European Accounting Review

ISSN: 0963-8180 (Print) 1468-4497 (Online) Journal homepage: https://www.tandfonline.com/loi/rear20

In Search of Management Accounting Theory

Teemu Malmi & Markus Granlund

To cite this article: Teemu Malmi & Markus Granlund (2009) In Search of
Management Accounting Theory, European Accounting Review, 18:3, 597-620, DOI:
10.1080/09638180902863779

To link to this article: https://doi.org/10.1080/09638180902863779

Published online: 04 Aug 2009.

Submit your article to this journal

Article views: 3673

View related articles

Citing articles: 123 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rear20
European Accounting Review
Vol. 18, No. 3, 597– 620, 2009

In Search of Management
Accounting Theory

TEEMU MALMI and MARKUS GRANLUND

 
Helsinki School of Economics, Finland and Turku School of Economics, Finland

ABSTRACT In this article we discuss the motivation for and role of theory in management
accounting. We argue that theories in an applied field such as management accounting
research should provide explanations that are useful for those we study – managers,
organizations and society. We evaluate the nature of theories currently used and
developed. Those theories that are considered theories by the research community are
largely imported from other disciplines, but have hardly anything that makes them
unique to management accounting. Those theories that are not currently regarded as
theories by many of our colleagues attempt to explain how to apply management
accounting to achieve superior performance. We argue that both forms of theories, at
present, largely fail to provide valid support for practitioners. We contend that
management accounting theory should help us to answer questions of what kind of
management accounting systems we should apply, how, in what circumstances, and
how to change them. We provide suggestions on how management accounting research
could proceed to produce better theories in this regard.

The Purpose of Management Accounting Theory

Why do we research management accounting (MA)?1 A recent debate provoked
by Zimmerman (2001) on the state of MA research and theory (Hopwood, 2002;
Ittner and Larcker, 2001, 2002; Luft and Shields, 2002; Lukka and Mouritsen,
2002; Zimmerman, 2001) provides some answers, but also raises a number of
further questions. The following is particularly puzzling: what is the purpose
and role of theory in MA research? Given that we can answer to this, the next
obvious questions are: do current theories fulfill this purpose? And how should
we, as an academic community, proceed in our theory development? These are
the issues we address in this paper.

Correspondence Address: Teemu Malmi, Helsinki School of Economics, PO Box 1210, FIN-00101,
Helsinki, Finland. Email: malmi@hse.fi

0963-8180 Print/1468-4497 Online/09/030597–24 # 2009 European Accounting Association

DOI: 10.1080/09638180902863779
Published by Routledge Journals, Taylor & Francis Ltd on behalf of the EAA.
598 T. Malmi & M. Granlund

In their response to Zimmerman, Luft and Shields (2002) refer to their

extensive literature review on theory consistent research in MA (Luft and
Shields, 2003). The review focuses on articles that explain the causes and
effects of MA. This is to say that the theory of, or theorizing in, MA is
about explaining its causes and effects. Hopwood (2002, p. 783) refers to
the emergence, functioning and impacts of MA practices as an object of
inquiry, thus extending the scope from causes and effects to include how
MA is practiced.2 Although not all accounting researchers subscribe these
views (see Baxter and Chua, 2003, for a review of some of this literature;
see also Chua 1986), this is the mainstream conception prevailing in today’s
MA research community. But why do we, as an academic community, try
to understand the causes, effects and functioning of MA? We believe that
the ultimate reason for this is to be able to use this understanding, or
theory, in creating better MA practices, both in terms of content and use
(cf. Chenhall, 2003, p. 159; Ittner and Larcker, 2001, p. 399). The definition
of ‘better’ is not crucial here as we believe that the goodness of any MA
practice depends on the objectives of the users of MA as well as the
organizational and social context in which MA practice takes place. The argu-
ment is that we undertake research and develop theories in MA to be used by
someone to accomplish something. However, in this regard the most radical
calls for practice-relevant MA research have had only a very limited impact
on research practice (Kaplan 1998; Kasanen et al., 1993; see also Mattessich,
1995a, b).
Zimmerman (2001) himself justifies the importance of accounting theory by
referring to a number of interest groups that have to make decisions about
external accounting reports. Moreover, Watts and Zimmerman (1986, p. 2)
argue that one important criterion for a theory’s success ‘is the value of the
theory to users’ (see also Demski et al., 1991). Hence, even the advocates of
positive accounting theory subscribe to the practical purpose for a theory in
accounting, although they beware of being normative.
Despite its practical purpose, explicit or implicit, MA research is often
criticized for not having an impact on practice, let alone leading it. Swieringa
(1998, p. 43) points to the fact that most academic researchers and their works
are not known outside the academia at all (cf. Lee, 2003):

They do research, publish the results, review the work of other researchers,
and build reputations. Academic researchers usually write for other
researchers and experts in their field. They are expected to be objective,
to meet high technical standards and to subject their work to rigorous
peer reviews, and they are concerned about the quality and cleverness of
their work and the reactions of colleagues.

There are also a number of concerns raised regarding the gap between research
and practice that it is suggested followed from the above:
 In Search of Management Accounting Theory 599

Are not the many contradictions between theory and practice vivid evi-
dence that in accounting we have not done enough to serve the practitioner,
the stockholder and, above all, society at large? (Mattessich, 1995b, p. 279,
emphasis in original)

There is anecdotal evidence to support the assertion that accounting

research, or what is alleged to be research, is of little or no value to the prac-
tice of accounting, nor to the development of accounting as an academic
discipline. (Inanga and Schneider, 2005, p. 227)

