The Fordham Environmental Law Review presents Corporate ... · dismissal. These findings provide novel insight into how CEOs’ career outcomes may be affected by earlier CSR decisions

CLE MATERIALSFriday, March 9, 2018 Fordham Law School

Gorman Moot Courtroom 150 W 62nd Street

New York, NY 10023

The Fordham Environmental Law Review presents

Corporate Sustainabilityin the Era of

Shifting Federal Priorities

Table of Contents

CLE Materials

Panel 1

The Rise of CSR & the Role of Lawyers as CSR Advisors

Panel 2

Sustainability & Big Business

Strategic Management JournalStrat. Mgmt. J., 38: 2255–2265 (2017)

Published online EarlyView 20 March 2017 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.2646Received 27 January 2016; Final revision received 7 November 2016

Higher Highs and Lower Lows: The Role of CorporateSocial Responsibility in CEO Dismissal

Timothy D. Hubbard,1* Dane M. Christensen,2 and Scott D. Graffin3

1 Mendoza College of Business, University of Notre Dame, Notre Dame, Indiana2 Lundquist College of Business, University of Oregon, Eugene, Oregon3 Terry College of Business, University of Georgia, Athens, Georgia

Research summary: Investing a firm’s resources in corporate social responsibility (CSR)initiatives remains a contentious issue. While research suggests firm financial performance is theprimary driver of CEO dismissal, we propose that CSR will provide important additional contextwhen interpreting a firm’s financial performance. Consistent with this prediction, our resultssuggest that past CSR decisions amplify the negative relationship between financial performanceand CEO dismissal. Specifically, we find that greater prior investments in CSR appear to exposeCEOs of firms with poor financial performance to a greater risk of dismissal. In contrast, greaterpast investments in CSR appear to help shield CEOs of firms with good financial performance fromdismissal. These findings provide novel insight into how CEOs’ career outcomes may be affectedby earlier CSR decisions.

Managerial summary: In this study, we examined a potential personal consequence for CEOsrelated to corporate social responsibility (CSR). We explored the role prior investments in CSRplay when a board evaluates the firm’s financial performance and considers whether or not to firethe CEO. Our results suggest that while financial performance sets the overall tone of a CEO’sevaluation, CSR amplifies that baseline evaluation. Specifically, our results suggest that greaterpast investments in CSR appear to (a) greatly increase the likelihood of CEO dismissal whenfinancial performance is poor, and (b) somewhat reduce the likelihood of CEO dismissal whenfinancial performance is good. Thus, striving to deliver profits in a socially responsible mannermay have both positive and negative personal consequences. Copyright © 2017 John Wiley &Sons, Ltd.

Introduction

A CEO’s primary objective is to generate eco-nomic returns for shareholders (Quigley & Ham-brick, 2015). Consistent with this idea, firm finan-cial performance is the primary metric by whichCEOs are evaluated (Graffin, Boivie, & Carpenter,2013) and it is also the strongest predictor of CEO

Keywords: CEO dismissal; corporate social responsibil-ity; financial performance; panel regression; stakeholders*Correspondence to: Timothy D. Hubbard, Mendoza College ofBusiness, University of Notre Dame, Notre Dame, IN 46556.E-mail: [email protected]

Copyright © 2017 John Wiley & Sons, Ltd.

termination (Finkelstein, Hambrick, & Cannella,2009). Some argue, however, that these economicreturns should be pursued in a socially responsi-ble manner (Freeman, 1984), thereby suggestingthat corporate social responsibility (CSR) is a sec-ondary objective on which CEOs should focus. As aresult, some CEOs have begun supporting CSR ini-tiatives, committing significant corporate resources(e.g., money, time, attention) to help address socialand environmental problems.

Whether or not CSR activities are in the bestinterest of the firm, however, remains a contentiousissue. Proponents of CSR suggest it can help

http://orcid.org/0000-0002-6029-7446

http://orcid.org/0000-0003-4841-7238

2256 T. D. Hubbard, D. M. Christensen, and S. D. Graffin

increase shareholder value and build support amongstakeholders, such as employees, customers, andcommunity members (Hillman & Keim, 2001). Incontrast, critics of CSR suggest it benefits thesestakeholders at the expense of shareholders, whichconflicts with meeting the primary goal of increas-ing shareholder value (e.g., Friedman, 1970; Mar-golis & Walsh, 2003; Petrenko, Aime, Ridge, &Hill, 2016). This controversy is aggravated by thefact that research has been unable to provide aclear, consistent link between CSR and firm finan-cial performance (e.g., McWilliams & Siegel, 2000;Waddock & Graves, 1997; Wright & Ferris, 1997).These inconclusive findings on the financial bene-fits of CSR thus provide no clear guidance regardinghow CSR may influence a CEO’s evaluation. Givensuch controversies and our limited knowledge ofthe personal consequences CEOs face as a result ofengaging in CSR, we explore the role prior invest-ments in CSR play when a board evaluates the firm’sfinancial performance and considers whether or notto fire the CEO.

While firm financial performance is of primaryimportance in evaluating a CEO, we theorize CSRis an important secondary objective that will influ-ence how those financial returns are interpretedby board members. Our focus on CSR as a sec-ondary goal stems from its unique characteristics,particularly its visibility, its likelihood of attractingscrutiny by outside groups, and its contestability. Interms of its visibility and scrutiny, CSR is widelytracked and interpreted by third-party rating agen-cies, such as MSCI (i.e., KLD), and over 4 trilliondollars are invested in socially responsible invest-ment funds (Social Investment Forum, 2014). Cor-porate social responsibility initiatives also receiveglobal attention, as over 8,000 firms from more than160 countries have signed onto the United NationsGlobal Compact, which commits firms to incorpo-rating CSR into their business practices (UN GlobalCompact, 2014). In response to greater demand forCSR information, firms now often issue standalonereports on their CSR performance (Christensen,2016). Despite this visibility and scrutiny, CSR ini-tiatives are more complex to evaluate than othertypes of corporate initiatives (e.g., R&D or capi-tal investments) because CSR may be motivated byconcerns unrelated to enhancing shareholder value(Wang, Tong, Takeuchi, & George, 2016). Theseother concerns may include reducing carbon out-put, promoting diversity, or building support amongstakeholders beyond shareholders. Measuring CSR

initiatives’ overall benefit is thus difficult, makingCSR more contestable than more traditional invest-ments, which are motivated almost exclusively byeconomic concerns. We suggest that the visibilityand scrutiny of CSR will make it an important com-ponent in assessing a CEO, while its contestabilitymeans that its association with CEO dismissal willlikely be more nuanced than direct.

We thus suggest that prior investments in CSRwill influence how financial performance is inter-preted, and thereby moderate the effect of finan-cial performance on CEO dismissal. Specifically,we argue that while financial performance sets theoverall tone of a CEO’s evaluation, CSR amplifiesthat baseline evaluation. On the one hand, a highlevel of prior investments in CSR can amplify thepositive framing of good financial performance andmake it appear that the CEO can not only achievethe primary goal of generating economic returns,but can also do so in a socially responsible mannerthat is salient to multiple stakeholders. On the otherhand, such investments may amplify the negativeframing of poor financial performance and makethe CEO appear to have dedicated too much of thefirm’s resources to the secondary goal of CSR andsupporting other stakeholders, and not enough tothe primary goal of generating economic returns forshareholders. Consistent with this prediction, ourresults suggest that greater past investments in CSRamplify the likelihood that CEOs of poorly per-forming firms are dismissed, whereas greater pastinvestments in CSR appear to reduce the likelihoodof dismissal for CEOs of well-performing firms.This article thus contributes to the literature by pro-viding a better understanding of how CSR maybe associated with CEO career outcomes. Specifi-cally, our results suggest that prior investments inCSR amplify the most well-documented relation-ship in this literature—firm financial performanceand CEO dismissal.

Theory and Hypothesis Development

A firm’s financial performance is the primarymetric by which its board assesses the CEO. Thetendency to attribute firm performance primarilyto the quality of a CEO results in the CEO’scontinuing employment prospects being heavilydependent upon firm performance (Finkelsteinet al., 2009). Indeed, the influence of firm financialperformance on CEO dismissal has been found

Copyright © 2017 John Wiley & Sons, Ltd. Strat. Mgmt. J., 38: 2255–2265 (2017)DOI: 10.1002/smj

The Role of Corporate Social Responsibility in CEO Dismissal 2257

in multiple samples across decades of research(Finkelstein et al., 2009).

While this conclusion has garnered consistentsupport over time, it explains only a small portionof the variance in CEO dismissals (Finkelstein et al.,2009). To help provide a better understanding ofwhen CEOs are more or less likely to be dismissed,we suggest that prior investments in CSR may mod-erate the well-established relationship between firmfinancial performance and CEO dismissal due to itsvisibility, scrutiny, and contestability. The visibilityand scrutiny associated with CSR ensure that suchinitiatives will be well known and thoroughly ana-lyzed, while its contestability suggests that the valueof CSR is open to interpretation.

In recent years, the visibility of CSR hasincreased substantially (Wang et al., 2016).Socially responsible investment funds now holdtrillions of dollars (Social Investment Forum, 2014)and various stakeholders have increased their focuson CSR (UN Global Compact-Accenture, 2010).The increasing visibility of CSR is thought tobe driven by growing mass media coverage ofCSR initiatives, the rise of advocacy groups, andincreased investments in CSR by large corporations(Wagner, Lutz, & Weitz, 2009). Further increasingits visibility is a recent spike in disclosures by firms;roughly three-quarters of all S&P 500 firms in theUnited States now publish an annual CSR report(Governance and Accountability Institute, 2015).

CSR has also been increasingly scrutinized andassessed by multiple stakeholders. Third-party rat-ing agencies, such as MSCI (i.e., KLD), Robe-coSAM (i.e., Dow Jones), and Thomson Reuters(i.e., Asset4), evaluate CSR regularly. Negative rat-ings by these agencies can lead to strong investorreactions, such as TIAA-CREF selling off over50 million shares of Coca-Cola stock following con-cerns raised by KLD (Chatterji, Levine, & Toffel,2009). Advocacy groups are also increasing theirscrutiny of CSR. For instance, nearly 40% of share-holder proposals now focus on social and environ-mental issues (Ernst & Young, 2013). Such actionssuggest that board members are quite aware ofthe degree to which nonfinancial stakeholders arepleased or frustrated with the firm’s prior invest-ments in CSR. Governmental agencies have alsoincreased their oversight of CSR. For example,India now mandates that corporations must investat least 2% of their net profit in CSR, while othercountries such as China, Denmark, Malaysia, andSouth Africa require some level of CSR reporting

(Wang et al., 2016). Thus, investors, activist groups,and governmental agencies collectively ensure thatCSR is highly scrutinized and, beyond shareholders,these other stakeholder groups can and do influenceorganizational outcomes in a visible manner.

Further, the value of CSR is contestable and opento interpretation by stakeholders because the overallimpact of CSR on firms is often difficult to quantify.Despite the fact that the relationship between CSRand financial performance has been widely studied,research has yielded conflicting results without aclear consensus. Some research suggests CSR hasa positive influence on firm financial performancebecause it can generate stronger relationships withstakeholders, increase customer loyalty, and posi-tively influence corporate reputation (Choi & Wang,2009; Hillman & Keim, 2001). Other research, how-ever, suggests that CSR initiatives hinder financialperformance (Aupperle, Carroll, & Hatfield, 1985;Jensen, 2002) and come at the expense of share-holders (Brammer & Millington, 2008; Navarro,1988). Still other research finds no relationshipbetween CSR and firm financial performance (e.g.,McWilliams & Siegel, 2000).

One potential cause for these disparate findingsmay be that quantifying the financial returns ofCSR is more difficult than quantifying the returnsfrom other forms of investments, such as capitalexpenditures, because the outcomes from CSR maysimply be more diffuse (Wang et al., 2016). As wenoted earlier, the motivations for engaging in CSRgo beyond exclusively economic considerations.While there may be positive economic returns toCSR, other benefits, such as a positive impact onemployees’ dignity, are more difficult to quantify.Thus, these complex motivations and more diffusebenefits make CSR a more contestable form ofinvestment than those that are purely economicallymotivated.

Financial Performance, Corporate SocialResponsibility, and CEO Dismissal

Given its unique characteristics, we argue that priorinvestments in CSR will not directly influenceCEO dismissal. Rather, we suggest that financialperformance—a major determinant of CEO dis-missal (Finkelstein et al., 2009) and a key referencepoint for decision makers (Greve, 1998)—willbe interpreted in light of the firm’s level of priorinvestments in CSR. Specifically, we argue thatfinancial performance will set the overall tone of



how the CEO is evaluated, and that CSR, due toits contestable nature, will amplify that baselineevaluation.

On one hand, good financial performance sup-ports the perception that the CEO is of high qual-ity. When this is combined with high levels ofprior investment in CSR, this suggests that, notonly was the CEO able to achieve the primarygoal of generating economic returns, but was alsoable meet the secondary objective of accomplishingthis in a socially responsible manner that benefitedother, nonfinancial stakeholders. We suggest that bysupporting these other stakeholders (e.g., employ-ees, community activists, customers) through priorinvestments in CSR, these groups will view the sit-ting CEO more positively. In turn, due to the vis-ibility of such stakeholders and their influence onfirm outcomes (e.g., Briscoe & Safford, 2008; Choi& Wang, 2009; Freeman, 1984; Hillman & Keim,2001), board members who are charged with retain-ing or dismissing the CEO, will evaluate the CEOeven more positively as a result of the approval ofthese nonfinancial stakeholders.

On the other hand, when a firm displays poorfinancial performance compared to other firms,this negatively influences perceptions of the CEO’squality. We suggest that such negative perceptionswill be amplified if the CEO has previously dedi-cated significant resources to CSR. Specifically, wesuggest that poor financial performance, when com-bined with high levels of prior investment in CSR,will lead the board to believe that the CEO hasinvested too much of the firm’s resources on the sec-ondary objective of CSR rather than on the firm’sprimary mission—generating economic returns forshareholders. Thus, this may lead board members toconclude that the CEO focused too much on satis-fying these other stakeholder groups at the expenseof shareholders. We thus suggest:

Hypothesis 1: CSR will moderate the negativerelationship between firm financial performanceand CEO dismissal, such that CSR amplifies thisrelationship.

Methods

Sample

Our sample consisted of Fortune 500 firms span-ning the years 2003–2008. We obtained financial

information from COMPUSTAT; executive char-acteristics, ownership, and pay from ExecuComp;CEO dismissal from an analysis of media collectedfrom LexisNexis; Corporate Social Responsibilitydata from Kinder, Lydenberg, and Domini (KLD);governance measures from Risk Metrics and Thom-son Reuters; and stock returns from the Center forResearch in Security Prices (CRSP). Private firmsand missing data reduced our sample to 441 firmsand 2,298 firm-year observations.

Dependent Variable

Our dependent variable, CEO dismissal, equals 1 ifthe CEO was fired, 0 otherwise. It was measuredusing Shen and Cannella’s (2002) method to codeinvoluntary turnover. We evaluated 339 successionsusing media articles from major U.S. newspapers inLexisNexis covering 1 year before to 1 year after thetransition. First, we eliminated successions in whichCEOs died, had health issues, were interim CEOs,accepted similar positions at other firms, or leftdue to a merger or acquisition. From the remainingsuccessions, we identified dismissals when CEOswere reported to have (a) been fired or forced out;(b) resigned immediately or unexpectedly due topoor performance, undisclosed personal reasons, ora desire to pursue other interests; (c) retired earlyamidst performance problems; or (d) left before age64 and also gave up their board seat. This methodidentified 104 CEO dismissals. After requiring con-trol variables, our final sample included 98 CEOdismissals occurring at 90 different companies.1

Independent Variable

Our independent variable is the interaction ofindustry-adjusted returns and CSR. We calcu-lated industry-adjusted returns as the firm’sindustry-adjusted annual stock return (includingdividends), where industries were classified basedon two-digit SIC codes; we found substantivelysimilar results with three-digit SIC codes. Wecalculated CSR as the net strength and weaknessrankings of five dimensions from KLD: employee,community, diversity, environment, and product(Choi & Wang, 2009; Hillman & Keim, 2001;Kang, 2013; Tang, Qian, Chen, & Shen, 2015).2

1 In our sample, no individual firm-year had more than one CEOdismissal.2 Similar to prior research, we assume that observable CSR rat-ings are positively correlated with CSR investment, which is



We used KLD ratings as they are “broadly regardedas the most comprehensive data available tomeasure CSR” (Petrenko et al., 2016).

Control Variables

We included several control variables to captureother factors that may influence the likelihood ofdismissal. First, we included firm size, measuredas the natural log of firm assets, to account forgreater expectations for CEOs at larger firms (Shen& Cannella, 2002). Second, we controlled for CEOcharacteristics that could influence the likelihood ofdismissal. CEO tenure and CEO duality help cap-ture the power and influence the CEO has in theboardroom (Shen & Cannella, 2002). CEO tenureis measured as the number of years the CEO hasbeen in office at the firm. CEO duality takes avalue of 1 if the CEO chairs the board, 0 otherwise(Shen & Cannella, 2002). CEO pay helps capturethe likelihood that the CEO will attract greater out-side scrutiny and pressure as pay increases, which ismeasured as the natural log of total current compen-sation (Cai, Jo, & Pan, 2011). Third, we controlledfor corporate governance characteristics using insti-tutional ownership, calculated as the percentage ofthe firm’s shares owned by institutional investors(Parrino, Sias, & Starks, 2003), and the numberof blockholders, measured as the count of own-ers with at least 5% of the firm’s shares. Finally,we controlled for the firm’s R&D intensity, capitalintensity, and market-to-book ratio, to capture dif-ferences in the firm’s operating strategies (Ioannou& Serafeim, 2015; McWilliams & Siegel, 2000).R&D intensity was calculated as R&D expensedivided by sales, and set to zero if missing; Capexwas calculated as capital expenditures scaled bysales; Market-to-book was calculated as the mar-ket value of equity divided by the book value ofequity.

generally unobservable. While there are a number of dimensionsof CSR available from KLD, the five that we chose are com-monly used in the literature, and are classified as stakeholdermanagement dimensions. Other dimensions—such as alcohol,tobacco, gambling—are typically considered social issue partic-ipation, and are commonly excluded from measures of CSR (cf.Hillman & Keim, 2001). We reran the analyses including all KLDdimensions in our CSR measure and found substantively similarresults. We also examined the individual dimensions of CSR andfound that no single dimension drives the overall results, suggest-ing the overall measure is most appropriate. Furthermore, using adecile ranked measure of CSR also produced similar results.

Estimation

We used random-effects probit regression with stan-dard errors clustered by firm to estimate our binaryCEO dismissal model.3 All regressions includedyear fixed-effects, and all independent and controlvariables have been lagged by 1 year.

Results

Table 1 provides descriptive statistics and correla-tions for the variables in our models. We also cal-culated variance inflation factors and found that allvalues were less than two, indicating that multi-collinearity was not an issue (Kennedy, 2008).

