Summary
The sustainability transition is making corporations increasingly tackle complex, ambiguous issues such as globalization, climate change, and inequality. To lead in the sustainability transition, corporations are laying out ambitious plan with aggressive targets, which may have negative implications in terms of corporate ethical behaviors. Moreover, these grand challenges are eminently political, and thus stakeholders and the broader public might hold divergent if not polarized views on them. Indeed, even on climate change and inequality, two core social issues which are increasingly central in the debate on the public role of corporations, the general public and corporate employees hold divergent views. We believe that the increasing need for companies to tackle grand challenges in the context of political polarization raises important, and so far, relatively neglected, questions on the role of corporate culture and political ideologies on corporate behavior. With this project we address the three following questions:
- What are the cultural and structural antecedents of corporate misconduct?
- How does political ideology and corporate purpose affect corporate environmental performance and corporate carbon emissions?
- How does political polarization affect corporate innovation?
In Study 1 we examine how the management’s excessive focus on goals and performance can lead to the development of a culture that brings constant performance pressure for employees. In this article, we theorize strain at the organizational level as the result of both organizational culture and organizational structure. We argue that organizations with a culture characterized by strong performance pressure are more likely to engage in misconduct. Furthermore, we build on the literature on bureaucracy and monitoring to argue that more formalized organizations, which arguably have more controls in place, have fewer negative effects of performance pressure. Likewise, more decentralized organizations, which tend to have less effective monitoring mechanisms, provide more fertile ground for performance pressure to have adverse effects.
To test these hypotheses, we analyze the regulatory and law violations of 880 publicly traded firms in the United States from 2008 to 2019 and measure organizational culture and structure through a natural language processing (NLP) analysis of the firms’ employee reviews on Glassdoor. We find that performance pressure significantly contributes to increasing the likelihood of organizational misconduct. We also find that the negative effect of performance pressure is stronger for companies that rely on a more decentralized structure, while the results for companies that rely on a more formalized structure are mixed.
In Study 2, we look at two specific aspects of organizational culture: organizational purpose (i.e., a set of beliefs about the meaning of a firm’s work beyond measures of financial performance) and political ideology (the prevailing beliefs among organizational members about how the social world operates) and how these constructs interact to explain differences in organizational actions toward the reduction of carbon emissions. Carbon neutrality entails a drastic technological and organizational change, and specific internal drivers may endorse its effectiveness. Recent literature proposes that organizational purpose can help advance organizational efforts towards sustainability; however, we do not know much about the condition under which this relationship can be effective. In this paper, we propose that organizational carbon emissions reduction may also depend on the interaction between the organizational purpose and the political ideology of the organization. Drawing on the organizations as polities literature to argue that the mediation of purpose and liberalism towards sustainability will further depend on the polarization of climate change beliefs at the community level. We gathered data from Glassdoor reviews to measure employees’ perceptions about the organization’s purpose, the Federal Election Commission (FEC), which provides individual contributions to political parties, and data from the Yale Climate Opinion Survey to measure community beliefs on climate change issues. The results show that organizational purpose and ideology (i.e., liberal vs. conservative firms) alone do not explain significant differences in CO2 reductions. However, when purpose is mediated by organizational liberalism, we find a high employee perception of purpose in liberal-leaning organizations to be associated with a substantial reduction in carbon emissions. Finally, organizations in communities with a more polarized view regarding climate change show weaker effects for the mediation of purpose and liberalism towards sustainability.
Finally, in Study 3, we examine whether and how political polarization affects organizational outcomes that heavily depend on collaborative behaviors, such as corporate innovation.