We believe the orientation towards colleagues, not external users of our theories
has partly contributed to the situation where academic MA research has had a
very limited effect on practice during the last few decades.3 As considerable
amounts of state and other funding are granted to perform research and education
in the area, Swieringa’s description of the role of accounting researchers is
not feasible in the long run (cf. Hambrick, 1994). It should not be in anybody’s
interest to generate research articles the results of which, if any, could not be of
less interest to practitioners.
In this paper we suggest that partly extending what we perceive as theory and the
role of theory in MA research may aid in making MA research relevant to a broad
audience. We aim to contribute to the MA literature by presenting a research
agenda to generate MA theories that are practice relevant and unique to our
field. We do not claim that we are saying something new about the problems we
face as an academic community – these have been widely discussed both in
accounting and related fields (Arrington and Schweiker, 1992; Quattrone, 2000;
see also Nagel, 1968; Woolgar, 1988). But we do try to offer one way ahead. As
Hopwood (2002) noted, there are no simple solutions. However, we intend to
provide one solution, but we are not claiming that the solution presented here
would be the only way ahead. Our solution is directed mainly to those whose
research is motivated by an attempt to find some ‘regularities’ in organizational
life that could be applied from one organization to another given their specific
context. We acknowledge the different interests of knowledge creation and the
ontological and epistemological assumptions of interpretative and critical research
traditions. Hence, our quest for MA theory is not primarily targeted to them.
In the next section we will define what we mean by theory. We then assess
briefly what type of theories we have at the moment to help us in building better
MA practices and what the main shortcomings of those are. This is followed by
what we see should be the core of MA theory, or a set of theories. We then provide
some avenues to develop such theories. Finally, we summarize the main arguments.

What is Theory?
Accounting academics seem to have very different perceptions of what is to be
regarded as theory. While, for some authors, theory seems to be all the existing
600 T. Malmi & M. Granlund

literature on management accounting and control, for others theory is rather an

orienting set of ideas or explanatory concepts (Ahrens and Chapman, 2006).
Sometimes accounting scholars refer to the theory being simply the psycho-
logical, economic, or social dynamics providing answers to why questions.
Similarly, some colleagues commenting on this manuscript argue that any idea
that can serve as a basis for a testable hypothesis is a theory. In more general
terms, social scientists have had troubles in defining what theory is (Llewelyn,
2003; Sutton and Staw, 1995). There is also great debate about what kind of the-
ories we should be looking for in social sciences in general (about the ‘science
wars’, see Flyvbjerg, 2001; Nagel, 1968; Oldroyd, 1986; Salmon, 1989). We
argue that all this has led to major problems in clearly identifying theoretical con-
tributions. It seems to be difficult to know when we have achieved a small/big/
relevant/significant theoretical contribution and made progress. Although the
identification of theoretical contributions is very difficult for senior researchers,
it is more so for doctoral students. Such a situation cannot be considered fruitful
for the progression of any discipline. Part of this problem derives from the
reluctance of MA researchers to define theory. They do discuss theory among
epistemological, methodological and other issues. However, they never seem
to define theory, but settle for discussing theorizing (see Llewellyn, 2003;
Quattrone, 2000).
We are aware that while discussing ways to conduct research we inexorably
also touch epistemological, methodological, paradigm (in)commensurability
and other issues and problems in the core of philosophy and sociology of
science. Yet such general concerns are beyond the scope of this paper. We are
unable to contribute in that regard, as the related debates within and around the
accounting academia seem not to proceed anywhere: there will always exist
different research camps arguing for the superiority of their own approach.
Therefore, we are not suggesting one best way to proceed (or that no useful
studies were made in the past), but rather argue for potential solutions through
which we could be more distinct, better identify theoretical contribution, and
also narrow down the gap between research and practice.4 Before that, we will
define our perception of the nature and content of theory to be further utilized
in the following sections.
Weick (1989) quotes Sutherland in defining theory – ‘an ordered set of
assertions about a generic behavior or structure assumed to hold throughout a
significantly broad range of specific instances’. This corresponds fairly well to
a standard dictionary definition: ‘a coherent group of general propositions used
as principles of explanation for a class of phenomena’ (Webster’s Dictionary;
emphasis added). Whetten (1989) argues that a complete theory must contain
four elements. The first is which factors (variables, constructs, concepts) logically
should be considered as part of the explanation of the phenomena of interest.
The second element is concerned with how they are related. What is included
(factors) and how they relate to each other together constitute the domain or
subject of the theory. In research papers providing formal models, constructs
 In Search of Management Accounting Theory 601

are boxes and relations are arrows.5 The third element relates to the question why.
What are the underlying psychological, economic, or social dynamics that justify
the selection of factors and the proposed causal relationships? The fourth element
focuses on the conditions that place limitations on the propositions generated
from a theoretical model. Questions such as who, where and when address the
temporal and contextual factors that set the boundaries of generalizability.
Whetten’s (1989) view is a more detailed account of what theory is than that of
Weick’s (1989) (these accounts are not contradictory in any way). Our view on
theory is akin to this. For us it appears that the view provided by Whetten is actu-
ally more or less the view held by most positive accounting researchers. For MA
theory the phenomena of interest could be, for example, financial performance of
the organization, its sustainability, or something else an organizational actor(s)
wants to achieve with the aid, or within the influence, of management accounting.
Accounting and many other constructs such as other forms of control, strategy,
environment and institutional setting are factors that can be part of the explanation
of the phenomena of interest. The relations between these factors and the rationale
of why these factors and their relations explain the phenomena of interest are part
of a theory. Finally, any MA theory should contain assertions on the conditions
that proposed explanations are assumed to hold.
Watts and Zimmerman (1986, pp. 7 – 9) make a clear distinction between posi-
tive and normative theory. Theory for them yields no prescription for accounting
practice. They argue that prescription requires the specification of an objective
and an objective function. In principle, we agree that prescription requires the
specification of an objective and an objective function (or value judgments,
see Mattessich, 1995b). However, we argue that most of the economics based
management accounting research assumes at least implicitly the objective of
economic efficiency or shareholder value maximization (cf. Mattessich, 1995b;
see also Boland and Gordon, 1992). Therefore, theories capable of explaining
and predicting how certain forms and uses of management accounting (1) lead
to decisions that are in line with these implicit objectives, (2) increase goal
congruence, and (3) ultimately produce financial gains are likely to provide
prescriptions relevant for business managers pursuing such goals. Moreover,
we argue that researchers should develop management accounting theories
devoted to value maximizing, theories devoted to social equality, theories
devoted to environmental sustainability, etc. In other words, we do not believe
that there can be only one base theory in management accounting we can build
on just by varying our objective function. Research on management accounting
would be better off if researchers would clearly assume and explicate an objective
and objective function, and build theories to support that objective.
To sum up, even if we call for theories to be useful in practice, we do not mean
that theories need to be normative as such. Theories capable of explaining and
predicting are likely to provide practical insights given that they deal with the
issues that are of practical interest, i.e. explaining and predicting phenomena
related to practitioners’ objectives.
602 T. Malmi & M. Granlund

