The Interplay between Learning and Change
Peter A. Murray
Senior Lecturer
School of Business
Division of Economic and Financial Studies,
Macquarie University, NSW. Australia.
Phone +61 2 98507787 Fax +61 2 98506065
Email pmurray@efs.mq.edu.au
Kimberley B. Boal
Trinity Professor of Organizational Studies
and Strategic Management,
Rawls College of Business Administration
Texas Technology University,
Lubbock. TX 79409
Phone: (806) 742-2150
E-mail: kimboal@ttu.edu
1
The Interplay between Learning and Change
Summary
Various types of change, change capacity, and change processes have often been presented as
single isolated events with a clear beginning and end. Learning similarly has been presented
as an end that is desirable within certain contexts, as a basis for strategic capability and
renewal, learning styles, and as a process influencing individuals, groups, and organizations
and how they learn. While scholars have discussed the connection between change and
learning on a broad scale, it remains unclear how learning behaviour contributes to change
success; the possibility of increasing the alignment between learning behaviour and different
types of change is not a common research topic. The study addresses in part some of these
gaps. The first section of the paper explores the interplay between learning and change. The
second examines institutional learning and relates this to institutional contexts. We explore
the effects of these multiple interplays on change success. The third section explores learning
and change relationships through an empirical study of the Australian construction industry.
2
Introduction
Emanating from the conception that strategic renewal is obtained by balancing exploitation
with exploration (March, 1991; Crossan, Lane, and White, 1999), it is almost impossible to
separate learning from this process. Only more recently have scholars pointed to the need to
differentiate between exploitation and exploration in terms of the type and amount of learning
(Gupta, Smith, and Shalley, 2006), and it is important not to overlook the reliabilityenhancing learning that results from human and organizational attempts to replicate past
routines’ (p. 695). While exploitation and exploration has been firmly grounded in studies of
strategic renewal, change scholars prefer to highlight renewal as a continuum with two ends.
One as a need to break from the past and challenge equilibrium through punctuated bursts of
activity (Gersick, 1991), the other as a basis for continuous change (Brown and Eisenhardt,
1997) where organisations are capable of shifts to different technological trajectories (Benner
and Tushman, 2003) by creating ambidextrous capabilities. The challenge for managers it
appears has been the need to break the fit or alignment between strategy, structure, and the
culture of past exploitation (Tushman and O’Reilly, 1996). For change scholars, achieving
exploration comes from pursuing both incremental and discontinuous innovation that hosts
multiple contradictory structures, processes, and cultures within the same firm (1996: p. 24).
The precise role of learning in these processes however has not been made explicit.
In this study, our central thesis is that the type and quality of learning underpins change
success. The purpose of the paper is to focus on the interplay between different learning
modes that influence change processes which in turn lead to change success, which is a
different emphasis. We do not see change processes as isolated events with a clear beginning
and end (Meyer and Stensaker, 2006), and we attempt to illustrate why the interplay between
learning and change deserves greater prominence. Similarly, we advance the idea that
learning is not a valuable and important means within itself; it only becomes valuable when it
is used to develop behaviours that lead to superior capabilities (Eisenhardt and Martin, 2000;
Murray and Chapman, 2003), that are beneficial in both exploitation and exploration pursuits.
We build on the connections between change and learning by going outside of existing studies
by providing empirical evidence of the importance of the relationship, and by creating a
framework that takes us closer to theory building. Not all change processes, tools, and
methods are discussed as this is beyond the scope of the study. We limit our discussion to
processes commonly linked to well known change types (Nadler and Tushman, 1989).
3
Similarly, following Miller (1996), and March (1991), we limit our focus on learning to
learning behaviours and learning modes rather than those with a broader compass reflecting
learning styles, learning contexts, and team learning, among others.
This is a study about learning and change in the Australian construction industry. The
industry can be highlighted by two important facts: 1) it has been subject to constant change
on the part of powerful buyers and institutional elites, and 2) it has strong shared values and
beliefs representing an organisation code (March, 1991), that is commonly linked across the
industry by many interdependent actors. The reality of learning occurring within and outside
the code is discussed. In much the same way that radical change outcomes are often different
from what was originally intended (Greenwood and Hinings, 1996; Hinings and Greenwood,
1988a), the industry has not benefited by change attempts that have been poorly aligned to the
quality of learning and the type of change required. The study examines these relationships.
Similarly, we examine the notion that tacit knowledge is never fully explicit (Miller, Zhao,
and Calantone, 2006) through the organisation code suggesting that different modes of
learning are engaged by individuals outside of standard shared meanings. In discussing the
interplay between learning and change, we recognize both the phenomenon of interest and the
underlying assumptions. Our analysis is at the macro-organisation level. The paper is divided
into three parts. The theoretical background is discussed first. This consists of the modes of
learning (discussed next), the institutional context, and change types, where the underlying
assumptions and relationships are outlined. The second section comprises the methodology
including the research questions and results, and the third section outlines the discussion and
conclusions.
Theoretical Background
A Case for Learning Behaviour
In terms of describing how organisations learn, there are as many methods as concepts from
individual learning (Argyris, 1993), to group learning (Marks, Mathieu, and Zaccaro, 2001;
Edmondson, Bohmer, and Pisano, 2001), to organisational learning (Hedberg, 1981; Kim,
1993; Easterby-Smith and Araujo, 1999), including studies of the connection between all
three (Crossan et al., 1999; Crossan and Berdrow, 2003). While some scholars have preferred
to categorize learning into routines, levels, and capabilities (Fiol and Lyles, 1985; Kim, 1993;
Eisenhardt and Martin, 2000; Murray and Donegan, 2003), others have preferred to broadly
4
conceptualise the concept in terms of exploring the environment and exploiting existing
learning (March, 1991, 2006), through various change techniques (Nadler and Tushman,
1989; Benner and Tushman, 2003). The interplay between change and learning, and between
thought and action, in most if not all of these studies however has been the concern for a
series of change steps, methods, tools, techniques, and change capacity (Nadler and Tushman,
1989; Benner and Tushman, 2003; Meyer and Stensaker, 2006), among others, rather than a
focus on specific learning behaviours required for achieving change success. So far,
behaviour for organisational learning scholars has been studied in respect of individuals,
groups, and organisation-wide processes (Hedberg, 1981; Kim, 1993; Easterby-Smith and
Araujo, 1999; Crossan and Berdrow, 2003), in response to both the internal and external
environment. Possible links between the dependence of change success on the one hand and
learning behaviour on the other has been less explicit. Conceptions for successful change
appear to be implied on the basis that capabilities exist in bundles of resources and that
learning routines are subsumed within these or are a simple matter of purposeful training
(Argyris, 1993; Eisenhardt and Martin, 2000).
Following Miller (1996), we simplify learning between thought and action into a typology of
two learning modes: methodical and emergent. Specifically, the modes can be explored in
terms of their sub-behaviours. The methodical mode has three: structural, analytical, and
experimental (p. 488). Both learning modes we contend paint a new face on the interplay
between learning and change. Scholars have attempted to link the two, often with a focus on
one leading to the other, by describing the relationship as exploitive innovation versus
exploratory innovation (Benner and Tushman, 2002), as a feed forward and feedback process
between intuiting, interpreting, integrating, and institutionalizing (Crossan et al., 1999), and
more recently as a focus between the type and amount of learning (Gupta et al., 2006).
Generally however, both learning and change have been studied as separate processes leading
to different ends. Hence, we discuss change attempts in terms of the exploitation and
exploration trade-off but with a focus on learning behaviour. More successful change we
define as the ability of an organisation to exploit the past yet explore and innovate for the
future by aligning the quality of learning behaviour to the type of change required. We
discuss later how poor alignment between the two impoverishes change success.
