Content uploaded by Patrick Haack
Author content
All content in this area was uploaded by Patrick Haack on Feb 11, 2020
Content may be subject to copyright.
1
A BAIT-AND-SWITCH MODEL OF
CORPORATE SOCIAL RESPONSIBILITY
Patrick Haack
(Corresponding author)
patrick.haack@unil.ch
University of Lausanne
Dirk Martignoni
dirk.martignoni@usi.ch
Università della Svizzera italiana
Dennis Schoeneborn
ds.msc@cbs.dk
Copenhagen Business School
Leuphana University of Lüneburg
Pre-print version
Article accepted for publication in the Academy of Management Review
Acknowledgements:
We thank handling editor Heli Wang and three anonymous reviewers for their guidance and
developmental feedback during the review process. We furthermore received, based on previ-
ous drafts, valuable comments and suggestions by Blagoy Blagoev, Lars T. Christensen, Andy
Crane, Peer Fiss, Mikkel Flyverbom, Mike Lounsbury, Jim March, Mette Morsing, Andreas
Rasche, Anna Stöber, and Klaus Weber. We are also grateful for the helpful feedback by par-
ticipants at the Academy of Management Annual Meeting 2015 as well as in research semi-
nars at the Universities of Lausanne, Stanford, and Zurich. This research has been supported
through the Governing Responsible Business (GRB) cluster at Copenhagen Business School.
2
A BAIT-AND-SWITCH MODEL OF
CORPORATE SOCIAL RESPONSIBILITY
ABSTRACT
The notion that transparency forces organizations to eschew decoupling and embrace substan-
tive adoption represents an important assumption in the corporate social responsibility (CSR)
literature. Conversely, research on learning and social control has considered opacity—
understood as a lack of transparency—to be conducive to substantive CSR adoption. These
opposing viewpoints highlight a fundamental tension: Is transparency good or bad for sub-
stantive adoption? This paper resolves this tension by asking an alternative question: When is
transparency good or bad, and why? We advance a dynamic perspective, which conceives
transparency and opacity as transitory phenomena, and we specify the boundary conditions
for which either enduring or transitory forms of transparency and opacity further the substan-
tive adoption of CSR. Our analyses reveal that, for circumstances under which the motivation
of ceremonial adoption is hypocritical (rather than opportunistic) and where both substantive
adoption and practice abandonment are difficult, the former can be maximized by first allow-
ing organizations to adopt a CSR practice ceremonially under opacity (“bait”), and then
prompting ceremonial adopters to become substantive adopters through a shift to transparency
(“switch”). Specifying this bait-and-switch mechanism and its underlying contingencies re-
veals a hitherto unexplored, and potentially more effective, pathway towards the institutional-
ization of CSR.
Keywords: corporate social responsibility, decoupling, formal modeling, institutional theory,
practice abandonment, practice adoption, opacity, transparency
3
The last two decades have witnessed the emergence of a plethora of principle-based initi-
atives, certifications, reporting and accountability frameworks, and other modes of industry
self- or co-regulation in the realms of human rights, social rights, and environmental protec-
tion (Rasche & Waddock, 2017). Organizations are under pressure to integrate such corporate
social responsibility (CSR) practices into their strategies and operations, and there is a grow-
ing research interest in the antecedents, outcomes, and processes of CSR (Aguinis & Glavas,
2012; Wang, Tong, Takeuchi, & George, 2016). Given the prevalence and growing signifi-
cance of CSR, management scholars have sought to identify the conditions that facilitate the
substantive adoption of CSR, that is, the actual implementation of a CSR practice in an organ-
ization’s core activities and processes such that it is likely to become an integral and enduring
part of the organization (e.g., Durand, Hawn, & Ioannou, 2019; Zeitz, Mittal, & McAulay,
1999).
In this context, research suggests that transparency, understood as conditions that make it
relatively easy for external observers to accurately evaluate the degree to which CSR is im-
plemented, can help reveal a lack of substantive adoption (e.g., Marquis & Qian, 2014). As a
consequence, transparency may threaten the organization’s legitimacy and subject it to strong
criticism and pressure to justify its failure to adopt a CSR practice substantively (Christmann
& Taylor, 2006; Lange & Washburn, 2012). Scholars thus conclude that transparency fosters
the substantive adoption of CSR and helps eschew ceremonial adoption and decoupling, the
merely symbolic pretense of following a practice without embedding it in the organization’s
core activities (Boiral, 2007). However, other scholars suggests that the call for transparency
is insufficient as a means to promote the substantive adoption of CSR and may even turn out
to be counterproductive for enhancing CSR in business operations (Gold & Heikkurinen,
2018). Indeed, research on learning and social control (Bernstein, 2012; 2017) demonstrates
that opacity, understood as conditions under which it is relatively difficult for external ob-
4
servers to accurately determine the degree to which a practice is implemented, is a necessary
condition for substantive adoption, because it enables organizations to explore, embrace, and
eventually enact the behavioral prescriptions enshrined in CSR policies and principles, a con-
clusion that is diametrically opposed to those who emphasize the beneficial effects of trans-
parency.
We seek to reconcile the fundamental tension between “transparentist” and “opacitist”
positions to advance theory on substantive CSR adoption. By developing a Markov chain
model, a formal modeling technique frequently applied in organization science (e.g., Pent-
land, Hærem, & Hillison, 2010), we demonstrate that the alleged contradiction between trans-
parency and opacity can be resolved in a dynamic perspective. The Markov model allows us
to consider the possibility that evaluation regimes (reflecting the degree to which the imple-
mentation of CSR practices is visible to external observers) may change, even drastically,
over time. We are particularly concerned with the question of under which circumstances en-
during or transitory forms of transparency and opacity further the substantive adoption of
CSR in a given field or industry. Specifically, we compare adoption outcomes for four ideal-
typical regime sequences: enduring opacity, enduring transparency, transitory opacity (i.e.
opacity followed by transparency), and transitory transparency (i.e. transparency followed by
opacity).
Our analyses reveal that different regime sequences can be optimal, depending on the
practice adoption rate (i.e., the likelihood that an organization adopts a CSR practice substan-
tively) and the practice abandonment rate (i.e., the likelihood that an organization abandons
substantive adoption). Importantly, if both of these rates are low, the share of substantive
adoption can be maximized under a regime sequence of transitory opacity. In such circum-
stances, opacity allows organizations to adopt the practice ceremonially in a first step (“bait”),
and then a switch to a regime of transparency prompts ceremonial adopters to become sub-
5
stantive adopters in a second step (“switch”). Specifying this “bait-and-switch” mechanism
and its underlying boundary conditions reveals an unexplored conundrum: Even though trans-
parency constitutes a central tenet of CSR research and has been shown to be essential to ad-
vance the organizational embedding of CSR, a regime sequence of enduring transparency may
result in a lower share of substantive CSR adoptions compared to transitory opacity.
Our paper makes three primary contributions. First, we employ an explicit processual
perspective that reveals that both transparentist and opacitist perspectives are valid (Poole &
Van de Ven, 1989). Reconciling conflicting viewpoints, we demonstrate that, for certain con-
ditions, a regime sequence of transitory opacity maximizes the extent of substantive adoption,
while enduring transparency is less effective. Our model thus challenges the “transparency
imperative” (Ringel, 2019)—that is, the largely taken-for-granted assumption that transparen-
cy is beneficial for the institutionalization of CSR. Second, our model has profound implica-
tions for institutional theory. Because decoupling policy from practice can be helpful to the
diffusion of ceremonial adoption, which then may trigger a switch from opacity to transparen-
cy that enforces substantive adoption, it can also be conducive to the institutionalization of
CSR. Hence, while decoupling is commonly conceived as a strategy to ensure legitimacy in a
context of institutional complexity (Meyer & Rowan, 1977), it can also be conceptualized as
an endogenous source of institutional change (Boxenbaum & Jonsson, 2017). Third, by rec-
ognizing the “value of method in advancing theory” (Greenwald, 2012: 106), our paper ex-
emplifies how researchers can use formal models to advance theory on CSR.
TRANSPARENCY VERSUS OPACITY
The role social evaluations play in CSR has recently attracted considerable scholarly at-
tention (e.g., Cuypers, Koh, & Wang, 2016); however, the relationship between social evalua-
tion and CSR adoption has not been theorized or studied systematically, thus far. In this pa-
per, we argue that a thorough understanding of the evaluation–adoption link can offer im-
6
portant insights on how to ensure that CSR becomes a fully integrated and widely embraced
practice. The interest in social evaluations is grounded in evidence that organizations and their
activities are increasingly assessed by a variety of evaluators, such as customers, investors,
and society at large. Such assessments are institutionalized over time into structural properties
of the external environment, which in turn shape organizations’ sensitivity to legitimacy pres-
sures (Bitektine & Haack, 2015). With the term “evaluation regime” we refer to the degree to
which the implementation of CSR practices is visible to external observers; thus, it is indica-
tive of the extent to which such observers are able to discern unambiguously whether an or-
ganization has adopted a practice ceremonially or substantively (Crilly, Zollo, & Hansen,
2012). Under an evaluation regime of transparency, disclosure data are widely available, and
it is relatively easy for external observers to accurately determine whether a CSR practice is
adopted substantively or ceremonially. Conversely, under opacity, observability is obstructed,
making it difficult for external observers to unequivocally determine the degree of practice
implementation.1
The prevalent assumption in the CSR literature is that transparency enhances substantive
CSR adoption, as it forces organizations to forgo ceremonial adoption and the habit of policy-
practice decoupling in favor of aligning rhetoric with action. Conversely, there is evidence
that opacity is a necessary condition for substantive CSR adoption, as it fosters organizational
learning and allows adopters to experiment with, and make sense of, the novel practices they
adopt. Table 1 summarizes the key differences of the ideal-type positions of transparentist and
opacitist views, which the next sections discuss in further depth.
--- INSERT TABLE 1 ABOUT HERE ---
1 For instance, transparency can be said to characterize the domain of gender diversity, as the application of gen-
der quotas in corporate boards is publicly available for stock-listed companies and thus can be unambiguously
determined. In contrast, the Swiss agricultural trading industry is characterized by a regime of opacity, since
industry members have been unable and/or unwilling to disclose specific information regarding the degree to
which UN policies and guidelines on human rights and environmental protection are implemented (Berne Decla-
ration, 2019). This example illustrates that opacity can be induced by business firms who are reluctant to disclose
data on CSR performance or who deliberately make disclosures complex and hard to understand (Fabrizio &
Kim, 2019).