The Types of Theories in Management Accounting

Research and practice in MA relies largely on two types of theories. Those theories
that are considered theories by the research community are imported mainly from
other social sciences, but have hardly anything that makes them unique to MA.
Those theories currently not regarded as theories attempt to explain how to
apply MA to achieve superior performance. Let us address them both in turn.

Theories that have Theory Status

Based on Luft and Shields (2003), a number of MA related issues in MA research
are explained by a number of theories. Typically, MA is either an independent
or a dependent construct, but the whole causal chain is not addressed. Theories
applied to explain causes, effects and various interrelationships of MA are
mainly from the fields of economics, organization theory (contingency theory),
sociology and psychology. There is nothing wrong with this; we need these
theories and many MA issues of interest can be explained by these theories.
However, we have three main concerns with current theorizing in MA.
First, none of these theories is something we could build on if we wish to dis-
tinguish MA theory from some other theory. For us, these are theories about MA,
not theories of MA (Humphrey and Scapens, 1996). MA theory, as judged based
on the publications in ‘top tier’ academic research journals, is determined by the
object of inquiry. Researchers use principal/agent theory, information economics
theory, structuration theory, actor-network theory, goal setting theory and a
countless number of other theories to explain issues of interest in MA. Sometimes
it seems that what is interesting is determined by the pre-adopted theory (and/or
method tool-box), not the accounting phenomena of interest.6 These same
theories can be used to explain a number of issues other than MA. We argue
that there is a need for theories of MA. This is not to say that we do not need
the explanations provided by and insights derived from these other theories.
We certainly do. But we argue there is room and need for both.
Why do we need unique theories of MA? Theories of MA could be problem-
focused (see Merchant et al., 2003), rather than disciplinary driven, providing
clear insights on practitioners’ concerns. For example, MA scholars could aim
to develop a theory of organizational incentive systems. Such a theory could
build on concepts derived from agency theory, expectancy theory, goal setting
theory and so on. The resulting theory would explain how to design and use
incentive systems to achieve superior performance (or some other outcome).
Such a theory would be one unique theory of MA as some of its constructs
consist of MA (e.g. performance measures) and relations between constructs
are explained using concepts derived from more than one ‘borrowed’ theory.
As should be evident, although we call for theories of MA, such theories
would hardly be totally new conceptualizations, but a coherent set of propositions
regarding some practitioner-related problem. We argue that, thereby, theories
 In Search of Management Accounting Theory 603

would be easier to communicate directly to various user groups, which would

again mean more influential research with direct practical effects.7
The overall consequence of having own practice relevant and useful MA
theories would be a stronger identity to MA as a discipline. This should
significantly enhance the acknowledgement and respect of MA research inside
and especially outside the academia. Furthermore, perhaps this would also
attract new good quality students to our field.
A limited insight for practitioners is the second concern we have with current
theorizing. Let us use some of the contingency research as an example of how
current ‘science’-based theories fall short in providing practice relevant guidance.
We use the contingency framework as a frame of reference as we believe it is fairly
explicit in its aim of providing insights to practice and as we see theories of
management accounting necessarily contingent (as opposed to covering laws).
As such we subscribe to the aim of the contingency-based research. The argument
is that this literature is not specific enough to provide much useful guidance to
practice (see Chapman, 1997, for a discussion of the trade-offs between simplicity,
accuracy and generalizability in variable definition).
Chenhall (2003, p. 138) summarizes the research findings relating the impact
of an uncertain external environment to MCS to a following proposition: the
more uncertain the external environment the more open and externally focused
the MCS. This proposition is based on research by Gordon and Narayanan
(1984), Chenhall and Morris (1986), Gul and Chia (1994) and Chong and
Chong (1997), among others. Gordon and Narayanan (1984) found that perceived
importance of external, non-financial and ex ante information is positively associ-
ated with increased perceived environmental uncertainty. External information
referred to events outside the domain of the organization. Chenhall and Morris
(1986) defined broad scope information in similar terms, and their construct
was used by Gul and Chia (1994) and Chong and Chong (1997). We may ask,
is it not obvious that if perceptions about the environmental uncertainty increase,
managers would value systems producing information about this external
environment? Moreover, we do not get any clues about the types of systems
that might serve these purposes. Would any system having this external focus
do? Does it not matter how systems are used? Our argument is that such prop-
ositions, or the underlying findings, are so general, and partly self-evident, that
they are of little use, or of incremental value, in practice. There is a need to
advance from this stage to be able to argue that as uncertainty increases,
certain forms of MCS used in a certain way would provide better decision-
making support, or more likely achievement of goal congruence.
Third, we do not have a clear picture of which one(s) of these theories pro-
vide(s) the strongest explanations for various MA forms and outcomes in a
given time frame and circumstances. For example, economics-based theories
explain the existence of MA practices as a means to enhance efficiency. Political
theories explain these practices by referring to various interest groups and their
struggle for power. Institutional theories provide signaling as an explanation as
604 T. Malmi & M. Granlund