Building on research from political science, we propose that political polarization in society penetrates firms’ boundaries as organizational members bring their political identities into the workplace, and theorize that the activation of this process depends on the degree of political polarization among organizational members – that is, the degree to which politically engaged organizational members in the workforce are divided into two opposing political coalitions rather than a single aligned coalition. We propose that firms with more polarized workforces are less innovative because political polarization increases animosity and decreases collaboration between members with opposing political stances. We also theorize that this negative relationship is stronger in workplaces where politics is more salient. We test these predictions using data on the political affiliations of employees of 394 U.S. publicly traded companies and their corporate patenting activity for the period 2008–2015, and provide evidence of the underlying mechanism using employee reviews from Glassdoor. The results support our theory and inform organization theory research, highlighting the importance of political ideology and suggesting that political polarization might affect corporate innovation, a central driver of value creation in the economy.
Data
We assembled a unique dataset mixing scraped data (political contributions, organizational culture) and third parties’ datasets (climate change opinion, political risk, reputational risk, carbon emissions, corporate regulatory violations, patents) together with traditional data on economic, financial and ESG performance. To study organizational culture, we followed recent studies of organizational culture (Corritore et al., 2019) and build upon employees’ perceptions of culture using free-response reviews gathered from Glassdoor.com. Glassdoor is among the largest available database of employee reviews. The service requires users to anonymously rate the company they are working for to access the contents of the web-page for free (called also give-to-get model), a strategy that proved to reduce polarization in the evaluations (Marinescu, Klein, Chamberlain, Smart, 2018). As the data mainly contain textual data, we used natural language processing (NLP) methods that permits us to separate between the different topics that employees touch upon in their reviews. To study the political ideology of organizational members we processed the following data sources: (1) data on individuals’ political contributions and committee recipients from FEC (Federal Election Commission) databases, which register all political donations equal or exceeding $200. (2) Additional data on individuals’ political contributions provided by the Center of Responsive Politics (www.opensecrets.org). FEC requires from the donor information such as name, the amount of the contribution, home address, occupation, and employer. Thus, the sample of companies used in our studies was identified by applying an approximate (fuzzy) string matching between the FEC information on each contributor’s employer and Compustat company names. Following the literature, we included individual contributions to party committees, candidates, campaign committees and PACs connected unambiguously to either the Democratic or Republican Party (Chin, Hambrick, and Treviño, 2013; Gupta, Briscoe, and Hambrick, 2017). The variables measuring organizational misconduct were gathered from Good Jobs First and, specifically, by the “Violation Tracker” dataset, which covers 71 types of legal and regulatory violations addressed by more than 300 federal regulatory agencies, including 412,000 civil and criminal cases with total penalties of $616 billion. In addition, we restricted our sample to publicly traded companies with matched CIK codes in Compustat, resulting in 1,293 companies affected by penalties for 34,203 violations. Data on GHG emissions were gathered from Trucost, a commercial provider of corporate carbon emission data and other sustainability measures. Trucost is a widely used source for firm carbon emission data and covers a broad cross-section of firms worldwide. The dataset offers data that is either publicly available (through different sources) or estimated by Trucost through environmental profiling of the organization when the data is not available. The profiling process calculates sustainability scores through information regarding the primary business activities, revenues, direct and supply chain, annual reports, websites, or expenditures data. Finally, we studied innovation by using patent data, working on (1) the DISCERN dataset, provided by Arora, Belenzon, and Sheer (2021), which contains publicly listed U.S. headquartered firms matched to assignees of patents from the U.S. Patent & Trademark Office (USPTO) for the period 1980-2015. This dataset contains firms that have at least one patent granted and one year of positive R&D expenditures during this time frame. (2) PatentsView (www.patentsview.org) datasets, containing organized data on patents applications, grants, and citations.
Main Results
After outlying our hypotheses, collecting and matching the data, and performing the analyses, we have generated the three following working papers:
Study 1: Cavicchini, Andrea; Ferraro, Fabrizio; Samila, Sampsa. “Under Pressure: Culture and Structure as Antecedents of Organizational Misconduct” –submitted.