7
The Transparentist Perspective
The call for transparency is the hallmark of what we refer to as the transparentist view in
CSR research (e.g., Gilbert, Rasche, & Waddock, 2011; Sethi & Schepers, 2014) as well as in
institutional theory (e.g., Boiral, 2007). Proponents of this view argue that monitoring and
public scrutiny foster the substantive adoption of CSR, since transparency is believed to help
expose “greenwashing,” that is, the hypocritical promotion of a “green” or socially responsi-
ble image (Delmas & Burbano, 2011). From the viewpoint of institutional theory, the notion
of greenwashing reflects that an organization engages in policy-practice decoupling, meaning
that the organization’s formal structure is only loosely linked to its actual activities (Bromley
& Powell, 2012; Meyer & Rowan, 1977). Such decoupling is equivalent to ceremonial adop-
tion and represents a strategy that allows organizations to maintain legitimacy in the face of
contradictory institutional demands, while also helping a policy to be widely diffused (Box-
enbaum & Jonsson, 2017; Bromley & Powell, 2012). However, the widespread diffusion of a
policy and increasing rates of adoption may reveal more about the growing uniformity in for-
mal policies than about the organizational embedding and integration into concrete work ac-
tivities (Schneiberg & Clemens, 2006; Zeitz et al., 1999). Accordingly, researchers in both
business ethics and institutional theory have concentrated on the question of whether organi-
zations actually implement CSR practices (Aravind & Christmann, 2011; Boiral, 2007) and
examined various antecedents and consequences of decoupling (Behnam & MacLean, 2011;
Christmann & Taylor, 2006). In the same context, low entry barriers for CSR adoption and
lax enforcement mechanisms and reporting requirements have been found to encourage shirk-
ing and free riding (Zeyen, Beckmann, & Wolters, 2016) and to institutionalize unethical and
opportunistic behavior within organizations (Weaver, Treviño, & Cochran, 1999).
On the basis of these findings, research has shown that firms are more likely to act in so-
cially responsible ways and to adopt CSR practices substantively when third parties monitor
8
and inspect their behavior (e.g., Marquis, Toffel, & Zhou, 2016). Indeed, institutional theory
assumes, albeit often implicitly, that transparency furthers the substantive adoption of CSR
(e.g., Behnam & MacLean, 2011). For instance, most research on policy–practice decoupling
rests on the assumption that observers are guided by the “logic of confidence and good faith”
(Meyer & Rowan, 1977: 357), according to which external observers trust organizations to act
in line with the legitimate goals of an adopted policy. Hence, observers typically abstain from
evaluation and do not scrutinize the degree of actual implementation (Zajac & Westphal,
2004). However, given that demands for accountability and transparency have intensified in
recent decades (Bromley & Powell, 2012; Wijen, 2014), scholars emphasize that decoupling
may be difficult to maintain (Marquis & Qian, 2014). External observers of corporations, such
as non-governmental organizations (NGOs), are skeptical as to whether rhetorical commit-
ments are indeed genuine, and often penalize decoupling through ‘naming and shaming’ cam-
paigns (den Hond & de Bakker, 2007). Likewise, internal constituents, such as employees,
tend to refuse to support ceremonial behavior or to engage in “Goffmanesque back-
stage/frontstage activities” (Tolbert & Zucker, 1996: 179). This particularly applies to CSR, a
context in which employees search for purpose and meaningfulness through their work and
thereby defy the formation of organizational façades (Carnahan, Kryscynski, & Olson, 2017;
Girschik, 2020). There is substantial support for the claim that the widespread demands for
transparency make it difficult for decoupling, and thus ceremonial adoption, to be perpetuated
in the context of CSR (Marquis et al., 2016). Hence, like their colleagues in business ethics,
institutional theorists suggest that promoting transparency, tightening requirements, and,
where necessary, penalizing ceremonial adoption and decoupling further the substantive adop-
tion and institutionalization of CSR.
The Opacitist Perspective
9
It has been pointed out that “rarely does one hear about any negative effects of transpar-
ency or problems stemming from too much transparency” (Bernstein, 2012: 182). Neverthe-
less, there is evidence that systems of surveillance reduce the intrinsic motivation of organiza-
tional members (Frey, Homberg, & Osterloh, 2013) and negatively affect learning and trust,
thus diminishing organizational performance (Gilbert & Behnam, 2013; Tenbrunsel &
Messick, 1999). Other works suggest that detailed behavioral prescriptions focusing on com-
pliance reduce the decision-making ability of employees regarding moral issues and impede
the discovery of solutions to moral dilemmas (Zhang, Gino, & Margolis, 2018).
Citing the negative effects of transparency, proponents of the opacitist view argue that a
lack of transparency can, in fact, facilitate substantive adoption. Under opaque conditions, it is
difficult for external observers to accurately assess organizational activities, outcomes, and
performance and to make sense of how these interrelate (Bernstein, 2012). Notwithstanding
this difficulty, opacitists argue that opacity can improve organizational learning and thus, by
implication, can increase the likelihood that organizations will adopt practices substantively
(Weick, 1995). From this viewpoint, opacity ensures that ceremonial adopters have sufficient
time and leeway to adapt CSR practices to ensure that they fit with the organization’s culture
and meaning structures (Ansari, Fiss, & Zajac, 2010). Hence, opacitists caution against im-
posing surveillance, as they believe that a focus on compliance can have a detrimental impact
on exploration and creativity and thus can hinder learning. Scholars also point out that trans-
parency may induce organizational members to establish “new backstages” (Ringel, 2019: 5)
by hiding their activities through encryption and other costly means (Bernstein, 2012). Weick
(1995: 183) summarized the opacitist rationale as follows:
If [managers] are forced to walk the talk, this may heighten accountability, but it
is also likely to heighten caution and inertia and reduce risk taking and innovation.
This outcome occurs not just because people are scared. It occurs because people
who are forced to walk the talk prematurely often forgo exploration and walk on
behalf of words that they barely understand. Because things that are poorly under-
10
stood are things that tend to be seen as uncontrollable, they seem like threats ra-
ther than opportunities. Innovation shuts down.
Thus, from an opacitist viewpoint, ceremonial adoption helps adopters understand the
need to honor their promises and to thoroughly implement CSR practices. Accordingly, opaci-
ty enables “bad” managers and firms to explore novel ways of “becoming good” (March,
1989: 263). This understanding is based on the premise that adoption goals are not fixed or
predetermined; rather, goals are newly discovered when certain means become available, spe-
cifically in the context of CSR, where problems are often “wicked” and defy easy solutions
(Haack & Schoeneborn, 2015). It follows that the most effective way to institutionalize CSR
is to encourage experimentation (Ferraro, Etzion, & Gehman, 2015) and allow organizations
to freely decide whether, when, and how to adopt a practice substantively (Ansari et al.,
2010).
From the same standpoint, organizations should not be automatically penalized for hy-
pocrisy and greenwashing. Instead, opacitists propose that decoupling should be tolerated to
some extent, on the grounds that it enables organizations to “talk the walk” (Schoeneborn,
Morsing, & Crane, 2020) and, in so doing, to explore, embrace, and eventually enact ethical
prescriptions (March, 1995; Weick, 1995). Christensen, Morsing, and Thyssen (2013) recon-
sidered the consequences of ceremonial adoption in the CSR context and suggested that “aspi-
rational talk” (i.e., the rhetorical pledges of ceremonial adopters to reduce the gap between
actual and projected behavior) may lead organizational members to reassess their self-
perception and goals. In this view, the ceremonial adoption of a practice can be conducive to
institutionalization, as public commitments to certain CSR goals create pressure for accounta-
bility, which makes organizational members more likely to “live up” to their promises and
shift their behavior toward substantive CSR adoption. These dynamics illustrate that aspira-
tional statements may entail requests for transparency, highlighting that evaluation regimes
can change endogenously.
11
DYNAMICS OF SOCIAL EVALUATION
If considered in isolation, the transparentist and opacitist perspectives each represent a
somewhat static description of social reality that assumes a unidirectional relationship be-
tween the given evaluation regime and an organization’s CSR adoption decision—that is, ei-
ther the ceremonial or substantive adoption of CSR. However, neither perspective has consid-
ered in sufficient depth the possibility that an industry can switch from a regime of opacity to
transparency, or vice versa, nor that such a switch may be triggered by practice adoption out-
comes at the field or industry level. This section explores the notion of changing evaluation
regimes in further depth and elaborates that regime switches can be triggered either exoge-
nously or endogenously. By doing so, we build an internally consistent and explicit processual
perspective on the relationship between social evaluation and practice adoption that aims to
solve the contradiction between transparentist and opacitist viewpoints (Poole & Van de Ven,
1989).
Exogenous triggers may occur through regulatory and legislative interventions that man-
date socially responsible behavior and policies intended to advance substantive CSR practice
adoption (Berger-Walliser & Scott, 2018). These measures can enforce a regime switch, either
from transparency to opacity or, more representative of the context of CSR, from opacity to
transparency. As an example of the latter type of switch, the European Union has recently
acknowledged the importance of transparency on environmental and social matters and thus
enforced the mandatory disclosure of non-financial information by large companies, request-
ing information on policies, risks, and results regarding matters of environmental protection,
human rights, social and labor rights, anti-corruption, and diversity on boards of directors
(European Commission, 2014). Other prominent examples include China’s 2008 mandate that
business firms must disclose their CSR activities (Chen, Hung, & Wang, 2018) and the United
States’ 2010 Dodd–Frank Wall Street Reform and Consumer Protection Act, which requires
12
disclosure of the use of “conflict minerals” in products manufactured by covered entities
(Woody, 2012). Besides such direct interventions, a change of evaluation regimes can be also
enforced indirectly, such as through regulators’ attempts to “nudge” organizations into desired
behavior through the design of regulatory choice architectures (e.g., Pilaj, 2017).
However, switches between evaluation regimes may also occur endogenously. This can
happen, for instance, when the growing prevalence of ceremonial adoptions provokes a self-
reinforcing social dynamic, which induces a switch in evaluation regimes at the industry level.
This notion matches evidence that the early participation of ceremonial adopters in CSR initi-
atives is better than the participation of only those firms that undertake substantive action
(Delmas & Montes-Sancho, 2010). This appears to be the case because growing adoption
numbers and support help CSR initiatives to reach a status of exteriority and facticity (Berger
& Luckmann, 1967), signaling that the prescriptions and principles of the respective practices
are valid and ought to be followed (Cashore, 2002). However, when ceremonial adoption is
ubiquitous in an industry because of low entry barriers and low demands for transparency,
both substantive and ceremonial adopters are categorized as similar members of the same in-
dustry and thus become “tarred with the same brush,” as opacity makes it hard for external
observers to discern differences in each organization’s mode of adoption (Barnett & King,
2008; Yu, Sengul, & Lester, 2008). In such cases, observers may erroneously extrapolate sin-
gle instances of ceremonial adoption to the whole population of organizations (Jonsson,
Greve, & Fujiwara-Greve, 2009). It follows that both ceremonial and substantive adopters
suffer from a negative spillover and are discredited in the eyes of external observers (Diestre
& Rajagopalan, 2014; Yu et al., 2008). Consequently, it is in the interest of substantive
adopters to differentiate themselves from ceremonial adopters and to seek to enforce transpar-
ency, such as by setting up stronger governance and self-regulatory institutions that monitor
compliance and sanction free riding (Desai, 2011). As a result, and fully in line with the trans-
13
parentists’ arguments, increasing numbers of organizations can be expected to switch from
ceremonial to substantive adoption, fueling a self-reinforcing process that facilitates substan-
tive adoption.2 To set such processes of differentiation and “ratcheting up” in motion, it may
be necessary to start from a state of opacity. This can allow the number of ceremonial
adopters to reach a “critical mass”—that is, a tipping point at which substantive adopters de-
mand that the operations of their competitors who adopt merely ceremonially be made ob-
servable and subject to social control at the field level.
