Introduction
Political ideology is normally defined as an interrelated set of attitudes, beliefs, opinions, and values about the proper goals of society (Tedin, Reference Tedin1987; Jost, Reference Jost2021). As political polarization intensifies in the United States, partisanship has also become one of Americans’ most salient social identities, significantly influencing individuals’ economic decision-making. However, the relationship between the partisanship of top management team (TMT) and corporate culture remains insufficiently explored. Corporate culture constitutes a system of beliefs, norms, and values that are widely shared and deeply ingrained throughout the organization. It complements formal governance mechanisms, which are codified in forms of rules and regulations (e.g., Cremer, Reference Crémer1993; O’Reilly and Chatman, Reference O’Reilly and Chatman1996; Li et al., Reference Li, Mai, Shen and Yan2021). Resent empirical evidence, drawn both from field studies and interviews, underscores the substantial impacts of culture on business decisions, ethical practices, and firm value (e.g., Guiso, Sapienza, and Zingales, Reference Guiso, Sapienza and Zingales2015; Jiang et al., Reference Jiang, Kim, Ma, Nofsinger and Shi2019; Dupont and Karpoff, Reference Dupont and Karpoff2020; Nadarajah, Atif, and Gull, Reference Nadarajah, Atif and Gull2022; Graham et al., Reference Graham, Grennan, Harvey and Rajgopal2022a, Reference Graham, Grennan, Harvey, Campbell and S. Rajgopal2023b; Lei et al., Reference Lei, Wang, Yu and Chan2022). These findings highlight the importance for further understanding what shapes corporate culture.
Previous research indicates that organizational cultures are largely created by and reflect the values, beliefs, and actions of the senior leaders (e.g., Baron and Hannan, Reference Baron and Hannan2002; Davis, Reference Davis1984; Kotter and Heskett, Reference Kotter and Heskett1992; Schein, Reference Schein1985). In the seminal book, Organizational Culture and Leadership, Schein (Reference Schein1985) asserts that the only thing of real importance that leaders do is to create and manage culture. When culture is viewed as a consensus on norms, the recurring behaviors of leadership serve as an informative source about the normative order within the organization (Bandura, Reference Bandura1986; Carroll and Harrison, Reference Carroll and Harrison1998). In this sense, the fundamental values and personalities of leaders may be the primary building blocks of organizational culture (Detert et al., Reference Detert, Schroeder and Mauriel2000; Fu et al., Reference Fu, Tsui, Liu and Li2010).
One important and pervasive trait of corporate leaders is their political ideology, which significantly affects the workplace dynamics (e.g., Johnson and Roberto, Reference Johnson and Roberto2018). Research in political science and political psychology posits that partisanship has evolved into a social identity (e.g., Bankert, Huddy, and Rosema, Reference Bankert, Huddy and Rosema2017; West and Iyengar, Reference West and Iyengar2022). Political ideology gives rise to divergent viewpoints and impact a wide array of social and economic behaviors exhibited by individuals (e.g., Wang, Devine, and Molina-Sieiro, Reference Wang, Devine and Molina-Sieiro2021). Corporate leaders also inject their partisan perceptions into the workplace, as suggested by the upper echelons theory (Hambrick and Mason, Reference Hambrick and Mason1984). For instance, Aiken, Ellis, and Kang (Reference Aiken, Ellis and Kang2020) show that Democratic politicians engage in more socially responsible investing compared to their Republican counterparts. Furthermore, mutual fund, hedge fund, and state pension fund managers, who lean toward the Democratic Party, tend to allocate their portfolios toward companies with robust corporate social responsibility (CSR) practices (Hong and Kostovetsky, Reference Hong and Kostovetsky2012; Hoepner and Schopohl, Reference Hoepner and Schopohl2020). At the firm level, Di Giuli and Kostovetsky (Reference Di Giuli and Kostovetsky2014) show that firms led by Democratic-leaning CEOs, founders and directors allocate greater resources on CSR activities and have higher CSR ratings than firms leaning toward the Republican Party. Conversely, Republican TMT also adopt more conservative corporate policies (Hutton, Jiang, and Kumar, Reference Hutton, Jiang and Kumar2014). Oc, Netchaeva, and Kouchaki (Reference Oc, Netchaeva and Kouchaki2021) find that conservative firm decision-makers exhibit gender bias, providing less positive descriptions of leadership positions for female candidates compared to male candidates. Therefore, we posit that corporate culture can be related to the partisan values of TMT, which we know of no studies have empirically explored this relationship.
We focus on specific dimensions of cultural values—namely, integrity, teamwork, innovation, respect, and quality, due to their prominence as the top cultural attributes advertised by firms (Guiso, Sapienza, and Zingales, Reference Guiso, Sapienza and Zingales2015).Footnote 1 Executives with Democratic (liberal) and Republican (conservative) perceptions might be associated with different corporate culture within U.S. firms for the following reasons. First, liberals prioritize equality and progress over conservatives (Jost, Reference Jost2006, Reference Jost2017). The Democratic Party advocates for social equality rather than hierarchy, which would foster collaboration and respect in the workplace. Moreover, this pro-social ideology can positively impact the quality of products and services delivered by firms. Second, liberals exhibit greater openness to change and are more comfortable with uncertainty, whereas conservatives value tradition and stability (Swigart et al. Reference Swigart, Anantharaman, Williamson and Grandey2020). To this extent, firms led by Democratic-leaning executives are expected to demonstrate higher levels of innovation compared to those with Republican-leaning executives. Third, although there is no direct link between partisanship and integrity, some evidence suggests that Democrats prioritize morality and ethics more than Republicans. A recent Pew Research Center survey reveals that Democrats are significantly more likely than Republicans to say that it is “very” important to have a president who personally lives a moral and ethical life (71% vs. 53%).Footnote 2 Therefore, we anticipate a stronger cultural emphasis on integrity within firms led by Democratic-leaning executives.