firms seek legitimacy from their stakeholders. Empirical research has

demonstrated that all these have had an impact on MA practices. But do we
know when each of these explanations is likely to be valid? Are these mutually
exclusive? And even if we can answer these questions, so what? In other words,
even if we understand that accounting is practiced for various reasons, what are
the implications of this knowledge? Do these various reasons or motives to adopt
practices have an impact on how these practices are implemented and what the
expected benefits are? In the case where we do not understand the relative
power of various theories explaining the same phenomena, we do not actually
understand the issues surrounding MA in sufficient depth (and breadth) to be
able to provide clues about how to apply MA. One may argue that this is only
a question of insufficient accumulation of research knowledge, and will be
defeated as the field matures. We doubt this as we seldom see attempts to
address the strengths and weaknesses of various theories in a single study
(Merchant et al., 2003; see for example Ittner et al., 2003, for an exception).
Accounting scholars seem to believe one theory at a time, and make few attempts
to modify or further develop, let alone reject those borrowed theories they use
(Reiter and Williams, 2002; there are exceptions, of course, see for example
Baiman and Demski, 1980, with regard to principal-agent theory; Covaleski
and Dirsmith, 1988, with regard to institutional theory). If we are not willing
to develop theories we borrow from elsewhere to better explain issues we are
interested in, and if we are not prepared to assess the relative strengths and
weaknesses of various theories, there is little hope for coming up with practice
relevant MA theories.8 We reject Zimmerman’s (2001) solution to rely on only
economics-based theories (see also Lukka and Mouritsen, 2002); for us it is
once again an example of putting the carriage before the horse. Rather, the
explanatory and predictive power of various theories needs to be assessed, and
theories, whatever the origin, need to be developed, modified or combined to
increase their explanatory power in a management accounting context.9

Theories that do not seem to have Theory Status

In MA we also have a number of ‘normative’ theories or constructs. These
include activity-based costing for assigning overheads, Balanced Scorecard for
control system design, Quality Costing framework to manage and reduce
quality costs and Value Based Management framework to guide decision-
making and control to ensure shareholder returns. These are not regarded as
theories by many academic researchers, although these theories instruct how
we should be doing something and why. Zimmerman (2001) argues this as
follows: ‘while it [VBM] resembles various theories, it is not a positive theory
in the sense that it neither explains nor predicts firm-related phenomena’.10
Even if we accept Zimmerman’s quest for a positive theory, we disagree with
his view that VBM neither explains nor predicts firm-related phenomena.
It could be argued that the VBM framework aims to explain firm performance,
 In Search of Management Accounting Theory 605

measured by shareholder returns. Adherence to six managerial steps forming the

core of VBM (Martin and Petty, 2000; Morin and Jarrell, 2001; Ittner and
Larcker, 2001; Zimmerman, 2001) explains why one of the otherwise similar
firms performs better, or why a performance of a certain firm improves over
time. This same logic can be extended to prediction, and these predictions can
be refuted by evidence.
Recall Sutherland’s definition of theory: a coherent set of propositions used as
principles of explanation for a class of phenomena assumed to hold throughout a
broad range of specific instances. But is this not exactly what VBM proponents
claim? There is a coherent set of propositions regarding how organizations
should be managed in order to maximize shareholder wealth, which are
assumed to hold in a broad range of instances. These propositions of VBM are
not tested, but it can still be seen as a preliminary theory. Watts and Zimmerman
(1986, p. 2) contend that the objective of accounting theory is to explain and
predict accounting practice. Sure, VBM is not a theory of accounting practice
but a theory of organizational performance, including accounting-related issues
as a mechanism of explaining outcomes. We can even argue, that as VBM litera-
ture explains the mechanism by which better performance is about to emerge,
it deserves a status of theory more than some other so-called theories, that we
use to predict outcomes (usually with low R squares), but which treat the
actual mechanism (i.e. firm) as a black box (e.g. S-curve models in innovation
diffusion literature).
How does VBM comply with the four elements of theory as discussed by
Whetten (1989)? We may think that the six managerial steps provide a frame
for a managerial construct, organizational performance being the other major
construct in this theory. This managerial construct consists of managerial
choices and actions, including management control related choices. We may
call the main choices and actions illustrated in VBM literature as sub-constructs.
Six managerial steps now suggest which values those sub-constructs may get in
firms practicing VBM. For example, Step 4 (see Ittner and Larcker, 2001) in the
VBM framework argues that action plans, performance measures and target
setting should be based on identified value drivers. In practice, performance
measures may be defined relying on for example value drivers, strategy, TQM
framework or they may be ad hoc. Performance measures can now be considered
a sub-construct and part of the managerial construct. Values for the performance
measure sub-construct may vary, including for example ‘measures based on
value drivers’ or ‘measures defined ad hoc’.
The relationship between the managerial construct and organizational
performance is simple: if all these managerial steps are taken as suggested or,
more precisely, if all values of sub-constructs are aligned with VBM assertions,
firm performance will improve. Selecting some alternative methods to accom-
plish these managerial tasks will produce less shareholder value.
Why does this relationship between these constructs then exist? We do not
suggest that this preliminary theory is unambiguous about why these steps are
606 T. Malmi & M. Granlund

to produce better performance; although we consider the explanation fairly

simple and straightforward. Considering the management control part of it, for
example, it seems that expected benefits are based on an assumption that you
get what you measure. Moreover, making sure that measures reflect value cre-
ation (either outcome or drivers) should ensure that the system creates right
incentives for managers to behave in the best interest of shareholders. Further,
given that bonuses and career development are linked to the achievement of
such targets, these measures should provide further assurance that managers
take actions in the best interest of shareholders. Is this an economic or psycho-
logical explanation? We think that should not be the main concern. The aim is
to develop an explanation that reflects reality, whatever the reasons are: rational,
emotional or something else.
Finally, and very importantly, there is no attention to conditions in which these
propositions may hold. Therefore, we would not consider it, as it stands, a com-
plete theory, as it does not comply with all four elements necessary for a complete
theory. We agree with Zimmerman (2001) that this theory is presented as if
all firms should follow it. This is why we regard it as of limited value for
practitioners as it stands. Rigorous research should now provide support for
these arguments, or show where its limits lie. This would lead to refinement or
rejection of this preliminary theory. It may be obvious that it becomes rejected
as a universal truth, but do these steps hold under particular circumstances?
Are all steps required to secure success? What are the necessary and sufficient
conditions for assuming that VBM would lead to better performance? That
type of theory explication and refinement would be useful for practice as well.
Zimmerman (2001) also makes a case against VBM being a theory by arguing
that it does not make predictions about when particular compensation schemes will
be used or what firms are most likely to adopt Activity Based Costing (ABC). We
are not quite sure what Zimmerman is expecting from a theory, but it seems that he
is after a single theory that could explain everything. We may need one theory to
explain why firms adopt ABC, another theory to explain how overheads are to be
assigned to enhance decision-making and yet another theory to explain which
managerial steps are required to enhance shareholder value. These all may be
based on economics, or not, but even in economics there are different theories to
explain different phenomena. This is also in line with Mattessich (1995b, p. 262)
who suggests that we should offer a broad range of alternative purpose-oriented
models to the users of accounting information.
Let us take another example from recent MA literature, ABC. What the
proponents said in the early normative writings is that poor performance may
be explained by the use of overly simplified cost accounting models. They
went on to show how allocation of overheads based on volume distorts product
costs and leads to wrong decisions. Hence, they explained the mechanism produ-
cing poor performance. A number of other things may explain poor performance,
but cost assignment issues are definitely something accounting researchers and
practitioners should be concerned about. They suggested assigning costs based
 In Search of Management Accounting Theory 607