MARKOV CHAIN MODELS AS A TOOL FOR THEORY BUILDING
The opposing positions of transparentists and opacitists are both valid and can be recon-
ciled, if reconsidered from a dynamic perspective that treats the evaluation regimes of opacity
and transparency as existing in a sequential succession. To contribute to the development of
such a perspective, and to gain a better understanding of the role of different evaluation re-
gimes in promoting substantive CSR adoption, we employ a formal modeling approach. For-
mal modeling can serve as a helpful tool for generating theory, especially when the focal phe-
nomenon is complex and involves multiple interacting processes, as well as when it is diffi-
cult to obtain large-scale empirical data (Adner, Pólos, Ryall, & Sorenson, 2009). In such con-
texts, formal modeling allows researchers to systematically experiment with assumptions and
constructs to refine and build theory (Harrison, Lin, Carrol, & Carley, 2007). Formal model-
ing also requires scholars to make their assumptions and constructs explicit, thereby reducing
the ambiguity of meaning associated with verbal theorizing (Suddaby, 2010).
2 A self-reinforcing and largely endogenous switch in evaluation regimes is corroborated by the finding that
public benchmarking and rankings enhance the environmental performance of low-performing adopters (Chat-
terji & Toffel, 2010) and the proposition that the differentiation of firms according to certain performance criteria
can lead to a “ratcheting up effect” in CSR performance (Overdevest, 2010). Examples of differentiation include
the tiered system of the Leadership in Energy and Environmental Design (LEED) certification program (Potoski
& Prakash, 2013) and the differentiation framework of the UN Global Compact (Baccaro & Mele, 2011). Differ-
entiation can also occur at the level of CSR initiatives through membership associations, such as the Internation-
al Social and Environmental Accreditation and Labeling (ISEAL) Alliance, which benchmark initiatives and
assess their compliance with best practices (Potoski & Prakash, 2013).
14
By offering precision with respect to the underlying assumptions and logical consistency
of verbal theorizing, as well as an ability to identify unanticipated implications, Markov mod-
els share the general benefits of formal modeling approaches. Markov models are probabilis-
tic models that allow researchers to represent the states of a process over time and to predict
the final state on the basis of the initial states and transition probabilities at each point in time
(Abbott, 1990). They are particularly suitable for tackling a major challenge when studying
the relationship between social evaluation and practice adoption. This is because the task of
assessing the share of substantive adopters within an industry requires a careful and longitudi-
nal investigation of both the diffusion of practice adoption and the quality of practice imple-
mentation. However, ceremonial behavior and decoupled practices are not easy to observe,
especially across many organizations, because organizations are often unwilling or unable to
disclose information about internal CSR activities (Aravind & Christmann, 2011; de Bakker,
Rasche, & Ponte, 2019). Given these impediments, previous research has primarily focused
either on practice diffusion at the field level (e.g., Delmas & Montes-Sancho, 2011) or on the
quality of practice implementation within individual organizations at one point in time (e.g.,
Humphreys & Brown, 2008). However, since adoption decisions involve dynamics at both the
organizational and the field levels (Hoffman, 2001), theory advancement is needed in order to
comprehend how a given evaluation regime affects the transition between different stages of
adoption at the organizational level and how single instances of adoption, in turn, scale up to
promote the diffusion and implementation of practices at the industry or institutional field
level.
Formal models based on Markov chains enable researchers to address this challenge in
two main ways. First, given that Markov models create their own virtual data and allow re-
searchers to compute an infinite number of counterfactual scenarios, they do not entail the
issues of distrust and inadequate disclosure. Second, Markov models help reveal how adop-
15
tion patterns scale up to produce higher-level outcomes, providing a viable method of creating
a “simple theory” of how CSR is institutionalized at the industry level (see Davis, Eisenhardt,
& Bingham, 2007: 481). The Markov approach is thus particularly effective in advancing the-
ory on the interplay between social evaluation and practice adoption.
ANALYTICAL APPROACH
To take first steps toward formalizing a theory of CSR practice adoption, we intentionally
build a parsimonious model. We model the CSR practice adoption process as a Markov pro-
cess and distinguish between three different states: non-adoption (NA), ceremonial adoption
(CA), and substantive adoption (SA). In each period, organizations may remain in their cur-
rent state or move to a different one.3 These moves are reflected in transition probabilities P:
=
For example, the probability that an organization moves from non-adoption to substan-
tive adoption is given by , while the probability that an organization moves from non-
adoption to ceremonial adoption is given by . Figure 1 provides an overview of the
Markov model that we use in this paper.
--- INSERT FIGURE 1 ABOUT HERE ---
If the initial share of organizations in the different states is known, we can compute the
corresponding shares for all subsequent periods and thus the extent of substantive adoption at
each point in time. Specifically, the distribution of states after H periods (i.e., after a certain
number of periods under the regime of transparency) is given by ()=(0), and the ini-
tial distribution of states is given by x(0) and the transition matrix P. Implicit in this formula-
3 Like any theoretical model, ours is a deliberate simplification that serves as point of departure for understand-
ing a much more complex reality. While we could have included more variables, doing so would have come at
the cost of comprehensibility. Thus, we chose to “start somewhere” (Repenning, 2002: 110) and to analyze the
processes we consider to be central to the relationship between social evaluation and practice adoption. While
there is always “one more variable, one more relation to include, one more variable to make endogenous, or one
more feature that someone may want to add to make [the model] more realistic,” the tendency to maximize real-
ism often fails to align with the model’s purpose, constituting a “reality trap” in formal modeling (Burton &
Obel, 2011: 1122).
16
tion is the assumption that the transition probabilities (and accordingly the evaluation regime)
do not change. In our analyses, however, we also consider changes to the evaluation regime—
that is, a switch from opacity to transparency or vice versa. For example, if the evaluation
regime switches from opacity to transparency after R periods, then the state distribution is
given by ()=(0). Similarly, if transparency is followed by opacity, then the
state distribution is given by: ()=(0). We compare the four different process
sequences presented in Table 2.
--- INSERT TABLE 2 ABOUT HERE ---
We set the regime switch to period R = 5 and assume that a switch occurs between two
time periods and is reflected in a change in the respective transition matrix.4 Throughout the
paper, we report analyses for a parsimonious model which does not take into account the pos-
sibility that adoption and abandonment decisions are affected by the decisions of other organ-
izations. In additional analyses we demonstrate that results are also robust to social influence
(see Appendix A).
MODEL SPECIFICATION
In our analyses, we seek to make as few assumptions as possible about the transition
probabilities and instead report the results on as many combinations of transition probabilities
as possible. However, given that each transition matrix has six degrees of freedom, the possi-
ble parameter space is 6*6 = 36-dimensional.5 Reporting its results and analyzing such a high-
dimensional parameter space are extremely difficult, if not impossible. Therefore, in a next
step, we made some plausible assumptions to reduce the dimensionality of the parameter
space and enhance our understanding of the model’s basic dynamics.6 Specifically, in a first
4 We tested the robustness of our key findings to changes in the parameter choices (including, for example, R)
and to alternative formulations of our model (including, for example, more gradual transitions between evalua-
tion regimes). Our key findings are unaffected by these changes (analyses available on request).
5 The row sum must be 100 percent, reducing the degrees of freedom from nine to six for each of the two matri-
ces, one for opacity and the other for transparency.
6 In additional analyses we relax these assumptions and show that our findings hold for a wide range of parame-
17
step, to reduce the parameter space, we hold constant the transition matrix for opacity and
vary only the parameters for the transparency matrix, thus reducing the number of parameters
to six. We chose the following transition probabilities for the opacity matrix:
=0.7 0.2 0.1
0.1 0.7 0.2
0.1 0.2 0.7
These choices for the transition probabilities under a regime of opacity reflect several im-
plicit assumptions. By definition, the evaluation regime of opacity is characterized by restrict-
ed observability and inaccurate observations. Accordingly, in the opacity regime, it is relative-
ly difficult to differentiate between ceremonial and substantive adoption, and the prevailing
conditions make ceremonial adoption (reflected in the entries in the central column) a viable
option. Furthermore, we assume that there is some inertia in organizations, such that main-
taining the status quo is often the most likely outcome (Hannan & Freeman, 1984). This as-
sumption is reflected in the fact the diagonal elements of the matrix are set to values of 0.7 for
all states—and in that sense larger than the probability of any change (1-0.7) = 0.3. Another
assumption is that if non-adopters adopt a practice under a regime of opacity, they are more
likely to adopt it ceremonially than substantively ( >), because an in-depth
implementation tends to be costly and time-consuming (Baumann-Pauly, Wickert, Spence, &
Scherer, 2013; Wickert, Scherer, & Spence, 2016). We also assume that firms that have adopt-
ed a practice only ceremonially are more likely to subsequently adopt it substantively than to
abandon it ( >). This assumption is consistent with the notion of coupling,
which refers to the gradual alignment of structure and actual activity in organizations. Cou-
pling can be driven by cohort replacement (Tilcsik, 2010) and committed employees who tend
to resist and actively work against decoupling (Hallett, 2010). The chosen transition probabili-
ties thus reflect the insight that decoupling is not stable (Tolbert & Zucker, 1996). This view is
in line with Bromley and Powell (2012) and Wijen (2014), who stress that ceremonial adop-
ter combinations (see Appendix B).
18
tion tends to evolve into substantive adoption, supporting the notion that regular exposure to
the goals of CSR practices may progressively lead to goal internalization and organizational
change (Christensen et al., 2013, 2020; Haack, Schoeneborn, & Wickert, 2012). Finally, we
assume that the regime of opacity does not preclude the existence of some “true believers”—
that is, organizations that are genuinely committed to CSR and that thus invest a significant
amount of resources to adopt CSR practices substantively (e.g., Boiral, 2007; Girschik, 2020).
This assumption is reflected in a non-zero value of the probability of becoming a substantive
adopter (i.e., = 0.1 and = 0.2). In additional sensitivity analyses, we excluded
the possibility of true believers under a regime of opacity ( = 0). Our results are not
affected by this more conservative assumption (analyses available on request).
For the regime of transparency, we vary all parameters and make only one assumption: In
a context of transparency, ceremonial adoption is not a viable option for organizations, since
they seek to avoid social disapproval and perceptions of corporate irresponsibility (Lange &
Washburn, 2012). This assumption is substantiated by evidence that, in a context of surveil-
lance and social control, the ceremonial adoption of a practice poses a significant risk for or-
ganizations, especially in the context of CSR (Marquis & Qian, 2014; Wijen, 2014). Indeed,
CSR and other “morally charged” practices mirror a special kind of institutional expecta-
tion—namely, that being legitimate requires being “good” and acting rightfully, not hypocriti-
cally. Consequently, in our model, under a regime of transparency, the probability of an or-
ganization moving to the state of ceremonial adoption or remaining in that state is zero
( = = = 0). In additional analyses, we relax this assumption and in-
stead assume that ceremonial adoption is viable even under a regime of transparency
( = = > 0). These analyses show that the results are not sensitive to
this alternative assumption, provided that the likelihood of ceremonial adoption is lower under
a regime of transparency than under a regime of opacity (analyses available upon request).