To test these hypotheses, we use a sample comprising 26,027 firm-year observations across 2,424 unique S&P 1500 firms spanning the period from 2001 to 2018. We obtain time-varying measures of cultural values for U.S. firms from Li et al. (Reference Li, Mai, Shen and Yan2021), which encompass the five dimensions of corporate culture, integrity, teamwork, innovation, respect, and quality. To gauge the political partisanship of firm executives, we follow Jiao and Ren (Reference Jiao and Ren2024) and calculate the tilt in executives’ historical individual political donations to either the Democratic Party or the Republican Party. Our findings reveal that firms led by Democratic-leaning executives exhibit significantly higher scores across all culture values. Furthermore, we explore the effect of corporate governance on the relationship between partisanship and corporate culture. Specifically, we observe that this relationship weakens when firms possess less entrenched management and locate in states characterized by stronger judicially accountable. Our results remain robust after employing the propensity score matching (PSM) method.
Our study makes a two-fold contribution to the existing literature. First, we contribute to the literature on organizational and corporate culture, demonstrating that political partisanship of senior leaders can be a key determinant of corporate culture. The personal traits exhibited by senior leaders play a pivotal role in molding organizational culture (e.g., Wilderom, van den Berg, and Wiersma, Reference Wilderom, van den Berg and Wiersma2012; O’Reilly et al., Reference O’Reilly, Caldwell, Chatman and Doerr2014; Devine, Holmes, and Wang, Reference Devine, Holmes and Wang2021). The past experience of senior leaders leaves indelible mark on how organizational culture is developed (e.g., Bernile, Bhagwat, and Rau, Reference Bernile, Bhagwat and Rau2017; Koch-Bayram and Wernicke, Reference Koch-Bayram and Wernicke2018; Navajas et al., Reference Navajas, Paula A., Rossi and Vazquez2022). In addition, the cultural heritage stemming from the ancestry country of senior managers can also exert an influence on corporate culture (e.g., Liu, Reference Liu2016; Pan, Siegel, and Wang, Reference Pan, Siegel and Wang2020). To the best of our knowledge, our study provides the first empirical evidence linking political partisanship to corporate culture across multiple dimensions.
Second, our study contributes to a growing literature that documents the economic and social outcomes driven by political partisanship. Democrats and Republicans diverge significantly in their liberalism and conservatism, leading to contrasting norms and perspectives (e.g., Hutton, Jiang, and Kumar, Reference Hutton, Jiang and Kumar2014; Jost, Reference Jost2017; Swigart et al., Reference Swigart, Anantharaman, Williamson and Grandey2020). These divergent ideologies wield substantial influence over individuals’ economic decision-making (e.g., Di Giuli and Kostovetsky, Reference Di Giuli and Kostovetsky2014; Gupta, Nadkarini, and Mariam, Reference Gupta, Nadkarini and Mariam2019; Pecot, Vasilopoulou, and Cavallaro, Reference Pecot, Vasilopoulou and Cavallaro2021; Fernandes et al., Reference Fernandes, Ordabayeva, Han, Jung and Mittal2022; Gohary et al., Reference Gohary, Madani, Chan and Tavallaei2023). At the firm level, partisanship and shared partisan views affect a wide range of corporate policies and stakeholder behaviors (e.g., Antonetti and Anesa, Reference Antonetti and Anesa2017; Northey and Chan, Reference Northey and Chan2020; Chow et al., Reference Chow, Louca, Christodoulos, Andreas and Procopiou2022, Reference Chow, Petrou and Procopiou2023; Ketron, Kwaramba, and Williams, Reference Ketron, Kwaramba and Williams2022; Jiao and Ren, Reference Jiao and Ren2024). Our study extends this line of literature by establishing a link between political partisanship and corporate cultural values.
The remainder of this paper is organized as follows. We begin by describing existing research on organizational and corporate culture, as well as the influence of political partisanship on firm decisions. Next, we develop our empirical hypothesis. In the methods section, we detail our data sources and sample construction process. Subsequently, we empirically investigate the relationship between the political partisanship of firm executives and various cultural values. The final section draws conclusions.
Literature review and hypotheses development
Senior leaders and corporate culture
Researchers have defined culture as a social control system of shared values and norms among organizational members (Schein, Reference Schein1985; O’Reilly and Chatman, Reference O’Reilly and Chatman1996; Chatman et al., Reference Chatman, Caldwell, O’Reilly and Doerr2014). This social control system arises, in part, from the personality, preferences, and actions of organizational leaders (O’Neill and O’Reilly, Reference O’Neill and O’Reilly2009; O’Reilly et al., Reference O’Reilly, Caldwell, Chatman and Doerr2014; O’Reilly, Chatman, and Doerr, Reference O’Reilly, Chatman and B. Doerr2021). Leaders can signal employees about the norms and values through their behaviors, communications, and managerial decisions, thereby shaping organizational culture (Bandura, Reference Bandura1986; Carroll and Harrison, Reference Carroll and Harrison1998).
