Description
Despite anecdotal evidence questioning the relevance of management accounting in firms pursuing “new”
mamtfacturing strategies, the exact nature and role of accounting information for controlling manut&xuring
activities has received little empirical attention
Pergamon
Accounting 0r#anization.r and Society, Vol. 20, No. 4. pp. 241-258, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0361-3682/95 S9.50+0.00
0361~3682(94)E0014-L
THE IMPACT OF MANUFACTURING FLEXIBILITY ON MANAGEMENT
CONTROL SYSTEM DESIGN*
MARGARET A. ABERNETHY
The University of Melbourne
and
ANNE M. LILLIS
Deakin University
Abstract
Despite anecdotal evidence questioning the relevance of management accounting in firms pursuing “new”
mamtfacturing strategies, the exact nature and role of accounting information for controlling manut&xuring
activities has received little empirical attention. Thii study examines the impact of manufacturing flexibility
on the design of management control systems. In particular, we examine the implications of flexibiity on
the use of efficiency-based performance measurement systems and the use of integrative liaison devices. We
collected data, through semi-structured interviews, from general managers in manufacturing firms in
Melbourne, Australia These interview data were used to develop measures for the constructs of interest.
Exploratory analysis of the data indicates that integrative liaison devices are a critical form of control in
managing the implementation of flexible manufacturing strategies. However, the role of accounting or other
efficiency-based performance measures declines in those firms with a commitment to manufacturing flexibility.
Confidence in traditional manufacturing manage-
ment accounting systems has eroded significantly
in the past five years. The relevance of account-
ing is most seriously questioned in lirms
implementing “new” manufacturing strategies
(Kaplan, 1990). While the accounting literature
provides anecdotal evidence that traditional
management practices are inappropriate in
these settings, there has been little systematic
empirical study of the link between manufactur-
ing strategy and control system design. This
paper examines the impact of strategic com-
mitments to manufacturing flexibility on two
critical aspects of a lirm’s management control
system (MCS): the performance measurement
system, and the structural arrangements required
to co-ordinate production activities.
The following section reviews the literature
which has examined the link between control
system design and strategic commitments,
concluding with a number of exploratory
propositions. Subsequent sections address
methodology, variable measurement, results,
concluding comments and directions for further
research.
*We wish to acknowledge contributions from participants at the ABO Conference (Bond University, July 1991) The
Management Accounting Research Conference (Atlanta, October 199 1) The Strategic Management Accounting Conference
(University of New South Wales, October 1992) and helpful comments from Peter Brownell, Rob Chenhall, Kenneth
Merchant, Frank Collins, Geoff Burrows and Christine Jubb. We are also grateful for the detailed advice given by Anthony
Hopwood and the anonymous referees. We wish to thank the Department of Accounting and Finance, Monash University,
Bowater Faculty of Business and the Monash Medical Centre for the tinancial support provided for this project.
241
242 M. A. ABERNE’ITN and A. M. LILLIS
CONTROL SYSTEM DESIGN AND
STRATEGIC COMMITMENT
Manufacturing flexibility is reflected in a firm’s
ability to respond to market demands by
switching from one product to another through
co-ordinated policies and actions (Nemetz &
Fry, 1988) and a willingness or capacity to offer
product variations (Buffa, 1980; Bowen et al.,
1989). It is a strategy which attempts to
maximize differentiation (Parthasarthy & Sethi,
1993). The manufacturing approach associated
with this strategy differs from that of a lirm
competing on the basis of cost leadership.
Typically, firms committed to cost leadership
have invested in manufacturing technology
which enables long production runs of standard-
ized products (Dean & Susman, 1989; Buffi,
1984; Chase, 1990). The characteristics of
traditional management accounting systems,
which have been designed to support mass
production technology, are potentially at odds
with a strategic commitment to manufacturing
flexibility (Kaplan, 1990). These systems empha-
size cost control through techniques such as
responsibility accounting, standard costing and
the promotion of production efficiency through
variance analysis. Management accounting
systems, with their focus on task segregation and
efficiency, are particularly well suited for mass
producers of standard products with unchang-
ing characteristics (Simons, 1990; Drucker,
1990; Bowen et al., 1989; Nemetz Bz Fry, 1988).
Manufacturing flexibility is a customer-
responsive strategy and is used as a “competitive
weapon” to create enduring competitive advan-
tages (Chrisman et al., 1988, p. 4 15). Respond-
ing to customer-driven demands requires control
systems which facilitate product customization
rather than product standardization. These
demands impose complex, multifunctional
problems which are not well managed by
systems focusing on efficiency and task segrega-
tion. To the extent that accounting performance
measurement systems focus on efficiency, they
are inconsistent with customization and flexibility
(Kaplan, 1990). Furthermore, the task segrega-
tion implicit in responsibility accounting and
variance analysis is antithetical to the cross-
functional co-ordination required to meet
customer-driven demands (Bowen et al, 1989).
The implications of manufacturing flexibility for
performance measurement system design and
the structural arrangements required to co-
ordinate multifunctional tasks are elaborated in
the following sections.
Manufacturing J exibility and performance
measurement systems
Few studies have examined empirically the
link between manufacturing flexibility and the
design of performance measurement systems in
manufacturing. Much of the research examining
the link between strategy and control systems
has done so at the strategic business unit (SBU)
level, rather than at the manufacturing level
(Govindarajan, 1988; Govindarajan & Gupta,
1985; Merchant, 1985; Simons, 1987, 1990).
This literature, however, suggests that the
role of accounting in measuring performance
depends on the strategic commitments of the
SBU. Formal cost control mechanisms, such as
standard costing systems, appear to be particu-
larly well suited for measuring performance in
manufacturing firms with high product standard-
ization, relatively stable production processes
and where the strategic priority is typically one
of cost minimization (Macintosh, 1985; Miller,
1988; Hirst, 1983; Brownell & Merchant, 1990;
Govindarajan, 1988).
Research at the SBU level argues that firms
competing on the basis of product differentia-
tion will move away from financial and other
efficiency-based measures of performance to
measures which support the achievement of
strategic priorities associated with differentia-
tion (Govindarajan, 1988; Simons, 1987). A
commitment to manufacturing flexibility is a
form of product ditferentiation and, as such,
creates “co-ordinative, technical and control
problems. . . ” distinct from those found in lirms
employing mass production technology (Miller,
1988, p. 281). Consequently, it is unlikely
that manufacturing activities can be effectively
controlled by assigning tasks, programming
desired behaviour and implementing efficiency-
MANUFACTURING FLEXIBILITY 243
based indicators to capture task performance
(Parthasarthy & Sethi, 1992). The absence of
standardization makes it increasingly difficult to
specify unambiguous performance standards, as
optimal relationships between inputs and out-
puts for production tasks are usually not known.
In this situation, financial measures of manu-
facturing performance become less relevant
(Brownell & Merchant, 1990; Hayes, 1977).
Successful implementation of manufacturing
flexibility requires cross-functional responsive-
ness to specific customer-initiated demands.
Effective performance is the result of the
combined efforts of those functional units
required to satisfy customer demands. This
suggests that effective performance measure-
ment will require a shift from measures which
focus on manufacturing efficiency to measures
which encourage interfunctional co-operation
and adaptation and which capture the critical
success factors related to customer-initiated
demands (Macintosh, 1985).
It is expected, therefore, that when firms
pursue flexibility there will be a shift “from
treating financial figures as the foundation for
performance measurement” to a broader set of
measures designed to support flexibility as a
strategic priority (Eccles, 199 1, p. 13 1). These
measures are likely to include qualitative and
quantitative indicators of customer responsive-
ness. For example, measures of cycle times on
product variations, delivery performance and
assessment of the ability of manufacturing to
vary product characteristics or develop new
products, are appropriate. Similarly, interfunc-
tional performance measures, such as those
which assess the strengths of interdepartment
relationships and levels of co-operation, will be
appropriate.
Manufacturing fl exi bi l i ty and structural
arrangements
A number of researchers have identified the
direct implications of manufacturing flexibility
on the structural arrangements required to
facilitate co-ordination of multifunctional tasks
(Bowen et al ., 1989; Nemetz & Fry, 1988;
Parthasarthy & Sethi, 1993). These structural
arrangements are in direct contrast to those
which are considered to be appropriate where
product standardization is high and production
processes relatively stable. These manufacturing
processes impose few co-ordination problems.
Indeed, it has often been argued that the most
efficient organizational form is one which
buffers manufacturing from outside influences
(Thompson, 1967) and which has attributes
similar to the mechanistic structure described
by Burns & Stalker ( 1961). A mechanistic
structure emphasizes high vertical and hori-
zontal task differentiation, centralized decision
making, vertical information flows and func-
tional groupings (Mint&erg, 1983; Miller, 1988).
This structure is considered efficient, as the
relationships between manufacturing and other
functional units are pooled or sequential and
the need for co-ordination between functional
units is minimal (Van de Ven et al ., 1976). Any
co-ordination between functional units can be
managed through formal standardized proce-
dures and processes (Parthasarthy & Sethi,
1993).
Structures that emphasize a high differentia-
tion of task and vertical information flows are
incongruent with a commitment to flexibility
(Parthasarthy & Sethi, 1993). A commitment to
manufacturing flexibility changes the way in
which manufacturing interacts both with external
customers and with other functional units
within the firm. To be customer responsive,
manufacturing can no longer be buffered from
the influence of customers and other functional
units (Bowen et al ., 1989). Increased inter-
actions are likely to emerge between customers
and manufacturing, and between manufacturing
and other functional units. It becomes increas-
ingly difficult to assign tasks to individual
functional units and to measure the performance
of individual tasks. Effective performance results
from “team” efforts. These teams can involve
multiple functional units and, in some cases,
customers (Larsson & Bowen, 1989). The
necessary interactions between members of the
“team” in their performance of interfunctional
tasks introduce higher levels of interdependen-
ties between manufacturing and other functional
244 M. A. ABEBNETHY and A. M. LILLIS
units. Interdependencies are likely to change
from being sequential and resemble the recipro-
cal or team interdependencies described by Van
de Ven et al . ( 1976). These forms of interdepen-
dencies can have direct implications on the
organizational arrangements required to co-
ordinate task performance effectively. Flexibility
requires “organizational arrangements to be
functionally more integrative, . . . i.e. the
structure should be organic” (Parthasarthy &
Sethi, 1993, p. 532). This organizational form
encourages co-operation and co-ordination
among functional subunits (Bruns & Waterhouse,
1975; Galbraith, 1973; Thompson, 1967).
in the field we attempt to answer the following
questions:
(1) Do lirms committed to manufacturing
flexibility use performance measures which de-
emphasize accounting and other efficiency
measures?
(2) Does a commitment to flexibility intluence
interfunctional structural arrangements? In
particular, do lirms increase their reliance on
integrative liaison devices?
It is expected that the functional interdepen-
dencies created when hrms pursue flexible
manufacturing strategies will increase the use
of integrative liaison devices to reduce the
rigidity of functional structures. Spontaneous
contact and “integrative liaison devices” (such
as task forces and committees) which allow
regular, personal, and intensive contact among
experts and decision makers of different depart-
ments, greatly facilitate collaboration among
functional units (Miller, 1988, p. 286). These
co-ordination mechanisms provide a means of
breaking down the functional barriers imposed
by mechanistic organizational structures. To the
extent that co-ordination requirements necessi-
tate the full integration of functional units, firms
may choose more “sophisticated and costly”
design alternatives such as product-line or
matrix structures (Galbraith, 1973, p. 47).
(3) Is the performance of the firm enhanced
when the manufacturing performance measure-
ment system or structural arrangements are
adapted to facilitate the implementation of
manufacturing flexibility?
Sampl e sel ecti on
Research questi ons
We have argued that Iirms will adapt their
control system to facilitate a commitment to
manufacturing flexibility. ’ Using data collected
The study is based on data collected in manu-
facturing firms located in Melbourne, Australia.
Firms were identified from the Victorian Manu-
Edcturing Directory using the Australian Standard
Industry Classification (ASIC). The selection of
industry categories was limited to those where
it was expected that there would be variation
on the flexibility dimension. Only lirms in the
Melbourne metropolitan area, and those with
more than 200 employees, were included in the
target sample.2 This resulted in an initial sample
of 81 business units. Forty-two lirms agreed to
participate, and Table 1 shows the number of
firms by size category for each industry. Using
the limited data provided by the Manufacturing
Directory, a comparison of the responding firms
to the target sample indicated no systematic
response bias.
METHODOLOGY
’ Following prior literature, we have assumed that strategic priorities are a response to a firm’s competitive environment
and that decisions concerning strategic direction also involve technological and administrative decisions to achieve strategic
priorities (Miller, 1988; Govindarajan, 1988; Parthasarthy & Sethi, 1992). It is recognized that the assumption concerning
the unidirectional nature of these relationships is simplistic. It is, however, diflicuh to establish causality with cross-
sectional data, although research in the general management literature is beginning to develop models for that purpose
(Miller et aL , 1988).
‘Several tirms included in the sample had reduced the number of employees at the time the interviews were conducted.
These firms were not excluded in the analysis.
MANUFACTURING FLEXIBILITY 245
Industry classification
TABLE 1. Sample description
Size of firms by number of employees
150-299 300-499 500-799 800-999 1000+
Paper products I 1 1
Fabricated metal products 5 6 2
Appliances, electrical equipment and machinery 7 8 2 3
Plastic and related products .)
