The influence of corporate misconduct on the relationship
between political ideology and different forms of CSR.
ABSTRACT
Corporate social responsibility (CSR) is crucial to the survival of a firm. Yet, past research has mainly focused on CSR as a whole. This study, therefore, focuses on the different components of which CSR exists and how they are influenced by political ideology. In addition, the influence of corporate misconduct on these relations was examined. The theory was tested in the context of 574 companies that were included in the S&P500 between 2000 and 2017. Results from the panel regression do not confirm the hypotheses, but include some interesting findings. The findings show that corporate misconduct increases both internal and external CSR. Furthermore, when political ideology comes into play these influences are strengthened. These findings provide both theoretical and practical implications.
Keywords: Corporate social responsibility, Internal CSR, External CSR, Political ideology,
Corporate misconduct, Liberalism, Conservatism.
1
INTRODUCTION
“Blackrock shakes up the business to focus on sustainable investing” (Financial Times, 2020), “Coca-Cola intends to create a 25% reduction in their carbon footprint by 2020” (Digital
Marketing Institute, 2020), “Google CEO stands up against social issues including Donald Trump’s anti-muslim comments” and “Google earned RI’s highest CSR score” (Digital
Marketing Institute, 2020). With some of the most influential organizations in the world going all in on corporate social responsibility (CSR), there is no denying that CSR is more actual than ever. But what made these organizations decide to do so? External pressure to do ‘the right thing’ is not a new phenomenon. Thus, what drove these organizations and what changed? Looking at most of the organizations of this size, the decision to pursue CSR practices is made in the upper echelon of the organization, mostly by the board of directors (Rao & Tilt, 2016).
The drivers or values of a person have long been associated with decision making (Fritzsche & Oz, 2007) and could therefore be a good guidebook of what decisions people would make. In this case, whether or not to engage in corporate social responsibility. However, whether that would result in actually taking action to improve CSR practices (internal CSR) or simply more communication to the outside world about the existing actions (external CSR) remains unclear.
2 Due to this proxy, it is possible to investigate whether liberal or conservative values influence a person’s CSR decisions, similar to what Chin et al. (2013) have done before with decision making regarding environmental sustainability. As liberals traditionally engage more in actual CSR activities than conservatives (Liston-Heyes & Ceton, 2007), this proxy should then reveal liberals as the more internal CSR-focused group of the two. On the other side it could reveal conservatives as less ‘walking than talking’, as Chin et al. (2013) argue that conservatives’ CSR efforts are linked to the recent firm performances. This ‘talking without
walking’ would be an example of external CSR.
However, as past research has already proven, a person’s values, or a person’s political ideology for that matter, is not the sole driver of engaging in CSR (Hofmann et al., 2012). Nor can it be excluded that the relationship with these drivers themselves are influenced by other phenomena (Grewatsch & Kleindienst, 2017). Grewatsch and Kleindienst (2017) argue that the current theory fails to deliver a satisfying answer to this question.
Recent times seem to expose an example of such a phenomenon that could influence the relationship with CSR. As the world struggles to fight the coronavirus, in some cases it becomes painfully clear that, for instance, the social part of CSR has been less prominent than it should have been and is rapidly becoming more and more important (The Asset, 2020). Or to put it in a more abstract perspective, this is an example of a crisis which leads to an increase in CSR practices. To answer why those organizations decided to pursue CSR
3 A form of crisis which could increase CSR practices is corporate misconduct. Du (2015) states that misconduct leads to (environmental) philanthropy in a setting of Chinese family-owned companies. If this setting would be transformed into a global one, it may very well be the case that the mechanism still holds. Philanthropy could in this case be similar to CSR, as both would arguably serve, to some extent, to the same end (Liston-Heyes & Ceton, 2007).
This research will try to fill two theoretical gaps. First of all, it will make a distinction between two forms of CSR, internal and external CSR, to shed new light on what parts of CSR are actually influenced by political ideology. Secondly, this research will try to fill a gap by suggesting and investigating the relatively novel phenomenon that could influence the relation of political ideology and corporate social responsibility, namely corporate
misconduct. As said before, there already has been a wide range of research regarding the different influences of political ideology, even on the relationship with CSR. To maximize the academic contribution, this research will keep its focus on the two forms of CSR and whether or not corporate misconduct influences the relationships between the political ideology and these forms of corporate social responsibility. This results in the following research question:
‘To what extent does corporate misconduct influence the relationships between political ideology and the different forms of corporate social responsibility?’
5
THEORETICAL BACKGROUND Corporate social responsibility
Corporate social responsibility (CSR) is best defined as combination of its five themes as stated by Scott (2007). Scott (2007) stated that CSR is a responsibility to the community and society, the promotion of democracy and citizenship, aiming to reduce poverty and the inequality between rich and poor, improving employee rights and working conditions and general ethical behaviour. CSR can be divided in two types; internal and external CSR. Internal CSR actions, are the actual actions undertaken with the goal to improve CSR practices (e.g. training, forming board committees) (Hawn & Ioannou, 2016; Marquis & Qian, 2014). External CSR, is the communication of the internal CSR measures to the environment (e.g. branding, disclosure, partnerships)( Hawn & Ioannou, 2016; Marquis & Qian, 2014). Hawn and Ioannou (2016) also state that to maximize outcomes of CSR, there must be alignment between both forms.
Supported by the claims of McElhaney (2009) and McWilliams et al. (2006) that CSR is part of a firm’s strategy and the claim of Judge and Zeithaml (1992) that boards of directors
are responsible for formulating those strategies, Rao and Tilt (2016) state that CSR measures are the outcome of board decisions. According to 90% of the top executives, these CSR decisions are essential to a firm’s competitiveness (Unruh et al., 2016). Next to being responsible for the measures, Rao and Tilt (2016) also state that the role of a board in these matters is among the most crucial roles regarding CSR measures. This justifies the assumption that boards influence CSR in their organization.