Recent studies provide empirical evidence supporting the impact of firm leaders on corporate culture and cultural-related activities. For example, Peterson et al. (Reference Peterson, Smith, Matorana and Owens2003) and Giberson et al. (Reference Giberson, Resick, Dickson, Mitchelson, Randall and Clark2009) show that firms led by more aggregable and extraversive leaders tend to have more cohesive cultures. Relatedly, O’Reilly et al. (Reference O’Reilly, Caldwell, Chatman and Doerr2014) link five dimensions of CEO personality attributes to various types of organizational culture. Berson, Oreg, and Dvir (Reference Berson, Oreg and Dvir2008) show that CEO self-directive values are associated with innovation-oriented cultures, security values correlate with bureaucratic cultures, and benevolence values are related to supportive cultures. Additionally, O’Reilly, Chatman, and Doerr (Reference O’Reilly, Chatman and B. Doerr2021) find that narcissistic leaders are more likely to create cultures that undermine collaboration and integrity. Furthermore, Dupont and Karpoff (Reference Dupont and Karpoff2020) provide a comprehensive review on the measures used to reflect cultural values and explore the impact of culture in the finance literature.
Political partisanship and corporate culture: hypothesis development
The topic of political partisanship has increasingly captured the interest of academic scholars, particularly in light of the heightened level of partisan conflict in the U.S. society (e.g., Mason, Reference Mason2013, Reference Mason2015). Political partisanship represents a schema of related values or beliefs, wherein certain behaviors are deemed preferable over others, and it has evolved into a social identity for individuals (e.g., Rokeach, Reference Rokeach1973; Jost, Reference Jost2006; Bankert, Huddy, and Rosema, Reference Bankert, Huddy and Rosema2017; West and Iyengar, Reference West and Iyengar2022). In the United States, the Democratic Party and the Republican Party are the two major political parties that has dominated the U.S. politics since the World War II. The Democratic Party’s platform is rooted in American liberalism, advocating for a range of issues including social programs, labor unions, consumer protection, workplace safety, disability rights, racial equality, environmental protection, social justices, humanity rights, immigration reform, and economic equality (e.g., Allen and Scruggs, Reference Allen and Scruggs2004; Alvarez, Garrett, and Lange, Reference Alvarez, Garrett and Lange1991; Engelberg et al., Reference Engelberg, Henriksson, Manela and Williams2023; Gilens, Reference Gilens2023; Hout and Maggio, Reference Hout and Maggio2021). In contrast, the Republican Party’s platform is based on American conservatism, emphasizing support for free market, deregulation, individualism, traditional values, obedience to authority, and varying economic rewards (e.g., Calhoun, Reference Calhoun2011; Goren, Reference Goren2005; Keller and Kelly, Reference Keller and Kelly2015; Iyengar et al., Reference Iyengar, Lelkes, Levendusky, Malhotra and Westwood2019). The ideological divide between conservatism and liberalism serves as a defining trait that distinguishes Republicans from Democrats in their beliefs and perspectives. According to a 2019 Pew Research survey, approximately half of Democratic and Democratic-leaning registered voters who describe their own political views as liberal.Footnote 3 A 2009 Gallup survey also show that 73% of Republicans self-identify as conservative.Footnote 4
A large strand of empirical literature examines the impact of political partisanship on individuals’ economic and social behaviors (e.g., Kaustia and Torstila, Reference Kaustia and Torstila2011; Claassen and Ensley, Reference Claassen and Ensley2016; Iyengar and Westwood, Reference Iyengar and Westwood2015; Gentzkow Reference Gentzkow2016; Huber and Malhotra, Reference Huber and Malhotra2017; McConnell et al., Reference McConnell, Margalit, Melhotra and Levendusky2018; Northey and Chan, Reference Northey and Chan2020; Pecot, Vasilopoulou, and Cavallaro, Reference Pecot, Vasilopoulou and Cavallaro2021; Gohary et al., Reference Gohary, Madani, Chan and Tavallaei2023; Mian, Sufi, and Khoshkhou, Reference Mian, Sufi and Khoshkhou2023). In the context of firms, researchers also demonstrate that partisan values of firm leaders influence various firm policies and practices. For example, Cohen et al. (Reference Cohen, Hazan, Tallarita and Weiss2019) find that firms led by Republican CEOs tend to be less transparent to investors. Di Giuli and Kostovetsky (Reference Di Giuli and Kostovetsky2014) show that firms with Democratic-leaning leaders spend more on CSR activities and have a higher CSR score than other firms. Hutton, Jiang, and Kumar (Reference Hutton, Jiang and Kumar2014) find that firms with Republican-leaning executives have more conservative financial policies. Hutton, Jiang, and Kumar (Reference Hutton, Jiang and Kumar2015) show that firms with a Republican culture are more likely to be the subject of civil rights, labor, and environmental litigation than Democratic firms. However, firms with a Democratic culture are more likely to be the subject of securities fraud and intellectual property rights violations than are Republican firms. Weng et al. (Reference Weng, Chuang, Zhang and Church2022) find that liberal CEOs are more likely to implement LGBT-friendly policies. Gupta, Nadkarini, and Mariam (Reference Gupta, Nadkarini and Mariam2019) show that Democratic-leaning CEOs are more likely to enact CSR practices, and Republican-leaning CEOs are more likely to engage in downsizing. Research has not explicitly examined the role of political partisanship in influencing broader organizational norms. In this study, we delve into the relationship between partisanship of firm leaders and corporate culture.
We propose that political values of the Democratic and the Republican Party lead to different ideologies among their supportive members, sequentially shaping divergent cultural values within the workplaces where these members are in charge. A key distinguishing feature between the two political orientations is that conservatism reflects a rightist belief system focusing on hierarchy and tradition, while liberalism reflects a leftist ideology prioritizing equality and progress (Jost, Reference Jost2017). Political conservatism is conceptualized composed of two core components: resistance to change and opposition to equality (Conover and Feldman, Reference Conover and Feldman1981; Jost et al. Reference Jost, Napier, Thorisdottir, Gosling, Palfai and Ostafin2007). Glasgow and Cartier (Reference Glasgow, Cartier and Wilson1985) show that individuals with a conservative ideology prefer familiar versus unfamiliar stimuli. Atieh, Brief, and Vollrath (Reference Atieh, Brief and Vollrath1987) find that conservative individuals prefer job security to task variety. Wilson (Reference Wilson and Wilson1973) show that conservative individuals exhibit greater aversion to ambiguity, uncertainty, and complexity and are more sensitive to the possibility of a loss. Swigart et al. (Reference Swigart, Anantharaman, Williamson and Grandey2020) also find that liberals are more open to change and more comfortable with uncertainty, but conservatives value tradition and stability.