19
With this simplifying assumption, the transparency matrix can be fully defined by three pa-
rameters (,, and ):
(,,)=1 0.0
1 0.0
0.0 1
These three parameters can be interpreted in conceptual terms: First, recall that under a
regime of transparency, ceremonial adoption is not feasible; thus, if a practice is adopted, it
can only be adopted substantively. The CSR adoption rate () then reflects the probabil-
ity that an organization that has not yet adopted a CSR practice adopts it substantively in a
given period. Second, in a context of transparency, organizations cannot decouple, and thus
transitions from substantive to ceremonial adoption are impossible ( = 0). However,
since we also consider cases in which a regime of transparency is preceded by a regime of
opacity, some firms may have adopted a practice ceremonially under opacity. Such a ceremo-
nial adoption is unmaintainable and lacks stability under a regime of transparency. The pa-
rameter can thus be understood as the coupling rate—that is, the probability that a
ceremonial adopter turns into a substantive adopter in a given period. Finally, the parameter
refers to the CSR abandonment rate—that is, the probability that a firm drops out of
substantive adoption and moves to non-adoption in a given period.
RESULTS
In Figure 2, we report for all possible combinations of adoption and abandonment rates,
which of the four process sequences (T/T, O/O, O/T, and T/O; see Table 2) maximize the
share of substantive adopters under transparency (averaged over periods t = 1 to t = 25). Each
point in a panel reflects a different combination of adoption rate (, x-axis) and aban-
donment rate (, y-axis). The different colors reflect the regime sequence that is optimal
for a specific combination of adoption and abandonment rate (black = T/T, dark gray = O/O,
white = T/O, light gray = O/T). The three panels reflect different coupling rates—that is, re-
20
spective values for of 0.1 (left panel), 0.5 (middle panel), and 0.9 (right panel). Figure
2 thus displays three different “slices” cut through the parameter space of the transparency
matrix.
--- INSERT FIGURE 2 ABOUT HERE ---
Two findings are noteworthy: First, varying the coupling rate does not change the
general pattern of results. It follows that dynamics related to coupling processes such as wide-
spread “aspirational talk,” which assumes that action eventually follows rhetorical pledges in
the context of ceremonial adoption (Christensen et al., 2013), can be a conducive condition
for a regime sequence to maximize the share of substantive adoption (i.e., it may increase the
parameter space for which a regime sequence is optimal), but it is not a sufficient condition.
The question of whether a particular regime sequence is optimal is determined primarily by
additional factors: the adoption and abandonment rate as well as the relationship between
these rates. Second, our results suggest that both the transparentist and opacitist perspectives
can be valid. For both enduring and transitory forms of transparency and opacity there are
areas for which a given regime sequence maximizes the share of substantive adoption and, in
that sense, “outperforms” the other sequences. In the following sections, we detail the bound-
ary conditions for which the different regime sequences are optimal and address the question
of whether all combinations of adoption and abandonment rates are equally relevant for the
context of CSR. We then explore the mechanisms that drive the optimality of a given regime
sequence.
Boundary Conditions and their Relevance
Enduring transparency (T/T). Enduring transparency (T/T) is the optimal regime se-
quence for the majority of combinations of adoption and abandonment rates. A sufficient
condition for enduring transparency to be optimal is that the adoption rate is relatively
high. Specifically, as long as the adoption rate is larger than 0.52, enduring transpar-
21
ency maximizes the share of substantive adoption, independent of the abandonment rate
. This boundary condition is reflected by line A in Figure 3: For all parameter combi-
nations to the right of line A, enduring transparency is optimal. The optimality of enduring
transparency for a wide range of parameter combinations is consistent with the transpar-
entists’ argument that organizations should seek to “walk the talk”—that is, to promote the
substantive integration of CSR principles into organizational structures and processes by es-
tablishing strict transparency rules to monitor goal achievement (e.g., Behnam & MacLean,
2011).
--- INSERT FIGURE 3 ABOUT HERE ---
To put the value of = 0.52 into perspective, however, recall that the combined
(both ceremonial and substantive) adoption rate under opacity is only + =
0.1 + 0.2 = 0.3. Such high values for imply that substantive adoption is easier under
transparency than under opacity. This implicit assumption seems problematic given that under
transparency, organizations cannot opt for the “easier” and less costly option of ceremonial
adoption, as decoupling is not possible under transparency. Thus, while we report results for
combinations where the substantive adoption rate is higher under transparency than even the
combined adoption rate under opacity ( > +), these combinations are
of limited practical and theoretical relevance. Indeed, high rates of substantive adoption can
be assumed to represent a rare exception in empirical reality due to managers’ hesitation to
make significant investments in the implementation of CSR, given the inconsistent evidence
regarding its impact on organizational performance (e.g., McWilliams & Siegel, 2000). Also,
due to significant implementation costs (Wickert et al., 2016), it is difficult for organizations
to translate CSR policies into practice instantaneously. Instead, CSR adoption is an extended
learning process that takes time (Zadek, 2004). For these reasons, rather than directly adopt-
ing a CSR practice substantively, organizations will often first implement it incompletely or
22
ceremonially. Substantive adoption rates of above a value of 0.5, implying that the
majority of the organizations immediately adopt the practice substantively, are thus rather
unlikely.
For lower adoption rates (to the left of line A in Figure 3), enduring transparency is opti-
mal as long as the adoption rate is larger than > 0.16 (to the right of line E). This par-
ticularly applies to contexts where the abandonment rate is not excessively high relative to the
adoption rate (i.e., below line D). This finding raises the question of whether it can be ex-
pected that the abandonment rate is higher under transparency than opacity ( >
) or even higher than the combined (both ceremonial and substantive) abandonment
rate under opacity ( > +). Under a regime of opacity, organizations that
prefer to abandon a once-adopted CSR practice but that also want to avoid associated risks to
legitimacy can decouple and move from substantive to ceremonial adoption. Accordingly,
abandonment should be easier under opacity than transparency, and, as a consequence, the
abandonment rate under transparency should not exceed the combined abandonment rate un-
der opacity. This condition is reflected by line C in Figure 3 ( = + =
0.1 + 0.2 = 0.3), and abandonment rates above line C are of limited practical and theoretical
relevance.
Optimality of enduring opacity (O/O). The regime sequence of enduring opacity (O/O)
maximizes the share of substantive adoption if the adoption rate is relatively low (i.e., signifi-
cantly lower than < 0.4) and the abandonment rate is relatively high ( > 0.3,
above line C). The optimality of enduring opacity for this combination of parameter condi-
tions corresponds to the contention that a lack of transparency may allow decision-makers to
experiment with a novel practice which increases the likelihood that practices are implement-
ed. In that sense, enduring opacity furthers substantive adoption (Bernstein, 2012). Experi-
mentation suggests that organizations may not only adopt a CSR practice incompletely or
23
ceremonially but may also easily abandon the practice later. Since there are low penalties for
abandonment, the abandonment rate will be rather high. However, a relatively high abandon-
ment rate under a regime of transparency ( > 0.3) seems problematic if compared to
the abandonment rate under a regime of opacity ( = 0.1 and + =
0.1 + 0.2 = 0.3), as it implies that abandoning a practice is easier under a regime of transpar-
ency than under opacity. In the context of CSR, it is hard to imagine that an organization
would be more likely to abandon substantive adoption under a regime of transparency than
under a regime of opacity, since opacity facilitates ceremonial adoption and thus allows or-
ganizations to reap the benefits of being cost-efficient and legitimate at the same time (see
Meyer & Rowan, 1977). Hence, most conditions for which enduring opacity is optimal are of
little theoretical or practical relevance.
For lower abandonment rates (below line C), the abandonment rate must be still relatively
high compared to the adoption rate (i.e., almost twice as high as the adoption rate; above line
D). In the event that organizations’ own experience with the practice is negative, practice
abandonment might outpace practice adoption and thus trigger practice decline (Gaba & Dok-
ko, 2015; Younkin, 2016). Hence, even though enduring opacity can be optimal for certain
combinations of adoption and abandonment rate, CSR can never fully institutionalize (and
thus will fail to become a widely endorsed and taken-for-granted practice in a given domain)
in situations in which the abandonment rate is higher than the adoption rate.
Optimality of transitory transparency (T/O). A necessary condition for transitory
transparency (T/O) to maximize the share of substantive adoption is a low-to-medium adop-
tion rate ( < 0.5) and a relatively high abandonment rate ( > 0.3). Such pa-
rameter combinations are reflected in the white area above line C and to the left of line D.
However, as elaborated above, such high abandonment rates, particularly compared to the
abandonment rates under opacity, imply that abandonment becomes easier under transparency
24
than under opacity. In general, it seems doubtful that the regime sequence of transitory trans-
parency has much bearing in the context of CSR, given that industries typically start under a
regime of opacity and only eventually face greater pressure for transparency (Berger-Walliser
& Scott, 2018). Thus, while there are combinations for which transitory transparency is opti-
mal, they are of little empirical and theoretical relevance.
Optimality of transitory opacity (O/T). Finally, a necessary condition for transitory
opacity (O/T) to maximize the share of substantive adoption is that the adoption rate is low
( < 0.16, to the left of line E) and that the abandonment rate is low ( < 0.3, be-
low line C) and is not more than two times higher than the adoption rate (below line D). As
elaborated above, these conditions represent typical features of the CSR context: High adop-
tion rates are unlikely, since CSR practices are at first implemented incompletely and their
integration is only advanced gradually over time (Zadek, 2004). At the same time, CSR con-
stitutes a morally charged practice that, once adopted, is difficult to abandon without public
criticism and penalties to the adopter’s legitimacy (Haack et al., 2012). Ceremonial adoption
can thus operate coercively by pushing organizations into moral or rhetorical “entrapment”
(Haack & Schoeneborn, 2015; Risse, 2000), making it difficult for them to withdraw from
substantive adoption.7 We conclude that the conditions for which transitory opacity is optimal
are of significant theoretical or practical relevance.
Optimality as a Function of Adopter Motivation
Figure 3 demonstrates that if both the adoption and abandonment rates are low, even
small variations in these rates can change which regime sequence is optimal. In order to refine
our understanding of what variations in the combination of low adoption and abandonment
rates imply, Figure 4 zooms into the lower left corner of Figure 3, with the small rectangular
7 The difficulty of practice abandonment is illustrated by Apple’s attempt to withdraw from the Electronic Prod-
uct Environmental Assessment Tool (EP E AT ), a certification program that purchasers use to assess the environ-
mental impact of electronic products. After massive public outrage, Apple swiftly reinstated its initial promise to
maintain and closely follow the program’s requirements (Kohl, 2016).
25
white box demarcating the parameter combinations that can be considered as being particular-
ly relevant for the context of CSR. Combinations outside the white box lack empirical and
theoretical validity, while combinations within the white box are valid and thus merit analyti-
cal attention. The size of the box is derived from two logical considerations. First, the adop-
tion (abandonment) rate under transparency does not exceed the combined adoption (aban-
donment) rate under opacity. In other words, given the additional option of ceremonial adop-
tion, substantive adoption (abandonment) does not become easier under transparency com-
pared to opacity. Thus, the upper bound for the adoption rate is given by:
+
(upper bound abandonment rate: +). Second, if
organizations adopt a practice substantively under opacity, they will do the same under a re-
gime of transparency. This consideration reflects the notion of “true believers” introduced
above (i.e., organizations that have confidence in in the meaningfulness and value of CSR and
therefore adopt CSR practices substantively independently of the evaluation regime). Thus,
the lower bound for the adoption rate is given by (lower bound abandon-
ment rate: ).
