Power, organization design and managerial behaviour

Description
This paper examines empirically the impact ofauthority structures on the use of accounting information systems
(AISs) for decision control and decision management. The model is designed to enable an assessment of the relative
impact of formal authority that stems from allocation of decision rights and informal authority that stems from individual
power and influence. The study is based on data collected from physician managers in large public teaching
hospitals in Italy. The results support the hypotheses and demonstrate the consequences ofpower on organizational
functioning. Our findings indicate that the delegation of formal authority to physician managers not only has a direct
impact on the use ofaccounting for decision control and decision management it also has an important effect on their
cost consciousness.

Power, organization design and managerial behaviour
Margaret A. Abernethy
a,
*, Emidia Vagnoni
b
a
Department of Accounting and Business Information Systems, The University of Melbourne, VIC 3010, Australia
b
Dipartimento di Economia, Istituzioni e Territorio, Universita` di Ferrara, Via del Gregorio, nn.13-15, 44100 Ferrara, Italy
Abstract
This paper examines empirically the impact of authority structures on the use of accounting information systems
(AISs) for decision control and decision management. The model is designed to enable an assessment of the relative
impact of formal authority that stems from allocation of decision rights and informal authority that stems from indi-
vidual power and in?uence. The study is based on data collected from physician managers in large public teaching
hospitals in Italy. The results support the hypotheses and demonstrate the consequences of power on organizational
functioning. Our ?ndings indicate that the delegation of formal authority to physician managers not only has a direct
impact on the use of accounting for decision control and decision management it also has an important e?ect on their
cost consciousness.
#2003 Elsevier Ltd. All rights reserved.
Introduction
Hospitals are implementing new authority
structures and accounting information systems
(AISs) primarily directed towards changing physi-
cian behaviour (Chow, Ganulin, Hadded, &
Williamson, 1998; Lee & Alexander, 1998). Physi-
cians are a key stakeholder in hospitals and their
involvement in resource management is seen to
be critical to hospital survival (Abernethy &
Stoelwinder, 1995). Physician resource manage-
ment behaviour is, however, likely to be directly
related to their power and in?uence within hospi-
tals. The in?uence of ‘‘power and politics’’ on
organizational behaviour is not new. Indeed, its
e?ect on organizational design choices has long
been recognized in the general management litera-
ture (Cyert & March, 1963; Hardy & Clegg, 1999;
Perrow, 1986; Pfe?er, 1992). The accounting lit-
erature provides numerous illustrations of how
management control systems are used to legitimize
and maintain systems of power as well as to
redistribute power among the various organiza-
tional actors (see for example Abernethy & Chua,
1996; Covaleski & Dirsmith, 1986; Kurunma¨ ki,
1999). There is, however, little broad-based
empirical literature examining the e?ect of power
on management controls systems or its e?ect on
organizational outcomes. Much of the empirical
research in managerial accounting, particularly
research drawing on economic theories of beha-
viour, assumes that individuals are rational and
that successful ?rms make the appropriate cost/
bene?t trade-o?s in determining organizational
0361-3682/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0361-3682(03)00049-7
Accounting, Organizations and Society 29 (2004) 207–225
www.elsevier.com/locate/aos
* Corresponding author. Tel.: +61-3-8344-5475; fax: +61-
3-9349-2397.
E-mail address: [email protected]
(M.A. Abernethy).
design choices.
1
The political model of behaviour
seriously challenges this assumption. It explicitly
recognizes that powerful coalitions within an
organization can resist attempts by management
to allocate decision rights and implement admin-
istrative systems to monitor agents’ behaviour.
Depending on the level of power it is possible to
circumvent, sabotage or manipulate elements of
the authority system and the accounting informa-
tion systems (AISs) implemented to complement
authority structures (Eisenhardt & Bourgeois,
1988; Pfe?er, 1992).
2
The purpose of this paper is to empirically
assess the impact of physician power on the use of
AISs. The model is designed to enable an assess-
ment of the relative impact of formal authority
that stems from the delegation of decision rights
and the informal authority that stems from indi-
vidual power and in?uence. While there is begin-
ning to emerge some empirical research
supporting the notion that formal authority struc-
tures are an important antecedent of the use of
AISs (Abernethy & Lillis, 2001; Chenhall & Mor-
ris, 1986; Wruck & Jensen, 1994), there is no
research, of which the authors are aware, examin-
ing the role of power in this relation. The model
examined here also enables us to assess the con-
sequences of these two forms of authority on
organizational outcomes and the in?uence of AISs
on this relation. Our study examines two roles of
AISs, namely, decision management and decision
control (Zimmerman, 1997).
3
By recognizing the
two roles of AIS we are able to develop a better
understanding of the conditions that in?uence the
role of accounting for supporting managerial
decision making and whether this role comple-
ments or con?icts with the decision control role.
Our empirical study is based on data collected
from physician managers in large public teaching
hospitals. We select this research setting for sev-
eral reasons. First, there is very little under-
standing of either the antecedents or the
consequences of AISs in this setting despite the
size of the health care sector and its importance
socially and economically (Evans, 1998). Second,
it is an ideal laboratory in which to examine
power. Power and politics are central to under-
standing the functioning of hospitals (Alexander
& Morlock, 2000; Succi, Lee, & Alexander, 1998;
Young & Saltman, 1985). As Mintzberg (1983)
and others (Alford, 1975; Freidson, 1975; Perrow,
1986) have argued, hospitals display the char-
acteristics that enable power to become the domi-
nant logic for behaviour. They have con?icting
goals, information for decision making is ambig-
uous, there is a complex division of labour, and
the cause and e?ect relation between actions and
outcomes are uncertain or unknown. Hospitals,
therefore, provide a setting in which to relax the
assumptions that dominate economic models of
behaviour (Harris, 1977).
Examining the model in this setting also enables
us to assess the e?cacy of some of the market-
based reforms currently being implemented in
hospitals. Traditionally hospital decision making
was dominated by the power and interests of phy-
sicians (Perrow, 1965). The power of physicians
within hospitals exacerbates goal con?ict and
potentially is problematic for implementing e?ec-
tive management control systems. This con?ict
arises when physicians pursue goals that increase
their status as a professional but which are not
congruent with achieving organizational goals that
are critical to maintaining the resource base of the
hospital (Abernethy & Stoelwinder, 1991). The
economic, political and social environment now
faced by hospitals is, however, changing the power
base of physicians. The increasing ?nancial, legal
and regulatory complexities associated with hos-
pital management have resulted in a shift away
from physician dominance in hospitals towards
professional management (Alexander & Morlock,
2000). This has been accompanied by a decrease in
the referent power traditionally bestowed on phy-
sicians by society (Freidson, 1975; Perrow, 1965).
Hospitals are increasingly moving towards
what Scott (1993) de?ned as a joint model where
1
It is interesting that Jensen is now integrating nonrational
aspects of human behaviour in a theory of organizational
behaviour (Jensen, 1998).
2
The political model is only one of a number of alternative
models that can be used to explain organizational behaviour
(see Luft & Shields, 2003, for a review of alternative models).
3
Much of the empirical research has focused on the decision
control role of AISs (Ittner & Larcker, 2003; Luft & Shields,
2003).
208 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
physicians and managers share formal authority
for the management of hospitals.
4
This study
enables us to assess the e?cacy of the organization
design adaptations implemented in order to for-
mally integrate physicians into hospital manage-
ment structures. There is, however, very little
evidence as to whether new authority structures or
the investment in new costing and budgeting sys-
tems are achieving their intended consequences
(Abernethy & Chua, 1996; Abernethy & Stoel-
winder, 1995; Jones & Dewing, 1997; Kurunma¨ ki,
1999). The desired outcome of these adaptations is
to create an organization that is not only respon-
sive to the demands of consumers but also one
that is ?nancially viable. Our study sheds some
light on whether these systems are achieving their
intended consequences.