A board of directors, like any group, consists of people who make their decisions based on their personal values (Badr et al., 1982). In addition, directors and executives tend to make choices through highly individualized lenses that are formed by the managers’ experiences,
6 the values of a person are difficult, if not impossible, to observe or measure, research proposes that the values of a person can be captured through their political ideology (Chin et al., 2013). Thus, political ideology can serve as a proxy for a person’s values.
Political ideology
Political ideology will cover the context of the continuum between conservative and liberal ideologies. It represents a person’s individual values, as value theorists have concluded that the conservatism-liberalism dimension is among the most centrally important constructs to measure such values (Schwartz, 2013) and hence, as stated, it serves as a proxy (Chin et al., 2013). Political conflicts can often be explained in conservative-liberal terms (Poole & Rosenthal, 1984). Goren et al. (2009) even state that in the United States of America, party identification is the most influential political predisposition a citizen can have.
7 and not with its stakeholders. Therefore, one can argue that conservatism tries to preserve the ‘old way’, while dismissing open-minded ideas of corporate responsibility of any kind.
At the other side of the spectrum, liberalism is characterized by values as freedom, equality and rationality (Blau, 1975), which are values that conservatism downgrades (Feather, 1979). Von Kuehnelt-Leddihn (1997) states that regardless of the form of government, the exercise of power should not prevent citizens from enjoying the greatest amount of liberty compatible with the common good. The fact that liberalism needs to be compatible with the common good, brings in an aspect of utilitarianism. As von Kuehnelt-Leddihn (1997) states, not even the most celebrated liberal has the right to drive 100 miles per hour through a village, as this would do more harm than good to society. It is because of this compatibility, that even though liberals favour freedom in general, liberalism in the United States is characterized by the promotion of substantial government intervention in the economy (Walters, 1977). Following that line of thought, as corporate social responsibility is compatible with the greater good and its benefits for all outweigh the costs for one firm, it can also be argued that liberalism is in favour of CSR practices.
The essential differences between conservatism and liberalism can be categorized in five foundations; harm/care, fairness/reciprocity, ingroup/loyalty, authority/respect and purity/sanctity (Graham et al., 2009). As liberalists are more focused on the harm dimension (Graham et al., 2009), one can argue that liberalists will engage more in internal CSR practices, as these are in principle intended to reduce harm in a given way (e.g. environmental harm, social harm).
8 specified, most of these papers did focus somewhat more on internal CSR, as it was mostly about the actual CSR actions that a firm undertook. Therefore, this study will assume this to be true as well, leading to the first hypothesis:
H1: Firms with a more liberal board, as opposed to a more conservative board, are likely to show higher levels of internal corporate social responsibility.
However, the fact that papers of Chin et al. (2013), Jiang et al. (2018) and Gupta et al. (2019) unintentionally focused on internal CSR, does not exclude external CSR from the statements of both researches, they merely didn’t use measurements for both forms of CSR.
The fact that these papers did not made a distinction between these forms might be a crucial difference. As this research aims to make this kind of distinction, additional evidence supporting external CSR in this claim is required. Specific evidence supporting or objecting the claim that political ideology indeed positively influences external CSR is not available yet. Although such claims would be preferable, the lack of them does not leave us clueless. Some of the more general CSR studies, like the ones from Jiang et al. (2018) and Chin et al. (2013), suggests that a more liberal political ideology might indeed positively influence CSR in all its forms. Although this suggestion might not be as strong as desired, it does provide direction.
On the other hand, Walters (1977) states that participants in the corporate social responsibility debate, frequently find that they have unlikely political allies and opponents. This would suggest that not in all cases a more liberal political ideology leads to more CSR practices and perhaps a more conservative political ideology could be desirable.
9 and the liberals are more ideologically divided than in the past (PEW, 2020). Therefore, the findings of Walters (1977) might be less applicable to the context of this study, than more actual articles might be. What also comes to the attention, is that not only in China (like Jiang et al. (2018) studied) the evidence showed that political ideology is positively linked to CSR practices, as the studies of Gupta et al. (2019), Chin et al. (2013) and Liston-Heyes and Ceton (2007) are based on respectively Fortune 500 firms, S&P1500 and S&P500 firms. These are companies which are located in the United States of America. The combination of these aspects tilts towards the assumption that liberalism will increase a firm’s external CSR practices. Unfortunately, a clear mechanism that explains this specific relation cannot be identified in either one of these studies. However, when put together, the papers of Hawn and Ioannou (2016), Liston-Heyes and Ceton (2007) and Graham et al. (2009) do provide such a mechanism. Hawn and Ioannou (2016) state that to maximize CSR outcomes, internal and external CSR must be aligned. In addition, as stated by Liston-Heyes and Ceton (2007), liberals are more inclined to pursue higher CSR outcomes than conservatives. As CSR in principle prevents harm in a certain way and liberals are more easily triggered than conservatives to prevent harm, as this foundation is most important to them (Graham et al., 2009), liberals are more inclined to increase their level of external CSR than conservatives. This leads to the second hypothesis:
H2: Firms with a more liberal board, as opposed to a more conservative board, are likely to show higher levels of external corporate social responsibility.