I 1 1
Data collection
We obtained data from semi-structured inter-
views with the general manager or managing
director of the firm (the questions included in
the interview protocol are provided in Appendix
A), with each interview lasting approximately
two hours. We adopted this form of data
collection, as the link between manufacturing
flexibility and control system design has received
relatively little empirical attention. At the time
of the study there were no established measure-
ment instruments which had been subjected to
sufficient psychometric assessment for us to be
confident in either the measures’ reliability or
their validity. Difficulties with using arms-length
questionnaire measures to capture these con-
structs continues to be a challenge for research
in this area (Parthasarthy & Sethi, 1993).
The semi-structured interview method en-
abled effective follow-up questioning where
necessary, thus giving a clearer understanding
of the variables and their organizational context.
These data were particularly useful in the
development of measures of manufacturing
flexibility and use of integrative liaison devices.
We taped all interviews and transcribed them
verbatim. Each of us then independently coded
the responses. The use of the interview tran-
scripts in this coding procedure is described in
the next section.
Measurement of variables
Codings on each variable were developed by
the researchers independently using the entire
interview transcripts and other information
collected during the site visits. Full utilization
of the data was particularly important in the
measurement of the flexibility and integrative
liaison devices constructs. These are both
nebulous constructs comprising a number of
dimensions (Parthasarthy & Sethi, 1993). While
specific questions from the interview protocol
were designed to focus on defined aspects of
these constructs, it was necessary to use the
complete data set to develop ratings for each
firm. Thus, for these two variables the scaled
responses to individual questions do not build
directly into measures of variables. The way in
which the data about each firm were used to
code the responses is reflected in the schemas’
developed for both constructs. We also provide
profiles for seven firms in Appendix B to
illustrate the nature of the data collected and to
demonstrate the multidimensional nature of
manufacturing flexibility as well as the different
structural arrangements adopted by firms to co-
ordinate manufacturing activities.
Manufacturing jZex9ility. Manufacturing
flexibility is reflected in a firm’s ability to
respond to market demands by switching from
one product to another through co-ordinated
policies and actions (Nemetz & Fry, 1988) and
a willingness or capacity to offer product
variations (Buffa, 1980; Bowen et al., 1989).
Respondents were asked questions about flex-
ibility, and the elaborated responses and site
inspections* allowed the researchers to rate
independently each firm on a five-point scale
‘We wish to acknowledge the suggestion, made by an anonymous referee, of elaborating and illustrating the schema used
in the coding process.
* Firms were either independent or autonomous divisions with profit centre responsibilities.
246 M. A. ABERNETHY and A. M. LILLIS
ranging from 1 (little or no flexibility) to 5 (a
great deal offlexibility). The following two items
from the interview protocol formed the basis of
the initial questions posed to each respondent:
(a) proportion of turnover from non-standard
product lines;
(b) extent to which the manufacturing pro-
cess provides flexibility to offer customers
product variations.
The data collected in the field resulted in the
identification of three key dimensions of the
flexibility construct. These dimensions were
subsequently used to develop a schema for
rating firms on flexibility:
(a) technological dilKculty in making product
changes;
(b) strategic commitment;
(c) turn-around time to meet customer
demands.
The following discussion and the firm profiles
(Appendix B) provide some insight into the way
the schema was developed and used. Firstly, the
simplicity of product changes emerged as a
critical first-order discriminant. For example,
the firms described in profiles 1 and 2 (Appendix
B) produce a large range of what the General
Managers referred to as non-standard products.
However, the interviews and the Iirms’ product
brochures indicated that these non-standard
products required only simple manufacturing
changes from standard products. For example,
the manufacturer of medical bandages (profile
1) was able to vary the adhesive placed on the
bandages to suit climatic conditions. This
adjustment to the manufacturing process did not
constitute flexibility. These lirms were given a
“ 1” for flexibility on the basis that flexibility was
largely Irrelevant in accommodating the customer
demands placed on them.
Secondly, firms were distinguishable in their
relative strategic commitment to standardiza-
tion and flexibility. It was quite straightforward
to rate a firm which was a mass producer of
standard product lines. The proportion of
turnover from standard products was usually
oCLlOO% and the manufacturing process did
not facilitate product variation. These firms were
given a “ 1” for flexibility.
It was also relatively easy to rate those firms
we are really committed to a mainstream group of
standard products basically, the whole trick I suppose
in terms of trying to get your costs down is in fact
to standardize as much as you can there is no
flexibility in essence in the system formally.
there is a standard range of colours and that’s it and
if you want something that’s outside that range, well
that’s too bad
which had achieved a great deal of flexibility
(see profile 7). Such firms rated highly on all
dimensions of flexibility.
every single item [is]. custom designed and custom
built.
every manufacturer has a dilferent size cabinet
there is no standardization
most products [could be made] in a flexible way
we can punch it differently fit different electrical
components it’s very little hassle at all
Between these extremes, firms were rated on
their actual flexibility reflected in the cycle time
required to meet customer-initiated product
variations. Generally, reductions in cycle time
were achieved through adoption of non-dedicated
tooling and other flexible production tech-
nology. For example, contrast the following
quotations:
nowhere near as flexible as we would like. We always
liken it to a battleship where it takes miles to turn it
around.
the place goes into a bit of turmoil when we have to
make a non-standard run.
all the work we are doing in terms of investment is
aimed at increasing flexibility and reducing change-over
times, minimizing run sizes
we have the capability to make the product perform
better by customizing it and we are, as a business,
learning how to use that in a market sense
Thus, ratings from 2 to 5 reflect degrees of
product change complexity, strategic commit-
ment to flexibility and the implementation of
processes and technology which enable manu-
facturing to switch from one product to another
with relative ease. The schema described above
is illustrated in Fig. 1.
MANlJFACTLJRING FLEXIBILITY
Step 1
247
Step 3
Turnaround (cycle) time for customer
Fig. 1. Manufacturing flexibility measurement schema.
The schema was useful in ensuring that the
interview data were interpreted reliably. Once
each of the researchers had independently
analysed the interview data and determined the
rating for each firm, an inter-rater coefficient was
computed. The coefficient for the flexibility
rating indicated a high level of convergence
between the two raters [r = 0.96, p < 0.00 11. In
the three cases where the two researchers gave
very different ratings, the researchers subsequently
reviewed the transcripts, discussed the results
and developed a consensus rating for that firm.
I ntegrative liaison devices. Based on instru-
ments used previously by McCann & Galbraith
( 1981) Mintzberg ( 1983) and Van de Ven et
aZ. (1976) respondents were asked about their
reliance on each of the following integrative
liaison devices:
-
spontaneous contact between department
managers;
-
regular meetings between departments;
-
interdepartmental task forces;
- permanent teams;
- permanent liaison positions.
Respondents were also asked to assess the
extent to which the work or the work process
within departments was governed by the use of
standard operating procedures and whether the
organizational structure was functional, product
line or matrix.
The interview data were used to rank each
firm on a five-point scale reflecting the extent
of reliance on integrative liaison devices. The
data indicated that there were only three forms
of integrative liaison devices in general use in
the firms studied: spontaneous contacts, regular
meetings, and teams and task forces.
Teams and task forces were combined, as
most firms using these forms of integrative
devices did not distinguish between them. In
almost all cases the only interfunctional liaison
role was vested in the firm’s general manager.
There were no cases of matrix organizations and
only three product-focused firms. The relative
248
Step 1
M. A. ABERNETHY and A. M. LILLIS
Step 2
Step 3
Fig. 2. Integrative liaison devices (ILD) measurement
schema.
absence of higher-level liaison devices appears
to be a function of the size of firms under study.
The small size of these ftrms is, in fact, a general
feature of Australian manufacturing industry.5
A schema for rating the use of integrative
liaison devices was developed as follows. We
considered a product-focused structure as a
formal first-order response to developing func-
tional integration. As only three firms had a
product-focused structure, it was then necessary
to assess alternative approaches used by Iirms
to achieve functional integration. Within func-
tional structures, the interviews established that
“flatness” in the managerial authority structure
was critical to effective integration. This “flat-
ness” was mentioned so frequently in our
interviews that it became the essential discri-
minant in our classifying schema for firms with
functional structures. Profiles 4, 6 and 7
demonstrate the notion of a flat management
structure. If a product-focused structure is
considered a formal first-order response to
developing functional integration, then flatness
is a second-order, less formal approach to
integration within a functional structure.
Flatness is designed to encourage integrative
problem solving, fast response (by reducing
bureaucratic barriers) and spontaneity. There
appeared to be a widespread belief among
respondents in the effectiveness of “flatness” in
achieving integration ~z’s&vi~ other formal
mechanisms such as meetings and teams. This
led us to believe that firms adopting such a flat
structure would be relatively more functionally
integrated than those making extensive use of
meetings and teams without such flatness. Thus,
we gave a high loading to flatness in coding the
use of integrative liaison devices. The steps
’ The majority (97% ) of Australian manufacturing firms are less than 199 employees in size. Based on the latest published
statistics there were only 884 firms in Australia with more than 200 employees. Only 60 firms (0.002% ) have more than
1000 employees (Australian Bureau of Statistics, 1990).
MANUFACTURING FLEXIBILI’IY 249
involved in coding each firm on the use of
integrative liaison devices are represented via
the schema in Fig. 2.
The schema and the profiles presented
in Appendix B demonstrate how the data
were used to develop a rating for each firm.
A “5” represented the adoption of a product-
focused structure to facilitate interfunctional co-
ordination. A “4” was given where a firm had a
functional structure but had deliberately created
a flat structure, relying extensively on sponta-
neous integrative liaison devices to manage co-
ordination requirements (see profile 4). Within
hierarchical functional structures, co-ordinative
requirements could be classified as minimal
through to substantial. Where interfunctional
contact was relatively infrequent (e.g. monthly
meetings) the rating was “1” (see profile 1).
Ratings of “2” and “3” reflected the differing
degrees of reliance on teams and meetings
required to co-ordinate activities between
functional units. The ratings between each
researcher were again compared, and the inter-
rater reliability coefficient of 0.94 0, < 0.001)
obtained for this variable indicated a high level
of convergence, with four cases subsequently
requiring review consensus.
Manufacturingperformance measures A list
of performance measures was developed for this
study, based on Kaplan ( 1983) Howell & Saucy
(1987) and Chase (1990). The list included
items related to cost efficiency, flexibility and,
for completeness, quality and dependability
measures. During the interviews, respondents
were asked to indicate the use and relative
importance of each measure. The questionnaire
items emerged as non-discriminatory. Most
firms reported extensive use of virtually all
measures. Respondents were then asked to
identify and rank their three most critical
manufacturing performance measures in order
of importance. Respondents had no difficulty in
identifying the three most critical measures. In
the analysis, we used only these rankings to rate
the relative importance of efficiency-based
measures in the manufacturing performance
measurement system. There were no Iirms
which did not include an efficiency-based
measure in the three most critical measures.
There were, however, differences in the ranking
attached to efficiency-based measures, thus
enabling lirms to be coded on a scale of 1 (least
important) to 3 (most important).
Pwformunce. Following Khandwalla (1972)
Brownell & Merchant (1990) and others,
general managers were asked to rate the
performance of the firm relative to that of
competitors on a five-point fully anchored
Likert-type scale.
RESULTS AND DISCUSSION
Flexibility and control q&em design
Since this study is essentially exploratory, we
used simple inferential statistics to examine the
relationship between manufacturing flexibility,
efficiency-based performance measures and use
of integrative liaison devices. In the first
instance, we examined the bivariate correlations
between the variables. As can be seen from the
correlation matrix provided in Table 2, the
relationship between flexibility and use of
integrative liaison devices is in the direction
predicted. That is, the pursuit of flexibility
increases the extent to which lirms use integra-
tive liaison devices to manage functional inter-
dependencies and to break down functional
barriers. The correlation between flexibility and
use of efficiency-based performance measures is
also significant and in the predicted direction.
The use of efficiency-based performance measures
in manufacturing declines as a firm’s commit-
ments to flexibility increases. We also split the
sample into those Iirms which pursued flexibility
and those which did not, and performed a
parallel analysis which focused on tests of
difference between the means. The results are
presented in Table 3 and are in the predicted
direction. That is, the non-flexible group placed
more importance on efficiency-based measures
and used integrative liaison devices to a lesser
extent than those firms in the flexible manufac-
turing group. The t-values indicate that the mean
scores are significantly different from each other
for both control elements. These results add
250 M. A. ABERNETHY and A. M. LLLLIS
TABLE.2. Pearson correlation (significance)* matrix (n = 42)
Variables 1 2 3
1. Flexibility
2. Integrative liaison devices
3. Etficiency-based PMS
4. Performance
* One-tailed test.
0.46 (0.001)
-0.24 (0.06) -0.10 (0.27)
0.13 (0.12) 0.28 (0.04) 0.08 (0.31)
TABLE 3. Means and t-test for use of efficiency-based performance measures and
integrative liaison devices for lirms with low flexibility and high flexibility
Group 1: Group 2:
Non-flexible ftrms Flexible firms t-value (sip)
(n = 20) (n = 22)
Mean (S.D.) Mean (S.D.)
Efficiency-based performance
measurement system
Use of integrative liaison
devices
2.60 (0.50) 2.27 (0.76) 1.62 (0.05)
2.30 (1.21) 3.45 (1.01) 3.35 (0.001)
l One-tailed test.
further support to the idea that firms committed
to flexibility use integrative liaison devices to a
signiIicantly greater extent and rely on efficiency-
based performance measures to a significantly
lesser extent.