The paper is structured as follows. The theore-
tical framework is developed in the following sec-
tion and is summarized in a set of research
questions. Method and results follow. The con-
cluding section discusses the results and the impli-
cations of these for further research.
Theoretical framework
Role of AIS
While the importance of AISs is rarely disputed
in the literature, considerable debate has emerged
concerning the role accounting plays in organiza-
tions.
5
According to orthodox theorists, account-
ing systems are implemented into organizations to
serve two functions: (a) to facilitate decision
making or what is often referred to as decision
management and (b) to control behaviour (Zim-
merman, 1997). Accounting, it is argued, serves
the decision management function by providing
information to reduce ex ante uncertainty. This in
turn enables decision makers to improve their
action choices with better informed e?ort (Kren,
1997). AISs support the formulation of strategy,
assist in strategy implementation, provide infor-
mation for co-ordination of organizational activ-
ities, and facilitate organizational learning
(Abernethy & Brownell, 1999; Bouwens & Aber-
nethy, 2000; Simons, 1995).
The importance of the control function of AISs
stems from the assumption that individuals do not
act in the organization’s best interests but rather
in their own. Control systems are thus imple-
mented by top management to increase the prob-
ability that individuals will behave in a manner
that will enable organization goals to be achieved
e?ciently and e?ectively (Flamholz et al., 1985).
AISs are purported to serve this purpose by pro-
viding information ex post about the action choi-
ces taken by subordinates. This information is
then used to measure and reward subordinates’
performance. The objective of such information is
to change subordinate behaviour or in?uence the
actions taken, so that organizational outcomes
can be e?ectively achieved.
We assess both roles of AISs by examining how
budget information is used. We examine use at the
subunit level. We assess the importance of the
decision management role to physicians who are
appointed to manage the day-to-day activities of
clinical units. We assess the decision control role
by examining how superiors use budget informa-
tion for controlling physician managers’ beha-
viour. We develop the model to include three
antecedent variables. Two of these relate to
authority structures: (1) the formal authority
structure that de?nes the decision rights of physi-
cian managers and (2) the informal authority
structure which is derived from the power and
in?uence of individual physicians operating within
the institutional setting. The third antecedent
relates to the design characteristics of the AIS.
Each is discussed in turn.
4
Some have argued that this reduces the power of physi-
cians as formal authority structures will curtail informal power
(Abernethy & Lillis, 2001), while others suggest that it may
increase their power (Alexander & Morlock, 2000). The extent
to which this occurs is not an issue directly addressed in this
paper. What is important here is that there is su?cient varia-
bility on our power measurement scale to explore the relations
of interest. We expect this to be the case.
5
This debate has drawn on the political and sociology lit-
erature and examined some of the more subversive roles of
accounting (see Chua, 1995; Miller & O’Leary, 1987; Preston,
1992). This paper does not address these roles. It adopts what is
often referred to as a functionalist positivist view of accounting.
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 209
Authority structures
We make a distinction between formal and
informal authority structures. The assignment of
decision rights represent formal authority in the
sense that this is a deliberate choice by top man-
agement to delegate particular types of decisions
to lower level management. The organization
chart generally captures formal authority. It
represents the o?cial system of accountability,
control and in?uence and is based on scalar prin-
ciples of authority. In other words, formal delega-
tion of decision rights is generally related to the
incumbent’s position in a ranked hierarchical
structure (Barnard, 1968). Furthermore, the roles
and responsibilities of the incumbent are de?ned
within this structural arrangement. The formal
authority structure, however, does not fully repre-
sent what transpires between the various actors
within the organization. We de?ne informal
authority as the ability of an individual or groups
of individuals to in?uence organization decisions
and activities in ways that are not sanctioned by
the formal authority system (Alexander & Mor-
lock, 2000). These decision rights are quite distinct
from formal assignment of decision rights from
superiors to subordinates. Informal decision rights
are ‘‘captured’’ by virtue of an individual’s (or
group of individuals’) expertise, where they stand
in the division of labour and/or their ability to
control the critical resources of a ?rm (Freidson,
1975; Pfe?er, 1992).
Formal authority structure and AISs
We examine the relation between the formal
authority structure and (a) the use of AIS for
controlling subordinates’ behaviour (the control
role) and (b) the use of AIS for facilitating deci-
sion making (the decision management role) at the
subunit level. Jensen and Meckling (1992) provide
a useful framework for linking formal authority
structure and the control role. They argue that
organizations have two problems—the decision
right assignment problem, and the control prob-
lem. The two are inter-related. Economic models
of behaviour assume that individuals can be
encouraged to achieve a common set of goals and
objectives. It assumes that if decision rights are
allocated to lower level managers, control systems
can be designed and implemented to encourage
employees to operate in the best interests of the
?rm (Zimmerman, 1997; Milgrom & Roberts,
1992; Jensen, 1998). AISs are one form of control.
These systems provide the measures by which
managers’ performance can be assessed, contracts
written and rewards determined. Much of the
economic-based literature has been focused on
modeling the optimal design characteristics of
performance measurement and/or incentive based
compensation system (Shields, 1997). While this
research has been primarily analytical, broad
based empirical research is beginning to emerge
(Ittner & Larcker, 2001). Empirical studies exam-
ining the control function of AIS are often based
on theories developed in the psychology and
organizational behavioural literatures (Luft &
Shields, 2003). This literature provides some evi-
dence that assignment of decision rights (the
behavioural literature tends to use the term
decentralization or autonomy) in?uences the use
of AISs for controlling behaviour at the subunit
level (Abernethy & Lillis, 2003; Wruck & Jensen,
1994). We, therefore, expect that there will be a
positive relation between the level of formal
authority delegated to physician managers in clin-
ical units and use of AISs by superiors for control
purposes.
There is little if any broad based empirical
research examining how formal authority struc-
tures in?uence the use of AISs for facilitating
decision management (Luft & Shields, 2003).
6
Early empirical studies on budgeting behaviour
provides evidence of how budgeting is used to
support the planning, co-ordination and manage-
ment role (Bruns & Waterhouse, 1975; Swieringa
& Moncur, 1975). Others have extended this
research and provided some evidence on factors
in?uencing the role of budgets for managing
activities at the subunit level (e.g. Macintosh &
Williams, 1992; Merchant, 1981). We add to this
rather limited set of empirical evidence, and
6
There are, however, numerous case studies that describe
how accounting facilitates decision making (see for example
Abernethy, Lillis, Brownell, & Carter, 2001; Dent, 1991;
Simons, 1991).
210 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
examine the relation between the level of formal
authority delegated to physician managers and
their use of AISs for managing and co-ordinating
activities at the subunit level. We expect to ?nd a
positive relation. Formal authority structures
de?ne roles and responsibilities and provide direct
signals as to what is important to the role incum-
bent. E?ective AIS are those that are designed to
reinforce or complement this role (Flamholz et al.,
1985; Abernethy & Lillis, 2001). Thus we would
expect that if managers are delegated authority to
manage subunit activities, the AIS will be designed
to support this role.
One of the major changes occurring in hospitals
is the creation of business units where physicians
are granted decision rights over both inputs and
outputs. These individuals are required to take on
managerial roles and are held accountable for the
?nancial management of their clinical units
(Abernethy & Lillis, 2001). Kurunma¨ ki (1999)
demonstrates how the introduction of new
accountability structures in public hospitals in
Finland in?uenced the use of budgeting and cost-
ing information by physicians. They became much
more concerned with monitoring and controlling
the costs of running their units. The Kurunma¨ ki
study, however, did not allow for di?erences in the
decision rights delegated to physicians. We are
interested in assessing the importance of this
structural feature on the use of AISs. We expect
that the use of accounting information for deci-
sion management will be related to the extent to
which physicians are delegated formal authority.