--- INSERT FIGURE 4 ABOUT HERE ---
Within the subset of valid combinations of adoption and abandonment rates (i.e., rates
within the lower and upper bounds), we can logically derive important differences regarding
the organizations’ motivation to adopt a practice only ceremonially. Some organizations can
be labeled as “hypocrites”; that is, they adopt the practice only when ceremonial adoption is
possible (i.e., under a regime of opacity) and would refrain from adopting it when ceremonial
adoption is impossible (i.e., under a regime of transparency). Other organizations can be la-
beled as “opportunists”; that is, they adopt the practice only ceremonially (rather than sub-
stantively) when possible, but they would adopt it substantively when ceremonial adoption is
not feasible (i.e., in a regime of transparency). In other words, hypocritical ceremonial
26
adopters may neither be willing nor able (because of resource constraints) to adopt a CSR
practice substantively and will thus fall back on non-adoption under transparency. In contrast,
opportunistic ceremonial adopters are not willing but are able to adopt a CSR practice sub-
stantively, and therefore they can be forced into substantive adoption under transparency.
That is, they seek to reduce costs related to implementation if possible, but can mobilize the
necessary resources when implementation is necessary (Durand et al., 2019).
We can elicit adopters’ motivations underlying ceremonial adoption by comparing their
adoption rates under opacity and under transparency. To illustrate this logic, consider a com-
bination in which the adoption and abandonment rates under a regime of transparency match
the corresponding values under an evaluation regime of opacity, i.e., = = 0.1
and = = 0.1. For such a combination, transitory opacity is optimal, despite the
fact that all organizations that would adopt the practice ceremonially under a regime of opaci-
ty would refrain from substantive adoption under a regime of transparency. These ceremonial
adopters can be said to act hypocritically rather than opportunistically. Compare this situation
to a combination of = + = 0.3 and = = 0.1: For such a
combination of adoption and abandonment rates, enduring transparency is optimal, since or-
ganizations that would adopt a CSR practice ceremonially under opacity would adopt it sub-
stantively under transparency. These organizations, in turn, can be said to act opportunistically
rather than hypocritically. In conceptual terms, an organization’s motivation (i.e., opportunism
versus hypocrisy) drives the ceremonial adoption of a CSR practice under opacity and that, in
turn, determines which evaluation regime sequence maximizes the share of substantive adop-
tion in a context of low adoption and abandonment rates. If the probability of opportunistic
behavior (or the share of opportunistic organizations) exceeds a particular threshold (reflected
by line E in Figure 4), enduring transparency is the optimal evaluation regime. Only if this
27
probability (or share) is lower than the threshold (i.e., most organizations are hypocrites rather
than opportunists) can the regime sequences of enduring or transitory opacity be optimal.
Process Dynamics and Mechanisms
In a next step, we identify the mechanisms and process dynamics that drive the optimality
of the different regime sequences. The identification of mechanisms, defined as “theoretical
explanations of why focal phenomena or effects occur” (Davis & Marquis, 2005: 336), is es-
sential for improving scholarly understanding of the interplay between social evaluation and
practice adoption. As elaborated above, in order for enduring (O/O) or transitory (O/T) forms
of opacity to be optimal, the primary motivation of ceremonial adopters must be hypocrisy
(i.e., is close to ), whereas enduring transparency (T/T) is optimal if the prima-
ry motivation is opportunism (i.e., is close to +). When opportunism is
the prevalent motivation, organizations take advantage of the chance to adopt the practice
ceremonially under opacity—at the expense of substantive adoption. A transparent evaluation
regime precludes this possibility, and thus some opportunists will adopt the practice substan-
tively. In contrast, if the primary motivation is hypocrisy, a different dynamic emerges. Spe-
cifically, in a first step, organizations may be lured into adopting a practice ceremonially in a
context of opacity. This “bait” is particularly strong when the rate of ceremonial adoption
() is high. In a second step, once the evaluation regime changes to transparency, these
ceremonial adopters then “switch” from ceremonial to substantive adoption.
The bait-and-switch mechanism explains why ceremonial adoption, especially when mo-
tivated by hypocrisy, can be an important stepping stone to the institutionalization of CSR. In
contrast, if the primary motivation for ceremonial adoption is opportunism, then ceremonial
adoption offers a welcome opportunity to reap the benefits of CSR adoption without incurring
its costs. These results clarify that the coupling rate has an amplifying effect on the
optimality of transitory opacity, but is not the only driver behind this optimality. Only if or-
28
ganizations are attracted or “pulled” into ceremonial adoption under a regime of opacity can
transitory opacity result in higher levels of substantive adoption than enduring transparency.
ROBUSTNESS ANALYSES
As mentioned above, we examined the robustness of our findings for different switching
points, gradual regime changes, and the absence of true believers. We also explored model
dynamics for scenarios in which ceremonial adoption is possible both under a regime of
transparency and under a regime of opacity. These analyses, which are available upon request,
demonstrate that the key findings remain essentially unaffected. Additionally, we tested the
robustness of our core results in a variety of alternative model formulations and parameter
choices. First, we examined how social influence may affect the share of substantive adopters
for the different regime sequences (T/T, O/O, T/O, and O/T). While organizations have some
agency and leeway in their adoption decisions, and may differ in their adoption motivations,
they are also embedded in a social context where the behavior of competitors and other field
actors has some bearing on adoption decisions. Modeling social influence acknowledges that
industries and markets are social structures in which organizations watch each other and use
signals from other organizations to guide their choices and actions (Kennedy, 2008). As noted
previously, under conditions of opacity, organizations face difficulties in observing whether,
and how substantively, other organizations adopt certain CSR practices (implying that indi-
vidual organizations are less affected by what their competitors do). Conversely, under condi-
tions of transparency, inter-organizational dynamics may have greater bearing on an organiza-
tion’s decision to adopt a given CSR practice. Our analysis demonstrates that social influence
expands the range of parameter combinations for which transitory opacity is optimal (see Ap-
pendix A). Second, we reran our analysis by generating both transparency and opacity matri-
ces randomly (i.e., we do not make any particular assumptions about the underlying transition
probabilities). The only assumption we make is that ceremonial adoption is less likely under a
29
regime of transparency than under one of opacity. Our analysis, based on one million pairs of
random matrices for opacity and transparency, suggests that the general pattern of findings is
robust to relaxing a vast majority of model assumptions (see Appendix B).
DISCUSSION
This paper reconciles the opposing viewpoints of transparentists and opacitists on how to
advance the substantive adoption and institutionalization of CSR in two main ways. First, we
resolve the tension by adopting a dynamic perspective that grasps both opacity and transpar-
ency not only as enduring but also as transitory phenomena. We theorize that changes in the
evaluation regime may occur, either exogenously due to regulatory measures or endogenously
because substantive adopters enforce transparency to avoid being misclassified as ceremonial
adopters and thereby to safeguard their legitimacy. We demonstrate that such regime switches
and the resulting transitory regime sequences can be conducive to the substantive adoption of
CSR. Second, we identify the boundary conditions under which enduring and transitory forms
of transparency and opacity maximize the share of substantive CSR adoption. We demon-
strate that the optimality of a regime sequence is determined by the relationship between
adoption and abandonment rates within and across transparency and opacity regimes. Overall,
our paper makes first steps in developing a systematic treatment of the interplay between so-
cial evaluation and practice adoption. In so doing, it opens up a novel and potentially im-
portant research area.
Implications for Research on CSR
A hallmark of the CSR literature has been the “transparency imperative” (Ringel,
2019)—that is, a widely endorsed and taken-for-granted assumption that the observability of
organizational actions is beneficial for substantive adoption (Christensen & Cheney, 2015;
Gold & Heikkurinen, 2018). Transparentists posit that, in the wake of public scrutiny and re-
quests for accountability, decoupling is not a viable option for adopters of CSR practices and
30
that, therefore, ceremonial adopters are inevitably pushed toward substantive adoption. It fol-
lows that the best way to institutionalize CSR is through promoting transparency, tightening
requirements, and, where necessary, disciplining ceremonial adopters.
However, while intuitively appealing, this logic ignores the fact that the very characteris-
tics of CSR can undermine the effectiveness of transparency. When implementation costs are
high and immediate substantive adoption of a CSR practice is unlikely (i.e., low adoption
rate) and adopters run the risk of limited chances to quit a once-established practice at a later
point in time (i.e., low abandonment rate), a regime of transparency can significantly limit the
number of organizations that engage with the practice, thus reducing the overall impact of
CSR.
For combinations of rather low substantive adoption and abandonment rates (i.e., condi-
tions that are highly relevant for the CSR context), our analyses reveal another, potentially
more effective, pathway to the institutionalization of CSR: transitory opacity (i.e., a phase of
initial opacity succeeded by a phase of transparency). Our analyses show that transitory opaci-
ty may outperform both enduring opacity and enduring transparency, depending on the moti-
vation underlying ceremonial adoption. More specifically, if ceremonial adoption is primarily
driven by an opportunistic motivation (i.e., an incentive to reap the benefits of CSR adoption
while minimizing its costs), then enduring transparency may be effective in leading to wide-
spread substantive adoption, as it shuts down the possibility of adopting a practice ceremoni-
ally. In contrast, if ceremonial adoption is primarily driven by a hypocritical motivation (i.e.,
an incentive to reap the benefits of CSR adoption without incurring any of its costs), then
transitory opacity can be effective in leading to widespread substantive adoption. The opti-
mality of transitory opacity is driven by the fact that organizations that adopt the practice cer-
emonially when possible (but would refrain from adoption when impossible under a regime of
transparency) will be forced into substantive adoption once the regime switches to transparen-
31
cy. It is therefore important to understand how organizations respond if their preferred option
of ceremonial adoption is not viable, and to theorize whether they would commit to substan-
tive adoption (“opportunists”) or would refrain from substantive adoption (“hypocrites”). As
such, transitory opacity constitutes an important but hitherto largely overlooked path for the
institutionalization of CSR.
These results shed new light on an important conversation in the field of CSR (see
Voegtlin & Pless, 2014): namely, whether the lack of implementation of CSR indicates a need
for stricter sanctions or whether decoupling should be tolerated and organizations provided
time and leeway to integrate CSR activities. For good reasons, scholars have questioned the
merit of low entry barriers and weak monitoring as means of achieving self-regulation in var-
ious industries (King & Lenox, 2000), given that the “lowest common denominator” (Sethi,
2002) tends to be conducive to opportunism and adverse selection. However, our model
demonstrates that the hypocritical adoption of CSR practices can have positive consequences
for their institutionalization. This is the case not only because decoupling in a single organiza-
tion can prove unstable (Tilcsik, 2010; Penttilä, 2020) but also because of the effects that a
change in the prevalent evaluation regime can have at the industry or institutional field level.
The bait-and-switch mechanism identified in our paper complements existing opacitist
perspectives, which highlighted that transparency may be detrimental to experimentation and
organizational learning (March, 1995; Weick, 1995), and may crowd out trust and intrinsic
motivation (Frey et al., 2013). Assumptions related to experimentation, learning and motiva-
tion, as well as to organizational-level dynamics of aspirational talk, goal displacement, and
identity change (e.g., Christensen et al., 2013) are reflected in our model in the coupling rate,
that is, the probability that an organization moves from ceremonial to substantive adoption.