These findings are consistent with our expecta-
tions and add weight to the integrative fiame-
works of control which argue that organizations
adapt their control system to meet the control
requirements of the situation (Flamholtz, 1983;
Govindarajan, 1988; Merchant, 1985). It would
appear, Tom the evidence provided here, that the
control needs imposed by flexibility result in firms
placing different emphasis on the elements which
form part of the control system package.
Pet$wrnunce, munuf&cturi ng~i bi l i ~ and
control system &si gn
The third research question of interest in this
paper is whether firms which adapt their control
systems to support their commitment to flexibility
perform better than those firms which do not.
To answer this question, we again used simple
inferential statistics.” The sample was split into
the same two groups as the previous analysis.
One group consisted of firms with a degree of
flexibility and the second group comprised
those Iirms which were not flexible. We
computed correlations between the two control
elements and performance for each group. If
adapting control systems to meet strategic
‘We also assessed whether the match between flexibility and each control element influenced tirm performance by fitting
two regression equations, each employing a multiplicative interaction term. The equations were of the same form as used
by Brownell (1982) and others (Govindarajan, 1988; Hirst, 1983). We examined the coefficients of the interaction
terms to assess if the appropriate “match” resulted in better performance. We expected performance to be atkcted
ditferently accordingly to the extent to which manufacturing flexibility is pursued. The results are presented in Appendix
C and are consistent with those presented in the paper. The reliance on efficiency-based measures has a progressively
more negative effect on performance as manufacturing flexibility increases in importance (as = -0.50, t = 2.13, p =
0.04). However, although we expected increased reliance on integrative liaison devices to have a progressively more
positive effect on performance as flexibility increases (i.e. a3 > 0), the coefficient was negative but not significant
MANUFACTURING FLEXIBILITY
TABLE 4. Test of the difference in the correlations between firm performance and use
of controls
251
Correlation (sig) between
efficiency-based performance
measures and firm performance
Group 1: Group 2: Difference in
Non-flexible firms Flexible firms correlations
(n = 20) (n = 22) t-value (sig)
0.44 (0.03) -0.08 (0.35) 1.60 (0.05)
Correlation (sig) between use of
integrative liaison devices and firm
performance
0.27 (0.12) 0.38 (0.04) 0.31 (ns.)
l One-tailed test.
commitments infhtences performance, the cor-
relation between the use of efficiency-based
measures and performance should be signilicantly
higher for ftrms which do not pursue flexibility
than for those which do. In contrast, we expect
the correlation between performance and use
of integrative liaison devices to be significantly
higher for flexible Iirms. The separate correla-
tions between performance and each control
element for the two groups is presented in
Table 4. These correlation coefficients indicate
whether the relationship between performance
and the control element is significantly different
from zero. What is of interest here, however, is
whether the coefficients for each group are
significantly ditferent from each other, as opposed
to horn zero. To test for differences between the
correlations we used the Fisher transformation
test (Winkler & Hayes, 1975, p. 653). The t-
values and significant levels which resulted from
this test are presented in Table 4 and provide
some support for our expectations. Not only
were the correlations between performance and
use of efficiency-based measures in the pre-
dicted direction for the two groups (the
correlation was positive for the non-flexible
firms and negative for the flexible firms), the
correlations also differed from each other at less
than the 0.05 level of significance (one-tailed
test). However, while the correlation between
use of integrative liaison devices and perform-
ance appeared to be larger for the flexible
manufacturing firms, the difference between the
coefficients for the two groups was not significant.
Although the difference in the correlation
coefficients between performance and the use
of integrative liaison devices was not significant
for the two groups, the relationship between
performance and use of integrative liaison
devices was positive and significant for firms
committed to flexibility. This supports the data
collected in the field. The expanded responses
by General Managers indicated that organizational
arrangements, and in particular, the develop-
ment of integrative liaison devices were, in fact,
the key facet of the organization control system
in the effective management of manufacturing
flexibility. It was considered more important to
get the structural arrangements right than the
design of the performance measurement system
for fums pursuing flexibility. Informal and
organic management structures, development
of lateral linkages, and the development of an
organization culture which encouraged indivi-
duals to identify with corporate goals, appeared
to carry strong control implications for effective
strategy implementation. Changes to the per-
formance measurement system appeared to be
less of an imperative to guarantee achievement
of strategic goals relating to flexibility.
Much of the contingency-type empirical
research in accounting has stressed the import-
ance of empirically examining the impact of an
appropriate match between control system
design and contextual factors. However, the data
collected in this study highlight the difficulties
identified by researchers in obtaining significant
results for “fit” hypotheses (Hopwood, 1976;
Merchant, 1981; Kim, 1988) and question the
practical significance of any such findings. It is
252 M. A. ABERNETFN and A. M. LILLIS
interesting to note the range of reported
influences on performance. The following are
some of the responses given. None of these are
related to the “fit” or “misfit” between control
system design and manufacturing strategy.
Responses are from general managers of low
performing firms as they naturally tended to
elaborate their responses on the performance
question.
performance, the qualitative data collected in
the field suggest the need for caution in
interpretation.
SOME CONCLUDING COMMENTS
- General Manager had resigned, set up with
a competitor, and taken much of the business.
- Changes in import regulations which
allowed competitors into the market. Another
firm had lost considerable market share due to
entry of an Australian competitor who was
constantly undercutting them on price.
This study sought to establish a link between
manufacturing strategy and two dimensions of
an organization’s management control system:
structural arrangements and the performance
measurement system. In particular, the study
focused on examining differences that flow from
the pursuit of flexibility as a manufacturing
strategy,
- One Erm had an appropriate “fit” between
strategy and control system design but was
pursuing a strategy which did not give them a
competitive advantage.
- Out-dated manufacturing technology and
absence of investment capital due to current
economic climate.
- Currently undergoing a major restructuring
of the firm.
- Economic downturn.
Attempts to measure the impact of “fit” on
performance also ignore the dynamic nature of
organizations. Several firms had only recently
changed, or were in the process of changing,
either their strategy or control system design.
These changes, however, were not reflected in
the firm’s performance in the current period.
One firm, for example, indicated that its
performance had significantly improved in the
past year, although overall performance was still
considered relatively low “. . . we were a
minus five last year, this year we are a plus
two .”
Our results clearly support the notion that
organizations adapt their structural arrange-
ments in order to implement manufacturing
flexibility. Increasing reliance on integrative
liaison devices in the form of teams, task forces,
meetings and spontaneous contacts reflects the
interconnectedness of functions that has been
theoretically argued to flow from customer-
responsive manufacturing strategies (Nemetz &
Fry, 1988; Bowen et al , 1989; Parthasarthy &
Sethi, 1993). The results also provide support
for the negative relationship between flexibility
and the role of efficiency-based performance
measures.
The study has potential limitations, particularly
in the area of measurement. The use of interviews
for data collection can be criticized on the basis
that managers’ responses, or the interpretation
by the researchers of the responses, are likely
to be biased. Following Merchant & Manzoni
( 1989) we attempted to overcome this limita-
tion by examining the interview transcripts in
depth for any inconsistencies in the interviewee’s
logic. Transcripts were also coded by each
researcher independently and compared for
inconsistencies or contradictions.’
While our results do provide some empirical
evidence that the appropriate match between
control system design and flexibility enhances
Offsetting these limitations is the potential
provided by a field study methodology to
make significant contributions to the literature.
‘Ideally we would have preferred to use a third independent researcher to code the transcripts. Limited funds and
difficulties in finding an adequately trained researcher for the task precluded this course of action.
MANUFACTURING FLEXIBILITY 253
This study contributes to the literature by
drawing on a rich body of data from 42 firms,
to provide an empirical perspective on the
management of new competitive priorities in
manufacturing through the development of
integrative co-ordinating mechanisms and per-
formance measurement systems. The study also
makes a methodological contribution to the
literature. Prior research has identified the
problems with measuring nebulous constructs
such as flexibility, or use of informal co-
ordination mechanisms such as integrative
liaison devices (Parthasarthy & Sethi, 1993).
This study used the knowledge gamed in the
field to build two measurement schemas for
both these constructs. The schemas provide
future researchers with a basis for coding field
data and a foundation on which to develop
survey instruments for measuring manufactur-
ing flexibility and use of integrative liaison
devices.
BIBLIOGRAPHY
Australian Bureau of Statistics (Canberra: Commonwealth Government Printer ACT, 1990).
Bowen, D. E., Siehl, C. & Schneider, B., A Framework for Analysing Customer Service Orientations in
Manufacturing. Acudemy of Management Review (1989) pp. 75-95.
Brownell, P., The Role of Accounting Data in Performance Evaluation, Budgetary Participation and
Organizational Effectiveness,journuI ofAccounting Research ( 1982) pp. 12-27.
Brownell, P. & Merchant, K A., The Budgetary and Performance Influences of Product Standardization
and Manufacturing Process AutomationJournal of Accounting Research (1990) pp. 388-397.
Bruns, W. J. & Waterhouse, J. H., Budgetary Control and Organization Structure,Journul of Accounting
Research (1975) pp. 177-203.
Buffa, E. S., Research in Operations Management,JournuZ of Operations Management (1980) pp. l-6.
Buffa, E. S., Meeting the Competitive Challenge (New York: Irwin, 1984).
Burns, T. and Stalker, G. M., The Management of Innovation (London: Tavistock, 1961).
Chase, R. B., Dimensioning the Service Factory, in Etdie, J. E., Burstein, M. & Flegenbaum, A. (eds),
Research Agenda fur the Next Decade (Boston: Kluwer, 1990).
Chrisman, J. J., Hofer, C. W. & Boulton, W. R., Toward a System for Classifying Business Strategies, Academy
of Management Review ( 1988) pp. 413-428.
Dean, J. W. & Susman, G. I., Strategic Responses to Global Competition, in Snow, C. (ed.), Strategy,
Organization Design and Human Resource Management (Greenwich, CT: JAI Press, 1989).
Drucker, P. F., The Emerging Theory of Manufacturing, Harvard Business Review (May-June 1990) pp.
94-102.
Eccles, R. Cr., The Performance Measurement Manifesto, Harvard Business Review (January-February
1991) pp. 131-137.
Flamholtz, E. G., Accounting, Budgeting and Control Systems in their Organizational Context, Accounting,
Organizations and Society (1983) pp. 153-169.
Galbraith, J. R., Designing Complex Organizations (Reading, MA: Addison-Wesley, 1973).
Govindarajan, V., A Contingency Approach to Strategy Implementation at the Business-Unit Level:
Integrating Administrative Mechanisms with Strategy, Academy of ManagementJ ournal ( 1988) pp.
828-853.
Govindarajan, V. J. & Gupta, A. K., Linking Control Systems to Business Unit Strategy: Impact on
Performance, Accounring Organizations and Society (1985) pp. 51-66.
Hayes, D. C., The Contingency Theory of Managerial Accounting, Accounting Review (1977) pp. 22-39.
Hirst, M. K., Reliance on Accounting Performance Measures, Task Uncertainty and Dysfunctional
Behaviour: Some ExtensionsJournal of Accounting Research (1983) pp. 596-605.
Hopwood. A. G., Accounting and Human Behaviour (Englewood Cliffs, NJ: Prentice-Hall, 1976).
Howell, R. A. & Soucy, S. R., Operating Controls in the New Manufacturing Environment, Management
Accountant (1987) pp. 25-31.
Kaplan, R. S., Measuring Manufacturing Performance: A New Challenge for Management Accounting
Research, Accounting Review (1983) pp. 686705.
254 M. A. ABERNETHY and A. M. LILLIS
Kaplan, R. S., Measures fw Manufacturi ng Excel l ence (Boston: Harvard Business School Press, 1990).
KhandwaIIa, P. N., The Effect of Different Types of Competition on the Use of Management Controls,
journal of Accounti ng Research (1972) pp. 275-285.
Kim, K. K., Organizational Co-ordination and Performance in Hospital Accounting Information Systems,
Accounti ng Revi ew (1988) pp. 22-39.
Iarsson, R. & Bowen, D. E., Organization and Customer: Managing Design and Coordination of Services,
Academy of Management Revi ew (1989) pp. 213-233.
Macintosh, N. B., The Soci al Software of Accounti ng and I nformation Systems (New York: Wiley, 1985).
McCann, J. & GaIbraith, J. R., Interdepartmental Relations, in Nystrom, P. & Starbuck, N. (cds), Handbook
of Organi zati onal Desi gn, Vol. 2, pp. 60-84 (New York: Oxford University Press, 1981).
Merchant, K. A., The Design of the Corporate Budgeting System: Inihtences on Managerial Behaviour and
Performance, Accounti ng Revi ew (1981) pp. 813-829.
Merchant, K A., Organizational Control and Discretionary Program Decision Making: A Field Study,
Accounti ng, Organi zati ons and Soci ety (1985) pp. 67-85.
Merchant, K A. & Manzoni, J., The Achievability of Budget Targets in Profit Centres: A Field Study,
Accounti ng Revi ew (1989) pp. 539-559.
Miller, D., Relating Porter’s Business Strategies to Environment and Structure: Analysis and Performance
Implications, Academy of Management J ournal (1988) pp. 280-308.
MiIIer, D., Droge, C. & Toulouse, J., Strategic Process and Content as Mediators between Organizational
Context and Structure. Academy of Management J ournal ( 1988) pp. 544-569.
Mintzberg, H., Structuri ng i n Fi ves (EngIcwood CliI%, NJ: Prentice-Hall, 1983).