Table 2 reports the results of our random-effectsprobit regressions predicting CEO dismissal. Model1 presents the results with only the control vari-ables, while Model 2 includes both componentsof our interaction term, and Model 3 presents thefull model with the interaction term. Hypothesis1 predicted that CSR would moderate the relationbetween financial performance and CEO dismissal.The results of Model 3 show that the coefficient onthe interaction term is negative with a high likeli-hood that its value differs from zero (𝛽 =−0.174,p= .002). We plot the interaction and provide sim-ple slopes in Figure 1.

At low levels of financial performance (one stan-dard deviation below the mean), increasing priorinvestments in CSR from one standard deviationbelow the mean to one standard deviation above themean is associated with an increased likelihood ofdismissal by 84% (from 4.15 to 7.64%, p= .040).At extremely low levels of financial performance(two standard deviations below the mean) increas-ing prior investments in CSR from one standarddeviation below the mean to one standard deviationabove the mean is associated with a 206% increasein the likelihood of dismissal (from 4.82 to 14.7%,p= .015).

At high levels of financial performance (one stan-dard deviation above the mean), increasing priorinvestments in CSR from one standard deviationbelow the mean to one standard deviation abovethe mean is associated with a 53% reduction in thelikelihood of a CEO’s dismissal (from 3.05% down

3 Indistinguishable results and conclusions are drawn usingrandom-effects logistic regression. Moreover, mean-centeringindustry-adjusted returns and CSR does not substantively changethe results.


2260T.D

.Hubbard,D

.M.C

hristensen,andS.D

.Graffin

Table 1Descriptive Statistics and Correlations

Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12

1. CEO dismissal 0.043 0.2022. CSR 0.398 3.161 0.0363. Industry-adjusted returns −0.027 0.282 −0.075 −0.0384. Industry-adjusted returns × CSR −0.045 0.821 −0.069 −0.146 0.0335. Firm size 9.632 1.371 0.045 0.165 −0.045 −0.0266. Capex intensity 0.056 0.078 −0.025 −0.132 −0.043 0.026 0.0697. R&D intensity 0.018 0.042 0.012 0.313 −0.041 −0.088 0.018 0.0468. Market-to-book 3.106 3.282 −0.002 0.170 0.076 −0.062 −0.110 −0.069 0.1289. CEO pay 8.862 0.925 −0.001 0.124 −0.012 −0.030 0.402 −0.009 0.080 0.04810. CEO tenure 5.582 6.062 0.009 −0.009 −0.028 0.007 −0.040 0.002 −0.024 −0.018 0.03111. CEO duality 0.717 0.450 −0.030 0.008 −0.016 0.022 0.140 0.040 −0.032 0.005 0.185 0.24012. Number of blockholders 1.891 1.454 −0.001 −0.133 −0.016 0.018 −0.310 −0.046 −0.102 −0.111 −0.137 0.001 −0.05813. Institutional ownership 0.741 0.162 −0.002 −0.048 0.066 −0.005 −0.297 −0.103 −0.084 0.000 −0.017 −0.008 −0.035 0.634

N = 2,298 firm-year observations.

to1.42%

,p=

.047).T

hisassociation

increasesto

an81%

reductionin

likelihoodof

dismissal

(from2.60%

down

to0.050%

,p=

.030)at

extremely

highlevels

offinancial

performance,

two

standarddeviations

abovethe

mean.

Overall,

thefindings

areconsistent

with

therelationship

suggestedby

Hypothesis

1. 4

SupplementalA

nalyses

We

conductedseveral

supplemental

analysesto

assessthe

robustnessof

ourfindings.

First,since

boardsm

aycom

parethe

firm’s

CSR

toother

firms

inthe

industryw

henevaluating

theC

EO

,w

ereran

ouranalysis

usingan

industry-adjustedC

SRm

easure.T

heresults

arereported

inM

odel4

ofTable

1,and

areconsistent

with

ourprior

results(𝛽

=−

0.163,p=.006).

Second,to

ruleout

thepossibility

thata

time-invariant

omitted

variablem

aybe

drivingour

results,w

ereran

ouranalyses

usingthree

models

thatinclude

fixed-effects.W

efirst

duplicatedour

primary

analysisincluding

controlsfor

industryfixed-effects.T

heresults

remain

unchanged(M

odel5

ofTable

2:𝛽=−

0.180,p=.003).N

ext,we

rana

conditionallogisticregression

with

clusteredrobust

standarderrors

includingfirm

fixed-effects(S.

J.L

ong,1997).This

producedresults

consistentwith

ourm

ainanalysis

(Model6

ofTable

2:𝛽=−

0.259,p=

.041).W

hileaccounting

forthe

dichotomous

dependentvariable,

thism

ethodreduced

oursam

pleto

onlythose

firms

thatexperienced

aC

EO

dismissal.

Toovercom

ethis

limitation,

we

rana

linearfixed-effects

regressionw

ithclustered

robuststandard

errors(K

ennedy,2008),

andagain

obtainedconsistent

results(M

odel8

ofTable

2:𝛽=−

0.012,p=

.036). 5T

heseanalyses

areconsis-

tentw

ithour

main

findingsin

thepresence

offirm

fixed-effects.T

hird,w

eran

two

teststo

assessthe

possi-bility

thata

time-varying

omitted

variablem

ay

4T

heseeffect

sizesw

ereof

consistenteconom

icm

agnitudeacross

estimation

methods.

Summ

arystatistics

froma

2×2

analysisof

dismissal

ratesbased

onhigh/low

CSR

andhigh/low

industry-adjustedreturns

yieldsim

ilarinferences.

Specifically,w

henfinancial

performance

isbelow

them

edian,and

CSR

isabove

(below)

them

edian,CE

Odism

issalratesare

6.6%(5.2%

).W

henfinancial

performance

isabove

them

edian,and

CSR

isabove

(below)

them

edian,CE

Odism

issalratesare

2.5%(2.8%

).5

Forreference,a

pooledlinear

modelw

ithoutfirmfixed-effects

isshow

nin

Model7

ofTable

2.

Copyright©

2017John

Wiley

&Sons,L

td.Strat.M

gmt.J.,38:2255

–2265

(2017)D

OI:10.1002/sm

j

The

Role

ofCorporate

SocialResponsibility

inC

EO

Dism

issal2261

Table 2Results of Regressions Predicting CEO Dismissal

Random-effects probit Conditional logit Linear probability model Survival model

Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9

Firm size 0.110 0.099 0.100 0.102 0.114 2.686 0.010 0.074 0.155(0.012) (0.020) (0.019) (0.016) (0.053) (0.019) (0.043) (0.001) (0.047)

Capex intensity −0.788 −0.852 −0.923 −0.949 −0.266 0.776 −0.060 0.066 −2.247(0.208) (0.184) (0.153) (0.140) (0.794) (0.854) (0.138) (0.738) (0.101)

R&D intensity 0.837 0.108 −0.203 −0.032 1.468 −2.847 0.008 −0.065 −0.404(0.432) (0.921) (0.854) (0.977) (0.248) (0.746) (0.945) (0.865) (0.881)

Market-to-book 0.005 0.006 0.004 0.005 −0.002 0.050 0.000 0.000 −0.003(0.732) (0.660) (0.746) (0.700) (0.916) (0.613) (0.714) (0.940) (0.924)

CEO pay −0.046 −0.050 −0.048 −0.048 −0.026 0.202 −0.005 −0.001 −0.067(0.348) (0.333) (0.354) (0.361) (0.678) (0.272) (0.273) (0.851) (0.557)

CEO duality −0.205 −0.206 −0.206 −0.206 −0.194 −0.967 −0.018 −0.050 −0.474(0.054) (0.053) (0.055) (0.054) (0.120) (0.012) (0.090) (0.012) (0.031)

Number of blockholders 0.028 0.014 0.013 0.013 −0.013 −0.071 0.001 −0.005 0.046(0.489) (0.724) (0.753) (0.756) (0.763) (0.598) (0.732) (0.328) (0.596)

Institutional ownership 0.184 0.259 0.250 0.260 0.559 3.756 0.023 0.158 −0.051(0.678) (0.556) (0.570) (0.553) (0.237) (0.030) (0.566) (0.070) (0.958)

CEO tenure 0.007 0.007 0.008 0.008 0.009 0.184 0.001 0.008 n/a(0.311) (0.312) (0.288) (0.291) (0.337) (0.133) (0.313) (0.000) n/a

Industry-adjusted returns −0.791 −0.730 −0.722 −0.803 −1.049 −0.050 −0.041 −1.688(0.001) (0.001) (0.001) (0.000) (0.057) (0.000) (0.009) (0.001)

CSR 0.013 −0.006 −0.010 −0.015 −0.040 0.001 −0.001 −0.015(0.457) (0.750) (0.627) (0.460) (0.638) (0.568) (0.796) (0.713)

Industry-adjusted returns×CSR (H1) −0.174 −0.163 −0.180 −0.259 −0.015 −0.012 −0.393(0.002) (0.006) (0.003) (0.041) (0.003) (0.036) (0.001)

Year fixed-effects Yes Yes Yes Yes Yes Yes Yes Yes YesIndustry-adjusted CSR No No No Yes No No No No NoIndustry fixed-effects No No No No Yes No No No NoFirm fixed-effects No No No No No Yes No Yes NoConstant −2.453 −2.387 −2.395 −2.428 −2.600 n/a −0.014 −0.759 n/a

(0.000) (0.000) (0.000) (0.000) (0.004) n/a (0.809) (0.001) n/aPseudo-R2/Adj. R2 0.038 0.083 0.093 0.090 0.163 0.152 0.017 0.219 n/aObservations 2,298 2,298 2,298 2,298 1,917 499 2,298 2,298 2,298Unit of Observation FirmYear FirmYear FirmYear FirmYear FirmYear FirmYear FirmYear FirmYear CEOYear# of unique firms/CEOs 441 441 441 441 375 90 441 441 682

Two-tailed p values are shown in parentheses below coefficient estimates. All models use robust standard errors clustered by firm. The survival model incorporates CEO tenure in the hazardfunction. Model 4 employs an industry-adjusted CSR measure, which was calculated using two-digit SIC codes. The variable of interest is shown in bold font.

Copyright©

2017John

Wiley

&Sons,L

td.Strat.M

gmt.J.,38:2255

–2265

(2017)D

OI:10.1002/sm

j


0.000

0.025

0.050

0.075

0.100

0.125

0.150

Lik

elih

ood

of C

EO

Dis

mis

sal

-0.60 -0.30 0.00 0.30 0.60

Financial Performance

Low CSR (–1 s.d.) Mean CSR High CSR (+1 s.d.)

Simple slope @ High CSR = –0.105 (p = 0.000)Simple slope @ Mean CSR = –0.062 (p = 0.001)Simple slope @ Low CSR = –0.019 (p = 0.311)

Figure 1. Interaction plot of marginal effects for financial performance (i.e., industry-adjusted returns) and corporatesocial responsibility (CSR) predicting CEO dismissal. Simple slopes are measured at mean levels of financial performanceand are shown with two-tailed p values testing whether the slope is statistically different from zero.

be driving our results. First, we reran our analy-ses with additional controls: stock return volatility,return on assets (ROA), strategic change and strate-gic deviation, board independence, CEO age, CEOownership, the entrenchment index, firm age, andleverage. Our conclusions remained unchanged.6

Next, we wanted to assess how strong a corre-lated omitted variable would have to be to overturnour results, so we calculated the impact thresholdof a confounding variable (ITCV) for our inter-action term (Frank, 2000).7 The results show anITCV of −0.011, which implies that partial corre-lations between Industry-adjusted returns*CSR andCEO dismissal with an omitted confounding vari-able would have to be about 0.104 (=

√0.011) to

overturn the results. To put this in perspective, itwould take a correlated omitted variable with animpact nearly as large as the strongest variable inthis model to overturn the results. Assuming thatwe have a reasonable set of control variables, thissuggests that the results are not likely driven by acorrelated omitted variable.

Fourth, we evaluated the potential endoge-nous nature of our two independent variables

6 Results of all untabulated supplemental analyses can be obtainedfrom the first author upon request.7 To calculate the ITCV (Frank, 2000; Larcker & Rusticus, 2010),we used an expanded version of Model 7 in Table 2, whichalso included additional interaction terms (Maroulis & Gomez,2008). Specifically, we interacted Industry-adjusted returns withall the control variables in the model (e.g., R&D intensity×Industry-adjusted returns) to provide a better benchmark againstwhich to compare the ITCV. Using this model, we found similarresults to those reported in Model 8 and calculated that the ITCVfor Industry-adjusted returns*CSR is −0.011.

by assessing whether (a) prior investments inCSR led to industry-adjusted returns or (b)industry-adjusted returns led to CSR in our sample.We ran two linear fixed-effects models—oneusing CSR to predict industry-adjusted returnsand one using industry-adjusted returns to predictCSR—including clustered robust standard errorsand the same control variables as our main models.The results suggest that CSR did not help explainindustry-adjusted returns (𝛽 =−0.002, p= .777),nor did industry-adjusted returns help explain CSR(𝛽 = 0.130, p= .318).

Fifth, we considered if the hypothesized prod-uct term—Industry-adjusted returns×CSR—isendogenous by using a two-stage least squaresinstrumental variables method proposed byWooldridge (2003), which accounts for endoge-nous product terms. We used the political leaningsof the state where the firm is headquartered, mea-sured as whether the state voted for a Democrat inthe prior presidential election, as the instrument.While no instrument is perfect, we suggest thatfirms in these states are likely to have higher levelsof CSR, while it is not clear why the local politicalviews would influence CEO dismissal, making it apotentially valid instrument. Further, it is predictiveof CSR (𝛽 = 1.136, p= .000), but not of CEOdismissal (𝛽 =−0.012, p= .909). Results from thesecond stage of this test show that the coefficient onthe interaction term is consistent with our primaryfindings (𝛽 =−0.032, p= .008).

Sixth, because the likelihood of dismissalchanges over CEO tenure, we reran our mainanalysis using a Cox proportional hazard event



history model, which incorporates CEO tenurein the hazard function (Jenter & Kanaan, 2015;Singer & Willett, 2003). As shown in Model 9 ofTable 1, we again found similar results (𝛽 =−0.393,p= .001).

Finally, as a placebo test, we considered vol-untary CEO turnover as a dependent variable, aswe would not expect to find any results in thissetting. We coded for voluntary CEO turnover byexcluding departures that were due to sickness,death, interim CEOs, or due to involuntary dis-missal, which identified 141 voluntary turnovers.When we reran our primary analyses with thisdependent variable, our results show the coefficientof industry-adjusted returns (𝛽 = 0.24, p= .675),CSR (𝛽 =−0.01, p= .596), and the interaction(𝛽 = 0.006, p= .902) all match our expectationsthat there is no meaningful effect. Thus, ourcombined main and supplemental analyses provideconsistent evidence suggesting that CSR moderatesthe relationship between firm financial performanceand CEO dismissal.

Discussion and Summary

In this study, we sought to explain the potentialpersonal consequences for CEOs related to corpo-rate social responsibility (CSR). Broadly, our find-ings suggest that prior investments in CSR amplifythe relationship between firm financial performanceand CEO dismissal. Specifically, if things are goingpoorly financially, greater prior investments in CSRappear to expose the CEO to an even higher riskof being fired. In contrast, if the firm is performingwell financially, higher levels of CSR appear to helpprotect the CEO from dismissal. It thus appears thatprior investments in CSR amplify directors’ assess-ments of financial performance and, in turn, the like-lihood of CEO dismissal.

Our findings extend the direct linkage betweenfinancial performance and CEO dismissal (Finkel-stein et al., 2009) by suggesting that prior invest-ments in CSR subsequently frame the assessment offinancial performance delivered during the CEO’swatch. Our results are consistent with prior researchasserting the primary importance of firm financialperformance on CEO assessment, but also sug-gest that earlier investments in CSR inform howfinancial performance is interpreted. Our resultsalso suggest that investing or not investing in CSRinfluences career outcomes for CEOs, but that this

influence is dependent upon how their firm is per-forming financially.

As with all studies, this article also has its lim-itations. One limitation is that we focused only onCEOs in the United States. Thus, our findings maynot generalize to other countries due to differingcultural norms or governance practices. Also, sinceour sample was limited to Fortune 500 firms, theresults may not generalize to smaller firms as theymay not face the same level of scrutiny as largerfirms. Additionally, although we employ numerousmethods to address endogeneity, it is difficult tofully rule out its influence when conducting empir-ical research. Finally, there is still debate regard-ing how to properly measure CSR. Consistent withprevious research (e.g., Barnett & Salomon, 2012;David, Bloom, & Hillman, 2007; Hillman & Keim,2001; Hull & Rothenberg, 2008; Ioannou & Ser-afeim, 2015), we used KLD ratings as our proxyfor earlier investments in CSR. While we recog-nize that measuring CSR is difficult, this measureis the most well-known and frequently studied mea-sure of CSR. Thus, despite questions about its reli-ability (e.g., Chatterji, Durand, Levine, & Touboul,2016), its widespread usage, among academics andin social investments funds, suggests it is a reason-able CSR proxy.

Acknowledgements

We would like to thank the editor, James West-phal, and two anonymous reviewers for their valu-able comments and guidance. We would also like tothank John Busenbark, Robert Campbell, and Tim-othy Quigley for their helpful comments and sug-gestions. We are grateful for the financial supportthat was provided by a Terry-Sanford research grantfrom the Terry College of Business at the Universityof Georgia.

References

Aupperle, K. E., Carroll, A. B., & Hatfield, J. D. (1985). Anempirical examination of the relationship between cor-porate social responsibility and profitability. Academyof Management Journal, 28(2), 446–463.

Barnett, M. L., & Salomon, R. M. (2012). Does it pay to bereally good? Addressing the shape of the relationshipbetween social and financial performance. StrategicManagement Journal, 33(11), 1304–1320.

Brammer, S., & Millington, A. (2008). Does it pay tobe different? An analysis of the relationship between



corporate social and financial performance. StrategicManagement Journal, 29(12), 1325–1343.

Briscoe, F., & Safford, S. (2008). The Nixon-in-Chinaeffect: Activism, imitation, and the institutionalizationof contentious practices. Administrative Science Quar-terly, 53(3), 460–491.

Cai, Y., Jo, H., & Pan, C. (2011). Vice or virtue? The impactof corporate social responsibility on executive compen-sation. Journal of Business Ethics, 104(2), 159–173.

Chatterji, A., Durand, R., Levine, D., & Touboul, S.(2016). Do ratings of firms converge? Implications formanagers, investors and strategy researchers. StrategicManagement Journal, 37(8), 1597–1614.

Chatterji, A., Levine, D., & Toffel, M. (2009). How welldo social ratings actually measure corporate socialresponsibility? Journal of Economics and ManagementStrategy, 18, 125–169.

Choi, J., & Wang, H. (2009). Stakeholder relations andthe persistence of corporate financial performance.Strategic Management Journal, 30(8), 895–907.

