QUESTION
‘Citibankers’ at Citigroup: A Study of the Loss of
Institutional Trust after a Merger
Steve Maguire and Nelson Phillips
McGill University, Montreal; Imperial College London
abstract In this paper, we present the results of a study of the loss of institutional trust
following a merger. Specifically, we focus on how issues of organizational identity and
identification processes contributed to the loss of institutional trust among a group of
employees of Citigroup after its creation through the merger of Citicorp and Travelers. Our
study makes two important contributions. First, we propose and demonstrate empirically that
institutional trust, like interpersonal trust, can be identity-based. Second, adopting a narrative
approach to organizational identity, we explore institutional trust in a post-merger context,
highlighting how institutional trust is initially undermined after a merger by the ambiguity
of the new organization’s identity; and how later, once the identity of the new organization
becomes less ambiguous, institutional trust can continue to be undermined by the absence of
employees’ identification with the new organization, especially among those who were
highly-identified with their legacy organizations.
INTRODUCTION
In this paper, we present the results of an empirical study of the loss of institutional trust
– an individual’s expectation that some organized system will act with predictability and
goodwill (Costigan et al., 1998; Rousseau et al., 1998; Shapiro, 1987; Zucker, 1986). We
show that institutional trust, like interpersonal trust (i.e. trust between individuals), can be
identity-based, and we explore the implications of this in the context of a merger.
Specifically, our study examines the merger of Citicorp and Travelers to form Citigroup.
Adopting a narrative approach to organizational identity (e.g. Brown, 2006), we find
that, immediately following a merger, institutional trust – the trust that members have in
their organization – suffers because of the ambiguity surrounding the identity of the
newly merged organization. Divergent views of those aspects of the new organization’s
nature that are central, enduring and distinctive in the stories told about the organization
mean that its actions are experienced as unpredictable, which undermines institutional
trust. Later, as ambiguity is reduced and new post-merger organizational identity
Address for reprints: Steve Maguire, Desautels Faculty of Management, McGill University, 1001 Sherbrooke
Street West, Montreal, Quebec H3A 1G5, Canada (steve.maguire@mcgill.ca).
© Blackwell Publishing Ltd 2008. Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
Loss of Institutional Trust 373
narratives become more convergent, some employees – those who were highly identified
with their legacy organization – can continue to experience low institutional trust
because they do not identify with – i.e. develop a psychological bond with (Benkhoff,
1997) – the new organization in which they find themselves.
Our study makes several contributions to our understanding of the link between trust
and identity, as well as to the literature on mergers and acquisitions. First, despite calls
for a ‘shift in research attention from trust in dyadic relations to trust in aggregate social
systems’ (McEvily et al., 2003a, p. 100), there are still few empirical studies of institutional
trust in the literature. Our study provides some initial insight into the nature of this
important form of trust as well as mechanisms through which it can be lost following a
merger. Second, we link identity to institutional trust and provide a framework for
understanding the role of organizational identity and individual identification processes
in its creation, maintenance and loss, pointing to an important area for future research.
This is a significant contribution because, unlike the literature on interpersonal trust
where several typologies have been presented, different possible bases upon which
institutional trust can develop remain unidentified and unexplored. Third, we connect
the literature on institutional trust (e.g. Bachmann, 2001; Costigan et al., 1998; Zucker,
1986) to discussions of trust in the mergers and acquisitions literature (e.g. BijlsmaFrankema,
2001; Nikandrou et al., 2002; Searle and Ball, 2004; Stahl and Sitkin, 2005;
Stahl et al., 2003). While the mergers and acquisitions literature has a well developed
discussion of trust, the concept of institutional trust and its difference from interpersonal
trust have not yet been brought to the fore.
We present our study in four steps. First, we provide an overview of the literature on
institutional trust and link it to related discussions in the mergers and acquisitions
literature. We then present our research methods and outline our case study: the merger
of Citicorp and Travelers. Next, we present our analysis and findings on institutional
trust after a merger. We conclude with a discussion of theoretical implications and
possible future research.
INSTITUTIONAL TRUST
We view trust as an ‘attitude or expectancy about other people and the social systems in
which they are embedded’ (Kramer, 1999, p. 575).
[1]
It is a set of ‘socially learned and
socially confirmed expectations that people have of each other, of the organizations and
institutions in which they live, and of the natural or moral social orders that set the
fundamental understandings of their lives’ (Barber, 1983, p. 164). It is also something
that is easily destroyed, and researchers have called for more research on how and why
this happens: trust is ‘a fragile entity yet few attend to how it unravels’ (Searle and Ball,
2004, p. 711).
The broader literature on trust is large, growing rapidly, and characterized by a
diversity of approaches, levels of analysis, and definitions of trust (e.g. Bachmann et al.,
2001; Dirks and Ferrin, 2001; Hosmer, 1995; Kramer, 1999; Lane and Bachmann,
1998; Mayer et al., 1995; McEvily et al., 2003b; Mollering et al., 2004; Rousseau et al.,
1998). Zucker (1986, p. 57) identifies two broad strands of trust research: ‘one asserts that
trust resides in actors’ assumptions that others in an exchange will put self-interest aside
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips374
in favour of “other-orientation” or “collectivity-orientation” ’; while ‘the other major
perspective asserts that trust resides in actors’ expectations of “things as usual” ’. With the
latter, trust rests solely on predictability, while with the former, benevolence (Schoorman
et al., 2007) or goodwill is also required, making it a stronger form of trust (Hardy et al.,
1998). In this paper, we build on the former strand of research to address trust in its
strong form and thus define trust as the expectation that some other will act with predictability and
benevolence.
[2]
Nature and Sources of Institutional Trust
Different types and sources of trust have been theorized. One important distinction stems
from the nature of the trustee: ‘one can have trust in particular people ( personal trust)
and trust in organized systems (institutional trust)’ (Lewicki and Benedict-Bunker, 1995,
p. 136). Our interest is in the latter, institutional trust, which ‘develops when individuals
must generalize their personal trust to large organizations made up of individuals with
whom they have low familiarity, low interdependence and low continuity of interaction’
(Lewicki and Benedict-Bunker, 1995, p. 137). Institutional trust can exist at the organizational
(Costigan et al., 1998; Rousseau et al., 1998), interorganizational (Zucker, 1986)
or societal level (Bachmann, 2001). It is also called ‘institutional-based trust’ (Zucker,
1986), ‘impersonal trust’ (Shapiro, 1987), and ‘system trust’ (Bachmann, 2001).
There is a growing consensus that institutional trust is critical for organizations to
retain valuable human resources. For example, Costigan et al. (1998) found support for
the hypotheses that employees’ trust in their organization was positively related to the
perceived effectiveness of the organization’s reward system, and negatively related to
their desire and intent to leave the organization. Other work suggests that institutional
trust establishes a context conducive to more trusting interpersonal relationships:
‘institution-based trust can ease the way to formulating both calculus-based and relational
trust’ (Rousseau et al., 1998, p. 400).
Despite growing recognition of its importance, organizational researchers have
devoted much less attention to institutional as compared to interpersonal trust. What
little research does exist suggests that institutional trust is ‘determined less on direct
observation but more on decision outcomes’ and ‘the efficiency and fairness of the
organization-wide systems created’ (Costigan et al., 1998, p. 309). In addition, the
actions of top management play a special role in generating institutional trust (Kramer,
1999); not only do individuals tend to generalize from vivid, highly salient events
involving institutional leadership when assessing the trustworthiness of institutions and
organizations (Zimmer, 1972), but ‘people may draw general inferences about institutional
trust from the behaviour of highly visible role models’ (Kramer, 1999, p. 589).
Given the lack of work exploring the basis for institutional trust, it is helpful to look at
related work on interpersonal trust (Six and Sorge, 2008). Several typologies and categorizations
of ‘personal’ or ‘interpersonal’ trust exist, organized, typically, around the
basis for the trusting relationship. Rousseau et al. (1998) identify two broad types –
‘calculative’ trust and ‘relational’ trust (also called ‘affect-based’ (McAllister, 1995) or
‘identity-based’ trust (Coleman, 1990) – while Shapiro et al. (1992), Lewicki and
Benedict-Bunker (1995), and Maguire et al. (2001) all work with a slightly more nuanced
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 375
set of categories – calculus-based, knowledge-based and identification-based trust.
Notably, within each of these typologies, the strong form of trust in which we are
interested derives from actors’ identities and identification processes, i.e. identity-based
or identification-based trust. These involve, in addition to the predictability of trustee
behaviour associated with both weak and strong trust, the extra feature of expected
trustee benevolence or goodwill. Some have gone so far as to refer to trust based on
identification as the ‘highest order of trust’ (Shapiro et al., 1992, p. 371).
We can see then that work investigating the different bases upon which trust can be
built suggests that institutional trust may be based in identity and identification. It is this
possibility that forms the basis of our research and, specifically, our focus on how
institutional trust evolves in the context of a merger where issues of the trustee’s identity,
i.e. organizational identity, are of particular importance.