on causality instead of relying solely on volume. ABC may be thought of as a

theory of cost accounting if we consider that following certain cost assignment
procedures accounting produces more useful information for managerial
decision-making. Improved decision-making is assumed to lead to better per-
formance. Hence, the theory of ABC, or more broadly, the theory of cost account-
ing explains how cost accounting should be done, and why, to ensure better
performance. Even at its normative phase (which was grounded on inductive
case observations and/or consulting assignments), ABC proponents claimed
that cost assignment principles explicated as ABC are suitable for, and eliminate
certain costing distortions in, organizations with a complex product mix, a
complex production process and a large proportion of overheads. This attention
to the boundaries of these propositions comes close to the fourth element Whetten
calls for, for a theory to be complete. Management accounting research has so far
provided a number of valuable insights on whether these limitations hold and
what are the other possible limitations. Based on research, we may today
predict which firms are most likely to adopt ABC (certain cost assignment pro-
cedures) as hoped for by Zimmerman (2001). Note also that based on current
understanding these adoptions are explained both by economic and institutional
factors, institutional factors clearly not being an issue in the early normative
writings (see Malmi, 1999).
We would think about ABC as a tool, not a theory. A situation where ABC
refers to both theory and a certain set of practices is not desirable. Therefore, it
might be better to refer to the theory of cost accounting, or theory of product
costing, instead of the theory of ABC. The theory of cost accounting would
explain how overheads are to be assigned in different circumstances to provide
best possible decision-support. It would also explain how overheads are to be
assigned under different circumstances given that resulting figures are used for
achieving goal congruence or to support signaling. Traditional ABC as a tool
would be one method for accomplishing this. Time Driven ABC (Kaplan
and Anderson, 2004) and plant-wide overheads, for example, would be
alternatives.
To sum up, we argue that an attempt to establish scientific identity for MA
by borrowing theories from other fields has led us to an identity crisis as a
scientific discipline. There is no theory unique to MA that is currently considered
scientific by the international research community. What we do not understand
is why the academic MA community cannot be proud of practice-oriented MA
theories?11 These are something that distinguishes us from economists on the
one hand and from organization theorists on the other. These are something
practitioners can pursue in their attempts to improve their organizations. This
is not to say that these theories should be accepted and followed as they were
initially presented. The boundaries of these theories need to be explored and
new theories developed to meet the requirements that society sets for an
applied science like ours – to facilitate developments within society, including
organizations.
608 T. Malmi & M. Granlund

What Should be the Core of MA Theory

So far we have argued for MA theory that would be both useful in deriving guide-
lines for practice and somehow unique to our field. What should be the core of such
theory? From the managerial perspective the answer is fairly straightforward. First,
theory should consist of assertions of what accounting and control methods we
should apply, how, and in what circumstances (see also Ittner and Larcker,
2002, p. 788). By ‘accounting and control methods’ we do not simply mean
ABC, BSC, VBM or some other existing or emerging preliminary normative
theories arguing that certain methods lead to better performance. Accounting
and control methods may refer to some commonly used categorizations such as
reliance on financial performance measures, non-financial measures or both;
reliance on technocratic controls, socio-ideological controls or both; using
simple or sophisticated cost assignment, etc. ‘How’ may refer to using these
systems diagnostically, interactively or both; using them for decision-making,
control, or both; using them as supplements or complements to other methods,
using them in the top of the hierarchy, in the bottom of the hierarchy or in the
whole hierarchy, etc. ‘Circumstances’ include the objectives of management
accounting (enhance efficiency, environmental compliance, etc), traditional contin-
gency variables, institutional forces, political issues, economic factors, individual
factors, historical factors, etc. The word ‘should’ above, suggests that there is an
assumed relationship to performance. Note that also performance may be
defined differently in different circumstances; different objectives require different
assessment of achievement and proper performance for government organizations
is likely to be different from that of private sector organizations.
In a sense, what we are claiming is that MA theory should be a set of prop-
ositions of how to organize accounting and control practices under given circum-
stances. If we take performance management and management control systems as
an example, Otley (1999) provides a general framework (not a theory) for how to
conduct research in the area as well as what are the questions we need to be able to
answer based on our research and theorizing. He poses the questions in a manage-
rial tone, which is akin to our argument about the role of theory in MA. Otley’s five
questions focus on organizational objectives and the evaluation of their achieve-
ment; strategies, plans, processes and activities to achieve those objectives as
well as the assessment of the success of these means; the level of performance
in the previous two as well as the way to set targets on those; rewards or penalties
that are associated with the achievement of these targets and the information flows
that enable learning and adaptation. If we could come up with a coherent set of
propositions on how to respond to each of these questions in a given situation,
we could have a managerially useful, yet disciplinary distinctive theory of how
to use MCS to achieve superior performance.
Second, theory should consist of assertions of how we should change manage-
ment accounting. Why a change might be appropriate is already covered by the
questions above, while here the focus is on implementation. The word ‘should’
 In Search of Management Accounting Theory 609