Given these contrasting views on the openness to change and equality, firms led by Democratic-leaning TMT are expected to foster more innovative and collaborative cultural values in the workplace compared to firms led by Republican-leaning executives. These conjectures are consistent with the recent survey evidence showing that Democrats are more likely than Republicans to express positive views regarding the importance and impact of racial and ethnic diversity than Republicans and that Democrats are more likely than Republicans to support increased federal spending for scientific research.Footnote 5
Furthermore, we expect firms with Democratic-leaning executives to be more likely to exhibit cultural values related to product quality. The inherent features of resistance to change and opposition to equality may hinder firms from enhancing their products and services to meet customer needs. Indirect evidence suggests that conservatives tend to be more satisfied with the products and services they consume than liberals because conservatives trust their own decisions (Fernandes et al., Reference Fernandes, Ordabayeva, Han, Jung and Mittal2022). From the perspective of producers, conservative leaders may also find satisfaction in their existing products and opt not to invest further efforts in quality improvement.
The cultural value of respect can correlate with fostering fairness and embracing diversity within employee relationships, aligning closely with the democratic ideology that champions the promotion of equality. In this context, firms led by Democratic-leaning TMTs are anticipated to prioritize and demonstrate greater respect toward their shareholders and stakeholders. However, it is worth noting that the respectful value may also be associated with concerns related to obedience to authorities and traditions, a perspective more representative of conservatives (e.g., Graham, Nosek, and Haidt, Reference Graham, Nosek and Haidt2012; Kugler, Jost, and Moorbaloochi, Reference Kugler, Jost and Moorbaloochi2014).
The relationship between political partisanship of executives and the value of integrity remains relatively underexplored in the literature. On one hand, the conservative view implies a personality structure of conscientiousness, which tends to prevent individuals from engaging in inappropriate behavior and misconduct (Carney et al., Reference Carney, Jost, Gosling and Potter2008). In this case, Republican-leaning executives are often perceived as more orderly, conventional, and better organized, contributing to an honest corporate culture. On the other hand, the characteristics of individualism and obedience to authority may foster optimism but also amplify agency problems between top executives and shareholders. As a result, firms led by Republican-leaning executives may exhibit a weaker cultural emphasis on integrity. A relevant study by Hutton, Jiang, and Kumar (Reference Hutton, Jiang and Kumar2015) examines the role of political culture in corporate misconduct. They find that firms with a Republican culture are more likely to be the subject of civil rights, labor, and environmental litigation, while firm with a Democratic culture are more likely to be prone to securities fraud litigation and property rights violations. Overall, the authors observe a significant higher litigation propensity for all cases in Republican-leaning firms compared to Democratic-leaning firms. Anecdotal evidence also suggests that Democrats are more likely to emphasize the importance of having a president who personally lives a moral and ethical life.Footnote 6
All these discussions above lead to our formal hypothesis that:
Hypothesis 1: Firms with executives leaning toward the Democratic Party are more likely to exhibit stronger cultural values of innovation, collaboration, respect, quality, and integrity compared to those with executives leaning toward the Republican Party.
Methods
Data and sample
Our sample comprises S&P 1500 firms covered by the Execucomp database between 2001 and 2018. The measures of corporate cultural values are from Li et al. (Reference Li, Mai, Shen and Yan2021).Footnote 7 We collect data on political contributions made by executives from the Federal Election Commission (FEC) records, starting in 1979. Firm financial information is obtained from Capital IQ Compustat. After merging all datasets, our final sample consists of 26,027 firm-year observations for 2,424 U.S. S&P 1500 firms spanning the period from 2001 to 2018. Detailed variable definitions are provided in Appendix 1.
Corporate cultural values
We adopt the measures for corporate cultural values developed by Li et al. (Reference Li, Mai, Shen and Yan2021). Using machine learning techniques and analyzing earnings call transcripts, Li et al. (Reference Li, Mai, Shen and Yan2021) construct a culture dictionary and score the five corporate cultural values: innovation, integrity, quality, respect, and teamwork for U.S. publicly traded firms during the period from 2001 to 2018. As Guiso, Sapienza, and Zingales (Reference Guiso, Sapienza and Zingales2015) documented, these five dimensions of cultural values are identified as the most prominently advertised values by S&P 500 firms on their corporate websites.Footnote 8
Li et al. (Reference Li, Mai, Shen and Yan2021) employ the word2vec machine-learning algorithm, which utilizes a neural network model to learn word association from the analyst call transcripts. The model is trained to create an expanded, context-specific dictionary for measuring each of the five cultural values at the firm-fiscal year level. Appendix 2 provides the ten most representative words associated with each cultural value in the culture dictionary. We obtain the final datasets containing culture scores from the authors. We also construct an aggregated corporate culture score by summing up the scores across the five dimensions.
We use Li et al. (Reference Li, Mai, Shen and Yan2021)’s culture measures for several reasons. First, their measures are more likely to capture the prevailing cultural values of a firm, rather than relying solely on cultures advertised by the firm. Extracting cultural values from earnings calls transcripts minimizes the risk of “cheap talk” compared to values advertised on a firm’s website or its press releases. Furthermore, they focus on the extemporaneous question-and-answer (QA) section of earnings calls, where firm executives directly respond to questions and topic discussions. This approach helps mitigate excessive self-promotion and provide a more genuine reflection of culture value. Second, the NLP methodology determines the most relevant words and phrases associated with specific cultural values. This methodology also assigns lower weights to frequently occurring words, addressing the potential bias introduced by “stated value”. Lastly, earnings calls offer dynamic information, which allows us to measure time-varying corporate cultural values.