Christensen, D. (2016). Corporate accountability reportingand high-profile misconduct. The Accounting Review,91(2), 377–399.

David, P., Bloom, M., & Hillman, A. (2007). Investoractivism, managerial responsiveness, and corporatesocial performance. Strategic Management Journal,28(1), 91–100.

Ernst and Young. (2013). Taking flight. Environmentalsustainability proposals gain more attention. Ernst &Young LLP.

Finkelstein, S., Hambrick, D. C., & Cannella, A. A.(2009). Strategic leadership: Theory and research onexecutives, top management teams, and boards. NewYork, NY: Oxford University Press.

Frank, K. (2000). Impact of a confounding variable ona regression coefficient. Sociological Methods andResearch, 29(2), 147–194.

Freeman, R. (1984). Strategic management: A stakeholderperspective. Englewood Cliffs, NJ: Prentice Hall.

Friedman, M. (1970, 13 September). The social respon-sibility of business is to increase its profits. New YorkTimes Magazine, pp. 32–33, 122, 124, 126.

Governance and Accountability Institute (2015). Flashreport: Seventy-five percent (75%) of the S&P 500index published corporate sustainability reports in2014. New York, NY: Governance and AccountabilityInstitute.

Graffin, S., Boivie, S., & Carpenter, M. (2013). ExaminingCEO succession and the role of heuristics in early-stageCEO evaluation. Strategic Management Journal, 34,383–403.

Greve, H. R. (1998). Performance, aspirations, and riskyorganizational change. Administrative Science Quar-terly, 9(2), 54–80.

Hillman, A. J., & Keim, G. D. (2001). Shareholder value,stakeholder management, and social issues: What’s thebottom line? Strategic Management Journal, 22(2),125–139.

Hull, C. E., & Rothenberg, S. (2008). Firm performance:The interactions of corporate social performance withinnovation and industry differentiation. Strategic Man-agement Journal, 29(7), 781–789.

Ioannou, I., & Serafeim, G. (2015). The impactof corporate social responsibility on investmentrecommendations: Analysts’ perceptions and shiftinginstitutional logics. Strategic Management Journal, 36,1053–1081.

Jensen, M. (2002). Value maximization, stakeholder the-ory, and the corporate objective function. BusinessEthics Quarterly, 12(2), 235–256.

Jenter, D., & Kanaan, F. (2015). CEO turnover and relativeperformance evaluation. Journal of Finance, 70(5),2155–2184.

Kang, J. (2013). The relationship between corporate diver-sification and corporate social performance. StrategicManagement Journal, 34(1), 94–109.

Kennedy, P. (2008). A guide to econometrics. Malden, MA:Blackwell.

Larcker, D., & Rusticus, T. (2010). On the use of instru-mental variables in accounting research. Journal ofAccounting and Economics, 49(3), 186–205.

Long, S. J. (1997). Regression models for categorical andlimited dependent variables. Thousand Oaks, CA: SagePublications.

Margolis, J. D., & Walsh, J. P. (2003). Misery lovescompanies: Rethinking social initiatives by business.Administrative Science Quarterly, 48(2), 268–305.

Maroulis, S., & Gomez, L. M. (2008). Does “connected-ness” matter? Evidence from a social network analysiswithin a small-school reform. Teachers College Record,110(9), 1901–1929.

McWilliams, A., & Siegel, D. (2000). Corporate socialresponsibility and financial performance: Correlationor misspecification? Strategic Management Journal,21(1), 603–609.

Navarro, P. (1988). Why do corporations give to charity?Journal of Business, 61(1), 65–93.

Parrino, R., Sias, R., & Starks, L. (2003). Voting with theirfeet: Institutional ownership changes around forcedCEO turnover. Journal of Financial Economics, 68(1),3–46.

Petrenko, O. V., Aime, F., Ridge, J., & Hill, A. (2016). Cor-porate social responsibility or CEO narcissism? CSRmotivations and organizational performance. StrategicManagement Journal, 37(2), 262–279.

Quigley, T., & Hambrick, D. (2015). Has the “CEO effect”increased in recent decades? A new explanation for thegreat rise in America’s attention to corporate leaders.Strategic Management Journal, 36(6), 821–830.

Shen, W., & Cannella, A. A. (2002). Power dynamicswithin top management and their impacts on CEOdismissal followed by inside succession. Academy ofManagement Journal, 45(6), 1195–1206.

Singer, J., & Willett, J. (2003). Applied longitudinal dataanalysis: Modeling change and event occurrence. NewYork, NY: Oxford University Press.

Social Investment Forum (2014). Report on US sustain-able, responsible and impact investing trends 2014.Washington, DC: Social Investment Forum.

Tang, Y., Qian, C., Chen, G., & Shen, R. (2015). HowCEO hubris affects corporate social (ir)responsibility.Strategic Management Journal, 36(9), 1338–1357.

UN Global Compact (2014). Guide to corporate sustain-ability. New York, NY.



UN Global Compact-Accenture. (2010). A new era ofsustainability. Retrieved from http://www.accenture.com/sitecollectiondocuments/pdf/accenture_a_new_era_of_sustainability_ceo_study.pdf

Waddock, S. A., & Graves, S. B. (1997). The corporatesocial performance-financial performance link. Strate-gic Management Journal, 18(4), 303–319.

Wagner, T., Lutz, R., & Weitz, B. (2009). Corporatehypocrisy: Overcoming the threat of inconsistentcorporate social responsibility perceptions. Journal ofMarketing, 73(6), 77–91.

Wang, H., Tong, L., Takeuchi, R., & George, G. (2016).Corporate social responsibility: An overview & newresearch directions. Academy of Management Journal,59(2), 534–544.

Wooldridge, J. (2003). Further results on instrumentalvariables estimation of average treatment effects inthe correlated random coefficient model. EconomicsLetters, 79(2), 185–191.

Wright, P., & Ferris, S. P. (1997). Agency conflict and cor-porate strategy: The effect of divestment on corporatevalue. Strategic Management Journal, 18(1), 77–83.


http://www.accenture.com/%20sitecollectiondocuments/pdf/accenture_a_new_era_of_sustainability_ceo_study.pdf

r Academy of Management Journal2017, Vol. 60, No. 4, 1582–1606.https://doi.org/10.5465/amj.2014.0691

IDEOLOGY AND THE MICRO-FOUNDATIONS OF CSR:WHY EXECUTIVES BELIEVE IN THE BUSINESS CASE FORCSR AND HOW THIS AFFECTS THEIR CSR ENGAGEMENTS

SEBASTIAN HAFENBRADLIESE Business School

Yale UniversityUniversity of Lausanne

DANIEL WAEGERWilfrid Laurier UniversityUniversity of Amsterdam

Existing research on executives’ belief in the business case for corporate social re-sponsibility (CSR) is built on two premises. The first is that, in order to believe in thebusiness case, executives need factual evidence that this business case indeed exists. Thesecond premise is that those executives who do believe in the business case will readilyinvest in CSR-related activities. The results from our four studies tell a different story.We show that managers, rather than focusing on factual evidence, believe in the busi-ness case because they espouse a fair market ideology—the tendency to justify andidealize the market economy system. At the same time, even thoughmanagers espousinga fair market ideology believe in the business case for CSR, they are not more inclined toengage in CSR than managers who do not hold such an ideology, because they alsoexperience weaker moral emotions when confronted with ethical problems. By drawingon system justification theory, we simultaneously explore antecedents and conse-quences of executives’ belief in the business case for CSR and of their moral emotions. Indoing so, we help advance knowledge about the micro-foundations of CSR.

Why do executives invest in activities related tocorporate social responsibility (CSR)? The literatureprovides a variety of explanations, ranging from ex-ternal factors such as pressures from activists(Briscoe, Gupta, & Anner, 2015; de Bakker, denHond, King, & Weber, 2013; Delmas & Toffel, 2008;McDonnell, King, & Soule, 2015), to idiosyncraticcharacteristics of executives such as their beliefs,values, and emotions (Agle, Mitchell, & Sonnenfeld,1999; Chin, Hambrick, & Treviño, 2013; Muller &

Kolk, 2010; Weaver, Reynolds, & Brown, 2014). Awell-known proposition is that executives are morelikely to invest in CSR activitieswhen they believe inthe business case for CSR or, in other words, ina positive relationship between corporate socialperformance (CSP) and corporate financial perfor-mance (CFP) (Bansal & Roth, 2000; Brønn &Vidaver-Cohen, 2009; Chin et al., 2013; Orlitzky, Schmidt, &Rynes, 2003). Yet, it has been asserted that manyexecutives find it difficult to believe in such a posi-tive link (Porter & Kramer, 2002, 2006; Stahl & DeLuque, 2014). This has spurred the emergence ofa large academic literature attempting to find outwhether or not CSP and CFP are indeed related(Eccles, Ioannou, & Serafeim, 2014; Flammer, 2015;Margolis, Elfenbein, & Walsh, 2009; Van Beurden &Gossling, 2008). It is often implied that finding evi-dence of a positive link would persuade executivesto believe in the business case and to ultimately in-vest more in CSR activities (Kang, Germann, &Grewal, 2016; Orlitzky et al., 2003; Waddock &Graves, 1997). Not surprisingly, the quest for evi-dence of a positive link between CSP and CFP has

Both authors contributed equally. We would like tothank Associate Editor Heli Wang for her exceptionalguidance and three anonymous reviewers for their helpfulcomments throughout the revision process. We are in-debted to Ulrich Hoffrage and Guido Palazzo for their un-wavering support for this project and toKlausWeber for aninvaluable suggestion. We further benefitted from en-lightening discussions with Nathan Betancourt, GregoireBollman, M.K. Chin, Joerg Dietz, Irina Feygina, ManuelGrieder, Don Hambrick, Florian Hoos, Nina Mazar,Sebastien Mena, Lamar Pierce, Leigh Plunkett Tost, Jan K.Woike, andChristianZehnder.Wegratefully acknowledgesupport by theSwissNational ScienceFoundation (SNSF).

1582

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

https://doi.org/10.5465/amj.2014.0691

thus been dubbed the “‘Holy Grail’ of the businessand society field of research” (Gond & Palazzo,2008: 1).

This literature relies on the premise that becauseexecutives do not have factual evidence of a positivelink, they also do not have a basis for believing in thebusiness case for CSR. However, a long-standingstream of research in psychology has highlightedthat individuals often form specific beliefs not somuch on the basis of factual information and evi-dence, but rather on the basis of their more generalbelief systems, worldviews, and ideologies (Allport,1954; Bobbio & Cameron, 1996; Converse, 1964;Knight, 2006). This psychological literature suggeststhat, in order to study executives’ beliefs aboutcompanies, we must account for executives’ moregeneral views on the economic system in whichthese companies operate. Individuals’ general viewson systemic arrangements—such as the economicsystem—are the focus of system justification theory(Jost, Banaji, & Nosek, 2004; Proudfoot & Kay, 2014).Therefore, we draw on system justification theory toexplain the psychological origin of executives’ beliefin the business case for CSR. Specifically, we hy-pothesize that the belief in the business case for CSRis grounded in executives’ fair market ideology—apositive ideological stance over themarket economysystem (Cichocka & Jost, 2014; Jost, Blount, Pfeffer, &Hunyady, 2003a).We find consistent support for thishypothesis in our four studies.

The finding that executives’ belief in the CSP–CFPlink originates in fair market ideology reveals animportant theoretical tension in our understandingof how the belief in the CSP–CFP link becomes ma-terialized in CSR activities at the firm level. On theone hand, it is logical to expect that executives whobelieve in the CSP–CFP link enhance their compa-nies’ CSR activities (Porter & Kramer, 2002, 2006;Vogel, 2005). On the other hand, prior research hasshown that fair market ideology also affects people’smoral values and emotions. Specifically, it limitsindividuals’ potential to feel morally outraged byethical problems stemming from corporate activities(Jost et al., 2003a;Wakslak, Jost, Tyler, &Chen, 2007).Such a lack of moral outrage should reduce execu-tives’ tendency to engage in CSR: if their moralemotions are not making them aware of a problem,then they are unlikely to take action against it. Thus,the theoretical tension arises because two contra-dictory elements (belief in the CSP–CFP link andlack of moral outrage) have their origin in the sameunderlying belief system (fair market ideology). Weexplore this theoretical tension by analyzing how the

belief in the CSP–CFP link and the lack of moraloutrage concurrently impact executives’ tendency toengage in CSR activities. Our findings show that thenegative impact of lacking moral outrage neutralizesthe positive impact of the belief in the CSP–CFP linkon executives’ CSR engagement.

The present paper makes several contributions.First, the existing literature argues that executiveswill engage inCSR if theybelieve in thebusiness case(Chin et al., 2013; Kang et al., 2016; Orlitzky et al.,2003). Our findings lead us to caution against thisassertion. We find that, even though they believe inthe business case for CSR, managers who hold a fairmarket ideology will not readily engage in CSR be-cause they experienceweaker emotional reactions toethical problems than managers who do not holda fair market ideology.

Second, our studies illustrate the complex andinterdependent nature of how executives’ idiosyn-cratic lenses impact their strategic choices. Specifi-cally, our research underlines that it is not onlyexecutives’ beliefs and moral emotions that matterbut also the psychological antecedents of these be-liefs and moral emotions. By showing how both thebelief in the business case and moral emotions orig-inate in fair market ideology, we reveal the contra-dictory ways in which they are connected andimpact managerial decisions. Uncovering thesecontradicting paths advances the literature, whichhas thus far conceptualized the belief in the businesscase and moral emotions or values as having eitherindependent (Agle et al., 1999; Brønn & Vidaver-Cohen, 2009) or mutually reinforcing (Chin et al.,2013) effects on CSR engagement. Our resultsthereby illustrate how we can improve our un-derstanding of CSR by investigating its micro-foundations (Aguinis & Glavas, 2012; Christensen,Mackey, & Whetten, 2014; Kourula & Delalieux,2016; Spiess, Mueller, & Lin-Hi, 2013) and also un-derscore the need for more research that heedsHambrick’s (2007) call to investigate the psy-chological antecedents of executives’ idiosyncraticcharacteristics.

Third, we have developed an innovative methodtomeasure the belief in the business case for CSR.Nosuch measure existed previously and upper eche-lons theorists have so far relied on inferring this be-lief from rather distant biographical proxies (Chinet al., 2013), such as executives’ educational back-ground. Such biographical proxies are problematicbecause we cannot be sure that the proxies correlatewith organizational outcomes for the reasons wehypothesize (Carpenter, Geletkanycz, & Sanders,

2017 1583Hafenbradl and Waeger

2004; Hambrick, 2007; Lawrence, 1997). If we wantto be able to make more grounded hypotheses, it isnecessary to open the “black box” of executives’ id-iosyncratic characteristics (Hambrick, Geletkanycz,& Fredrickson, 1993), and thus to measure the beliefin the business case more directly than via bio-graphical proxies.Wedo this in the present article bymeasuring this belief through an original predictiongame.

The remainder of this article is structured as fol-lows. In the next section, we develop our hypotheseson the relationships between fair market ideology,the belief in the business case forCSR,moral outrage,executives’ biographical proxies, and CSR engage-ment. We then report the methods and results fromour four studies. We conclude by discussing theimplications of our studies and possible avenues forfuture research.

THEORY AND HYPOTHESES

Executive Characteristics and CSR Engagement

As firms’ activities in the area of corporate socialresponsibility have expanded over the past decade(Etzion & Ferraro, 2010; Scherer, Rasche, Palazzo, &Spicer, 2016), researchers have started examiningthe factors that lead companies to engage in CSR.Traditionally, the bulk of this research has empha-sized the importance of environmental factors, suchas pressures stemming from activists (de Bakkeret al., 2013; McDonnell et al., 2015; Mena &Waeger,2014), the general institutional environment(Ioannou& Serafeim, 2012; Roulet & Touboul, 2015),a company’s industry (Barley, 2007; Weber, Rao, &Thomas, 2009), or its peers (Campbell, 2006). Therelative dearth of individual-level investiga-tions into CSR has led to a call for increased ef-forts to understand the micro-foundations of CSR(Aguinis & Glavas, 2012; Christensen et al., 2014).Consequently, scholars have recently started tofocus increasingly on firm-internal factors, such asindividual-level managerial characteristics, that canexplain why some executives are more likely to en-gage in CSR than others (Lewis, Walls, & Dowell,2014;Maak, Pless, &Voegtlin, 2016; Petrenko, Aime,Ridge, & Hill, 2016). This is reflected in theemergence of the responsible and ethical leadershipliterature (Brown, Treviño, & Harrison, 2005;Waldman & Balven, 2014) and in the growing in-terest of upper echelons theorists in the topic of CSR(Deckop, Merriman, & Gupta, 2006; Slater & Dixon-Fowler, 2010).

At this individual level, the literature hints at twobroad motives that drive executives’ CSR engage-ments. One part of the literature emphasizes execu-tives’belief in the business case forCSR (Orlitzky et al.,2003; Porter & Kramer, 2002, 2006; Stahl & De Luque,2014;Waddock & Graves, 1997). The other underlinestheir moral values and emotions (Agle et al., 1999;Weaver et al., 2014). These two broad motives havebeenmostly investigated as separate and independentdrivers of CSR. The possibility that they may both begrounded in the same general worldview orideology has not been considered thus far.

In the present article, we build on prior research insocial psychology, which suggests that specific be-liefs and moral orientations are embedded in andshaped by more general, relatively coherent beliefsystems or ideologies. Such belief systems or ideol-ogies are the focus of system justification theory (Jost& Hunyady, 2005). Therefore, we draw on systemjustification theory to theorizehowexecutives’beliefin the business case for CSR as well as their moralemotions are shaped.We startwith executives’beliefin the business case for CSR to build our theoreticalframework. Figure 1 provides an overview of thetheoretical framework and the correspondinghypotheses.

System Justification Theory and the Belief in theBusiness Case for CSR

System justification theory (Jost & Banaji, 1994;Jost et al., 2004) explores how individuals justify andidealize systemic social arrangements. Examples ofsuch social systems include the Indian caste systemand the institution of slavery (Jost, Liviatan, van derToorn, Ledgerwood, Mandisodza, & Nosek, 2010),but also themore recentlydeveloped systems suchasthe modern democratic political system (Feygina,Jost, & Goldsmith, 2010) or the market economysystem (Jost et al., 2003a). In the present article, wefocus on how individuals justify and idealize thislatter market economy system and the activities ofthe principal actors operating in this system—

private companies. While most people have beenfound to engage in system justification (Feyginaet al., 2010; Jost & Hunyady, 2005), research hasshown that some individuals have a stronger ten-dency to do so than others (Jost et al., 2003a; Jost &Thompson, 2000). These individual differences arecaptured in the concept of fair market ideology,which is defined as the extent to which individualsjustify and idealize the market economy system(Cichocka & Jost, 2014; Jost et al., 2003a).