refers here to performance implications or success of such change attempts. Various

implementation success- and failure-related factors provide a good basis for such
theory, although there is a lot more work to be done to link these factors to each
other and various circumstances in which these change attempts may take place.
We suggest that MA researchers attempt to develop not a single theory of MA,
but a set of theories able to explain the form, use and change of accounting and
control methods in a range of circumstances. We would like to emphasize once
more that we do not mean that all MA researchers should subscribe to what we
say, or should shift to develop and test the type of MA theories we propose. But
we do think that, in general, more emphasis should be devoted to this type of
theorizing than is currently taking place. It is important also to note that we
are not saying anything about the suitable theoretical starting points towards
these types of theories, nor do we have any particular view of preferred
methodologies as such. The main argument is simple: let us try to build MA
theories that are useful for accounting practice – preferably in the very form in
which they are published.

How Research Should Proceed to Produce Better Management

Accounting Theories?
We can identify three main avenues to produce theories that are eventually useful
in practice. The first one is to alter some current practices when conducting
traditional research based on ‘scientific’ theories. The second is to take
normative, preliminary theories as a starting point. The third is to rely more on
the interventionist research approaches.

Traditional Research
We used contingency literature (Chenhall, 2003) above to illustrate how current
research falls short in providing meaningful advice to practice. We argued that
there is a need, for example, to be able to argue that as uncertainty increases,
certain forms of MCS used in a certain way would provide better decision-
making support, or more likely lead to the achievement of goal congruence.
Such claims could be based on theories, which explain and predict the outcomes
of various forms and uses of MA in given circumstances. There is nothing new in
this argument. However, as results of this research paradigm have not been
compelling so far, we might need to re-think how we are approaching it.
There are seven suggestions for traditional research to progress. One of these
we have raised already above. We argue that researchers should assume objec-
tives and objective functions, and build theories that support the attainment
of these objectives. Second, research needs to incorporate the link between
accounting choice and performance to be complete and useful. This is exactly
what many researchers have done already and what both Ittner and Larcker
(2002) and Luft and Shields (2003) argue.
610 T. Malmi & M. Granlund

Third, we would also urge researchers to utilize more comprehensive manage-

rial or control systems, such as VBM discussed above or control systems packages
(e.g. Abernethy and Chua, 1996; Otley, 1980; Malmi and Brown, 2008)12 as a unit
of analysis. The underlying reason why a certain MCS element in isolation is
expected to have certain effects is probably not the most puzzling issue. Instead,
why certain combinations work together and others do not, in various circum-
stances, needs more attention. In other words, we are suggesting that instead of
trying to provide economic, psychological or some other explanation for atomistic
relationships, e.g. between MCS elements and performance, we could try to build
up theories containing more complex MCS constructs and explanations of why
those combinations are likely to produce certain outcomes in certain
circumstances. These explanations may build on various established theories
borrowed from related fields. However, as our MA constructs get more
complex, and hence realistic, these combinations of established theories required
to explain the outcomes are likely to become unique to MA.
Fourth, we should think carefully about the situations in which the study of a
random sample of organizations produces sufficient ground for a useful theory.
Hume’s guillotine postulates that from how things are we cannot infer how
things ought to be. This seems to be true when it comes to observed accounting
practices. Frequent changes in exogenous variables accompanied by the fact that
the time required for major changes in accounting and control systems has been
shown can be long, and the fact that most ideas, methods or principles in MA may
be applied in a number of ways, and the evidence of accounting failures, all
suggest that assuming that all firms optimize is bold at best and dangerous at
worst (see also Luft and Shields, 2002). If we cannot assume that an average
firm has optimal accounting and control methods, conclusions derived from
empirical studies regarding the usefulness of certain systems in certain circum-
stances are doomed to be biased. This is not to say that cross-sectional studies
are useless, or that there are never or even seldom situations where optimizing
behavior may be assumed. What we argue is that these issues should be explicitly
discussed in empirical papers.
Fifth, we should devote more attention to accounting and control practices in
successful organizations. It seems that there is a bias towards studying problems
and shortcomings of accounting, especially among those using the case method.
Case-based analysis of successful practice is seldom published. Similarly, unsuc-
cessful organizations could provide a basis for theorizing what systems not to use,
or how not to use them.
Sixth, if what we suggest above as a fruitful avenue for research seems a huge
task, there is no logical reason why we should continue to conduct MA research
individually or in small groups. Large research groups and projects may help to
overcome the problem of addressing overly simple relationships. In addition, we
suggest that research conducted in such groups could apply methodological tri-
angulation, i.e. apply more than just one research method (Modell, 2005). Such
triangulation possesses great potential for enhancing the validity of research
 In Search of Management Accounting Theory 611

results, which again facilitates the assessment of theoretical contribution.

Moreover, MA theory building is likely to benefit from more repetitive studies.
Decreasing somewhat the application of social meta theories providing new
‘insights’ on MA practice, and increasing to some extent the conduct of
repetitive studies in different contexts, would be a fruitful way towards develop-
ing a cumulative body of knowledge. We admit that for some the idea of
repetitive studies may sound boring. However, are we here to enlighten ourselves
or to develop a body of knowledge that could help managers and other actors to
improve their practices? It should be noted that we are not suggesting that
studies conducted in one country should be repeated as such with data from
other countries. Pure repetition without explicit a priori consideration of theory
contribution is definitely not what we are suggesting.
Seventh, according to the agenda put forward here, the object of inquiry should
be motivated and positioned by linking it to the underlying main research
question, i.e. cause and effect of MA or how to implement MA, and how that
is supposed to help in building a MA theory. This would help readers to assess
the contribution. Finally, the concluding sections of research papers should
make both the theory tested or developed and its practical implications explicit.
Moreover, we encourage researchers to speculate more in discussion sections
above the data (cf. Hinings and Greenwood, 2002). Even if the research is in
the early stage, and no definitive conclusions or prescriptions can be derived,
the discussion section provides us with an opportunity to speculate on the
meaning of the particular findings, or those assumed to be available after the
later stages of the research program have been finished, for practice.