Corporate misconduct
10 already negatively influence firm performance (Carberry et al., 2018). In the case of actual corporate misconduct, investors perceive the risk to invest in the firm as higher, which negatively influences a firm’s stock market value (Carberry et al., 2018; Murphy et al., 2009). Also the firm’s profits might be threatened (Murphy et al., 2009) and its reputation damaged (Davies & Olmedo-Cifuentes, 2016). To the question of what triggers corporate misconduct, different answers arise. One that stands out is the study from Li et al. (2019), as they state that personality traits of a CEO could influence the level of environmental misconduct a firm commits. Past research also provided evidence that CSR activities are influenced by corporate misconduct (Du, 2015; Bae et al., 2020; Park et al., 2020).
In their study, Park et al. (2020) state that conservative boards, when confronted with a case regarding financial misconduct, tend to act more often in alignment with CSR practices than their liberal counterparts. In principle, this would lead to liberal boards engaging more in CSR practices (as hypothesis one assumes), but when misconduct comes into play, conservative boards might take over as leading boards on CSR practices. The rationale behind this finding, as stated by Park et al. (2020), is that politically conservative boards address responsibility in a different way than politically liberal boards do. Conservatives tend to address more responsibility to the individual responsible for the corporate misconduct. On the other hand, liberals tend to address more responsibility towards situational factor, which might be softening (Park et al., 2020). In cases of misconduct, conservatives will therefore act more decisive than liberals, resulting in increased internal CSR practices.
11 2003), conservatives are expected to be more inclined to act upon corporate misconduct than liberals are, leading to the third hypothesis:
H3: Corporate misconduct weakens the positive relationship between a more liberal board and internal CSR.
On the other hand, Du (2015) states that corporate misconduct positively influences philanthropy. The fact that these studies seem somewhat contradicting might be, because all the studies that are mentioned investigate corporate misconduct or CSR, but none completely combines both in the same study. Therefore, it cannot be said with certainty what the influence of corporate misconduct in this setting will be, but they do provide guidance. As this study makes the distinction between internal and external CSR, this somewhat contradicting evidence can be split in two parts, which make more sense separately. For example, one can see that the study from Bae et al. (2020) states that -especially external- CSR is positively influenced by corporate misconduct. This would lead to the assumption that when put in a moderating role, corporate misconduct would positively influence relationships of factors with external CSR. A closer look at the rationale that was used in the study of Bae et al. (2020) suggests it could be used in the context of this study as well. Bae et al. (2020) state that when there is corporate misconduct, a firm’s reputation is damaged and when CSR increases, a firm’s reputation
12 on security (Feather, 1979), they are expected to engage less in such security-threatening practices.
Based on the fact that two out of the three papers regarding CSR advocate a positive influence of corporate misconduct on CSR in general, of which one is more focused on external CSR, the mechanism that they used is also applicable in this context and the assumption that conservatives will engage less in external CSR due to risk avoidance, the fourth hypothesis will be:
13
METHODOLOGY Data collection
As no available database offered all the required data, different databases were used in this research. The required data was then combined in a new dataset. To gather data about internal
CSR, external CSR and corporate misconduct of the companies, the Thomson Reuters Eikon
database was used.
To get data about the political ideology of board members, a new dataset had to be made. As this dataset had to consist of a clear division between conservatism and liberalism, a suitable country with such a division was sought after. Perhaps the most influential country with a political spectrum that is dominated by the conservatism-liberalism dimension, is the United States of America. As the Republicans (more conservative) and the Democrats (more liberal) are more ideologically divided than in the past in the United states of America (PEW, 2020), this country offered the perfect opportunity to study topics like political ideology. In addition to the fortunate circumstances in the country’s political spectrum, the United Stated also have federal legislation that forces every donation of $200,- or more made to a political campaign to be publically registered. On the assumption of ‘putting money where your mouth is’, these
donations imply the political ideology of corporate executives in the United States of America. Therefore, this research has the political ideology of American corporate executives as the basis of the variable political ideology.
14 demographics. This check was done by at least two persons for each name, to rule out personal errors and disagreements were settled with additional background checks. This process was repeated 3 times to fine-tune the outcomes. To transfer this data into a political ideology score, the method as used by Chin et al. (2013) was repeated. This method creates a combined liberalism score based on four scores that each evaluate a different aspect of political giving. These four scores include “the number of donations towards Democrats divided by the number of donations to recipients of both parties, the dollar amount of donations to Democrats divided by the amount of donations to both parties, the number of years the executive made donations to democrats divided by the number of years donations were made to either party and the number of distinct Democratic recipients to which the executive made donations divided by the total number of distinct recipients of both parties” (Chin et al., 2013, page 208). This combined
liberalism score of the executives was then linked to the companies of which the executives were board members. These scores were then used to create the average liberalism score per company board of S&P500 firms. The data regarding the control variables gender, board size,
age difference, firm size, firm profitability and industry were collected from the Boardex
database.
Variables
As not all constructs will be measured in the same way, each construct will be elaborated on individually. To measure internal CSR, external CSR and corporate misconduct, the example of Hawn and Ioannou (2016) was followed. They measured their constructs as the totals of a set of variables. For internal CSR a set of 42 variables1 were combined into an index score that indicated the amount of policies regarding internal CSR a company had in a given year. For
external CSR a set of 32 variables1 were combined into an index score. This totalled the amount
15 of report opportunities that were used by a company for external CSR purposes in a given year.
Corporate misconduct was combined out of a set of 12 variables1. This total indicated whether or not a company had encountered corporate misconduct exposure in a certain year. In the years in which companies had corporate misconduct exposure, the amount of misconducts were included as well.
Political ideology was measured as the average political ideology of a board, ranging
from 0 to 1. A completely conservative board would have a score of 0 and a completely liberal board would have a score of 1, with 0,5 meaning that the board is equally balanced between conservatism and liberalism.