INSTITUTIONAL TRUST AFTER A MERGER
Mergers and acquisitions are important strategic options with significant consequences
for firms (Hunt, 1990; Jemison and Sitkin, 1986; Larsson and Finkelstein, 1999;
Lubatkin, 1987; Seth, 1990) and for the individuals within them (e.g. Cartwright and
Cooper, 1990, 1996; Daniel and Metcalf, 2001; Napier, 1989; Schmidt, 2002). Firms
adopt merger and acquisition strategies to achieve various objectives, including: increasing
market share; being a first-mover in industry consolidation; enhancing brand
strength or reputation; reducing operating costs; entering new industries; and accessing
talent, technologies, or manufacturing capabilities (Schmidt, 2002). Much of the early
research in this area focused on questions such as which firms, when combined, had the
highest chance of achieving these objectives and of consequently improving their financial
performance (Chatterjee et al., 1992; Fowler and Schmidt, 1989; Kusewitt, 1985;
Lubatkin, 1987).
Empirically, however, success has been elusive, with ‘fewer than one quarter of
mergers and acquisitions achiev[ing] their financial objectives, as measured in ways
including share value, return on investment and postcombination profitability’ (Marks
and Mirvis, 2001, p. 80). Thus, more recent research has tended to focus on identifying
reasons for these dismal outcomes (Hunt, 1990; Pablo, 1994). Reviewing research into
the performance of mergers and acquisitions, Schmidt (2002, p. 8) concludes that the top
seven obstacles are ‘directly or indirectly related to the strategic management of people’,
which is consistent with most practitioner-oriented discussions (e.g. DePamphlis, 2001;
Marks and Mirvis, 1998; Schweiger, 2002) as well as other reviews (e.g. Napier, 1989).
For instance, Vaara (2002, p. 213) writes that ‘an overall theme in these studies [examining
employees’ reactions to mergers and acquisitions] has been that human-resourcerelated
problems usually receive too little attention, or are under-managed’, while
Schmidt (2002, p. ix) similarly concludes that ‘the so-called soft issues – managing the
vast change that will occur in both organizations, retaining key talent, and getting the
best of both organizations’ cultures –…arethevery issues that…cause mergers to
fail’. This shift towards understanding the ‘postcombination’ (Marks and Mirvis, 1998) or
‘integration’ phase (Schweiger, 2002) of mergers and acquisitions has resulted in a
significant volume of research into merging organizational cultures and the challenges
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips376
of acculturating employees as causes of integration problems (Buono et al., 1985;
Cartwright and Cooper, 1996; Larsson, 1993; Nahavandi and Malekzadeh, 1988;
Weber et al., 1996).
With the monetary stakes so high, firms have an obvious incentive to avoid these
integration problems, especially the loss of valuable human resources. As DePamphlis
(2001, p. 224) writes:
If done correctly, the integration process can help to mitigate the loss of key talent or
managers and the potential deterioration in employee morale and productivity. The
potential loss of this ‘human capital’ is perhaps one of the greatest risks associated with
mergers and acquisitions.
Mergers and Trust
One central issue in research on post-merger integration and human resource chal-
lenges, including the retention of key talent or managers, is that of trust. In the context
of a merger, trust ‘fosters integration, co-operation and commitment to shared goals’
(Bijlsma-Frankema, 2001, p. 199), so understanding and addressing issues of trust are
crucial (Daniel and Metcalf, 2001; Marks and Mirvis, 1998; Napier, 1989; Schmidt,
2002). For example, Xerox Vice-President John Vester (Schweiger, 2002, p. 38), in
describing ‘lessons learned about integrating acquisitions’, explains how he and his
organization actively sought ‘to build a culture and environment of trust’ and why
creating an ‘environment of trust…wassoimportant for our success’.
One reason for this is because interpersonal trust and dialogue are related in a virtuous
circle:
Trust in managers and colleagues, in turn, brings about psychological safety, a necessary
condition for openness on which dialogues and monitoring are thriving. These
factors seem to influence each other in a cycle of growing trust. (Bijlsma-Frankema,
2001, p. 205)
The absence of trust, on the other hand, represents a barrier to sharing knowledge
between individuals from different legacy organizations and, thus, to finding solutions to
post-acquisition problems (Bijlsma-Frankema, 2004).
A number of features of mergers (and acquisitions) have been associated with lower
levels of trust. Some relate to the context of the merger, the pre-merger organizations,
and their histories (e.g. an uncertain economic context; absence of positive history of
interactions between the organizations); while others relate to the post-merger integration
process (e.g. infrequent or poor quality communications; erosion of autonomy) and
its outcomes (e.g. absence of reward and job security enhancement; low levels of perceived
organizational justice). These factors from the literature are summarized in
Table I.
Despite these notable contributions to understanding trust in the context of mergers,
this work has important limitations and thus presents opportunities for further research.
First, it is common in this body of research, and sometimes even within a single study, to
© Blackwell Publishing Ltd 2008
Table I. Features of mergers and acquisitions associated with lower levels of trust
Features of pre-merger organizations and context
Hostile mode of takeover Stahl et al. (2003, 2006)
Stahl and Sitkin (2005)
Feelings of betrayal due to pre-merger secrecy Schweiger (2002)
Low perceived attractiveness of dominant organization’s human resource and
reward systems
Stahl et al. (2003, 2006)
Low tolerance to change in less dominant organization Nikandrou et al. (2002)
Absence of positive history of interactions between organizations Stahl et al. (2003, 2006)
Stahl and Sitkin (2005)
Negative reputations of individual managers Searle and Ball (2004)
Cultural distance between organizations Stahl et al. (2003, 2006)
Stahl and Sitkin (2005)
Power inequality between organizations Stahl and Sitkin (2005)
Uncertainty in the broader economic context Nikandrou et al. (2002)
Features of post-merger integration process
Loss of Institutional Trust 377
Erosion of autonomy Stahl et al. (2003, 2006)
Stahl and Sitkin (2005)
Absence of dialogue leading to shared knowledge, norms and goals Bijlsma-Frankema (2001)
Absence of cultural tolerance and sensitivity Stahl and Sitkin (2005)
Infrequent or poor quality communications Nikandrou et al. (2002)
Stahl and Sitkin (2005)
Perceived violations of the psychological contract between employees and
their firms
Searle and Ball (2004)
Slow-paced integration Stahl and Sitkin (2005)
Poor employee relations Nikandrou et al. (2002)
Absence of reward and job security enhancement Stahl and Sitkin (2005)
Low levels of perceived organizational justice Searle and Ball (2004)
use the same label of ‘trust’ to refer to both interpersonal and institutional trust, despite the
important differences between them and calls for more nuanced research: ‘operationalisations
of the abstract concept of trust need to be sensitive to the specific characteristics
of the trustors and trustees concerned’ (Mollering et al., 2004, p. 557). Second, most
empirical work on trust refers, often implicitly, to interpersonal rather than institutional
trust; in other words, the literature on mergers mirrors the broader organizational
literature in that institutional trust receives far less attention. Thus, despite it demonstrated
importance, little research has been undertaken that aims at understanding
institutional trust in the context of mergers.
Mergers, Organizational Identity and Identification Processes
If we are to explore the relation between institutional trust and identity in the context of
mergers, we need to understand the nature of organizational identity – it is this special kind
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips378
of identity that characterizes the potential trustee – as well as the identification processes
of individual organizational members who are potential trustors. Very broadly, the
concept of organizational identity refers to those attributes of the organization that
members perceive as central, enduring, and distinctive (Albert and Whetten, 1985;
Ashforth and Mael, 1996). It is closely related to but distinct from the concept of culture.
Like Hatch and Schultz (1997, p. 357), we view culture as ‘a context within which
interpretations of organizational identity are formed’. In other words, culture is the
broad context within which processes of identity formation and identification take
place.
A coherent and attractive organizational identity influences the degree to which
members identify with the organization, assimilate its core values and accept its goals as
their own (Cheney, 1983; Dutton et al., 1994; Scott and Lane, 2000). While there has
been much interest in organizational identity as a concept and an empirical focus of
study since Albert and Whetten’s seminal article, there is little consensus in terms of how
organizational identity should be studied. In fact, the organizational identity literature is
characterized by ‘amazing theoretical diversity’ (Pratt and Foreman, 2000, p. 141) with
very different theoretical and methodological lenses being brought to bear by different
groups of researchers. Capturing and organizing this diversity in a recent review, Cornelissen
(2006) describes six distinct ‘research traditions’ in the literature on organizational
identity: the organizational communication, organizational behaviour, cognitive
framing, discursive psychology, institutional theory, and social identity traditions.
These different approaches vary in terms of ‘where “organization” is located and
wherein it is constituted: in language, cognition and/or behaviour’ (Cornelissen, 2006,
p. 696).
Our approach falls into what Cornelissen (2006, p. 698) refers to as the ‘discursive
psychology’ tradition (cf. Humphreys and Brown, 2002; Maguire et al., 2001) in which
‘organization’ is constituted ‘not only through discursive acts (i.e. language), but also
through the sensemaking of the members of the organization as interactants (i.e. cognition)’;
and ‘organizational identity’ refers to ‘collective meaning that is discursively
(re)constructed in a social context’. Research in this tradition challenges conceptualizations
of identity as objective and essential, pointing instead to its processual, dynamic,
constructed and contested nature; organizational identity is precarious and fluid,
and is produced discursively over time through the production, distribution and consumption
of texts which make sense of the organization and its actions (Cornelissen,
2006).