1584 AugustAcademy of Management Journal

The starting point for system justification theory isthe commonplace assertion that powerful socialsystems exert great control over the lives of in-dividuals, whereas individuals have little or nocontrol over the systems. According to system justi-fication theory, if individuals had to acknowledgethat these powerful and hard-to-change systems areunstable, arbitrary, or flawed, theywould experiencepsychological threat and anxiety (Jost & Hunyady,2005; Proudfoot & Kay, 2014). Individuals strive toavoid such threat and anxiety in order to maintaina sense of psychological stability and safety—an ex-istential human need (Jost et al., 2010; Jost, Glaser,Kruglanski, & Sulloway, 2003b; Jost & Hunyady,2005; Lerner, 1980). To maintain this sense of psy-chological stability and safety, individuals are mo-tivated to justify and idealize the status quo in thesocietal systems surrounding them.Conversely, theyaremotivated to oppose changes to the status quo, assuch changes would be indicative of an unstable,arbitrary, or flawed system (Jost & Hunyady, 2005;Wakslak et al., 2007).

Thismotivation to psychologically bolster the statusquo (Jost et al., 2004; Proudfoot & Kay, 2014), in turn,systematically affects people’s reasoning. Motivatedreasoningmeans that an individual relies on a “biasedset of cognitive processes” (Kunda, 1990: 480) to rea-son inaway that allowshimorher to concludewhatheor she wanted to believe all along (Detert, Treviño, &Sweitzer, 2008; Paharia, Vohs, & Deshpande, 2013).Hence, when engaging inmotivated reasoning, peoplesearch for, pay more attention to, and put more em-phasis on information that supports what theywant tobelieve. At the same time, they avoid, discount, anddiscredit information that does not support what theywant to believe (Paharia et al., 2013).

Individuals with a high fair market ideology wantto uphold the belief that companies’ activities in themarket economy systemare not random, arbitrary, orillegitimate, but instead serve the system’s purposeof generating economic value (Boltanski &Thevenot,2006; Cichocka & Jost, 2014; Friedland & Alford,1991; Jost et al., 2003a). To uphold this belief, theseindividuals idealize companies’ activities by using“information about how things are currently done toinform their beliefs about how things should bedone” (Proudfoot & Kay, 2014: 176 emphasis added;see also Kay et al., 2009). In other words, fair marketideologists idealize the activities that companies dopursue as activities that companies should pursue asactors operating in the market economy system.Thus, they search for, pay more attention to, and putmore emphasis on information indicating that theactivities companies do engage in have economicvalue, and they avoid, discount, and discredit in-formation indicating that this is not the case (Elsbach& Kramer, 1996; Shepherd & Kay, 2012).

This reasoning is relevant to the present articlewhen applied to companies’ activities related toCSR. Indeed, asmore andmore companiesdo engagein CSR activities (Etzion & Ferraro, 2010; Marquis &Qian, 2013;McDonnell et al., 2015), individualswhoscore high on fair market ideology are motivated toidealize these CSR activities as having economicvalue. Accordingly, they search for, pay more at-tention to, and put more emphasis on informationindicating that CSR does indeed have economicvalue, and they avoid, discount, and discredit in-formation indicating that CSR is not of economicvalue. As a result, individuals with high fair marketideology believe in the business case for CSR. Bycontrast, individuals who score low on fair market

FIGURE 1Overview of the Theoretical Framework

H1: B (+)H2: A + B (+)H3: B + C (+) H4: D + E (–)

Educational background(Biographical proxy)

Belief in theCSP–CFP link

Fair marketideology

Tendency for CSRengagement

Moral outrageD (–)

C (+)(+)

A (+) B

(+)

E (+

)


ideology will be less likely to idealize companies’CSR activities as having economic value and aretherefore also less likely to believe in the businesscase for CSR. Stated formally:

Hypothesis 1. The higher individuals score onfair market ideology, the more strongly they be-lieve in the business case for CSR.

Recent scholarship has emphasized that the ten-dency to engage in system justification is stronger un-der specific circumstances (Day, Kay, Holmes, &Napier, 2011; Kay et al., 2009; Shepherd, Kay,Landau, & Keefer, 2011). In particular, it has beenfound that individuals’ system justification motive isstronger themore theysee the systeminwhich theyareliving or operating as inescapable (Kay & Friesen,2011). The intuition behind this finding is that thereason individuals engage in system justification in thefirst place—a need for psychological stability andsafety—is enhanced under conditions of system ines-capability. In other words, “when people find them-selves in a system they cannot escape, it is particularlypsychologically threatening for them to acknowledgethat system’s flaws” (Proudfoot & Kay, 2014: 178).Therefore, individuals who perceive their system asinescapable would engage more strongly in systemjustification. Prior researchhas found evidence for thisargument. For instance, in an experiment, Kay et al.(2009) manipulated participants’ perceptions that ithad become more difficult (respectively less difficult)to emigrate from their home country. After this ma-nipulation, participants were told that politicians intheir home country were disproportionately wealthy.Participants in the inescapability condition werefound to view this indication of inequality to be morejustified than the other participants.

With regard to the market economy system, a nat-ural way to explore the degree to which individualsare exposed to a feeling of system inescapability is bylooking at their educational background. Upperechelons theorists also often use educational back-ground as a proxy for underlying psychological ori-entations (Carpenter et al., 2004; Hambrick &Mason,1984). More specifically, research in the upper ech-elons tradition has argued that individuals with aneducational background in business, economics, orlaw hold similar attitudes when compared to in-dividuals with other educational backgrounds(Barker & Mueller, 2002; Wiersema & Bantel, 1992).These similarities among business, economics, andlaw students have been found to be particularlystrong with respect to general worldviews on themarket economy and the role of corporations therein

(Fiss & Zajac, 2004), which are the focal interest ofthe present article.

In all of these curricula, an important part of edu-cation is dedicated to teaching about firms as profit-seeking entities that operate within an economicsystemgrounded inmarket-basedexchanges. Suchanemphasis on the market economy system is accom-panied by a de-emphasis on other systems—or “or-ders of worth” (Boltanski & Thevenot, 2006; Patriotta,Gond, & Schultz, 2011). Thus, the inner logic of sys-tems other than the market economy is less presentand available to individuals with an educationalbackground in business, economics, and law. Thisleads to a perception that themarket economy systemlacks alternatives. In turn, such a sense that there isa lack of alternatives enhances feelings of inescap-ability1 (Kay & Friesen, 2011). For upper echelonstheorists, this is relevant because it indicates that ed-ucational background can be used as a biographicalproxy for fair market ideology. And since we expecthigher levels of fair market ideology to be associatedwith the belief in the business case for CSR (see Hy-pothesis 1 above), it follows that an educationalbackground in business, economics, and law is asso-ciated with such a belief in the business case for CSR.In other words, there is a relationship between edu-cational background and the belief in the businesscase for CSR; the mechanism that explains this re-lationship is fair market ideology. Stated formally:

Hypothesis 2. There is a positive indirect effectof educational background in business, eco-nomics, and law on the belief in the businesscase for CSR. This indirect effect is mediated byfair market ideology.

If we find evidence to support our first two hy-potheses, then fair market ideology can explain whyindividuals believe in the business case for CSR.Such a finding, in turn, would enable us to draw in-ferences about how fair market ideology impactsindividuals’ tendency to engage in CSR activities.Researchers looking into the determinants of com-panies’CSRengagement have for a long time focused

1 Note that our point does not depend on whether in-dividuals who had stronger feelings of inescapability (ora higher fair market ideology) self-selected into studyingbusiness, economics, or law, or whether individuals whostudy these academicdisciplines develop stronger feelingsof inescapability (or a higher fair market ideology). Ourpoint is that the consequence is always the same: an edu-cational background in business, economics, and law isassociated with higher levels of fair market ideology.


on macro-level factors, such as pressure from activ-ists or industry affiliations (Durand & Vergne, 2015;Weber et al., 2009). However, recent scholarly in-terest has started exploring the more micro-level as-pects or company-internal drivers of CSR (Bridoux,Stofberg, & Den Hartog, 2016; Crilly, Zollo, &Hansen, 2012; Maak et al., 2016; Muller & Kolk,2010). Of particular relevance for our purposes in thepresent article is that researchers have proposed, butnever tested, the intuition that managers would bemore likely to lead their firms toward increased CSRengagements if they believed in a positive relation-ship between CSP and CFP (Chin et al., 2013;Orlitzky et al., 2003). Underlying this intuition is theacknowledgment that the managers of private com-panies feel pressured by various powerful stake-holders, such as investors and regulators, and by thenotion of fiduciary responsibility, to focus squarelyon financial performance (Margolis & Walsh, 2003;Porter & Kramer, 2006). Therefore, if managers per-ceive CSR activities as an opportunity to increasetheir firms’ financial performance—in other words,if they believe in the business case for CSR—theyshould be motivated to lead their firms toward en-gaging in CSR (Orlitzky et al., 2003).

Conversely, if managers see CSR engagementsmerely as a cost, and thus as potentially reducing fi-nancial performance, they will refrain from pursuingCSR activities. In line with this argument, academicshave long viewed the presumed difficulty that exec-utiveshave inbelieving in thebusiness case forCSRasan important obstacle to increased engagement oftheir firms in CSR (Porter & Kramer, 2002; Stahl & DeLuque,2014;Waddock&Graves, 1997).However, thisobstacle is lifted for executives who believe in thebusiness case, which means they should have a com-paratively stronger tendency to engage in CSR activi-ties (Chin et al., 2013). Asweproposed in our first twohypotheses, we expect fair market ideology to bepredictive of executives’ belief in the business case forCSR. This implies an indirect relationship betweenfair market ideology and executives’ tendency to en-gage inCSR. If fairmarket ideology is predictive of thebelief in thebusinesscase forCSRand thisbelief in thebusiness case for CSR is predictive of executives’tendency to engage in CSR activities, then fair marketideology should have a positive indirect effect on thetendency to engage in CSR. Stated formally:

Hypothesis 3. There is a positive indirect effectof fair market ideology on tendency to engage inCSR activities. This indirect effect is mediatedby the belief in the business case for CSR.

Existing research has—seemingly in contrast toHypothesis 3—found evidence that system justifica-tion in general and fair market ideology in particularare associatedwith attitudes and actions that ought torun counter to an increased tendency for CSR en-gagements. For instance, Feygina et al. (2010) foundthat individuals who engage more in system justifi-cation are also more likely to deny the existence ofglobalwarmingand, as a consequence thereof, are lesslikely to engage in pro-environmental behavior.Morespecifically, with respect to the context of corpora-tionsoperating inamarket economy, Jost et al. (2003a)investigated the attitudes of MBA students towardcorporate scandals. They found that MBA studentswho scored higher on fair market ideology were lessconcerned about ethical violations and dubious ac-counting practices at companies in the United States.

How can these findings be explained? As reviewedabove, foundational to system justification theory isthe contention that system-justifying individuals aremotivated to psychologically bolster the status quo insocietal systems (Jost et al., 2004; Proudfoot & Kay,2014). In response to their fundamental need forpsychological stability and safety, individualswant toperceive such systems as stable, flawless, and legiti-mate (Jost & Banaji, 1994; Jost et al., 2004). This mo-tivation affects their reasoning: individuals high onsystem justification search for, pay more attention to,and put more emphasis on information that supportstheir belief in the stability, flawlessness, and legiti-macyof the systems (Nam, Jost, &VanBavel, 2013).Atthe same time, they avoid, discount, and discreditinformation that could point to problems in that sys-tem (Proudfoot & Kay, 2014; Shepherd & Kay, 2012,2014). For instance, Shepherd and Kay (2012, 2014)provided evidence that individuals who score higheron system justification tryharder to avoid informationabout economic issues during a recession or aboutcompanies involved in environmental scandals.When individuals high on system justification cannotavoid information about problems in the system, theyinstead discount and discredit such information andrefuse to see the status quo as problematic (Jost et al.,2003a). As we already alluded to when developingHypothesis 1, these individuals tend to idealize “thecurrent status quo as the most desirable and reason-able state of affairs” (Kay et al., 2009: 421) by using“information about how things are currently done toinform their beliefs about how things shouldbe done”(Proudfoot & Kay, 2014: 176 emphasis added). Kayet al. (2009) conducted an experiment to illustrate thisphenomenon. After reading factual information stat-ing that women were underrepresented in high-level


business positions (information about how thingscurrently are), individuals in the high system justifi-cation condition were found to be more favorable to-ward gender inequality (how things should be) thanindividuals in the control condition.

When determining what is normatively justifiedon the basis of what they observe in the status quo,system justifiers guard themselves from using moreintrinsic yardsticks of what should be, such as theirmoral values, intuitions, or emotions. That does notmean, of course, that high system justifiers cannotpossess these moral values and emotions that priorresearch associated with higher CSR engagements(Brown & Treviño, 2006; Shao, Aquino, & Freeman,2008). Rather, it means that high system justifiersrefrain from relying on their moral values and emo-tions when assessing problems that occur in thecurrent status quo of doing business, as if they wereethically blind (Palazzo, Krings, & Hoffrage, 2012).Specifically, Wakslak et al. (2007) showed that sys-tem justification leads to a significant reduction inboth inward-focused moral emotions like guilt orfrustration and in the outward-focused moral emo-tion known as moral outrage (Montada, Schmitt, &Dalbert, 1986). Moral outrage is especially relevantfor the present article because it has been shown topredict behavioral intentions (Wakslak et al., 2007)and has already been studied in the context of CSR(Antonetti & Maklan, 2014). Moral outrage is de-termined primarily by the “perception that a moralstandard or principle has been violated” (Batson,Chao, &Givens, 2009: 155).Hence,when individualswho are high in system justification guard them-selves from feeling moral outrage, they limit theirpotential to perceive and detect moral problems inthat system. For instance, using both correlationaland experimental designs, Tan, Liu, Huang, Zhao,and Zheng (2016) found that lower levels of moraloutrage predicted lower awareness of corruption.Such insensitivity to problems, in turn, leads in-dividuals to be less supportive of activities andpractices that aim to address these problems. Ac-cordingly, Wakslak et al. (2007) showed that in-dividualswho lackmoral outrage are less supportiveof helping disadvantaged segments of the pop-ulation. CSR activities of companies are often con-ceived of as addressing social, environmental, orothermoral problems in themarket economy system(Carroll, 1999; Garriga &Mele, 2004). As individualsexperiencing lower levels of moral outrage are lesslikely to acknowledge the existence of such prob-lems, we expect these individuals to be less sup-portive of CSR engagements.

In sum, research has shown that system justifica-tion limits individuals’ potential to feel outrage to-ward moral problems (Wakslak et al., 2007). On thisbasis, we argue that individuals with higher levels offair market ideology will experience less outrage to-ward social, environmental, or othermoral problemsin the market economy system. In turn, lower levelsof moral outrage imply that individuals are lesssensitive to theseproblems andare therefore also lessmotivated to solve them via CSR engagements. Inother words, if fair market ideology is predictive ofa lack of moral outrage and this lack ofmoral outragedecreases individuals’ tendency to engage in CSRactivities, then fair market ideology should have anindirect negative effect on the tendency to engage inCSR. Stated formally:

Hypothesis 4. There is a negative indirect effectof fair market ideology on tendency to engage inCSR activities. This indirect effect is mediatedby moral outrage.

Overview of Studies

This series of hypotheses converges in the theo-retical framework depicted in Figure 1. The center-piece is Path B, which connects fair market ideologyand the belief in the CSP–CFP link, as proposed inHypothesis 1. In Study 1, we first test this path witha sample of executives. We then test this path ex-perimentally in Study 2 in order to rule out concernsabout reverse causality or omitted variables. Path Aconnects educational background, a proxy oftenused in upper echelons research, to fair marketideology. Paths A and B combined describe the in-direct effect of educational background on the beliefin the CSP–CFP link via fair market ideology, whichwe propose in Hypothesis 2. We test this indirectpath in Study 3, exploiting the natural variation ineducational backgrounds in a student sample. PathsC, D, and E bring the tendency for CSR engagementinto the picture. Path C connects the belief in theCSP–CFP link to CSR engagement. Together, Paths Cand B illustrate the indirect positive effect of fairmarket ideology on CSR engagement via the belief inthe CSP–CFP link proposed in Hypothesis 3. Path Dleads from fair market ideology to moral outrage.Path E describes the positive link between moraloutrage and CSR engagement. Together, Paths D andE constitute the negative effect of fair market ideol-ogy on CSR engagement via moral outrage, as pro-posed in Hypothesis 4. In Study 4, we investigatethese two indirect effects (PathsB1CandPathsD1E)concurrently with a sample of executives.


STUDY 1: EXECUTIVES’ BELIEF IN THE CSP–CFPLINK AND FAIR MARKET IDEOLOGY

The purpose of Study 1 is to test Hypothesis 1 andto investigate the link between fair market ideologyand executives’ belief in the CSP–CFP link. To es-tablish the relevance of fair market ideology for de-cision makers who could potentially influence theircompany‘s policies, we followed Hambrick’s (2007)suggestion to use executive MBAs (EMBAs).

Sample

We recruited 59 executives from an EMBA class ata large Swiss university. We excluded 12 partici-pants because they did not complete the question-naire, respectively did not understand the instructionsor the incentive structure of their task (i.e., they failedto correctly answer comprehension check questions).This resulted in a final sample of 47participants, 12ofwhom (26%) were women and 35 (74%) were men.On average, the executiveswere 37 years of age (SD55.1) and had 11 (SD 5 4.8) years of managerial expe-rience. Twenty percent worked for small companies(1–50 employees), 23% formedium-sized companies(51–500 employees), and 57% for large compa-nies (more than 500 employees).

Measures

Belief about the link between CSP and CFP. Inorder to measure our subjects’ belief about the link be-tween CSP and CFP, we developed a prediction game.Participants predictedCFPbased on information aboutprior financial performance and prior social perfor-mance. Using a prediction game enabled us to set in-centives for participants to make their judgmentsaccording to their actual and true beliefs and therebyavoid social desirability bias. Appendix 1 provides anexample of how the predictions were made.

In the prediction game, participants were giveninformation about the social and financial perfor-mance of a company at one point in time (time T).Based on these two pieces of information, partici-pants had to predict the financial performance of thiscompany two years later (time T12).

All the information we presented to the re-spondents was based on real data. To operationalizesocial performance, we used data from CovalenceEthicalQuote (www.ethicalquote.com) for the years2002–2006 for a total of 183 companies taken fromthe Dow Jones Sector Titans Index, an index of thelargest companies in important industries. CovalenceEthicalQuote is a rating agency based in Geneva,

Switzerland that specializes in assessing externalinformation about the social and environmentalperformance of companies. The rating agency’smethodology is based on the difference between allpositive and all negative pieces of information aboutthe social and environmental consequences of thecompanies’ activities reported in the news mediaworldwide (in English, Spanish, French, and Ger-man). As a measure of financial performance, we ob-tained the return on equity (ROE) for the previous 12months from the COMPUSTAT database. ROE is anaccounting-based measure from corporations’ bal-ance sheets that is obtained by dividing net incomeafter taxby shareholder equity. ROEexpresses a firm’sefficiency in generating profits and is therefore gen-erally used as an indication of how profitably a com-pany has operated over a year. Participants wereinformed in detail about the measures for social andfinancial performance.