While previous works from an opacitist view have focused primarily on coupling dynamics
and hence the “switch” part of the bait-and-switch mechanism (Christensen et al., 2013;
32
Haack et al., 2012), the analysis of the “bait” part, especially as the starting point of a dynam-
ic process comprising an exogenous or endogenous switch towards a regime of transparency,
has been largely neglected, thus far. Our model offers a crucial extension of extant opacitist
works by pointing out that achieving a large number and “critical mass” of ceremonial adop-
tions is a critical prerequisite to institutionalizing CSR across different domains and indus-
tries. Conversely, when enforcing transparency too early, only a few organizations will be
pulled into ceremonial adoption. Coupling dynamics will thus be limited to a small set of or-
ganizations and CSR will fail to gain prominence and exert enduring influence at a larger
scale.
Our proposed bait-and-switch model of CSR adoption also has significant practical im-
plications for NGOs and civil society activists. Specifically, given that initial opacity and non-
penalized decoupling can be a viable way to institutionalize CSR, NGOs and civil society
activists should not categorically criticize or punish organizations for ceremonial adoption.
Instead, they should allow for a stage of transformation during which ceremonial adoption can
spread, marked by an (exogenous or endogenous) switching point that leads to a regime
change and puts an end to widespread ceremonial adoption and decoupling. At the same time,
given that transitory opacity is only optimal for circumstances in which the probability of
practice abandonment is low, practitioners and activists should criticize substantive adopters
who diminish or slow down their implementation efforts. Through such social disapproval,
NGOs and civil society activists can send a strong signal that falling back on ceremonial
adoption or even non-adoption is unacceptable. In addition, since the type of adopter motiva-
tion (hypocrisy vs opportunism) reveals which regime sequence is optimal, NGOs and activ-
ists should try to gauge the share of hypocrites and opportunists in a given domain or industry,
for instance, through continuous dialogue and exchange with relevant actors (Palazzo &
Scherer, 2006) .
33
Implications for Institutional Theory
This paper offers important insights to institutional theory. Specifically, it contributes to a
better understanding of the relationship between decoupling (i.e., ceremonial adoption), diffu-
sion, and institutionalization, thereby complementing works that have integrated the topic of
practice diffusion with aspects of adaptation and variation in practice implementation (Ansari
et al., 2010; Lounsbury, 2001). Although decoupling, diffusion, and institutionalization con-
stitute foundational concepts in institutional theory (Greenwood, Oliver, Lawrence, & Meyer,
2017), their precise relationship and the character of their interdependence has not been dis-
cussed in much detail. Extant research tends to equate the diffusion of a practice with its insti-
tutionalization or to treat diffusion as a result of successful institutionalization (Schneiberg &
Clemens, 2006; Zeitz et al., 1999). Colyvas and Jonsson (2011), however, proposed disentan-
gling diffusion from institutionalization and argued that diffusion can function as an anteced-
ent of institutionalization. This view is consistent with our model, which indicates that prac-
tice diffusion, even if ceremonial in character, can facilitate the institutionalization (i.e., wide-
spread substantive adoption) of CSR practices due to an exogenous or endogenous switch in
evaluation regimes. Given that, under an evaluation regime of opacity, decoupling can be
beneficial to the diffusion of instances of ceremonial adoption, which then trigger a switch
from opacity to transparency that enforces substantive adoption, decoupling can be said to be
conducive to institutionalization. In this view, decoupling has important institutional conse-
quences and can be conceptualized as an endogenous source of institutional change (Boxen-
baum & Jonsson, 2017).
Another important implication of our paper is that institutional scholars need to theorize
and systematically study the evaluation-decoupling nexus. Previous works on decoupling fol-
lowed the arguments of Meyer and Rowan (1977) and assumed that external observers do not
monitor organizational activities. In that view, ceremonial behavior is not criticized or con-
34
demned but actually desired and, thus, can be said to be equivalent to institutionalized behav-
ior (Zajac & Westphal, 2004). This assumption, however, is at odds with a phenomenological
understanding of institutions as tightly coupled systems that are consolidated through contin-
uous social interaction (Berger & Luckmann, 1967). Indeed, the assumption that “myth and
ceremony” are equivalent to institutionalized behavior has ignored the growing trend towards
an “audit society” (Power, 2007), which reveals that external observers lack “confidence and
good faith” in the symbolic actions of organizations. External observers instead seek to in-
spect the type of practice adoption and enforce transparency and social control, meaning that
decoupling poses a significant reputational risk for organizations (Crilly et al., 2012; Marquis
& Qian, 2014). It follows that ceremonial behavior cannot be equated with institutionalized
behavior. This proposition seems to particularly apply to the context of CSR, where employ-
ees search for purpose and meaningfulness through their work and are not willing to tolerate
moral inconsistencies (Tolbert & Zucker, 1996). Moreover, our analyses reveal that ceremo-
nial behavior can be based on hypocritical and opportunist motives, a distinction which ac-
counts for profound differences in institutional effects (i.e., with respect to maximizing the
share of substantive adoption). Therefore, in order to better understand institutional outcomes,
institutional theorists need to conceptualize the micro-level diversity in motivations that often
gets ignored in view of the pervasive homogeneity of behavior at more macro levels (Tolbert
& Darabi, 2020).
Implications for Methodology
Crane, Henriques, and Husted (2018) identified considerable methodological challenges
in the CSR literature that hamper its development. They also expounded the problems of a
significant bias toward a relatively small number of methods of data collection and analysis.
Morsing and Spence (2019), in turn, demanded greater sophistication in the analysis of the
links between theoretical and practice-based aspects of CSR research. In reviewing the criti-
35
cism of CSR research, Wang, Gibson, and Zander (2020) discussed the frequent perception
that it suffers from weak theoretical foundations and lack of theory development, rarely pay-
ing attention to underlying mechanisms and boundary conditions, and consequently fails to
inform practice adequately. Embracing methodological pluralism and method triangulation
can speak to these challenges. While formal modeling and simulation approaches have gained
currency in institutional theory (e.g., Colyvas & Maroulis, 2015) and CSR research (e.g., Kaul
& Luo, 2018), to our knowledge they have not been employed in the more specific context of
CSR practice adoption. This underutilization is unfortunate, since formal modeling and simu-
lation approaches allow researchers to gain a thorough understanding of the role of social
evaluation in decoupling and practice adoption. The quality of practice implementation at the
industry level and the evolution of practice adoption over time are naturally difficult to exam-
ine empirically by means of qualitative research, which tends to be more suitable for in-depth
study of adoption types and degrees of practice implementation in a single firm. In that re-
gard, our model allows for transcending paradigmatic boundaries by showing that some of the
considerations of interpretivist and qualitative research can be translated into formalized lan-
guage, thereby sharpening the conceptual clarity and specificity of extant CSR research.
36
APPENDIXES
Appendix A: Social Influence
Under an evaluation regime of opacity, organizations face difficulties in observing whether,
and how substantively, other organizations adopt certain CSR practices, implying that indi-
vidual organizations are less affected by what their competitors do. Conversely, under a re-
gime of transparency, inter-organizational dynamics may have greater bearing on an organiza-
tion’s decision to adopt certain CSR practices. We therefore examined how social influence
may affect the share of substantive adopters for the four different regime sequences (T/T,
O/O, T/O, and O/T). Arguably, actors can be influenced by other actors in a myriad of ways.
For illustrative purposes, we report the outcomes for one particular form of social influence:
the adoption rate (), abandonment rate (), and coupling rate () are mod-
eled as depending on the share of organizations that are adopters (i.e., (3)) or non-adopters
(i.e., (1)) operating in the same domain or industry. Specifically, we assume that, in each
period, the transition probabilities for transparency T' are adjusted, with
=
(3)+(1), with social as a weighting factor for the social influence (here
defined as the share of organizations that have already adopted the practice substantively). If,
for example, the social influence is 0.0, then the share of existing adopters does not affect the
adoption rate. If the social influence is 0.5, the share adjusted adoption rate is
= 0.5
(3) + 0.5 . Thus, if the share of substantive adopters (3) is lower than the adop-
tion rate (), the adjusted rate of adoption (
) is lower than the actual rate
(). In contrast, if the share of substantive adopters is larger, then so is the adjusted rate.
Social influence may thus decrease or increase the adoption rate, depending on the share of
substantive adopters (relative to the unadjusted adoption rate). A similar logic applies to the
abandonment rate, which is given by
=(1)+(1), and
the coupling rate, which is given by
=()
()()+(1). In
37
Figure A1, we report the results for social = 0.1 (left panel) and social = 0.5 (right panel).
Overall, the general pattern of results remains largely unaffected; however, within the subset
of valid parameter combinations, the prominence of transitory opacity (O/T) increases.
Figure A1: The Role of Social Influence
Appendix B: Relaxing Assumptions Underlying the Opacity Matrix
Our main analyses draw on a particular transition matrix for opacity (see above our elabora-
tion in the section ‘model specification’). In this analysis, we relax the assumptions that in-
formed these parameters and generate, in a first step, a random transition matrix for the re-
gime of opacity. Such a random matrix may look as follows:
=0.07 0.24 0.69
0.62 0.08 0.30
0.12 0.29 0.59
In a next step, we generate a random transition matrix for the regime of transparency. In so
doing, we ensure that the transition matrix is valid by assuming
+; that is, the (substantive) adoption rate under transparency is located in a
range between the substantive adoption rate under opacity and the combined (both substantive
38
and ceremonial) adoption rate under opacity. Thus, for the random transition matrix of opacity
described above, the adoption rate under transparency can be 0.69 0.69 +
0.24 = 0.93. We then randomly pick one value in this range (e.g., 0.77). For the abandonment
rate, we follow the same procedure and pick a random value (e.g., 0.34) from
+, or 0.12 0.12 + 0.29 = 0.41. Finally, we pick a ran-
dom coupling rate within the range of 0 to 1 (e.g., 0.19). The resulting transition ma-
trix for transparency looks like this:
=0.23 0.00 0.77
0.81 0.00 0.19
0.34 0.00 0.66
We make only one assumption in our analysis—namely, that ceremonial adoption is less like-
ly under a regime of transparency than under one of opacity. We then calculate the optimal
regime sequence—that is, the sequence that maximizes the share of substantive adopters aver-
aged over t = 1 to 25. We repeat the process of (1) generating random matrices O and T and
(2) determining the optimal regime sequence one million times. To make results comparable,
we normalize the adoption and abandonment rates under transparency to a range between 0
and 1. In our example above, a value of = 0.77 is at the lower end of the range of
[0.69, 0.93] and translates to a normalized value of ..
. = 0.33. We normalize the aban-
donment rate in the same way: ..
. = 0.76. In Figure A2, we report the probability that
any of the four evaluation regime sequences is optimal for different normalized rates of adop-
tion (x-axis) and abandonment (y-axis). These normalized rates equal the set of valid combi-
nations, highlighted through a white box in the main analyses reported in the manuscript. The
color of each point in these panels reflects the probability that the corresponding evaluation
regime is optimal, with lighter colors indicating lower probabilities and darker colors indicat-
ing higher probabilities.