More specifically, we see organizational identity as an ongoing achievement of social
construction that happens through narrative – ‘stories about organizations that actors’
author in their efforts to understand, or make sense of, the collective entities with which
they identify’ (Brown, 2006, p. 734). Organizational identities are constituted by the
narratives that organizational members, and sometimes external parties, author about
them: organizational identity is a discursive construct ‘and “resides” in the collective
identity stories that, for example, people tell to each other in conversations, write into
corporate histories, and encode on websites’ (Brown, 2006, p. 734). As a result, the study
of organizational identities becomes the study of the identity narratives that characterize
an organization.
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 379
Organizational identities are not necessarily coherent and widely shared. At times,
multiple narratives may coexist, each proposing an alternative interpretation of the
organization, its history and its nature (Boje, 1995; Humphreys and Brown, 2002; Pratt
and Rafaeli, 1997). Nor are organizational identities necessarily stable. They can be the
target of intentional change efforts (Fiol, 2002); or they can be transformed as a side effect
of major organizational changes, including mergers, and preliminary research indicates
that this involves a transition period of ‘identity ambiguity’ (Corley and Gioia, 2004).
Ultimately, ‘the extent to which an organization’s identity is in flux or enduring is an
issue to be decided empirically’ (Brown, 2006, p. 744).
Similar to the production of organizational identity, identification of individuals with
their organization is also largely linguistic (Maguire and Hardy, 2005) as ‘[t]he process
of identification is conducted primarily with language, and the product of identification
is expressed primarily with language’ (Cheney and Thompkins, 1987, p. 11). Identification
can be ‘situated’ or ‘deep-structure’ (Rousseau, 1998), with the former referring
merely to organizational members’ perception of common interest with their organization,
and the latter referring to more profound changes wherein an organization is
incorporated into members’ very understanding of themselves. In other words, deepstructure
identification reflects ‘the perception of oneness with or belongingness to an
organization, where the individual defines him or herself in terms of the organization(s)
in which he or she is a member’ (Mael and Ashforth, 1992, p. 104; emphasis
in original). Deep-structure identification, the focus of our paper, is important because
it can lead to more cooperative behaviour, support for the organization, and
lower turnover among members (Abrams et al., 1998; Dutton et al., 1994; Tyler,
1999).
In the literature on mergers, there is a small but growing body of work addressing
organizational identity and the identification of organizational members (Haunschild
et al., 1994; Hogg and Terry, 2000; Terry and O’Brien, 2001; Terry et al., 2001; van
Knippenberg et al., 2002; van Leeuwen et al., 2003; Zaheer et al., 2003).
Mergers
disrupt identification because they represent a possible threatening change in identity
(Haunschild et al., 1994): ‘if the [new post-merger] identity is not perceived as being able
to offer employees a positive social identity, they might reject it and cling to the old,
which is likely to engender negative intergroup responses and feelings that may jeopardize
the success of the merger’ (Terry et al., 2001, p. 267). However, case study (van
Knippenberg et al., 2002) and experimental (van Leeuwen et al., 2003) research suggests
that this is moderated for individuals from the legacy organization that emerges as
dominant in the merger.
Despite this progress, there is a need for more research: although it is strictly no longer
true that ‘no empirical study to date has directly investigated how reactions to a merger
are rooted in the pre-merger group identity’ (van Leeuwen et al., 2003, p. 679), such
work is still quite rare (see Björkman et al., 2003 for a notable exception) and no studies
have yet explicitly explored the connections between identification processes and institutional
trust. Our goal in this study is to begin to fill this gap and our research questions
are as follows: (a) what leads to low levels of institutional trust among organizational
members after a merger?; and (b) what role do organizational identity and identification
processes play?
[3]
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips380
METHODS
In this section, we describe the empirical study upon which this paper is based. We
discuss the research design, the selection of our research site, the data collection process,
as well as the steps in our data analysis.
Research Design and Site: A Study of Citigroup
Marshall and Rossman (1995) suggest that qualitative studies are most appropriate for
studying poorly understood phenomena. In a similar vein, Lee (1999, p. 43) argues that
qualitative research is appropriate when ‘(a) contextualization, (b) vivid description, (c)
dynamic (and possible causal) structuring of the organizational member’s socially constructed
world, and (d) the worldviews of the people under study’ are important. A
detailed, qualitative investigation of a single case study is therefore appropriate for our
purpose because what leads to low levels of institutional trust in post-merger contexts is
not well understood; because we are interested in the worldviews of organizational
members experiencing low institutional trust; and because our study is exploratory and
aimed at theory building (e.g. Eisenhardt, 1989; Yin, 2003).
Our choice of a qualitative study is also consistent with calls in the trust literature for
more qualitative work to test the validity of the extensive theoretical work done to
date. For example, in the introduction to their special issue on trust, Mollering et al.
(2004, p. 560) conclude that the existing literature is overly quantitative and they
‘would like to see new conceptual development and more qualitative work’. Our
study is one step in redressing this ‘overbalance’ and in performing ‘collective “reality
checks” on concepts that have been discussed for a long time’ (Mollering et al., 2004,
p. 560).
Our study focused on the merger of Citicorp and Travelers in 1998 to form Citigroup.
We selected this case study for several reasons. First, our selection has ‘rare or unique’
qualities that make it a logical candidate for ‘theoretical sampling’ ( Eisenhardt, 1989;
Yin, 2003). Preliminary research revealed that issues of trust contributed significantly to
a range of post-merger complications that we suspected could highlight the dynamics of
trust in the context of a merger. Second, the merger was the largest ever completed
among financial institutions up to that point. This ensured considerable press interest
and the availability of a tremendous range of information on the merger. Third, prior to
the merger, Citicorp had a history of being receptive and helpful to academic researchers.
Our entry into the newly merged company was facilitated and supported by a legacy
department of Citicorp that had dealt with academic researchers and which continued to
do so for some time after the merger. We therefore had excellent access to people and
archival material for the study. Fourth, the merger had happened just prior to the time
we began the study, which meant that the events surrounding the merger were still fresh
in the minds of interviewees, and issues of trust, particularly among former Citicorp
employees, were prominent.
Our initial interest when we approached Citigroup was simply the dynamics of trust
following a merger. The company was very interested in this research due to their recent
experiences and agreed to allow us access. After discussions with senior managers, we
© Blackwell Publishing Ltd 2008
identified parts of the organization that suffered from continuing problems with trust
following the merger. Based on the guidance of these managers, we focused on two
departments at the head office in New York (which must remain unnamed to respect
our confidentiality commitments). This choice maximized the chance of identifying
processes that led to low levels of trust.
Data Collection
We collected data over a 24 month period starting in March 1999. We began by
familiarizing ourselves with the case. We assembled and reviewed a wide range of
primary and secondary sources dating from prior to the merger up to when we gained
access, and continued to amass documentation from both inside and outside the
company over the course of the project. Primary sources included annual reports,
company newsletters, corporate videos, press releases, communications to shareholders,
and some internal reports from the Human Resources function. Secondary sources
included a wide range of journalists’ accounts, stock analysts’ reports and news
reports.
Next, we interviewed Citigroup employees who were members of the departments
identified by the senior managers as having significant problems with trust after the
merger. As it turned out, these individuals had been members of Citicorp prior to the
merger. While pre-merger organizational membership was not a part of our sampling
methodology, based on our discussions with senior managers it was apparent that
problems with trust were disproportionately represented among Citicorp legacy
employees.
As mentioned above, our research project was funded and sponsored by the Citigroup
Behavioral Science Research Council, an organizational unit devoted to facilitating
organizational research within Citigroup. Our primary contacts there facilitated introductions
to members of the departments that we had identified as being of interest,
ensuring good cooperation. In total, 15 Citicorp legacy employees participated in our
study.
[4]
Loss of Institutional Trust 381
Several were interviewed on more than one occasion during the term of the
study. Our claims in this paper therefore refer to the people we interviewed who
experienced a loss of trust in their organization; we do not generalize to all employees of
Citigroup.
Interviews were taped with permission, and later transcribed. Quotations used in this
paper were drawn from the interview transcripts. Interviewees were asked to recount the
story of the merger, how it affected their relationship with others and with the organization
in general, and why. In most cases, interviewees raised the topic of trust themselves;
where they did not, specific ‘probe’ questions about trust were asked at the end of
the interview. In researching trust in this way we are consistent with the definition of trust
as an expectation (and hence a mental state that can be effectively asked about) and
other empirical studies on the topic. Mollering et al. (2004, p. 563), in their review of
approaches to the study of trust, argue that one important ‘approach to measuring
trust(worthiness) requires respondents to report how they perceive a specific other actor
(target)’. Other examples of this kind of approach include Atkinson’s (2004) where she
asks her respondents to talk about their own understanding of trust and develops a
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips382
nuanced theory of interpersonal trust based on their responses. Similarly, Stahl et al.
(2003, 2006) asked employees to respond to a series of hypothetical scenarios by indicating
the extent to which they would trust the acquiring firm’s management. Research
on trust is notoriously difficult, and careful, indirect questioning is an effective method for
bringing the expectations that underlie trust to the surface.
Data Analysis
The analysis of the data comprised four main stages. In the first stage, we organized the
case study data into an ‘event history database’ (Van de Ven and Poole, 1990). This was
done by chronologically ordering descriptions of events taken from the raw data –
interview transcripts, interview and field notes, primary sources such as Citigroup documents,
and secondary sources such as journalists’ accounts – and juxtaposing multiple
accounts against each other to ascertain the degree of convergence.