In other words, we expect to observe a direct
relation between the delegation of decision making
to physician managers and their use of AISs for
managing activities within clinical units.
Informal authority structures and Accounting
Information Systems (AISs)
The importance of informal authority, derived
through the power and in?uence of dominant
coalitions, has long been recognized in beha-
vioural theories of the ?rm (Cyert & March,
1963). Power, however, is not well operationalized
or researched in the organizational literature
(Alexander & Morlock, 2000). Pfe?er (1981) is one
of the few researchers who has attempted to
empirically assess the impact of power on organi-
zational functioning. While there are many models
of power, we de?ne power as the ability of an
individual to in?uence organization decisions and
activities in ways that are not sanctioned by the
formal authority of the system (Alexander &
Morlock, 2000; Kotter, 1985). We focus on the
power of physicians, as this is the group that has
traditionally been the dominant coalition in hos-
pitals. Their power stems from their ability to
control revenue generation and also because of
their claims to the specialized knowledge and
skills that are critical to the functioning of the
hospital. Physicians are central to the functioning
of the hospital due to their monopolistic control
over a specialized body of knowledge. The orga-
nization is dependent on the co-operation of
physicians and it is this dependence that has
enabled them to demand and achieve consider-
able power within hospitals (Pfe?er & Davis-
Blake, 1987).
Physician power is manifested in their control
over signi?cant resources without any formal
accountability for the use of those resources
(Abernethy & Lillis, 2001). Unlike formal author-
ity structures where decision rights are delegated
and individuals are held accountable for those
decision, physicians have been able to use their
power to in?uence decision making at all levels
within hospitals (Alexander & Morlock, 2000;
Weiner, Maxwell, Sapolsky, Dunn, & Hsiao, 1987;
Young & Saltman, 1985).
7
They use this informal
authority to bypass the authority systems imple-
mented by senior management. Furthermore, their
power has enabled them to avoid accountability
for the resource management of clinical units. We
are interested in exploring how power in?uences
the two roles of AIS, decision control and decision
management. The most direct e?ect of power is in
relation to the use of AIS by top management to
7
Physicians have been able to retain their power to a much
greater extent than other professionals associated with provid-
ing social services. Llewellyn (1999, p. 42) attributes this to
their success in de?ning the ‘‘true nature of their domain of
activities’’.
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 211
control the behaviour of physicians. This power
often grants physicians informal authority at the
strategic level of decision making enabling them to
‘‘end run’’ direct to the board and thus bypass any
attempts of senior management to use AISs to
control their behaviour (Young & Saltman, 1985).
Coupled with their control over core operating
activities they are in a position to either ignore or
sabotage AISs. Thus, we predict that the ability of
top management to use AISs to monitor and/or
evaluate physician behaviour will be inversely
related to the degree of power held by physicians.
When physicians become powerful and are able to
pursue their own goals rather than that of the
organization, there is evidence that this will lead to
a strong resistance to AISs implemented by top
management to control or curtail their behaviour
(Abernethy & Stoelwinder, 1995).
The relation between power and the use of AIS
for decision management is not nearly so clear.
There is some support for the argument that phy-
sicians with power will resist any attempts by top
management to implement administrative systems
(Abernethy & Stoelwinder, 1995). However, these
arguments relate to resistance to systems designed
for control purposes. There is no a priori theore-
tical or empirical rationale for predicting a simi-
larly negative relation between power and use of
AIS for decision management. Observations in the
?eld suggest that physicians with power may not
view the information provided by the accounting
system as relevant for decision making. This stems
from their reluctance to embrace the managerial
role. Their frame of reference relates to profes-
sional issues within the clinical unit rather than
resource management issues. It is the absence of a
managerial orientation that will in?uence their use
of AIS for decision management (Abernethy &
Stoelwinder, 1991). They simply do not see the
relevance of budget information for managing
clinical activities. If this is the case, then the level
of physician power may in itself have no in?uence
on how physicians use AIS for managing clinical
activities. At best, it will have a small negative
e?ect. As it is di?cult to predict the nature of the
relation between informal authority and the use of
AIS for decision management, we will allow the
empirics to shed light on this issue.
Design characteristics of the AIS
Our third antecedent variable captures the
design characteristics of the AISs. This variable
can be considered as a control variable in the
model.
8
Anthony (1965) recognized the impor-
tance of the design characteristics of AIS in his
seminal work. He distinguished between di?erent
dimensions of AIS, namely, whether the informa-
tion was ?nancial/non?nancial, internal/external
or historical/future orientated. Anthony’s frame-
work also described the di?erent criteria critical in
the design of AIS, namely, relevance, timeliness,
accuracy and the format of the information pre-
sented. The accounting literature has focused pri-
marily on the dimensions of AIS (e.g. Bowens &
Abernethy, 2000; Chenhall & Morris, 1986). Little
empirical attention has been devoted to the
importance of the design criteria. In contrast,
there is a signi?cant body of empirical research in
the information systems literature examining the
in?uence of these criteria on the use of and satis-
faction with information systems (e.g. Ang &
Koh, 1997; Doll & Torkzadeh, 1988; McHanney
& Cronan, 1998). We expect these design criteria
to be particularly important to physicians in the
use of this information for managing clinical
activities. To the extent that physicians perceive
that the information is relevant and accurate for
decision making they will use it for managing
clinical activities. In other words the use of
accounting information is dependent on physi-
cians’ perceptions of the design criteria associated
with the system. We also expect that the design
characteristics of the system will in?uence the
importance placed on AISs in controlling beha-
viour of physician managers by top management.
If the information relating to managerial actions is
too late, inaccurate, or does not capture the desired
set of behaviours, superiors are unlikely to rely on
this information for measuring subordinate
8
The focus of the paper is on assessing the impact of
authority structures on AIS. However, as the use of AIS is
likely to be in?uenced by the design characteristics of the AIS,
it is important to control for this variable. The analytical model
used in the study allows us to test for the direct e?ect of each
variable after partially out the e?ect of the other antecedent
variables.
212 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
behaviour (Milgrom & Roberts, 1992). Thus, we
expect there to be a positive relation between the
design characteristics of the information (in terms
of relevance, accuracy, format and timeliness) and
its use for control purposes.
Organizational outcomes
We assess the consequences of organizational
design choices by examining their e?ect on the
cost consciousness of physician managers. We
select cost consciousness as our outcome variable
for two reasons. First, increasing the cost con-
sciousness of physicians has been the major justi-
?cation for adaptations to internal management
structures and AISs in hospitals (Kurunma¨ ki,
1999). Second, much of the prior empirical litera-
ture that has attempted to assess the consequences
of organizational design choices on organizational
outcomes has used managerial or organizational
performance as the criterion variable (Ittner &
Larcker, 2001). Researchers have had great di?-
culty establishing empirical relations between
organizational design choices and performance
outcomes due to the rather tenuous theoretical
links among these variables (Briers & Hirst, 1990).
We avoid this problem by adopting an outcome
variable that is theoretically more closely linked to
our antecedent variables. We adopt the concept of
cost consciousness developed by Shields and
Young (1994). The construct does not attempt to
capture the trade-o? between ‘‘costs’’ and ‘‘car-
ing’’ described by Llewellyn (1998) but rather focu-
ses narrowly on the extent to which physician
managers are concerned with the cost consequences
of clinical decision making.