To account for the possibility that other factors might influence the outcomes, this study controlled for the most important ones. First of all, the gender of the corporate executives was included, as Kassinis et al. (2016) state that a more gender diverse board tends to engage more in environmental sustainability policies than a more gender homogeneous board. Which on its turn, leads to an increase in CSR practices. With the inclusion of gender in the model, this study excludes the possible effect gender could have on the decision to whether or not engage in forms of corporate social responsibility. For the variable gender a ratio was used, ranging from 0 to 1. In this ratio 0 means a board consisting of only female members and 1 means a board with only male members.
It could also be the case that a board with a large age difference increases the CSR practices of a company (Post & Rubow, 2011). Although there are few studies on board age diversity, Kang et al. (2007) state that it is among the most important observable backgrounds of diversity issues regarding the board of directors. Therefore, there was controlled for age
difference within boards. Age difference was measured as the standard deviation of the average
16 Past research has also found firm size to influence corporate social responsibility (Baumann-Pauly et al. 2013; Wickert et al., 2016; Oktavianawati, 2018). To exclude the influence of firm size on CSR as a whole, or as stated by Baumann-Pauly et al. (2013), on either
internal CSR or external CSR, this study controlled for it. The firm size was measured as the
log of a company’s total market value, to account for skewness.
Board size is also a factor that has been found to influence CSR (Oktavianawati, 2018).
Oktavianawati (2018) states that, as the size of the board increases, so does the corporate social responsibility disclosure. Thus, this research controlled for board size. The board size was measured as the total number of directors on a board.
The industry in which a company operates can be of influence to the extent of CSR practices firms tend to have (Wanderley et al., 2008), whether competitors in a certain market engage in CSR might stimulate firms to do the same. Therefore, this study controlled for
industry. Industry was measured as the GIC industry code.
Lastly, Firm profitability is also found to be of influence regarding CSR (Oktavianawati, 2018). To exclude the possibility that a firm’s profit is the driver of its CSR practices, this study
17
Table 1: Univariate descriptive statistics of the variables
Variable Mean
(Variance)
Minimum Maximum N total
Internal CSR 17.14 (12.26) 0 40 8,924 External CSR 5.99 (6.14) 0 27 8,924 Political ideology 0.43 (0.21) 0.01 0.99 8,630 Corporate misconduct 0.23 (0.66) 0 7 8,924 Firm size 22840.94 (48205.40) 71.25 790050.10 8,232 Board size 10.58 (2.54) 2 33 8,924 gender 0.85 (0,10) 0,38 1 8,924 Industry 8,924 Age difference 7.11 (2,07) 1.5 16.7 8,924 Firm profitability 5076.42 (11257.99) -76735 130978 8,819 Analysis methods
As the proposed model in this study actually consists of two separate statistical models which are combined for interpretation, these two first need to be treated as separate models to test all the hypothesis. To test whether fixed or random effects where in play, a Hausman test was conducted. This was done using the ‘xtreg’ command in Stata, as this study used panel data. This test had for both the internal CSR and the external CSR model a p<0.01(i.e. significant), thus a fixed effect model was used for both. Next to the Hausman test, a test was done to check if time-fixed effects were needed. This was the case, as both models were significant (p<0.01). The panel regression was lagged for one year, as it takes time to translate cases of misconduct into policies and reports. Additionally, there was tested for heteroscedasticity, autocorrelation, normality and collinearity.
18 control variables gender, firm size, board size, age difference, firm profitability and industry. Afterwards, to test whether or not hypothesis one could be accepted, political ideology was added. To see how corporate misconduct would fit in the model as a direct influence on internal
CSR, it was added to the model as well. Lastly, to test whether or not hypothesis two could be
accepted, the interaction term political ideology*corporate misconduct was included.
The second model that was tested included external corporate social responsibility as the dependent variable and tested the third and fourth hypotheses. To get a clear impression of this setting, an entity and time fixed effects panel regression was conducted, including external
CSR and the control variables gender, firm size, board size, age difference, firm profitability
and industry. Afterwards, to test whether or not hypothesis three could be accepted, political
ideology was added. To test the direct influence of corporate misconduct on external CSR, it
19
RESULTS Descriptive statistics
To get a global impression of underlying relations between all the variables in this study, bivariate analysis are addressed. These correlations are displayed in table 2. In the table one can see that the variables have significant correlation amongst them, with only one exception (i.e. between age difference and board size). Next to this exception, there is another correlation that stands out. The correlation between board size and industry is the only correlation that is less significant than the others, but still significant for p<0.05. A noticeable remark that can be made when observing the table, is that political ideology and the other key variables are weakly correlated when compared to the correlations among the other key variables.
Table 2: Correlations among variables
20
Diagnostics
To account for the reliability of the results, multiple tests were conducted. These tested for normality, heteroscedasticity, serial correlation and collinearity in both models.
To test for normality, a Shapiro Wilk W test was used. As the key variables internal CSR,
external CSR and corporate misconduct are assembled out of a wide range of variables, with
21
Hypothesis testing
The results of the entity and time fixed-effects panel regression analysis with internal CSR as dependent variable are presented in table 3. Model 1a introduces the setting in which the hypotheses will be tested, including internal CSR and the control variables firm size, board size,
firm profitability, gender and age difference. Model 2a then reports the first main effect (between political ideology and internal CSR), in the setting as described before. Model 2a also represents hypothesis one, as it reports that political ideology positively, but not significantly (β = 0.716, p = 0.217), affects internal CSR. Therefore, the hypothesis “Firms with a more
liberal board, as opposed to a more conservative board, are likely to show higher levels of internal corporate social responsibility” cannot be supported.
