of Empirical Corporate Finance, 3-36.
Liston-Heyes, C., Ceton, G.C., 2007. Corporate social performance and politics. Journal of Corporate Citizenship 25, 95-108.
MacKinlay, A. Craig., 1997. Event studies in economics and finance. Journal of Economic Literature 35, 13-19.
Malkiel, B. G., Fama, E. F., 1970. Efficient capital markets: a review of theory and empirical work. The Journal of Finance 25, 383-417.
Mobarak, A. M., Purbasari. D. P., 2006. Corrupt protection for sale to firms: evidence from Indonesia. Unpublished working paper. University of Colorado at Boulder and Gadjah Mada University.
Oliver, C., Holzinger, I., 2008. The effectiveness of strategic political management: a dynamic capabilities framework. Academy of Management Review 33, 1-68.
Ordeshook, P., 1969. A theory of political exchange: economic reasoning in political analysis. By R. L. CurryJr., and L. L. Wadf. (Englewood Cliffs: Prentice-Hall, Inc., 1968. pp. 130.). American Political Science Review 63, 1294-1296.
38
Pfeffer, J., Salancik, G., 1978. The external control of organizations: a resource dependence perspective. New York: Harper & Row.
Preston, L. E., O’Bannan, D. P., 1997. The corporate social-financial performance relationship: a typology and analysis. Business & Society 36, 419-429.
Qu, T., Harris, R., 2019. Does support from the government help firms survive? Evidence on financial and political assistance in China, 1998-2007. Journal of Applied Economics 51, 528-541.
Richter, B. K., Samphantharak, K. and Timmons, J. F., 2009. Lobbying and taxes. American Journal of Political Science 53, 893-909.
Sánchez, C. M., 2000. Motives for corporate philanthropy in EI Salvador: altruism and political legitimacy. Journal of Business Ethics 27, 363-375.
Shen, C. H., Lin, Y. C., Wang, Y. C., 2015. Do strong corporate governance firms still require political connections and vice versa? International Review of Economics & Finance 39, 107-120.
Schipper, K., Thompson, R., 1983. Evidence on the capitalized value of merger activity for acquiring firms. Journal of Financial Economics 11, 85-119.
Shleifer, A. ,1998. State versus Private Ownership. Journal of Economic Perspectives 12, 133-150.
Shon, J., 2006. Do stock returns vary with campaign contributions? The Bush versus Gore 2000 presidential elections. Working paper, City University of New York.
Sim, G.S., 2005. A political theory of corporate social responsibility. Miller Center for Public Affairs. Available at: www.americanpoliticaldevelopment.org.
Tian, L., Estrin, S., 2008. Retained state shareholding in Chinese PLCs: does government ownership always reduce corporate value? Journal of Comparative Economics 36, 74-89.
Wang, Q., Wu, S., 2008. Political connection’s impact on company performance—based on China’s listed companies’ political influence index. International Symposium on Empirical Accounting in China.
Wooldridge, J., 2009. Introductory Econometrics: A Modern Approach (4th Ed.). Australia, South-Western 88.
39
Wu, W., Wu, C., Zhou, C., Wu, J., 2012. Political connections, tax benefits and firm performance: evidence from China. Journal of Accounting and Public Policy 31, 277-300. Yi, F., Xu, E., 2014. The relationship between the political connection of top managers and corporate social responsibility—an empirical study of Chinese listed companies. Research on Economics and Management 5, 5-13.
Yu, F., Yu, X., 2011. Corporate lobbying and fraud detection. Journal of Financial and Quantitative Analysis 46, 1685-1891.
Zhang, C., 2017. Political connections and corporate environmental responsibility: adopting or escaping? Energy Economics 68, 539-547.
Zhou, J. Q., Peng, M. W., 2012. Does bribery help or hurt firm growth around the world? Asia Pacific Journal of Management 29, 907-921.
8. Appendix
Appendix 1: Distribution of Different Classification of Former US Congress Member Between 1939 and 2009.
Appendix 2: Timeline of Event Study. 44 23 39 29 1 0 5 10 15 20 25 30 35 40 45 50
Distribution of former US Congress Member (1939-2009)
40
Appendix 3: Logic Tree of Hypothesis 2.
41
In this graph, it compares the performance of social and corporate governance between politically connected firms and unconnected firms based on the total ESG score, and shows how they change over time.
Appendix 5: Information of Data Used in This Paper. Data Name Period Frequency Source
Stock Price of S&P 500 2003-- 2019 Daily DataStream Market Index of S&P 500 2003-- 2019 Daily DataStream Boards Announcement 2000--2019 Daily WRDS US Former Politician 1930--2010 NA https://www.govtrack.us/congress/members/all Market Value 2003--2019 Yearly DataStream Three Factors of Fama French 2003--2018 Daily WRDS Financial Transparency 2010--2016 Yearly DataStream Fair Competition 2010--2016 Yearly DataStream Bribery and Corruption Training 2010--2016 Yearly DataStream 65 70 75 80 85 90 2010 2011 2012 2013 2014 2015 2016
Average Total ESG Score
42 Corporate Governance 2010--2016 Yearly DataStream Equal Weight Rating of FESG (Total ESG) 2010--2016 Yearly DataStream
Appendix 6: Average Abnormal Returns of Politically Connected Firms With an Event Window (-10, +10).
Appendix 7: T Statistics of Abnormal Returns (ARs) With an Event Window (-3,+3).
Event Window
N Mean Std. Err. T Statistics
t =-3 129 -0,0007 0,00128 -0,529 t =-2 129 -0,0020 0,00135 -1,457* t =-1 129 0,0000 0,00131 0,007 t = 0 129 0,0004 0,00135 0,301 t =+1 129 0,0033 0,00165 2,005** t =+2 129 -0,0013 0,00161 -0,803 t =+3 129 0,0023 0,00114 1,986**
*,**,*** denote 10%, 5%, 1% significance level respectively; Std. Err. stands for standard error mean. In this table, it presents the results of stock reaction (ARs) to the nomination between US former politicians and the board of director in a company. The estimation window is from 120 to 30 days prior to the event window, and the event windows are chosen three days before and after the announcement date.