Tomake theprediction task as intuitive aspossiblefor our participants, the financial performance andthe social performance were given as a rank amongthe entire set of 183 companies. A low number in-dicated a good rank (1st was the best) and a highnumber indicated a poor rank (183rd was the worst).Ranks are indicative of how good a company iscompared to the other companies. This enabled ourparticipants to consider the relationship betweensocial performance and financial performance in-dependently from factors that affected the economyas a whole. To underline this point, we did not in-dicate the precise years for which the participantswere making their predictions.

We selected seven of the 183 companies for theprediction game. The selection procedure for theseven companies was as follows: we chose compa-nies based on their social and financial performanceat timeT. To avoid a “regression to themean” effect,2

we chose companies whose rank was closest to themiddle rank for financial performance—the

2 This effect describes a situation in which a variable ismeasured multiple times. When the first measurement ofthe variable returns an extreme value, the second mea-surement will tend to be closer to the variable’s true mean.As a consequence, had we given our participants compa-nies with extreme ranks at timeT (i.e., close to 1 or 183) forthe dimension they had to predict (i.e., financial perfor-mance), and had they predicted a rank closer to the meanrank at time T12, then we would not have been able toseparate two possible explanations for such predictions:(1) beliefs concerning the link between CSP and CFP thatallow us to test our hypotheses, and (2) correct intuitionsconcerning statistical regression toward the mean.


http://www.ethicalquote.com

dimension that participants had to predict. At the sametime, these companies should be as extreme as possibleonsocialperformance—thedimensionthatparticipantsdid not have to predict. We selected seven companiesthat matched our criteria. Their financial performanceranks were in the middle (between 78 and 111), andtheir social performance ranks were split: four of thecompanies had a high rank (between 168 and 183) andthree of the companies had a low rank (between 1 and17). This procedure enabled us to investigate if, and inwhich way, the information about social performanceinfluenced the prediction of financial performance.

Participants were given the rank for financial per-formanceand the rank for socialperformanceat timeT,and they predicted the financial performance at timeT12 (two years later). We calculated our main de-pendent variable by estimating how the given socialperformance ranks influenced the predicted financialperformance ranks across all seven predictions, whilecontrolling for the influence of the given financialperformance ranks.More specifically,we fitteda linearregression for each participant separately, with thepredicted financialperformance rankas thedependentvariable and the given social performance rank as theindependent variable, while the given financial per-formance rank was used as the control variable. Aparticipant’s coefficient for theeffect of the givensocialperformance rank on the predicted financial perfor-mance rank isourmeasureof thisparticipant’sbelief inthe CSP–CFP link. This coefficient measures the ex-pected change in the predicted financial performancerank when the given social performance changes byone rank. A positive coefficient indicates that the par-ticipant predicted a positive association betweensocial performance at T and financial performance atT12, whereas a negative coefficient indicates that theparticipant predicted a negative association.

Participants predicted the financial performanceof the companies at point T12, which was between2004 and 2006. We evaluated the accuracy of theirpredictions by comparing their predicted rank withthe actual ranks at that point in time. The ten mostaccurate participants were each awarded 50 Swissfrancs (approximately USD 46.50 at the time of thestudy). We introduced this incentive to increaseparticipants’ motivation and to counteract potentialsocial desirability biases—that is, that the partici-pants would predict a stronger link between socialand financial performance than they believed to ac-tually exist. In this incentive scheme, participantsmaximize their chances of winning the 50 Swissfrancs by stating their true beliefs about how socialperformance impacts future financial performance.

Fair market ideology. To measure the extent towhich participants engage in justification and ide-alization of themarket economy system,we used thesystemic fair market ideology scale. This scale wasdeveloped and tested by Jost et al. (2003a), whoaimed to measure individual differences regardingthe ideological tendency “to believe that the existingfree market system is fair, ethical and legitimate”(Jost et al., 2003a: 66). Example items are “In manymarkets, there is no such thing as a true ‘fair’marketprice” (reverse-coded) and “In free market systems,people tend to get the outcomes that they deserve.”This scale has been shown to strongly correlate withother general and economic system justificationscales (Jost et al., 2003a). Participants answered onan 11-point scale ranging from 25 (completely dis-agree) through 0 (neither agree nor disagree) to 15(completely agree). The 15 items were averaged intoa fair market ideology score (Cronbach’s a in thissample 5 0.69, M 5 20.2, SEM 5 0.16).

Demographics and additional measures. Priorresearch has found a relationship between the politicalorientation of CEOs and their companies’ CSR engage-ment andhas explained this finding by arguing that it isdriven by CEOs’ belief in the business case for CSR(Chin et al., 2013). Therefore, we included executives’political orientation as the control variable (on twoseven-point scales, one ranging from liberal to conser-vative and one ranging from left-wing to right-wing).

We were also interested in exploring whether thebeliefs about the CSP–CFP link differed systemati-cally between executives with different demographiccharacteristics. Therefore, we included several de-mographic control variables that are regularly in-cluded inbehavioral research inmanagement, suchasupper echelons research. In addition to gender andage, these demographic variables were the number ofyears of work experience, level of education (with thecategories: primary school, secondary school, com-pleted high school, undergraduate degree, graduatedegree,PhD), and rank in the organizationalhierarchy(in terms of number of hierarchy levels below theCEO). For ease of interpretation, we coded both thelevel of education and the hierarchy level as contin-uous, althoughall results are robust to using a dummyvariable for each level of these variables.

Results and Discussion Study 1

Table 1 summarizes the descriptive statistics andcorrelations for Study 1. Participants’ belief in theCSP–CFP link is significantly correlated with partici-pants’ fair market ideology (r5 0.36, p5 0.01). As can


beseeninTable2,whenregressing fairmarket ideologyon the belief in the CSP–CFP link, the coefficient of fairmarket ideology remains significant (b5 0.06, SEM50.02, t 5 3.07, p 5 0.004), even after controlling forgender, age, education level, political orientation,workexperience, and thehierarchydistance to theCEO.Thisanalysis is also robust to the exclusion of the controlvariables. Thus, we find support for Hypothesis 1: in-dividuals’ belief in a positive link between CSP andCFP is correlated with their ideological tendency tojustify the market economy system.

STUDY 2: EXPERIMENTALLY ESTABLISHINGTHE CAUSAL LINK BETWEEN FAIR MARKET

IDEOLOGY AND THE BELIEF IN THECSP–CFP LINK

The purpose of Study 2 is to establish that the linkbetween system justification of the market economysystemand thebeliefs about theCSP–CFP link is causaland in the proposed direction. Specifically, we hy-pothesized that we could prompt participants to be-lieve in a stronger (weaker) link between CSP and CFPby making fair market ideology more (less) salient. Ex-perimentally manipulating the accessibility of fairmarket ideology and measuring the effect of this ma-nipulationon the belief in theCSP–CFP link enablesusto exclude alternative explanations, such as reversecausality or omitted variables.

Sample

We recruited 95 business and economics studentsfrom a large Swiss university. Twenty participants

were excluded because they did not complete thequestionnaire, respectively did not understand theinstructionsor the incentive structure (i.e., they failedto correctly answer comprehension check ques-tions). Sixty of the participants were women (81%),14 (19%) were men, and one participant did not

TABLE 1Descriptive Statistics and Correlations from Study 1

Variable Mean SD 1 2 3 4 5 6 7 8

1. Belief in the CSP–CFP link 0.08 0.152. Fair market ideology 20.20 1.12 0.36*3. Gender 0.74 0.44 0.14 0.124. Age 36.87 5.08 20.09 20.19 20.045. Political orientation

(liberal–conservative)21.23 1.13 0.16 20.02 20.12 0.02

6. Political orientation (left–right) 0.17 1.27 20.07 0.27 0.27 20.11 20.147. Work experience 11.26 4.82 20.12 20.17 20.07 0.83*** 0.00 0.018. Level of education 4.96 0.91 20.04 0.11 20.03 20.05 20.20 0.04 20.169. Hierarchical distance to CEO 2.64 1.90 20.23 0.14 20.42** 0.07 0.07 20.07 20.03 0.07

Notes: N 5 47.*p , 0.05

**p , 0.01***p , 0.001

TABLE 2Regression Analysis with the Criterion Belief in the

CSP–CFP Link from Study 1

Variables Model 1 Model 2

Constant 0.13 0.13(0.25) (0.23)

Gender 0.03 0.003(0.06) (0.06)

Age 0.002 0.004(0.009) (0.008)

Political orientation(liberal–conservative)

0.02 0.02

(0.02) (0.02)Political orientation (left–right) 20.009 20.02

(0.02) (0.02)Work experience 20.006 20.006

(0.009) (0.008)Level of education 20.02 20.03

(0.01) (0.01)Hierarchical distance to CEO 20.001 20.008

(0.03) (0.02)Fair market ideology 0.06**

(0.02)

R-Squared 0.11 0.29

Notes: SEM in parentheses; N 5 47.*p , 0.05

**p , 0.01***p , 0.001


provide gender information. The average age was21.04 (SD5 1.70) years, and all participants were inthe second year of their bachelor studies.

Experimental Design

The experiment consisted of two parts. In the firstpart, we manipulated the salience of system justifi-cation by inducing participants to conceive of themarket economy as either fair or unfair. In the secondpart, participants played the same prediction game asin Study 1. To manipulate the salience of systemjustification, we adopted an unscramble sentencesprocedure from Feinberg and Willer (2011). We pre-sented participants with a set of scrambled sentencesand instructed themtounscramble these sets ofwordsto form coherent sentences. Participants were ran-domly assigned to three conditions in which theywere presented different sets of sentences. In the highsystem justification condition, eight unscrambledsentences described the market economy system asjust and fair. These sentences were taken from thesystemic fair market ideology scale by Jost et al.(2003a). Example items include “The free marketsystem is a just system” and “In free market systems,people tend to get what they deserve.” These scram-bled sentences were mixed with six filler sentences,which were not associated with the topic. In the lowsystem justification condition, participants were pre-sented with eight scrambled sentences that describedthe market economy system as unjust and unfair

(e.g., “The freemarket system is anunjust system,” “Afree market does not guarantee that people get whatthey deserve”) and the six filler items. In the controlcondition, participantshad tounscramble only the sixfiller items.

Measures

Belief about the link between CSP and CFP.Participants played a prediction game similar to theone in Study 1 (described above); the only differencewas that participants made ten predictions instead ofseven.Wecalculated thevariablemeasuring thebeliefin the CSP–CFP link in the same way as in Study 1.

Fair market ideology. As a manipulation check,participants completed the systemic fair marketideology scale at the end of the questionnaire(Cronbach’s a in this sample 5 0.66, M 5 –0.68,SEM 5 0.13). An ANOVA with experimental condi-tion as the independent variable and fair marketideology as the dependent variable reveals the maineffect of the experimental condition on fair marketideology (F (2,72)5 4.81,p5 0.01), implying that themanipulation did indeed work.


The results, detailed in Figure 2, provide additionalsupport for Hypothesis 1. An ANOVA with experi-mental condition as the independent variable and thebelief in the CSP–CFP link as the dependent variable

FIGURE 2Belief in the CSP–CFP Link in the Three Experimental Conditions, Study 2

0

0.05

0.1

0.15

0.2

0.25

0.3

Bel

ief

in t

he

CS

P–C

FP

lin

k (+

/–S

EM

)

High system justification Low system justificationExperimental condition

Control


reveals that there is a main effect of the experi-mental condition on the belief in the CSP–CFP link(F (2,72) 5 3.13, p 5 0.0499). The belief in theCSP–CFP link for participants in the high systemjustification condition is significantly higher thanfor participants in the low system justificationcondition (high system justification condition:M50.257, SEM 5 0.039; low system justification con-dition: M 5 0.133, SEM 5 0.035, t (49) 5 2.35, p 50.023). Thus, we find additional support for Hy-pothesis 1; namely, that there is a link between fairmarket ideology and the belief in a positive linkbetween CSP and CFP. The belief in the CSP–CFPlink for participants in the control condition con-taining only the filler sentences is not significantlydifferent from the belief in the CSP–CFP link forparticipants in the high system justification condi-tion (control condition: M 5 0.241, SEM 5 0.042, t(46) 5 0.27, p 5 0.79.) and is only marginally sig-nificantly different from participants in the low sys-tem justification condition (t (49) 5 1.98, p 5 0.054).This indicates that being assigned to the low systemjustification condition had a larger effect than beingassigned to the high system justification condition. Apossible explanation for this could be that our samplefor Study2 consisted solely of business and economicsstudents. As our results from Study 3 (see below) in-dicate, business and economics students justify themarket economy toahigherdegree thanother samples,whichmeans itmightbemoredifficult to increase theirlevels of system justification even further.

STUDY 3: EDUCATIONAL BACKGROUND, FAIRMARKET IDEOLOGY, AND BELIEFS ABOUT THE

CSP–CFP LINK

The purpose of Study 3 is to test Hypothesis 2 andto investigate the role of educational background forindividuals’ fair market ideology and, ultimately, forindividuals’ beliefs in the CSP–CFP link. Educa-tional background is often used in upper echelonsresearch as a proxy for the individualized lenses ofexecutives, and specifically to explain their beliefs.

Sample

We chose a student sample for Study 3 in order toexploit the natural variation in exposure to reasoningabout the market economy system in different edu-cational backgrounds, which allows us to test Hy-pothesis 2. We recruited 124 students from a largeSwiss university. Twenty-two participants wereexcluded because they did not complete the

questionnaire, respectively did not understand theinstructions or the incentive structure (i.e., they failedto correctly answer comprehension check questions).We targeted students from two groups: (a) business,economics, and law; and (b) sociology, psychology,and philosophy. Forty-eight of the participants werewomen (47.6%), 53 (50.5%) were men, and one par-ticipant did not give gender information. The averageage was 22.54 (SD5 2.35) years, and the participantshad studied for an average of 3.33 (SD5 0.97) years.

Measures

Belief about the link between CSP and CFP.Participants played the same prediction game as inStudy 2 (described above). We calculated the vari-able measuring the belief in the CSP–CFP link in thesame way as in Studies 1 and 2.

Robustness check for the way financial perfor-mance is measured. In the prediction game, we ran-domly assigned two different measures of financialperformance to participants. The goal of this manipu-lation was to test whether our results would be sensi-tive to the type of financial performance used todetermine the companies’ financial performanceranks. Approximately half of the students (n5 53) re-ceived rank information based on ROE (as in Studies 1and 2). The remaining students (n5 49) received rankinformationbasedon the relative change in sharepriceover thepast 12months.Contrary toROE, thismeasureis based on the valuation of the company on the stockmarket. We expected similar results from both groups.

Educational background. To measure partici-pants’ exposure to reasoning about the marketeconomy system, participants reported their field ofstudy. We created a dummy variable that was coded1 for business, economics, and law students, and0 for sociology, psychology, and philosophy stu-dents. Because participants’ belief in the CSP–CFPlink could potentially also be influenced by knowl-edge about the CSP–CFP link acquired in a businessethics or an ethics course, we also asked participantswhether they had taken such a course.

Fair market ideology.As in Study 1, participantscompleted the systemic fair market ideologyscale developed and tested by Jost and colleagues(2003a) (Cronbach’s a in this sample 5 0.70, M 5–0.21, SEM 5 0.11).


Table 3 summarizes the descriptive statistics andcorrelations. Table 4 summarizes the results from


our mediation analysis, and Figure 3 illustratesthese results graphically. For the direct path,studying business, economics, or law is positivelyassociatedwith the belief in the CSP–CFP link (c50.105, p 5 0.016). In the indirect path, fair marketideology is significantly greater for participantsstudying business, economics, or law (a 5 1.02,p, 0.001) than for other participants. Holding theacademic field constant, fair market ideology sig-nificantly increases the belief in the CSP–CFP link(b 5 0.043, p 5 0.02).

To test for the hypothesized mediation, we fol-lowed the bootstrap procedure proposed byPreacher and Hayes (Preacher & Hayes, 2004, 2008;see also, Zhao, Lynch, & Chen, 2010). The meanindirect effect is positive and significant (a x b 50.044; the bootstrapped bias corrected 95% confi-dence interval [0.0102, 0.1013] does not include 0,which indicates a significant effect). When the me-diator is included in the regression, the direct effect

is no longer significant (c950.061,p50.187). Sincea x b is significant and c9 is not, the effect can becategorized as an indirect-only mediation. The re-sults from this mediation analysis support Hy-pothesis 2, namely that fair market ideologymediates the relationship between participants’educational background and their belief in theCSP–CFP link.

Our results are not sensitive to the way financialperformance is defined. We did not find a signifi-cant effect of the experimentalmanipulation of thefinancial performance measure (ROE versus rela-tive change in the share price), either overall or asa control variable in all the reported results. Also,our results are not sensitive to age or gender orwhether participants have taken a course in ethicsor business ethics as control variables. Thus,knowledge about the CSP–CFP link that partici-pants could have acquired during these coursesdoes not seem to affect our results.


Variable Mean SD 1 2 3 4 5 6 7

1. Belief in the CSP–CFP link 0.12 0.172. Fair market ideology 20.21 1.09 0.31**3. Educational background 0.75 0.44 0.22* 0.49***4. Age 22.54 2.35 0.04 0.16 0.055. Gender 0.52 0.50 0.04 0.32*** 0.34*** 0.006. Course in business ethics 0.24 0.43 20.04 0.11 0.32*** 20.16 0.21*7. Course in ethics 0.14 0.35 0.18 0.02 0.04 20.01 0.16 0.25*8. Financial performance given in ROE or

share price0.48 0.50 20.05 20.13 20.07 0.16 20.02 20.21* 20.21*

Notes: N 5 102.*p , 0.05

**p , 0.01***p , 0.001

TABLE 4Mediation Analysis from Study 3

Regression paths B SE p

Mediation a path (Educational background on Fair market ideology) 1.020 0.24 ,0.001Mediation b path (Fair market ideology on Belief in the CSP–CFP link) 0.043 0.18 0.020Total effect, c path (Educational background on Belief in the CSP–CFP link; No

mediator)0.105 0.43 0.016

Direct effect c’ (Educational background on Belief in the CSP–CFP link includingFair market ideology as mediator)

0.061 0.046 0.187

Indirect effect (c – c’) with bootstrapped bias-corrected 95% CI 0.044 [0.0102, 0.1013]

Notes: N 5 102, B 5 unstandardized coefficient; CI 5 confidence interval.Because the indirect effect may not be normally distributed, the CI is derived by a bootstrap procedure (here 10,000 resamples).As the CI does not include zero, the criterion for mediation has been meet (Preacher & Hayes, 2004).