In model 3a the main effect is presented, including the control variables and the variable corporate misconduct. Corporate misconduct is added to test its direct effect on
internal CSR. This model reports a positive and significant (β = 0.675, p < 0.01) influence of corporate misconduct on internal CSR. In the complete model, model 4a, also the interaction-term political ideology*corporate misconduct is included. In this model, the influence of
political ideology on internal CSR is positive, but not significant (β = 0.326, p = 0.575). Both
the influences of corporate misconduct and the interaction-term are found to be positive and significant (respectively, β = 0.440, p < 0.01 and β = 2.428, p < 0.01). As the main effect ,on
which the moderating effect is based, is not significant and the direction of the coefficient differs from the one that is hypothesized, hypothesis three “Corporate misconduct weakens the
positive relationship between a more liberal board and internal CSR” cannot be supported. In
addition, the complete model’s explained variance (Model 4a, R2 = 0.661) is significantly (p <
22
Table 3:Results of the four entity and time fixed effects regression-analyses regarding internal CSR. Includes the results of hypotheses 1&3, including the moderator, interaction-term and the control variables.
Model 1a Model 2a Model 3a Model 4a
Variable b (se) p b (se) p b (se) p b p (constant) -9.553 (1.654) 0.001** -11.422 (1.723) 0.001** -11.349 (1.719) 0.001** -11.096 (1.717) 0.001** Political ideology (PI) 0.716 (0.581) 0.217 0.650 (0.579) 0.262 0.326 (0.581) 0. 575 Corporate misconduct (CM) 0.675 (0.123) 0.001** 0.440 (0.136) 0.001** PI*CM 2.428 (0.471) 0.001** Firm size 0.565 (0.154) 0.001** 0.680 (0.157) 0.001** 0.679 (0.157) 0.001** 0.687 (0.156) 0.001** Board size 0.044 (0.050) 0.379 0.055 (0.050) 0.266 0.070 (0.050) 0.160 0.076 (0.050) 0.128 Firm profitability 0.533 (0.170) 0.002** 0.487 (0.170) 0.004** 0.468 (0.169) 0.006** 0.437 (0.169) 0.010* Gender -0.093 (1.235) 0.940 0.428 (1.241) 0.730 0.447 (1.238) 0.718 0.503 (1.236) 0.684 Age difference
Year fixed effects
-.0.017 (0.049) Included 0.727 0.012 (0.049) Included 0.807 0.003 (0.049) Included 0.948 -0.007 (0.049) Included 0.888 R2 (adj) 0.639 0.649 0.656 0.661 F-value 21.694** 22.674** 23.674** 30.673** N 7,535 7,327 7,327 7,327
a dependent variable: Internal CSR
Significant for p<0.05*, Significant for p<0.01**
Table 4 presents the results of the entity and time fixed-effects panel regression analysis with external CSR as the dependent variable. Model 1b introduces the setting of the second model, including external CSR and the same control variables firm size, board size, firm
profitability, gender and age difference. Model 2b introduces the second main effect of this study, namely the relation between political ideology and external CSR. In this model one can see that political ideology has a positive, but not significant (β = 0.595, p = 0.072), influence on external CSR. Therefore, hypothesis 2 “Firms with a more liberal board, as opposed to a
23 Model 3b includes the variables of model 2b, with the addition of corporate misconduct as a direct influence on external CSR. This model reports that corporate misconduct positively and significantly (β = 0.819, p < 0.01) affects external CSR. The complete model, model 4b, includes the interaction-term political ideology*corporate misconduct as well. This model reports a positive, but not significant (β = 0.267, p = 0.415) influence of political ideology on
external CSR. Moreover, it reports a positively and significantly (β = 0.580, p < 0.01) effect of corporate misconduct on external CSR and a positively and significantly (β = 1.722, p < 0.01)
effect of the interaction-term. Although the interaction-term is significant and the direction of the relation is as expected, hypothesis four “Corporate misconduct strengthens the positive
relationship between a more liberal board and external CSR” cannot be supported, as the main
24
Table 4:Results of the four entity and time fixed effects regression-analyses regarding external CSR. Includes the results of hypotheses 2&4, including the moderator, interaction-term and the control variables.
Model 1b Model 2b Model 3b Model 4b
Variable b (se) p b (se) p b (se) p b p (constant) 0.456 (0.931) 0.624 -0.957 (0.981) 0.329 -0.868 (0.971) 0.371 -0.650 (0.967) 0.501 Political ideology (PI) 0.595 (0.331) 0.072 0.514 (0.327) 0.117 0.267 (0.328) 0.415 Corporate misconduct (CM) 0.819 (0.070) 0.001** 0.580 (0.076) 0.001** PI*CM 1.477 (0.265) 0.001** Firm size 0.101 (0.087) 0.244 0.188 (0.089) 0.036* 0.187 (0.088) 0.034* 0.189 (0.088) 0.032* Board size -0.058 (0.028) 0.040* -0.059 (0.028) 0.038* -0.041 (0.028) 0.141 -0.034 (0.028) 0.224 Firm profitability -0.081 (0.096) 0.397 -0.069 (0.097) 0.472 -0.092 (0.096) 0.335 -0.123 (0.095) 0.197 Gender -0.806 (0.695) 0.246 -0.611 (0.706) 0.387 -0.588 (0.699) 0.400 -0.519 (0.696) 0.456 Age difference
Year fixed effects
0.030 (0.027) Included 0.278 0.049 (0.028) Included 0.082 0.038 (0.028) Included 0.171 0.030 (0.028) Included 0.277 R2 (adj) 0.411 0.429 0.458 0.468 F-value 21.694** 22.674** 23.674** 30.673** N 7,535 7,327 7,327 7,327
b dependent variable: External CSR
25
DISCUSSION
Firms nowadays target their corporate social responsibility as one of the most important practices to survive (McElhaney, 2009; McWilliams et al., 2006). Yet, very little research has been done to the influence of corporate misconduct on CSR practices. Even fewer make the distinction between different forms of CSR in this context. Therefore, our understanding of the role of corporate misconduct in this context is very limited. To expand literature and our understanding of the topic, this study examined how corporate misconduct influences the relations between political ideology and both internal and external CSR.