Nemetz, P. L. & Fry, L. W., Flexible Manufacturing Organizations: Implications for Strategy Formulation
and Organization Design, Academy of Management Revi ew (1988) pp. 627-638.
Parthasarthy, R. & Scthi, S. P., The Impact of Flexible Automation on Business Strategy and Organizational
Structure, Academy of Management Revi ew (1992) pp. l-26.
Parthasarthy, R & Scthi, S. P., Relating Strategy and Structure to Flexible Automation: A Test of Fit and
Performance Implications, Strategi c Management J ournal ( 1993) pp. 529-549.
Simons, R., Accounting Control Systems and Business Strategy: An Empirical Analysis, Accounti ng
Organi zati ons and Soci ety (1987) pp. 357-374.
Simons, R., The Role of Management Control Systems in Creating Competitive Advantage: New Perspective,
Accounti ng, Organi zati ons and Soci ety (1990) pp. 127-143.
Thompson, J. D., Organi zati ons i n Acti on (New York: McGraw-Hill, 1967).
Van de Ven, A. H., Delbecq, A. L. & Koenig, R. Jr, Determinants of Co-ordination Modes within
Organizations, Ameri can Soci ol ogi cal Revi ew (1976) pp. 322-338.
WinkIer, R. L. & Hayes, W. L., Stati sti cs: Probabi l i ty, I nference and Deci si on, 2nd Edn (New York: Holt,
Rinehart & Winston, 1975).
APPENDIX A: STRUCTURED INTERVIEW QUESTIONNAIRE
Note: AI1 items used a fully anchored five-point scale ranging from one to five
1. What proportion of your turnover comes from non-standard product lines?
2. Does your manufacturing process provide the flexibility to offer customers product variations?
Sh-uctural arrangements (i ntegrati ve I i ai son devi ces)
1. Would you describe your organization as having a functional structure, i.e. one where responsibilities
are divided primarily by function, e.g. marketing function, manufacturing function?
2. Would you describe your organization as having a product-focused structure, i.e. where aI1 functional
departments working on one product are grouped together?
3. To what extent is there a need to co-ordinate the work of the manufacturing department with other
departments, e.g. marketing, engineering?
4. How often are regular meetings held between departments?
5. Does your organization use inter-departmental task forces (usuaIIy having a limited and relatively
short life), for problem solving, system design, department of product lines, manufacturing and co-
ordinating product lines?
MANUFACTURING FLEXIBILITY 255
6. Are teams (more permanent and having a longer time orientation) used for these purposes?
7. Are there any permanent iiaison positions within this organization, whose roles and responsibilities relate to co-
ordinating manufacturing activities with other departments?
8. Is the co-ordination between departments (e.g. marketing, engineering, manufacturing) primarily by spontaneous
contact between managers of the various departments?
9. To what extent is the work within departments governed by the use of standardized operating procedures or work
processes?
Performance measu- t system
Respondents were asked to rate the extent to which each of the following measures were used for perlbrmance evaluation
in the manuktcturlng department of their organization.
1. On time delivery performance records.
2. Number of customer complaints.
3. Iabour utilization/efficiency statistics.
4. Incidences of product defects.
5. Number of warranty claims.
6. Inventory turnover ratio.
7. Reports on whether standard product costs are met.
8. Survey of customer satisfaction.
9. Measurement of cost reduction due to quality improvements.
10. Evaluation of whether people in manufacturing are co-operative and responsive to the demands of other departments
within the organization.
11. Customer satisfaction with problem solving and sales assistance.
12. Rate of introduction of new products.
13. Evaluation of the ability to vary product characteristics.
14. Length of cycle time from order to delivery (for standard products).
15. Length of cycle time for customer-requested product variations.
16. Material purchase price variance.
17. Rate of material scrap loss.
18. Measurement of machine utilization and down time.
256 M. A. ABERNETHY and A. M. LILLIS
APPENDIX B: MEASUREMENT OF FLEXIBILITY AND INTEGRATIVE LIAISON DEVICES
PROFILE 1
‘ROFILE 2
‘ROFILE 3
‘AOFILE 4
-
FLEXIBILITY
A manufactL/rer of medical supplies with 420
emloyees. Ninety per cent of turnover comes
from the production of standard lines. The
non-standard products only differ in
specifications as to the adhesive or weight of
cloth used in production.
Aluminium can producer with 210 employees.
Mass producer of standard lines. Products
onty vary in logos. P/ant runs 24 hours per
day. High off-the-shelf product avaitabitity is
strategba/& critical and the manufacturing
process does not e&/y facilitate changes in
product demands. “We’ve got 200 different
/abets and it takes half-an-hour to an hour to
change a machine from one label to the other
. ..so we carry buffer stocks.. For short-run
jobs we’// carry, in some instances, 72 months
stock for customers, because 8s economical
for us to do if that way”.
A manufacturer of nails. nuts, bat& and
screws with 850 employees. High off-the-
shelf availabilify is more important than any
other strategk priority. The Product mix is 60
per cent standard and 40 per cent
nonstandard. However, “in a lot of cases they
(the non-standard products) are very similar to
a standard product, they want a longer thread
or whatever...in some cases they are different
products”. The manufacturing process is
inflexible and usua//y invokes long production
runs. ‘We are nowhere near as flexible as we
would like it to be we always liken it to a
battleship where it takes miles to turn it around
.
Aluminium can manufacturer employing 145.
The firm is reelatively flexible. It produces
highly decorated cans with special fittings,
shapes and sizes. :..k tends to be more
flexible [than the division described in Profile
t], it tends to be slower as well and the capital
investment is lower.’ Current investment is
aimed at increasing f/exbi/ity and reducing
changeover times. However, decorating “set
up times for a smaN run of cans costs more
than the actual running casts through the tine.
As a result stock ho/dings are still relatively
high (2.8 months of cover). We err on the
side of holding stock” to ensure that customer
demands are met on time.
Rating
1
1
INTEGRATIVE LIAISON DEVICES
The firm has a high& functional and
bureaucratti structure. The relationship
between sa/es and production is sequential
and there was evidence of functional barriers
between the departments (ie separate lunch
rooms). Marketing determines the sales
forecasts and production schedules are
based on that forecast. AM of the interface
between functional departments is through
formal monthly meetings. Task forces are
used maybe twice a year to solve a specific
orobtem. A task force recentlv assessed the
causes of increased material usage.
There is a strong need for co-ordination
between sales and production because of the
high customer service (delivery) commitment.
The need to respond to daity order changes
(depending, for exan@e, on the weather and
associated demand for SOI? drinks)
discourages reliance on forma/ co-ordinative
mechanisms and standard operating
procedures. Spontaneous verbal
communications are heavily retied upon.
Special purpose %orrective action teams”ere
used for problem-solving in a// areas (not
necessarity production), including, for
exan@e, at the time of this study, workers
The firm has recentfy reorganized into what
can best be described as a product-line
structure for the manufacturing segment of
the business. Sales and distribution remain a
separate functional department responsible to
the GM. There are three separate product
lines. The relationship between sales and
manufacturing remains sequential. They
have a materials management group which
acts as the linchpin between sales and
manufactur/ng. Sales and manufactuting
meet regularly two to three times a week for
non-standard products and once monthly for
standard products.
The management structure is functional but
%ompressed”and in the process of changing
from a more formal meeting-based structure
to an informal structure, “We have meetings
and meetings and that is part of the probtem.
We’re changing a// the time, setting up
communication vehicles . ..sncouraging
people to pick up the phone, go and talk to
someone to get it done. * Teams are used,
with lessening invotvement of managers and
accountants.
-
Ming
MANLJFACTURING FLEXIBILITY
PROFILE 5 Punp rnanufact&m?r enqioyk&7 280. Seventy- 4
~o_~pj;“at;on wi?hin the functional structure is 4
five per cent oi turnover is custom-buik
assisted by product managers within
puw,motor and base p/ate combinations sold
marketing who manage price information,
on order. The remaining 25 per cent of
technical data, stock avai/abi/ity, sale.5
turnover is sfandard base punp sates supplied
forecasts and projects and so on. These
from stock The manufacturing process is
product managers determine product
very ffexibk in the contexl of identified technical specifications so that they are
markets: V’s not very fJexib/e. /t’s flexible to
‘market-dtien rather than manufacturing-
a degree to cover the markets that we clearly driven”. Standard operating procedures are
identify . ..but in term of being able to make heavily relied upon to manage engineered
things other than puws andpuw sets-no, product changes (resu/ting from order
we are not cut out to do that. We can make a changes). Co-ordhative devices are used
wide variety of concepts in any shape or form extensively and include forma/ meetings,
within genera/ limitations that punlp
teams and spontaneous contacts for problem
manufacturers can envisage For the market solving
that we sewice, I think rhal our organization
has more flexibility than mosl of our
competitors...”
PROFILE 6 A /age cab/e manufacturer emfoying 1,400 4 The structure is messy and a result of a 4
en@oyees. It has two production plants. One number of mergers and takeovers ..“ii’s never
p/ant produces standard pmducts...“it’s our been structured ..A just happened.’ Co-
sausage factory, ft just makes virfua/ty ordination between funct!ona/ units is
continuous products’! Seventy per cent of achieved by... “having regular meetings,
turnover comas from a second plant which is irrpromptu meetings, grabbing three or four
“more like a jobbing shop .., more of the extra guys and bringing them together to discuss a
high voltage and the specials, it is certainly not problem. Task forces are used to a “very
run as a process factory”. They make great extent” for problem solving and product
everyihing to order . ..“you can’t have that stuff development. Senior management has made
., sitting around, we// you could, but would go a deliberate effort to ensure that the
broke _.. we need about 6.3 tonnes of copper a organization is f/at and informal. . ..V’d rather
day and we are paying $3750 a lonne . ..you not call a meeting ._. I mean, rea//y once a
don’t have too much stock iying around”. month to discuss the resu/ts, maybe .we do
Flexibility is necessary for survival Yapart not write memos to each other. As soon as I
from the contract stuff, everything is virtual/y a see a memo . ..someooe is going for political
speciai and that’s why _..you can’l afford not to points __, that is when we real& kick ass. I get
be flexible... .you’re dead if you’re not flexible them up here and say you, you and you. sort
it out and that’s it”. Liaison between sales
and manufacturing is dons through informal
m&-action .._ “if I find out we’ve got a problem
and these blokes haven’l even been tatking
to each other, a// he// breaks loose. They
interact . ..they get the job done and no cheap
points ,just get it done or we wi// get
someone who can . ..a// departments interact
or we fix them”. The GM summanied the
structure as . ..“/‘ve got to say, this is a veiy, I
think. a fairl), informal management structure.
it’s mn a lilt/e bit like a big milk bar”,
PROFILE 7 The second largest /~ght finings manufacturer 4 The structure of the firm is functional but
4
in Austra/ia em/eying 290. /t manufacturers a management has ensured thal the structure
range of standard light fillings but a lol of the is f/at . ..“the conpany is a fair/y f/at structure, I
work is special right fittings. They have mean we haven‘t got a who/e heap of
adopted flexibility as a strategy priority :..a tot management /eve/s and things like that, so jf
of the machinery we have deliberalsly we are looking at new products you get
selected because it’s f/sxible machinery. We someone from the factory, someone from the
have avoided wherever possible going to design office, someone from the sales office:
dedicated tooling, hard tooling, that isoY Production and sales ‘get together and work
flexible”. They have “very flexible machines out what they want. They do that fist thing
even with standard products...’ so that even in the morning over coffee”. Two pennanenl
standard products can be varied to meet teams are in p/ace. One involves
customer demands. “Most of our product is manufacturing and engineering who are
made in a very flexible way we can in fact continually %oking at our manufacturing
punch it dinerent/y .., fit different electrical strategy”, the other team includes
conyDonents . . . is very /in/e hassle at all.’ Even
the production schedule is fkuible !..as f said
engineering, finance and manufacturing and
looks at prioriiies in capital expenditure.
it b a very informal scheduling system,
flexibifity is the woti, I mean we mint things
both ways, it is love& to have a schedule. but
we also want to be abte to walk out into the
factory andput a job on at a moment’s notice.
And that is the way the p/ace operates.*
257
258 M. A. ARERNETHY and A. M. ULLIS
APPENDIX C: RESULTS OF TWO REGRESSIONS
The regressions are of the form
Y = 6 + a,F + a&, + a&__,
where Y = firm performance, F = manufhcturing flexibility, C = control elementj, j = 1, 2 ( 1 = efficiency based PMS;
2 = integrative liaison devices), FC = interaction between manufacturing flexibility and control element j.