STUDY 4: FAIR MARKET IDEOLOGY, THEBELIEF IN THE CSP–CFP LINK, MORALOUTRAGE, AND CSR ENGAGEMENT

Study 4 was designed to investigate the effect ofboth the belief in the CSP–CFP link and fair marketideology on CSR engagement. While we expect thatfair market ideology has a positive indirect effect onCSR engagement via the belief in the CSP–CFP link(Hypothesis 3),we alsohypothesized that fairmarketideology has a negative indirect effect on CSR en-gagement via moral outrage (Hypothesis 4). Thepurpose of Study 4 was to test these hypotheses.

Sample

For study 4,we recruitedmanagers and executivesin class settings. Specifically, we recruited 83MBAsfrom a large Swiss business school, 32 EMBAs froma large Swiss university, as well as 12 EMBAsfrom a large Dutch university.3 We excluded 22participants because they did not complete thequestionnaire, respectively did not understand theinstructions or the incentive structure of their task(i.e., they failed to correctly answer comprehensioncheck questions). In addition to using the exact sameexclusion criteria that we used in Studies 1–3, weexcluded four participants who only submitted theirquestionnaires after the debriefing, and five partici-pants who took multiple days between starting andfinishing the questionnaire. (All our results are ro-bust to the inclusion of excluded participants.) Thisresulted in a final sample of 96 participants, 30 of

whom (31%) were women and 66 (69%) were men.They had an average age of 34 (SD 5 5.7) and anaverage of 10 (SD 5 5.0) years of managerial experi-ence. Twelve percent worked for small companies(1–50 employees), 18% for medium-sized compa-nies (51–500 employees), and 70% for large compa-nies (more than 500 employees).

Measures

Moral outrage.Tomeasureparticipants’potentialto feel outraged by injustices or violations of moralprinciples, we used the moral outrage scale de-veloped by Montada et al. (1986), which was linkedto system justification by prior research (Wakslaket al., 2007). Example items include “I feel morallyoutraged by social injustice” and “I rarely feel bur-dened by the unfairness of this world” (reverse-coded). Participants answeredon a seven-point scaleranging from –3 (“That is not at allwhat I am thinkingor feeling”) to13 (“That is exactlywhat I am thinkingor feeling”). The ten itemswere averaged into amoraloutrage score (Cronbach’s a in this sample 5 0.82,M 5 1.16, SEM 5 0.09).

Tendency for CSR engagement. To measureparticipants’ tendency for CSR engagement, we re-lied on the corporate stakeholder responsibility scaledeveloped by El Akremi, Gond, Swaen, De Roeck,and Igalens (2015). The items on this scale describevarious “actions and policies designed to enhancethe welfare of various stakeholder groups” (ElAkremi et al., 2015: 2). We instructed participantsto imagine that theywere theCEOof a large companyand then asked them to what extent they would en-sure that their company engaged in the actions andpolicies described in the items of El Akremi et al.’s(2015) scale. Specifically,we selected the three itemswith the highest factor loadings, as reported by ElAkremi et al. (2015), for the domains of community-oriented CSR, natural environment-oriented CSR,

FIGURE 3Fair Market Ideology Mediates the Relationship Between Educational Background and Belief in the CSP–CFP

Link, Study 3

A: Educationalbackground

B: Belief in theCSP–CFP link

M: Fair marketideology

c = 0.105*

c'= 0.061,n.s.

b = 0.043*a = 1.02***

3 We recruited multiple samples to ensure sufficientstatistical power. The pattern of results is similar in allthree samples, and all our hypothesis tests also reach sta-tistical significance when only considering the managersand executives recruited in Switzerland. Nevertheless, weused a dummy variable for each sample to control forsample fixed effects in our analyses.


employee-oriented CSR, and supplier-oriented CSR,and averaged these 12 items into a CSR engagementscore (Cronbach’s a in this sample5 0.85,M5 2.95,SEM 5 0.11). Example items are “As the CEO, I willensure that my company invests in humanitarianprojects in poor countries,” “As the CEO, I will en-sure that my company makes investments to im-prove the ecological quality of its products andservices,” “As the CEO, I will ensure that my com-pany promotes the safety and health of its em-ployees,” and “As the CEO, I will ensure that mycompany makes sure that its suppliers (and sub-contractors) respect justice rules in their own work-places.” Participants responded on a scale rangingfrom –5 (completely disagree) through 0 (neitheragree nor disagree) to 15 (completely agree).

Belief about the link between CSP and CFP.Participants played the same prediction game as inStudies 2 and 3 (making ten predictions). As in theother studies, we incentivized participants to makeaccurate decisions. Specifically, the 10% of theparticipants who made the most accurate pre-dictions received 50 Swiss francs each (for theEMBAs and MBAs recruited in Switzerland), or 50euros (approximately USD 50 at the time of thestudy) (for the EMBAs recruited in the Netherlands).To confirm the validity of our prediction game, wealso included a direct question for what participantsbelieve is the effect of social performance on futurefinancial performance,with answers ranging from–3(“strong negative influence”) through 0 (“no influ-ence”) to 13 (“strong positive influence”). The an-swer to this question is significantly correlated withour measure of the CSP–CFP belief based on theprediction game (r 5 0.33, p 5 0.011).

Fair market ideology. As in studies 1, 2, and 3,participants completed the systemic fair marketideology scale developed and tested by Jost andcolleagues (2003a) (Cronbach’s a in this sample 50.70,M 5 0.21, SEM 5 0.12).

Demographics and additional measures. Par-ticipants were asked to indicate their gender andage. As a control variable for any potential effect ofsocial desirability, we asked participants to com-plete the three items of the sincerity subscale fromthe honesty–humility scale, which is part of theHEXACO inventory, developed by Ashton and Lee(2009). (Cronbach’s a in this sample 5 0.60, M 51.40, SEM 5 0.23.) As in Study 1, we also includedparticipants’ political orientation as a control vari-able (on two seven-point scales, one ranging fromliberal to conservative, and one ranging from left-wing to right-wing); their level of education (with the

following categories: primary school, secondaryschool, completed high school, undergraduate de-gree, graduate degree, PhD); the number of years ofwork experience; and their rank in theorganizationalhierarchy (in terms of whether they consideredthemselves non-managerial, lower-, middle-, orupper-management). For the ease of interpretation,we coded both the level of education and the hier-archy level as continuous variables, but all results arerobust to using a dummy variable for each level ofthese variables.


Table 5 summarizes the descriptive statistics andcorrelations for Study 4.

We tested Hypotheses 3 and 4 following the rec-ommendations from Preacher and Hayes (2008) formediation analysis with multiple mediators. In linewith these recommendations, we estimated threeregression equations simultaneously, using theseemingly unrelated regression method (Zellner &Huang, 1962). In all regressions, we include gender,age, political orientation, work experience, level ofeducation, hierarchy level, and social desirability, aswell as a dummy variable for each of the differentcourses from which we recruited participants ascontrol variables. All the results reported below arerobust to the exclusion of these control variables.Figure 4 illustrates the different paths in the media-tion analysis, while Table 6 summarizes its results.

First, we regressed the independent variable, fairmarket ideology, on the first mediator, belief in theCSP–CFP link (a1 5 0.04, p 5 0.015). Thus, we rep-licate the effect of fairmarket ideologyon thebelief inthe CSP–CFP link thatwe established in Studies 1–3.Second, we regressed the independent variable, fairmarket ideology, on the second mediator, moraloutrage. Fairmarket ideology significantly decreasesmoral outrage (a2 5 20.18, p 5 0.011). Third, weregressed both mediators and the independent vari-able on the dependent variable, tendency for CSRengagement (paths b1 and b2). The belief in theCSP–CFP link (b1 5 1.68, p 5 0.001) and moraloutrage (b25 0.52, p, 0.001) significantly affect thetendency for CSR engagement.

The bootstrap analysis reveals that the indirecteffect of fair market ideology on tendency for CSRengagement mediated by the belief in the CSP–CFPlink is positive and significant (indirect path a1b150.070; bootstrapped bias-corrected 95% CI: [0.007,0.195]), thereby supporting Hypothesis 3. The boot-strap analysis also reveals that the indirect effect of



Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11

1. Belief in the CSP–CFP link 0.19 0.212. Fair market ideology 0.21 1.15 0.25*3. Moral outrage 1.16 0.91 0.13 20.184. Tendency for CSR engagement 2.95 1.13 0.31** 20.08 0.54***5. Gender 0.69 0.47 0.11 0.18 20.16 20.21*6. Age 33.60 5.70 20.05 20.14 20.02 0.01 20.167. Political orientation

(liberal–conservative)20.55 1.39 0.13 20.07 0.02 0.06 0.01 0.19

8. Political orientation (left–right) 20.07 1.24 20.13 0.22* 0.03 20.05 0.07 20.12 0.019. Work experience 9.56 5.03 0.08 20.07 0.03 0.08 20.14 0.86*** 0.29** 20.1510. Level of education 4.51 1.18 0.13 20.01 0.22* 0.07 20.11 0.16 0.05 20.07 0.1111. Hierarchy level 2.30 1.00 0.01 0.04 20.12 20.04 20.04 0.33*** 0.15 0.19 0.37*** 0.0912. Social desirability 1.40 2.29 0.00 0.03 0.36*** 0.36*** 20.37*** 0.14 20.02 20.16 0.23* 20.03 0.06

Notes: N 5 96.*p , 0.05

**p , 0.01***p , 0.001

20171597

Hafen

bradlan

dWaeger

fair market ideology on tendency for CSR engage-ment mediated by moral outrage is negative andsignificant (indirect path a2b2 5 –0.096; boot-strapped bias-corrected 95% CI: [–0.200, 20.026]),thereby supporting Hypothesis 4.

The total indirect effect of fair market ideologyon tendency for CSR engagement (the sum of thetwo indirect effects reported above) is negative anddoes not reach statistical significance (total indirecteffect 5 20.025; bootstrapped bias-corrected 95%CI: [–0.150, 0.123]).This means that the positive in-direct effect on tendency for CSR engagement,whichstems from the path via the belief in the CSP–CFPlink, is neutralized by the negative indirect effect ontendency for CSR engagement, which stems from thepath via decreased moral outrage.

GENERAL DISCUSSION

Recent literature has focused on the individualcharacteristics of company executives that influencethese executives’ tendency to invest in CSR-relatedactivities. One characteristic that has receivedmuch

attention is the extent to which executives believe inthe business case for CSR. Two premises are perva-sive in existing research on the belief in the businesscase. First, to believe in the business case, executivesneed factual evidence that this business case actuallyexists (Kang et al., 2016; Margolis et al., 2009;Orlitzky et al., 2003; Waddock & Graves, 1997). Andsecond, those executives who do believe in thebusiness case will readily invest in CSR-related ac-tivities (e.g., Bansal & Roth, 2000; Brønn & Vidaver-Cohen, 2009; Chin et al., 2013). The results from ourfour studies tell a different story. On the one hand,rather than focusing on factual evidence, we showthat managers believe in the business case becausethey espouse a fair market ideology. On the otherhand, even though managers who espouse a fairmarket ideology believe in the business case for CSR,they are not more inclined to engage in CSR becausethey also experience weaker emotional reactions toethical problems than managers who do not holdsuch an ideology.

We developed a theoretical framework with thecentral contention that the belief in the business casefor CSR is grounded in fair market ideology; that is,individuals’ tendency to justify and idealize the mar-ket economy system. While this contention mightseem like an inconspicuous departure from prior re-search, our four studies show that it carries importantimplications. In Study 1, we demonstrated that exec-utives’ belief in the business case is indeed groundedin fair market ideology; we confirmed this finding inStudy 2 with an experimental design. In Study 3, wefound a relationship between individuals’ educa-tional background and their belief in the business casefor CSR and that this relationship is mediated by fairmarket ideology. In Study 4, we showed that even

FIGURE 4Overview of the Different Paths in the Mediation

Analysis, Study 4

Belief in theCSP–CFP link

Fair marketideology

Moral outragea2

b1 Tendency for CSRengagement

b2a1

c’

TABLE 6Mediation Analysis from Study 4

Regression paths B SE p

Mediation a1 path (Fair market ideology on Belief in the CSP–CFP link) 0.04 0.017 0.015Mediation a2 path (Fair market ideology on Moral outrage) 20.18 0.064 0.011Mediation b1 path (Belief in the CSP–CFP link on Tendency for CSR engagement) 1.68 0.494 0.001Mediation b2 path (Moral outrage on Tendency for CSR engagement) 0.52 0.117 ,0.001Indirect a1b1path (viaBelief in theCSP–CFP link)with bootstrappedbias-corrected95%CI 0.070 [ 0.007, 0.195]Indirect a2b2 path (via Moral outrage) with bootstrapped bias-corrected 95% CI 20.096 [–0.200, 20.026]Total indirect effect with bootstrapped bias-corrected 95% CI 20.025 [–0.150, 0.123]Direct effect c’ (Fair market ideology on Tendency for CSR engagement including both

mediators)20.08 0.087 0.36

Notes: N 5 96, B 5 unstandardized coefficient; CI 5 confidence interval.Because the indirect effects may not be normally distributed, the CI is derived by a bootstrap procedure (here 10,000 resamples).When the CI does not include zero, the criterion for mediation has been meet (Preacher & Hayes, 2004).


though managers espousing a fair market ideologybelieve in the business case forCSR, they are notmoreprone to engage in CSR because they also experienceweaker emotional reactions to ethical problems thanmanagers who do not hold such an ideology.

Contributions to the Micro-foundations of CSR

Existing research on the micro-foundations ofCSR has argued that both managers’ belief in thebusiness case for CSR and their moral inclinationscan explain why managers are prone to engage inCSR (Chin et al., 2013). However, this existing re-search has treated the belief in the business case andmoral inclinations as independent factors (Agleet al., 1999; Brønn & Vidaver-Cohen, 2009). In ourtheoretical framework, we propose that the two arelinked together by a common antecedent, fair mar-ket ideology. Figure 1 above summarizes the hy-potheses of our theoretical framework, forwhichwefound support in our four studies. Specifically,Figure 1 illustrates that we would draw highlymisleading conclusions if we did not study fairmarket ideology as a psychological antecedent toboth the belief in the business case for CSR and tomoral outrage. For instance, if our knowledge wasrestricted to the relationships among the threeboxes in the upper part of Figure 1, we would con-tend that selecting executives with an educationalbackground in business, economics, or law wouldlead to higher CSR engagements because their ed-ucational background leads them to believe in theCSP–CFP link (which then leads to CSR engage-ment). It is only when we also consider fair marketideology that we can see that selecting executiveswith an educational background in business, eco-nomics, and law will not increase a firm’s CSRengagement—even though executives with such aneducational background believe in the businesscase for CSR. This is so because the relationshipbetween educational background and the belief inthe CSP–CFP link is driven by fair market ideology.In turn, fair market ideology impacts executives’tendency for CSR engagement via two competingpaths. It has an indirect positive impact on thetendency for CSR engagement via the belief in theCSP–CFP link, but it also has an indirect negativeimpact on the tendency for CSR engagement vialack of moral outrage. Importantly, our findingsfrom Study 4 show that the negative impact of a lackof moral outrage neutralizes the positive impact ofthe belief in the CSP–CFP link on executives’ CSRengagement.

These findings highlight the significant potentialthat lies in the study of micro-level foundations foradvancing our knowledge about CSR (Aguinis &Glavas, 2012; Christensen et al., 2014; Kourula &Delalieux, 2016). They further underline theneed formore research that heeds Hambrick’s (2007) call toinvestigate the psychological antecedents of execu-tives’ beliefs, values, and emotions. Our studies alsoillustrate the importance of inferring executivecharacteristics, not only through biographical prox-ies, such as educational background, but also viamore precise measures (Carpenter et al., 2004;Lawrence, 1997), because the rather distant bio-graphical proxies upon which much of the researchon upper echelons relies might not adequatelycapture the underlying hypothesized constructs(Hambrick, 2007; Hambrick et al., 1993). Indeed, it isonly because we have developed a prediction gameto measure the belief in the CSP–CFP link as pre-cisely as possible that we could investigate thecomplex web of relationships illustrated in Figure 1.

Contributions to the Literature on the BusinessCase for CSR

Our results also have implications for the manystudies that have aimed to answer the question ofwhether there is a link between CSP and CFP at thecompany level (Allouche & Laroche, 2005; Margoliset al., 2009; Margolis & Walsh, 2003; Orlitzky et al.,2003). Much of the research in the area implies that,a priori, executives do not believe in the businesscase for CSR, but that scientific evidence of the ex-istence of the business case could convince them tobelieve in it (Baird, Geylani, & Roberts, 2012;Margolis et al., 2009; Orlitzky et al., 2003; VanBeurden & Gossling, 2008). However, we show that,rather thanwaiting for scientific evidence, managersbelieve in the business case for CSR for ideologicalreasons. We also find that, rather than needing to beconvinced, themajority of executives believes in thislink. In fact, 80% of the executives who participatedin our studies believe in the business case for CSR, asexpressed in a positive coefficient in ourmeasure forthe belief in the business case.

The business case for CSR also occupies a centralrole among researchers advocating for an in-strumental viewonCSR (McWilliams&Siegel, 2001;Siegel, 2009; Sundaram & Inkpen, 2004). Scholars inthis tradition suggest that executives should invest inCSR activities when such activities enhance a firm’sprofitability. The underlying assumption in this ar-gument is that executives are rational actors who


apply cost–benefit calculi to determine whethera business case exists. However, our results indicatethat executives believe in the existence of the busi-ness case on the basis of ideological rather than ra-tional considerations.

More critical scholars have noted that executivesendorsing such an instrumental view would refrainfrom investing inCSRwhen theyhavedifficulty seeinga business case for CSR (Banerjee, 2008; Crane,Palazzo, Spence, & Matten, 2014; Hahn, Preuss,Pinkse, & Figge, 2014; Marques & Mintzberg, 2015).However, under the premise that executives endorsingan instrumental view on CSR also hold a fair marketideology, the results from our four studies suggest thateven though these executives believe in the businesscase, theynevertheless refrain from investing inCSR. Itis thus not so much the difficulty of believing in thebusiness case, but rather their lack of moral emotionsthat prevents executives with an instrumental viewfrom investing inCSR.Hence, our results point towarda potential new micro-level explanation for why ex-ecutives with an instrumental CSR orientation mayrefrain from investing in CSR activities.

Contributions to System Justification Theory

Our results show that those individuals who studybusiness, economics, or law, and therefore have thehighest level of formal knowledge about companiesand the economic system, are also those with thehighest level of ideology about the market economysystem. Thereby, our results support system justifi-cation theorists’ proposition that more knowledgedoesnotnecessarilymake ideologiesobsolete, but caninstead contribute to their continuing existence. Forinstance, Shepherd and Kay (2012) suggested thatindividuals closely tied to a systemwill increase theiridealization of that system and subsequently avoidsearching for information that could challenge thisidealized view. This notion, that actors who areclosely tied to an institutionalized systemareunlikelyto look for and perceive alternatives to that system, isin line with neo-institutionalists’ argument thatchange in highly institutionalized settings is lesslikely to come from actors deeply ingrained in an in-stitution. Rather, change initiatives are more likely tooriginate fromeitherperipheral actors (Faulconbridge& Muzio, 2016; Greenwood, Raynard, Kodeih,Micelotta, & Lounsbury, 2011) or from actors operat-ing across multiple systems (Greenwood & Suddaby,2006; Smith, Gonin, & Besharov, 2013). Institutionalscholars argue that such actors are well positioned toinitiate change efforts because they are less subject to

institutional pressures for conformity or because theyhave access to prevalent ideas in multiple in-stitutional orders. A system justification perspectivewould add that such actors have a lower tendency toidealize the existing order and thus be more open toalternative arrangements.