We are interested in assessing both the direct
relation between authority structures and cost
consciousness as well as the indirect relations that
occur via the AISs. We expect that the use of AIS
will increase the cost consciousness of physician
managers. The over-riding purpose of AISs is to
reinforce the importance of resource management
and draw attention to the cost consequences of
clinical decision making. We expect a direct and
positive relation between formal authority struc-
tures and cost consciousness but predict the
reverse e?ect with informal authority structures.
Assigning physicians formal decision rights over
inputs and outputs would by itself be expected to
in?uence their commitment to system goals asso-
ciated with e?ciency (Steers, 1977). These new
structural forms are designed to encourage physi-
cians to embrace resource management (Aber-
nethy & Stoelwinder, 1995). Informal authority,
gained through power, however, is likely to have a
negative e?ect on the cost consciousness of physi-
cians. Physicians have traditionally viewed the
hospital as a workshop and the maintenance of
their power simply allows them to use resources as
they see ?t with no concern for the cost con-
sequences of these decisions on the ?nancial via-
bility of the hospital (Weiner et al., 1987).
Summary and research questions
Our model is summarized in Fig. 1. The model
enables us to explore two forms of authority—
formal authority captured in terms of the decision
rights delegated by superiors to subordinates and
informal authority captured in terms of the power
of individuals within an organization. We explore
the impact of these two antecedent variables on
the use of AISs by physicians who manage clinical
units in hospitals. We also assess the importance
of the design characteristics of AISs on the use of
these systems for controlling physician behaviour
and facilitating decision making. Furthermore, we
assess if AISs have any impact on physician man-
agers’ cost consciousness. While our arguments
are primarily related to the relation between deci-
sion rights and power on AISs use we also exam-
ine if there is any direct impact of informal and
formal authority structures on cost consciousness.
Given that this study is primarily exploratory we
summarize our arguments in the form of ?ve
research questions:
1. What is the relation between the formal
assignment of decision rights and the use of
AISs for decision control and decision
management?
2. What is the relation between the informal
authority structure that operates via indi-
vidual power andthe use of AISs for decision
control and decision management?
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 213
3. What is the relation between the design
characteristics of AISs and the use of the
system for decision control and decision
management?
4. Is there any relation between the use of
AISs and physician managers’ cost con-
sciousness?
5. Is there any direct relation between the
formal authority structure and cost con-
sciousness? Between the informal authority
structure and cost consciousness?
The research study
Questionnaire data were collected from physi-
cian managers in large teaching hospitals in Italy.
To ensure that hospitals had similar characteristics
and faced similar political, economic and reg-
ulatory environment it was necessary to limit our
sample to the two large hospitals in one region in
Italy.
9
Both hospitals were university teaching
hospitals and had the same funding arrangements,
similar internal formal structures and accounting
information systems. There are several institu-
tional features worth noting. The Italian health
care system is primarily publicly funded through
compulsory national insurance. Hospitals are fun-
ded by the regional authority who has responsi-
bility for the provision of health services in each
region. A residual amount of funds come from
direct contributions by patients (e.g. pharmaceu-
ticals purchased by outpatients, laboratory service
costs and specialist services). It is the regional
authority’s responsibility to implement National
Health Plan policies and it is held directly
accountable for the e?ciency and e?ectiveness of
the health care delivery. It is thus the regional
authority who initiates changes in accounting
information systems and/or has the incentive to
Fig. 1. Summary of research questions, the empirical model.
9
One hospital had 950 beds and a budget of approximately
US$150 million and the other had 1800 beds and a total oper-
ating budget of US$290 million. We tested each of the relations
speci?ed in Fig. 1 to ensure that the results were not a?ected by
‘‘hospital’’. There was no evidence that this was the case.
214 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
encourage improvements in accountability struc-
tures within hospitals. Regional authorities are,
however, heavily regulated by the State although
there is some ?exibility in the regulations to
‘‘experiment’’ with new managerial structures and
accounting systems. Of particular interest in our
study is the hospital/physician employment
arrangement. Physicians in university hospitals are
appointed both by the hospital and by the uni-
versity and their employment arrangement is gov-
erned by regulation. There are penalties imposed
on physicians who opt for part-time rather than
full-time employment status. Only physicians who
are full-time employees can be appointed to man-
agerial positions. Our study is based on this group
of physicians.
We collected archival and interview data. The
archival data included annual reports and regula-
tions in?uencing the industrial, ?nancing, report-
ing and administrative arrangements of hospitals.
As well as using the archival data to gain an
appreciation of the institutional arrangements
associated with the hospitals, we also conducted
interviews with a variety of stakeholders asso-
ciated with hospitals in the region. This included
interviews with key informants at the regional
o?ce, the local health authority, as well as the
General Director, Medical Director, Adminis-
trative Director, Nursing Director and physicians
working in one of the research sites. These inter-
views were audio recorded and transcribed verba-
tim. The interview and archival data facilitated the
development of the questionnaire. It also ensured
that we captured the appropriate set of respon-
dents, i.e. that the respondents were physicians
who were responsible for the management of
clinical units and that the hospitals routinely
distributed budget reports.
10
The O?ce of the
General Director at each hospital provided the
names of physicians with managerial responsi-
bilities. A total of 135 questionnaires were dis-
tributed with a letter from the GD asking
physicians for their support in the research study.
Con?dentiality was guaranteed and respondents
were asked to return the questionnaire directly to
the researchers.
The purpose of the questionnaire was to assess
both the antecedent conditions in?uencing the use
of AISs and the consequences in terms of cost
consciousness. Measurement instruments were
based on those used in prior studies. The ques-
tionnaire was administered in Italian and we fol-
lowed the back-translation procedure (Behling &
Law, 2000).
11
We used multi-items to capture each
construct and employed seven-point Likert-type
scales. Some of the items were reverse-coded to
minimize the potential for acquiescence error.
12
These items were recoded prior to the creation of the
measurement scale. The measures for the variables of
interest in this study are described in turn. The
instruments used in the study are in the Appendix.
Measurement of variables
Formal authority
Athree-itemmeasure based on the Govindarajan
(1988) instrument was used to capture the decision
rights delegated to physician managers. The
instrument focuses on the delegation of decisions
relating to inputs and outputs and required physi-
cians to indicate the extent to which they had been
delegated these types of decisions. The three items
were summed for use in the analysis. Factor analy-
sis indicated that the scale was unidimensional and
the Cronbach (1951) a coe?cient of 0.70 provides
support for the use of the summed measure.
Informal authority
The power of physicians in hospitals is best repre-
sented by capturing their in?uence over strategic
11
One of the authors was a native Italian speaker but also
spoke English ?uently. She translated the instrument from
English into Italian. The other author was a native English
speaker and familiar with Italian. She back-translated the
instrument. A third bilingual translator, not associated with the
project, independently translated the instrument from English
into Italian. This was also back-translated. Di?erences in the
two translations were compared until a consensus was
achieved.
12
There was no evidence that the use of the reverse-coding
in?uenced the reliability of the instrument. Assessment of the
factor structure for each scale did not indicate that the reverse
coded items were a problem for respondents.
10
Our ?eld visits also enabled us to assess the formal
authority structures existing in the hospitals. This was impor-
tant for the measurement of informal authority.
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 215
level decisions. It is at this level where physicians
often exert their power and in?uence without any
formal sanctions to do so (Young & Saltman,
1985). It is also important to note that no physi-
cian manager in our sample had formally been
delegated decision rights at this level. We capture
informal authority by adapting the instrument
developed by Succi et al. (1998). This instrument
was designed to capture the relative power of
physicians and managers. It included physician
in?uence on strategic priorities as well as in?uence
over the use of important strategic resources, e.g.
investment in technology, determining clinical
privileges within the hospital, allocation of clinical
sessions across clinical programs, macro-level
decisions concerning clinical practices and poli-
cies. These decisions represent strategic level deci-
sion making as they a?ect the operation of all
clinical programs. They are generally taken at
board level. We measured power by asking physi-
cian managers to indicate their in?uence over
seven areas of strategic in?uence. Factor analysis
indicated that two of the items (items e and g)
loaded on a di?erent factor. We, therefore, exclu-
ded these two items. The Cronbach a coe?cient
for the remaining ?ve items was 0.84. We summed
these ?ve items to form the scale.