Political partisanship of TMT
The Execucomp dataset provides the biographical information on top executives of S&P 1500 firms. To identify the partisanship of executives, we collect individual political contribution data from the FEC bulk data, which includes the name, employer, occupation, and address of the contributor. Following the approach of Jiao and Ren (Reference Jiao and Ren2024), we match executives from Execucomp to FEC individual political contributions based on first name, last name, employer information, and zip code. We further identify the party affiliation of political committees and calculate the amount that corporate executives donate to each political party during each congressional cycle. For an executive in an election cycle, we compute the total donation amount to each of the two parties from 1979 (the 96th congress) to the current election cycle. We then measure the partisanship of each executive as a ratio of the difference in cumulative dollar amount to the Republican Party and the Democratic Party over the total cumulative contribution amount the executive has made. Importantly, our measure of partisanship for each executive avoids look-ahead bias by relying solely on historical political contribution information.
To construct a firm-year level measure of political partisanship index, TMT REP-DEM, we calculate the weighted average of partisanship of each executive based on their compensations. This weighting reflects their relative importance or ranking within the firm. The measure ranges from negative one to one, where a positive value indicates that the TMT leans toward the Republican Party, a negative value indicates that the TMT leans toward the Democratic Party, and zero indicates that the TMT is politically neutral or apolitical. In addition to the continuous measure for political partisanship at the firm-year level, we create two dummy variables, Republican Leaning and Democratic Leaning, denoting firms led by Republican-leaning executives and Democratic-leaning executives, respectively. Specifically, we define Republican-leaning firms as those with TMT political partisanship greater than zero and Democratic-leaning firms as those with TMT political partisanship less than zero.
Control variables
We include several firm-level variables that can affect corporate culture, drawing from prior literature (e.g., Hutton, Jiang, and Kumar, Reference Hutton, Jiang and Kumar2014; O’Reilly et al., Reference O’Reilly, Caldwell, Chatman and Doerr2014). Specifically, we control for Firm Size (defined as the natural logarithm of total assets), Leverage Ratio (defined as total debts divided by total assets), Tobin’s Q (defined as (book value of assets + (market value of equity—book value of equity))/book value of assets), ROA (defined as income before extraordinary items divided by total assets), Capex Ratio (defined as total capital expenditures divided by total assets). In addition, we include Firm PAC (defined as the natural logarithm of total firm PAC donations) and Lobbying (defined as the natural logarithm of total corporate lobbying expenditure). All firm-level variables are winsorized at the 1% and 99% percentiles.
Model specification
To test our hypothesis, we regress measures of corporate cultural values on political partisanship of a firm’s TMT. Specifically, we use the following model specification:
$$\eqalign{ & Culture\;Scor{e_{i,t}} = {\beta _0} + {\beta _1}TMT\;Partisanship\;Measure{s_{i,t}} + \;{\beta _2}Firm\;Siz{e_{i,t}} + \;{\beta _3}Leverage\;Rati{o_{i,t}} \cr
& \quad \quad \quad \quad \quad \quad \quad \quad + \;{\beta _4}Tobin's\;{Q_{i,t}} + {\beta _5}RO{A_{i,t}} + \;{\beta _6}Capex\;Rati{o_{i,t}} + \;FEs + {\varepsilon _{i,t}} \cr} $$
where i indexes firm and t indexes year. The dependent variable, Culture Score, denotes one of the corporate culture measures, Total Culture Score, Integrity Score, Teamwork Score, Innovation Score, Respect Score, and Quality Score, for a firm in a fiscal year. TMT REP-DEM denotes one of the measures for political partisanship of a firm’s TMT. We use both a continuous measure, TMT REP-DEM, and two indicators, Republican Leaning and Democratic Leaning, denoting firms with executives leaning toward the Republican Party and the Democratic Party, respectively. A negative (positive) estimated coefficient on the TMT REP-DEM variable suggest firms with Republican-leaning TMT tend to have a lower (higher) culture score. For the model specifications with indicators, a negative (positive) estimated coefficient on Republican Leaning suggests that firms with executives leaning toward the Republican Party have lower (higher) culture scores.
We consider several fixed effects, including industry fixed effect to control for the potential commonality in culture for firms from the same industry, state fixed effect to control for the commonality in culture for firms located in the same state due to similar external political, business, and social environment, year fixed effect to control for common effects on culture for observations from the same year, and firm fixed effect to control for the persistence in a firm’s culture over time. We report model specifications using different fixed effects combinations to gauge the robustness of our results.
Empirical results
Summary statistics
Table 1 presents summary statistics for firm-year level variables. Consistent with Li et al. (Reference Li, Mai, Shen and Yan2021), the mean values of Integrity Score, Teamwork Score, Innovation Score, Respect Score, and Quality Score are 0.53, 0.74, 1.85, 0.99, and 1.32, respectively. The mean value of TMT REP-DEM is 0.12, which aligns with Jiao and Ren (Reference Jiao and Ren2024), indicating that more firm executives lean toward to the Republican Party than the Democratic Party. Approximately 53.5% of sample observations are Republican-leaning and 26.8% are Democratic-leaning, which are consistent with the distribution observed in previous literature (e.g., Cohen et al., Reference Cohen, Hazan, Tallarita and Weiss2019; Fos, Kempf, and Tsoutsoura, Reference Fos, Kempf and Tsoutsoura2023; Jiao and Ren, Reference Jiao and Ren2024). Table 2 presents the correlation matrix for key variables. The pairwise correlations among the five culture scores are all positive and the magnitude are similar to those in Li et al. (Reference Li, Mai, Shen and Yan2021). The correlations between TMT REP-DEM and culture scores are all negative. Republican Leaning and culture scores are all negatively associated, and Democratic Leaning and culture scores are all positively associated. These correlations analyses suggest that firms with TMT leaning toward the Democratic Party tend to score higher on the five dimensions of corporate cultural values.