39
Figure A2: Optimality for Normalized Rates
Consistent with our manuscript’s findings, enduring transparency (T/T, top left panel) is the
optimal regime for many combinations of adoption and abandonment rates. However, if the
normalized adoption rate is rather low (i.e., the primary motivation for ceremonial adoption is
hypocrisy rather than opportunism), then transitory opacity (O/T) or enduring opacity (O/O)
become more likely to be the optimal regime sequence (a finding that is also consistent with
the manuscript’s findings). Also consistent with our arguments, transitory opacity (T/O) is
very unlikely to be optimal (within the subset of valid combinations of adoption and aban-
donment rates). In sum, given the complete randomness of both transition matrices, the results
are obviously less clear-cut than those of the manuscript’s analyses. Nonetheless, even if we
relax most of the assumptions about both transition matrices, our key findings still hold.
40
REFERENCES
Abbott, A. 1990. A primer on sequence methods. Organization Science, 1: 375-392.
Adner, R., Pólos, L., Ryall, M. D., & Sorenson, O. 2009. The case for formal theory. Academy
of Management Review, 34: 201-208.
Aguinis, H., & Glavas, A. 2012. What we know and don’t know about corporate social re-
sponsibility: A review and research agenda. Journal of Management, 38: 932-968.
Ansari, S. M., Fiss, P. C., & Zajac, E. J. 2010. Made to fit: How practices vary as they diffuse.
Academy of Management Review, 35: 67-92.
Aravind, D., & Christmann, P. 2011. Decoupling of standard implementation from certifica-
tion: Does quality of ISO 14001 implementation affect facilities' environmental perfor-
mance? Business Ethics Quarterly, 21: 73-102.
Baccaro, L., & Mele, V. 2011. For lack of anything better? International organizations and
global corporate codes. Public Administration, 89: 451-470.
Barnett, M. L., & King, A. A. 2008. Good fences make good neighbors: A longitudinal analy-
sis of an industry self-regulatory institution. Academy of Management Journal, 51: 1150-
1170.
Baumann-Pauly, D., Wickert, C., Spence, L., & Scherer, A. G. 2013. Organizing corporate
social responsibility in small and large firms: Size matters. Journal of Business Ethics,
115: 693-705.
Behnam, M., & MacLean, T. L. 2011. Where is the accountability in international accounta-
bility standards? A decoupling perspective. Business Ethics Quarterly, 21: 45-72.
Berger, P. L., & Luckmann, T. 1967. The social construction of reality: A treatise in the soci-
ology of knowledge. New York: Anchor.
Berger-Walliser, G., & Scott, I. 2018. Redefining corporate social responsibility in an era of
globalization and regulatory hardening. American Business Law Journal, 55: 167-218.
41
Berne Declaration, 2019. July 9: https://www.publiceye.ch/en/news/detail/a-pioneering-
analysis-of-swiss-agricultural-traders-responses-from-firms-lack-substance.
Bernstein, E. S. 2012. The transparency paradox: A role for privacy in organizational learning
and operational control. Administrative Science Quarterly, 57: 181-216.
Bernstein, E. S. 2017. Making transparency transparent: The evolution of observation in man-
agement theory. Academy of Management Annals, 11: 217-266.
Bitektine, A., & Haack, P. 2015. The macro and the micro of legitimacy: Towards a multi-
level theory of the legitimacy process. Academy of Management Review, 40: 49-75.
Boiral, O. 2007. Corporate greening through ISO 14001: A rational myth? Organization Sci-
ence, 18: 127-146.
Boxenbaum, E., & Jonsson, S. 2017. Isomorphism, diffusion, and decoupling: Concept
evolution and theoretical challenges In R. Greenwood, T. B. Lawrence, C. Oliver, & R.
Meyer (Eds.), The Sage Handbook of Organizational Institutionalism: 59-113. London:
Sage.
Bromley, P., & Powell, W. W. 2012. From smoke and mirrors to walking the talk: Decoupling
in the contemporary world. Academy of Management Annals, 6: 483-530.
Burton, R. M., & Obel, B. 2011. Computational modeling for what-is, what-might-be, and
what-should-be studies—and triangulation. Organization Science, 22: 1195-1202.
Carnahan, S., Kryscynski, D., & Olson, D. 2017. When does corporate social responsibility
reduce employee turnover? Evidence from attorneys before and after 9/11. Academy of
Management Journal, 60: 1932-1962.
Cashore, B. 2002. Legitimization and the privatization of environmental governance: How
non-state market-driven (NSMD) governance systems gain rule-making authority. Gov-
ernance: An International Journal of Policy, Administration, and Institutions, 15: 503-
529.
42
Chatterji, A. M., & Toffel, M. 2010. How firms respond to being rated. Strategic Manage-
ment Journal, 31: 917-945.
Chen, Y-C., Hung, M., & Wang, Y. 2018. The effect of mandatory CSR disclosure on firm
profitability and social externalities: Evidence from China. Journal of Accounting and
Economics, 65: 169-190.
Christensen, L. T., & Cheney, G. 2015. Peering into transparency: Challenging ideals, proxies,
and organizational practices. Communication Theory, 25: 70-90.
Christensen, L. T., Morsing, M., & Thyssen, O. 2013. CSR as aspirational talk. Organization,
20: 372-393.
Christensen, L. T., Morsing, M., & Thyssen, O. 2020. Talk-action dynamics: Modalities of
aspirational talk. Organization Studies, forthcoming.
Christmann, P., & Taylor, G. 2006. Firm self-regulation through international certifiable
standards: Determinants of symbolic versus substantive implementation. Journal of
International Business Studies, 37: 863-878.
Colyvas, J. A., & Jonsson, S. 2011. Ubiquity and legitimacy: Disentangling diffusion and in-
stitutionalization. Sociological Theory, 29: 27-53.
Colyvas, J. A., & Maroulis, S. 2015. Moving from an exception to a rule: Analyzing mecha-
nisms in emergence-based institutionalization. Organization Science, 26: 601-621.
Crane, A., Henriques, I., & Husted, B. W. 2018. Quants and poets: Advancing methods and
methodologies in business and society research. Business & Society, 57: 3-25.
Crilly, D., Zollo, M., & Hansen, M. 2012. Faking it or muddling through? Understanding de-
coupling in response to stakeholder pressures. Academy of Management Journal, 55:
1429-1448.
Cuypers, I., Koh, P-S., & Wang, H. 2016. Sincerity in corporate philanthropy, stakeholder
perceptions and firm value. Organization Science, 27: 173-188.
43
Davis, J. P., Eisenhardt, K. M., & Bingham, C. B. 2007. Developing theory through simula-
tion methods. Academy of Management Review, 32: 580-599.
Davis, G. F., & Marquis, C. 2005. Prospects for organization theory in the early twenty-first
century: Institutional fields and mechanisms. Organization Science, 16: 332-343.
de Bakker, F. G. A., Rasche, A., & Ponte, S. 2019. Multi-stakeholder initiatives on sustainabil-
ity: A cross-disciplinary review and research agenda for business ethics. Business Ethics
Quarterly, 29: 343-383.
Delmas, M., & Burbano, V. C. 2011. The drivers of greenwashing. California Management
Review, 54: 64-87.
Delmas, M., & Montes-Sancho, M. J. 2010. Voluntary agreements to improve environmental
quality: Symbolic and substantive cooperation. Strategic Management Journal, 31: 575-
601.
Delmas, M., & Montes-Sancho, M. J. 2011. An institutional perspective on the diffusion of
international management system standards: The case of the environmental management
standard ISO 14001. Business Ethics Quarterly, 21: 103-132.
den Hond, F., & de Bakker, F. G. A. 2007. Ideologically motivated activism: How activist
groups influence corporate social change activities. Academy of Management Review, 32:
901-924.
Desai, V. M. 2011. Mass media and massive failures: Determining organizational efforts to
defend field legitimacy following crises. Academy of Management Journal, 54: 263–-
278.
Diestre, L., & Rajagopalan, N. 2014. Toward an input-based perspective on categorization:
Investor reactions to chemical accidents. Academy of Management Journal, 57: 1130-
1153.
Durand, R., Hawn, O., & Ioannou, I. 2019. Willing and able: A general model of organiza-
44
tional responses to normative pressures. Academy of Management Review, 44: 299-230.
European Commission. 2014. Press release 15 April. http://europa.eu/rapid/press-
release_STATEMENT-14-124_de.htm.
Fabrizio, K. R., & Kim, E-H. 2019. Reluctant disclosure and transparency: Evidence from
environmental disclosures. Organization Science, 30: 1207-1231.
Ferraro, F., Etzion, D., & Gehman, J. 2015. Tackling grand challenges pragmatically: Robust
action revisited. Organization Studies, 36: 363-390.
Frey, B. S., Homberg, F., & Osterloh, M. 2013. Organizational control systems and pay for
performance in the public service. Organization Studies, 34: 949-972.
Gaba, V., & Dokko, G. 2016. Learning to let go: Social influence, learning, and the abandon-
ment of corporate venture capital practices. Strategic Management Journal, 37: 1558-
1577.
Gilbert, D. U., & Behnam, M. 2013. Trust and the United Nations Global Compact: A net-
work theory perspective. Business & Society, 51: 135-169.
Gilbert, D. U., Rasche, A., & Waddock, S. 2011. Accountability in a global economy: The
emergence of international accountability standards. Business Ethics Quarterly, 21: 23-44.
Girschik, V. 2020. Shared responsibility for societal problems: The role of internal activists in
reframing corporate responsibility. Business & Society, 59: 34-66.
Gold, S., & Heikkurinen, P. 2018. Transparency fallacy: Unintended consequences of stake-
holder claims on responsibility in supply chains. Accounting, Auditing & Accountability
Journal, 31: 318-337.
Greenwald, A. G. 2012. There is nothing so theoretical as a good method. Perspectives on
Psychological Science, 7: 99-108.
Greenwood, R., Oliver, C., Lawrence, T. B., & Meyer, R. 2017. Introduction: Into the fourth
decade. In R. Greenwood, T. B. Lawrence, C. Oliver, & R. Meyer (Eds.), The Sage Hand-
45
book of Organizational Institutionalism: 1-23. London: Sage.
Haack, P., & Schoeneborn, D. 2015. Is decoupling becoming decoupled from institutional
theory? A commentary on Wijen. Academy of Management Review, 40: 307-310.
Haack, P., Schoeneborn, D., & Wickert, C. 2012. Talking the talk, moral entrapment, creeping
commitment? Exploring narrative dynamics in corporate responsibility standardization.
Organization Studies, 33: 815-845.
Hallett, T. 2010. The myth incarnate: Recoupling processes, turmoil, and inhabited institutions
in an urban elementary school. American Sociological Review, 75: 52-74.
Hannan, M. T., & Freeman, J. 1984. Structural inertia and organizational change. American
Sociological Review, 49: 149-164.
Harrison, J. R., Lin, Z., Carroll, G. R., & Carley, K. M. 2007. Simulation modeling in organi-
zational and management research. Academy of Management Review, 32: 1229-1245.
Hoffman, A. 2001. Linking organizational and field level analyses: The diffusion of corporate
environmental practice. Organization & Environment, 14: 133-156.
Humphreys, M., & Brown, A. D. 2008. An analysis of corporate social responsibility at credit
line: A narrative approach. Journal of Business Ethics, 80: 403-418.