Theoretical implications
As part of a company’s strategy, CSR is critical to a company’s survival on the long term (McElhaney, 2009; McWilliams et al., 2006). Therefore, firms must properly act upon their CSR goals and communicate the CSR practices that they undertake to their environment (Hawn & Ioannou, 2016). This study found that, to increase internal CSR practices, multiple factors are important. However, one that was hypothesized to be of influence, political ideology, was not found to be supported by the data. As this hypothesis was partly based on past research (Chin et al., 2013; Jiang et al., 2018; Gupta et al. 2019), an important question to ask is: why do these findings differ? The reason for this difference might either lie within political ideology or internal CSR. As the procedure to measure political ideology stems from one of the same studies that was used to support this hypothesis, Chin et al. (2013), it is more plausible that the difference has to do with internal CSR. However, it does not exclude the possibility that the difference is due to political ideology. Chin et al. (2013) focused only on newly appointed CEO’s to measure “their firms’ CSR profiles from the outset of their tenures” (Chin et al., 2013,
26 could suggest that the strategical decision to increase internal CSR, although still made in the upper echelons of the organization, is determined on a more day-to-day business level, hence, by the CEO. When including the papers of Jiang et al. (2018) and Gupta et al. (2019) as well, a more broad explanation must be sought. One possibility is that, although the papers did have a tendency towards internal CSR, it simply might not be the case that a more liberal board of directors increases the internal CSR practices in a firm. This adds to the existing literature, as there has been no research before that tested the effect of political ideology on internal CSR specifically. Other studies might benefit from the fact that there seems to be a difference between the influence of political ideology on CSR as a whole and different parts of CSR separately.
However, when adding the influence of corporate misconduct to the relation, this statement becomes far from truth. Although the influence of political ideology on internal CSR still isn’t supported by the findings and therefore, the second hypothesis has to be rejected, the combined influence of political ideology and corporate misconduct does influence internal CSR. This might suggest that the mechanism should be turned around, with political ideology as moderator of the relation between corporate misconduct and internal CSR. This would mean that in cases of corporate misconduct, the political ideology of a board becomes relevant for the company’s internal CSR. Or put differently, when a company has to deal with corporate
27 might explain the different findings. Secondly, the assumption was used that based on the value ‘security’ the conservatives would act more upon misconduct, as this value is more important
to conservatives than to liberals (Feather, 1979). It might be the case that this assumption does not hold. To regain security in times of corporate misconduct, conservatives might focus more on core processes of their firm, to increase firm performance again. In doing so, it could be the case that they neglect internal CSR practices more than before.
A crucial assumption for the statement that the mechanism should be turned around, would be that corporate misconduct influences internal CSR directly. An assumption that is supported by the findings. This finding contributes to the existing literature of the influences on internal CSR, as up to this point, research regarding the influence of corporate misconduct on CSR has not specified the form of CSR. Future research might use this finding to hypothesize more specifically on CSR topics. As there are a few papers that studied the influence of corporate misconduct on CSR as well, this addition might aid in underlining the importance it can have on CSR. Other factors that were found to influence the internal CSR practices of a company are the firm size and the firm profitability. Both positively influence the internal CSR practices.
28 in similar settings, executives should have a certain knowledge of CSR management. Another conclusion could be that more liberal boards simply do not increase the external CSR practices of a company. To slightly jump ahead, this latter conclusion seems less plausible, as when corporate misconduct is added to the relation between political ideology and external CSR, this becomes of influence. Even stronger than solely the influence of corporate misconduct on external CSR.
Another finding of this study was the direct influence of corporate misconduct on external CSR. Although the fourth hypothesis could not be supported, due to the lack of support for the influence of political ideology on external CSR, this positive influence of corporate misconduct was expected. Bae et al. (2020), whose paper formed the basis of this hypothesis, stated in their paper that corporate misconduct damages firm reputation and firm reputation could be fixed by an increase in CSR practices. The missing link in this assumption that managers would act in ways that increase would increase a firm’s reputation was provided by Tadelis (1999), but was not tested in the same research settings yet. This study provides that missing link, adding to the existing literature.
Of most importance in the second model, is the finding that when corporate misconduct is added to the relation between political ideology and external CSR, this relation becomes of influence. Similar to the model including internal CSR, the assumption that the mechanism should be turned around seems the most plausible. When done, this turned model can be supported by the evidence. When a company encounters corporate misconduct, a more liberal board of directors would strengthen the increase in external CSR practices. This can be reasoned to be the case on basis of the ‘harm’ value, which is more important to liberals than to
29 board to act in ways to resolve this harm. The resolving of this harm could include external CSR as well.
The combination of the two models report some interesting implications as well. Two of which are most notable. First of all, the direct influence of corporate misconduct seems stronger on external CSR than on internal CSR. This would imply that when corporate misconduct is encountered, a firm’s increased CSR practices could include more
communicating about the practices than actual improvements in CSR actions. Other research must therefore be aware of this difference, before hypothesizing that overall increases in CSR practices means an increase in actual CSR actions. A critical review of the company’s increases
in CSR following corporate misconduct is suggested.