While exploration is about exploring new possibilities, exploitation is concerned with
exploiting old certainties (March, 1991; Crossan et al., 1999). Where the links between
5
exploitation, exploration, and strategic renewal are well known, less is known about the role
learning plays in this process. What we do know is that individuals learn from the
organizational code of shared languages, beliefs, and practices, but that conformity to the
code drives out interpersonal heterogeneity leading to lower long-run performance (Miller et
al., 2006; March, 2006). This would suggest that a firm’s response to its environment will be
compromised by its organizational code. Moreover, while mutual learning occurs between
individuals and the organization, learning also occurs outside of the code through one-to-one
exchanges and interpersonal networks (Miller et al., 2006). Hence, further empirical
observations between change and learning will help to distil the theoretical interplay between
learning behaviour and change types, not least between individuals and/or groups as actors
engaged in strong network relationships.
The Methodical Mode
First, the structural behaviours of the methodical mode codify prior learning by specifying
how to carry out tasks and roles (Miller, 1996, p. 495). Structure guides behaviours and
perceptions; it also provides habit and redundancy and can constrain the range of experiences.
Analytical behaviours by comparison are used to perform market analysis and to develop and
monitor strategic plans. At best, they assist managers to think critically about assumptions
related to markets, where the right foot follows the left in a mechanically rigid fashion, but
are possibly more suited in the implementation of strategy (Mintzberg, Ahlstrand, and
Lampel, 1998; Miller, 1996, p. 491). Scholars have found for instance that analytical
behaviours were typical of moderately dynamic markets where change was predictable
(Eisenhardt and Martin, 2000, Murray and Donegan, 2003), but less likely under conditions
of continuous change (Brown and Eisenhardt, 1997). The experimental behaviour (the third
type), is useful in exploiting the capacities for learning that already exist. While the behaviour
is useful in adaptation to discover better ways in doing things (Miller, 1996), mostly through
small iterations and workplace experiments (Bessant, 2004, Bessant and Caffyn, 1996), it is
not used as a basis to explore new opportunities and to overcome long-term efficiency
problems (Boal & Shultz, in Press).
Since all three types of learning behaviour are methodical, they are less likely to underpin
change attempts that rely on more dynamic or radical learning. We discuss later how dynamic
change requires a break from existing structures and systems and a more robust change effort
(Brown and Eisenhardt, 1997; Benner and Tushman, 2003). Methodical learning suits
6
exploitation and adaptive change. The methodical mode represents the integration of learning
by teams and individuals as they strive for shared meaning. Yet, the strengths of the mode
limit new learning. Methodical routines commonly exploit known behaviours from the
organizational code, and are less beneficial in exploration. What is learned at the
organizational level is what has been embedded in the organizational code where learning
feeds back to the individual before the feed forward process repeats itself from the individual,
to the team, to the organisation (Crossan and Berdrow, 2003; Crossan et al., 1999).
Workplace change success in these circumstances will be influenced by adaptations of the
past and learning behaviour will quite possibly be restrained.
The Emergent Mode
The flip side to a methodical approach is the emergent mode. The emergent mode can be
interpreted on the basis of transforming change as illustrated in high velocity markets since it
is geared more to exploration than exploitation (Eisenhardt and Martin, 2000; March, 1993;
Miller, 1996). Here, organizational members constantly challenge system processes and
search for novel and new ideas. The emergent mode has three sub-behaviours: synthetic,
interactive, and institutional. The synthetic behaviour is the creative capacity to detect
systems and configurations where others see only a jumble of events (Miller, 1996, p. 492).
Where core rigidities are evident such as past capabilities that are a stumbling block to change
(Leonard-Barton, 1992), synthetic behaviours are more likely to enable managers to collect
information free of extraneous details and to discover critical resources. Similarly, learning
occurs in an implicit way by exchanging information and evaluating transactions.
Interactive behaviours also lead to different ways an organizational event is interpreted (LeeKelley and Blackman, 2005). Interaction is the product of spontaneous local adjustments to
local opportunities that foster realistic collaboration. Interactive learning leads to new and
different interpretations as a result of speaking up and creating psychological safety for team
members (Edmondson et al., 2003). Interpersonal signals such as gestures, facial expression,
and physical presence are important components of interactive behaviour (Murray and
Blackman, 2006), and participants socially construct meaning as they constantly swap
dialogue and share interpretations (Oswick et al., 2000). While synthetic behaviours examine
novel ideas and patterns of relationships, interactive one’s are more about bargaining and
trading through daily encounters, and accommodating and negotiating local agreements,
7
where knowledge transfer is a critical outcome of face-to-face communication (Orlikowski,
2002).
Institutional behaviours by comparison are also emergent but they allow for little voluntarism
on the part of learners (Miller, 1996). Institutional learning relates to inductive processes
where firms assimilate values, ideologies, and practices such that knowledge is widely
diffused. However, one emergent outcome is indoctrination often by an elite group in the
pursuit of enforcing institutional values that may be particularly invasive in change attempts.
For instance, associations of accountants and law professionals institutionalize procedures in
such a way that new learning (e.g. synthetic behaviour) can be difficult to implement
(Greenwood and Hinings, 1996; Hinings and Greenwood, 1988a). Change agendas become
deflected or thwarted from a specific change track or direction and change implementation is
difficult in such circumstances.
We argue that exploration skills will be more explicit under the learning conditions of the
emergent mode. However, as scholars have pointed out, the balance between both
exploitation and exploration involves at least some learning as organisations go down the
learning curve (Gupta et al., 2006). Scholars contend that exploration is at the heart of
strategic renewal (Crossan, Lane, and White, 1999; Crossan and Berdrow, 2003;
Kontoghiorghes, Awbrey, and Feurig, 2005), and that the justification for exploration is the
myopia from adaptive processes. Indeed, the latter pose problems for exploration and longrun viability (March, 2006: p. 206), if firm actions converge around well known learning
behaviours at the expense of establishing emergent ones that lead to exploration practices.
Between the juxtaposition of the methodical and emergent learning modes lies the need to
intuit and interpret in new and novel ways. Popular conceptions such as double-loop learning
(Argyris, 1993) have often been used in the past to explain how organisational members need
to search for new and novel creations by challenging existing ones. The construction industry
in Australia for instance is an industry calling for continuous change through new learning yet
is deeply entrenched in methodical behaviours that stifle new learning. It is also troubled by
institutional dominance. Change attempts in the industry have resulted in mixed messages
and outcomes. Change has been more often viewed as a nuisance and interruption to current
learning behaviours and an organization code that reinforces them.
8
Change efforts in the industry are similar to the theoretical explanation of long periods of
stability punctuated by compact periods of metamorphic change described by Gersick (1991)
and others as punctuated equilibrium (Tushman and O’Reilly, 1996; Burgelman, 2003).
Sudden bursts in the past however have seldom been successful leaving the researchers in the
current study to question the type of change being implemented by powerful buyers, not least
the influence of learning behaviour. In much the same way that Xerox was able to encourage
service representatives to share experience-based tips on servicing techniques in a deliberate
learning fashion (Arthur and Huntley, 2005), our initial findings indicate that shared learning
has been undervalued in the industry. So that learning represents the ability to engage in rapid
and continuous change (Brown and Eisenhardt, 1997; Benner and Tushman, 2003), further
research is needed to determine how the current quality and quantity of learning supports
change. The discussion thus far leads to the first research question.
RQ 1: What is the dominant mode of learning in the construction industry and what role does
this play in exploitation and exploration?
The next section examines the institutional contexts.
Institutional Contexts
Institutional learning behaviour emanates from the emergent mode and is common in
assimilating beliefs and values (Schein, 2004; Greenwood and Hinings, 1996; Schneider and
Barsoux, 1997). Institutional behaviour commonly represents routines of domination (Clegg,
2000) and is influential across a broader context (Gordon, 2002; Greenwood and Hinings,
1996). While in a positive sense such behaviour provides direction, articulates strategy, and
reinforces desirable ends (Schein, 2004), from time to time, it can also be contested, or have a
negative effect on emerging workplace change success (Miller, 1996; Hinings and
Greenwood, 1988a). The assimilation of institutional learning by powerful players, whether
by institutions themselves or some other means, will often clash with the existing
organization code and existing learning behaviour. This is especially the case when the new
form of institutional learning wants to replace the old one. Leader attempts for instance in the
medical industry are typical of such conflicts with professional groups (Denis et al., 2000;
Gordon, 2002) such as medical professors, law and accounting professionals, or semiprofessionals who embody a set of structures, processes and systems unique to their
institution. Any form of institutional change interrupts the existing learning behaviour.