Design characteristics of the AIS
There are few studies in the accounting litera-
ture examining the design characteristics of AISs.
We were interested in assessing physician man-
ager’s perceived satisfaction with the accuracy,
relevance, format and timeliness of the budget
information provided to them on a routine basis.
The information systems literature has devoted
considerable e?ort in developing measures of
information satisfaction. We adapted an instru-
ment developed by Doll and Torkzadeh (1988).
The instrument has been used repeatedly in the
literature and there is considerable support for its
psychometric properties (Doll, Xia, & Torkzadeh,
1994; McHanney & Cronan, 1998). The adapted
instrument included eight separate items capturing
the content, accuracy, format and timeliness of the
budgeting information provided to physician
managers. The instrument also included one over-
all question that asked managers to indicate their
satisfaction with the information provided by the
budgeting system. A factor analysis of the instru-
ment indicated that there was one factor. The
Cronbach a coe?cient of 0.95 provides support
for its use as a uni-dimensional scale. In addition,
we used regression analysis to assess the extent to
which the eight separate items explained the var-
iance in the overall measure. The results (not pro-
vided here) indicated that 0.89% of the variance in
the overall measure was captured by the eight
separate dimensions. We were, therefore, satis?ed
that the eight item summed measure captured
overall level of satisfaction.
Use of budget information
We adapted the budget-related behaviour
instrument developed originally by Swieringa and
Moncur (1975) to measure the extent to which
budgeting information is used for decision man-
agement and decision control. The instrument has
been used repeatedly in the literature and its psy-
chometric properties are well documented (see
Abernethy & Stoelwinder, 1991, 1995; Macintosh
& Daft, 1987; Merchant, 1981). We used four
items to capture the use of budgeting information
by physicians for managing clinical unit activities
(i.e. the decision management role). Factor analy-
sis provided support for the four-item measure
and the Cronbach a statistic (0.68) supports the
use of an additive scale (Van de Ven & Ferry,
1980).
We captured the decision control role using four
items. These items related to the extent to which
physician managers were required to report bud-
get variance information to their superior and the
extent to which the information was used to eval-
uate the performance of the clinical unit. Factor
analysis revealed that the four items represented
one construct and the a coe?cient (0.60) provided
reasonable support for the use of the summed
measure in the analysis.
Cost consciousness
We used the instrument developed by Shields
and Young (1994). The instrument includes six
items relating to cost conscious behaviour and one
overall item designed to capture cost conscious-
ness. Our factor analysis of the six items indicated
216 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
that it was a uni-dimensional scale. Reliability of
the measure was 0.86 and regression analysis indi-
cated that the six items explained more than 55%
of the variance in the overall measure. Based on
these results, the six items were summed for use in
the analysis.
Analytical method and results
A total of 70 questionnaires were returned pro-
viding an overall response rate of 52%. There
were, however, only 56 useable questionnaires.
The physician managers in our sample had been in
their current position for an average of 10 years
and had practiced as a medical practitioner in the
hospital for an average of 24 years. We tested for
response bias in our sample by assessing if there
were any di?erences in the mean responses for
each of the variables of interest between the early
respondents and the late respondents.
13
There was
no evidence of response bias at conventional levels
of signi?cance. Table 1 provides the descriptive
statistics (mean and standard deviation) for the
sample and the correlations among the variables.
To allow for comparability between the variables,
the mean value for each multi-item scale repre-
sents the average score (i.e. the multi-item scale
was divided by number of items in the scale).
The relations to be explored in this study are
summarized in Fig. 1. We tested these relations
using structural equation modelling techniques.
We used LISREL to estimate the standardized
path coe?cients, the associated standard errors
and to provide an assessment of the ?t of the
model to the sample data. We undertook the ana-
lysis in a series of steps. The ?rst step was to test
the model as speci?ed in Fig. 1. We then examined
the diagnostics (namely the modi?cation indices)
to determine if the ?t of the model could be
improved. These diagnostics suggested that we
should allow the error terms associated with the
two roles of accounting to co-vary. While the use
of modi?cation indices to adjust the model should
only occur based on theoretical grounds, it does
seem reasonable that these two terms would, over
time, be correlated. We thus adjusted the model
accordingly. The ?t statistics indicated that our
data ?t the model well (w
2
=0.157, P=0.692,
df=1, AGFI=0.980, NFI=0.997). While these ?t
statistics more than meet the cut-o? criteria
necessary for a good ?t (0.90 and 0.80, respec-
tively), we adopted an approach widely accepted
in the general management literature of using nes-
ted models to establish the most parsimonious
Table 1
Descriptive statistics and Pearson correlations
Mean
(S.D.)
Pearson correlations (signi?cance levels)
Formal Design
characteristics
Informal Decision
control
Decision
management
Cost
conscious
Formal 5.48 1.00 0.39 À.11 0.37 0.31 0.13
(1.13) (0.00) (ns) (0.00) (0.02) (0.34)
Design 4.62 1.000 À0.05 0.20 0.32 0.07
Characteristics (1.64) (ns) (ns) (0.01) (ns)
Informal 2.24 1.00 À0.12 0.09 À0.26
(1.18) (ns) (ns) (0.05)
Decision control 4.43 1.00 0.55 0.219
(1.17) (0.00) (0.10)
Decision
management
3.75 1.00 0.31
(1.18) (0.02)
Cost conscious 5.22 1.00
(1.26)
13
We do this comparison on the assumption that late
respondents have similar characteristics to the non-respondents.
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 217
model (see Anderson & Gerbing, 1988; Medsker,
Williams, & Holahan, 1994; Wayne, Shore, &
Linden, 1997).
14
Starting with the model in Fig. 1,
we assessed the series of nested models through
sequential chi-square di?erence tests (not pre-
sented here) until we were satis?ed that we had
achieved the most parsimonious model.
15
This is
depicted in Fig. 2 with the signi?cant path coe?-
cients noted. All of the measures normally
employed to test the ‘‘?t’’ of structural equation
models indicate that the modi?ed model ?ts the
data very well. The a
2
value is insigni?cant (0.714,
P=0.98). The adjusted goodness-of-?t-index
(0.98) and the normed ?t index (0.961) more than
meet the cut-o? criteria necessary for a good ?t
(0.90 and 0.80, respectively).
The results of our model support our expecta-
tions. There is a signi?cant relation between for-
mal authority and the use of the AIS for decision
management and decision control. The formal
allocation of decision rights to physicians results
in a signi?cantly greater use of AISs by physicians
in the management of clinical activities (0.24,
P=0.06). In addition, there is an increased use of
the system by superiors for monitoring and mea-
suring physician behaviour (0.37, P=0.00) when
decision rights are delegated to physician man-
agers. There is also some support for the idea that
the use of AISs for decision management increases
when the relevance of the system for decision mak-
ing increases. There is, however, no signi?cant rela-
tion between the design characteristics of the system
and its use for controlling physician behaviour.
The relation between informal authority and use
of AIS is not signi?cant. Recall that we expected a
negative relation between informal authority and
the decision control role of AIS but had no theo-
retical priors concerning the relation with the
decision management role. Our results relating to
the relation between the role of AIS and cost con-
sciousness supports our expectations. Informal
power has a negative impact on cost consciousness
(À0.29, P=0.02). In contrast, we ?nd a very
strong and positive relation (0.33, P=0.01)
between the use of AIS for decision management
and the cost conscious behaviour of physicians.