Limitations and Future Research

There are certain limitations to this paper, whichwe suggest could be addressed in future research.First, our studies took place in a controlled envi-ronment (i.e., in the classroom). This allowed us totake clean measurements and to exploit randomassignment tomake causal claims, but it might limitthe external validity of our findings. At the sametime, as prior research already provided evidencefor the external validity of the relationship betweenexecutive characteristics and CSR engagement(Chin et al., 2013; Lewis et al., 2014; Petrenko et al.,2016), our explicit goal was to focus on un-derstanding the psychological underpinnings ofsuch executive characteristics and specifically oftheir beliefs and their moral emotions. Further-more, we have established (Study 1) and replicated(Study 4) ourmain findingwith different samples ofexperienced executives—the population that wewant to generalize to –and have developed an in-novative methodology to measure their beliefs re-liably. Nevertheless, further investigations—using,for instance, field data to measure actual CSRengagement—would strengthen the generalizabil-ity of our results. An ideal study would follow ex-ecutives and aim to detect changes in their fairmarket ideology, and then analyze whether suchchanges would manifest themselves in their beliefabout the business case for CSR, in their moraloutrage about ethical problems, and ultimately intheir companies’ CSR engagements.

Second, to make the belief in the business case forCSR measureable and to make our analysis feasible,we represented the belief within one single re-gression coefficient. Thus, we have assumed thatexecutives conceive of the relationship betweenCSPand CFP as linear. While existing research on exec-utives’ belief in the CSP–CFP link implies such lin-earity (e.g., Chin et al., 2013), it would be interestingto explicitly test this assumption in the future. Forinstance, executives might believe that there are di-minishing returns to CSR engagements and assumethat, while going from zero CSR engagementsto moderate levels of CSR increases financialperformance, going from very high levels of CSR to


even higher levels will at some point decrease fi-nancial performance (Flammer, 2015).

Third, the present article focused on beliefs thatare, by definition, at the level of the individual ex-ecutive. As executives usually make decisions inteams (Carpenter et al., 2004; Hambrick, 2007;Hambrick &Mason, 1984), teammembersmight notonly be diverse in terms of their beliefs but also interms of their educational background, their fairmarket ideology, and their level of moral outrage.Potentially, such diversity could boost CSR en-gagement, as diverse teams might include bothmemberswho are sensitive to the ethical dimensionof corporate activities because they feel outraged bymoral problems and those who believe that it paysto engage in CSR.

Fourth, while we investigated the beliefs of execu-tives with the goal of contributing to the micro-foundations of CSR, executives are not the only groupwhose decisions affect companies and the economicsystem. For instance, consumers often evaluate com-panies and their products. Such evaluations are notonly based on the actions of the companies and theattributes of their products, but alsoon theunderlyingmotivation that is assumed to have driven the com-panies’ actions. Specifically, consumers believe thatproducts of companies with CSR engagements havebetter quality and functionality (Chernev & Blair,2015), unless they believe that the company in-tentionally focused on the product’s social value(Newman, Gorlin, & Dhar, 2014). In this case, con-sumers assume that the company sacrificed productquality by directing limited resources towardmakingthe product socially beneficial. Other research hasshown that people evaluate companies less favorablywhen those companies benefit economically frominvesting in social or environmental initiatives, evencompared to companies that do not engage in suchinitiatives at all (Makov&Newman, 2016; Newman&Cain, 2014). More generally, high profits are seen asindicators of low social value (Bhattacharjee, Dana, &Baron, in press). Linking these findings to our studies,it would be interesting to investigate whether peopleevaluate companies differently if they believe in thebusiness case for CSR. Because evaluators are oftenunaware that there is a business case for CSR ina specific company, or where the company’s profitscome from, it would be fruitful for future research toinvestigate the relationships between fair marketideology, the belief in the business case for CSR, andthe evaluations of corporate activities by consumers,or other third parties such as investors, regulators, orthe general public.

CONCLUSION

In this article,we focusedon the ideological origin ofexecutives’belief in thebusiness case forCSRand theiremotional reactions to ethical problems.We found thatexecutives holding a fair market ideology are morelikely to believe in the business case but are less likelyto be morally outraged by ethical problems. Thisideological foundation has important consequencesfor executives’ tendency to engage in CSR activities.Indeed, our results show that even though they believein thebusiness case forCSR, executiveswhoholda fairmarket ideology will not readily engage in CSR be-cause they experience weaker emotional reactions toethicalproblems thanexecutiveswhodonotholda fairmarket ideology. Hence, while existing research con-tends that executives will readily invest in CSR activ-ities if they believe in the business case for CSR (Bairdet al., 2012;Orlitzkyet al., 2003), our findings leadus tocaution against this assertion.

REFERENCES

Agle, B. R., Mitchell, R. K., & Sonnenfeld, J. A. 1999. Whomatters to CEOs? An investigation of stakeholder at-tributes and salience, corporate performance, andCEO values. Academy of Management Journal, 42:507–525.

Aguinis, H., & Glavas, A. 2012. What we know and don’tknow about corporate social responsibility a reviewand research agenda. Journal of Management, 38:932–968.

Allouche, J., & Laroche, P. 2005. A meta-analytical exam-ination of the link between corporate social and fi-nancial performance. Revue Française de GestionDes Ressources Humaines, 57: 18–41.

Allport, G. W. 1954. The historical background of modernpsychology. In G. Lindzey (Ed.), Handbook of socialpsychology, vol. 1: 3–56. Reading, MA: Addison-Wesley.

Antonetti, P., & Maklan, S. 2014. An extended model ofmoral outrage at corporate social irresponsibility.Journal of Business Ethics, 24: 1–16.

Ashton, M. C., & Lee, K. 2009. The HEXACO–60: A shortmeasure of the major dimensions of personality.Journal of Personality Assessment, 91: 340–345.

Baird, P. L., Geylani, P. C., & Roberts, J. A. 2012. Corporatesocial and financial performance re-examined: In-dustry effects in a linear mixed model analysis. Jour-nal of Business Ethics, 109: 367–388.

Banerjee, S. B. 2008. Corporate social responsibility: Thegood, the bad and the ugly. Critical Sociology, 34:51–79.


Bansal, P., & Roth, K. 2000. Why companies go green: Amodel of ecological responsiveness. Academy ofManagement Journal, 43: 717–736.

Barker, V. L., III, &Mueller, G. C. 2002. CEO characteristicsand firm R&D spending. Management Science, 48:782–801.

Barley, S. R. 2007. Corporations, democracy, and thepublic good. Journal of Management Inquiry, 16:201–215.

Batson, C. D., Chao, M. C., & Givens, J. M. 2009. Pursuingmoral outrage: Anger at torture. Journal of Experi-mental Social Psychology, 45: 155–160.

Bhattacharjee, A., Dana, J., & Baron, J. In press. Anti-profitbeliefs: How people neglect the societal benefits ofprofit. Journal of Personality andSocial Psychology.

Bobbio, N., & Cameron, A. 1996. Left and right: The sig-nificance of a political distinction. Chicago, IL: Uni-versity of Chicago Press.

Boltanski, L., & Thevenot, L. 2006. On justification:Economies of worth. Princeton, NJ: Princeton Uni-versity Press.

Bridoux, F., Stofberg, N., & Den Hartog, D. 2016. Stake-holders’ responses to CSR tradeoffs: When other-orientation and trust trump material self-interest.Frontiers in Psychology, 6: 1–18.

Briscoe, F., Gupta, A., & Anner,M. S. 2015. Social activismand practice diffusion: How activist tactics affectnon-targeted organizations. Administrative ScienceQuarterly, 60: 300–332.

Brønn, P. S., & Vidaver-Cohen, D. 2009. Corporatemotivesfor social initiative: Legitimacy, sustainability, or thebottom line? Journal of Business Ethics, 87: 91–109.

Brown, M. E., & Treviño, L. K. 2006. Ethical leadership: Areview and future directions. The Leadership Quar-terly, 17: 595–616.

Brown,M. E., Treviño, L.K., &Harrison,D.A. 2005. Ethicalleadership: A social learning perspective for constructdevelopment and testing. Organizational Behaviorand Human Decision Processes, 97: 117–134.

Campbell, J. L. 2006. Institutional analysis and the paradoxof corporate social responsibility. The American Be-havioral Scientist, 49: 925–938.

Carpenter,M.A.,Geletkanycz,M.A., &Sanders,W.G. 2004.Upper echelons research revisited: Antecedents, ele-ments, and consequences of top management teamcomposition. Journal of Management, 30: 749–778.

Carroll, A. B. 1999. Corporate social responsibility: Evo-lution of a definitional construct. Business& Society,38: 268–295.

Chernev, A., & Blair, S. 2015. Doing well by doing good:Thebenevolenthalo of corporate social responsibility.The Journal of Consumer Research, 41: 1412–1425.

Chin, M. K., Hambrick, D. C., & Treviño, L. K. 2013. Polit-ical ideologies of CEOs: The influence of executives’values on corporate social responsibility. Adminis-trative Science Quarterly, 58: 197–232.

Christensen, L. J., Mackey, A., & Whetten, D. 2014. Takingresponsibility for corporate social responsibility: Therole of leaders in creating, implementing, sustaining,or avoiding socially responsible firm behaviors. TheAcademy of Management Perspectives, 28:164–178.

Cichocka, A., & Jost, J. T. 2014. Stripped of illusions? Ex-ploring system justification processes in capitalist andpost‐Communist societies. International Journal ofPsychology, 49: 6–29.

Converse, P. E. 1964. The nature of belief systems in masspublics. In D. Apter (Ed.), Ideology and discontent:206–261. New York, NY: Free Press.

Crane, A., Palazzo, G., Spence, L. J., & Matten, D. 2014.Contesting the value of “creating shared value.” Cal-ifornia Management Review, 56: 130–153.

Crilly, D., Zollo, M., & Hansen, M. T. 2012. Faking it ormuddling through? Understanding decoupling in re-sponse to stakeholder pressures. Academy of Man-agement Journal, 55: 1429–1448.

Day, M. V., Kay, A. C., Holmes, J. G., & Napier, J. L. 2011.System justification and the defense of committedrelationship ideology. Journal of Personality andSocial Psychology, 101: 291–306.

de Bakker, F. G., den Hond, F., King, B., & Weber, K. 2013.Social movements, civil society and corporations:Taking stock and looking ahead. Organization Stud-ies, 34: 573–593.

Deckop, J. R.,Merriman,K.K., &Gupta, S. 2006.The effectsof CEOpay structure on corporate social performance.Journal of Management, 32: 329–342.

Delmas, M. A., & Toffel, M. W. 2008. Organizational re-sponses to environmental demands: Opening theblack box. Strategic Management Journal, 29:1027–1055.

Detert, J. R., Treviño, L. K., & Sweitzer, V. L. 2008. Moraldisengagement in ethical decision making: A study ofantecedents and outcomes. The Journal of AppliedPsychology, 93: 374–391.

Durand, R., & Vergne, J.-P. 2015. Asset divestment as a re-sponse to media attacks in stigmatized industries.Strategic Management Journal, 36: 1205–1223.

Eccles, R. G., Ioannou, I., & Serafeim, G. 2014. The impactof corporate sustainability on organizational pro-cesses and performance. Management Science, 60:2835–2857.

El Akremi, A., Gond, J.-P., Swaen, V., De Roeck, K., &Igalens, J. 2015. How do employees perceive corpo-rate responsibility? Development and validation of


amultidimensional corporate stakeholder responsibilityscale. Journal ofManagement. https://doi.org/10.1177/0149206315569311.

Elsbach, K. D., & Kramer, R. M. 1996. Members’ responsesto organizational identity threats: Encountering andcountering the Business Week rankings. Administra-tive Science Quarterly, 41: 442–476.

Etzion, D., & Ferraro, F. 2010. The role of analogy in theinstitutionalization of sustainability reporting. Orga-nization Science, 21: 1092–1107.

Faulconbridge, J., & Muzio, D. 2016. Global professionalservice firms and the challenge of institutional com-plexity: “Field relocation” as a response strategy.Journal of Management Studies, 53: 89–124.

Feinberg, M., & Willer, R. 2011. Apocalypse soon?: Diremessages reduce belief in global warming by contra-dicting just-world beliefs. Psychological Science, 22:34–38.

Feygina, I., Jost, J. T., & Goldsmith, R. E. 2010. Systemjustification: The denial of global warming, and thepossibility of “system- sanctioned change.” Person-ality and Social Psychology Bulletin, 36: 326–338.

Fiss, P. C., & Zajac, E. J. 2004. The diffusion of ideas overcontested terrain: The (non-)adoption of a shareholdervalue orientation among German firms. Administra-tive Science Quarterly, 49: 501–534.

Flammer, C. 2015. Does corporate social responsibilitylead to superior financial performance? A regressiondiscontinuity approach. Management Science, 61:2549–2568.

Friedland, R., & Alford, R. 1991. Bringing society back in:symbols, practices and institutional contradictions. InW. Powell & P. Dimaggio (Eds.), The new institution-alism in organizational analysis: 232–263. Chicago,IL: University of Chicago Press.

Garriga, E., &Mele, D. 2004.Corporate social responsibilitytheories: Mapping the territory. Journal of BusinessEthics, 53: 51–71.

Gond, J.-P., & Palazzo, G. 2008. The social construction ofthe positive link between corporate social and finan-cial performance. Academy of Management BestPaper Proceedings.

Greenwood, R., Raynard,M., Kodeih, F., Micelotta, E. R., &Lounsbury, M. 2011. Institutional complexity and or-ganizational responses. The Academy of Manage-ment Annals, 5: 317–371.

Greenwood, R., & Suddaby, R. 2006. Institutional en-trepreneurship in mature fields: The big five ac-counting firms. Academy of Management Journal,49: 27–48.

Hahn, T., Preuss, L., Pinkse, J., & Figge, F. 2014. Cognitiveframes in corporate sustainability: Managerial

sensemaking with paradoxical and business caseframes. Academy of Management Review, 39:463–487.

Hambrick, D. C. 2007. Upper echelons theory: An update.Academy of Management Review, 32: 334–343.

Hambrick, D. C., Geletkanycz, M. A., & Fredrickson, J. W.1993. Top executive commitment to the status quo:Some tests of its determinants. Strategic Manage-ment Journal, 14: 401–418.

Hambrick, D. C., &Mason, P. A. 1984. Upper echelons: Theorganization as a reflection of its top managers.Academy of Management Review, 9: 193–206.

Ioannou, I., &Serafeim,G. 2012.What drives corporate socialperformance? The role of nation-level institutions.Journal of InternationalBusinessStudies, 43: 834–864.

Jost, J. T., & Banaji, M. R. 1994. The role of stereotyping insystem-justification and the production of false con-sciousness. British Journal of Social Psychology, 33:1–27.

Jost, J. T., Banaji, M. R., & Nosek, B. 2004. A decade ofsystem justification theory: Accumulated evidence ofconscious and unconscious bolstering of the statusquo. Political Psychology, 25: 881–919.

Jost, J. T., Blount, S., Pfeffer, J., & Hunyady, G. 2003a. Fairmarket ideology: Its cognitive-motivational un-derpinnings. Research in Organizational Behavior,25: 53–91.

Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, F.2003b. Political conservatism as motivated socialcognition. Psychological Bulletin, 129: 339–375.

Jost, J. T., & Hunyady, O. 2005. Antecedents and conse-quences of system-justifying ideologies. Current Di-rections in Psychological Science, 14: 498–509.

Jost, J. T., Liviatan, I., van der Toorn, J., Ledgerwood, A.,Mandisodza, A., & Nosek, B. A. 2010. System justifi-cation: How dowe know it’s motivated. In R. Bobocel,A. C. Kay, M. P. Zanna & J. M. Olson (Eds.), The psy-chology of justice and legitimacy: The Ontariosymposium, vol. 11: 173–203. Hillsdale, NJ: LawrenceErlbaum.

Jost, J. T., &Thompson, E. P. 2000.Group-baseddominanceand opposition to equality as independent predictorsof self-esteem, ethnocentrism, and social policy atti-tudes among African Americans and EuropeanAmericans. Journal of Experimental Social Psy-chology, 36: 209–232.

Kang, C., Germann, F., & Grewal, R. 2016. Washing awayyour sins? Corporate social responsibility, corporatesocial irresponsibility, and firm performance. Journalof Marketing, 80: 59–79.

Kay, A. C., & Friesen, J. 2011. On social stability and socialchange understanding when system justification does


https://doi.org/10.1177/0149206315569311

https://doi.org/10.1177/0149206315569311

and does not occur. Current Directions in Psycho-logical Science, 20: 360–364.

Kay, A. C., Gaucher, D., Peach, J. M. , Laurin, K., Friesen, J.,Zanna, M. P., & Spencer, S. J. 2009. Inequality, dis-crimination, and the power of the status quo: Directevidence for a motivation to see the way things are asthe way they should be. Journal of Personality andSocial Psychology, 97: 421–434.

Knight, K. 2006. Transformations of the concept of ideol-ogy in the twentieth century. The American PoliticalScience Review, 100: 619–626.

Kourula, A., & Delalieux, G. 2016. The micro-level foun-dations and dynamics of political corporate social re-sponsibility: Hegemony and passive revolutionthrough civil society. Journal of Business Ethics, 135:769–785.

Kunda, Z. 1990. The case for motivated reasoning. Psy-chological Bulletin, 108: 480–498.

Lawrence, B. S. 1997. Perspective: The black box of organi-zational demography.Organization Science, 8: 1–22.

Lerner, M. J. 1980. The belief in a just world: A funda-mental delusion. New York, NY: Plenum.

Lewis, B. W., Walls, J. L., & Dowell, G. W. S. 2014.Difference in degrees: CEO characteristics and firmenvironmental disclosure. Strategic ManagementJournal, 35: 712–722.

Maak, T., Pless, N. M., & Voegtlin, C. 2016. Businessstatesman or shareholder advocate? CEO responsibleleadership styles and the micro‐foundations of polit-ical CSR. Journal of Management Studies, 53:463–493.

Makov, T., & Newman, G. E. 2016. Economic gains stimu-late negative evaluations of corporate sustainabilityinitiatives. Nature Climate Change, 6: 844–846.

Margolis, J. D., Elfenbein, H. A., & Walsh, J. 2009. Does itpay to be good? A meta-analysis and redirection ofresearch on the relationship between corporate socialand financial performance. Working paper, HarvardBusiness School, Cambridge MA.