Control element
Regression I (j = 1) Regression II (j = 2)
Efficiencv-based PMS Integrative liaison devices
a, = Constant (SE.)
t-value (sig)
a, = Flexibility coefficient (S.E.)
t-value (sig)
a2
= Control element coefficient (S.E.)
t-value (sig)
as = Interaction coefficient (S.E.)
t-value (sig)
R2
Sip of F f one-tailed test)
1.77 (0.89)
1.99 (0.05)
0.45 (0.19)
2.34 (0.02)
0.5 1 (0.30)
1.69 (0.10)
-0.50 (0.23)
-2.13 (0.04)
0.13
1.96 (cO.05)
2.37 (0.55)
4.33 (0.00)
0.19 (0.20)
0.95 (0.35)
0.37 (0.19)
1.96 (0.06)
-0.23 (0.20)
-1.17 (0.25)
0.11
1.54 (n.s)
doc_921478334.pdf
Despite anecdotal evidence questioning the relevance of management accounting in firms pursuing “new”
mamtfacturing strategies, the exact nature and role of accounting information for controlling manut&xuring
activities has received little empirical attention
Pergamon
Accounting 0r#anization.r and Society, Vol. 20, No. 4. pp. 241-258, 1995
Copyright 0 1995 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0361-3682/95 S9.50+0.00
0361~3682(94)E0014-L
THE IMPACT OF MANUFACTURING FLEXIBILITY ON MANAGEMENT
CONTROL SYSTEM DESIGN*
MARGARET A. ABERNETHY
The University of Melbourne
and
ANNE M. LILLIS
Deakin University
Abstract
Despite anecdotal evidence questioning the relevance of management accounting in firms pursuing “new”
mamtfacturing strategies, the exact nature and role of accounting information for controlling manut&xuring
activities has received little empirical attention. Thii study examines the impact of manufacturing flexibility
on the design of management control systems. In particular, we examine the implications of flexibiity on
the use of efficiency-based performance measurement systems and the use of integrative liaison devices. We
collected data, through semi-structured interviews, from general managers in manufacturing firms in
Melbourne, Australia These interview data were used to develop measures for the constructs of interest.
Exploratory analysis of the data indicates that integrative liaison devices are a critical form of control in
managing the implementation of flexible manufacturing strategies. However, the role of accounting or other
efficiency-based performance measures declines in those firms with a commitment to manufacturing flexibility.
Confidence in traditional manufacturing manage-
ment accounting systems has eroded significantly
in the past five years. The relevance of account-
ing is most seriously questioned in lirms
implementing “new” manufacturing strategies
(Kaplan, 1990). While the accounting literature
provides anecdotal evidence that traditional
management practices are inappropriate in
these settings, there has been little systematic
empirical study of the link between manufactur-
ing strategy and control system design. This
paper examines the impact of strategic com-
mitments to manufacturing flexibility on two
critical aspects of a lirm’s management control
system (MCS): the performance measurement
system, and the structural arrangements required
to co-ordinate production activities.
The following section reviews the literature
which has examined the link between control
system design and strategic commitments,
concluding with a number of exploratory
propositions. Subsequent sections address
methodology, variable measurement, results,
concluding comments and directions for further
research.
*We wish to acknowledge contributions from participants at the ABO Conference (Bond University, July 1991) The
Management Accounting Research Conference (Atlanta, October 199 1) The Strategic Management Accounting Conference
(University of New South Wales, October 1992) and helpful comments from Peter Brownell, Rob Chenhall, Kenneth
Merchant, Frank Collins, Geoff Burrows and Christine Jubb. We are also grateful for the detailed advice given by Anthony
Hopwood and the anonymous referees. We wish to thank the Department of Accounting and Finance, Monash University,
Bowater Faculty of Business and the Monash Medical Centre for the tinancial support provided for this project.
241
242 M. A. ABERNE’ITN and A. M. LILLIS
CONTROL SYSTEM DESIGN AND
STRATEGIC COMMITMENT
Manufacturing flexibility is reflected in a firm’s
ability to respond to market demands by
switching from one product to another through
co-ordinated policies and actions (Nemetz &
Fry, 1988) and a willingness or capacity to offer
product variations (Buffa, 1980; Bowen et al.,
1989). It is a strategy which attempts to
maximize differentiation (Parthasarthy & Sethi,
1993). The manufacturing approach associated
with this strategy differs from that of a lirm
competing on the basis of cost leadership.
Typically, firms committed to cost leadership
have invested in manufacturing technology
which enables long production runs of standard-
ized products (Dean & Susman, 1989; Buffi,
1984; Chase, 1990). The characteristics of
traditional management accounting systems,
which have been designed to support mass
production technology, are potentially at odds
with a strategic commitment to manufacturing
flexibility (Kaplan, 1990). These systems empha-
size cost control through techniques such as
responsibility accounting, standard costing and
the promotion of production efficiency through
variance analysis. Management accounting
systems, with their focus on task segregation and
efficiency, are particularly well suited for mass
producers of standard products with unchang-
ing characteristics (Simons, 1990; Drucker,
1990; Bowen et al., 1989; Nemetz Bz Fry, 1988).
Manufacturing flexibility is a customer-
responsive strategy and is used as a “competitive
weapon” to create enduring competitive advan-
tages (Chrisman et al., 1988, p. 4 15). Respond-
ing to customer-driven demands requires control
systems which facilitate product customization
rather than product standardization. These
demands impose complex, multifunctional
problems which are not well managed by
systems focusing on efficiency and task segrega-
tion. To the extent that accounting performance
measurement systems focus on efficiency, they
are inconsistent with customization and flexibility
(Kaplan, 1990). Furthermore, the task segrega-
tion implicit in responsibility accounting and
variance analysis is antithetical to the cross-
functional co-ordination required to meet
customer-driven demands (Bowen et al, 1989).
The implications of manufacturing flexibility for
performance measurement system design and
the structural arrangements required to co-
ordinate multifunctional tasks are elaborated in
the following sections.
Manufacturing J exibility and performance
measurement systems
Few studies have examined empirically the
link between manufacturing flexibility and the
design of performance measurement systems in
manufacturing. Much of the research examining
the link between strategy and control systems
has done so at the strategic business unit (SBU)
level, rather than at the manufacturing level
(Govindarajan, 1988; Govindarajan & Gupta,
1985; Merchant, 1985; Simons, 1987, 1990).
This literature, however, suggests that the
role of accounting in measuring performance
depends on the strategic commitments of the
SBU. Formal cost control mechanisms, such as
standard costing systems, appear to be particu-
larly well suited for measuring performance in
manufacturing firms with high product standard-
ization, relatively stable production processes
and where the strategic priority is typically one
of cost minimization (Macintosh, 1985; Miller,
1988; Hirst, 1983; Brownell & Merchant, 1990;
Govindarajan, 1988).
Research at the SBU level argues that firms
competing on the basis of product differentia-
tion will move away from financial and other
efficiency-based measures of performance to
measures which support the achievement of
strategic priorities associated with differentia-
tion (Govindarajan, 1988; Simons, 1987). A
commitment to manufacturing flexibility is a
form of product ditferentiation and, as such,
creates “co-ordinative, technical and control
problems. . . ” distinct from those found in lirms
employing mass production technology (Miller,
1988, p. 281). Consequently, it is unlikely
that manufacturing activities can be effectively
controlled by assigning tasks, programming
desired behaviour and implementing efficiency-
MANUFACTURING FLEXIBILITY 243
based indicators to capture task performance
(Parthasarthy & Sethi, 1992). The absence of
standardization makes it increasingly difficult to
specify unambiguous performance standards, as
optimal relationships between inputs and out-
puts for production tasks are usually not known.
In this situation, financial measures of manu-
facturing performance become less relevant
(Brownell & Merchant, 1990; Hayes, 1977).
Successful implementation of manufacturing
flexibility requires cross-functional responsive-
ness to specific customer-initiated demands.
Effective performance is the result of the
combined efforts of those functional units
required to satisfy customer demands. This
suggests that effective performance measure-
ment will require a shift from measures which
focus on manufacturing efficiency to measures
which encourage interfunctional co-operation
and adaptation and which capture the critical
success factors related to customer-initiated
demands (Macintosh, 1985).
It is expected, therefore, that when firms
pursue flexibility there will be a shift “from
treating financial figures as the foundation for
performance measurement” to a broader set of
measures designed to support flexibility as a
strategic priority (Eccles, 199 1, p. 13 1). These
measures are likely to include qualitative and
quantitative indicators of customer responsive-
ness. For example, measures of cycle times on
product variations, delivery performance and
assessment of the ability of manufacturing to
vary product characteristics or develop new
products, are appropriate. Similarly, interfunc-
tional performance measures, such as those
which assess the strengths of interdepartment
relationships and levels of co-operation, will be
appropriate.
Manufacturing fl exi bi l i ty and structural
arrangements
A number of researchers have identified the
direct implications of manufacturing flexibility
on the structural arrangements required to
facilitate co-ordination of multifunctional tasks
(Bowen et al ., 1989; Nemetz & Fry, 1988;
Parthasarthy & Sethi, 1993). These structural
arrangements are in direct contrast to those
which are considered to be appropriate where
product standardization is high and production
processes relatively stable. These manufacturing
processes impose few co-ordination problems.
Indeed, it has often been argued that the most
efficient organizational form is one which
buffers manufacturing from outside influences
(Thompson, 1967) and which has attributes
similar to the mechanistic structure described
by Burns & Stalker ( 1961). A mechanistic
structure emphasizes high vertical and hori-
zontal task differentiation, centralized decision
making, vertical information flows and func-
tional groupings (Mint&erg, 1983; Miller, 1988).
This structure is considered efficient, as the
relationships between manufacturing and other
functional units are pooled or sequential and
the need for co-ordination between functional
units is minimal (Van de Ven et al ., 1976). Any
co-ordination between functional units can be
managed through formal standardized proce-
dures and processes (Parthasarthy & Sethi,
1993).
Structures that emphasize a high differentia-
tion of task and vertical information flows are
incongruent with a commitment to flexibility
(Parthasarthy & Sethi, 1993). A commitment to
manufacturing flexibility changes the way in
which manufacturing interacts both with external
customers and with other functional units
within the firm. To be customer responsive,
manufacturing can no longer be buffered from
the influence of customers and other functional
units (Bowen et al ., 1989). Increased inter-
actions are likely to emerge between customers
and manufacturing, and between manufacturing
and other functional units. It becomes increas-
ingly difficult to assign tasks to individual
functional units and to measure the performance
of individual tasks. Effective performance results
from “team” efforts. These teams can involve
multiple functional units and, in some cases,
customers (Larsson & Bowen, 1989). The
necessary interactions between members of the
“team” in their performance of interfunctional
tasks introduce higher levels of interdependen-
ties between manufacturing and other functional
244 M. A. ABEBNETHY and A. M. LILLIS
units. Interdependencies are likely to change
from being sequential and resemble the recipro-
cal or team interdependencies described by Van
de Ven et al . ( 1976). These forms of interdepen-
dencies can have direct implications on the
organizational arrangements required to co-
ordinate task performance effectively. Flexibility
requires “organizational arrangements to be
functionally more integrative, . . . i.e. the
structure should be organic” (Parthasarthy &
Sethi, 1993, p. 532). This organizational form
encourages co-operation and co-ordination
among functional subunits (Bruns & Waterhouse,
1975; Galbraith, 1973; Thompson, 1967).
in the field we attempt to answer the following
questions:
(1) Do lirms committed to manufacturing
flexibility use performance measures which de-
emphasize accounting and other efficiency
measures?
(2) Does a commitment to flexibility intluence
interfunctional structural arrangements? In
particular, do lirms increase their reliance on
integrative liaison devices?
It is expected that the functional interdepen-
dencies created when hrms pursue flexible
manufacturing strategies will increase the use
of integrative liaison devices to reduce the
rigidity of functional structures. Spontaneous
contact and “integrative liaison devices” (such
as task forces and committees) which allow
regular, personal, and intensive contact among
experts and decision makers of different depart-
ments, greatly facilitate collaboration among
functional units (Miller, 1988, p. 286). These
co-ordination mechanisms provide a means of
breaking down the functional barriers imposed
by mechanistic organizational structures. To the
extent that co-ordination requirements necessi-
tate the full integration of functional units, firms
may choose more “sophisticated and costly”
design alternatives such as product-line or
matrix structures (Galbraith, 1973, p. 47).
(3) Is the performance of the firm enhanced
when the manufacturing performance measure-
ment system or structural arrangements are
adapted to facilitate the implementation of
manufacturing flexibility?
Sampl e sel ecti on
Research questi ons
We have argued that Iirms will adapt their
control system to facilitate a commitment to
manufacturing flexibility. ’ Using data collected
The study is based on data collected in manu-
facturing firms located in Melbourne, Australia.
Firms were identified from the Victorian Manu-
Edcturing Directory using the Australian Standard
Industry Classification (ASIC). The selection of
industry categories was limited to those where
it was expected that there would be variation
on the flexibility dimension. Only lirms in the
Melbourne metropolitan area, and those with
more than 200 employees, were included in the
target sample.2 This resulted in an initial sample
of 81 business units. Forty-two lirms agreed to
participate, and Table 1 shows the number of
firms by size category for each industry. Using
the limited data provided by the Manufacturing
Directory, a comparison of the responding firms
to the target sample indicated no systematic
response bias.
METHODOLOGY
’ Following prior literature, we have assumed that strategic priorities are a response to a firm’s competitive environment
and that decisions concerning strategic direction also involve technological and administrative decisions to achieve strategic
priorities (Miller, 1988; Govindarajan, 1988; Parthasarthy & Sethi, 1992). It is recognized that the assumption concerning
the unidirectional nature of these relationships is simplistic. It is, however, diflicuh to establish causality with cross-
sectional data, although research in the general management literature is beginning to develop models for that purpose
(Miller et aL , 1988).
‘Several tirms included in the sample had reduced the number of employees at the time the interviews were conducted.
These firms were not excluded in the analysis.
MANUFACTURING FLEXIBILITY 245
Industry classification
TABLE 1. Sample description
Size of firms by number of employees
150-299 300-499 500-799 800-999 1000+
Paper products I 1 1
Fabricated metal products 5 6 2
Appliances, electrical equipment and machinery 7 8 2 3
Plastic and related products .)