In the second stage, we coded the interview transcripts and notes for references to
trust, organizational identity and identification. Coding proceeded, initially, on two
levels: (1) using a lexicon of concrete terms grounded in the data (e.g. ‘Citibanker’,
‘cost-oriented’, ‘mistrust’, etc); and (2) using a lexicon of more abstract terms arising
from ‘the a priori specification of constructs’ (Eisenhardt, 1989) found in the literature
(e.g. ‘central, distinctive and enduring feature of organization’, ‘identity narrative’,
‘institutional trust’, etc). As the coding proceeded, we were careful to differentiate
between interpersonal and institutional trust: when a reference to trust was linked to
an individual as the trustee, we coded it as an instance of interpersonal trust; conversely,
if the reference referred to a collective actor or organized system as the trustee
then it was coded as an instance of institutional trust. We also coded for changes over
time.
In the third stage of data analysis, we complemented interviewees’ narratives of the
merger and resulting organization from the previous stage with organizational identity
narratives embedded in internally- and externally-produced texts and combined these to
track the evolution of identity over time. We coded speeches and announcements from
top management, annual reports, newsletters, etc as well as media accounts of the
merger, consistent with established interpretive methods for assessing organizational
identity (e.g. Ravasi and Schultz, 2006) and for studying mergers and acquisitions (e.g.
Vaara et al., 2006).
The fourth stage of analysis involved axial coding – the search for and identification
of relationships between and among first order codes – and clustering theoretical categories
into a few key aggregate dimensions (cf. Corley and Gioia, 2004). This analysis
continued iteratively, moving among data, emerging patterns, and existing theory and
research until the patterns were refined into adequate conceptual categories (Eisenhardt,
1989). This permitted a synthesis anchored both empirically in our data and theoretically
in the literature. Figure 1 shows our data structure (cf. Corley and Gioia, 2004), highlighting
the theoretical categories that make up our framework relating organizational
identity, individuals’ identification processes and loss of institutional trust following a
merger.
© Blackwell Publishing Ltd 2008
First-Order Codes
Theoretical Categories
Aggregate Theoretical
Dimensions
Statements that Citicorp and Travelers were very
different companies
Media reports contrasting Citicorp and Travelers
Statements about Citigroup becoming more like
Travelers
Press releases offering rationale for Citigroup as “One
stop financial shopping…”
Questioning whether acquisition or merger
Media reframing merger
: “Now its Sandygroup…”
Internal communications using “Just married!”
analogy suggesting merger of equals
Appointment of co-CEOs
Maintaining then changing HR policies
Making headcount reductions
Exit of John Reed
References to self as “Citibanker”
Citibankers’ descriptions of Citicorp
Citibankers asking “who are we?” following merger
Citibankers’ descriptions of Citigroup
Statements about trusting
and distrusting specific
individuals
Descriptions of how interaction between co-workers
from different legacy organizations builds
interpersonal trust between them.
Statements about trusting or
distrusting organizations
or “the system”
Figure 1. Overview of data structure
Loss of Institutional Trust 383
Organizational Identity
Narratives
Post-merger Change
Narrative
Organizational Actions
referred to in Post-merger
Identity Narratives
Self Categorization
Distinctive Features of
Organization
Interpersonal Trust
Institutional Trust
CITIGROUP
On 8 October 1998, Travelers Group merged with Citicorp to form Citigroup. The
merger was, at the time, the largest merger ever completed. The merged firm had over
US$700 billion in assets, a market value of nearly US$84 billion, 170,000 employees, and
over 100 million customers worldwide. By the first quarter of 2000, the merged company
was the most profitable company in the world (Economist, 2000, p. 59).
In 1997, just prior to the merger, Citicorp was the largest US based financial services
firm. It operated in over 100 countries, had a market value of US$67.4 billion,
and
profits of US$3.6 billion. Since its founding as the First National City Bank in 1812, the
bank had a long history of innovation, organic growth and stability. Its core business unit,
Citibank, was the first bank to offer compound interest on savings accounts, to offer
negotiable certificates of deposit, and, in 1977, to introduce ATM’s as a mass market
service. ‘Citibankers’, as many employees of Citicorp referred to themselves, prided
themselves on their innovativeness and customer service.
Travelers Group, on the other hand, was a diversified, integrated financial services
company engaged in a wide variety of services including asset management, life insurance,
property casualty insurance, and consumer lending. In 1997, Travelers Group had
a market value of US$69.7 billion and profits of US$3.1 billion. Its operating companies
included Salomon Smith Barney, Travelers Life & Annuity, Primerica Financial Services,
and Commercial Credit. Travelers Group first emerged in 1986 when Commercial
Credit was spun off from its parent, Control Data Corporation. In 1988 it acquired
Primerica Corporation and added life insurance and brokerage to its lines of business. In
1993, Travelers acquired the retail and asset management business of Shearson Lehman
Brothers and combined them with Smith Barney. In November 1997, Travelers
acquired Salomon Inc. and merged it with Smith Barney to create Salomon Smith
Sources of
Organizational
Identity
Ambiguity
Elements of
Identification
Types of Trust
[5]
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips384
Barney, one of the world’s largest global investment and securities brokerages. The
history of Travelers, therefore, in contrast to the organic growth and innovation that
characterized Citicorp, was one of repeated acquisitions, rapid growth and continual
change. The focus in the company was on efficiency rather than innovation, and the
company’s strategy was to buy up inefficient business and rapidly integrate and overhaul
them.
The merger’s logic was to create ‘one-stop financial shopping for consumers, offering
Citicorp’s strengths of traditional banking, consumer finance, and credit cards along with
insurance and brokerage services from Travelers and its units’ (CNNfn, 1998). Sandy
Weill and John Reed, the CEOs of Travelers and Citicorp respectively, described the
merger in a joint statement as follows (Citigroup, 1998, p. 1):
Citicorp and Travelers Group bring together some of the best people in the financial
services business, creating resources for customers like no other – a diversified global
consumer financial services company, a premier global bank, a leading global assets
management company, a pre-eminent global investment banking and trading firm,
and a broad-based insurance capability. Our ability to serve consumers, corporations,
institutions, and government agencies, domestic and foreign, will be without parallel.
The logic of the merger, and the political pull of the two companies, was such that they
successfully managed to convince the American government to set aside the Glass–
Stegall Act preventing the amalgamation of insurance and financial services in one
company and allowing the creation of Citigroup.
Following the merger, Weill and Reed were appointed co-CEOs of the new firm. This
approach to managing transition in the new company was adopted in the organization,
and two managers, one from Citicorp and one from Travelers, manned many critical
posts. Initially, the merger was presented as one of equals, with communications circulated
internally and externally sometimes using the metaphor and imagery of marriage,
announcing that the merger partners had ‘Just married’. Over time, however, despite
technically being a merger, the fusion of the companies was increasingly viewed as an
acquisition of Citicorp by Travelers by organizational members as well as outside
observers in the business press. Indeed, the shared CEOship continued until July 1999
only, when Weill and Reed announced that their duties would be split, and in early 2000
Reed formally retired, prompting Business Week to declare in a headline, ‘Now, it’s
Sandygroup’ (Timmons, 2000).
By early 2000, the merger was characterized as a success in economic terms, but at the
same time, major problems of integration still remained to be solved. Bringing together
two organizations with such different histories, systems, and leadership styles created an
atmosphere of suspicion and an absence of trust that was taking years to work itself out.
Questions of trust and identity were constant topics of conversation as employees tried
to make sense of the actions and communications of top management, attempting to
construct and comprehend the new company. The challenge facing managers was to
manage the ongoing merger process effectively and to reduce the uncertainty of this
arrangement; the challenge facing everyone in the company was to make sense of this
new organization.
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 385
LOSS OF IDENTIFICATION AND INSTITUTIONAL TRUST AFTER A
MERGER
In this section, we present the results of our study of the loss of institutional trust after a
merger (see Table II for a summary). Our analysis points to the crucial role of organizational
identity and processes of identification in influencing the level of institutional
trust experienced by employees before and after the merger. Prior to the merger in
October, 1998, the respondents in our study, who continued to self-categorize themselves
as ‘Citibankers’ years after the merger, experienced their legacy organization,
Citicorp, as having a relatively clear and unambiguous organizational identity, reflected
in convergent narratives of the organization. They also found the distinctive features of
Citicorp appealing, and through a process of identification, experienced ‘a sense of
oneness or belongingness’ (Mael and Ashforth, 1992, p. 104) and, therefore, high levels
of institutional trust which our study shows to be identity-based.
After October 1998, however, our findings suggest that the high degree of ambiguity
about the new post-merger organization’s identity initially undermined institutional
trust: at this time Citibankers could not identify with Citigroup because competing and
divergent narratives of the new organization, along with a narrative of the change
process as being a merger of equals, made it difficult to know or predict the new
organization’s central, distinctive and enduring features. By mid-2000, however, as
narratives of Citigroup became increasingly convergent and coherent, and as a narrative
of the change process as being an acquisition by Travelers became dominant, the identity
of the new organization became less ambiguous but institutional trust continued to be
undermined due to an absence of identification with the new organization: at this time
Citibankers did not identify with Citigroup as its central, distinctive and enduring features
differed substantially from the legacy identity of Citicorp with which they remained
identified, as summarized in Table II.
