Content uploaded by Laura Empson
Author content
All content in this area was uploaded by Laura Empson on Jun 20, 2019
Content may be subject to copyright.
1
Empson, L,, & Alvehus, J. (2019) “Collective leadership dynamics among professional peers:
Co-constructing an unstable equilibrium”, Organization Studies. DOI:
10.1177/0170840619844291
Collective leadership dynamics among professional peers:
Co-constructing an unstable equilibrium
Laura Empson
Cass Business School
City, University of London
106 Bunhill Row
London EC1Y 8TZ, United Kingdom
Tel: +44 (0)20 7040 5172
e-mail: laura.empson@city.ac.uk
Johan Alvehus
Department of Service Management and Service Studies
Lund University
Universitetsplatsen 2
SE-25225 Helsingborg, Sweden
Tel: +46 709 740474
e-mail: johan.alvehus@ism.lu.se
Acknowledgements
We would like to thank the following scholars who have made a substantial contribution
to the development of this paper: Ann Langley, Vivian Sergi, David Collinson, and Mary
Uhl-Bien, also our editors Trish Rea and David Arellano-Gault, and three anonymous
reviewers.
Funding
This work was supported by the Economic and Social Research Council, grant number RES-
062-23-2269.
2
Collective leadership dynamics among professional peers:
Co-constructing an unstable equilibrium
Abstract
Professional service firms (PSFs) are characterised by contingent and contested power
relations among an extended group of professional peers. Studies of such firms can therefore
yield important insights for the literatures on collective leadership and leader–follower
relations. Yet to date PSF scholars have neglected the topic of leadership, and leadership
scholars have neglected the context of PSFs. Based on 102 interviews across the consulting,
accounting, and legal sectors, we identify three relational processes through which
professional peers co-construct collective leadership: legitimising, negotiating, and
manoeuvring. We demonstrate how the relational processes taken together constitute an
unstable equilibrium, both in the moment and over time, emphasising how leadership in PSFs
is inherently contested and fragile. Our model contributes to theories of collective leadership
and leader–follower relations by foregrounding the power and politics that underlie collective
leadership. We highlight the significance of the individual leader within the collective. We
challenge assumptions concerning the binary nature of leadership and followership, by
showing how colleagues may grant leadership identities to their peers without necessarily
granting them leadership authority, and without claiming follower identities for themselves.
Key words: Collective leadership, Leader–follower relations, Contingent authority, Unstable
equilibrium, Leadership dynamics, Politics, Professional service firms.
3
Introduction
The leading role that professionals, such as accountants, lawyers, and consultants, play in
shaping societal, economic, and governmental institutions has been well-documented in this
journal and elsewhere (Scott, 2008). Yet recent years have been marked by intense public
criticism of the leaders of professional service firms (PSFs1) for their failure to prevent the
global financial crisis and the various corporate scandals that preceded and followed it (e.g.
Mueller, Carter, and Whittle, 2015). While leaders of PSFs have been extensively criticised,
very little is known about how leadership is actually achieved in PSFs. To date, PSF scholars
have neglected the topic of leadership and leadership scholars have neglected the context of
PSFs (Empson and Langley, 2015).
PSFs are characterised by contingent and contested power relations among an extended group
of professional peers; in the largest firms this group may comprise several hundred partners
(Greenwood and Empson, 2003). These senior professionals claim extensive autonomy and
grant leadership authority to their colleagues on a contingent basis (Empson and Langley,
2015), resulting in highly politicised internal competitions for power, protracted processes of
consensus-based decision making, and failures to execute decisions once they have been
agreed (Hinings, Brown, and Greenwood, 1991; Lawrence, Malhotra, and Morris, 2012;
Morris, Greenwood and Fairclough, 2010). Conventional leadership studies, predicated as
they are on clearly defined roles and hierarchical relationships between leaders and followers,
therefore fail to capture the complex power relations and collective nature of leadership which
is characteristic of PSFs (Empson, 2017).
4
However, two recent and related developments in leadership research, the emerging
literatures on collective leadership and leader–follower relations, represent useful lenses
through which to develop a more nuanced understanding of leadership dynamics among
professional peers. Scholars of collective leadership (such as Cullen and Yammarino, 2014;
and Ospina, Foldy, Fairhurst, and Jackson, 2018) focus on the process of leadership as co-
constructed among an extended group of colleagues, rather than the actions of individual
leaders alone. Empirical studies of collective leadership “emphasize processual ‘how’
questions, aimed at understanding how leadership is produced and performed” (Denis,
Langley, and Sergi, 2012: 255). Consistent with this approach some scholars of leader–
follower relations (Chreim, 2015; DeRue and Ashford, 2010; Uhl-Bien, Riggio, Lowe, and
Carsten, 2014) add an important perspective to studies of collective leadership by
emphasising that leadership and followership reflect a “relational process co-created by
leaders and followers in context” (Fairhurst and Uhl-Bien, 2012, p. 1024). Problematising the
conventional leadership literature’s focus on the primacy and agency of individuals in
leadership positions, Uhl-Bien and Pillai (2007) state, “if leadership involves actively
influencing others, then followership involves allowing oneself to be influenced” (p. 196).
In building our analysis we draw together these bodies of literature. In so doing, we address
some of their key limitations. First, the literature on PSFs focuses on governance at a
relatively abstract and depersonalised level and is generally silent on the subject of leadership
dynamics (Empson and Langley, 2015). Second, the collective leadership literature presents
an unproblematised view of power relations so there is a “need to pay greater attention to the
underlying pattern of power relationships” (Denis et al., 2012: 270) that affect the interactions
and outcomes of leadership processes; as Bolden (2011: 260) argues, much current research
on collective forms of leadership “takes insufficient consideration of the dynamics of power
5
… within which it is situated”. Third, the leader–follower relations literature tends to assume
positions of leader and follower are dichotomised and stable (Collinson, 2005; Chreim, 2015;
Fairhurst, 2016). There is therefore a need for more research that explores how individuals
navigate these relational dynamics, both collectively and individually.
Based on a qualitative study of 102 senior professionals in three accounting, consulting and
law firms, we ask: How is collective leadership co-constructed among professional peers? To
explore this broad question we begin by asking, 1) What are the dynamics through which
professional peers co-construct collective leadership? We use cross-sectional analysis of
interviews to identify the relational processes and underlying power relations that mutually
constitute collective leadership. Identifying relational processes is an important first step but
to focus on them in isolation risks ignoring or misrepresenting their interplay over time. We
therefore go on to ask, 2) How do changes in underlying power relations among professional
peers serve to stabilise and destabilise collective leadership dynamics over time? We address
this second question through a longitudinal case analysis of one firm over a five-year period
The paper proceeds as follows. We begin by examining what is already known about
leadership and power in PSFs and then highlight key issues in the literatures on collective
leadership and leader–follower relations which merit particular attention in the context of our
study. We outline the research design and explain our process of analysis. We go on to
identify and describe the three interdependent relational processes by which collective
leadership is co-constructed among professional peers (research question 1). We then focus
on the interplay of these relational processes over time (research question 2). We illustrate
how the destabilisation of any one of the relational processes will likely destabilise the
6
collective leadership dynamics as a whole, and argue that the relational processes taken
together constitute an unstable equilibrium, both in the moment and over time.
In addition to representing a novel study of leadership in PSFs, the model we develop
contributes to current leadership theories on collective leadership and leader–follower
relations. We identify and explicate the power relations and political behaviours underlying
collective leadership dynamics, and identify the significance of the individual leader within
collective leadership. We challenge taken for granted assumptions concerning the binary
nature of leadership and followership within the leader–follower relations literature by
showing how colleagues may grant leadership identities to their peers, without necessarily
granting them leadership authority, and without claiming follower identities for themselves.
Leadership and power in PSFs
PSFs typically adopt a partnership form of governance, with ownership and profits shared
among professional peers (Greenwood and Empson, 2003). Whereas many PSFs have
abandoned partnership as a legal form, they often continue to emulate its organisational
characteristics (Empson, 2007; Pickering, 2015). Chief among these characteristics are
extensive autonomy and contingent authority (Empson & Langley, 2015), which give rise to
distinctive power relations among professional peers.