I 1 1
Data collection
We obtained data from semi-structured inter-
views with the general manager or managing
director of the firm (the questions included in
the interview protocol are provided in Appendix
A), with each interview lasting approximately
two hours. We adopted this form of data
collection, as the link between manufacturing
flexibility and control system design has received
relatively little empirical attention. At the time
of the study there were no established measure-
ment instruments which had been subjected to
sufficient psychometric assessment for us to be
confident in either the measures’ reliability or
their validity. Difficulties with using arms-length
questionnaire measures to capture these con-
structs continues to be a challenge for research
in this area (Parthasarthy & Sethi, 1993).
The semi-structured interview method en-
abled effective follow-up questioning where
necessary, thus giving a clearer understanding
of the variables and their organizational context.
These data were particularly useful in the
development of measures of manufacturing
flexibility and use of integrative liaison devices.
We taped all interviews and transcribed them
verbatim. Each of us then independently coded
the responses. The use of the interview tran-
scripts in this coding procedure is described in
the next section.
Measurement of variables
Codings on each variable were developed by
the researchers independently using the entire
interview transcripts and other information
collected during the site visits. Full utilization
of the data was particularly important in the
measurement of the flexibility and integrative
liaison devices constructs. These are both
nebulous constructs comprising a number of
dimensions (Parthasarthy & Sethi, 1993). While
specific questions from the interview protocol
were designed to focus on defined aspects of
these constructs, it was necessary to use the
complete data set to develop ratings for each
firm. Thus, for these two variables the scaled
responses to individual questions do not build
directly into measures of variables. The way in
which the data about each firm were used to
code the responses is reflected in the schemas’
developed for both constructs. We also provide
profiles for seven firms in Appendix B to
illustrate the nature of the data collected and to
demonstrate the multidimensional nature of
manufacturing flexibility as well as the different
structural arrangements adopted by firms to co-
ordinate manufacturing activities.
Manufacturing jZex9ility. Manufacturing
flexibility is reflected in a firm’s ability to
respond to market demands by switching from
one product to another through co-ordinated
policies and actions (Nemetz & Fry, 1988) and
a willingness or capacity to offer product
variations (Buffa, 1980; Bowen et al., 1989).
Respondents were asked questions about flex-
ibility, and the elaborated responses and site
inspections* allowed the researchers to rate
independently each firm on a five-point scale
‘We wish to acknowledge the suggestion, made by an anonymous referee, of elaborating and illustrating the schema used
in the coding process.
* Firms were either independent or autonomous divisions with profit centre responsibilities.
246 M. A. ABERNETHY and A. M. LILLIS
ranging from 1 (little or no flexibility) to 5 (a
great deal offlexibility). The following two items
from the interview protocol formed the basis of
the initial questions posed to each respondent:
(a) proportion of turnover from non-standard
product lines;
(b) extent to which the manufacturing pro-
cess provides flexibility to offer customers
product variations.
The data collected in the field resulted in the
identification of three key dimensions of the
flexibility construct. These dimensions were
subsequently used to develop a schema for
rating firms on flexibility:
(a) technological dilKculty in making product
changes;
(b) strategic commitment;
(c) turn-around time to meet customer
demands.
The following discussion and the firm profiles
(Appendix B) provide some insight into the way
the schema was developed and used. Firstly, the
simplicity of product changes emerged as a
critical first-order discriminant. For example,
the firms described in profiles 1 and 2 (Appendix
B) produce a large range of what the General
Managers referred to as non-standard products.
However, the interviews and the Iirms’ product
brochures indicated that these non-standard
products required only simple manufacturing
changes from standard products. For example,
the manufacturer of medical bandages (profile
1) was able to vary the adhesive placed on the
bandages to suit climatic conditions. This
adjustment to the manufacturing process did not
constitute flexibility. These lirms were given a
“ 1” for flexibility on the basis that flexibility was
largely Irrelevant in accommodating the customer
demands placed on them.
Secondly, firms were distinguishable in their
relative strategic commitment to standardiza-
tion and flexibility. It was quite straightforward
to rate a firm which was a mass producer of
standard product lines. The proportion of
turnover from standard products was usually
oCLlOO% and the manufacturing process did
not facilitate product variation. These firms were
given a “ 1” for flexibility.
It was also relatively easy to rate those firms
we are really committed to a mainstream group of
standard products basically, the whole trick I suppose
in terms of trying to get your costs down is in fact
to standardize as much as you can there is no
flexibility in essence in the system formally.
there is a standard range of colours and that’s it and
if you want something that’s outside that range, well
that’s too bad
which had achieved a great deal of flexibility
(see profile 7). Such firms rated highly on all
dimensions of flexibility.
every single item [is]. custom designed and custom
built.
every manufacturer has a dilferent size cabinet
there is no standardization
most products [could be made] in a flexible way
we can punch it differently fit different electrical
components it’s very little hassle at all
Between these extremes, firms were rated on
their actual flexibility reflected in the cycle time
required to meet customer-initiated product
variations. Generally, reductions in cycle time
were achieved through adoption of non-dedicated
tooling and other flexible production tech-
nology. For example, contrast the following
quotations:
nowhere near as flexible as we would like. We always
liken it to a battleship where it takes miles to turn it
around.
the place goes into a bit of turmoil when we have to
make a non-standard run.
all the work we are doing in terms of investment is
aimed at increasing flexibility and reducing change-over
times, minimizing run sizes
we have the capability to make the product perform
better by customizing it and we are, as a business,
learning how to use that in a market sense
Thus, ratings from 2 to 5 reflect degrees of
product change complexity, strategic commit-
ment to flexibility and the implementation of
processes and technology which enable manu-
facturing to switch from one product to another
with relative ease. The schema described above
is illustrated in Fig. 1.
MANlJFACTLJRING FLEXIBILITY
Step 1
247
Step 3
Turnaround (cycle) time for customer
Fig. 1. Manufacturing flexibility measurement schema.
The schema was useful in ensuring that the
interview data were interpreted reliably. Once
each of the researchers had independently
analysed the interview data and determined the
rating for each firm, an inter-rater coefficient was
computed. The coefficient for the flexibility
rating indicated a high level of convergence
between the two raters [r = 0.96, p < 0.00 11. In
the three cases where the two researchers gave
very different ratings, the researchers subsequently
reviewed the transcripts, discussed the results
and developed a consensus rating for that firm.
I ntegrative liaison devices. Based on instru-
ments used previously by McCann & Galbraith
( 1981) Mintzberg ( 1983) and Van de Ven et
aZ. (1976) respondents were asked about their
reliance on each of the following integrative
liaison devices:
-
spontaneous contact between department
managers;
-
regular meetings between departments;
-
interdepartmental task forces;
- permanent teams;
- permanent liaison positions.
Respondents were also asked to assess the
extent to which the work or the work process
within departments was governed by the use of
standard operating procedures and whether the
organizational structure was functional, product
line or matrix.
The interview data were used to rank each
firm on a five-point scale reflecting the extent
of reliance on integrative liaison devices. The
data indicated that there were only three forms
of integrative liaison devices in general use in
the firms studied: spontaneous contacts, regular
meetings, and teams and task forces.
Teams and task forces were combined, as
most firms using these forms of integrative
devices did not distinguish between them. In
almost all cases the only interfunctional liaison
role was vested in the firm’s general manager.
There were no cases of matrix organizations and
only three product-focused firms. The relative
248
Step 1
M. A. ABERNETHY and A. M. LILLIS
Step 2
Step 3
Fig. 2. Integrative liaison devices (ILD) measurement
schema.
absence of higher-level liaison devices appears
to be a function of the size of firms under study.
The small size of these ftrms is, in fact, a general
feature of Australian manufacturing industry.5
A schema for rating the use of integrative
liaison devices was developed as follows. We
considered a product-focused structure as a
formal first-order response to developing func-
tional integration. As only three firms had a
product-focused structure, it was then necessary
to assess alternative approaches used by Iirms
to achieve functional integration. Within func-
tional structures, the interviews established that
“flatness” in the managerial authority structure
was critical to effective integration. This “flat-
ness” was mentioned so frequently in our
interviews that it became the essential discri-
minant in our classifying schema for firms with
functional structures. Profiles 4, 6 and 7
demonstrate the notion of a flat management
structure. If a product-focused structure is
considered a formal first-order response to
developing functional integration, then flatness
is a second-order, less formal approach to
integration within a functional structure.
Flatness is designed to encourage integrative
problem solving, fast response (by reducing
bureaucratic barriers) and spontaneity. There
appeared to be a widespread belief among
respondents in the effectiveness of “flatness” in
achieving integration ~z’s&vi~ other formal
mechanisms such as meetings and teams. This
led us to believe that firms adopting such a flat
structure would be relatively more functionally
integrated than those making extensive use of
meetings and teams without such flatness. Thus,
we gave a high loading to flatness in coding the
use of integrative liaison devices. The steps
’ The majority (97% ) of Australian manufacturing firms are less than 199 employees in size. Based on the latest published
statistics there were only 884 firms in Australia with more than 200 employees. Only 60 firms (0.002% ) have more than
1000 employees (Australian Bureau of Statistics, 1990).
MANUFACTURING FLEXIBILI’IY 249
involved in coding each firm on the use of
integrative liaison devices are represented via
the schema in Fig. 2.
The schema and the profiles presented
in Appendix B demonstrate how the data
were used to develop a rating for each firm.
A “5” represented the adoption of a product-
focused structure to facilitate interfunctional co-
ordination. A “4” was given where a firm had a
functional structure but had deliberately created
a flat structure, relying extensively on sponta-
neous integrative liaison devices to manage co-
ordination requirements (see profile 4). Within
hierarchical functional structures, co-ordinative
requirements could be classified as minimal
through to substantial. Where interfunctional
contact was relatively infrequent (e.g. monthly
meetings) the rating was “1” (see profile 1).
Ratings of “2” and “3” reflected the differing
degrees of reliance on teams and meetings
required to co-ordinate activities between
functional units. The ratings between each
researcher were again compared, and the inter-
rater reliability coefficient of 0.94 0, < 0.001)
obtained for this variable indicated a high level
of convergence, with four cases subsequently
requiring review consensus.
Manufacturingperformance measures A list
of performance measures was developed for this
study, based on Kaplan ( 1983) Howell & Saucy
(1987) and Chase (1990). The list included
items related to cost efficiency, flexibility and,
for completeness, quality and dependability
measures. During the interviews, respondents
were asked to indicate the use and relative
importance of each measure. The questionnaire
items emerged as non-discriminatory. Most
firms reported extensive use of virtually all
measures. Respondents were then asked to
identify and rank their three most critical
manufacturing performance measures in order
of importance. Respondents had no difficulty in
identifying the three most critical measures. In
the analysis, we used only these rankings to rate
the relative importance of efficiency-based
measures in the manufacturing performance
measurement system. There were no Iirms
which did not include an efficiency-based
measure in the three most critical measures.
There were, however, differences in the ranking
attached to efficiency-based measures, thus
enabling lirms to be coded on a scale of 1 (least
important) to 3 (most important).
Pwformunce. Following Khandwalla (1972)
Brownell & Merchant (1990) and others,
general managers were asked to rate the
performance of the firm relative to that of
competitors on a five-point fully anchored
Likert-type scale.
RESULTS AND DISCUSSION
Flexibility and control q&em design
Since this study is essentially exploratory, we
used simple inferential statistics to examine the
relationship between manufacturing flexibility,
efficiency-based performance measures and use
of integrative liaison devices. In the first
instance, we examined the bivariate correlations
between the variables. As can be seen from the
correlation matrix provided in Table 2, the
relationship between flexibility and use of
integrative liaison devices is in the direction
predicted. That is, the pursuit of flexibility
increases the extent to which lirms use integra-
tive liaison devices to manage functional inter-
dependencies and to break down functional
barriers. The correlation between flexibility and
use of efficiency-based performance measures is
also significant and in the predicted direction.
The use of efficiency-based performance measures
in manufacturing declines as a firm’s commit-
ments to flexibility increases. We also split the
sample into those Iirms which pursued flexibility
and those which did not, and performed a
parallel analysis which focused on tests of
difference between the means. The results are
presented in Table 3 and are in the predicted
direction. That is, the non-flexible group placed
more importance on efficiency-based measures
and used integrative liaison devices to a lesser
extent than those firms in the flexible manufac-
turing group. The t-values indicate that the mean
scores are significantly different from each other
for both control elements. These results add
250 M. A. ABERNETHY and A. M. LLLLIS
TABLE.2. Pearson correlation (significance)* matrix (n = 42)
Variables 1 2 3
1. Flexibility
2. Integrative liaison devices
3. Etficiency-based PMS
4. Performance
* One-tailed test.
0.46 (0.001)
-0.24 (0.06) -0.10 (0.27)
0.13 (0.12) 0.28 (0.04) 0.08 (0.31)
TABLE 3. Means and t-test for use of efficiency-based performance measures and
integrative liaison devices for lirms with low flexibility and high flexibility
Group 1: Group 2:
Non-flexible ftrms Flexible firms t-value (sip)
(n = 20) (n = 22)
Mean (S.D.) Mean (S.D.)
Efficiency-based performance
measurement system
Use of integrative liaison
devices
2.60 (0.50) 2.27 (0.76) 1.62 (0.05)
2.30 (1.21) 3.45 (1.01) 3.35 (0.001)
l One-tailed test.
further support to the idea that firms committed
to flexibility use integrative liaison devices to a
signiIicantly greater extent and rely on efficiency-
based performance measures to a significantly
lesser extent.