Pre-Merger Identity-based Institutional Trust
Our first finding represents a contribution to the broader literature on trust as well as to
the literature on trust after a merger. As discussed in our literature review, while several
bases of interpersonal trust have been proposed, the bases of institutional trust have
remained largely unexplored. Our study suggests that, like interpersonal trust, institutional
trust can be identity-based. Our respondents’ references to trust in their premerger
and post-merger organizations revealed a clear pattern: this type of trust was
associated with coherent and convergent organizational identity narratives combined
with individuals’ ability to identify with the organizational identity constructed through
these narratives. As our interviewees discussed their trust in their legacy organization and
lack of trust in Citigroup, they made repeated references to who they were as individuals,
to what they believed to be defining features of their pre-merger and post-merger
organizations, and the relationship among these (see Table II).
We consistently found that the individuals experiencing problems of institutional trust
to whom upper management directed us were Citicorp legacy employees. Throughout
the interviews, they referred to themselves with obvious pride as ‘Citibankers’, and they
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips386
Institutional Trust? Yes No – initial loss of institutional trust No – continued loss of institutional
trust
associated expectation of absence
of benevolence
towards actions consistent with
Travelers’ legacy identity, with
Citigroup: characterized by
growing predictability but
negative results
No – comparison of organizational
and individual identity yields
hierarchical; cost-conscious;
control-oriented; ‘lean’ HR
Citigroup: unclear Citigroup: increasingly like
Travelers – centralized;
employees
• Headcount reductions
• Extension of Travelers’ HR
policies to Citicorp legacy
Acquisition of Citicorp by Travelers
Citigroup: increasingly convergent
and coherent
Citigroup: characterized by absence
of predictability
individual identity due to high
organizational identity ambiguity
No – inability to perform
comparison of organizational and
policies
Identification with organization? Yes – comparison of organizational
Organizational identity ambiguity? Low High Low
Self-categorization Citibanker Citibanker Citibanker
reductions
Travelers’ HR policies
• Stated policy of no headcount
Expectations re organizational actions Citicorp: characterized by
predictability and benevolence
and individual identity yields
positive results
policies; safe and comfortable
Distinctive features of organization Citicorp: decentralized; democratic;
flexible; innovative; generous HR
Not applicable • Appointment of co-CEOs • Exit of John Reed
• Retention of both Citicorp’s and
Dominant post-merger change narrative Not applicable Merger of Citicorp and Travelers
as equals
Organizational identity narrative(s) Citicorp: convergent and coherent Citigroup: multiple, competing and
divergent
Prior to October, 1998 October, 1998 to mid-2000 After mid-2000
Organizational actions referred to in
post-merger identity narratives
Table II. Relationships among organizational identity, identification and institutional trust of ‘Citibankers’
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 387
spoke fondly, and often with a sense of loss, about their legacy organization. Interviewees
made repeated reference to differences between their legacy organization with its ‘Citi
way’, and the post-merger organization, and they made it clear that they preferred the
former. For example, one employee, reflecting on their legacy organization of Citicorp,
stated that Citibankers had been ‘spoiled’ as compared to the post-merger situation in
which they now found themselves.
This finding suggests that institutional trust can be identity-based; its creation, maintenance
and loss depend upon organizational identity and identification of organizational
members. More specifically, we propose that a high level of institutional trust
experienced by members of an organization depends on the existence of convergent and
coherent organizational identity narratives (i.e. an absence of identity ambiguity) and on
members’ ability to identify with the central, distinctive and enduring features constructed
in these narratives. Our second and third findings, which describe mechanisms
involved in the loss of institutional trust, support this conclusion.
Post-Merger Organizational Identity Ambiguity and Initial Loss of
Institutional Trust
Our second finding is that organizational identity ambiguity undermines institutional
trust. That mergers lead to periods of ambiguity and uncertainty for organizational
members is well-documented (Marks and Mirvis, 1998; Schweiger, 2002). However,
little attention has been paid specifically to organizational identity ambiguity and its
consequences. Our study shows that, immediately following the merger of Citicorp and
Travelers, several factors contributed to ambiguity as to the identity of the new postmerger
organization, Citigroup: a dominant change narrative of ‘mergers of equals’ that,
combined with the quite different and strong legacy identities of Citicorp and Travelers,
left Citibankers wondering whether any coherent organization would emerge; early
post-merger organizational actions that were open to multiple interpretations and divergent
sensemaking; and the circulation of multiple and competing Citigroup identity
narratives. As a result, Citibankers did not know the features of Citigroup that were (or
would turn out to be) central, distinctive and enduring, so they could not identify with it.
They experienced the new organization as unpredictable and had less trust in it than
they had in their legacy organization, Citicorp. These relationships are summarized
in Table II.
Interviewees consistently reported a loss of trust beginning when the merger was
announced, and many perceived it as pervasive among Citibankers, especially those at
the former head office: ‘The closer you come to New York the scarcer a commodity trust
is. I’d say it is very poor’. Citibankers were suddenly suspicious of the organization in
which they found themselves:
There was a lot of suspicion in the beginning, there were people who said ‘oh
these terrible people. You know these Salomon Smith Barney people they are
awful blah blah blah. You know they don’t care about this, they don’t care about
that…’
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips388
Interviewees stated clearly that ‘there was a lot of mistrust on the part of employees’
and ‘people didn’t trust anybody…They were sure things were going to be really bad
and there were a lot of issues’. One was ‘sure’ that the new organization was ‘out to
screw’ them.
Moreover, this initial loss of trust in their new organization occurred at the same time
as organizational identity ambiguity increased:
So that is the problem right now – that we are trying to figure out: Who are we? What
are we going to call ourselves to the public? What are we going to call ourselves to
ourselves?
This identity ambiguity meant that Citibankers lost the identification that they had with
Citicorp; they could not identify with Citigroup as there were divergent views as to its
central, distinctive and enduring features, in the narratives told about it. For instance,
one manager summarized:
We don’t know what Citigroup is. I mean we knew what Citicorp was. It was a bank
holding company…butit[the identity of Citigroup] has not been articulated in the
same way and people don’t identify with it, with Citigroup.
They also experienced their new organization as unpredictable: there was ‘a lot of
uncertainty’ which led to, in their words, ‘anxiety and tension on the part of the
Citibankers’ that were ‘looking out, wondering what is happening to me and my unit’.
Others used narratives peppered with psychological metaphors of dysfunction:
It [Citigroup] is schizophrenic, and that’s true and that bothered a lot of people. They
didn’t know who they were and it bothered managers. They wanted to have people as
one thing and we are just not there yet.
Some interviewees attributed these problems to the absence of established norms and
standards, stating that they and their colleagues were unsure ‘what you could do and not
do’. One manager described how expectations formed in the legacy organizations went
repeatedly unmet in the newly merged organization because of differences in operating
policies and procedures, and how this led to low levels of trust:
There was a lot of mistrust on the part of employees who were sure that somebody was
trying to cheat them. They [new parts of the organization] weren’t really, they just
didn’t operate that way. They operated in very different ways so they [the mistrusting
employee] didn’t understand.
In addition, interviewees explicitly linked low institutional trust to the absence of predictability,
explaining, for example, that when ‘you don’t know which end is up, it is hard
to trust’.
Several factors contributed to ambiguity as to the organizational identity of Citigroup.
One important one was the initial post-merger change narrative of the transaction
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 389
offered to employees, regulators and the press, which presented the change as a ‘merger
of equals’ where two companies of similar size and value and with complementary
products and service lines joined to form the first truly global, full-service financial
services firm. Despite an official narrative of a merger of equals – or ‘marriage’ of equals
(early Citigroup internal documentation used the metaphor and imagery of marriage
prominently) – Citibankers wondered whether and how such divergent identities could
be reconciled ‘as equals’. Thus, the ambiguity surrounding the new post-merger organizational
identity was also compounded by the very strong but different identities of
Travelers and Citicorp. Identity narratives of the two pre-merger organizations were
relatively convergent and coherent, and interviewees were acutely conscious of the
differences that separated the two pre-merger companies. Citibankers were especially
suspicious of Travelers, given its identity and how different it was from their own highly
trusted legacy organization. As they recounted the story of the merger to us, they stressed
that ‘there were differences there’ and that the companies had ‘very strongly different
ways of operating day to day’. Employees underlined, especially, ‘a definite difference in
culture’ and ‘very different philosophies and appreciations’.
Citicorp, although technically a holding company, was viewed as a ‘bank’ because of
its heritage, having been created by Citibank in 1960, a unit that dominated its portfolio.
Its growth strategy had emphasized internal growth by developing new businesses, by
offering new products and services to the markets it already served or by entering new
markets. Thus, the identity of Citicorp was that of a bank, organizations that ‘have a
tendency to…being safer and more comfortable places to work’ according to one
employee. Citicorp was viewed as being ‘very big and laborious’ and ‘not necessarily cost
conscious’ – an organization where ‘you wouldn’t find the most economic way of doing
something’, as it operated with ‘lots of process, lots of bureaucracy, lots of controls, stuff
like that’. Citicorp was ‘a decentralized shop…[where,] at corporate [head office], you
couldn’t make a lot of decisions on your own; if you needed to do something you would
have to get approval’. Although they recognized that they worked within ‘more of a
laborious decision making process’, Citicorp’s employees prided themselves on being
‘decentralized’, ‘democratic’, and ‘a big global company that is quite successful in
understanding cultural differences and recognizing the need to let the regions out there
do what they need to do’. This had implications for the types of people who were
comfortable within the organization. Employees stressed that ‘you had to be very flexible
to work here’ and that ‘you had to kind of make your own way.’ Individuals identified
themselves and their coworkers as ‘Citibankers’, an identity associated with selfconfidence
(if not arrogance, mentioned by several interviewees), intelligence and a
sensitivity towards ‘process’ issues.