There is no direct relation between formal
authority and cost consciousness. It appears that
the e?ect of formal authority on cost conscious-
ness is an indirect one via the use of AIS in
managing clinical activities.
Fig. 2. Results of model, standardized path coe?cients, (non-signi?cant paths not shown).
14
See Abernethy and Lillis (2001).
15
A similar method was used by Abernethy and Lillis (2001).
218 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
Discussion and concluding comments
This is an exploratory study designed to assess
how authority structures in hospitals in?uence the
use of accounting information by physicians
appointed to manage clinical units. We examine
the relative importance of formal authority dele-
gated by senior management to physician man-
agers and informal authority derived from power
and in?uence that physicians hold within the hos-
pital. Understanding the consequences of these
two forms of authority is important for several
reasons. First, hospitals are currently attempting
to integrate physicians into formal management
structures in an e?ort to reduce the power that
physicians have traditionally held over hospital
decision making (Abernethy & Chua, 1996). For-
mal integration of physicians is argued to increase
their commitment to resource management. The
power of dominant physicians in hospitals, how-
ever, has been argued to seriously limit attempts
by hospitals to implement strategies directed
towards improved resource management by phy-
sicians. While there has been much anecdotal evi-
dence on the adverse consequences of physician
dominance on hospital outcomes, this is the ?rst
study of which we are aware that provides broad
based evidence of unsanctioned physician author-
ity, particularly the impact on attitudes to
resource management. This study provides evi-
dence on the direct e?ect of adaptations to formal
authority structures and physician power on the
cost consciousness behaviour of physicians. Sec-
ond, we shed light on the role played by AISs in
this setting. Changes in internal management
structures in hospitals have been accompanied by
signi?cant investment in the implementation of
new and improved accounting information sys-
tems. The e?cacy of these developments, however,
will depend on their e?ect on physician behaviour.
Our ?ndings indicate that the formal delegation
of authority to physicians has a direct impact on
the use of accounting information for decision
control and decision management. What is parti-
cularly important is the e?ect of AIS on the cost
consciousness of physicians. It would appear that
the choice to implement a structure that formally
integrates physicians into the management
structures has a positive e?ect on cost conscious-
ness but this operates via the AISs. The use of AIS
for managing clinical activities provides a means
of reinforcing the formal delegation of authority.
It compliments the formal authority structure by
articulating the value set associated with a com-
mitment to resource management (Comerford &
Abernethy, 1999). This is particularly important in
this setting as physician managers experience con-
siderable role con?ict between their professional
goal set and the goals and values associated with
their managerial role (Abernethy & Stoelwinder,
1995). Accounting systems de?ne ?nancial
responsibilities and thus can serve to reduce the
role ambiguity associated with the managerial role
(Chenhall & Brownell, 1988). It is interesting to
note that a similar e?ect did not occur with respect
to the decision control role of the AIS. The dele-
gation of authority to physicians in?uenced the
use of AISs for control purposes but this did not
in?uence the cost consciousness of physicians. Our
experience in the ?eld suggests that physicians do
not pay much attention to the control role of AISs
as the information does not fully re?ect their
performance or the performance of the clinical
unit. If this is the case it will not impact on cost
consciousness.
As expected, informal authority of physicians is
not a signi?cant antecedent to the use of AIS. It
was, however, very signi?cant in explaining the
cost conscious behaviour of physicians. It would
appear that the consequences for hospitals are
signi?cantly and adversely a?ected by physician
power. The higher the level of power of physicians
the less they are likely to be committed to using
resources e?ciently. This supports much of the
anecdotal evidence concerning physician beha-
viour and the impact of this behaviour on the
?nancial viability of hospitals (Shortell & Conrad,
1996). And lastly, our ?ndings support the impor-
tance of designing AIS that are relevant for the
users of the system. This is particularly important
for the physicians who use these systems for
managing activities within their clinical units.
As with all exploratory research the study has
some potential limitations. First, we examine a
relatively simple model to enable us to develop a
parsimonious model and assess the impact of
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 219
authority structures as antecedents to the use of
AISs. We also focussed narrowly on the functional
role of AISs. However, we recognize that AISs can
be used to serve other purposes. Anecdotal data
collected through interviews with physicians illu-
strated how physicians use accounting informa-
tion for legitimizing and rationalizing decision
making (see Burchell et al., 1980; Covaleski &
Dirsmith, 1986). This was particularly the case
with physicians with strong political connections
outside the hospital. They use the information as
‘‘ammunition’’ to obtain additional resources not
only from the hospital’s budget but also to gain
funds from other external constituents (e.g. the
regional authority, the university hospital). The
complexities associated with power and alternative
roles of accounting are best examined using in-
depth case studies. For example, Abernethy and
Chua (1996) use a longitudinal case study in one
large public teaching hospital to demonstrate how
‘‘new’’ accounting innovations are in?uenced by
shifts in power within a hospital as well as changes
in societal values and norms. Accounting systems
become an instrument to serve di?erent purposes
for various stakeholders in the organization.
While our interview data drew our attention to
these complexities, this paper is unable to fully
explore them. Further research adopting ?eld-
based methodology would provide an in-depth
understanding of the complexities associated with
power and accounting information systems.
Our measurement instruments require further
testing. We drew on prior literature to develop our
measurement instruments, however, further
research is required to provide support for the
psychometric properties of these instruments. Our
study used data collected from Italian hospitals.
While this may limit the generalizability of our
?ndings there is no reason to believe that this
particular setting in?uences the ?ndings. Further-
more, as much of our literature is dominated by
data collected primarily from English speaking
countries, broadening our study of AISs to other
institutional environments will further enhance
our understanding of these systems in a more glo-
bal setting. Further research could be directed
towards testing this model in di?erent institutional
environments. And lastly, caution is required in
the interpretation of the results. It is not possible
to infer causality among our variables at test as
our data are collected contemporaneously and any
implied causality must stem from the theoretical
position taken (Cook & Campbell, 1979).
Despite these potential limitations, this study is
the ?rst to provide empirical evidence of the con-
sequences on physician behaviour due to the dual
in?uences of formal and informal authority. The
?ndings demonstrate the importance of imple-
menting new accountability arrangements and
AISs designed to encourage physicians to become
e?ectively integrated in hospital management
structures. It is only when AISs are designed and
implemented to support physician managers that
it is possible to create a culture in hospitals where
the major stakeholders are committed to provid-
ing good quality care while at the same time
maintaining ?nancial viability.
Acknowledgements
We wish to acknowledge the funding provided
by The Faculty of Economics and Commerce, The
University of Melbourne, and the Facolta` di
Economia, Universita` di Ferrara. We also extend
our appreciation to all of the physicians who par-
ticipated in the study and to Dott. Enrico Bracci
for his research assistance. The paper has also
received constructive comments from participants
at the Department of Accounting, University of
Melbourne Seminar, University of Nyenrode, the
EIASM Workshop on Performance Measurement
and Management Control and those from Jennifer
Grafton, Jan Bouwens, Frank Selto and the two
anonymous referees.
220 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
Appendix. Measurement instruments
Formal Authority
Strongly
Disagree
Strongly
Agree
–
a. I am held responsible for the costs incurred in my unit. 1 2 3 4 5 6 7
b. I am responsible for managing throughput in my unit. 1 2 3 4 5 6 7
c. My contract with the hospital holds me accountable
for achieving my budget targets and also for achieving
output targets.
1 2 3 4 5 6 7
Informal Authority
Indicate the extent to which you have in?uence over the
following types of decisions within the hospital.
To a great
extent
To a little
extent
a. Adding or expanding a clinical service within the
hospital.