These findings are consistent with our expecta-
tions and add weight to the integrative fiame-
works of control which argue that organizations
adapt their control system to meet the control
requirements of the situation (Flamholtz, 1983;
Govindarajan, 1988; Merchant, 1985). It would
appear, Tom the evidence provided here, that the
control needs imposed by flexibility result in firms
placing different emphasis on the elements which
form part of the control system package.
Pet$wrnunce, munuf&cturi ng~i bi l i ~ and
control system &si gn
The third research question of interest in this
paper is whether firms which adapt their control
systems to support their commitment to flexibility
perform better than those firms which do not.
To answer this question, we again used simple
inferential statistics.” The sample was split into
the same two groups as the previous analysis.
One group consisted of firms with a degree of
flexibility and the second group comprised
those Iirms which were not flexible. We
computed correlations between the two control
elements and performance for each group. If
adapting control systems to meet strategic
‘We also assessed whether the match between flexibility and each control element influenced tirm performance by fitting
two regression equations, each employing a multiplicative interaction term. The equations were of the same form as used
by Brownell (1982) and others (Govindarajan, 1988; Hirst, 1983). We examined the coefficients of the interaction
terms to assess if the appropriate “match” resulted in better performance. We expected performance to be atkcted
ditferently accordingly to the extent to which manufacturing flexibility is pursued. The results are presented in Appendix
C and are consistent with those presented in the paper. The reliance on efficiency-based measures has a progressively
more negative effect on performance as manufacturing flexibility increases in importance (as = -0.50, t = 2.13, p =
0.04). However, although we expected increased reliance on integrative liaison devices to have a progressively more
positive effect on performance as flexibility increases (i.e. a3 > 0), the coefficient was negative but not significant
MANUFACTURING FLEXIBILITY
TABLE 4. Test of the difference in the correlations between firm performance and use
of controls
251
Correlation (sig) between
efficiency-based performance
measures and firm performance
Group 1: Group 2: Difference in
Non-flexible firms Flexible firms correlations
(n = 20) (n = 22) t-value (sig)
0.44 (0.03) -0.08 (0.35) 1.60 (0.05)
Correlation (sig) between use of
integrative liaison devices and firm
performance
0.27 (0.12) 0.38 (0.04) 0.31 (ns.)
l One-tailed test.
commitments infhtences performance, the cor-
relation between the use of efficiency-based
measures and performance should be signilicantly
higher for ftrms which do not pursue flexibility
than for those which do. In contrast, we expect
the correlation between performance and use
of integrative liaison devices to be significantly
higher for flexible Iirms. The separate correla-
tions between performance and each control
element for the two groups is presented in
Table 4. These correlation coefficients indicate
whether the relationship between performance
and the control element is significantly different
from zero. What is of interest here, however, is
whether the coefficients for each group are
significantly ditferent from each other, as opposed
to horn zero. To test for differences between the
correlations we used the Fisher transformation
test (Winkler & Hayes, 1975, p. 653). The t-
values and significant levels which resulted from
this test are presented in Table 4 and provide
some support for our expectations. Not only
were the correlations between performance and
use of efficiency-based measures in the pre-
dicted direction for the two groups (the
correlation was positive for the non-flexible
firms and negative for the flexible firms), the
correlations also differed from each other at less
than the 0.05 level of significance (one-tailed
test). However, while the correlation between
use of integrative liaison devices and perform-
ance appeared to be larger for the flexible
manufacturing firms, the difference between the
coefficients for the two groups was not significant.
Although the difference in the correlation
coefficients between performance and the use
of integrative liaison devices was not significant
for the two groups, the relationship between
performance and use of integrative liaison
devices was positive and significant for firms
committed to flexibility. This supports the data
collected in the field. The expanded responses
by General Managers indicated that organizational
arrangements, and in particular, the develop-
ment of integrative liaison devices were, in fact,
the key facet of the organization control system
in the effective management of manufacturing
flexibility. It was considered more important to
get the structural arrangements right than the
design of the performance measurement system
for fums pursuing flexibility. Informal and
organic management structures, development
of lateral linkages, and the development of an
organization culture which encouraged indivi-
duals to identify with corporate goals, appeared
to carry strong control implications for effective
strategy implementation. Changes to the per-
formance measurement system appeared to be
less of an imperative to guarantee achievement
of strategic goals relating to flexibility.
Much of the contingency-type empirical
research in accounting has stressed the import-
ance of empirically examining the impact of an
appropriate match between control system
design and contextual factors. However, the data
collected in this study highlight the difficulties
identified by researchers in obtaining significant
results for “fit” hypotheses (Hopwood, 1976;
Merchant, 1981; Kim, 1988) and question the
practical significance of any such findings. It is
252 M. A. ABERNETFN and A. M. LILLIS
interesting to note the range of reported
influences on performance. The following are
some of the responses given. None of these are
related to the “fit” or “misfit” between control
system design and manufacturing strategy.
Responses are from general managers of low
performing firms as they naturally tended to
elaborate their responses on the performance
question.
performance, the qualitative data collected in
the field suggest the need for caution in
interpretation.
SOME CONCLUDING COMMENTS
- General Manager had resigned, set up with
a competitor, and taken much of the business.
- Changes in import regulations which
allowed competitors into the market. Another
firm had lost considerable market share due to
entry of an Australian competitor who was
constantly undercutting them on price.
This study sought to establish a link between
manufacturing strategy and two dimensions of
an organization’s management control system:
structural arrangements and the performance
measurement system. In particular, the study
focused on examining differences that flow from
the pursuit of flexibility as a manufacturing
strategy,
- One Erm had an appropriate “fit” between
strategy and control system design but was
pursuing a strategy which did not give them a
competitive advantage.
- Out-dated manufacturing technology and
absence of investment capital due to current
economic climate.
- Currently undergoing a major restructuring
of the firm.
- Economic downturn.
Attempts to measure the impact of “fit” on
performance also ignore the dynamic nature of
organizations. Several firms had only recently
changed, or were in the process of changing,
either their strategy or control system design.
These changes, however, were not reflected in
the firm’s performance in the current period.
One firm, for example, indicated that its
performance had significantly improved in the
past year, although overall performance was still
considered relatively low “. . . we were a
minus five last year, this year we are a plus
two .”
Our results clearly support the notion that
organizations adapt their structural arrange-
ments in order to implement manufacturing
flexibility. Increasing reliance on integrative
liaison devices in the form of teams, task forces,
meetings and spontaneous contacts reflects the
interconnectedness of functions that has been
theoretically argued to flow from customer-
responsive manufacturing strategies (Nemetz &
Fry, 1988; Bowen et al , 1989; Parthasarthy &
Sethi, 1993). The results also provide support
for the negative relationship between flexibility
and the role of efficiency-based performance
measures.
The study has potential limitations, particularly
in the area of measurement. The use of interviews
for data collection can be criticized on the basis
that managers’ responses, or the interpretation
by the researchers of the responses, are likely
to be biased. Following Merchant & Manzoni
( 1989) we attempted to overcome this limita-
tion by examining the interview transcripts in
depth for any inconsistencies in the interviewee’s
logic. Transcripts were also coded by each
researcher independently and compared for
inconsistencies or contradictions.’
While our results do provide some empirical
evidence that the appropriate match between
control system design and flexibility enhances
Offsetting these limitations is the potential
provided by a field study methodology to
make significant contributions to the literature.
‘Ideally we would have preferred to use a third independent researcher to code the transcripts. Limited funds and
difficulties in finding an adequately trained researcher for the task precluded this course of action.
MANUFACTURING FLEXIBILITY 253
This study contributes to the literature by
drawing on a rich body of data from 42 firms,
to provide an empirical perspective on the
management of new competitive priorities in
manufacturing through the development of
integrative co-ordinating mechanisms and per-
formance measurement systems. The study also
makes a methodological contribution to the
literature. Prior research has identified the
problems with measuring nebulous constructs
such as flexibility, or use of informal co-
ordination mechanisms such as integrative
liaison devices (Parthasarthy & Sethi, 1993).
This study used the knowledge gamed in the
field to build two measurement schemas for
both these constructs. The schemas provide
future researchers with a basis for coding field
data and a foundation on which to develop
survey instruments for measuring manufactur-
ing flexibility and use of integrative liaison
devices.
BIBLIOGRAPHY
Australian Bureau of Statistics (Canberra: Commonwealth Government Printer ACT, 1990).
Bowen, D. E., Siehl, C. & Schneider, B., A Framework for Analysing Customer Service Orientations in
Manufacturing. Acudemy of Management Review (1989) pp. 75-95.
Brownell, P., The Role of Accounting Data in Performance Evaluation, Budgetary Participation and
Organizational Effectiveness,journuI ofAccounting Research ( 1982) pp. 12-27.
Brownell, P. & Merchant, K A., The Budgetary and Performance Influences of Product Standardization
and Manufacturing Process AutomationJournal of Accounting Research (1990) pp. 388-397.
Bruns, W. J. & Waterhouse, J. H., Budgetary Control and Organization Structure,Journul of Accounting
Research (1975) pp. 177-203.
Buffa, E. S., Research in Operations Management,JournuZ of Operations Management (1980) pp. l-6.
Buffa, E. S., Meeting the Competitive Challenge (New York: Irwin, 1984).
Burns, T. and Stalker, G. M., The Management of Innovation (London: Tavistock, 1961).
Chase, R. B., Dimensioning the Service Factory, in Etdie, J. E., Burstein, M. & Flegenbaum, A. (eds),
Research Agenda fur the Next Decade (Boston: Kluwer, 1990).
Chrisman, J. J., Hofer, C. W. & Boulton, W. R., Toward a System for Classifying Business Strategies, Academy
of Management Review ( 1988) pp. 413-428.
Dean, J. W. & Susman, G. I., Strategic Responses to Global Competition, in Snow, C. (ed.), Strategy,
Organization Design and Human Resource Management (Greenwich, CT: JAI Press, 1989).
Drucker, P. F., The Emerging Theory of Manufacturing, Harvard Business Review (May-June 1990) pp.
94-102.
Eccles, R. Cr., The Performance Measurement Manifesto, Harvard Business Review (January-February
1991) pp. 131-137.
Flamholtz, E. G., Accounting, Budgeting and Control Systems in their Organizational Context, Accounting,
Organizations and Society (1983) pp. 153-169.
Galbraith, J. R., Designing Complex Organizations (Reading, MA: Addison-Wesley, 1973).
Govindarajan, V., A Contingency Approach to Strategy Implementation at the Business-Unit Level:
Integrating Administrative Mechanisms with Strategy, Academy of ManagementJ ournal ( 1988) pp.
828-853.
Govindarajan, V. J. & Gupta, A. K., Linking Control Systems to Business Unit Strategy: Impact on
Performance, Accounring Organizations and Society (1985) pp. 51-66.
Hayes, D. C., The Contingency Theory of Managerial Accounting, Accounting Review (1977) pp. 22-39.
Hirst, M. K., Reliance on Accounting Performance Measures, Task Uncertainty and Dysfunctional
Behaviour: Some ExtensionsJournal of Accounting Research (1983) pp. 596-605.
Hopwood. A. G., Accounting and Human Behaviour (Englewood Cliffs, NJ: Prentice-Hall, 1976).
Howell, R. A. & Soucy, S. R., Operating Controls in the New Manufacturing Environment, Management
Accountant (1987) pp. 25-31.
Kaplan, R. S., Measuring Manufacturing Performance: A New Challenge for Management Accounting
Research, Accounting Review (1983) pp. 686705.
254 M. A. ABERNETHY and A. M. LILLIS
Kaplan, R. S., Measures fw Manufacturi ng Excel l ence (Boston: Harvard Business School Press, 1990).
KhandwaIIa, P. N., The Effect of Different Types of Competition on the Use of Management Controls,
journal of Accounti ng Research (1972) pp. 275-285.
Kim, K. K., Organizational Co-ordination and Performance in Hospital Accounting Information Systems,
Accounti ng Revi ew (1988) pp. 22-39.
Iarsson, R. & Bowen, D. E., Organization and Customer: Managing Design and Coordination of Services,
Academy of Management Revi ew (1989) pp. 213-233.
Macintosh, N. B., The Soci al Software of Accounti ng and I nformation Systems (New York: Wiley, 1985).
McCann, J. & GaIbraith, J. R., Interdepartmental Relations, in Nystrom, P. & Starbuck, N. (cds), Handbook
of Organi zati onal Desi gn, Vol. 2, pp. 60-84 (New York: Oxford University Press, 1981).
Merchant, K. A., The Design of the Corporate Budgeting System: Inihtences on Managerial Behaviour and
Performance, Accounti ng Revi ew (1981) pp. 813-829.
Merchant, K A., Organizational Control and Discretionary Program Decision Making: A Field Study,
Accounti ng, Organi zati ons and Soci ety (1985) pp. 67-85.
Merchant, K A. & Manzoni, J., The Achievability of Budget Targets in Profit Centres: A Field Study,
Accounti ng Revi ew (1989) pp. 539-559.
Miller, D., Relating Porter’s Business Strategies to Environment and Structure: Analysis and Performance
Implications, Academy of Management J ournal (1988) pp. 280-308.
MiIIer, D., Droge, C. & Toulouse, J., Strategic Process and Content as Mediators between Organizational
Context and Structure. Academy of Management J ournal ( 1988) pp. 544-569.
Mintzberg, H., Structuri ng i n Fi ves (EngIcwood CliI%, NJ: Prentice-Hall, 1983).