Experienced professionals require, or at least expect, extensive autonomy to carry out their
work (Thomas and Hewitt, 2011). The customised nature of professional work means that
professionals are afforded considerable autonomy to make finely tuned judgements about how
best to apply their expertise to client work (Freidson, 2001). While performance pressures and
7
regulatory change have to some extent constrained the autonomy of professionals (Cooper,
Hinings, Greenwood, and Brown, 1996; Smets, Morris, von Nordenflycht, and Brock, 2017),
the partnership form of governance has persisted, thus institutionalising senior professionals’
expectations of autonomy (Alvehus, 2017; Empson, Muzio, Broschak, and Hining, 2015;
Robertson and Swan, 2003).
With regard to contingent authority, a senior executive in a PSF “maintains power only as
long as the professionals perceive him or her to be serving their interests effectively”
(Mintzberg, 1989: 181). In PSFs most professionals are selected, and often elected, by their
peers to formal leadership positions for a fixed term and can be deposed if they fail to retain
their support (Empson, 2007; 2017; Greenwood, Hinings and Brown, 1990). Authority is
“collegial and fragile” (Hinings et al., 1991) and deemed to rest with the professional peer
group rather than any individual.
Power relations in PSFs are therefore contingent and potentially contested among peers. The
extent to which power relations in PSFs differ from conventional hierarchical organisations
can be highlighted by applying French and Raven’s (1959) classic conceptualisation of the
bases of power, originally developed as a typology of power relations in conventional
hierarchical organisations. In PSFs “expert” power does not constitute a basis of power, as all
partners are experts. “Formal” power is typically constrained by the contingent nature of
authority. Similarly “reward” and “coercive” power are limited because professionals in
senior leadership positions often lack the power to set the pay of their peers2 and are
prevented by complex internal procedures from being able to dismiss them (Empson, 2017).
“Referent” power, however, represents a highly significant source of power among senior
professionals (Empson and Langley, 2015) as professionals who role model desired
8
behaviours gain credibility, and with it power, to demand the same of their peers (Muhr,
2011).
Because of contingent and contested power relations in PSFs, “leadership is a matter of
guiding, nudging, and persuading” (Greenwood et al., 1990: 748) and “is typically a
collective endeavor” (Empson and Langley, 2015: 179). PSFs are, therefore, “uniquely
political environments”; “many traditional professional partnerships are consensus-based
democracies but, as such, are subject to the lobbying, scheming, and bargaining which occur
in any other political arena to achieve agreement” (Morris et al., 2010: 297). In order to build
consensus, professionals in leadership positions need to display the full range of political
skills (Empson, 2017), combining networking ability and social astuteness with interpersonal
influence and apparent sincerity (Ferris et al., 2007). In such a context political behaviour
represents an essential means of achieving consensus (Kirk and Broussine, 2000), rather than
a negative and illegitimate pursuit of self-interest, as conventionally depicted within the
leadership literature (Ammeter, Douglas, Gardner, Hochwarter, and Ferris, 2002; Silvester,
2008).
Given the significance of PSFs to the global economy, and high-profile examples of
leadership failure (Empson et al., 2015), it is problematic that the PSF literature is largely
silent on the topic of leadership (Empson and Langley, 2015). The PSF literature analyses
governance as an archetypal organisational form at the level of structures, systems and
interpretive schema (Cooper et al., 1996; Greenwood et al., 1990; Greenwood and Empson,
2003) and is therefore somewhat abstract and impersonal. When PSF scholars do address the
topic of leadership it is to represent it somewhat simplistically, for example by referring to it
dismissively as “cat herding” (von Nordenflycht, 2010).
9
Collective leadership and leader–follower relations
Conventional leadership research is predicated on clearly defined hierarchical relationship
between leaders and followers and is ill-suited to studying contested and contingent power
relations among professional peers. Since mainstream leadership and PSF researchers both
have little to say about leadership in PSFs, we must turn to emerging bodies of leadership
research to help frame our study of leadership dynamics in PSFs: specifically the literature on
collective leadership and leader–follower relations.
In the past decade the study of leadership has moved beyond heroic, trait-based studies of
individuals in conventional hierarchical organisations (Bligh, Kohles and Pillai, 2011;
Collinson, Grint, and Smolovic-Jones, 2018) to explore this phenomenon of collective
leadership. Scholars within this emerging field view leadership “not as a property of
individuals ... but as a collective phenomenon that is distributed or shared among different
people, potentially fluid, and constructed in interaction” (Denis et al., 2012: 212), therefore
emphasising leadership processes and interactions. A variety of overlapping terms have
proliferated to describe this broad phenomenon, including collective (Ospina et al., 2018),
plural (Denis et al., 2012), and relational (Uhl-Bien and Ospina, 2012), as well as related
concepts such as shared (Pearce and Conger, 2003), distributed (Gronn, 2002), and
complexity leadership (Uhl-Bien, Marion, and McKelvey, 2007). In the context of PSFs, the
most pertinent form of collective leadership is that described by Denis et al. (2012: 231) as
“pooling leadership at the top to direct others”.
In pluralistic organisations, where objectives are divergent and power is diffuse, collective
leadership is often institutionally prescribed as a means of accommodating multiple
10
perspectives and managing internal political relationships (Denis et al., 2001; Reid and
Karambayya, 2009). No single individual is vested with power to lead unilaterally; it is only
though the collective interactions of multiple individuals in leadership positions that
leadership can be accomplished (Empson and Langley, 2015). The limited number of studies
of this “pooling” form of collective leadership have focused on leadership dyads, triads, and
constellations. They typically take the formal authority of those in leadership positions for
granted, neglecting the extended network of informal power relations that underlie the
leadership dynamics (Chreim, 2015; Denis et al., 2001; Denis et al., 2012). This is potentially
problematic in the context of PSFs, where the extended leadership group may encompass
several hundred professional peers, who may or may not occupy formal leadership positions
but are nevertheless involved in the process of the co-construction of leadership through their
informal political interactions and formal involvement in the election of their peers (Empson,
2017).
Thus, given the contested and contingent nature of power relations among professional peers,
our study of leadership in PSFs necessarily involves addressing issues of power, and doing so
in a way which captures the subtlety and nuance in how it is exercised. It therefore provides
an opportunity to address the collective leadership literature’s unproblematised view of power
(Denis et al., 2012), by examining power relations among an extended group of peers. Our
study recognises that, whereas leadership may be represented by formal positions,
professionals who occupy such positions will have been placed there by their peers and are
entirely dependent on colleagues’ on-going support to perform their role.
11
In parallel, and consistent with, the emerging literature on collective leadership, leadership
scholars’ increased interest in followership has shifted focus from individual leaders to those
supposedly under their influence (Bligh et al., 2011). As Uhl-Bien et al. (2014: 83)
emphasise, it is “accepted wisdom that there is no leadership without followers.” Yet, in the
past, followers “have been viewed as unproblematic and predictable cogs in the (leadership)
machine” (Collinson, 2005: 1424) and many followership studies adopt a conventional
positional approach to leadership and followership. However some scholars emphasise that
leadership and followership are not fixed and immutable positions and that followership is co-
created through a process of interaction between leaders and followers (Uhl-Bien et al., 2014).
For example, DeRue and Ashford’s (2010) theoretical model proposes that leader–follower
identities are co-constructed through the relational processes of an individual coming to see
themselves, and being seen by others, as a leader or follower (the claiming and granting of
leader and follower identities). Given the contingent and contested nature of power among
professional peers, this perspective on leader–follower relations represents a useful lens for
studying leadership dynamics in PSFs. While recognising that leadership and followership are
co-constructed through relational processes, most leader–follower studies nevertheless
assume that concepts of leadership and followership are not only co-dependent but mutually
exclusive. Some scholars argue, therefore, that many studies of leader–follower relations
perpetuate the overly oppositional binaries of leadership and followership common in
conventional leadership studies (Fairhurst, 2001; Collinson, 2005).
Collinson (2006: 1436) instead argues that leader and follower identities and power relations
are not clear-cut and homogeneous but “blurred, multiple, ambiguous, and contradictory”. He
suggests that followers are “knowledgeable agents ... proactive, self-aware and knowing
12
subjects who have at their disposal a repertoire of possible agencies within the workplace”
(Collinson, 2005: 1422). In this way some leader–follower studies potentially overcompensate
in their critique, ascribing a disproportionately high degree of agency to followers, and
neglecting the significance of formal authority and leadership positions. As Alvehus (2018:
17) emphasises, leadership “is not an organizational quality shared democratically, but a
question of contestation and negotiation, depending on power asymmetries.”