1 2 3 4 5 6 7
b. The strategic priorities of the hospital. 1 2 3 4 5 6 7
c. Determining of doctor’s clinical privileges within the
hospital.
1 2 3 4 5 6 7
d. Decisions relating to the allocation of beds in clinical
units throughout the hospital
1 2 3 4 5 6 7
e. Purchase of major pieces of medical equipment in the
hospital.
1 2 3 4 5 6 7
f. Appointment of new medical sta? in the hospital. 1 2 3 4 5 6 7
g. Clinical policies and practices throughout the hospital. 1 2 3 4 5 6 7
Design Criteria of Accounting Information System
Almost
never
Almost
always
a. Do your budget reports provide you with the precise
information you need?
1 2 3 4 5 6 7
b. Does the information content of these reports meet
your needs?
1 2 3 4 5 6 7
c. Do the reports provide su?cient information? 1 2 3 4 5 6 7
d. Is the information received accurate? 1 2 3 4 5 6 7
e. Are you satis?ed with the accuracy of the information
in the budget reports?
1 2 3 4 5 6 7
f. Do you think the budget reports are presented in a
useful format?
1 2 3 4 5 6 7
g. Is the information clear? 1 2 3 4 5 6 7
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 221
Almost
never
Almost
always
h. Do you get the information you need in time? 1 2 3 4 5 6 7
j. Overall, how would you rate your satisfaction with
the information provided by the budgeting system?
1 2 3 4 5 6 7
Use of AIS
1. Decision management role To a great
extent
To a little
extent
a. To what extent do you investigate items which
are ‘‘overspent’’ in the budget?
1 2 3 4 5 6 7
b. To what extent do you stop activities when
budget funds are used up?
1 2 3 4 5 6 7
c. To what extent do you trace the cause of
budget variances to groups or individuals within
the unit?
1 2 3 4 5 6 7
d. To what extent does the budget enable you to be a
better manager of the unit?
1 2 3 4 5 6 7
2. Decision control rule
To a great
extent
To a little
extent
a. To what extent is meeting the budget for your
unit of great importance to the person to whom
who are responsible?
1 2 3 4 5 6 7
b. To what extent are you evaluated on budget
performance?
1 2 3 4 5 6 7
c. To what extent are you help personally accountable
for budget variances occurring in your unit?
1 2 3 4 5 6 7
d. To what extent are you required to report actions
taken to correct causes of large budget variances?.
1 2 3 4 5 6 7
Cost Consciousness
Please indicate the extent of your agreement with
the following statements.
Strongly
Agree
Strongly
Disagree
a. In general, I know how much I have to spend
in operating my unit.
1 2 3 4 5 6 7
b. I have good knowledge of the way my unit’s
budget is spent.
1 2 3 4 5 6 7
c. I make sure those who work in my unit know
the spending goals and limits.
1 2 3 4 5 6 7
222 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
References
Abernethy, M. A., & Brownell, P. (1999). The role of account-
ing in organizations facing strategic change: an empirical
analysis in hospitals. Accounting Organizations and Society,
24, 1–23.
Abernethy, M. A., & Chua, W. F. (1996). A ?eld study of
control system ‘‘redesign’’: the impact of institutional pro-
cesses on strategic choice. Contemporary Accounting
Research, 13, 596–606.
Abernethy, M. A. & Lillis, A. M. (2001). Interdependencies in
organization design: a test in hospitals. Journal of Manage-
ment Accounting Research, 13, 107–130.
Abernethy, M. A., Lillis, A. M., Brownell, P., & Carter, C.
(2001). Product diversity and costing system design choice:
?eld study evidence. Management Accounting Research,
12(3), 261–279.
Abernethy, M. A., & Stoelwinder, J. U. (1991). Budget use,
task uncertainty, system goal orientation and subunit per-
formance: a test of the ‘‘?t’’ hypothesis in not-for-pro?t
hospitals. Accounting, Organizations and Society, 16(2), 105–
120.
Abernethy, M. A., & Stoelwinder, J. U. (1995). The role of
professional control in the management of complex organi-
zations. Accounting, Organizations and Society, 20(1), 1–17.
Alexander, J. A., & Morlock, L. L. (2000). Power and politics in
health services organizations, in health care management:
organization design and behavior (4th ed.). Albany, NY:
Thomson Learning.
Alford, R. R. (1975). Healthcare politics. Chicago and London:
The University of Chicago Press.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation
modeling in practice: a review and recommended two-step
approach. Psychological Bulletin, 103, 411–423.
Ang, J., & Koh, S. (1997). Exploring the relationships between
unser satisfaction and job satisfaction. International Journal
of Information Management, 17(3), 169–177.
Anthony, R. A. (1965). Planning and control system: a frame-
work for analysis. Boston: Harvard University Press.
Barnard, C. I. (1968). The functions of the executive (30th
anniversary edition). Cambridge: Harvard University Press.
Behling, O., & Law, K. S. (2000). Translating questionnaires
and other research instruments: problems and solutions.
Thousand Oaks: Sage Publications.
Bouwens, J., & Abernethy, M. A. (2000). The consequences of
customization of management accounting system design.
Accounting, Organizations and Society, 24(2), 221–241.
Briers, M., & Hirst, M. (1990). The role of budgetary informa-
tion in performance evaluation. Accounting, Organizations
and Society, 15, 373–398.
Bruns, W. J., & Waterhouse, J. H. (1975). Budget control and
organization structure. Journal of Accounting Research,
13(2), 177–203.
Burchell, S., Clubb, C., Hopwood, A., & Hughes, A. (1980).
The roles of accounting in organizations and society.
Accounting Organizations and Society, 5, 5–27.
Chenhall, R. H., & Morris, D. (1986). The impact of structure,
environment and interdependence on the perceived useful-
ness of management. The Accounting Review, 61(2), 263–272.
Chow, C. W., Ganulin, D., Haddad, K., & Williamson, J.
(1998). The balanced scorecard: a potential tool for energiz-
ing and focusing healthcare organizational management.
Journal of Healthcare Management, 43(3), 263–280.
Chua, W. F. (1995). Experts, networks and inscriptions in the
fabrication of accounting images: a story of the representa-
tion of three public hospitals. Accounting, Organizations and
Society, 20, 111–146.
Comerford, S., & Abernethy, M. A. (1999). Budgeting and the
management of role con?ict in hospitals. Behavioral Research
in Accounting, 11, 93–110.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation:
design and analysis issues for ?eld settings. Chicago: Rand
McNally College Publishing Company.
Covaleski, M. A., & Dirsmith, M. W. (1986). The budgetary
process of power and politics. Accounting, Organizations and
Society, 11(3), 193–214.
Cronbach, L. J. (1951). Coe?cient alpha and the internal
structure of tests. Psychometrika, 16(3), 297–334.
Almost
never
Almost
always
d. I am very con?dent of my ability to manage
costs in this unit.
1 2 3 4 5 6 7
e. I put a lot of e?ort into reducing costs. 1 2 3 4 5 6 7
f. When I decide to purchase new supplies or
equipment I focus heavily on how much it
costs.
1 2 3 4 5 6 7
g. I am very conscious of how actions in this unit
in?uence overall hospital costs.
1 2 3 4 5 6 7
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 223
Cyert, R. M., & March, J. G. (1963). A behavioral theory of the
?rm. Englewood Cli?s, NY: Prentice-Hall.
Dent, J. F. (1991). Accounting and organizational cultures: a
?eld study of the emergence of a new organizational reality.
Accounting, Organizations and Society, 16(8), 705–732.
Doll, W. J., & Torkzadeh, G. (1988). The measurement of end-
user computing satisfaction. MIS Quarterley, 12(2), 259–274.