Nemetz, P. L. & Fry, L. W., Flexible Manufacturing Organizations: Implications for Strategy Formulation
and Organization Design, Academy of Management Revi ew (1988) pp. 627-638.
Parthasarthy, R. & Scthi, S. P., The Impact of Flexible Automation on Business Strategy and Organizational
Structure, Academy of Management Revi ew (1992) pp. l-26.
Parthasarthy, R & Scthi, S. P., Relating Strategy and Structure to Flexible Automation: A Test of Fit and
Performance Implications, Strategi c Management J ournal ( 1993) pp. 529-549.
Simons, R., Accounting Control Systems and Business Strategy: An Empirical Analysis, Accounti ng
Organi zati ons and Soci ety (1987) pp. 357-374.
Simons, R., The Role of Management Control Systems in Creating Competitive Advantage: New Perspective,
Accounti ng, Organi zati ons and Soci ety (1990) pp. 127-143.
Thompson, J. D., Organi zati ons i n Acti on (New York: McGraw-Hill, 1967).
Van de Ven, A. H., Delbecq, A. L. & Koenig, R. Jr, Determinants of Co-ordination Modes within
Organizations, Ameri can Soci ol ogi cal Revi ew (1976) pp. 322-338.
WinkIer, R. L. & Hayes, W. L., Stati sti cs: Probabi l i ty, I nference and Deci si on, 2nd Edn (New York: Holt,
Rinehart & Winston, 1975).
APPENDIX A: STRUCTURED INTERVIEW QUESTIONNAIRE
Note: AI1 items used a fully anchored five-point scale ranging from one to five
1. What proportion of your turnover comes from non-standard product lines?
2. Does your manufacturing process provide the flexibility to offer customers product variations?
Sh-uctural arrangements (i ntegrati ve I i ai son devi ces)
1. Would you describe your organization as having a functional structure, i.e. one where responsibilities
are divided primarily by function, e.g. marketing function, manufacturing function?
2. Would you describe your organization as having a product-focused structure, i.e. where aI1 functional
departments working on one product are grouped together?
3. To what extent is there a need to co-ordinate the work of the manufacturing department with other
departments, e.g. marketing, engineering?
4. How often are regular meetings held between departments?
5. Does your organization use inter-departmental task forces (usuaIIy having a limited and relatively
short life), for problem solving, system design, department of product lines, manufacturing and co-
ordinating product lines?
MANUFACTURING FLEXIBILITY 255
6. Are teams (more permanent and having a longer time orientation) used for these purposes?
7. Are there any permanent iiaison positions within this organization, whose roles and responsibilities relate to co-
ordinating manufacturing activities with other departments?
8. Is the co-ordination between departments (e.g. marketing, engineering, manufacturing) primarily by spontaneous
contact between managers of the various departments?
9. To what extent is the work within departments governed by the use of standardized operating procedures or work
processes?
Performance measu- t system
Respondents were asked to rate the extent to which each of the following measures were used for perlbrmance evaluation
in the manuktcturlng department of their organization.
1. On time delivery performance records.
2. Number of customer complaints.
3. Iabour utilization/efficiency statistics.
4. Incidences of product defects.
5. Number of warranty claims.
6. Inventory turnover ratio.
7. Reports on whether standard product costs are met.
8. Survey of customer satisfaction.
9. Measurement of cost reduction due to quality improvements.
10. Evaluation of whether people in manufacturing are co-operative and responsive to the demands of other departments
within the organization.
11. Customer satisfaction with problem solving and sales assistance.
12. Rate of introduction of new products.
13. Evaluation of the ability to vary product characteristics.
14. Length of cycle time from order to delivery (for standard products).
15. Length of cycle time for customer-requested product variations.
16. Material purchase price variance.
17. Rate of material scrap loss.
18. Measurement of machine utilization and down time.
256 M. A. ABERNETHY and A. M. LILLIS
APPENDIX B: MEASUREMENT OF FLEXIBILITY AND INTEGRATIVE LIAISON DEVICES
PROFILE 1
‘ROFILE 2
‘ROFILE 3
‘AOFILE 4
-
FLEXIBILITY
A manufactL/rer of medical supplies with 420
emloyees. Ninety per cent of turnover comes
from the production of standard lines. The
non-standard products only differ in
specifications as to the adhesive or weight of
cloth used in production.
Aluminium can producer with 210 employees.
Mass producer of standard lines. Products
onty vary in logos. P/ant runs 24 hours per
day. High off-the-shelf product avaitabitity is
strategba/& critical and the manufacturing
process does not e&/y facilitate changes in
product demands. “We’ve got 200 different
/abets and it takes half-an-hour to an hour to
change a machine from one label to the other
. ..so we carry buffer stocks.. For short-run
jobs we’// carry, in some instances, 72 months
stock for customers, because 8s economical
for us to do if that way”.
A manufacturer of nails. nuts, bat& and
screws with 850 employees. High off-the-
shelf availabilify is more important than any
other strategk priority. The Product mix is 60
per cent standard and 40 per cent
nonstandard. However, “in a lot of cases they
(the non-standard products) are very similar to
a standard product, they want a longer thread
or whatever...in some cases they are different
products”. The manufacturing process is
inflexible and usua//y invokes long production
runs. ‘We are nowhere near as flexible as we
would like it to be we always liken it to a
battleship where it takes miles to turn it around
.
Aluminium can manufacturer employing 145.
The firm is reelatively flexible. It produces
highly decorated cans with special fittings,
shapes and sizes. :..k tends to be more
flexible [than the division described in Profile
t], it tends to be slower as well and the capital
investment is lower.’ Current investment is
aimed at increasing f/exbi/ity and reducing
changeover times. However, decorating “set
up times for a smaN run of cans costs more
than the actual running casts through the tine.
As a result stock ho/dings are still relatively
high (2.8 months of cover). We err on the
side of holding stock” to ensure that customer
demands are met on time.
Rating
1
1
INTEGRATIVE LIAISON DEVICES
The firm has a high& functional and
bureaucratti structure. The relationship
between sa/es and production is sequential
and there was evidence of functional barriers
between the departments (ie separate lunch
rooms). Marketing determines the sales
forecasts and production schedules are
based on that forecast. AM of the interface
between functional departments is through
formal monthly meetings. Task forces are
used maybe twice a year to solve a specific
orobtem. A task force recentlv assessed the
causes of increased material usage.
There is a strong need for co-ordination
between sales and production because of the
high customer service (delivery) commitment.
The need to respond to daity order changes
(depending, for exan@e, on the weather and
associated demand for SOI? drinks)
discourages reliance on forma/ co-ordinative
mechanisms and standard operating
procedures. Spontaneous verbal
communications are heavily retied upon.
Special purpose %orrective action teams”ere
used for problem-solving in a// areas (not
necessarity production), including, for
exan@e, at the time of this study, workers
The firm has recentfy reorganized into what
can best be described as a product-line
structure for the manufacturing segment of
the business. Sales and distribution remain a
separate functional department responsible to
the GM. There are three separate product
lines. The relationship between sales and
manufacturing remains sequential. They
have a materials management group which
acts as the linchpin between sales and
manufactur/ng. Sales and manufactuting
meet regularly two to three times a week for
non-standard products and once monthly for
standard products.
The management structure is functional but
%ompressed”and in the process of changing
from a more formal meeting-based structure
to an informal structure, “We have meetings
and meetings and that is part of the probtem.
We’re changing a// the time, setting up
communication vehicles . ..sncouraging
people to pick up the phone, go and talk to
someone to get it done. * Teams are used,
with lessening invotvement of managers and
accountants.
-
Ming
MANLJFACTURING FLEXIBILITY
PROFILE 5 Punp rnanufact&m?r enqioyk&7 280. Seventy- 4
~o_~pj;“at;on wi?hin the functional structure is 4
five per cent oi turnover is custom-buik
assisted by product managers within
puw,motor and base p/ate combinations sold
marketing who manage price information,
on order. The remaining 25 per cent of
technical data, stock avai/abi/ity, sale.5
turnover is sfandard base punp sates supplied
forecasts and projects and so on. These
from stock The manufacturing process is
product managers determine product
very ffexibk in the contexl of identified technical specifications so that they are
markets: V’s not very fJexib/e. /t’s flexible to
‘market-dtien rather than manufacturing-
a degree to cover the markets that we clearly driven”. Standard operating procedures are
identify . ..but in term of being able to make heavily relied upon to manage engineered
things other than puws andpuw sets-no, product changes (resu/ting from order
we are not cut out to do that. We can make a changes). Co-ordhative devices are used
wide variety of concepts in any shape or form extensively and include forma/ meetings,
within genera/ limitations that punlp
teams and spontaneous contacts for problem
manufacturers can envisage For the market solving
that we sewice, I think rhal our organization
has more flexibility than mosl of our
competitors...”
PROFILE 6 A /age cab/e manufacturer emfoying 1,400 4 The structure is messy and a result of a 4
en@oyees. It has two production plants. One number of mergers and takeovers ..“ii’s never
p/ant produces standard pmducts...“it’s our been structured ..A just happened.’ Co-
sausage factory, ft just makes virfua/ty ordination between funct!ona/ units is
continuous products’! Seventy per cent of achieved by... “having regular meetings,
turnover comas from a second plant which is irrpromptu meetings, grabbing three or four
“more like a jobbing shop .., more of the extra guys and bringing them together to discuss a
high voltage and the specials, it is certainly not problem. Task forces are used to a “very
run as a process factory”. They make great extent” for problem solving and product
everyihing to order . ..“you can’t have that stuff development. Senior management has made
., sitting around, we// you could, but would go a deliberate effort to ensure that the
broke _.. we need about 6.3 tonnes of copper a organization is f/at and informal. . ..V’d rather
day and we are paying $3750 a lonne . ..you not call a meeting ._. I mean, rea//y once a
don’t have too much stock iying around”. month to discuss the resu/ts, maybe .we do
Flexibility is necessary for survival Yapart not write memos to each other. As soon as I
from the contract stuff, everything is virtual/y a see a memo . ..someooe is going for political
speciai and that’s why _..you can’l afford not to points __, that is when we real& kick ass. I get
be flexible... .you’re dead if you’re not flexible them up here and say you, you and you. sort
it out and that’s it”. Liaison between sales
and manufacturing is dons through informal
m&-action .._ “if I find out we’ve got a problem
and these blokes haven’l even been tatking
to each other, a// he// breaks loose. They
interact . ..they get the job done and no cheap
points ,just get it done or we wi// get
someone who can . ..a// departments interact
or we fix them”. The GM summanied the
structure as . ..“/‘ve got to say, this is a veiy, I
think. a fairl), informal management structure.
it’s mn a lilt/e bit like a big milk bar”,
PROFILE 7 The second largest /~ght finings manufacturer 4 The structure of the firm is functional but
4
in Austra/ia em/eying 290. /t manufacturers a management has ensured thal the structure
range of standard light fillings but a lol of the is f/at . ..“the conpany is a fair/y f/at structure, I
work is special right fittings. They have mean we haven‘t got a who/e heap of
adopted flexibility as a strategy priority :..a tot management /eve/s and things like that, so jf
of the machinery we have deliberalsly we are looking at new products you get
selected because it’s f/sxible machinery. We someone from the factory, someone from the
have avoided wherever possible going to design office, someone from the sales office:
dedicated tooling, hard tooling, that isoY Production and sales ‘get together and work
flexible”. They have “very flexible machines out what they want. They do that fist thing
even with standard products...’ so that even in the morning over coffee”. Two pennanenl
standard products can be varied to meet teams are in p/ace. One involves
customer demands. “Most of our product is manufacturing and engineering who are
made in a very flexible way we can in fact continually %oking at our manufacturing
punch it dinerent/y .., fit different electrical strategy”, the other team includes
conyDonents . . . is very /in/e hassle at all.’ Even
the production schedule is fkuible !..as f said
engineering, finance and manufacturing and
looks at prioriiies in capital expenditure.
it b a very informal scheduling system,
flexibifity is the woti, I mean we mint things
both ways, it is love& to have a schedule. but
we also want to be abte to walk out into the
factory andput a job on at a moment’s notice.
And that is the way the p/ace operates.*
257
258 M. A. ARERNETHY and A. M. ULLIS
APPENDIX C: RESULTS OF TWO REGRESSIONS
The regressions are of the form
Y = 6 + a,F + a&, + a&__,
where Y = firm performance, F = manufhcturing flexibility, C = control elementj, j = 1, 2 ( 1 = efficiency based PMS;
2 = integrative liaison devices), FC = interaction between manufacturing flexibility and control element j.
Control element
Regression I (j = 1) Regression II (j = 2)
Efficiencv-based PMS Integrative liaison devices
a, = Constant (SE.)
t-value (sig)
a, = Flexibility coefficient (S.E.)
t-value (sig)
a2
= Control element coefficient (S.E.)
t-value (sig)
as = Interaction coefficient (S.E.)
t-value (sig)
R2
Sip of F f one-tailed test)
1.77 (0.89)
1.99 (0.05)
0.45 (0.19)
2.34 (0.02)
0.5 1 (0.30)
1.69 (0.10)
-0.50 (0.23)
-2.13 (0.04)
0.13
1.96 (cO.05)
2.37 (0.55)
4.33 (0.00)
0.19 (0.20)
0.95 (0.35)
0.37 (0.19)
1.96 (0.06)
-0.23 (0.20)
-1.17 (0.25)
0.11
1.54 (n.s)
doc_921478334.pdf