Results from our empirical analysis generally support our hypotheses. Performance pressure shows a positive relationship with the amount of dollars (in millions) the organization pays in fines the subsequent year. In subsequent analyses, we found that this effect is robust across different misconduct measures. The relationship between formalization and performance pressure provides limited support for our hypotheses. Our results confirm that performance pressure affects the dollar amount for the violations for companies with low formalization, though the effect is weakly significant, and we do not find a significant negative effect on the subset with high formalization.
The results on decentralization, in contrast, strongly support our hypotheses. Consistent with the idea that the delegation of responsibilities in decentralized organizations leads to less stringent control, we found that decentralized organizations are more likely to misconduct when performance pressure is present, paying more in fines for the violations. Taken together, our results show that a decentralized organizational structure can create conditions that enable the development of a toxic culture, with negative consequences for the organization. This result extends McKendall and Warner’s (1997) findings that an ethical climate negatively mediates the effects of decentralization on environmental misconduct by demonstrating that performance-oriented cultures could create conditions that increase the negative impact of misconduct
Study 2: Cavicchini, Andrea. “Organizational Purpose, Ideology, and Carbon Emissions” September 2022 -submitted.
I find evidence supporting the hypothesis that the interaction between purpose and liberalism plays a key role in explaining organizational actions toward reducing carbon emissions. Furthermore, I find a stronger effect regarding more carbon-intensive industries. Moreover, I find that the effect of the mediation between purpose and ideology is also dependent on community beliefs, and specifically polarization, regarding climate change. Communities in which there is polarization regarding climate change do not show the negative effects of purpose and ideology towards carbon emissions.
Study 3: Castiglia, Ludovica. “Porous Polities: Organizational Political Polarization and Firm Innovation.” September 2022 – Under review.
In this paper, we investigate the internal political composition of organizations to understand whether and how political polarization relates to their innovation outcomes. The empirical findings confirm our hypotheses and point toward an alarming fact: on average, firms with higher levels of political polarization were less innovative in the period under investigation. Specifically, the results suggest that a 10% increase in organizational political polarization above the mean is associated with a 7% reduction in a firm’s number of patents, on average. Political polarization in society seems to penetrate the boundaries of firms with polarized workforces, influencing their members’ affective experiences and hindering their collaborative behaviors. Moreover, political identity becomes more salient under circumstances that make politics more prominent in the workplace, further strengthening the negative relationship between organizational political polarization and innovation.
Impact and outputs
Conference presentations
The three studies have been presented at several management conferences and invited research seminars.
Study 1:
- 2022 Brown Bag Sessions, IESE Business School
- 2022 The Wharton School of the University of Pennsylvania (invited seminar)
- 2021 American Sociological Association Annual Meeting
- 2021 Academy of Management Annual Meeting
- 2021 EGOS Conference, Amsterdam, Holland
- 2021 London Business School (Invited Seminar)
- 2021 Society for Institutional & Organizational Economics (SIOE) Conference
- 2021 Society for the Advancement of Socio-Economics (SASE) Conference
- 2021 Università degli Studi di Trento (invited seminar)
- 2020 IRiSS Conference on Computational Sociology, Stanford University
Study 2:
- 2022 BrownBag Sessions, IESE Business School
- 2022 Academy of Management Annual Meeting
Study 3:
- 2022 Annual People & Organizations Conference, The Wharton School, Philadelphia, Pennsylvania.
- 2022 Academy of Management Annual Meeting, Seattle, United States.
- 2022 EGOS Colloquium, Vienna, Austria.
- 2022 Medici Summer School, HEC, Paris, France.
- 2021 Brown Bag Sessions, IESE Business School, Barcelona, Spain.
Awards
- Best Paper of the 2022 Annual Meeting Academy of Management Proceedings (top 10% of accepted papers) – for the article “Porous polities: Organizational political polarization and firm innovation”
Publications
An abridged version of the article Porous polities: Organizational political polarization and firm innovation has been published in the Academy of Management Annual Meeting Proceedings (2022) (https://journals.aom.org/doi/abs/10.5465/AMBPP.2022.65)