Travelers, on the other hand, was understood to be a dynamic ‘holding company’ that
pursued growth aggressively through acquisitions, subsequently paying for them through
asset stripping and cost-cutting programmes. Unlike Citicorp, it eschewed big, consultative
processes and programmes. It was ‘very focused, direct, to the point, practical,
getting all the bullshit out of the way, economical’. Power was much more centralized –
‘the corporate structure was much more empowered as an entity, if they wanted to get
things done’ – and structures were more ‘hierarchical’, so ‘big decisions got done faster’.
Unlike the ‘Citibankers’, Travelers employees prided themselves on being ‘very cost- and
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips390
very control-oriented’, and were comfortable with ‘a much leaner human resources set
of policies including pensions’.
As one senior manager described, while comparing the two companies:
From the Citibank side, the functions operated, I think, as much more of a player at
the table, much more as an advisor, consultant and it was quite democratic in the sense
that if you had something to say you had a place at the table, forget[ting] about rank
– you know, what have you. On the Travelers side of the business, it’s very different.
It’s much more hierarchical, more action-oriented than the [Citibank] thoughtfulness.
. . much more order taking as opposed to consulting, suggesting about things, so
those are very strongly different ways of operating day to day.
Also contributing to ambiguity as to the identity of Citigroup were several key organizational
actions that were repeatedly referenced in narratives of the new organization.
For example, the decision to maintain two co-CEOs seemed to reinforce the narrative of
‘schizophrenia’. Additionally, it was rumoured that John Reed was thinking about
retirement at the time of the merger (he denied it) but that his continued presence was
required because federal regulators were not comfortable with the idea of someone of
Sandy Weill’s limited experience running the largest bank in the USA. And this ‘Noah’s
Ark’ approach to management, as one media source put it, didn’t stop at the top as all
through the organization, two managers, one from Citicorp and one from Travelers,
manned critical posts. Similarly, the maintenance of parallel HR systems immediately
after the merger also contributed to identity ambiguity, as stated by one manager
interviewed: ‘it is these, you know – the lack of integration of systems, the lack of
integration of policies – that inhibit the identity’.
Finally, the existence of multiple plausible identity narratives immediately following
the merger also contributed to identity ambiguity. Citibankers realized that only a
limited number of possible outcomes could occur: (1) the new firm might more closely
resemble Citicorp (i.e. approximating a Citicorp acquisition); (2) it might more closely
resemble Travelers (i.e. approximating a Travelers acquisition); (3) it might develop an
new, unique identity (i.e. a true ‘merger of equals’), but the nature of this identity that
would reconcile the clear differences in pre-merger identities was unclear; or (4) the units
of the new company might continue to coexist independently, each with their own strong
identity (i.e. a holding company arrangement). Importantly, the business rationale publicly
offered to justify the merger made the last trajectory very unlikely: the merger had
been motivated by potential synergies to be achieved through some degree of integration
of activities across the various businesses. In addition, the pre-merger identity and
reputation of Travelers rendered the second identity trajectory quite plausible to many
observers both inside and outside Citigroup, and a Travelers acquisition narrative
circulated widely immediately following the announcement of the merger. This made
Citicorp employees suspicious and nervous. It also added to identity ambiguity as it
conflicted with the official story of a merger of equals.
In summary, identity ambiguity makes it difficult for individuals to compare their own
identity with that of their new organization to see if there is a fit, thus undermining
pre-merger identification. As a result, the institutional trust they had pre-merger is lost.
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 391
This finding contributes to the literature by further developing the link between institutional
trust, i.e. trust in an organization, and organizational identity. One benefit of a
strong, coherent, unambiguous organizational identity is that it facilitates the prediction
of organizational actions, and predictability of a trustee by a trustor is an important
precondition for both weak and strong trust (Hardy et al., 1998). Ambiguity of organizational
identity thus undermines the basis of institutional trust in even its weak form.
Reduction of Organizational Identity Ambiguity but Absence of
Identification and Continued Loss of Institutional Trust
Our third finding is that institutional trust is not necessarily restored once organizational
identity ambiguity is reduced and organizational members’ narratives begin to converge
in their descriptions of the new organization’s central, distinctive and enduring features.
Institutional trust can continue to be undermined if individual employees do not identify
with the post-merger organization. In our study, we found that by mid-2000 Citigroup’s
identity was less ambiguous than in the period immediately following the merger transaction.
Several factors contributed to this emerging clarity: a dominant narrative of the
change undertaken as one of ‘an acquisition by Travelers’ rather than a merger of equals;
post-merger organizational actions that were consistent with this narrative, including the
retirement of John Reed as well as efficiency-driven decisions to make headcount reductions
and to increase the hours in a standard work week; and increasingly convergent and
coherent Citigroup identity narratives that highlighted central, distinctive and enduring
features that more closely resembled those of Travelers rather than of Citicorp. As a
result, Citibankers could once again engage in individual-organizational identity comparisons
to assess their fit, as they could prior to the merger. But whereas this process had
yielded positive results, i.e. identification, prior to the merger, by mid-2000 this process
yielded negative results, i.e. absence of identification. Citigroup’s actions were increasing
in predictability, but towards actions more consistent with Travelers’ rather than Citicorp’s
legacy identity, which Citibankers did not associate with benevolence or goodwill.
Thus, institutional trust in its strong form continued to be undermined, as summarized
in Table II.
An interesting pattern emerging from our data related to shifting reasons why our
interviewees, who as we mentioned above continued to call themselves Citibankers years
after the merger, did not have trust in their new organization, Citigroup. Over time,
institutional trust remained low despite less Citigroup identity ambiguity but in parallel
with an emerging theme of the absence of identification with their new organization. As
compared to those at Citicorp, Citibankers at Citigroup described ‘a very different
feeling’ in the new organization and claimed, disapprovingly, that it was ‘almost like
being in a different company’. One put it explicitly: ‘people don’t identify with it, with
Citigroup’.
This coincided with the rise to dominance of a change narrative in which the story of
Citigroup was one of an acquisition by Travelers, replacing the change narrative of a
merger of equals. In the identity narratives of Citigroup we encountered while conducting
our interviews, several key organizational actions were referenced in telling this story.
The first involved a decision to reduce headcount in the new organization, despite public
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips392
statements at the time of the merger that efficiencies achieved through the elimination of
redundant activities were not an important part of the business logic driving the merger.
As one executive put it when describing the content of internal and external communications
at the time of the merger announcement: ‘So many of the mergers at that time
were based really on cost savings,…reduced headcount, but we maintained that that
was not the driving force’. Despite maintaining this position publicly for some time after
the merger, Citigroup nevertheless eventually announced plans to reduce its headcount
substantially. According to one executive, ‘in the end we really did cut tremendously’ but,
he went on to clarify, this was necessitated by the persistence of the Asian financial crisis
and the appearance of financial problems in Russia; it was not, according to them,
planned when the merger was announced. But no amount of reassurances to the
contrary could eliminate the suspicions of Citibankers who were convinced that they had
been lied to and, as a result, they could not trust the organization in which they now
found themselves.
Similarly, loss of institutional trust was also reinforced by a policy decision that
effectively increased the workweek of ex-Citicorp employees in the New York area by 14
per cent, when employees who were used to a 35-hour work week were obligated to
increase it to 40 hours, in accordance with previous Travelers policy. As one interviewee
described, this led to a situation where ‘we had a lot of angry Citibankers who felt that
we cheated them’ and ‘there are still some people who feel they are still being treated
unfairly’.
A third and highly symbolic action was the termination of the policy of co-CEOship,
announced in mid-1999, and, related, the retirement of John Reed, the co-CEO from
the Citicorp legacy side of the merger, in early 2000, leaving Sandy Weill as CEO of
Citigroup. As a result, increasingly the identity narratives of Citigroup converged and
cohered around features viewed as central, distinctive and enduring that resembled
those of the Travelers legacy organization. For instance, upon Reed’s retirement Business
Week declared in a headline ‘Now, it’s Sandygroup’ (Timmons, 2000), and other commentators
described the ‘acquisition’ as complete: ‘Weill’s cost-cutters hacked away at
Reed’s bureaucracy’, wrote one observer, as ‘the merger of equals became a takeover’
(Lowenstein, 2000).
These cost-cutting policy decisions and the exit of Reed were especially difficult for
Citibankers because of what they implied about Citigroup’s emerging organizational
identity – it was moving towards the Travelers’ legacy identity and away from the
Citicorp legacy identity with which they were highly identified. Additionally, these
efficiency-driven policy decisions left little room for interpretations of the new organization’s
identity as one where benevolence or goodwill would characterize the relationship
between organization and individual, directed as they clearly were at organizational
interests at the expense of individuals’ careers and pocketbooks.