Jonsson, S., Greve, H., & Fujiwara-Greve, T. 2009. Undeserved loss: The spread of legitima-
cy loss to innocent organizations in response to reported corporate deviance. Administra-
tive Science Quarterly, 54: 195-228.
Kaul, A., & Luo, J. 2018. An economic case for CSR: The comparative efficiency of for‐profit
firms in meeting consumer demand for social goods. Strategic Management Journal, 39:
1650-1677.
Kennedy, M. T. 2008. Getting counted: Markets, media, and reality. American Sociological
Review, 73: 1021-1021.
King, A. A., & Lenox, M. J. 2000. Industry self-regulation without sanctions: The chemical
46
industry’s responsible care program. Academy of Management Journal, 43: 698-716.
Kohl, K. 2016. Becoming a sustainable organization: A project and portfolio management
approach. Boca Raton, FL: CRC Press.
Lange, D., & Washburn, N. 2012. Understanding attribution of corporate irresponsibility.
Academy of Management Review, 37: 300-326.
Lounsbury, M. 2001. Institutional sources of practice variation: Staffing college and universi-
ty recycling programs. Administrative Science Quarterly, 46: 29-56.
MacLean, T. L., & Behnam, M. 2010. The dangers of decoupling: The relationship between
compliance programs, legitimacy perceptions, and institutionalized misconduct. Academy
of Management Journal, 53: 1499-1520.
March, J. C. 1989. Decisions and organizations. Oxford, UK: Blackwell.
March, J. C. 1995. A primer on decision making: How decisions happen. New York: Free
Press.
Marquis, C., & Qian, C. 2014. Corporate social responsibility reporting in China: Symbol or
substance? Organization Science, 25: 127-148.
Marquis, C. W., Toffel, M. W., & Zhou, Y. 2016. Scrutiny, norms, and selective disclosure: A
global study of greenwashing. Organization Science, 27: 483-504.
McWilliams, A., & Siegel, D. 2000. Corporate social responsibility and financial perfor-
mance: correlation or misspecification? Strategic Management Journal, 21: 603-609.
Meyer, J. W., & Rowan, B. 1977. Institutionalized organizations: Formal structure as myth
and ceremony. American Journal of Sociology, 83: 340-363.
Morsing, M., & Spence, L. J. 2019. Corporate social responsibility (CSR) communication and
small and medium sized enterprises: The governmentality dilemma of explicit and implicit
CSR communication. Human Relations, 72: 1920–1947..
Overdevest, C. 2010. Comparing forest certification schemes: The case of ratcheting stand-
47
ards in the forest sector. Socio-Economic Review, 8: 47-76.
Palazzo, G., & Scherer, A. G. 2006. Corporate legitimacy as deliberation: A communicative
framework. Journal of Business Ethics, 66: 71-88.
Pentland, B. T., Hærem, T., & Hillison, D. 2010. Comparing organizational routines as recur-
rent patterns of action. Organization Studies, 31: 917-940.
Penttilä, V. 2020. Aspirational talk in strategy texts: A longitudinal case study of strategic epi-
sodes in corporate social responsibility communication. Business & Society, 59: 67-97.
Pilaj, H. 2017. The choice architecture of sustainable and responsible investment: Nudging
investors toward ethical decision-making. Journal of Business Ethics, 140: 743-753.
Poole, M. S., & Van de Ven, A. H. 1989. Using paradox to build organization and manage-
ment theories. Academy Management Review, 15: 562-578.
Potoski, M., & Prakash, A. 2013. Green clubs: Collective action and voluntary environmental
programs. Annual Review of Political Science, 16: 399-419.
Power, M. 2007. Organized uncertainty: Designing a world of risk management. Oxford,
UK: Oxford University Press.
Rasche, A., & Waddock, S. 2017. Standards for CSR: Legitimacy, impact and critique. In: A.
Rasche, M. Morsing, & J. Moon (Eds.), Corporate Social Responsibility: Strategy, com-
munication, governance: 163-186. Cambridge, UK: Cambridge University Press.
Repenning, N. P. 2002. A simulation-based approach to understanding the dynamics of inno-
vation implementation. Organization Science, 13: 109-127.
Ringel, L. 2019. Unpacking the transparency-secrecy nexus. Frontstage and backstage behav-
ior in a political party. Organization Studies, 40: 705-723.
Risse, T. 2000. ‘Let’s argue!’ Communicative action in world politics. International Organi-
zation, 54: 1-39.
Schneiberg, M., & Clemens, E. S. 2006. The typical tools for the job: Research strategies in
48
institutional analysis. Sociological Theory, 24: 195-227.
Schoeneborn, D., Morsing, M., & Crane, A. 2020. Formative perspectives on the relation be-
tween CSR communication and CSR practices: Pathways for walking, talking, and
t(w)alking. Business & Society, 59: 5-33.
Sethi, S. P. 2002. Standards for corporate conduct in the international arena: Challenges and
opportunities for multinational corporations. Business and Society Review, 107: 20-40.
Sethi, S. P., & Schepers, D. H. 2014. United Nations Global Compact: The promise-
performance gap. Journal of Business Ethics, 122, 193-208.
Suddaby, R. 2010. Editor’s comment: Construct clarity in theories of management and organ-
ization. Academy of Management Review, 35: 346-357.
Tenbrunsel, A. E., & Messick, D. M. 1999. Sanctioning systems, decision frames, and cooper-
ation. Administrative Science Quarterly, 44: 684-707.
Tilcsik, A. 2010. From ritual to reality: Demography, ideology, and decoupling in a post-
communist government agency. Academy of Management Journal, 53: 1474-1498.
Tolbert, P. S., & Darabi, T. 2020. Bases of conformity and institutional theory: Understanding
organizational decision-making. In P. Haack, J. Sieweke, & L. Wessel (Eds.), Microfoun-
dations of Institutions (RSO, Vol. 65A), 269-290. Bingley, UK: Emerald.
Tolbert, P. S., & Zucker, L. G. 1996. The institutionalization of institutional theory. In S. R.
Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of Organization Studies: 175-190.
London: Sage.
Vo e g tlin, C. & Pless, N. 2014. Global governance: CSR and the role of The UN Global Com-
pact. Journal of Business Ethics, 122: 179-191.
Wang, H., Gibson, C., & Zander, U. 2020. Editors’ comments: Is research on corporate social
responsibility undertheorized? Academy of Management Review, 45: 1-6.
Wang, H., Tong, L., Takeuchi, R., & George, G. 2016. Corporate Social Responsibility: An
49
overview and new research directions. Academy of Management Journal, 59: 534-544.
Weaver, G. R., Treviño, L. K., & Cochran, P. L. 1999. Integrated and decoupled corporate
social performance: Management values, external pressures, and corporate ethics practices.
Academy of Management Journal, 42: 539-552.
Weick, K. E. 1995. Sensemaking in Organizations. Thousand Oaks, CA: Sage.
Wickert, C., Scherer, A. G., & Spence, L. 2016. Walking and talking corporate social respon-
sibility: Implications of firm size and organizational cost. Journal of Management Stud-
ies, 53: 1169-1196.
Wijen, F. 2014. Means versus ends in opaque institutional fields: Trading off compliance and
achievement in sustainability standard adoption. Academy of Management Review, 39:
302-323.
Woody, K. E. 2012. Conflict minerals legislation: The SEC’s new role as diplomatic and hu-
manitarian watchdog. Fordham Law Review, 81: 1315-1351.
Younkin, P. 2016. Complicating abandonment: How a multi-stage theory of abandonment
clarifies the evolution of an adopted practice. Organization Studies, 37: 1017-1053.
Yu, T., Sengul, M., & Lester, R. H. 2008. Misery loves company: The spread of negative im-
pacts resulting from organizational crisis. Academy of Management Review, 33: 452-472.
Zadek, S. 2004. The path to corporate responsibility. Harvard Business Review, 82: 125-132.
Zajac, E., & Westphal, J. D. 2004. The social construction of market value: Institutionaliza-
tion and learning perspectives on stock market reactions. American Sociology Review, 69:
433-457.
Zeitz, G., Mittal, V., & McAulay, B. 1999. Distinguishing adoption and entrenchment of
management practices: A framework for analysis. Organization Studies, 20: 741-776.
Zeyen, A., Beckmann, M., & Wolters, S. 2016. Actor and institutional dynamics in the devel-
opment of multi-stakeholder initiatives. Journal of Business Ethics, 135: 341-360.
50
Zhang, T., Gino, F., & Margolis, J. D. 2018. Does “could” lead to good? On the road to moral
insight. Academy of Management Journal, 61: 857-895.
51
TABLES AND FIGURES
Tabl e 1: Comparison of the Transparentist and Opacitist Perspectives
Transparentists
Opacitists
Key assump-
tions
Transparency nurtures sub-
stantive adoption.
Opacity nurtures substantive
adoption.
Assessment of
ceremonial
adoption
Unfavorable, because it im-
pedes substantive adoption.
Favorable, because it can further
substantive adoption.
Practical
implications
Enforce substantive adoption
by raising requirements and
imposing sanctions.
Promote substantive adoption by
providing time and leeway for
exploration/experimentation.
Key metaphor
Walk the talk.
Talk the walk.
Disciplines and
exemplary
publications
Institutional theory: Boiral
(2007), Meyer & Rowan
(1977), Wijen (2014).
Business ethics research: Ara-
vind & Christmann (2011),
Behnam & MacLean (2011).
Learning and social control:
Bernstein (2012), Tenbrunsel &
Messick (1999).
Organizational communication:
Christensen et al. (2013), Haack
et al. (2012).
Tabl e 2: Possible Regime Sequences
Phase 1 (t=1..R)
Phase 2
(t=R+1..T)
Literature
Enduring opacity
(O/O) Opacity Opacity
Opacitists: e.g., Bernstein
(2012)
Enduring trans-
parency (T/T) Transparency Transparency
Transparentists: e.g., Mac-
Lean & Behnam (2010)
Transitory
opacity (O/T) Opacity Transparency Largely unexplored
Transitory
transparency
(T/O)
Transparency Opacity Largely unexplored
52
Figure 1: The Markov Chain Model
Figure 2: Optimal Regime Sequences
53
Figure 3: Boundary Conditions (Coupling Rate =.)
Figure 4: Motivation Underlying Ceremonial Adoption (Coupling Rate =.)
54
Patrick Haack is Professor of Strategy at HEC Lausanne, University of Lausanne. His cur-
rent research focuses on social evaluations, practice adoption, and the application of experi-
ments to the study of institutionalization and legitimation. He currently serves on the editorial
boards of the Academy of Management Review, Business & Society, Journal of Management
Studies, and Organization Studies.
Dirk Martignoni is an assistant professor of strategy at Università della Svizzera italiana
(USI), Lugano. He received his PhD from the University of St. Gallen, Switzerland. His re-
search focuses on organizational learning and adaptation.
Dennis Schoeneborn is Professor of Organization, Communication, and CSR at Copenhagen
Business School as well as Visiting Professor of Organization Studies at Leuphana University
of Lüneburg. In his current research, he focuses on the constitutive role of communication for
phenomena of organization, organizing, and organizationality. He is editorial board member
of Management Communication Quarterly and Organization Studies. Furthermore, he serves
as Associate Editor at Business & Society.