9
Different change types (discussed next) we suggest will also play an important role. If a
radical type change is required, this may conflict with the organization code leading to poor
alignment to current learning behaviour. Here, new institutional learning behaviour may be
superimposed on existing ones causing confusion. Commitment to the change itself by actors
will most likely be negative since actors are more familiar with existing learning. Workplace
change attempts may then become an anathema to their context. Change processes that result
in wholesale upheaval for example are seldom successful when poorly matched to the context
in which change is intended (Greenwood and Hinings, 1996; Tushman, Newman, and
Romanelli, 1986; Benner and Tushman, 2003).
We argue that rapid change suits the emergent modes of learning, typically synthetic and
interactive learning behaviour, and possibly institutional learning when actors are committed
to the change. We expand on the latter below. Emergent learning promotes open
communication and information sharing together with new idea promotion and resource
availability (Miller et al., 2006; Kontoghiorghes et al., 2005). We suggest that the interplay
between learning and change will be richer in the emergent mode as a result. With the added
imposition of the institutional context, cognition and action will be severely restricted. Miller
(1996) describes institutional behaviour as an environment of elite members that ‘teach’ and
that what is taught is often ideological (1996: p. 495).
The final point noteworthy here is the connection between institutional contexts and the
exploitation/exploration tradeoff. Institutional dominance may strongly influence the
organization code limiting scope between thought and action (voluntarism versus
determinism), by either reinforcing the existing set of structures and systems, or
superimposing them with a new set. In such circumstances, successful change attempts will
not be based on exploring new possibilities but rather exploiting the past and reinforcing
methodical learning behaviour. If an organization was attempting to make change attempts
that relied on some form of new learning, this may be difficult if not impossible. The code
reinforces the convergence of existing practice (March, 2006). These interplays however are
better synthesized when we include a more detailed description of the types of change. The
second research question is now possible:
10
RQ 2: What is the effect of institutional learning behaviour on industry learning generally and
on change success?
Types of Change
In turning to different types of change, Nadler and Tushman (1989) indicate four: tuning,
adaptation, reorientation, and recreation. Our discussion is limited to these. Both tuning and
adaptation are consistent with the methodical mode of learning while reorientation and
recreation are more emergent learning forms. Our central hypothesis is that both the
methodical and emergent modes of learning underpin the four change types.
First, tuning is an incremental condition made in anticipation of future events by
concentrating on efficiency (1989: p. 196). In terms of learning behaviour, tuning is a process
that makes small scale adjustments to decisions on the basis that learning behaviours are
standardized. The process has often been described by scholars as lower-level learning or
single-loop learning (Fiol and Lyles, 1985; Kim, 1993; Murray and Donegan, 2003).
Adaptation is also incremental in response to external events (e.g. competitive attacks) but
does not involve fundamental change throughout the organisation. Instead, learning
behaviour is based on small adaptations to products in line with competitive attacks and
environmental adjustments that are a source of variation from environmental signals
(Hedberg, 1981). Both tuning and adapting are consistent with the methodical mode by
modifying a firm’s response to both its internal and external environment through analytical,
structured, and experimental learning probes. Learning behaviour is both confined and
restricted. Here, a firm deals with ‘one goal at a time while emphasizing short-run reaction
and short-run feedback’ (Romanelli and Tushman, 1994: 1143; Benner and Tushman, 2003);
work practices are driven by standard information processing (Miller, 1996: p. 488), and
learning behaviour is well known. Both tuning and adaptation is consistent with exploitation
since the change process relies on learning behaviours that are structured, analytical, and
experimental.
By comparison, reorientation is a strategic change. Reorientation anticipates external events
that ultimately require change (Nadler and Tushman, 1989: p. 196). While such changes may
not involve fundamental redirection, they may involve widespread organization change
within the context of emphasizing continuity with the past and maintaining current values.
Changing structure by dismantling a business unit would be an example of an internal change
11
supporting an external strategic reorientation. Recreation by comparison is necessitated by
external events that threaten the very existence of the organization (p. 196); see also Benner
and Tushman, 2003). Here, radical departures are common with a complete break from
traditional values by destroying the alignment between strategy, structure, and systems, and
are more risky. Both reorientation and recreation are consistent with the emergent mode since
they require greater amounts of synthetic, interactive, and institutional learning behaviour. To
achieve change in the emergent mode, decisions will be more spontaneous and intuitive,
‘dependant on instinct and impressions, learning tacitly and inductively, garnering insight by
interpreting unstructured impressions and streams of data’ (Miller, 1996: p. 487; see also
March, 1991).
Figure 1 about here
In order for a firm to embrace wholesale change such as reorientation and be committed to it,
there would be a greater need for voluntarism and flexibility on the part of decision makers.
Major changes in learning behaviour consisting of new procedures for instance, or new work
patterns that dissolve the previous ones, or new systems, are far reaching in their
consequences since they upset both the material and technical support functions (Clegg,
2000; Denis et al., 2000; Edmondson, 2003). Here, learning behaviour that has been based on
well rehearsed work routines and functions, including social relationships, is suddenly
disrupted. The pace of reorientation for instance would demand new work routines or
changes to technical arrangements, and possibly new structures associated with
organisational hierarchy. Accordingly, learning behaviour that can reflect and cope with the
change would more likely rely on greater amounts of synthetic and interactive learning;
testing and challenging assumptions and providing forums for presenting new knowledge
would be an example of this.
The existence of context suggests that a simple connection however between the type of
change and the mode of learning needs further analysis. There are a number of issues to
establish here: 1) the source of power of those proposing the change 2) the level of
commitment to the change by the current workforce and/or other actors and 3) the alignment
between the type of change and the current mode of learning. The latter would suggest that
both the code that expresses all current learning behaviour and the balance that an
12
organisation
achieves
between
exploiting
versus
exploring
would
be
important
considerations.
For the first point, powerful actors representing particular institutions (e.g. government
departments, professional groups) may impose change on organisational actors. When the
change is not accepted, threats over non compliance may result. The latter may lead to a
reactive reprisal by the workforce and/or actors where commitment to the change may be
negative (the second point). For both points 2 and 3, we posit that change commitment will
be positive when there is a greater alignment between the type of change and the existing
organisation code. The code is strong when there is a strong tie to existing structures and
systems that dictate learning behaviours whereas the opposite is the case when the tie is
weak. We suggest that providing change commitment by actors is positive, it matters little
whether the code is strong or weak since actors are receptive to change. The worse case
scenario however would occur when the code was strong yet change commitment negative.
Here we would expect conflicts to be accelerated on the basis of a poor match between
learning and change. Figure 1 reflects the relationships thus far and outlines the basis of a
theoretical framework.
We contend that both the type of change and learning behaviour are in alignment when
change commitment is positive and the current code of learning can be changed without a
dramatic negative effect on the existing material and technical systems. When both change
and learning are in alignment, we posit that workplace change efforts will be more successful.
Conversely, when not, change commitment will be negative and workplace change efforts
less successful. So, conflicts between learning and change may result when reorientation or
recreation as a change type confronts a methodical mode of learning preoccupied by strong
shared beliefs in existing structures. If the change type is calling for fundamental new
direction yet actors are restricted in cognition and action, change efforts will clash with the
existing organization code that promotes channelled and constrained behaviour particularly in
situations where change commitment is lacking. Effectively, reorientation may require too
much exploration at the expense of upsetting the balance of learning in the workplace which
is currently achieved by exploiting well known learning behaviours. This may well be the
case for tuning and adapting as well if the code is challenged and methodical learning
routines replaced by new ones. The successful interplay between learning and change is very
much dependent on the acceptance of both the type of change and the quality and amount of
13
learning in the form of the current or new learning behaviour, and how this influences the
balance between exploitation and exploration. Figure 2 outlines these relationships leading to
the third research question.