30
Practical implications
As on a managerial level, corporate social responsibility bears a lot of importance (Rao & Tilt, 2016; McElhaney, 2009; McWilliams et al., 2006), the findings of this study can be of certain use for managerial decision-making. An important note is that managers in some cases should be aware of the consequences which their decisions could have in the future, if a factor changes. The most important example in this study of such a risky decision, is when board member adjustments are made.
Although it probably isn’t the intention of a company, it might expect to deal with corporate misconduct in the current or coming years. When such cases occur and there is an opportunity to influence the political ideology of the company’s board (either adding/removing
a member or by replacing one), there should be chosen in favour of a more liberal board if the intention is to increase the CSR actions of the company in times of crisis. If there is no corporate misconduct exposure and there is a strong suspicion that corporate misconduct will not be of influence in the coming years, the fact whether a board member is liberal or conservative does not influence the company’s CSR actions.
Other results regarding internal CSR that were found to be of influence, are the influences of firm size and firm profitability. Both the firm size and firm profitability, as in its market value and gross profit, have been found to be of positive influence regarding the company’s actual CSR actions. As for most managers it includes their core job description to
try their best to increase these factors, it is hard to say if these factors will influence decision making. What they nevertheless can provide, is guidance regarding the drivers of internal CSR.
On the other side of CSR, the external one, the findings are similar. The findings show that when a company encounters corporate misconduct, the communication about the company’s CSR action increases. This increase is larger when the company’s board of directors
31 political ideology can be influenced , it should be influenced in favour of a more liberal board to increase CSR communication towards the environment in times of crisis. Another finding regarding the external CSR practices that stands out is the influence of firm size on external CSR. A larger firm size, measured as the market value of a company, is found to increase a company’s external CSR practices.
Limitations and future research
Although the findings are interesting, there are some drawbacks to the results. As heteroscedasticity and first-order autocorrelation were found in the data, the given standard errors in the models can differ from the true standard errors. Regarding the normal distribution of the data, even though it was expected to be non-normally distributed, some tests/ regressions are based on assumptions of normal distribution, therefore it has to be noticed that these results are on the basis of non-normally distributed data. Collinearity was not found to be an issue for the key variables in this study, but as past research has stated that the kind of industry could be relevant in the context of this study (Wanderlay et al., 2008) it might be desirable for future research to explore ways to include it.
The setting of this study might have some drawbacks as well. To test the hypotheses, data about the largest firms in the United States of America has been studied. This does not necessarily mean that the findings are generalizable across the globe. It could be the case in other parts of the world or in other cultures, political ideology or corporate misconduct have different influences on the forms of CSR. Findings might also differ if research would not only look at the largest firms in a country, but would take a more representative sample of the country’s firms. Lastly, a remark has to be made about the way in which corporate misconduct
32 firms can be expected to try to cover up cases of misconduct. This would mean that not all actual cases are represented in this study.
33
CONCLUSION
Research argues that corporate social responsibility is of great importance to a firm’s survival
(Rao & Tilt, 2016; McElhaney, 2009; McWilliams et al., 2006). Especially when its two components, internal and external CSR, are aligned (Hawn & Ioannou, 2016). This study shows that a more liberal political ideology does not necessarily have to increase internal and / or external CSR practices, as no such evidence was found. The findings also indicate that corporate misconduct positively influences both internal and external CSR, with a slightly stronger influence on external CSR. In addition, this study indicates that when corporate misconduct is included in the relation between political ideology and both internal and external CSR, these relations become significant. This might suggest that the mechanism should be turned around, with political ideology as moderator of the relation between
corporate misconduct and both internal and external CSR. In case the mechanism was turned around, the findings support the assumption that a more liberal board strengthens the
34
REFERENCES
Badr, H. A., Gray, E. R., & Kedia, B. L. (1982). Personal values and managerial decision making: Evidence from two cultures. Management International Review, 65-73. Bae, J., Choi, W., & Lim, J. (2020). Corporate social responsibility: An umbrella or a puddle
on a rainy day? Evidence surrounding corporate financial misconduct. European
Financial Management.
Baumann-Pauly, D., Wickert, C., Spence, L. J., & Scherer, A. G. (2013). Organizing corporate social responsibility in small and large firms: Size matters. Journal of
Business Ethics, 115(4), 693-705.
Blau, J. H. (1975). Liberal values and independence. The review of economic studies, 42(3), 395-401.
Brennan, M. C. (1995). Turning right in the sixties: The conservative capture of the GOP. University of North Carolina Press.
Carberry, E. J., Engelen, P. J., & Van Essen, M. (2018). Which firms get punished for unethical behavior? Explaining variation in stock market reactions to corporate misconduct. Business Ethics Quarterly, 28(2), 119-151.
Carlos, W. C., & Lewis, B. W. (2018). Strategic silence: Withholding certification status as a hypocrisy avoidance tactic. Administrative Science Quarterly, 63(1), 130-169.
Chin, M. K., Hambrick, D. C., & Treviño, L. K. (2013). Political ideologies of CEOs: The influence of executives’ values on corporate social responsibility. Administrative
Science Quarterly, 58(2), 197-232.
Colgate Palmolive, viewed on 28-04-2020. Retrieved from:
https://www.colgatepalmolive.com/en-us/about/governance/independent-board-candidate-qualifications
Davies, G., & Olmedo-Cifuentes, I. (2016). Corporate misconduct and the loss of trust.
European Journal of Marketing.
Du, X. (2015). Is corporate philanthropy used as environmental misconduct dressing? Evidence from Chinese family-owned firms. Journal of Business Ethics, 129(2), 341-361.
Digital marketing Institute, viewed on 28-04-2020. Retrieved from:
https://digitalmarketinginstitute.com/blog/corporate-16-brands-doing-corporate-social-responsibility-successfully
Feather, N. T. (1979). Value correlates of conservatism. Journal of Personality and Social
Psychology, 37(9), 1617.