Normative Theories
We see the development and testing of normative theories as a fruitful avenue
to produce more practice relevant research and theories of MA. It is easy to
argue that normative theories are usually presented as universal truths and
hence do not provide sufficient understanding of their potential limitations.
Moreover, they are typically presented in polemic style, lacking the precision
required from scientific theories (Lukka and Granlund, 2002). There is a need
to explicate used constructs, their relationships and underlying reasons and to
develop more contingent claims about their applicability. Moreover, an
additional complexity in studying normative theories is their changing
nature. Balanced scorecard, for example, was introduced as a more compre-
hensive information system for managerial decision-making, then as a tool
to translate strategy into action, essentially relying on the logic of management
by objectives, and finally as a tool to clarify and communicate strategies with
the aid of strategy maps. This development of ideas, especially in terms of
which practical problems they are about to solve, requires researchers to be
clear which variation of the idea they are addressing, both conceptually and
empirically.
612 T. Malmi & M. Granlund

We used ABC above as an example of an alternative method for overhead

assignment within the theory of cost accounting, illustrating how research has
produced valuable insights on circumstances in which to use it. From a practical
point of view, however, it is far from enough to explicate the circumstances in
which the use of ABC to assign overheads is likely to provide benefits. There
are a large number of practical considerations that are likely to have an impact
on how the functioning and results of a cost accounting model are understood,
how easy it is to maintain, how accurate the information it produces is, and so
on. These practical considerations include how to define activities, and what
type and how many cost drivers to use, to name just two. Theory of cost account-
ing should ultimately advance towards answering these questions. It appears to us
that too much emphasis is devoted to understanding why companies adopt ABC,
and too little to understanding how it should be applied to serve the purposes it is
adopted for. We believe accounting researchers could provide valuable insights
for practice by addressing these how questions.
Let us use BSC as another example. There are at least three different types of
scorecards (KPI-, stakeholder- and strategy scorecards) used in practice (Kaplan
and Norton, 2001). They seem to serve a different purpose as well. If we would
like to theorize about the use of BSC as a managerial control system, and develop
a theory of management control systems further, addressing the following BSC-
related questions are likely to be of help. We believe that the business community
applying scorecards would also warmly welcome answers to these questions:

(1) What are the necessary conditions for strategy scorecards to translate strategy
into action?
. For example, do measures need to be derived following assumed cause and
effect relationships?
. Do rewards need to be linked to achieving targets set for these measures?
(2) In which circumstances does strategy mapping as a method for deriving
scorecard lead to a proper set of measures?
. What is the nature of strategy; are there major changes taking place inside
the organization?
. In which circumstances should some other logic of deriving measures
be used?
(3) In which circumstances we should not use strategy scorecards at all?
. These may include factors traditionally studied in contingency studies, the
implications of existing control practices on the applicability of BSC, etc.13

Answers to these questions would help us towards a theory of how to apply BSC
in order to get desired results. An alternative approach would be to follow the lead
by Otley (1999) as discussed above. Research could assess how BSC is used and
ought to be used to facilitate each of these five steps. The specific questions above
as well as the role of BSC in each of these generic steps would provide
propositions as how to apply BSC and under which circumstances, and on the
 In Search of Management Accounting Theory 613

other hand, when not to apply, as the expected outcome is likely to be negative or
non-existent. One could also assume that many of the observations generated in
studying BSC are applicable to other existing and emerging management
accounting practices, hence providing elements for a more general theory of
MCS. As a result, we might not have a theory of BSC at the end, but a theory
of management control, where BSC is one option, or value, that some of the
sub-constructs of that theory may take.

Interventionist Studies
An alternative approach to create theories useful for practice is to solve practical
problems with practitioners and synthesize the novel solutions to a more general
form. In MA, such an approach has been suggested by Kasanen et al. (1993)
under the name of constructive research approach and by Kaplan (1998) under
the name of innovation action research. Recently, based on their screening of
how knowledge tends not to accumulate in the MA domain in a fruitful
manner, Lukka and Granlund (2002) suggested increasing reliance on the con-
structive approach.
The studies applying the constructive research approach (e.g. Malmi et al.,
2004; Tuomela, 2005; see also Labro and Tuomela, 2003) commit to strong
interventionist actions in the case organizations, and the researchers actively
participate in the innovation process of new management control constructs.
The action research tradition is not far from the constructive approach, although
typically implying weaker intervention and having no particular innovative
element (see Jönsson and Lukka, 2007).
Kaplan (1998) promotes the innovation action research approach, and starts his
piece quoting Kurt Lewin: ‘If social scientists truly wish to understand certain
phenomena, they should try to change them. Creating, not predicting is the
most robust test of validity-actionability’. Both the innovation action research
and the constructive research approach share a common core in this regard.
However, they are also very demanding research approaches, embodying
certain risks, not least ones related to the studied organizations (commitment
to the project over time, personnel replacements, etc). Of these two the approach
proposed by Kaplan (1998) can be considered as even more demanding, and may
probably not be even executable by most researchers. However, the risks and
potential problems associated with the constructive research approach, in particu-
lar, have been discussed, as well as guidelines and tactics to control them (Labro
and Tuomela, 2003; Lukka, 2000).
It is likely that positive accounting researchers would reject these approaches,
because researcher intervention does not belong in their agenda. Our position is
favorable to interventionist approaches, as in our view the potential of generating
directly applicable, yet theoretically informed solutions to practitioners is important
to pursue. However, it is true that in some sense the role of the academic researcher
is at stake here and should be solved: how normative can we be without becoming
614 T. Malmi & M. Granlund

labeled as consultants? As Bazerman (2005, p. 29) notes: ‘As a graduate student in