RQ 3: How does the type of change together with the quality and amount of learning
influence change success?
We now turn to the methodology to test the relationships of the frameworks (Figure 1 and 2).
Methodology
Background
This study examines the implementation of change in the construction industry. It
investigates the responses of managers to reorientation as a change type. It does this by
observing 1) the existence of the two learning modes and their sub-behaviours, and 2) by
aligning these to the type of change. The institutional context is explored also in relation to
these two points. Ten firms are examined. The value of the study lies in the empirical
observations between change and learning adding to existing scholarly interpretations
between both phenomena (Gupta et al., 2006; Miller et al., 2006; Benner and Tushman,
2003). It also seeks to make a contribution between the exploitation/exploration trade-off
(Crossan et al., 1999), by advancing the connection and interplay between learning and
change. In addition, the study tests the framework (Figure 1) in an effort to add to theory
building and as a basis for future research. This section describes the industry, explores the
research design, and explains how the data was collected, tabulated, and analysed.
Figure 2 about here
The Industry
The construction industry in Australia is represented by a myriad of large buyers including
government authorities and independent firms, and clusters of small to medium sized
contractors, subcontractors, consultants, associations, unions, and supporting businesses.
Typically, large contractor turnover is in excess of $100m dominated by very large firms.
Small businesses are represented by both medium sized firms with turnovers in excess of
$10m to ‘Mum and Dad’ businesses from bricklaying, to concreting, carpentry, and other
14
miscellaneous. Consultants are typically represented by professionals including engineering
and architectural firms. Collectively, rivalry is intense and powerful buyers often hold sway
in dictating terms and conditions.
Research Design
The research is based on the interpretive paradigm (Burrell and Morgan, 1979) and qualitative
observations of managers in the workplace. The paradigm focuses on the person and their
actual lived experiences, social actions encountered, and the meanings that lie behind the
various events and phenomena under question (Weick 2001; Berger and Luckman, 1966;
Schwandt, 1994). To measure the data, the research adopted both causal network analysis
and a thematic conceptual matrix (Miles and Huberman, 1994). A causal network is a display
of the most important independent and dependent variables in a field study and the
relationships among them (p. 153), while a thematic conceptual matrix is represented by the
broad themes, the participants and their comments, and possible connections and associations
(Miles and Huberman, 1994). The latter is particularly useful in theory building.
The Data
Primary data was obtained from 42 unstructured manager interviews and observations in
practice conducted within the ten firms during 2003 (Table 1). All firms had experienced
significant change during the previous 5 years of the study. Data was tabulated and arranged
using Nud*ist according to the two learning modes and the types of change evident (Figure
1). Data was also transported to thematic matrixes in order to recognize the learning
behaviours and common associations by matching work routines to both the mode of learning
and the type of change. The researchers expected to observe either positive or negative
commitment to the change on the basis of the observed work routines, and strong or weak
code on the basis of existing beliefs. In addition, institutionalized behaviours were recognized
by work routines imposed and again matched to the type of change (Table 2).
Table 1 about here
Triangulation of the data was achieved by adopting multiple research methods in support of
an interpretive approach including unstructured interviews (Table 3), multi-site observation,
and focus groups. Data was obtained through direct participation of the researchers on
construction sites and comments recorded. Focus group analysis was used to observe the
15
comments of groups of managers represented by different stakeholders. Research conducted
across multiple sites is consistent with Van Maanen’s (1988) conception that qualitative work
is about commonality and things shared in the social world, not so much differentiation and
things not so shared. Focus group research provided a further test of validity. Respondent
validity was noted by allowing focus group members to comment on the broad findings.
Table 2 about here
First, the data was analysed by developing research nodes of key parent disciplines as is
common using Nud*ist as a tool in causal network analysis. Each node was related to the
learning behaviours and the types of change. From here, the researchers examined the data by
building associations, developing sub-nodes of the assumptions, and testing the relationships
and effects of the associations including positive and negative change commitment.
Managerial interpretations of the effects of change were analysed one-by-one by constructing
manually 42 A3 sheets (population N= 42) summarised into thematic conceptual tables. This
also included one set of data from the focus group and also coded secondary data using the
causal nodes selected. In concert with the research design, the researchers carefully examined
the alignment between the mode of learning and the type of change with the actual
experiences of the managers. The following discussion examines the interplay between the
institutional context, the learning modes and their underlying behaviours, and the influence of
change type on change success.
Results
The Institutional Context
As an emergent mode of learning, institutional learning was observed on the basis of the
imposed systems and procedures represented by contractual obligations. The most powerful
buyer for construction services was the State Government’s Department of Public Works and
Services. Institutional work routines consisted of pre-designed standard quality frameworks
and a desire by the Government to overhaul old workplace practices to increase long-term
efficiency. This meant replacing current methods mostly by reorienting existing procedures
and tuning and adapting others. The Government sold its workplace change initiatives
through a new contractual tendering procedure that radically changed the existing one.
16
Supplier failure to adhere to these meant that contracts were only awarded to those who met
the criteria. Future contracts depended on the degree of cooperation displayed by all actors in
the supply chain from large contractors to subcontractors, engineers, architects, and all other
actors. Dominant issues related to the type of contract being negotiated since much of the
industry was highly price sensitive. Contracts were divided into two types: lump sum and
design/construct. The lump sum contract was imposed by Government buyers on a fixed price
commodity and all subsequent planning had to fall within the price.
Table 3 about here
Quite distinct from the lump sum process, design/construct was based on traditional beliefs by
combining both methodical and emergent learning. Here, clients (e.g. a large contractor or
engineering firm as opposed to the Government), handed much of the design work to either
architects or engineers who in turn subcontracted other supply chain activities. The old system
of design/construct however was inconsistent with the new radical procedure proposed by the
Government. Actors lamented the good old days consisting of long-standing traditions handed
down from one generation to the next. Mutually beneficial economic relationships, high trust,
and high supply chain collaboration was compromised by price as a common denominator by
which all new contracts were negotiated. In a highly competitive industry, the new radical
form of contracting led to short-term tradeoffs between quality and profit and a heightened
sense of competitive rivalry. This led to large-scale claims at completion as a result of
changing the systems.
The Department of Public Works and Services as a powerful buyer strictly enforced the new
systems and contractors had little opportunity to participate or voice their ongoing concerns:
‘The buyer is very set in their ways, very hierarchy set, very much public service’ (Contracting
manager). Observed work routines followed a standard response: ‘They [Public Works and
Services] ram a contract down your throat’ …’You cannot do any work here unless you have a
AWP, which is an approved work plan, right, none of us have ever come up against that
before’ (Subcontractor manager).
17
The researchers found two competing stories. While Government officials were adamant that
contracts were not awarded simply on price, nearly all contractor responses disagreed:
‘Everything gets down too price no matter what they say. It’s getting harder each time round’
(Contracting manager). Long-standing traditions were now replaced by rules and increased
bureaucracy: ‘I had 3 folders on that wall 5 years ago, now I’ve got 23’ (Contracting
manager). While the new tendering and contracting procedures were designed to increase
efficiency by setting strong rules and procedures related to quality, price, time, and
occupational health and safety, most contractors struggled with endless reporting procedures:
‘Again, the majority of people aren’t across the paperwork, there is simply too much of it, and
the average man in the street, especially small contractors which make up about 90% of
workers, it is too much’ (Subcontractor manager). Traditions for employing university
graduates were also replaced as engineers and architects cut costs in response to the new
systems. With the old system, most businesses employed between 10 and 50 new apprentices.
The new systems however led to more people leaving the industry than those entering.
Contractors blamed the Government for not creating the right conditions for employers to hire
new workers, while the Government blamed contractors for a lack of long-term employee
planning:
‘The Governments done nothing to improve this in fact we have a situation with the
same number of apprentices coming into the industry in 2001 as 1991, that’s ten
years of zero growth’ (Union official)….’We want to see evidence that firms are
training. You know there should be one apprentice for every four workers’ (Client
manager).