Regardless of what approach is taken, research on leader–follower relations, therefore,
typically excludes, or at least does not sensitise us to, more subtly interpersonal political
processes amongst colleagues. By examining in detail how peers claim and grant leadership
identities and authority to each other on an on-going basis, our study represents an
opportunity to examine the underlying power relations involved in the co-construction of
collective leadership dynamics and to interrogate a foundational assumption of leadership
research that there can be no leadership without followership.
Research design
Numerous scholars have emphasised the need for qualitative studies that “capture the truly
dynamic qualities of leadership” (Gordon and Yukl, 2004: 364), which cannot be studied
through the survey-based methods that dominate the leadership literature (Bryman, 2004).
Whereas observation is a popular qualitative method adopted by many practice-based studies
of leadership, it is not well suited to studying the more political aspects of leadership, which
may involve the withholding or concealment of actions. Accordingly, we developed an
interview-based research design that drew on current developments in grounded theory (e.g.
Charmaz, 2014). Our research questions evolved as we gathered data and developed
interpretations, and our structured approach as outlined below helped to safeguard against
13
interpreter bias and facilitated “systematic data collection in order to develop theories that
address the interpretive realities of actors” (Suddaby, 2006: 634).
Sampling and data collection
As part of a broader research project, the first author conducted 102 interviews in fifteen
countries with senior professionals in global PSFs in the accounting, consulting and legal
sectors. The project studied leadership in large, global PSFs across a range of sectors, to avoid
focusing on idiosyncratic sector-specific dynamics. Interviewees were selected from firms
that were among the largest and most preeminent in their respective sectors. Roughly equal
numbers of interviews were conducted in each firm (34, 31, and 37 in Firms A, B, and C
respectively). The firms ranged in size from 200 to 600 senior professionals/partners and
generated revenue ranging from US$ 750m to US$ 2,250m. All were organised as
partnerships, though one was legally a corporation and “mimicked” a partnership (Empson,
2017). Each, therefore, had several hundred senior professionals sharing in the ownership and
profits and electing their senior leaders for a fixed term.
Preliminary interviews were conducted with the two most senior executives (Chairman and
CEO or Senior and Managing Partner), alongside reviews of internal documentation (e.g.
partnership agreements, minutes of board meetings, transcriptions of partner meetings), to
identify formal structures of governance. Thereafter a “snowball” method of sampling
(Heckathorn, 1997) was employed to identify individuals recognised by each other as being
involved in the senior leadership group. This was typically around 30 professionals within
each firm, who might or might not have formal leadership positions but were recognised as
highly influential. Interviewees included: heads of practices, geographic regions, market
14
sectors, and functional areas. The interview phase was concluded when the same names were
consistently identified as part of the senior leadership group and no new names emerged3.
Interviews typically lasted for about 90 minutes and were recorded and transcribed verbatim.
Interviews explored professionals’ perspectives on leadership, asking how they interpreted
and engaged in leadership within their firms. Lines of questioning began broad but became
more focused. Example questions included: “What does the concept of leadership mean to
you in the context of [Firm]?”, “Do you consider yourself a leader of [Firm]?”, “Please
explain why you think that?”. Interviewees were then asked to identify colleagues they
considered part of the senior leadership group and to explain their reasoning.
Data analysis
The following description presents the highly complex and iterative analysis process in a
simplified and linear form – the process was inevitably “messy” and “nonlinear” (Suddaby,
2006: 637). The analysis took two distinct forms: cross-sectional analysis of all interviews
and longitudinal analysis of a single case. The cross-sectional analysis enabled us to address
our first research question and develop our model of collective dynamics among professional
peers (Figure 2). Having identified the leadership dynamics, we then explored them in depth
through a longitudinal case analysis of a single firm. This enabled us to address our second
research question, by identifying how changes in the underlying power relations among
professional peers serve to stabilise and destabilise the collective leadership dynamics over
time (Figure 3).
15
Our cross-sectional analysis followed a three-stage process of theoretically informed coding
(Charmaz, 2004): focused codes (stage 1), from which we developed theoretical codes (stage
2), which we then integrated into aggregate dimensions (stage 3). Throughout the analytical
process, the primary coder (first author) checked, challenged, and discussed the coding with
the second author (see Figure 1 for a representation of our coding, which includes illustrative
quotes.)
-------------------------------
Insert Figure 1
-------------------------------
Initially almost 100 focused codes and sub-codes were developed, covering themes such as:
Who are the leaders in the firm? Why are they seen as such? How has this changed over time?
Following Golden-Biddle and Locke’s (1993) guidelines, the focused coding enabled us to
identify a strong theoretical plot: the relational processes and power relations that underlie
collective leadership. We then refined the focused codes into theoretical codes. By drawing
on works such as Collinson (2005, 2006) and DeRue and Ashford (2010), we shifted our
emphasis from the actions of individuals (e.g. “enabling autonomy while maintaining
control”) to interactions among peers (e.g. one professional “asserting control” while others
see themselves as “exercising autonomy”). Through this we identified three aggregate
dimensions: the relational processes of legitimising, negotiating, and manoeuvring. For
example, “asserting control” and “exercising autonomy” came together in the relational
process “negotiating”. The interplay among these three relational processes is represented in
our model (see Figure 2).
For the longitudinal case analysis we focused on one particular firm in our study because,
very unusually, it had three Managing Partners over a five-year period. This suggested it
16
would provide a particularly clear illustration of how changes in underlying power relations
among professional peers serve to stabilise and destabilise collective leadership dynamics
over time. We started by creating a time line based on interviews and formal documentation,
and identified the main actors. We then analysed the events, actions and reactions of key
actors in terms of our model (represented by Figure 3). For example, we looked at how the
first Managing Partner was initially perceived as exerting excessive control, which created
opportunities for another partner to rise to power by political actions. We also analysed the
reactions among the firm’s partners, for example in response to the second Managing
Partner’s relaxation of control. Step-by-step we analysed the interplay of the three relational
processes (i.e. legitimising, negotiating, and manoeuvring) over a five-year period. In this
way, we demonstrated the value of the empirically derived model in understanding changing
power relations and the destabilisation and stabilisation of leadership dynamics over time.
Co-constructing collective leadership dynamics among professional peers
-------------------------------
Insert Figure 2
-------------------------------
Our empirically derived model (Figure 2) identifies three distinct and interrelated relational
processes: legitimising, negotiating and manoeuvring. Taken together they represent the
leadership dynamics by which collective leadership is co-constructed among professional
peers. We use the term “legitimising” to express the relational process whereby professionals
develop a leader identity, by succeeding in the market while their peers infer that they have
leadership ability. However, to be granted not just a leadership identity but leadership
authority, professionals must engage in two further relational processes: negotiating and
manoeuvring. We use the term “negotiating” to express the relational process whereby
17
professionals assert control while their colleagues exercise autonomy. This is an overt
interaction as it involves professionals contesting manifestations of formal authority. By
contrast, we use the term “manoeuvring” to express a more covert relational process, whereby
professionals seek to exert informal influence over their peers by behaving politically while
their colleagues perceive them as having integrity. As explained in detail in the following
three sections, and explored through the subsequent case analysis, these relational processes
express three different aspects of the contested and contingent power relations within PSFs,
whereby professional peers act and react to each other by claiming and/or temporarily
granting leadership and followership among themselves.
Legitimising: Succeeding in market and inferring leadership ability
To be accepted by colleagues as a leader you’ve got to show you are big in the market
and can win work. (C364)
Success in the market is not a direct manifestation of leadership ability but a necessary
precondition for professionals to gain the legitimacy to lead (Table I). An individual
professional’s success in the market causes them to stand out from their peers and be seen as a
sufficiently credible role model, thus increasing respect and signalling a shift in underlying
power relations.
You can’t be a leader without having credibility as a practitioner in this firm … You
have to have a demonstrated track record of having delivered. (B27)
However, as another interviewee says, this “doesn’t necessarily and mean you’re a good
leader” (A16).
18
------------------------------
Insert Table I
------------------------------
In this context, market success is characterised as a professional’s ability to generate work for
themselves and their colleagues, to be “outstanding with clients” (C7) and to “enhance the
value of the partnership” (C14); “everything else is icing on the cake” (A4). “If you bring in
lots of business you will always generate respect.” (A13). One interviewee explains:
I’ve been the partner on [major client] having won that … And, even now I’ve become
head of [major practice], I’m just going onto [major client]. … So if I put my little ego
hat on for a moment I was sort of, you know, seen to be a good partner. You know, one
of the top partners, I suppose, and one that can get out there and win work. (C18)
After being raised to a leadership position, professionals must continue to demonstrate that
they are succeeding in the market, to show that they can “still cut it” (A11), to ensure their
colleagues continue to infer that they have leadership ability.