Table 1. Summary statistics
This table presents the descriptive statistics for our sample of 26,027 firm-year observations for 2,424 unique S&P 1500 firms from 2001 to 2018. All variable definitions are in the Appendix 1.
Table 2. Correlation matrix
This table presents the pairwise variable correlations.
Figure 1 visualizes the relationship between political partisanship and corporate culture over time. We display the time-series mean culture scores for the Republican-leaning firms (in red with triangular markers) and Democratic-leaning firms (in blue with circular markers). First, we observe an upward trend in all culture scores over our sample period, which is consistent with Li et al. (Reference Li, Mai, Shen and Yan2021). Second, for the total culture score and scores for each cultural dimensions, we observe that blue lines are almost always above red lines, suggesting that firms led by Democratic-leaning executives tend to possess higher corporate cultural scores throughout the whole sample period from 2001 to 2018.
Figure 1. Partisan divides of corporate cultural values. These plots present the time-series average cultural value scores for Republican-leaning (in red with triangular markers) and Democratic-leaning (in blue with circular markers) firms, separately.
Additionally, in Figure 2, we compare the average TMT REP-DEM for firms with high and low cultural scores. Specifically, we categorize firms into high or low score groups depending on whether they have cultural scores greater than the median value. The time-series average TMT REP-DEM value for high-score firms is in blue color with circular markers, and that for low-score firms is in red color with triangular markers. Consistent with Figure 1, we find that firms with high cultural scores consistently exhibit smaller TMT REP-DEM (indicating more Democratic-leaning) compared to firms with low cultural scores.
Figure 2. TMT REP-DEM by corporate cultural values. These plots present the time-series average TMT REP-DEM for firms with high cultural value (in blue with circular markers) and firms with low cultural value (in red with triangular markers), separately.
Both figures provide direct support for our hypothesis that firms with executives leaning toward the Democratic Party are more likely to exhibit cultural values of innovation, collaboration, respect, quality, and integrity compared to those with executives leaning toward the Republican Party.
Baseline regressions
Table 3 presents the baseline regressions of corporate cultural values on the political partisanship of firm TMT. We report results for the aggregated culture score in columns (1) and (2), integrity score in columns (3) and (4), teamwork score in columns (5) and (6), innovation score in columns (7) and (8), respect score in columns (9) and (10), and quality score in columns (11) and (12). The odd columns include industry, state, and year fixed effects, and the even columns include firm and year fixed effects with t-statistics in the parentheses and standard errors clustered at the firm level. We observe that the estimated coefficients on TMT REP-DEM are all negative and statistically significant for all model specifications except for column (9), which suggest that firms with TMT leaning toward the Democratic Party have higher scores on all five dimensions of cultural values than those with TMT leaning toward the Republican Party.
Table 3. Political partisanship and corporate culture: baseline results
This table reports the regression results of corporate culture scores on political partisanship of firm top executives and other control variables. All variable definitions are in the Appendix 1. Industry, year, and state fixed effects are included in odd columns. Firm and year fixed effects are included in even columns. Standard errors are clustered by firm in all specifications. Robust t-statistics are in the parentheses. ***, **, and * denote statistical significance at the 0.1, 1, and 5 percent levels, respectively.
The coefficient on Total Culture Score is −0.372 using firm and year fixed effects specification. In terms of economic significance, we find that when TMT REP-DEM increases by one standard deviation, the Total Culture Score of a firm decreases by 0.13, which is 2.4% at the mean. Similarly, the estimated coefficients on Integrity Score, Teamwork Score, Innovation Score, Respect Score, and Quality Score are −0.034, −0.048, −0.159, −0.6, and −0.071. A one standard deviation increase in TMT REP-DEM are associated with 2.3%, 2.3%, 3%, 2.1%, and 1.9% decreases at the mean value of Integrity Score, Teamwork Score, Innovation Score, Respect Score, and Quality Score, respectively.
It is worth noting that the partisan gaps on Integrity Score in the figures appear to be comparatively smaller. As discussed in the hypothesis development section, liberals tend to prioritize civil rights, labor, and environmental protection, whereas conservatives focus more on concerns related to securities fraud and property rights. In the construction of the integrity measure, accountability, ethic, integrity, responsibility, and transparency are the most representative terms within the machine learning dictionary. However, directly correlating these broad words that encompass integrity values to specific integrity issues poses a challenge, resulting in a weaker empirical connection between the political partisanship of executives and the cultural value of integrity.
Moreover, we observe a weaker correlation between TMT REP-DEM and the Respect Score. The cultural value of respect, as identified by Li et al. (Reference Li, Mai, Shen and Yan2021), appears to be more closely linked to aspects such as diversity, job satisfaction, fairness, and camaraderie, rather than mere obedience to authority. In line with our argument regarding the potential influences of partisanship, we find that firms led by Democratic-leaning executives have higher Respect Score. Despite a lower level of statistical significance, the association between having Democratic-leaning executives and a higher respect score remains robust. These results indicate that TMT REP-DEM holds some influence in fostering a culture of respect tied to diversity and employee satisfaction, and this discrepancy endures over time.