The new systems and routines governed what was learned and how contracting routines were
negotiated and traded. The imposed practices led to large scale rebellion with little
understanding of the change effects: ‘The right hand doesn’t know what the left is doing mate’
…’Your men will leave you and go and work elsewhere, you know, you can’t afford to lose
men you know’ (Subcontractor manager). Shorter decision timelines and increased pressure to
follow the new practices led to an increase in abuse by some contractors over contract
variations and increased risk: ‘They are never brought to account for their shortcomings and
errors’….’The buyer is getting rid of [their risk] onto the builder, and the builder passes it on
to the subcontractor, it’s a nightmare’ (Association manager). The new systems replaced the
more ‘negotiated’ form of learning practices previously employed and embedded in the
18
organisation code of the supply chain. Accordingly, institutional learning was poorly received
with little commitment shown by all actors.
Change and the Organisation Code
With the new emphasis on lump-sum contracts, the organisation code of traditional beliefs
where learning was experimental in the design/construct tendering process mattered little to
large buyers. This may be partly due to the intense rivalry resulting from an unusually large
demand for construction projects in Australia in the late 1990s and beyond such as the
Olympic Village in Sydney, where the power of buyers was all-embracing. It should be noted
that the industry had a long-standing history of traditional supply chain relationships and new
radical type reorientation threatened strongly valued traditions in the existing organization
code. Secondary data revealed that a call for increased efficiency by Government departments
coupled with reduced funding for public works fuelled the discrepancies between traditional
contracting and the new procedures. The new change was a form of reorientation. While the
latter was meant to radically transform and break the alignment between existing structures
and systems (Nadler and Tushman, 1989), by transforming organisational arrangements, this
had the effect of installing a whole new system of work structures effectively dismantling the
methodical structures of the past.
The existing organisation code was based not only on methodical learning but also a network
of semi-structured relationships that resembled learning routines common to the emergent
mode. These relationships made up for the restrictions in existing systems that exploited the
past by exploring through mutual consent and bargaining projects of the future. In a sense,
interdependent contractors recognised the inadequacies of the existing systems by trading,
negotiating, and adapting their work systems for the benefit of the whole supply chain. While
fine tuning and adaptation was common, reorientation was achieved through constant
interaction and synthesis. While exploitation of existing systems occurred through daily
interaction and adjustment, exploration was also observed on the basis of long-standing
relationships and mutual commitment to new innovations in building quality. The new
institutional system threatened the organisation code however by attacking not only
methodical learning behaviours but also emergent learning by compromising the synthetic and
interactive routines that allowed for continuous exploration similar to notion of continuous
change highlighted by Brown and Eisenhardt (1997). Consequently, conflict increased
between construction agencies, and claims and counterclaims became the order of the day.
19
Most contractors felt ‘obliged’ or ‘bullied’ into changing their procedures to reflect the new
practices. Consistent with Schein’s (2004) emphasis on ‘value shaping behaviour’, powerful
buyers replaced old customs and traditions which led to negative change commitment on
behalf of supply chain actors. The double-edged sword here was non-compliance. Future
contracts were negated if actors failed to adhere to the change and firm survival threatened.
While the Government could not be expected to stand aside and allow traditional practices to
downgrade improved efficiencies when these were at fault, the radically new overhauled
practices faced widespread criticism and negative commitment: ‘Down the bottom line, this
makes ideas and everything else just go’ …’You talk of learning, there is no time the way
paperwork is forced on to business’ (Subcontracting manager).
The original perception of the main buyer (the Department of Public Works) related to
changing for the better existing procedures. Context documentary analysis revealed new
government manifestos from 1997 through to 2002 outlining the importance of a ‘seamless’
industry. The institutional logic here was to increase efficiency from the furthest back state of
production to the final stage by overhauling work practices. Many official reports suggested
that this would lead to more advanced creativity, increased innovative efforts, and increased
efficiency. Each stage of contracting in the supply chain was meant to reflect the new dynamic
learning. Other new procedures were also written related to occupational health and safety,
quality standards, and contracting procedures for doing business. The researchers noted that
most initiatives were devised separate from other stakeholders. The resulting institutional
learning was an emergent form but one incorrectly aligned to the existing organisation code
combining long-standing methodical traditions and emergent learning behaviours of informal
yet highly effective negotiated agreements between actors. The negative commitment to the
change by all interdependent actors was not what was intended.
Learning Modes
The organisation code that dominated the traditional supply chain with a delicate balance
between exploitation and exploration, method versus emergence, was suddenly replaced by
new rules and procedures, endless amounts of paperwork, and a learning mode dominated by
new analytical and structured work processes. The dramatic increase in structured routines (as
distinct from a balance between both methodical and emergent) led to heightened litigation.
Professional architects noted that some professionals would be ‘driven out of the industry’
because of the cost of professional liability. While in the old days, contractor collaboration
20
was largely a matter of discussing problems over ‘a few beers at the end of the day,’ the new
practices were driving the industry towards greater, rather than less, litigation: ‘We have
enormous professional liability problems, professional ability insurance, enormous, $200,000
for this year; it is an incredible amount of money’ (Consulting architect).
An analysis of the thematic data revealed a plethora of new structured work routines related to
tendering, quality control, safety reports, risk procedures, and work practices. With the old
system, the client managed the whole process reflecting much evidence of careful design and
strong consideration of other viewpoints. The methodical practices had been well established
over dozens of years. Designs reflected an interactive process between all players in the
industry and learning was both methodical yet emergent through intuitive and novel
relationships where the whole was greater than the sum of its parts (Miller, 1996). New
possibilities to existing designs were thus common reflecting a more emergent mode of
learning. While price was important in design/construct, more emphasis was placed on the
actual design itself. Consequently, fees reflected the degree of planning.
On the basis of the benefits that were promoted as part of the new change, the researchers
expected to observe a more emergent learning mode that was aligned to the reorientation
change type. We expected that this would be evident in at least three common strategies: 1)
The Government as the dominant buyer and rewarder of contracts, would assess the current
mode of learning in the industry by extensive analysis, 2) the degree to which the industry was
‘change ready’ would be determined, and 3) some form of training would be available to equip
managers to deal with the change. The researchers carefully analysed the thematic data by
extracting important relationships and refining and building associations which is consistent
with theory building (Miles and Huberman, 1993), in an effort to detect the strategies. Instead,
each matrix revealed the winding back of both the interactive and synthetic learning elements
by replacing these with an even more rigid and highly structured set of methodical learning
routines.
The previous emergent mode that led industry players to interact in new and improved ways to
discover core rigidities in business practices had now disappeared. At best, reorientation led to
improved efficiency in buyer procedures from the buyer’s perspective. At worst, the change
process was poorly aligned to the organisation code and the existing learning behaviours with
the effect of replacing the existing code with a new form of methodical learning. The net result
21
was less interaction and more anxiety. Greater interaction in the old system may have led to an
increase in new opportunities and novel ideas for solving problems that at least in part
matched reorientation efforts. Instead, the new practices mainly reinforced a bureaucratic
system that perpetuated routine practices in a way that impoverished future learning: ‘Control
means that you now have managed creativity. Be creative until 3 o’clock in the afternoon and
your creation must deliver a building design’ (Consulting engineer). A more emergent learning
environment might have resulted in increased productivity and new innovation but observed
work routines found the opposite: ‘What innovation is happening in architects offices,
probably bugger [sic] all, because the fees have been cut, and innovation takes time’
(Consulting architect).
If anything, the new learning routines resembled a highly structured learning environment
where procedures and rules held sway over new and novel ways of doing business:
‘It is no longer good enough to just do a safety walk on a job and issue instructions to
people. You’ve now got to go back afterwards and sign off what was checked and
fixed on such and such a job at such and such a time. It’s a big problem really as we
don’t have time to constantly tick off things’ (Contracting manager).
In an industry wishing to rid itself of inefficient practices, the researchers were surprised to
observe a learning environment that increased the level of inefficiency as a result of the
change. Quite clearly, the type of change imposed was poorly aligned to the existing learning
mode and the interplay between learning and change was both constrained and impoverished.