Joe, who’s very, very good, made a classic mistake of cutting right back on his practice
and becoming full-time management. And that doesn’t work in a firm like ours. You do
lose credibility doing that. (A11)
As the case analysis will demonstrate, once professionals decide that their colleague is no
longer succeeding in the market, they may conclude that he or she has lost the legitimacy to
lead them, prompting a shift in power relations and destabilising the collective leadership
dynamics overall.
Negotiating: Asserting control and exercising autonomy
19
It’s not about following in that sense. It’s about leaders enabling and directing, giving
people outlets. … Because frankly nobody has to follow anyone. (C9)
Interviewees consistently claim to be autonomous and resist any sense that they are being
controlled by colleagues whom they perceive to be their peers (Table II). This perception is
confirmed by professionals in leadership positions. As an interviewee explains:
You can’t really tell people what to do. You can say what you’re going to do and then
hope people will agree with it … and the people you can least tell what to do are those
who are most important for the success of the business. Because they are the ones who
control the client relationships. (A17)
------------------------------
Insert Table II
------------------------------
To achieve their objectives, therefore, leaders must negotiate with powerful peers when
seeking to impose formal controls. Professionals acknowledge the latitude they have in
relation to individuals in leadership positions. The CEO and other senior colleagues “are
really coordinating or communicating … but definitely not controlling, not managing, not
policing” (B29). This can be frustrating:
At times I’d just like to say, “You know something, just go and do it, because we
haven’t got time to debate it just do it.” But that’s not our culture. (C27)
Yet, professionals in leadership positions do exercise a degree of control, but in a context
where leadership “sort of happens” (A7), they must demonstrate an acute understanding of the
20
underlying power relations. They must “walk a tightrope” (B32) between their need to assert
control and their colleagues’ desire to exercise autonomy. As one Senior Partner explains:
Partners say “you’re too tight” and they say “get looser”. So you get looser and they say
“it’s chaotic, get tighter”. If the money is going up, you can do what you like. If the
money is going down, you can’t do anything. But the money going up or down isn’t
within control of the Senior Partner. (C1)
As the case analysis will demonstrate, if a professional in a leadership position is “too tight”
in asserting control and peers feel overly constrained in exercising autonomy, or he or she is
“too loose” and peers start to press for greater controls on colleagues autonomy, power
relations will shift and the collective leadership dynamics will become destabilised.
Manoeuvring: Behaving politically and perceiving integrity
I think the political types in this firm are playing too many games, getting too clever for
their own good, constantly thinking about tactics. But I do that all the time with my
clients…. My last transaction, there were seven different stakeholder groups, all with
complex demands, all wanting their own way, and my job was to get everybody across
the line without them punching each other … I’m really good at working out
strategically how to move something forward. I feel passionately about the firm and I
think there is a difference between me and the real political game players here who are
not always doing it for the common good. (C32)
This quote captures the complexity of the third relational process, manoeuvring, and the
power relations that underlie it. To be accepted by their peers as leaders, professionals must
21
engage in subtle political behaviour, while their peers perceive them to be apolitical and
behaving with integrity. Table III presents this manoeuvring as a set of distinct but
intertwined beliefs and behaviours.
------------------------------
Insert Table III
------------------------------
Overt political behaviour (Table III:1). The interview guide did not explicitly address
politics, yet many interviewees raised the topic unprompted. Interviewees express abhorrence
of apparent political behaviour, e.g. “I hate politics” (B12), colleagues “are not in the business
of politics” (B32). A professional’s reputation can be damaged if they are seen by colleagues
to be politically “pushy” or “ambitious” (A2), or “leading for their own advantage, just
grabbing opportunities for themselves” (A26); “people get found out” (C32). Interviewees
typically describe politics in terms of behaviours that are “self-serving” (C27).
By “too political” what I mean is someone who looks like they are constantly
manoeuvring for their own benefit as opposed to the wider firm benefit. (C32)
Yet, as interviewees’ comments reveal, partners adopt overtly political language: “It’s a bit
like being part of a political party” (C35). Politics is integral to their firm’s leadership
dynamics and is enshrined within their governance structures. To be admitted to partnership
professionals must be elected by the partners. Thereafter elections are held for senior
leadership positions, requiring candidates to issue “manifestos”, give speeches at “candidates’
debates”, and engage in “run-off elections”. Senior and Managing Partners refer to their
“electorate” or “constituents” who provide them with “mandates”. “We have our constituents
… we have to help them, be seen to be helpful” (A34). So, whereas abhorring what they view
22
as politics (i.e. self-serving behaviours) professionals are dependent upon political behaviours
for the effective functioning of their firm’s leadership dynamics.
Subtle political behaviour (Table III:2). Once elected to leadership positions, professionals
cannot simply represent the wishes of their electorate but must pay attention to their most
powerful peers – to “build a consensus amongst a key group of people” (A17). They need to
make trade-offs between competing interest groups and offer incentives to colleagues in
private to persuade them to lend their support in public. Interviewees talk of “promises being
made” behind “closed doors” in “smoke filled rooms” (C6).
The following two quotes from the same individual demonstrate how professionals in
leadership positions need the political skill to judge how to deal with different colleagues in
different contexts – knowing when to treat them like “adults” and when to challenge them for
childish behaviour:
This is about dealing with adults, very bright adults and consequently I think the
consummate leadership skill in this firm [is] the ability to drive change through
influence rather than by telling. (C27)
I spend the vast bulk of my time just trying to resolve internecine conflicts between
partners … To be able to hold up a mirror and say, “Guys, you look like two kids
scrapping in the playground. Is that really what we want? I mean we’re all partners, for
goodness sake.” (C27)
Perception of integrity (Table III:3). To continue to grant a colleague leadership authority,
professionals must to perceive that a colleague is acting for the “common good” (C32).
23
The partnership as a whole trusts our leader, that he’s going to resolve it for the best –
for the good of the partnership. As long as they believe that and trust that, then there’s
no reason to put any other kind of hold on him. (B23)
Professionals in leadership positions must be seen by their colleagues to be “genuine and
clean” rather than “creeping” and “slightly sinister” (A19); “[the firm] is stamped on their
heart if you open them up” (AI2). They need to be able to “come in on [a] white horse” (B5),
to be recognised as “trying to do the right thing” (C12), and “not appear to be having to work
politically to keep their position” (C4). One Senior Partner is praised by his colleagues for
being “very modest, quite self-effacing” and not being “a player” (A19). Of course a
professional such as he, operating in a highly-politicised environment, may in fact be
deploying highly sophisticated political skills – whether utilising them consciously or not, or
to fulfil his ambitions for the firm of for himself.
The one who does [leadership] more naturally, partners will recognise that, and he is
more likely to sustain the role. And the others are more likely to crash and burn. (C4)
If colleagues become aware of colleagues’ political behaviour and begin to question their
integrity, they may no longer be willing to grant those individuals leadership authority over
them.
We have, therefore, addressed our first research question by identifying the three relational
processes through which collective leadership dynamics are co-constructed among
professional peers and the power relations that underlie them. Power is most explicit within
the relational process of negotiating. Formal power is expressed as professionals in leadership
positions seek to assert control while their colleagues seek to exercise autonomy. The
24
relational process of legitimising illustrates the importance of referent power in PSFs. When a
professional succeeds in the market, colleagues may start to see them as role model and, in so
doing, be willing to elevate them to leadership status. Power is less apparent but nevertheless
present within the most covert relational process of manoeuvring, whereby an individual acts
politically while their peers infer that they have integrity. Because professionals’ formal
power is so circumscribed in this context, political behaviour becomes a necessary component
of leadership through exerting informal influence. But, as the case analysis will demonstrate,
once professionals recognise a colleague is behaving politically, they will no longer infer that
her or she has integrity, leading to a shift in power relations and destabilising the collective
leadership dynamics overall. In the following section, we address the second research
question and explore how changes in underlying power relations among professional peers
serve to stabilise and destabilise collective leadership dynamics over time.
Collective leadership dynamics in a PSF: an unstable equilibrium
This case analysis explores leadership dynamics as they unfold in a specific PSF, a firm
comprising more than 400 partners.5 We demonstrate how the complex interplay of the three
relational processes, and the changing power relations that underlie them, serve to destabilise
and stabilise the firm’s collective leadership dynamics over a five year period (Figure 3). The
following, which represents an inevitably simplified and reductive narrative of a complex and
nuanced process, is presented in four distinct phases, each representing a period of
stabilisation or destabilisation. It demonstrates how these leadership dynamics represent an
unstable equilibrium, and that stability can never be more than temporary. Instead leadership
dynamics are in perpetual flux, moving through periods of destabilisation and stabilisation,
but never achieving a stable equilibrium.