We further conduct two robustness tests. First, we use the inverse of executive compensation ranks as the weight in constructing the executive partisanship measure, and results are reported in Appendix 3. Second, we repeat the baseline regression analysis with the lagged partisanship measure, and results are reported in Appendix 4. Notably, the results remain qualitatively consistent.
Overall, the baseline regression results provide supporting evidence on our hypotheses: firms led by Democratic-leaning executives score higher in culture values. These findings suggest that firms with political partisanship leaning toward the Democratic Party are more likely to exhibit honest, collaborative, innovative, respectful, and customer-oriented corporate cultural values compared to those leaning toward Republican Party.
Interaction effects
In this section, we explore the moderating factors that strengthen or weaken the effect of political partisanship on corporate culture. We conjecture that the extent to which a leader can imprint their personal belief into a firm is related to the firm’s governance structure and overall environment. Corporate governance serves, in part, to guide leaders’ behavior and policies they advocate, with aim of benefiting stakeholders rather than merely reflecting personal convictions (Shleifer and Vishny, Reference Shleifer and Vishny1997). Moreover, within the context of trust and accountability in financial markets and business practice, Dupont and Karpoff (Reference Dupont and Karpoff2020) propose the Trust Triangle theory. According to this theory, reputational capital, legal institutions, and culture are three interactive pathways in affecting the development of trust. Therefore, we examine and discuss how internal governance provisions and external legal environment affect the imprinting of firm leaders’ political ideologies onto firm culture values.
For internal corporate governance structure, Bebchuk, Cohen, and Ferrell (Reference Bebchuk, Cohen and Ferrell2009) construct an entrenchment index (E index) to measure the strength of corporate governance provisions.Footnote 9 The E index is based on six provisions of corporate governance: staggered boards, limits to shareholder bylaw amendments, poison pill, golden parachutes, and supermajority requirements for mergers and charter amendments. A higher E index value indicates weaker shareholder governance and more entrenched management, implying reduced discipline or monitoring effects on managerial decisions and behaviors. Therefore, we hypothesize that the relationship between leaders’ partisanship and corporate cultural values will be more pronounced in firms with higher entrenchment level. In addition, the purpose of corporate governance for shareholders is to maximize firm value. Thus, we expect that cultural values of innovation and product quality, which are more directly related to value creation, will be more affected. In Panel A of Table 4, we examine the moderating effect of governance entrenchment on the effect of political partisanship on cultural values by interacting E index with TMT REP-DEM. Our results show that the coefficients of all interaction terms are negative, supporting our conjecture. Specifically, the coefficients in columns (1), (4), and (6) are statistically significant, which suggests that entrenchments especially hurt corporate culture in the dimensions of innovation and quality. In other words, stronger corporate governance mechanisms can protect innovation and product quality cultural values from the negative partisan influence.
Table 4. Political partisanship and corporate culture: interaction tests
This table reports the regression results of corporate culture scores on interaction effects between political partisanship of firm top executives and E index and state judicial accountability. All variable definitions are in the Appendix 1. Firm and year fixed effects are included in all specifications. Standard errors are clustered by firm in all specifications. Robust t-statistics are in the parentheses. ***, **, and * denote statistical significance at the 0.1, 1, and 5 percent levels, respectively.
For external governance environment, we focus on local judicial accountability which captures the effectiveness of state legal framework. Our data on state judicial accountability is drawn from the 2015 State Integrity Investigation, conducted by the Center for Public Integrity (CPI). This investigation assesses the strengths and weaknesses of the key institutional governance and anti-corruption mechanisms related to openness, transparency, and accountability in each state.Footnote 10 The CPI provides the detailed assessment results and final state judicial accountability scores. Given that state courts handle a vast majority of civil and criminal cases within the U.S. judicial branch, effective judicial accountability and discipline can help deter misconducts and foster an ethical climate and justice perceptions for firms headquartered in these states. We reasonably assume that when the effect of external governance environment is at work, the effect of leaders’ political partisanship on corporate culture becomes less pronounced. We thus expect the effect of partisanship on cultural values will be attenuated in firms located in states with effective judicial accountability. In Panel B of Table 4, we examine the interaction effect of external governance environment on the relationship between political partisanship of TMT and corporate culture by interacting State Judicial Accountability with TMT REP-DEM. We find that the coefficients of all interaction terms are positive, indicating that the relationships between political partisanship of executives and cultural values become weaker if firms located in states with more effective judicial accountability. More specifically, the coefficients in columns (1), (2), (3), and (5) are significant, suggesting that lower cultural values of integrity, teamwork, and respect in firms with Republican-leaning executives are improved in states with effective judicial accountability. The coefficients in columns (4) and (6) are close to significant, which suggests that the effect of external governance is less strong in the aspects of innovation and teamwork cultural values.
Overall, the empirical findings in this section provide supporting evidence that corporate governance plays a critical role in moderating the relationship between the political partisanship of firm executives and corporate cultural values. When a firm has stronger governance mechanisms shaping the firm’s culture and policies, the effect of management teams’ political ideology becomes weaker.
PSM methodology
One potential concern in our study is that firms with Democratic-leaning executives and those with Republican-leaning executives may be fundamentally different in firm characteristics that affect cultural values. To address this issue and reaffirm the effect of political partisanship on corporate culture, we perform PSM methodology in our analysis. For this analysis, we only include firms with Democratic-leaning executives and those with Republican-leaning executives and exclude firms with TMT REP-DEM measure equal to zero. In each sample year and for each firm with Democratic-leaning executives, we match it with a firm with Republican-leaning executives from the same 2-digit SIC industry, the same year, and based on the propensity score generated by all control variables in the main regressions. After obtaining the matched sample, we statistically compare the mean values of each control variable to examine whether control variables are not significantly different between the two groups of firms with different political partisanship.