Discussion
Research Question 1 – Learning Modes
For RQ 1, the dominant form of learning prior to the change was based on both the methodical
and emergent mode. The latter was not formally recognised throughout the industry but rather
informally acted out in all supply chain activities leading to an unexpected finding. When left
to their own devices before the imposition of the institutional element, construction actors had
developed a set of work practices and procedures largely based on the social construction of
reality (Weick, 2001: Burrell and Morgan, 1979). All actors socially crafted a set of work
routines consistent with the emergent mode of learning. Interpretations of complex systems
22
and structures such as engineering and architectural designs were based on a mix of
experimental and interactive learning by changing and incrementally adjusting plans in the
light of new information. Hundreds of minute details were often scrutinised by actors using the
synthetic elements to search for new connections and assumptions in such a way that key
patterns and hidden meanings surfaced (Miller, 1996; Murray and Chapman, 2003). This
occurred not only in relation to building and infrastructure designs but also in the intricate
network of relationships that were needed to action large construction projects. Notably, these
routines were crafted free of intervention and formal management. They became part of the
informal established systems and structures adhered to in the organisation code yet not
formally recognised by dominant institutional actors, or as part of a formal system. This
appears to confirm what Gupta et al. (2006) suggest that reliability-enhancing learning that
results from human and organizational attempts to replicate past routines is often overlooked,
and that interpersonal learning frequently occurs outside of the code (Miller et al., 2006). It
also confirms previous research that both exploitation and exploration are needed in the
pursuit of new learning (March, 2006). Indeed, the emergent mode is consistent with higherlevel learning behaviours common in dynamic interactions in modern workplaces (Murray and
Donegan, 2003; Miller et al., 2006).
While structured learning in the methodical mode is perhaps the best way to describe the
dominant learning routines used, reorientation led to more layers of structure based on highly
patterned and rigid procedures. The dismantling of the emergent mode to accommodate an
even stronger methodical form was an interesting observation. What this suggests is that
while reorientation is at the vanguard of strategic change, a poor understanding of the code
led to more exploitation by ramping up the methodical learning routines. So reorientation
efforts which were designed to be exploratory led to more exploitation. At no time did
reorientation lead to change success. All actors interviewed expressed great disenchantment
with the new system and negative change commitment. What this suggests is that both
learning modes have to be tied to change processes and that it is both necessary and important
to determine current learning behaviours (Fiol and Lyles, 1986; Murray and Donegan, 2003;
Crossan and Berdrow, 2003). That is, the current mode of learning should be assessed to
determine whether actors: 1) have the skills and capabilities to make the change, 2) have a
level of knowledge consistent with the type of change required, and 3) understand the
importance of the change so that existing ties to the organisation code can be relaxed, and
change successfully implemented. The interplay between learning and change should have
23
been better understood by aligning change processes with current learning behaviour.
Importantly, learning behaviours embedded within the existing beliefs and practices
commonly shared in the code should have been assessed (see Figure 1).
Consistent with theory, organisations exploit past practices and build on them by exploring
new possibilities (March, 2006). If new methodical routines however completely overhaul the
old ones, this has the effect of dampening morale and increasing inefficiency by destroying
the balance between voluntarism and determinism, method versus emergence (Miller, 1996).
The previous code had already reflected the balance between the methodical and emergent
modes. However, the new structured routines had the effect of stifling new learning and
effectively impoverishing the previous learning modes if not removing them altogether. This
was acted out through increased conflicts and frustrations placing even further pressure on an
already strained supply chain. For RQ1, the quantity and quality of learning are critical
determinants of systems success. When these are not well understood, this has the effect of
dismantling or replacing the existing learning behaviours stifling new explorations and
placing under threat the exploitation of existing routines.
Research Question 2 – Institutional Learning
In moving to RQ 2, institutional hegemony led to large-scale disruptions in the supply chain
meaning that the new institutional learning behaviour had a large negative effect on both
change and learning. All actors appeared to be affected. The previous institutional context
was one that tolerated the status quo where actors interacted and socially constructed an
environment where the whole was greater than the sum of its parts. The intersection of actor
inputs with the new system of change however was severely disrupted and changed by the
dominant power broker represented by the Government as buyer. The old institutional logic
was completely overhauled. While the new institutional change was designed to increase
efficiency and productivity, the opposite occurred with large cost overruns, claims and
counterclaims, and increased litigation and risk.
The findings for RQ 2 confirm the importance of understanding the effects of institutional
learning. The interpretation that powerful actors assimilate their values on organisation/s, and
that radical change efforts often lead to unresolved change tracks are important previous
findings (Hinings and Greenwood, 1988; Greenwood and Hinings, 1996). However, since
change success is very much dependent on institutional dominance whose influence is
24
manifest by new institutional learning behaviours, the effects of these on existing work
practices needs to be better defined and understood. New institutional learning behaviours
need to be carefully determined and matched to the existing ones before the change process
can be successful. Moreover, an assessment of institutional learning would need to ensure
several things: 1) that actors affected by the change will be committed to it, 2) that the new
change has some likelihood of success and 3) that as a result of the new learning, work
practices will be better off and improve efficiency. While actors needed to obey the new laws
and systems for economic survival which is common in institutional control, increased
tradeoffs between quality and cost, and learning and change, were experienced throughout the
supply chain. This left the researchers to ponder the effects of institutional change on an
industry clearly not ready for it.
Research Question 3 – Type of Change
For RQ 3, the existing organisation code coupled with the dominant learning mode was not
aligned to reorientation as a change type. The interplay between learning and change was
impoverished. Reorientation was designed to break the current association between strategy,
structure, and systems by installing a completely new contracting process. As already noted,
the old system was deeply grounded in the code underpinned by long-standing traditions
reflected in methodical routines and interactions and synthetic learning which often occurred
outside the code (Miller et al., 2006). Commitment to the change by all actors was negative.
Since the code was strong and reactions to the change negative (see Figure 2), any further
attempts by the buyer to enforce change was confronted by an even stronger resolve to resist
on the part of actors. Change success was less likely under such circumstances.
For the most part, it is difficult to conceive of change success in the current study when the
balance between exploiting and exploring was compromised. Interestingly, a strong balance
between exploitation and exploration occurred frequently before the change yet was
spectacularly interrupted after it. Before the change attempts, the industry reflected
ambidextrous capabilities (Tushman and O’Reilly, 1996; Benner and Tushman, 2003), in
such a way that organisational architecture managed both the multiple and conflicting
demands between exploiting and exploring such that the code reflected an ability to adjust to
continuous change efforts (Eisenhardt and Martin, 1997). As Meyer and Stensaker (2006)
suggest however, ‘managers consistently neglect or underestimate the adverse effects of
implementing change’ (2006, p. 219), and that a possible cause is the failure of management
25
generally to recognise the balance between resources deployed in change processes as well as
daily operations.
Contribution to Theory
Our contribution in this study has been to elevate the role of learning behaviour in change
implementation by pointing to the importance of alignment. What need to be better aligned
are the actual learning behaviours and the type of change and how it is implemented. The
need to understand the interplay between learning and change is thus important and
necessary. While a long list of scholars have noted the tensions between exploitation and
exploration (March, 1991; 2006; Crossan and Berdrow, 2003; Gupta et al., 2006; Siggelkow
and Rivkin, 2006), our findings confirm that both methodical and emergent learning
behaviour are important in understanding these tensions. The interplay between the modes
and change in terms of renewal attempts adds a different emphasis to existing studies.
Organisations embracing change may overlook the unintended consequences of destroying
the logic of past successes by dismantling systems of learning that lead to successful
exploitation. Similarly, change failures may reflect a lack of understanding of the interplay
between learning behaviour and change implementation. As a result of this study, future
change success might be determined by a number of factors: 1) a better understanding of
existing learning processes 2) greater understanding of the quantity and quality of learning
required, 3) greater realisation of the consequences of the type of change selected, 4) greater
understanding of the existing organisation code, 5) stronger appreciation of the relationships
between the commitment to change and the existing organisation code, and 6) a better
understanding of the tensions between exploitation and exploration embedded within existing
learning systems. These points are not meant as a prescriptive list or a panacea; rather, they
might form part of an objective analysis in understanding the junction or intersection between
learning behaviours and types of change.