Margolis, J. D., & Walsh, J. 2003. Misery loves companies:Rethinking social initiatives by business. Adminis-trative Science Quarterly, 48: 268–305.

Marques, J. C., &Mintzberg, H. 2015.Why corporate socialresponsibility isn’t a piece of cake. MIT Sloan Man-agement Review, 56: 8–11.

Marquis, C., & Qian, C. 2013. Corporate social re-sponsibility reporting in China: Symbol or substance?Organization Science, 25: 127–148.

McDonnell, M.-H., King, B. G., & Soule, S. A. 2015. A dy-namic process model of private politics activist tar-geting and corporate receptivity to social challenges.American Sociological Review, 80: 654–678.

McWilliams, A., & Siegel, D. 2001. Corporate social re-sponsibility: A theory of the firm perspective. Acad-emy of Management Review, 26: 117–127.

Mena, S., & Waeger, D. 2014. Activism for corporate re-sponsibility: Conceptualizing private regulation op-portunity structures. Journal of ManagementStudies, 51: 1091–1117.

Montada, L., Schmitt, M., & Dalbert, C. 1986. Thinkingabout justice and dealing with one’s own privileges.In H. W. Bierhoff, R. L. Cohen & J. Greenberg (Eds.),Justice in social relations: 125–143. New York, NY:Plenum Press.

Muller, A., & Kolk, A. 2010. Extrinsic and intrinsic driversof corporate social performance: Evidence from for-eign and domestic firms in Mexico. Journal of Man-agement Studies, 47: 1–26.

Nam,H.H., Jost, J. T., &VanBavel, J. J. 2013. “Not for all thetea in China!” Political ideology and the avoidance ofdissonance-arousing situations. PLoS One, 8: 1–8.

Newman,G.E., &Cain,D.M.2014.Tainted altruism:Whendoing some good is evaluated as worse than doing nogood at all. Psychological Science, 25: 648–655.

Newman, G. E., Gorlin, M., & Dhar, R. 2014. Whengoing green backfires: How firm intentions shape theevaluation of socially beneficial product enhance-ments. The Journal of Consumer Research, 41:823–839.

Orlitzky, M., Schmidt, F. L., & Rynes, S. L. 2003. Corporatesocial and financial performance: A meta-analysis.Organization Studies, 24: 403–441.

Paharia, N., Vohs, K. D., & Deshpande, R. 2013. Sweatshoplabor iswrong unless the shoes are cute: Cognition canboth help and hurt moral motivated reasoning. Or-ganizational Behavior and Human Decision Pro-cesses, 121: 81–88.

Palazzo, G., Krings, F., & Hoffrage, U. 2012. Ethical blind-ness. Journal of Business Ethics, 109: 323–338.

Patriotta, G., Gond, J.-P., & Schultz, F. 2011. Maintaininglegitimacy: Controversies, orders of worth, and publicjustifications. Journal of Management Studies, 48:1804–1836.

Petrenko, O. V., Aime, F., Ridge, J., & Hill, A. 2016. Cor-porate social responsibility or CEO narcissism? CSRmotivations and organizational performance. Strate-gic Management Journal, 37: 262–279.

Porter, M. E., & Kramer, M. R. 2002. The competitive ad-vantage of corporate philanthropy.Harvard BusinessReview, 80: 57–68.

Porter, M. E., & Kramer, M. R. 2006. Strategy and society:The link between competitive advantage and corpo-rate social responsibility. Harvard Business Review,84: 78–92.


Preacher, K. J., & Hayes, A. F. 2004. SPSS and SASprocedures for estimating indirect effects in simplemediation models. Behavior Research Methods,Instruments, & Computers, 36: 717–731.

Preacher, K. J., & Hayes, A. F. 2008. Asymptotic andresampling strategies for assessing and comparingindirect effects in multiple mediator models. Behav-ior Research Methods, 40: 879–891.

Proudfoot, D., & Kay, A. C. 2014. System justification inorganizational contexts: How a motivated preferencefor the status quo can affect organizational attitudesand behaviors. Research in Organizational Behav-ior, 34: 173–187.

Roulet, T. J., &Touboul, S. 2015.The intentionswithwhichthe road is paved: Attitudes to liberalism as de-terminants of greenwashing. Journal of BusinessEthics, 128: 305–320.

Scherer, A. G., Rasche, A., Palazzo, G., & Spicer, A. 2016.Managing for political corporate social responsibility:New challenges and directions for PCSR 2.0. Journalof Management Studies, 53: 273–298.

Shao, R., Aquino, K., & Freeman, D. 2008. Beyond moralreasoning: A review of moral identity research and itsimplications for business ethics. Business EthicsQuarterly, 18: 513–540.

Shepherd, S., & Kay, A. C. 2012. On the perpetuation ofignorance: System dependence, system justification,and the motivated avoidance of sociopolitical in-formation. Journal of Personality and Social Psy-chology, 102: 264–280.

Shepherd, S., & Kay, A. C. 2014. When government confi-dence undermines public involvement in modern di-sasters. Social Cognition, 32: 206–216.

Shepherd, S.,Kay,A.C., Landau,M. J., &Keefer, L.A. 2011.Evidence for the specificity of control motivations inworldview defense: Distinguishing compensatorycontrol from uncertainty management and terrormanagement processes. Journal of ExperimentalSocial Psychology, 47: 949–958.

Siegel, D. S. 2009. Green management matters only if ityields more green: An economic/strategic perspec-tive.TheAcademyofManagement Perspectives, 23:5–16.

Slater, D. J., & Dixon-Fowler, H. 2010. The future of theplanet in the hands of MBAs: An examination of CEOMBA education and corporate environmental perfor-mance. Academy of Management Learning & Edu-cation, 9: 429–441.

Smith,W.K., Gonin,M., &Besharov,M. L. 2013.Managingsocial-business tensions: A review and researchagenda for social enterprise. Business Ethics Quar-terly, 23: 407–442.

Spiess, S.-O., Mueller, K., & Lin-Hi, N. 2013. Psychologicalfoundations of corporate social responsibility: Theimportance of “avoiding bad.” Industrial and Orga-nizational Psychology: Perspectives on Science andPractice, 6: 383–386.

Stahl,G.K., &DeLuque,M. 2014.Antecedents of responsibleleader behavior: A research synthesis, conceptualframework, and agenda for future research. The Acad-emy of Management Perspectives, 28: 235–254.

Sundaram, A. K., & Inkpen, A. C. 2004. The corporate ob-jective revisited.Organization Science, 15: 350–363.

Tan, X., Liu, L., Huang, Z., Zhao, X., & Zheng, W. 2016.The dampening effect of social dominance orienta-tion on awareness of corruption: Moral outrageas a mediator. Social Indicators Research, 125:89–102.

Van Beurden, P., & Gossling, T. 2008. The worth of values:A literature review on the relation between corporatesocial and financial performance. Journal of BusinessEthics, 82: 407–424.

Vogel, D. J. 2005. Is there a market for virtue? The businesscase for corporate social responsibility. CaliforniaManagement Review, 47: 19–45.

Waddock, S. A., & Graves, S. B. 1997. The corporate socialperformance–financial performance link. StrategicManagement Journal, 18: 303–319.

Wakslak, C. J., Jost, J. T., Tyler, T. R., & Chen, E. S. 2007.Moral outrage mediates the dampening effect of sys-tem justification on support for redistributive socialpolicies. Psychological Science, 18: 267–274.

Waldman, D. A., & Balven, R. M. 2014. Responsible leader-ship: Theoretical issues and research directions. TheAcademy of Management Perspectives, 28: 224–234.

Weaver, G. R., Reynolds, S. J., & Brown, M. E. 2014. Moralintuition connecting current knowledge to future or-ganizational research and practice. Journal of Man-agement, 40: 100–129.

Weber, K., Rao, H., & Thomas, L. G. 2009. From streets tosuites: How the anti-biotech movement affected Ger-man pharmaceutical firms. American SociologicalReview, 74: 106–127.

Wiersema, M. F., & Bantel, K. A. 1992. Top managementteam demography and corporate strategic change.Academy of Management Journal, 35: 91–121.

Zellner, A., & Huang, D. S. 1962. Further properties ofefficient estimators for seemingly unrelated regres-sion equations. International Economic Review, 3:300–313.

Zhao, X., Lynch, J. G., & Chen, Q. 2010. ReconsideringBaron and Kenny: Myths and truths about mediationanalysis. The Journal of Consumer Research, 37:197–206.


Sebastian Hafenbradl ([email protected]) joined IESEBusiness School as an assistant professor, after conductingpostdoctoral studies at the School of Management, YaleUniversity. He received his PhD in management from theFaculty of Business and Economics (HEC) at the UniversityofLausanne.His researchon judgment anddecisionmakinglies at the intersection of psychology and economics. Inparticular, he focuses on managerial contexts, ethics, andthe role played by social and institutional forces.

Daniel Waeger ([email protected]) is an assistant professorat Wilfrid Laurier University. Before joining Wilfrid Lau-rier, Daniel was an assistant professor at the University ofAmsterdam. He received his PhD from the University ofLausanne and conducted post-doctoral studies at North-western University. Daniel’s research interests includesocial movement theory and organization theory aswell as

empirical phenomena such as corporate responsibility andcorporate governance.

APPENDIX 1EXAMPLE PREDICTION

Your prediction for the financial performance two yearslater:

Rank ______ out of 183

Company AFinancial performance:

Rank 95 out of 183(among middle rank performers)

Social performance:Rank 180 out of 183(among the worst 10%)


mailto:[email protected]

mailto:[email protected]

Hubbard, Timothy D., Dane M. Christensen, and Scott D. Graffin. "Higher Highs and Lower Lows: The Role of Corporate Social Responsibility in CEO Dismissal." Strategic Management Journal 38.11 (2017): 2255-2265.

The Private Governance Response to Climate Change

Fordham Environmental Law Review SymposiumFordham Law School

New York, NYMarch 9, 2018

Who Did This?

“XXXX Announces Goal to Eliminate 20 Million Metric Tons of Greenhouse Gas Emissions”

• February 25, 2010• Exceeded reductions target by end of 2015• Joint announcement with EDF• New billion ton initiative in 2017

Private Climate Governance

• Framing: Climate beliefs and mitigation policy support are driven not just by the framing of the problem, but the framing of the solution.

• Governance: Private climate initiatives can contribute a billion tons per year of additional carbon dioxide emissions reductions per year over the next decade. This is not a complete response, but it will buy time, reduce total mitigation costs and reduce the risk of exceeding important thresholds.

Examples of Household ActionsU.S. Residential Electricity Consumption

Lucas Davis, Evidence of a Decline in Electricity Use by U.S. Households, Haas Energy Institute, (May 8, 2017) at https://energyathaas.wordpress.com/2017/05/08/evidence-of-a-decline-in-electricity-use-by-u-s-households/,

based on Davis, Lucas W. “Evidence of a Decline in Electricity Use by U.S. Households,” Economics Bulletin, 2017, 37(2), 1098-1105

https://energyathaas.wordpress.com/2017/05/08/evidence-of-a-decline-in-electricity-use-by-u-s-households/

https://energyathaas.wordpress.com/2017/05/08/evidence-of-a-decline-in-electricity-use-by-u-s-households/

http://www.accessecon.com/Pubs/EB/2017/Volume37/EB-17-V37-I2-P96.pdf

Federal Environmental Regulations(photo courtesy of Michael Gerrard)

Major Pollution Control Statutes 1970-2015(Vandenbergh 2013, 2014)

Included: Excluded:1970 Clean Air Act

National Environmental Policy Act 1972 Federal Water Pollution Control Act

Coastal Zone Management Act1974 Safe Drinking Water Act1976 Resource Conservation and Recovery Act

Toxic Substances Control Act

1977 Surface Mining Control and Reclamation ActClean Air Act AmendmentsClean Water Act

1980Comprehensive Environmental Response, Compensation and Liability Act

1984 Hazardous and Solid Waste Amendments (to RCRA) 1986 Emergency Planning and Community Right-to-Know Act

SARA Amendments (to CERCLA)1990 Oil Pollution Act

Clean Air Act Amendments

1986 SDWA Amendments1987 Water Quality Act1988 FIFRA Amendments

1996 FQPA, SDWA Amendments2002 CERCLA Amendments

Climate Science and Climate MitigationThe Effect of Worldview

• Confirmation bias• People seek out and remember

information that fits with their existing beliefs

• Motivated reasoning• Reasoning that is motivated by desire to

maintain existing beliefs, rather than desire for accuracy

WorldviewBig Government

http://www.gallup.com/poll/166535/record-high-say-big-government-greatest-threat.aspx

• “Record High in U.S. Say Big Government Greatest Threat” (Gallup, 2013)

• 72% of U.S. respondents said that big government was a greater threat to the future of the country than big business or big labor (Gallup 2013)

• 30 States do not have GHG Targets (49% of population/64% of GHG emissions)(RMI, 2017)

• Solution Aversion (Campbell & Kay, 2014)

Solution AversionEffect of Ideology or Worldview on Climate Beliefs

Campbell & Kay, 2014

Campbell & Kay, 2014

IPCC: Likelihood that Observed Climate Change is Mostly Anthropogenic

Climate ChanteInfluences on Climate Beliefs: Worldview

(Vandenbergh, Raimi & Gilligan 2014)

Private Environmental GovernanceConceptual Shift

(Vandenbergh, 2013)

• Actor: Government Private Organization• Action: Law, Policy, Program Private Initiative• Examples

• MSC, FSC• Target & Walmart Toxics Standards• Environmental Supply Chain Requirements• Climate Change

Private Climate GovernanceVandenbergh & Gilligan (2015, 2017)

• Bypass Solution Aversion with Private Sector Responses• Buy Time -- ~ Gigaton of CO2 Per Year• Reduce Worldview-Based Resistance to Climate Science• Examples of Household and Corporate Actions • Theory: Drivers of Private Behavior• New Initiatives

• Corporations (~500 million tons)• Households (~500 million tons)• Other Actors and Actions

Buying TimeThe Paris Gap: Best Case

Reducing Worldview-Based Resistance to Climate SciencePrediction Market -- iPredict

Is the Corporate Sector Taking Major Steps?Examples of Corporate Actions

(Sources: David Gardiner & Associates, 2017; Rocky Mountain Institute, 2017; Power Forward 3.0)

• Targets -- Nearly half of the companies in the Fortune 500 and 63 percent of the Fortune 100 have set targets to reduce GHG emissions, improve energy efficiency, and increase renewable energy.

• Internal Carbon Taxes – Over 400 companies have an internal price on carbon.

• RE100 -- 111 companies have committed to 100% renewable power (e.g., GM, Mars, Unilever, VF Corporation, Nike, and Apple).

• Corporate Renewable Energy Buyers’ Principles -- 65 companies have signed on (over 48 million MWh of annual demand by 2020). (

• Corporate-NGO Initiatives -- Over 2,700 Companies are involved in at least one of nine climate platforms.

Is Our Billion Ton Claim Plausible?Theory: A Model of Private Climate Governance

Motivations for Emissions Reductions• Self Interest

• Financial Gains from Accelerating Efficiency• Market Failures (e.g., airlines, shipping, potato chips)• Behavioral Failures (e.g., idling)

• Third Party Pressure• Organizations (e.g., investors, lenders, insurers)• Individuals (e.g., social norms)

• Beyond Self Interest (e.g., personal norms)

TheoryDrivers of Corporate Emissions Reductions

• Accelerate Efficiency• Brand Reputation • Retail Customers• Corporate Customers• Investors • Lenders• Employee Morale• Assure Supply• Manager Norms

• Reduce Government Regulation

• Shape Government Regulation

• Raise Rivals’ Costs

Driver: Brand ReputationPrivate Sector Demand for Renewable Power in the SE

“Big tech companies like Google and Facebook are on the way to powering their data centers with renewable energy, but an obstacle stands in their path: the biggest utilities in Virginia and North Carolina, according to a new report from Greenpeace….”

David Ferris and Kristi E. Swartz, Southern utilities stand in the way of making the Internet greener – report, Energy Wire (May 12, 2015)

Regulators

Advocacy Groups

Investors

Retail Customers

Lenders

Employees

Retailer Manufacturer

Supplier

Supplier

Supplier

• Profit Motive: Efficiencies− Market Failures− Behavioral Failures

• Climate Mitigation Motive− Managers− Employees

Cross-Border EffectsGlobal Supply Chain Initiatives

Vandenbergh & Gilligan (2015, 2017)

Theory of Private Climate GovernanceOther Actors and Actions

• Examples• Religious Organizations• Insurers• Small Businesses• Universities• Rice Cultivation• Food Waste• Recycling

Private Climate Governance

• Framing: Climate beliefs and mitigation policy support are driven not just by the framing of the problem, but the framing of the solution.

• Governance: Private climate initiatives can contribute a billion tons per year of additional carbon dioxide emissions reductions per year over the next decade. This is not a complete response, but it will buy time, reduce total mitigation costs and reduce the risk of exceeding important thresholds.

Private Climate Governance Policy Gap

• Multi-Stakeholder Initiative• Identify Overall Opportunity• Prioritize Most Promising Immediate

Opportunities• Identify Private Actors to Address Them• Identify Sources of Financial Support• Identify Basic and Applied Areas Where

Research is Needed• Assess and Report on Outcomes

Private Climate Governance Project

• MICHAEL P. VANDENBERGH & JONATHAN M. GILLIGAN, BEYOND POLITICS: THE PRIVATEGOVERNANCE RESPONSE TO CLIMATE CHANGE (Cambridge University Press) available at https://www.amazon.com/Beyond-Politics-Governance-Response-Business/dp/1316632482/

• Michael P. Vandenbergh & Jonathan M. Gilligan, Beyond Gridlock, 40 COLUM. ENVTL. L.J. 217 (2015) available at http://ssrn.com/abstract=2533643

• Climate Change Research Network Website at http://law.vanderbilt.edu/academics/academic-programs/environmental-law/climate-change-network/

• Behavioral Wedge Website at https://my.vanderbilt.edu/behavioralwedge/

• Beyond Politics Website at www.beyondpoliticsbook.com

• Buying Time TEDx Talk at http://youtu.be/2bXNcEQ6QX0

https://www.amazon.com/Beyond-Politics-Governance-Response-Business/dp/1316632482/

http://ssrn.com/abstract=2533643

http://law.vanderbilt.edu/academics/academic-programs/environmental-law/climate-change-network/

https://my.vanderbilt.edu/behavioralwedge/

http://www.beyondpoliticsbook.com/

http://youtu.be/2bXNcEQ6QX0

Andrew Shakalis Supplemental Reading: Taken from Unilever’s public global website -

• https://www.unilever.com/ • https://www.unilever.com/sustainable-living/

https://www.unilever.com/sustainable-living/our-strategy/

https://www.unilever.com/

https://www.unilever.com/sustainable-living/

https://www.unilever.com/sustainable-living/our-strategy/

Documents

The Fordham Environmental Law Review presents Corporate ... · dismissal. These findings provide novel insight into how CEOs’ career outcomes may be affected by earlier CSR decisions