Panel B of Table 5 reports comparisons of mean values of control variables in the matched sample. The number of observations for firms with Republican-leaning executives and firms with Democratic-leaning executives are both 6,454. We find none of the difference in control variables between two groups is statistically significant, which indicates that PSM removes all observable variations in explanatory variables other than the partisanship of firm executives.
Table 5. Political partisanship and corporate culture: propensity score matching
This table reports the propensity score matching results of the effect of results of political partisanship of firm top executives on corporate culture scores. All variable definitions are in the Appendix 1. Each firm-year observation with a Democratic-leaning TMT is matched to a firm-year observation with a Republican-leaning TMT, by 2-digit SIC industry, year, and all other control variables. Panel A presents the regression results after PSM. Panel B presents the comparison of firm characteristics after PSM. Firm and year fixed effects are included in all specifications. Standard errors are clustered by firm in all specifications. Robust t-statistics are in the parentheses. ***, **, and * denote statistical significance at the 0.1, 1, and 5 percent levels, respectively.
Panel A of Table 5 presents the results of the PSM tests. All model specifications include firm and year fixed effects, and the standard errors are clustered at firm level. We find that the culture scores are significantly different when firms have TMT leaning toward different political parties. The coefficients on Republican Leaning are negative in all columns and they are all statistically significant except for column (5), which confirms our findings in Table 3 and supports our hypothesis. The results are more straightforward to interpret because we use indicators and exclude politically neutral firms in the sample. On average, firms led by Republican-leaning executives have an integrity score 0.028 lower, a teamwork score 0.041 lower, an innovation score 0.072 lower, a respective score 0.035 lower, and a quality score 0.049 lower than firms led by Democratic-leaning executives.
Conclusion
Corporate culture has been viewed as a critical feature of corporate America that significantly impacts firm practices, performance, and shareholder value. Extending the existing studies exploring the association between leader characteristics and corporate cultural values, we focus on the political partisanship of firm executives. Using a sample of 26,017 firm-year observations for 2,424 unique S&P 1500 firms from 2001 to 2018, we examine whether there are partisan gaps in corporate cultural values. Our analysis centers on five dimensions: integrity, teamwork, innovation, respect, and quality, which are the mostly advertised values by U.S. large firms. Our findings reveal significant differences in cultural values between firms with executives leaning toward different political parties. Specifically, firms with executives leaning toward the Democratic Party consistently exhibit higher culture scores in all dimensions compared to those with executives leaning toward the Republican Party. Moreover, we demonstrate that the relationship between partisanship and corporate cultural values weakens when firms have less entrenched management and are located in states with stronger judicial accountability.
Our paper contributes to the growing literature on organizational culture. Adding to previous studies that identify senior leaders’ personal traits, past experience, and original cultural heritage as determinants of corporate cultural values, we demonstrate that the political partisanship of TMT can be an important factor in shaping corporate culture. Our paper also contributes to an extensive field of research that explores the economic outcomes of political partisanship by highlighting its role in the workplace.
There are several avenues for future research extensions. First, future research could consider other dimensions of corporate culture. Guiso, Sapienza, and Zingales (Reference Guiso, Sapienza and Zingales2015) list nine categories of cultural values and we only include the top five based on Li et al. (Reference Li, Mai, Shen and Yan2021). The other four values of safety, citizenship, communication, and hard work are also worth of examination. Moreover, different studies define and categorize cultural values differently. For example, Graham et al. (Reference Graham, Grennan, Harvey and Rajgopal2022a) use a list of cultural value of adaptability, collaboration, community-oriented, customer-oriented, detail-oriented, integrity, and results-oriented. As such, future research can extend our study by investigating more measures of corporate culture. Second, future research could examine the potential role of other firm participants besides senior leaders, such as the board of directors and stakeholders, in corporate culture. Most studies on how partisanship affects firm outcomes focus on executives because they operate firms and make corporate decisions at a daily basis (e.g., Fos, Kempf, and Tsoutsoura, Reference Fos, Kempf and Tsoutsoura2023, Jiao and Ren, Reference Jiao and Ren2024). However, it would also be interesting to investigate from the perspective of other parties. Third, future research can link the partisanship of firm executives to culture and corporate culture to firm performance and practices. In doing so, we obtain a comprehensive picture of the origins of corporate culture and illustrate how culture consequentially affects firm value. Lastly, future research could seek to replicate our results in international studies to examine and generalize the impacts of liberalism vs. conservatives on corporate culture in other countries.
Financial statement
Anqi Jiao acknowledges the financial support from the Capital University of Economics and Business [grant no. QNTD202301]. Honglin Ren acknowledges the financial support from the National Natural Science Foundation of China [grant no. 72202225].
Appendix 1. Variable definition
Appendix 2. Ten most representative words in the culture dictionary by Li et al. (Reference Li, Mai, Shen and Yan2021)
Appendix 3. Robustness tests - alternative measure of partisanship based on inversed rank of executive compensation
This table reports the robustness test results of baseline model using the alternative measure of executive partisanship based on the inverse rank of executive compensation. Firm and year fixed effects are included in all columns. Standard errors are clustered by firm in all specifications. Robust t-statistics are in the parentheses. ***, **, and * denote statistical significance at the 0.1, 1, and 5 percent levels, respectively.
Appendix 4. Robustness tests - lagged executive partisanship measure
This table reports the robustness test results of baseline model using the lagged measure of executive partisanship. Firm and year fixed effects are included in all columns. Standard errors are clustered by firm in all specifications. Robust t-statistics are in the parentheses. ***, **, and * denote statistical significance at the 0.1, 1, and 5 percent levels, respectively.