Finally, a further contribution is to challenge the idea that change and learning respectively
are independent phenomena and communities of knowledge with their own beginning and
end. We have illustrated the importance of the interplay between each in such a way that
change efforts may be better determined on the basis of learning. Similarly, learning as a
means in itself matters little. It requires as Hedberg (1981) suggests both cognition and action,
and as Miller (1996) contends an understanding between determinism and voluntarism in
terms of thought and action. The cognition of change is perhaps the idea that different change
26
types will be successful simply on the basis that change is needed. Action however is required
for the change to be successful and the importance of determining learning behaviour is a
useful way to think about change attempts. Here we suggest a greater need for managers to
determine the current modes of learning and how these might be best aligned to change
implementation. We also add to institutional learning knowledge by bringing to the forefront
the learning behaviours themselves. Similarly, on the basis of our discussions, we have
attempted to illustrate what occurs in practice when the organisational code is changed and
how this upsets the balance between exploitation and exploration. While the latter is a well
known finding of previous studies, our contribution is to underline the critical interplay of the
actual learning behaviours used in exploitive and exploratory change attempts. Figure 1
reflects the interplay between learning and change and is the basis of a framework for
subsequent studies.
Directions for Future Research
While the findings of the study might be used to explore similar learning activities across the
industry, future research might explore a broader group of managers at different levels within
the same industry. The research design for this paper may be replicable to other industries.
Other scholars might explore the findings using the same or similar constructs. It is expected
that other industries may be less dependant on supplier firms and that learning modes will be
implemented differently. Positive change commitment is indeed expected in other industries
where the dual realities of learning and change are more evident. The computer and
electronics industries are a case in point where continuous reorientation and transformation is
necessary (Brown and Eisenhardt, 1997). The interplay between learning and change when
change commitment is positive and the organisation code strong would be useful in building
on the ideas of the current study. Future studies might also examine different respondents
other than managers such as field staff and office personnel potentially exploring rich sources
of data at different organizational levels. Other studies that examine the longitudinal effects of
learning modes and change are likely to be significant.
It should be noted that the methodology is limited and confined by the two learning modes
and indeed by the four change types. Other types of learning such as learning styles or
learning levels might be useful approaches for future studies. Similarly, specific change
processes and tools as distinct from types of change may form useful approaches in
subsequent work. Future studies might also expand the institutional context from a learning
27
perspective leading to a greater understanding of institutional change. We have attempted to
highlight the links in this study between institutional dominance on the one hand and learning
behaviours on the other. Studies that single out teams in the feed back or feed forward process
may be a useful approach between learning and change. More specific research is needed to
examine in more detail the actual workplace routines that underpin each of the learning
modes and their elements. One idea might be to explore routines in terms of their actual
implementation in workplace settings rather than across an industry. We hope that other
scholars however may find the approach in this study important for further research and
theory building.
Conclusion
Previously, organisational learning as a community of knowledge appears to form separate
yet connected studies related to individual learning processes (Argyris, 1993), knowledge
management (Orlikowski, 2002), and organisational learning (Huber, 1991; Crossan and
Berdrow, 2003; Easterby-Smith and Araujo, 1999), among others. Similarly, change has often
been studied as a process with a clear beginning and end (Nadler and Tushman, 1989; Meyer
and Stensaker, 2006). Scholars of strategic renewal have attempted to link the two concepts
through exploitation and exploration by concentrating on building an ambidextrous
organisational architecture (Bennis and Tushman, 2003; Crossan and Berdrow, 2003; Gupta
et al., 2006), and learning and change has been clearly linked to performance (Arthur and
Huntley, 2005). What is important in all studies is the need to improve how an organisation
learns and increases its knowledge to do things better or different by exploiting the past and
exploring the future. However, the interplay between learning and change has not been clear.
Strong associations exist between all research questions in this study to suggest that both the
quantity and quality of learning is important for successful change and that the type of change
should be aligned to learning behaviours. Indeed, it is difficult to conceive of a situation in
which change success is not influenced by the dominant learning mode, and the actual change
process itself. We started out by pointing to the value of developing a framework which might
move us closer to theory building. Figure 1 illustrates this framework and the discussion links
the variables. In particular, the interplay between learning and change in this study has been
designed to address, at least in part, an important research gap between the two.
28
References
Argyris, C. (1993). ‘Education for leading-learning’. Organizational Dynamics. 21 (3): pp 517.
Arthur, J.B. Huntley, C.L. (2005), ‘Ramping up the organizational learning curve: Assessing
the impact of deliberate learning on organizational performance under gainsharing’.
Academy of Management Journal, 48, 6, 1159-1170.
Benner, M.J. Tushman, M.L. (2003), ‘Exploitation , exploration, and process management:
The productivity dilemma revisited’. Academy of Management Review, 2: 238-256
Berger, P.L. & T. Luckmann (1966), The Social Construction of Reality: A Treatise in the
Sociology of Knowledge, Doubleday, New York.
Bessant, J. (2004). ‘Managing Innovation beyond the steady state’. Continuous innovation:
Strategic Priorities for the Global Knowledge Economy, 5th International CINET
Conference, September, Sydney.
Bessant J, & Caffyn S. (1996). ‘High involvement innovation through continuous
improvement’. International Journal of Technology Management 14 (1): 7-28.
Boal, K.B. & Schultz, P.L. (In Press), ‘Storytelling, Time, and Evolution: The Role of
Strategic Leadership in Complex Adaptive Systems’, The Leadership Quarterly
(forthcoming).
Brown, S.L. Eisenhardt, K.M. (1997), ‘The art of continuous change: Linking complexity
theory and time-paced evolution in relentlessly shifting organizations’. Administrative
Science Quarterley, 42, 1-34
Burgelman, R.A. (2002), ‘Strategy as vector and the inertia of co-evolutionary lock-in’.
Administrative Science Quarterly, 47: 325-357.
Burrell, G. & Morgan, G. (1979). Sociological paradigms and organisational analysis.
London: Heinemann.
Clegg, S. R. (2000) Frameworks of Power and Resistance, London: Sage
Crossan, M.M. Lane, H.W. White, R.E. (1999), ‘An organizational learning framework: From
intuition to institution’. Academy of Management Review, Vol. 24, (3), pp522-537.
Crossan, M.M. & Berdrow, I. (2003), ‘Organizational learning and strategic renewal’.
Strategic Management Journal. 24: 1087-1105.
DeGeus, A.P. (1988), ‘Planning as learning’. Harvard Business Review. 66 (2): 70-74.
Denis, J-L, Langley, A. Pineault, M. (2000), ‘Becoming a Leader in a Complex
Organization’. Journal of Management Studies, 37: 8, 1063-1099.
29
Easterby-Smith, M. Araujo, L. (1999), ‘Organizational learning: Current debates and
opportunities’. In M. Easterby-smith, J. Burgoyne, and L. Araujo (Eds), Organizational
learning and the learning organization: Developments in theory and practice. 1-21,
Thousand Oaks, CA:Sage.
Edmondson, A.C., Roberto, M.A. & Watkins, M.D. (2003), ‘A dynamic model of top
management team effectiveness: managing unstructured task streams’. Leadership
Quarterly, vol. 14, no. 3, pp. 297-325.
Edmondson, A.C. Bohmer, R.M. Pisano, G.P. (2001), ‘Disrupted routines: Team learning and
new technology implementation in hospitals’. Administrative Science Quarterly, 46: 685716.
Eisenhardt, K.M. & Martin. J.A. (2000), ‘Dynamic capabilities: What are they’? Strategic
Management Journal, 21: 1105-1121.
Fiol, C.M. & Lyles, M.A. (1985). ‘Organizational learning’. Academy of Management
Review, 10 (4): 803-813.
Gersick, C.J.G. (1991), ‘Revolutionary change theories: A multilevel exploration of the
punctuated equilibrium paradigm’. Academy of Management Review, 16 (1), 10-36.