25
-------------------------------
Insert Figure 3
-------------------------------
[1]6 Destabilising
Adam was Global Managing Partner of the firm. During his four-year term of office he
introduced a range of controls designed to improve financial performance. However, his
somewhat confrontational personal style prompted many partners to become concerned that
these controls represented a systematic attempt by him and his senior leadership group to
increase their power and inhibit partner autonomy. Their fears appeared justified when Adam
attempted to introduce new stringent partner performance management systems, prompting
the partners to reject the proposals.
When Adam announced his retirement as Managing Partner, he indicated his preferred
successor. As partners would not accept a faît accompli, so nine additional partners put their
names forward as candidates for election.
Bob was one of these candidates. He was unknown to most partners; as the practice he led
was relatively small and he was not a member of the senior leadership group. He therefore
campaigned intensively and met most of the more than 400 partners individually. Partners
were impressed by his success in building his small practice and his strong belief in the firm
and in them personally. He suggested he could unleash their “innate entrepreneurialism” by
removing some of the recently imposed controls. He explicitly criticised the track record of
candidates who had served with Adam as part of the senior leadership group. In his campaign
speeches he promised to “give the partnership back to the partners” and introduce a
26
significant programme of investment. Bob was elected with a substantial majority, in spite of
being a rank outsider at the start of the election.
Analysing the unstable equilibrium. During this phase Adam was seen by partners as
asserting excessive control, which impinged on their autonomy (negotiating). He lacked the
political skills to persuade his colleagues that he was acting in their best interests, causing his
colleagues to question his integrity (manoeuvring). By contrast Bob’s leadership campaign
revealed his significant political skills (specifically networking ability and apparent sincerity),
while partners perceived that he had integrity due to his rhetoric of commitment to the
partnership (manoeuvring). Bob’s success at building his practice also prompted partners to
infer he had leadership ability (legitimising). The shift in the underlying power relations
among Adam and the partners as a whole, caused by his clumsy attempt to assert control and
his lack of political skills, destabilised the collective leadership dynamics and created an
opportunity for Bob to rise to power.
[2] Stabilising
Bob had made campaign promises to influential colleagues (such as Chris who was leader of
the firm’s largest practice) to win their support. He now consolidated his position by inviting
them to join his new senior leadership group. Bob also acted on his election promises by
announcing a new programme of investment and involving the entire partnership in
nominating and prioritising investment projects. Partners appreciated the relaxation of
controls and began to enjoy their increased autonomy; they were reassured that Bob had the
integrity to deliver on his promises.
27
At the partner conference on the first anniversary of the election, Bob presented the year-end
financial results. While the presentations of the senior leadership group were emphatically
upbeat, some partners expressed concerns about the dramatic rise in expenditure and on-going
stagnation in revenue. They suggested that Bob’s electoral promises were unrealistic and that
he lacked the ability and integrity to lead the firm effectively. The majority of partners,
however, remained supportive.
Analysing the unstable equilibrium. During this phase collective leadership dynamics were
stabilised as partners were reassured by two factors: Bob had demonstrated integrity by
fulfilling his election promises (manoeuvring) and, as a result, partners had gained in
autonomy (negotiating). However, as financial performance deteriorated, some partners began
to question Bob’s leadership ability (legitimising).
[3] Destabilising
In spite of declining financial performance Bob held back from introducing more stringent
controls on partners’ spending as this would have been contrary to his electoral mandate (and
potentially undermined his reputation for integrity). Profits continued to decline steadily and,
towards the end of Bob’s second year, members of the senior leadership group began
expressing serious concerns. They put pressure on Bob to appoint a COO, with the brief to
reduce costs. The new COO, who was not a partner and came from outside the firm, cut costs
in areas that did not affect partners directly. As a result, partners were broadly supportive of
these enhanced controls.
28
Whereas initially supportive, Chris (the head of the firm’s largest practice) had become
privately critical of Bob, telling colleagues within the senior leadership group that he and the
new CEO were failing to “get a grip” on declining performance. He argued that the COO, as
an external hire with no track record of success in the market, lacked legitimacy among
partners, and was therefore unable to take the tough actions required to turnaround declining
performance. By contrast, Chris was exceptionally successful, both as an individual
professional and as a practice leader.
Chris had previously resisted all encouragement from colleagues to stand for election as
Managing Partner, professing his commitment to his practice. He now let it be known within
the senior leadership group that, for the sake of the firm, he might be persuaded to take over
as COO. Meanwhile, as the annual partner conference approached, a growing number of
partners were threatening a vote of no confidence in Bob.
A delegation from the senior leadership group met with Bob to argue that he should appoint
Chris COO. They believed Chris’s success in the market would give him the legitimacy
among partners for them to accept more stringent controls on their performance. Bob could
retain the title Managing Partner but focus exclusively on an “ambassadorial” role with
clients. Bob reluctantly agreed, fired his COO, and appointed Chris in his place.
Analysing the unstable equilibrium. During this phase three factors brought about a shift in
the underlying power relations among Bob, Chris, and the senior leadership group. First was
Bob’s reluctance to introduce controls that would impinge on partner autonomy (negotiating).
Second was Bob’s appointment of an external COO without a record of market success
(legitimising). Third was Chris’s political behaviour in undermining Bob while reassuring
29
colleagues in the senior leadership group that he had integrity through his stated reluctance to
run for Managing Partner (manoeuvring). The combination of these three factors served to
destabilise the collective leadership dynamics and led to Chris gaining power at the expense
of Bob.
[4] Stabilising
After three years of watching their profits decline, partners welcomed the “strong leadership”
that Chris offered as COO. Rather than complaining about infringements on their autonomy,
partners appreciated Chris for halting the precipitous decline in performance. By contrast,
Bob appeared weak. Towards the end of his four-year term as Managing Partner, Bob took
soundings among the partners to assess his probability of re-election. He realised he had very
little support so announced he would not run for a second term. Chris put his name forward
for election to Managing Partner, strongly supported by the members of the senior leadership
group who had helped engineer his appointment as COO. As no rival candidate challenged
him, he was elected unopposed.
Analysing the unstable equilibrium. During this phase Chris consolidated power at the
expense of Bob. Initially it was the members of the senior leadership group who perceived
Chris positively but now the partners as a whole came to share this view. They saw his
success in the market as evidence of his leadership ability (legitimising). They were unaware
of his political behaviour so perceived him as behaving with integrity (manoeuvring). As a
result, when Chris asserted control, partners did not resent the constraints on their autonomy
(negotiating) and the collective leadership dynamics as a whole were stabilised. Given the
inherently unstable equilibrium, this stability would inevitably prove temporary.
30
Our second research question asked how do changes in underlying power relations among
professional peers serve to stabilise and destabilise collective leadership dynamics over time?
Our case study points to various factors likely to trigger this destabilisation. First, declining
success in the marketplace may cause professionals to infer that a colleague no longer has the
ability to lead them. Second asserting too much control may cause professionals to infer that a
colleague is preventing them from exercising their autonomy, and asserting too little control
may cause professionals to judge them as ineffective. Third, being seen to act politically may
cause professionals to infer that a colleague lacks integrity. The destabilisation of any one of
these relational processes will likely destabilise the collective leadership dynamics as a
whole. For example, if individuals become less successful in the marketplace and lose the
legitimacy to lead, or are perceived to lack integrity as a result of overt political manoeuvring,
they will find it more difficult to assert control over their autonomous colleagues.
Conclusions
To answer our first research question – What are the dynamics through which professional
peers co-construct collective leadership? – we have identified three relational processes:
legitimising, negotiating, and manoeuvring. Legitimising expresses the relational process
whereby professionals develop leader identity. Negotiating and manoeuvring express how
professionals develop leader authority; the former is concerned with overt actions to enhance
formal authority and the latter is concerned with more covert political actions to enhance
informal influence. These express three different aspects of contested and contingent power
relations within PSFs. Taken together they explain how collective leadership is co-constructed
among professional peers. At any point in time, these relational processes may coexist in
31
equilibrium, as professional peers claim and temporarily grant leadership and followership
among themselves. However, as explained below, that equilibrium is inherently unstable.