As ambiguity surrounding Citigroup’s identity was reduced, it became clear to Citibankers
that they did not identify with the new organization in which they found
themselves. For instance, one Citibanker who, in a staff role, strongly identified with the
consultative and democratic processes at Citicorp, found the new, more hierarchical
processes and expectations that managers would function in ‘order-taking’ mode very
difficult to accept as these ran counter to their sense of self – their identity:
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 393
I sat in meetings in the beginning around a particular topic where I was saying no that
is not going to work and this is the reason why and meeting a lot of da, da, da, da and
got poopoed and booboed and I said fine, I am not going to continue knocking my
head against the wall wasting my time or energy around something is not working….
That is unnatural behaviour for me because as a staff person I think I am paid to give my
opinion and that’s what I am paid to do, absolutely. (emphasis added)
Sources in Human Resources reported that Citicorp legacy employees were more
likely to exit the organization than Travelers legacy employees, stating that turnover
levels were ‘very high’ and that they’d increased ‘hugely, hugely’ since the merger, and
that they were ‘skewed to one legacy group – Citibankers’. Other accounts of Citigroup
confirm an exodus of Citicorp legacy talent. For instance, Schweiger (2002, p. 197) writes
about Citigroup that: ‘Senior management battled, and so did everyone else. This led to
the departure of many good people who felt they did not fit in the new organization.’
In summary, an initial loss of institutional trust due to organizational identity ambiguity
can continue even once ambiguity is reduced, due to the absence of identification
with the new organization. Although the prerequisite for weak institutional trust, predictability,
may be present, thus allowing organizational members to form expectations
about their organization, if members expect an absence of benevolence, this undermines
strong institutional trust.
These findings are consistent with and extend – by highlighting the role of organizational
identity ambiguity, identification processes, predictability and expectations of
goodwill – Fiol’s (2002) model of identity transformation in organizations. She found
that, subsequent to an identity transformation process, highly identified organizational
members experience a phase of ‘deidentification’ and ‘trust-breaking’ due to the reciprocal
relationship between trust and identity. As Fiol (2002, p. 659) argues, ‘[p]eople’s
willingness to identify with a social group is predicated on trust, that itself is based on a
common or shared identity’, and this means that ‘once people trust, identity changes can
be seen as violations of that trust’. The reduction of ambiguity surrounding Citigroup’s
new identity and convergence of Citigroup identity narratives around stories in which
Citigroup was more like Travelers rather than Citicorp were experienced as a significant
organizational identity transformation by the Citibankers we studied, thus their trust in
the new organization was low.
CONCLUSION
In this paper, we contribute to the growing literature examining issues of trust following
a merger. More specifically, we focus on institutional trust, show that it can be identitybased,
and identify mechanisms related to organizational identity and individuals’ identification
processes that result in a sustained loss of institutional trust following a merger.
Initially, immediately following a merger, the identity of the new organization is ambiguous
and this undermines institutional trust. Employees do not experience high levels of
trust because, in the absence of convergent narratives of the new organization, there are
divergent views of which features are central, distinctive and enduring such that the
organization’s actions are experienced as unpredictable. Later, as narratives of the new
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips394
organization become increasingly coherent and convergent, some employees may continue
to experience low levels of institutional trust if they are unable to identify with the
features of the new organization constructed as central, distinctive and enduring in the
stories told about it. Our findings suggest that this is more likely to occur when the new
organization’s identity is very different from employees’ legacy organization in which
there were high levels of identification and institutional trust. Although those features of
the new post-merger organization constructed as central, distinctive and enduring may
become increasingly clear and its actions more predictable, if employees experience an
absence of fit between their sense of self and the new organization’s identity, they will not
expect benevolence or goodwill and, as a result, institutional trust remains undermined.
While it is clear that more research is required to come to a complete understanding
of the dynamics of institutional trust following a merger, our study has pointed to a
promising new line of inquiry linking institutional trust, organizational identity and
identification processes, and it makes a number of important contributions. First, our
study has important ramifications for understanding institutional trust. It is one of few
empirical studies of institutional, as opposed to interpersonal, trust in an organizational
context and provides an identity-based framework for understanding the creation, maintenance
and loss of institutional trust in an organization. Given that little empirical work
has been done on this topic, our case study provides an important first empirical look into
the basis of institutional trust as well as the process through which it is lost.
Second, our research provides insights into the underlying mechanisms leading to the
sustained loss of institutional trust following a merger. In particular, our work shows that
issues of institutional trust need to be considered as theoretically and practically distinct
from issues of interpersonal trust after a merger or acquisition. While there is a significant
body of work looking at trust in the context of mergers and acquisitions, it has not
differentiated clearly between individual and institutional trust; not trusting one’s new
coworker and not trusting one’s new organization are two very different phenomena
and restoring these different types of trust requires quite different actions. Our findings
highlight the importance of identity narratives for generating institutional trust and,
combined with extant research on trust in the context of mergers, suggest steps recently
merged firms can take to increase institutional trust. For example, low levels of perceived
organizational justice as well as perceived violations of psychological contracts with
employees have been linked to low levels of institutional trust (Searle and Ball, 2004) so
reducing post-merger identity ambiguity and establishing an organizational identity as a
‘just’ place to work where psychological contracts are honoured, through the circulation
of stories that highlight this, would appear to be very important.
Actions, because they are interpreted and incorporated into identity narratives told by
organizational members, are also key in this regard, especially the allocation of resources,
positions and functions to members of different legacy organizations in a manner which
can be constructed as fair (Meyer, 2001). In other words, recently merged firms need to
be especially attentive to the symbolism of major policy decisions, which are monitored
closely by employees in their efforts to assess which legacy organization is more dominant
and thus more likely to influence the new organization’s identity, post-merger – will the
new organization in which they find themselves be one where they can expect justice?
Similar to Vaara et al.’s (2005) finding that the decision to make Swedish the official
© Blackwell Publishing Ltd 2008
Loss of Institutional Trust 395
language of a company formed through the merger of Finnish and Swedish organizations
was interpreted in terms of power and dominance, Citigroup’s efficiency-driven
decisions to make headcount reductions and to increase the hours in a work week were
interpreted as signs that Travelers was emerging as the more dominant partner and that
Citigroup would resemble that legacy organization rather than Citicorp. Future research
could investigate whether and how national stereotypes impact on institutional trust in
cross-border mergers, as well as which symbolic actions, if any, can be taken to mitigate
negative impacts of national stereotypes on organizational identity and institutional trust.
In addition to fairness and dominance, actions are also monitored for their (in)congruence
with official corporate identity narratives as employees engage in sensemaking
(Weick, 1995) about the new post-merger organization in which they suddenly find
themselves. Our Citigroup study suggests that the sensegiving efforts of top management
can be undermined by actions interpretable by organizational members as contrary to
the organizational identity promoted in official narratives. Moreover, if post-merger
sensemaking results in employee perceptions of hypocrisy, then this likely serves as an
additional barrier to identification and building institutional trust. Future research could
therefore explore the relationship between sensegiving and sensemaking in the context of
a merger as a particular type of strategic change. Gioia and Chittipeddi (1991, p. 446)
advocate ‘ambiguity-by-design’ – ‘intentional creating, but carefully managing, the
ambiguity that comes with redefining an image of the organization’ – as a ‘more
manageable approach’ to launching change, yet merger contexts, because they bring
together legacy organizations with different identities, may yield ambiguity inherently
less amenable to design – and thus less manageable – than ambiguity generated when
strategic change is instigated in a single organization.
In addition, our research demonstrates how organizational identity ambiguity undermines
institutional trust, a notable risk to which managers, if they do act as designers
of ambiguity, should be attentive when implementing strategic change in any context.
Because research on organizational identity in the ‘discursive psychology’ tradition, like
our study, occupies an intermediate rather than extreme position in which only one of
behaviour, cognition or language is emphasized (Cornelissen, 2006), it appears particularly
well suited to exploring sensemaking and sensegiving during episodes of strategic
change, including mergers and acquisitions, where actions, understandings and narratives
are all relevant to organizational identity ambiguity and its reduction.
Third, our research also has important ramifications for the trust literature. Our study
of trust after a merger allowed us to establish an important aspect of institutional trust:
it depends on organizational identity. Researchers have invoked the concepts of identity
and identification to explain many organizational phenomena, including loyalty (Mael
and Ashforth, 1992), commitment (Foreman and Whetten, 2002), cooperation and
citizenship (Dutton et al., 1994) but there has been little work linking organizational
identity to institutional trust. More specifically, our research links the interaction of
individual and organizational identities to an important outcome – institutional trust –
and lends empirical support from the increasingly important context of mergers to
arguments that ‘[i]ndividual and organizational identities may both support and undermine
one another, so their interactions are critical, especially during times of radical
change’ (Fiol, 2002, p. 653).
© Blackwell Publishing Ltd 2008
S. Maguire and N. Phillips396
Fourth, our findings also shed light on mechanisms underpinning ‘the paradox of a
highly identified work force’: high levels of identification and a strong organizational
identity contribute to organizational success, but they also ‘can become a barrier to
success, particularly if the identity must change’ (Fiol, 2002, p. 655), as it often must
during mergers and acquisitions. Although identification of members with their organization
is typically linked to positive organizational outcomes such as cooperation, altruism,
involvement, commitment and loyalty (Foreman and Whetten, 2002), our research
shows that strong identification can become a double-edged sword in the context of a
merger, especially for those individuals in the dominated as compared to dominant
merger partner. Our study suggests that the loss of identity-based institutional trust plays
an important part in this paradox.