Gordon, R.D. (2002), ‘Conceptualizing leadership with respect to its historical-contextual
antecedents to power’. The Leadership Quarterly, 13: 151-167.
Greenwood, R. & Hinings, C.R. (1996), ‘Understanding Radical Organizational Change:
Bringing together the old and the new institutionalism’. Academy of Management Review,
October, 21, 4, 1022-1054.
Gupta, A.K. Smith, K.G. Shalley, C.E. (2006), ‘The interplay between exploration and
exploitation’. Academy of Management Journal, 49 (4), 693-706.
Hedberg, B. (1981). ‘How organizations learn and unlearn’. In P.C. Nystrom & W.H.
Starbuck (Eds.), Handbook of organizational design. London: Oxford University Press. pp
8.27.
Hinings, C.R. & Greenwood, R. (1988a), The dynamics of strategic change. Oxford, England:
Basil Blackwell.
Huber, G.P. (1991), ‘Organizational Learning: The Contributing Processes and the
Literatures’. Organization Science, Vol. 2, No 1.
Kim, D.H. (1993), ‘The Link between Individual and Organizational Learning’, Sloan
Management Review, Fall.
Kontoghiorghes, C. Awbrey, S.M. Feurig, P.L. (2005), ‘Examining the relationship between
learning organization characteristics and change adaptation, innovation, and organizational
performance’. Human resource development quarterly, 16, (2), pp. 185-211.
Lee-Kelley, L. & Blackman, D. (2005). ‘More than shared goals: the impact of mental
30
Models on team innovation and learning’. Journal of Innovation and Learning, 2 (1): 11-25.
Leonard-Barton, D. (1992), ‘The factory as a learning laboratory’. Sloan Management
Review, Fall.
March, J.G. (2006), ‘Rationality, foolishness, and adaptive intelligence’. Strategic
Management Journal, 27: 201-214.
March, J.G. (1991), ‘Exploration and exploitation in organizational learning’. Organization
Science, 2: 71-87.
Marks, M.A. Mathieu, J.E. Zaccaro, S.J. (2001), ‘A temporally based framework and
taxonomy of team processes’. Academy of Management Review, 26 (3), 356-376.
Meyer, C.B. Stensaker, I.G. (2006), ‘Developing capacity of change’. Journal of Change
Management, 6 (2), 217-231.
Miller, D. (1996). ‘A Preliminary Typology of Organizational Learning: Synthesizing the
Literature’. Journal of Management, 22 (3): 485-505
Miller, K.D. Zhao, M. Calantone, R.J. (2006), ‘Adding interpersonal learning and tacit
knowledge to March’s exploration-exploitation model’, Academy of Management Journal,
49 (4), 709-722.
Miles, M.B. & Huberman, A.M (1994), Qualitative Data Analysis. 2nd Edition, London:Sage.
Mintzberg, H. Ahlstrand, B. Lampel, J. (1998), Strategy safari: A guided tour through the
wilds of strategic management. Free Press: New York.
Murray, P. & Blackman, D. (2006), ‘Managing innovation through social architecture,
learning and competencies: A new conceptual approach’. Knowledge and Process
Management. Volume 13, Issue 3
Murray, P. & Donegan, K. (2003). ‘Empirical linkages between firm competencies and
organizational learning’. The Learning Organization: An International Journal, 10 (1): 5162.
Murray, P. & Chapman, R. (2003). ‘From continuous improvement to organizational
learning’. The Learning Organization, 10: pp.272-282.
Nadler, D.A. & Tushman, M.L. (1989), ‘Organizational frame bending: Principles for
managing reorientation’. The Academy of Management Executive, Vol. 3, No. 3. 194-204.
Orlikowski, W.J. (2002), ‘Knowing in practice: Enacting a collective capability in distributed
organization’. Organization Studies, 13: 249-273.
Oswick, C., Anthony, P., Keenoy, T., Mangham, I. & Grant, D. (2000) ‘A dialogic analysis of
organizational learning’. Journal of Management Studies, 37 (6): 887-901.
31
Romanelli, E. Tushman, M.L. (1994), ‘Organizational transformation as punctuated
equilibrium: An empirical test’. Academy of Management Journal, 5: 1141-1166.
Schein, E. (2004) Organizational Culture and Leadership, Third Edition. New York: Wiley
Publishers.
Schneider, S.C. & Barsoux, JL. (1997), Exploring Culture. Prentice Hall. London.
Schwandt, T.A. (1994), ‘Introduction: Entering the Field of Qualitative Research’. In
Handbook of Qualitative Research, N.K. Denzin and Y.S. Lincoln, (Eds), Sage
Publications, Thousand Oaks, pp. 118-37.
Tushman, M.L. O’Reilly, C.A. (1996), ‘Ambidextrous Organizations: Managing evolutionary
and revolutionary change’. California Management Review, Vol. 38, 4. 8-30.
Tushman, M.L. Newman, W.H. Romanelli, E. (1986), ‘Convergence and upheaval: Managing
the unsteady pace of organizational evolution’. California Management Review, 29: 1 2944.
Wageman, R. (1997). ‘Critical success factors for creating superb self-managing teams’.
Organizational Dynamics, Summer: 49-61.
Weick, K. (2001), Making Sense of Organization, Blackwell, Oxford.
Van Maanen, J. (1988), Tales of the field: On writing ethnography. Chicago: University of
Chicago Press.
32
TABLES AND FIGURES
Table 1: Interview Respondents by Contractor Type
Type of
Respondent
Management Position and Number of Respondents
Total
companies
Contractor
Managers
Site engineers (1), Contracting Managers (7), site managers (4),
occupational health and safety managers (1), employee relations
managers (1), chief estimators (1), general managers (1),
environmental managers (1)
Total = 16 managers
Directors – working managers (8)
Total = 8 managers
3 (CM 1CM3)
State Contracting Manager (1), Manager Policy Division (1),
General Manager (1) Director (1), Manager Procurement and
Industry Policy (1)
Total = 5 managers
Directors (8), Chief Executive Officer (2)
Total = 10 managers
1 (C1)
NSW Director (1), Managing Director (1)
Total = 2 managers
Chief Executive Officer (1)
Total = 1 manager
1 (A1)
Sub-contractor
managers
Clients
Consultants
Associations
Unions
2 (SM 1-SM2)
2 (CT 1-CT 2)
1 (U1)
Table 2: Thematic Conceptual Matrix
Learning
Modes
Change
Types
Tuning
Adapting
Reorientation
Recreation
Learning
Elements
Structured
Methodical
Analytical
Experimental
Synthetic
The relationship between learning
and change determined by observing
the current work routines of actors
Emergent
Interactive
Institutional
33
Table 3: Unstructured Questions
Broad Theme
Organisation code
Questions






Learning modes






Change type






In relation to contracting, tell me about how you have
done things in the past?
In what ways do these beliefs influence your current
work environment?
Can you explain the recent changes to tendering and
contracting?
Do you prefer the new or old system? Why?
To what extent do the old practices still exist?
Is tradition valued? Why?
How do the new changes influence the systems and
procedures of your work?
Tell me about the Government’s motives?
In what ways have learning processes changed?
Tell me about any new and novel learning initiatives?
How do suppliers interact in the supply chain?
Tell me about standard work methods and practices.
How do they influence current work practices?
Has change been incremental and gradual?
What do you think about the way the change was
implemented?
Has the industry been committed to the change?
How has your workplace supported the change?
If little support, why?
What were common change processes in the past?
Figure 1: A Framework of Learning and Change
analytical, structured,
experimental
Institutional
Contexts
Methodical
Tuning
Adapting
Learning
Modes
Change
Types
Emergent
Reorientation
Recreation
Change Commitment
(Positive or negative)
Workplace
Change
Organisation
Code
(Strong or weak)
synthetic, interactive,
institutional
34
Figure 2: Change Relationships and the Organisation Code
Organisation Code
Change Commitment
Positive
Negative
Strong
Weak
Quadrant
B
A
C
D
35