In answering our second research question – How do changes in underlying power relations
among professional peers serve to stabilise and destabilise collective leadership dynamics
over time? – we have demonstrated how the collective leadership dynamics constitute an
unstable equilibrium, as a result of the instability in the relational processes and power
relations that underlie them. At any moment, each relational process encompasses a
potentially destabilising tension between how a professional in a leadership position acts, and
how their peers experience, interpret and react to these actions, and this can potentially trigger
a shift in power relations. Over time, the destabilisation of one relational process potentially
destabilises the collective leadership dynamics as a whole. The co-construction of leadership
in PSFs is, therefore, unstable both at a moment in time and over time.
We have demonstrated the processes by which power may become manifest in professional
peer relations, and the impact this may have on leadership. When collective leadership
dynamics are destabilised, the most visible outcome is some form of change in formal
leadership positions. For example, individuals may replace colleagues in leadership positions,
new colleagues may join the senior leadership group, and existing leadership positions may be
redefined. In time the equilibrium can be re-established (albeit temporarily). Leadership
positions are confirmed and collective leadership dynamics are stabilised, as colleagues adjust
to new power relationships, and engage in the process of mutually claiming and granting
leadership and followership among themselves on a contingent and temporary basis.
However, our study goes further than previous studies of leadership and power in professional
service firms to explicitly identify the nature of these power relationships, and how processes
32
of claiming and granting leadership and followership are co-constructed among professional
peers. As outlined below, this contributes to the understanding of both collective leadership
processes in general, and in PSFs in particular.
Collective leadership and leader–follower relations
Our study contributes to the literature on collective leadership, which has been criticised for
treating power as implicit and taken for granted, ignoring the extended network of power
relations which underlie the interactions of those in formal leadership positions (Bolden,
2011; Denis et al., 2012). We have shown in detail the intricate, nuanced, and dynamic
relational processes by which power relations can play out. Each relational process embodies
a power relationship. This is most obvious within the relational process of negotiating (where
professionals seek to exert control by formal means), but power is also implicit in legitimising
(where professionals confer leader identity on a peer) and manoeuvring (where professionals’
engage in covert political behaviour in order to influence their peers).
Our study explains how power can be lost and gained by those who aspire to formal
leadership positions, and emphasises the inherently political nature of collective leadership.
We therefore extend established theories by placing politics as well as power at the heart of
our examination of collective leadership dynamics. In so doing our approach is consistent
with critique of much contemporary management research for neglecting the reality of power
relations within organizations (Bolden, 2011; Pfeffer, 2013). Whereas collective leadership
for some authors holds the promise of more democratic forms of organizing (Raelin, 2016),
our study shows that, just as with democracy in general, collective leadership may appear to
33
distribute power equally while conferring advantages on those who engage in political
activity.
Our study of collective leadership also speaks to theories of followership. Whilst theories of
collective leadership represent a valuable antidote to conventional approaches to leadership,
which overly privilege leadership agency, there is a risk that contemporary followership
studies place too great an emphasis on follower agency in isolation. Our study highlights the
extent to which professionals in non-leader positions actively constitute or refute both the
legitimacy and authority of those in positions of power, wielding a ‘repertoire of possible
agencies’ as Collinson (2005: 1422) suggests. However, our study emphasises that these
agencies are not detached from the agencies of leaders. Rather, as our model illustrates, the
processes are relational and reciprocal, co-constructing each other. In other words, we cannot
understand leadership and followership as separate phenomena (cf. Fairhurst, 2001;
Collinson, 2005); they are co-dependent. Thus, our study emphasises that the often-
perpetuated oppositional binaries of leader–follower positions (Collinson, 2006) do not offer a
good starting point for understanding processes of collective leadership.
In other words, our study argues that the “accepted wisdom” that there can be no leadership
without followers (Uhl-Bien et al, 2014: 83) is potentially simplistic and even misleading.
Instead we argue for a conceptualisation of the ways in which individuals can move in and out
of leader–follower relations, subtly and continuously, as they engage in the co-constructing
collective leadership. We believe this represents an important new direction in leadership
research and provides a fertile grounding for more studies that explore the challenging
processes of compromising and consensus building, in which multiple individuals engage in
co-constructing collective leadership.
34
At the same time, by identifying the significance of an individual leader’s market success and
political skills in gaining and exercising leadership legitimacy and authority, our study
encourages collective leadership scholars to acknowledge the significance of individual
agency – for leaders and followers alike – and the extent to which they are co-dependent.
Ultimately, collective leadership is co-constructed through the actions and reactions of
specific individuals. In studies of collective leadership there is a tendency to over-compensate
by emphasising the processual and constructed nature of leadership and followership to such
an extent that these studies neglect the significance of formal leadership positions altogether
(Alvehus, 2018). By contrast, our longitudinal case study analysis has demonstrated that
formal positions do matter in the co-construction of leadership, as it is peers’ political
processes of selecting, electing, undermining and deposing colleagues in leadership positions
which signals shifts in underlying power relations.
By identifying and examining the inherent instabilities in contested and contingent power
relations, our study calls into question a further taken for granted assumption within the
leadership literature – that leader identity and authority are mutually constitutive. We show,
instead, that colleagues may grant leadership identities to their peers, without necessarily
granting them leadership authority, and without claiming follower identities for themselves.
This deceptively straightforward statement initially appears somewhat counter-intuitive.
However, we argue that whereas at an abstracted level we can understand processes of
claiming and granting leader and follower positions to be unambiguous (DeRue and Ashford,
2010), in practice these relationships are considerably less clear than has previously been
portrayed. For example, individuals may identify themselves as being in leadership positions
whereas at the same time being bereft of authority. Claims and grants do not only happen in
35
the overt relational processes of legitimizing and negotiating, but also in the covert process of
manoeuvring. We suggest that future research on leader–follower relations will achieve a
deeper level of insight and more closely reflect reality, if it acknowledges that leadership and
followership are temporary and precarious, as colleagues engage in on-going processes of
interaction by claiming and challenging each other’s claims to leadership.
Leadership and power in PSFs
Finally, our study represents a novel study of leadership dynamics in PSFs, by combining
detailed systematic cross-sectional and longitudinal case analysis in multiple sectors to
investigate a phenomenon previously neglected within the PSF and leadership literatures. It
has gone beyond the somewhat abstract and impersonal representations of power relations
which characterise studies of PSF governance (e.g. Greenwood and Empson, 2003) to identify
and examine in detail the specific relational processes through which professional peers co-
construct collective leadership. In so doing we challenge the somewhat simplistic about PSF
leadership (von Nordenflycht, 2010). In fact, rather than talking in terms of “non-leadership”
(Alvesson and Sveningsson, 2003) we find that leadership is being achieved in PSFs, and may
be highly effective, but it is nuanced and complex, political and unstable.
We argue that what is deemed to be effective or ineffective leadership will vary according to
the specific configuration of circumstances, the actions of each professional in leadership
positions, and the interpretation that their colleagues place on those actions. For example,
actors’ explanations of cause and effect in firm performance play a key role in constituting
legitimacy. In our longitudinal study there is no clear pattern underlying the stabilising and
destabilising processes. This may in itself represent an important finding, suggesting that, like
36
weather patterns, it remains unpredictable and constantly changeable. However, further
research would be needed, into multiple longitudinal case studies of changing power relations
in PSFs to test this tentative finding. We suggest that there is much more to be discovered
from researching the subtleties of leadership dynamics in PSFs, in particular by studying such
processes in real-time, avoiding actors’ inevitable post-facto rationalizations of events and
intents.
Given the significance of PSFs to the global economy, our study has potentially disturbing
implications when it comes to leadership, governance, and accountability of PSFs. It
highlights two particularly vulnerabilities. First, given the complex and unstable political
dynamics of these firms, it may be difficult to precisely locate relevant responsibility for
inappropriate actions by individual professionals and to hold locally powerful leaders to
account. Ultimately the most senior leaders are dependent upon the ongoing support of their
powerful peers, suggesting that there may be circumstances where they are constrained in
their ability to engage in effective oversight of their actions. This leads on to the second
vulnerability. We have demonstrated that professionals can be elevated by their peers to
senior positions of leadership through their success at selling professional work, and not
necessarily as a result of their leadership experience and expertise. Also, to borrow from the
language of political leadership, leaders of PSFs may be very good at campaigning but not
good at governing, finding themselves in office but not in power. Both these underlying
vulnerabilities of in PSFs suggests that, given the exceptional complexity and inherent
instability of leadership dynamics, high-profile examples of leadership failings in PSFs are
likely to remain commonplace.