Doll, W. J., Xia, W. D., & Torkzadeh, G. (1994). A con-
?rmatory factor-analysis of end-user computing satisfaction
instrument. MIS Quarterly, 18(4), 453–461.
Eisenhardt, K. M., & Bourgeois, L. J. (1988). Politics of strategic
decision making in high-velocity environments: toward a mid-
range theory. Academy of Management Journal, 31(4), 737–770.
Evans, J. H. (1998). Cost management and management con-
trol in healthcare organizations: research opportunities.
Behavioral Research in Accounting, 10, 78–93.
Flamholz, E., Das, T. D., & Tsui, A. S. (1985). Toward an
integrative framework of organizational control. Accounting,
Organizations and Society, 10(1), 35–50.
Freidson, E. (1975). Doctoring together: a study of professional
control. Chicago: University of Chicago Press.
Govindarajan, V. (1988). A contingency approach to strategy
implementation at the business-unit level: integrating
administrative mechanisms with strategy. Academy of
Management Journal, 31(4), 828–853.
Hardy, C., & Clegg, S. R. (1999). Some dare call it power.
In S. R. Clegg, & C. Hardy (Eds.), Studying organization.
London: Sage Publications.
Harris, J. E. (1977). The internal organization of hospitals:
some economic implications. The Bell Journal of Economics,
8(2), 467–482.
Ittner, C. D., & Larcker, D. F. (2001). Assessing empirical
research in managerial accounting: a value-based manage-
ment perspective. Journal of Accounting and Economics, 32,
349–410.
Jensen, M. C. (1998). Foundations of organizational strategy.
Cambridge, MA: Harvard University Press.
Jensen, M. C., & Meckling, W. H. (1992). Speci?c and general
knowledge and organizational structure. In L. Werin, &
H. Wijkander (Eds.), Contract economics. Cambridge:
Blackwell Publishers.
Jones, C. S., & Dewing, I. P. (1997). The attitudes of NHS
clinicians and medical managers towards changes in
accounting controls. Financial Accountability and Manage-
ment, 13(3), 261–280.
Kotter, J. P. (1985). Power and in?uence: beyond formal
authority. New York: Free Press.
Kren, L. (1997). The role of accounting information in organi-
zational control: the state of the art. In V. Arnold, &
S. G. Sutton (Eds.), Behavioral accounting research, founda-
tions and frontiers. American Accounting Association: USA.
Kurunma¨ ki, L. (1999). Professional vs ?nancial capital in the ?eld
of health care—struggles for the redistribution of power and
control. Accounting Organizations and Society, 24(2), 95–124.
Lee, S. Y. D., & Alexander, J. A. (1998). Using CEO succession
to integrate acquired organizations: a contingency analysis.
British Journal of Management, 9(3), 181–197.
Llewellyn, S. (1998). Boundary work: costing and caring in the
social services. Accounting, Organizations and Society, 23(1),
23–48.
Luft, J., & Shields, M. (2003). Mapping management account-
ing: making structural models from theory-based empirical
research. Accounting, Organizations and Society, 28, 169–250.
Macintosh, N. B., & Daft, R. L. (1987). Management control
systems and departmental interdependencies: an empirical
study. Accounting, Organizations and Society, 12(1), 49–61.
Macintosh, N. B., & Williams, J. J. (1992). Managerial roles
and budgeting behaviour. Rehavorial Research in Accounting,
4, 23–48.
McHanney, R., & Cronan, T. P. (1998). Computer simulation
success: on the use of the end-user satisfaction instrument.
Decision Sciences, 29(2), 525–536.
Medsker, G. J., Williams, L. J., & Holahan, P. J. (1994). A
review of current practices for evaluating causal models in
organizational behaviour and human resources management
research. Journal of Management, 439–464.
Merchant, K. A. (1981). The design of the corporate budgeting
system: in?uences on managerial behaviour and perfor-
mance. The Accounting Review, 56, 813–829.
Milgrom, P., & Roberts, J. (1992). Economics, organization and
management. Englewood Cli?s, NJ: Prentice-Hall International.
Miller, P., & O’Leary, E. (1987). Accounting and the construc-
tion of the governable person. Accounting, Organizations and
Society, 12, 235–265.
Mintzberg, H. (1983). Power in and around organizations.
Englewood Cli?s, NJ: Prentice-Hall.
Perrow, C. (1965). Hospitals: technology, structure and goals.
In J. G. March (Ed.), Handbook of organizations. Chicago:
Rand McNally.
Perrow, C. (1986). Complex organizations (3rd ed). New York:
Random House.
Pfe?er, J. (1981). Power in organizations. Marsh?eld, MA:
Pitman Publishing.
Pfe?er, J. (1992). Managing with power: politics and in?uence in
organizations. Boston: Harvard Business Press.
Pfe?er, J., & Davis-Blake, A. (1987). Understanding organiza-
tional wage structures: a resource dependence approach.
Academy of Management Journal, 30, 437–455.
Preston, A. (1992). The birth of clinical costing: a study of the
emergence and transformations of discourses on costs and
practices of accounting in US hospitals. Accounting, Organi-
zations and Society, 17(1), 63–100.
Scott, W. R. (1993). The organization of medical care services:
towards an integrated theoretical model. Medical Care
Review, 271–303.
Shields, M. D. (1997). Research in management accounting by
North Americans in the 1990s. Journal of Management
Accounting Research, 9, 3–61.
Shields, M., &Young, S. M. (1994). Managing innovation costs:
a study of cost conscious behavior by R&D Professionals.
Journal of Management Accounting Research, 6, 175–196.
Shortell, S. M., & Conrad, D. A. (1996). Integrated health sys-
tems: promise and performance. Frontiers of health services
management, 13(1), 3–40.
224 M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225
Simons, R. (1991). Strategic orientation and top management
attention to control systems. Strategic Management Journal,
12, 49–62.
Simons, R. (1995). Levers of control: how managers use innova-
tive control systems to drive strategic renewal. Boston:
Harvard Business School Press.
Steers, R. M. (1977). Organizational e?ectiveness: a behavioral
view. Santa Monica: Goodyear.
Succi, M. J., Lee, S. D., & Alexander, J. A. (1998). Trust
between managers and physicians in community hospitals:
the e?ects of power over hospital decisions. Journal of
Healthcare Management, 43(5), 397–415.
Swieringa, R. J., & Moncur, R. H. (1975). The relationship
between managers’ budget related behavior and selected
attitude, position, size and performance measures. Empirical
Research in Accounting: Selected Studies. Supplement to
Journal of Accounting Review, 10, 194–205.
VanDe Ven, A. H., & Ferry, D. L. (1980). Measuring
and assessing organizations. New York: John Wiley and
Sons.
Wayne, S. J., Shore, L. M., & Linden, R. C. (1997). Perceived
organizational support and leader-member exchange: a
social exchange perspective. Academy of Management
Journal, 40, 82–111.
Weiner, S. l., Maxwell, J. H., Sapolsky, H. M., Dunn, D. L., &
Hsiao, W. C. (1987). Economic incentives and organizational
realities: managing hospitals under DRGs. Milbank Memor-
ial Fund Quarterly, 65(4), 463–487.
Wruck, K., & Jensen, M. C. (1994). Science, speci?c knowl-
edge, and total quality management. Journal of Accounting
and Economics, 18(3), 247–287.
Young, D. W., & Saltman, R. B. (1985). The hospital power
equilibrium, physician behavior and cost control. Baltimore:
Johns Hopkins University.
Zimmerman, J. (1997). Accounting for decision making and
control. Boston: Irwin/McGraw Hill.
M.A. Abernethy, E. Vagnoni / Accounting, Organizations and Society 29 (2004) 207–225 225

doc_179282502.pdf