Financial Times, viewed on 18-01-2020. Retrieved from:
35 Fritzsche, D., & Oz, E. (2007). Personal values’ influence on the ethical dimension of
decision making. Journal of Business Ethics, 75(4), 335-343.
Global rates, viewed on 20-01-2020. Retrieved from: https://www.global-rates.com/interest-rates/central-banks/central-banks.aspx
Goren, P., C. M. Federico, and M. C. Kittilson (2009) Source cues, partisan identities, and political value expression. American Journal of Political Science, 53: 805–820. Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets
of moral foundations. Journal of personality and social psychology, 96(5), 1029. Grewatsch, S., & Kleindienst, I. (2017). When does it pay to be good? Moderators and
mediators in the corporate sustainability–corporate financial performance relationship: A critical review. Journal of Business Ethics, 145(2), 383-416.
Gupta, A., Nadkarni, S., & Mariam, M. (2019). Dispositional sources of managerial discretion: CEO ideology, CEO personality, and firm strategies. Administrative
Science Quarterly, 64(4), 855-893.
Hawn, O., & Ioannou, I. (2016). Mind the gap: The interplay between external and internal actions in the case of corporate social responsibility. Strategic Management Journal, 37(13), 2569-2588.
Hofmann, K. H., Theyel, G., & Wood, C. H. (2012). Identifying firm capabilities as drivers of environmental management and sustainability practices–evidence from small and medium‐sized manufacturers. Business Strategy and the Environment, 21(8), 530-545. Jiang, F., Zalan, T., Herman, H. M., & Shen, J. (2018). Mapping the relationship among
political ideology, CSR mindset, and CSR strategy: A contingency perspective applied to Chinese managers. Journal of Business Ethics, 147(2), 419-444.
Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F. J. (2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129(3), 339–375.
Kassinis, G., Panayiotou, A., Dimou, A., & Katsifaraki, G. (2016). Gender and environmental sustainability: A longitudinal analysis. Corporate Social Responsibility and
Environmental Management, 23(6), 399-412.
Kekes, J. (1997). What is conservatism?. Philosophy, 72(281), 351-374.
Li, D., He, J., & Jiang, J. (2019, July). Big Egos Can Cause Natural Disasters: How CEO Overconfidence Leads to Environmental Misconducts. In Academy of Management Proceedings (Vol. 2019, No. 1, p. 13261). Briarcliff Manor, NY 10510: Academy of
Management.
Liston-Heyes, C., & Ceton, G. C. (2007). Corporate social performance and politics: Do liberals do more?. Journal of Corporate Citizenship, (25), 95-108.
Liu, X. (2016). Corruption culture and corporate misconduct. Journal of Financial
36 Marquis, C., & Qian, C. (2014). Corporate social responsibility reporting in China: Symbol or
substance?. Organization science, 25(1), 127-148.
McArdle, B. (1994). Multivariate analysis: a biologist's guide. Department of Marine Biology, James Cook University, Australia.
Murphy, D. L., Shrieves, R. E., & Tibbs, S. L. (2009). Understanding the penalties associated with corporate misconduct: An empirical examination of earnings and risk. Journal of
Financial and Quantitative Analysis, 55-83.
NPR, viewed on 28-04-2020. Retrieved from:
https://www.npr.org/2016/04/30/475794063/why-are-highly-educated-americans-getting-more-liberal?t=1588071857717
Oktavianawati, L. (2018). The Factors that Influence the Disclosure of Corporate Social Responsibility (CSR). Accounting Analysis Journal, 7(2), 119-126.
Park, U. D., Boeker, W., & Gomulya, D. (2020). Political ideology of the board and CEO dismissal following financial misconduct. Strategic Management Journal.
PEW, viewed on 20-01-2020. Retrieved from:
https://www.people-press.org/2014/06/12/political-polarization-in-the-american-public/
Poole, K. T., and H. Rosenthal (1984). The polarization of American politics. Journal of
Politics, 46: 1061–1079.
Post C, Rahman N, Rubow E. 2011. Green governance: Boards of directors’ composition and environmental corporate social responsibility. Business & Society 50(1): 189–223. Rao, K., & Tilt, C. (2016). Board composition and corporate social responsibility: The role of
diversity, gender, strategy and decision making. Journal of Business Ethics, 138(2), 327-347.
Schwartz, S. (2013). Value priorities and behaviour: Applying. In The psychology of values: The Ontario symposium (Vol. 8).
Scott, S. (2007). Corporate social responsibility and the fetter of profitability. Social
Responsibility Journal.
Tadelis, S. (1999). What's in a Name? Reputation as a Tradeable Asset. American Economic
Review, 89(3), 548-563.
The Asset, viewed on 27-04-2020. Retrieved from:
https://esg.theasset.com/ESG/40157/coronavirus-raises-the-focus-on-social-responsibility-in-the-asia-pacific
von Kuehnelt-Leddihn, E. (1997). Liberalism in America. The Intercollegiate Review, 33(1), 44.
Walters, K. D. (1977). Corporate social responsibility and political ideology. California
37 Wanderley, L. S. O., Lucian, R., Farache, F., & de Sousa Filho, J. M. (2008). CSR
information disclosure on the web: a context-based approach analysing the influence of country of origin and industry sector. Journal of business ethics, 82(2), 369-378. Wickert, C., Scherer, A. G., & Spence, L. J. (2016). Walking and talking corporate social
responsibility: Implications of firm size and organizational cost. Journal of
38
APPENDIX A: Internal CSR variables
40
APPENDIX A: External CSR variables
41
APPENDIX A: Corporate misconduct variables