Notes
37
1. The primary activity of professional service firms, such as consulting, accounting, and
law firms, is the “application of specialist knowledge to the creation of customised
solutions to clients’ problems” (Empson et al., 2015: 6). The core assets of these firms,
i.e. knowledge and clients, are intangible, not directly owned or controlled by the firm
and embodied in the senior professionals who have developed their technical
knowledge and client relationships over many years.
2. PSFs traditionally operated a lockstep system whereby profits were shared among
partners on the basis of time-served rather than individual performance. More
commonly now partner remuneration is determined by a remunerations committee,
elected by the partners, and operating independently of the Managing or Senior
Partner.
3. In each firm no more than two to three females were identified by interviewees as part
of the extended leadership group of 30 or more individuals. While there has been an
increase in the number of female partners in professional service firms in recent years
this change has predominantly been limited to more junior levels.
4. Letter and number in brackets denote firm and interviewee number.
5. As this case engages in analysis of very sensitive political activities, we have changed
minor details in the following text in order to ensure anonymity. These details do not
change the overall character of the leadership dynamics described.
6. Numbers in square brackets relate to numbers at top of Figure 3.
38
References
Alvehus, J. (2017). Clients and cases: Ambiguity and the division of labour in professional
service firms. Baltic Journal of Management, 12, 408–426.
Alvehus, J. (2018). Emergent, distributed, and orchestrated: Understanding leadership through
frame analysis. Leadership Online First. doi:10.1177/1742715018773832
Alvesson, M. & Sveningsson, S. (2003). Good visions, bad micro-management and ugly
ambiguity: Contradictions of (non-)leadership in a knowledge-intensive organization.
Organization Studies, 24, 961–988.
Ammeter, A. P., Douglas, C., Gardner, W. L., Hochwarter, W. A. & Ferris, G. R. (2002).
Toward a political theory of leadership. The Leadership Quarterly, 13, 751–796.
Bligh, M. C., Kohles, J. C. & Pillai, R. (2011). Romancing leadership: Past, present, and
future. The Leadership Quarterly, 22, 1058–1077.
Bolden, R. (2011). Distributed leadership in organizations: A review of theory and research.
International Journal of Management Reviews, 13, 251–269.
Bryman, A. (2004). Qualitative research on leadership: A critical but appreciative review. The
Leadership Quarterly, 15, 729–769.
Charmaz, K. (2014). Constructing grounded theory. 2nd edition. London: Sage.
Chreim, S. (2015). The (non)distribution of leadership roles: Considering leadership practices
and configurations. Human Relations, 68, 517–543.
Collinson, D. (2005). Dialectics of leadership. Human Relations, 58, 1419–1442.
Collinson, D. (2006). Rethinking followership: A post-structuralist analysis of follower
identities. The Leadership Quarterly, 17, 179–189.
39
Collinson, D., Jones, O. S., & Grint, K. (2018). ‘No more heroes’: Critical perspectives on
leadership romanticism. Organization Studies, 39, 1625–1647.
Cooper, D.J., Hinings, B., Greenwood, R. & Brown, J.L. (1996). Sedimentation and
transformation in organizational change: The case of Canadian law
firms. Organization Studies, 17, 623-647.
Cullen, K. & Yammarino, F. (2014). Special issue on collective and network approaches to
leadership. The Leadership Quarterly, 25, 180–181.
Denis, J.-L., Lamothe, L. & Langley, A. (2001). The dynamics of collective leadership and
strategic change in pluralistic organizations. Academy of Management Journal, 44,
809–837.
Denis, J.-L., Langley, A. & Sergi, V. (2012). Leadership in the plural. Academy of
Management Annals, 6, 211–283.
DeRue, D. S. & Ashford, S. J. (2010). Who will lead and who will follow? A social process
of leadership identity construction in organizations. Academy of Management
Review, 35, 627–647.
Empson, L. (2007). Managing the modern law firm: New challenges, new perspectives.
Oxford, UK: Oxford University Press.
Empson, L. (2017). Leading professionals. Power, Politics, and Prima Donnas. Oxford:
Oxford University Press.
Empson, L., & Langley, A. (2015). Leadership and professionals. In L. Empson, D. Muzio, J.
P. Broschak, & B. Hinings (Eds), The Oxford handbook of professional service firms
(pp. 163–188). Oxford: Oxford University Press.
Empson, L., Muzio, D., Broschak, J. & Hinings, B. (2015). Researching professional service
firms: An introduction and overview (pp. 1–22). In L. Empson, D. Muzio, J. P.
40
Broschak, & B. Hinings (Eds), The Oxford handbook of professional service firms.
Oxford, UK: Oxford University Press.
Fairhurst, G. (2001). Dualisms in leadership research. In F. M. Jablin & L. L. Putnam (Eds),
The new handbook of organizational communication (pp. 379-439). Thousand Oaks,
CA: Sage.
Fairhurst, G. T. & Uhl-Bien, M. (2012). Organizational discourse analysis (ODA): Examining
leadership as a relational process. The Leadership Quarterly, 23, 1043–1062.
Fairhurst, G. (2016). Leadership process. In A. Langley & H. Tsoukas (Eds). The SAGE
Handbook of Process Organization Studies, Sage Publications Ltd.
Ferris, G. R., Treadway, D. C., Perrewe, P. L., Brouer, R. L., Douglas, C. & Lux, S. (2007).
Political skill in organizations. Journal of Management, 33, 290–320.
Freidson, E. (2001). Professionalism. The third logic. Chicago, IL: The University of Chicago
Press.
French, J., & Raven, B. (1959). The bases of social power. In D. Cartwright (Ed.), Studies in
social power (pp. 151-164). Ann Arbor, MI: Institute for Social Research, University
of Michigan.
Golden-Biddle, K. & Locke, K. (1993). Appealing work: an investigation of how
ethnographic texts convince. Organization Science, 4, 595–616.
Gordon, A. & Yukl, G. (2004). The future of leadership research: Challenges and
opportunities. Zeitschrift fur Personalforschung, 18, 359–365.
Greenwood, R. & Empson, L. (2003). The professional partnership: Relic or exemplary form
of governance? Organization Studies, 24, 909–933.
Greenwood, R., Hinings, C., & Brown, J. (1990). “P2-form” strategic management: Corporate
practices in professional partnerships. Academy of Management Journal, 33, 725-
755.
41
Gronn, P. (2002). Distributed leadership as a unit of analysis. The Leadership Quarterly, 13,
423–452.
Heckathorn, D. (1997). Respondent-driven sampling: A new approach to the study of hidden
populations. Social Problems, 44, 174–199.
Hinings, C. R., Brown, J. & Greenwood, R. (1991). Change in an autonomous professional
organization. Journal of Management Studies, 28, 376–393.
Kirk, P. & Broussine, M. (2000). The politics of facilitation. Journal of Workplace Learning,
12, 13–22.
Lawrence, T. B., Malhotra, N. & Morris, T. (2012). Episodic and systemic power in the
transformation of professional service firms. Journal of Management Studies, 49,
102–143.
Mintzberg, H. (1989). Mintzberg on management: Inside our strange world of organizations.
New York, NY: Free Press.
Morris, T., Greenwood, R. & Fairclough, S. (2010). Decision making in professional service
firms. In P. Nutt & D. Wilson (Eds) Handbook of decision making (pp. 275–306).
Chichester, UK: John Wiley & Sons.
Mueller, F., Carter, C., & Whittle, A. (2015). Can Audit (Still) be Trusted? Organization
Studies, 36(9), 1171-1203.
Muhr, S. L. (2011). The leader as cyborg: Norm-setting, intimidation and mechanistic
superiority. In M. Alvesson & A. Spicer (Eds.), Metaphors We Lead By (pp. 138–
161). London: Routledge.
Ospina, S.M., Foldy, E. G., Fairhurst, G. T, & Jackson, B. (2018). Collective dimensions of
leadership: The challenges of connecting theory and method. Human Relations.
Special issue call for papers.
42
http://www.tavinstitute.org/humanrelations/special_issues/LeadershipCollect
iveDimensions.html
Pearce, C. L., & Conger, J. A. (2003). Shared Leadership: Reframing the Hows and Whys of
Leadership. Thousand Oaks, CA: SAGE.
Pfeffer, J. (2013). You’re still the same: Why theories of power hold over time and across
contexts. The Academy of Management Perspectives, 27, 269-280.