Finally, our research also points to the importance of separating discussions of culture
from discussions of identity in understanding issues of trust in mergers:
In a sense, the cultural perspective has become the dominant paradigm in studies
focusing on organizational issues in M&A integration….Atthesame time, it seems
that ‘culture’ has become a broad umbrella encompassing beliefs, norms, values,
cognitions, emotions, practices, rules or routines, often without clear specification.
This presents a problem when trying to go further in any analyses related to culture:
its all-embracing nature and poor definition lead to obfuscation. (Angwin and Vaara,
2005, p. 1447)
The addition of the more focused concept of identity, therefore, can help to clarify the
complex dynamics that follow a merger. Our findings highlight that institutional trust
depends on organizational identity. At the same time, organizational culture provides the
context in which organizational identities develop and are maintained (Hatch and
Schultz, 1997). Further research could therefore explore in more depth the relationship
between organizational culture and identity and, in particular, how cultural differences
between legacy organizations constrain and enable the emergence of post-merger organizational
identity.
Our study does, however, have limitations. First, it is a single case study. While this
allows us to explore the richness of organizational members’ experiences and provides a
sound basis for theorizing, it also means that generalizations must be made with caution.
Second, while the number of respondents in our study is similar to that found in many
other studies of trust, it is still a narrow window on such a large organization. Third, our
study focused on disaffected organizational members in an attempt to understand the
sources of their disaffection. Future research could be designed to compare between
more and less disaffected organizational members to explore whether and how identification
processes contribute to their differences. Fourth, we relied on interviews to assess
and track trust. Although this is consistent with and even encouraged by other trust
researchers (e.g. Atkinson, 2004; Mollering et al., 2004), it is possible that respondents
were not entirely forthcoming because of the sensitive nature of the topic. Fifth, our
assessment that the disaffected organizational members to whom we spoke were highly
identified with pre-merger Citicorp is based on post-merger interviews rather than
pre-merger interviews or observations; it is possible that Citibankers’ descriptions of their
© Blackwell Publishing Ltd 2008
pre-merger organization and their relation to it were coloured somewhat by nostalgic
retrospective sensemaking. Sixth, the pre-merger situation in Citicorp was characterized
by a highly identified workforce. While this worked to highlight the dynamics of identification
in a merger situation, it also means that we need to be cautious in generalizing
the results of this study to cases where the pre-merger situation is not characterized by the
same high levels of identification.
Finally, our study examined a particular kind of merger: while this case was technically
a merger, it increasingly came to be called an acquisition both inside and outside the
company. This provided interesting material for our study but also raises the question of
how mergers and acquisitions, as well as different kinds of mergers, may differ from the
point of view of institutional trust. One hypothesis is that the identity of an organization
resulting from a merger will be more ambiguous than that of one resulting from an
acquisition as, with the latter, it is reasonable to assume that the acquiring organization’s
identity will dominate over that of the acquired one. As a result, pre-existing institutional
trust may be more likely to be undermined in mergers than in acquisitions, and in the
case of an acquisition the institutional trust of employees from the acquiring legacy
organization may not be undermined at all, although there may still be a problem with
institutional trust among members from the acquired legacy organization, especially if
they do not identify with the acquiring organization. This proposition could be tested.
Similarly, the identity of an organization resulting from a merger of firms of equal size
will likely be characterized by more ambiguity than that of an organization resulting
from the merger of firms of very different size as, with the latter, it is reasonable to assume
that the identity of the much larger firm will dominate. As a result, pre-existing institutional
trust may be more likely to be undermined in mergers of firms of equal rather than
very different sizes, and it is possible that the institutional trust of employees from the
larger legacy organization may be eroded to a less degree than that of employees from
the smaller legacy organization. This testable hypothesis is consistent with case study (van
Knippenberg et al., 2002) and experimental (van Leeuwen et al., 2003) findings suggesting
that the rejection of post-merger organizational identity is moderated for individuals
from the legacy organization that emerges as dominant, and extends this work by
pointing to the specific mechanisms involved. In summary, our study, by showing that
institutional trust can be identity-based and thus undermined by organizational identity
ambiguity or organizational members’ absence of identification, lays the groundwork for
interesting future research
SOLUTION
Theoretical Development and Contribution of the Paper
In this academic paper which studies the case of Citibank done by Steve Macguire and Nelson Phillips they have tried to capture the main crux of loss of institutional trust whenever organisations go for a merger. They have noticed that the literature studies and those too empirical studies on the institutional trust after mergers are very less. AT the same time they have tried to create a framework by linking organisational identity with the role of individual identification process which helps in creating an area of future work and research on this matter.
The author has not been able to put forth all the latest literature and researches done on this topic because it has been found in several latest literatures that post-merger integration issues have been receiving greater attention in relation to identity related phenomenon (Misra, 2008) (Kroon, Noorderhaven and Leufkens, 2009). In fact Zaherer et al (Zaheer, Shomaker and Genk, 2003) has even noticed that organisational identity matter arising between the two merging organisations are more effective that the cultural issues which are being highlighted while trying to understand the problems arising during integration.
The paper does try to bring into light a new and fresh managerial phenomenon which says that if the mergers and the issues related to mergers are handled properly there would not be loss of valuable talent and the morale and productivity of the employee will also not deteriorate at the same time( DePamphilis,2001). Some researchers have tried to take a closer look at the human cost which the mergers incur in terms of conflict, loss, job insecurity , stress and identity disruption (Seo and Hill, 2005).
Paper’s Development of empirical Evidence
The paper is certainly based on the case of Citibanker at the CitiGroup and talks about its merger with Travellers Group to form the Citigroup. This case was called merger but later it was discovered that although technically it was called merger but it was famously referred to as acquisition. Moreover during this research the case study of just one single organisation is used so generalisation=s cannot be done promptly. Moreover the interview of the citibankers is just post merger so the case study is not much relevant to the whole research.
The study or research is based on the surveys or interviews conducted on the Citibankers are totally based on the post-merger interviews and not on the pre-merger interviews. Thus there is possibility that the views given by the Citibankers might be affected or influenced by the pre-merger picture of Citibank and they might be a bit biased. The questions which were asked to the respondents are quite effective and rational as helpful in assembling the evidences to draw conclusions regarding the institutional trust and organisational identity.
Although this case study has given relevant and significant insight into the mergers but what about acquisitions and generalising the views for other sectors mergers will not be practical taking the data and eth analysis from this particular case study. However it has been noticed that the data has been analysed in perfect manner and that’s how the conclusions have been established related to the institutional trust and organisational identity. Thus finally from the data analysis and case study we can say that institutional trust is something which is based on identity and it gets undermined by the individuals ;loss of identification or organisational identity.
Contribution to better management practices
This paper clearly signifies the importance of organisational identity and social identity
(Hogg and Terry, 2000) for organisations in case of mergers. It has been shown in other researches too that the post-merger identification does have major consequences and impact on the organisational commitment of the employees which cannot be explained in terms of economic concerns rather they are concerns like uncertainty and job insecurity (Ullrich, Wieseke and Van Dick, 2005). The study suggests that mergers do pose identity challenge as they create impact on the group status which is being perceived in the organisation.
However the paper does not bring forth the identity challenges or loss of institutional trust in other sectors of business. It also lacked to identify the challenges which are being faced due to mergers like pressure for sameness, changed group status and isolation.
There are some very important facts and relationships which have been indicated by the research for the organisations, society and management. It has thrown a deeper insight into the institutional trust through the ramifications for organisations. It has also helped in developing an identity-based framework which is very helpful in creating along with maintenance and loss of institutional trust in relation to any organisation. It will also help in finding out the mechanism which is responsible for the loss of trust in relation to institution post-merger. It has also helped in developing a relationship with the organisational identity.
Bibliography
Hogg, M.A. and Terry, D.J. (2000) ‘Social identity and self-categorization processes in organizational contexts’, Academy of Management Review, vol. 25, no. 1, pp. 121-140.
Kroon, D., Noorderhaven, N. and Leufkens, A. (2009) ‘Organizational identification and cultural differences: Explaining employee attitudes and behavioral intentions during postmerger integration’, Advances in Mergers and Acquisitions, vol. 8, pp. 19-42.
Misra, S.&.S.M.A. (2008) ‘In the aftermath of an acquisition: Triggers and effects on perceived organizational identity’, Journal of Applied Behavioral Science, vol. 44, no. 4, pp. 422-444.
Seo, M.G. and Hill, N.S. (2005) ‘Understanding the human side of merger and acquisition: An integrative framework’, Journal of Applied Behavioral Science, vol. 41, no. 4, pp. 422-443.
Ullrich, J., Wieseke, J. and Van Dick, R. (2005) ‘Continuity and change in mergers and acquisitions: A social identity case study of a German industrial merger’, Journal of Management Studies, vol. 42, no. 8, pp. 1549-1569.
Zaheer, S., Shomaker, M. and Genk, M. (2003) ‘Identity versus culture in mergers of equals’, European Management Journal, vol. 21, no. 2, pp. 185-191.
JA95
But you can order it from our service and receive complete high-quality custom paper. Our service offers “Management” essay sample that was written by professional writer. If you like one, you have an opportunity to buy a similar paper. Any of the academic papers will be written from scratch, according to all customers’ specifications, expectations and highest standards.”