Pickering, M. E. (2015). An exploratory study of organizational governance in publicly-
quoted professional service firms. Organization Studies, 36, 779–807.
Raelin, J.A. (2016). Imagine there are no leaders: Reframing leadership as collaborative
agency. Leadership, 12, 131-58.
Reid, W., & Karambayya, R. (2009). Impact of dual executive leadership dynamics in
creative organizations. Human Relations, 62(7), 1073-1112.
Robertson, M., & Swan, J. (2003). ‘Control – What control?’ Culture and ambiguity within a
knowledge intensive firm. Journal of Management Studies, 40(4), 831-858.
Scott, W. R. (2008). Lords of the dance: Professionals as institutional agents. Organization
Studies, 29, 219–238.
Silvester, J. (2008). The good, the bad, and the ugly: Politics and politicians at work.
International Review of Industrial and Organizational Psychology, 23, 107–148.
Smets, M., Morris, T., von Nordenflycht, A., & Brock, D. M. (2017). 25 years since ‘P2’:
Taking stock and charting the future of professional firms. Journal of Professions
and Organization, 4, 91–111.
Suddaby, R. (2006). What grounded theory is not. Academy of Management Journal, 49,
633–642.
Thomas, P., & Hewitt, J. (2011). Managerial organization and professional autonomy: A
discourse-based conceptualization. Organization Studies, 32, 1373–1393
43
Uhl-Bien, M., Marion, R. & McKelvey, B. (2007). Complexity leadership theory: Shifting
leadership from the industrial age to the knowledge era. The Leadership Quarterly,
18, 298–318.
Uhl-Bien, M., & Ospina, S. (Eds). (2012). Advancing Relational Leadership Research: A
Dialogue among Perspectives. Charlotte, NC: Sage.
Uhl-Bien, M., & Pillai, R. (2007). The romance of leadership and the social construction of
followership. In B. Shamsie, R. Pillai, M. Bligh, & M. Uhl-Bien (Eds) Follower-
centered perspectives on leadership: A tribute to the memory of James R. Meindl.
Charlotte, NC: Informational Age Publishers.
Uhl-Bien, M., Riggio, R. E., Lowe, K. B. & Carsten, M. K. (2014). Followership theory: A
review and research agenda. The Leadership Quarterly, 25, 83–104.
von Nordenflycht, A. (2010). What is a professional service firms? Toward a theory and
taxonomy of knowledge-intensive firms. Academy of Management Review, 35, 155-
174.
44
Table I: Legitimising
Succeeding in market and Inferring leadership ability
One of the reasons I think why people were happy to have me as Senior Partner was because I was
perceived to be a successful fee-earning partner… I ran a lot of our key relationships and so on.
(A7)
A lot of the leadership in a firm is linked to your practice and the position in the market because
those people have the credibility, those people are in the market, they actually know the client and
so on and in the end all that we do is ultimately directed towards clients. (A1)
Leadership in this firm is a function of one’s credibility as a professional, in what we do – I don’t
think I necessarily have been the best professional in our firm but I am respected, and particularly
respected as a client relationship person – without that I cannot lead. (B11)
The people on the global executive committee are the people who are perceived within the firm as
being very strong performers in their market or in their practice or whatever. (B2)
They earn the right to get into one of those [leadership] positions through ... doing fantastic stuff
with clients, through enhancing the value of the partnership because they’ve either developed a
client relationship that we didn’t have before, or won a major piece of work and done that
consistently ... They have the respect of the partners because they’re recognised for doing that.
(C14)
People were drawn to me, I had followers, before I went into a leadership role because I was the
lead partner on [Company X] which was a big client for the firm. So I created opportunities for
people, I was successful, they enjoyed working with the client. So all those things, I think, create
followers. (C27)
45
Table II: Negotiating
Asserting control and Exercising autonomy
Because we are the size we are now, we have to have a more regulated set of rules. … So there has
to be checks and balances and I think that stops a bit of entrepreneurialism. (A9)
Interviewer: Does anyone have power over you?
Partner: Not as far as I’m concerned no.
Interviewer: Does anyone think they have power over you?
Partner: I don’t think so. (A8)
Being Chairman is like walking a tightrope of helping my partners feel like owners, helping them
feel involved, helping them be engaged, not dominating them, not getting out in front, not having a
huge ego which makes them feel like the Chairman’s kind of off on his own trip. At the same time
being strong and providing them with a sense of confidence that we’re going somewhere. (B32)
The organisation has thrived on giving people a high degree of independence, allowing them to
operate within certain constraints, which are mainly ethically driven and sort of quality driven, to do
things in their own way. And has allowed people to experiment, quite freely, very freely I would
say. (B13)
Even though all the partners are peers, there do need to be decisions made. There do need to be
tough things done … knowing when you have to step in and say “no”, “yes”. (C14)
You take as much responsibility as you want. And people will allow it. But they don’t offer it to
you. (C6)
46
Table III: Manoeuvring
Behaving politically and Perceiving integrity
Part 1: Overt political behaviour – how it is understood, manifested, and criticised
There are people who are clearly very ambitious in the firm, who will say from quite an early age to
you, particularly over a beer or over a meal or over a chat, “do you think I’m in the frame to be
Managing Partner or Senior Partner, what is it I need to do?”… And they’re regarded as more
political in terms of what they think people will want to hear and what they think people will want to
vote for. (A2)
To me politics smacks of alliance building in the corridors, in offices behind the scenes. It smacks of
people engineering agendas, which creep up on the firm and deliver faît accompli or behave in ways
that become disruptive. Or politics could manifest as someone undermining another person. I would
like to think we don’t have those behaviours in this firm. (B11)
I’m not naïve in thinking there’s an amount of manipulation, manoeuvring, etc. involved in leading
this firm, but if, underneath all of that, there’s no authenticity about why we’re doing it, then people
get found out. And that’s what I mean by political. And I think sometimes that political tag is unfairly
ascribed. A very good friend of mine was unjustly perceived to be too political, and suffered as a
result…. I think he does genuinely care about the firm. But unfortunately, sometimes he gives off that
political vibe. (C32)
Part 2: Subtle political behaviour – how it is manifested, explained, and accepted
If there’s an idea, a sort of major initiative, it will start with the inner circle. Then it will get the air
cover from the Board… As it develops and it’s got some legs [the Senior Partner] will involve one or
two other heads of offices…and there will just be a sort of few people here and there. Then, if the
thing is going to happen, there will be a call of all the heads of major offices and all the heads of the
other practice groups to get their views before launching it on the partnership. (A12)
You need to build alliances with your colleagues, you need to build interest around the issue, you
need to show someone that this is a successful story that they would like to be associated with. Plus
you have to play a few kind of international things, alongside the local office things, and once you do,
it goes very smoothly. (B29)
The previous election for Senior Partner there was a bit of a power struggle between two individuals
who were dominant characters within the organisation and (person) held sway on the basis of support
from a number of the more senior partners including the outgoing Senior Partner who had quite a lot
of influence…. It was all closed doors – smoke filled rooms – lots of politicking etc. etc. etc. – a
variety of promises being made to various senior people to get them to support [person]. (C6)
Part 3: Perception of integrity – how leaders are seen to be working for “common good”
Sometimes my sense would be [Senior Partner] doesn’t necessarily always understand how
influential he is. He’s very modest about it, quite self-effacing, and he himself doesn’t attach such
great importance to some of those things that might be under the heading of creeping as in slightly
sinister. He is not himself a player in that way at all. … it’s simply because his own motivations in
this world are so, I think, very genuine and clean. (A19)
47
[Before we elected him Chairman] he didn’t say a lot at partner conferences. But on some key
discussions he was the compelling speaker. I’d say once a year on a particular issue, he would come
in on his white horse and say “well, this is what I think” and do it very well and with a good ability to
sort of link the discussion that had been happening to some prime principle of the firm, some prime
value of the firm. (B5)
[The Chair’s] got a really remarkable ability to operate on his feet, without a script and not mumble,
and he’s hugely eloquent. His ability to deal with questions is almost unrivalled. He’s very softly
spoken and sometimes you just want him to rah-rah a bit, but that’s not his style. He comes across as
having very high integrity. (C31)
48
SEE SEPARATELY UPLOADED FILES FOR CLEARER REPRESENTATIONS OF
GRAPHICS
Figure 1. Coding structure
49
Figure 2. Co-constructing collective leadership dynamics among professional peers
50
Figure 3. Collective leadership dynamics in a PSF: Events, actions and reactions over a 5-year period