the late 1970s, I was trained to be descriptive; prescription was for consultants, not
for serious researchers. I now believe that this attitude is wrong, not only for the
field of management but for society as a whole.’ Kasanen et al. (1993) provide
functioning guidelines in this regard. According to them the most important
element of constructive research is its constant theory connection, which
demarcates research from consultancy.14 Moreover, the strengths of such
approaches seem to be evident. By acting as experts in real-life development
projects, we can simultaneously produce research results that are both practically
(practice relevance guaranteed) and theoretically interesting.
The interventionist approaches presented above subscribe to the idea of
pragmatic truth theory distinctly, although to varying degrees (Kasanen et al.,
1993; see also James, 1955).15 Overall, this implies that we should always first
consider relevance and then truth, rather than vice versa. An underlying assumption
is that in problem solving science researchers can and should be prescriptive and
the validity of the results is tested through implementation: what works in practice
is true. The fundamental question then is: how to relax the dominant view
embedded in most conventions of MA research that avoids intervention and
sticks strictly to the very traditional criteria of science, such as statistical
generalizability only and objectivity. Regarding the former, Lukka and Kasanen
(1995) argue – while demonstrating problems with statistical generalizations
and uncertainty embedded in all inductive reasoning – that we can also pursue
other forms of generalizability and suggest contextual (mainly in case studies)
and constructive (in constructive research approach) rhetorics of generalizability
to be applied beside statistical ones. Regarding objectivity, we would follow
Mattessich (e.g. 1995b, p. 279) who suggests that the major criterion of objectivity
in an applied discipline is the clear revelation of its value judgments (or objectives/
objective functions) and extensive empirical testing of its prescriptions.
An important thing to realize is that when succeeding in producing a construct
to solve a practical problem that is of theoretical interest, and if the construct is
shown to be working, we have built a MA theory (Malmi et al., 2004). In that case
there is no need for any other kind of theory building. The functionality of the
theory should then of course be tested in other similar types of organizations,
and further under very different circumstances. The scope and applicability of
the new theory would thus be examined.

Summary and Conclusion

In this paper we have argued that one purpose of research and theorizing in MA
should be on determining which MA practices work and in which circumstances.
We argue that there is a need for MA theories addressing what systems or
techniques to use, how and in which circumstances. We also need theories
explaining how to change MA practices. Such theories would be unique to our
field and useful for practice. We believe that what we argue is not only an ‘I
 In Search of Management Accounting Theory 615

have a dream’ type of manifesto: the discussion has roots in academic research
traditions. What the acceptance of our suggestions necessitates in the first
place is acknowledgment of MA as an applied science. If this cannot be agreed
on, then we would like to see explicit and justified argumentation for denial:
why MA should be seen as a pure science, such as physics (cf. engineering as
an applied science) or biology (cf. medicine as an applied science; see Mattes-
sich, 1995a, b), or a social science in the form where the only objective would
be to understand how social systems work.
We argue that there are two types of theories currently used and developed.
Theories that are used by the research community are borrowed from related
fields. Although capable of explaining a number of issues of interest, they are
seldom – or in conjunction with MA practices – helpful in explaining what
systems to use, how and in which circumstances. We also have a number of nor-
mative theories that are not regarded as theories by the academic community.
These theories aim to give guidance to practice, but seldom address the potential
shortcomings and inherent limitations.
We outlined three major avenues for further research. The first one is to
develop traditional research approaches. It appears that we need to assume an
objective function to be able to build meaningful theories. We also need to
relax, at least partly, the assumption that firms are optimizing their management
accounting and control systems. Hence, studies should address the performance
implications of various practices. Moreover, we believe that more complex man-
agement control system constructs should be studied. The second avenue is to
develop existing ‘branded’ practice theories, such as ABC or BSC, to more com-
plete theories by specifying constructs, relationships and underlying mechanisms
more clearly and addressing their limitations. The third avenue is to rely on inter-
ventionist research approaches, such as the constructive research approach or
innovation action research, and get involved in theory building, testing, and
refinement by creating new practices.
We believe that the research agenda suggested here would help us as a research
community to provide more assistance to the organizations and societies than is
currently the case. It could also solve at least some concerns related to the gap
between research and practice, and address the conjectures regarding the
failure of MA research to produce a cumulative body of knowledge. Finally,
we believe it would differentiate our field of inquiry from related fields, and
give MA a stronger identity enhancing its acknowledgment and respect inside
and outside academia.

Acknowledgments
We would like to thank Thomas Ahrens, Chris Chapman, Trevor Hopper, Jari
Huikku, Marko Järvenpää, Taru Lehtonen, Sari Lounasmeri, Kari Lukka, Kai
Luotonen, Falconer Mitchell, Jan Mouritsen, Vesa Partanen, Robert Scapens,
Frank Selto, Jani Taipaleenmäki, Tuija Virtanen and the participants of research
616 T. Malmi & M. Granlund

seminars at the University of New South Wales, University of Technology

Sydney and University of Queensland, as well as the participants of the
Second Global Management Accounting Research Symposium (Sydney, June
2005), the Workshop ‘When management control tools meet reality’ (Aarhus,
Denmark, December 2005), and session participants in the 29th Annual
Congress of the EAA (Dublin, March 2006) and the 2006 AFAANZ Conference
(Wellington, NZ, July 2006) for helpful comments. The comments by three
anonymous referees and the Editor have clearly improved our argumentation.

Notes
1
We define Management Accounting following CIMA (1996): ‘the process of identification,
measurement, accumulation, analysis, preparation, interpretation and communication of infor-
mation used by management to plan, evaluate and control within an entity and to assure appro-
priate use of and accountability for its resources’. In addition, Management Accounting refers
to a scientific discipline (see Quattrone, 2000).
2
Functioning may be captured in studying causes and effects by the variance in MCS as a result
of causes, or as a reason for variance in effects. This interpretation would suggest these authors
use different words for similar objects of inquiry.
3
A more complete account of various reasons why research has had limited impact on practice is
beyond the scope of this paper. Lee (2003) provides one recent and persuasive account on this
(see also Lee, 1989, 1997, 1999).
4
For an ample analysis on epistemology in accounting research and a potential solution for
related challenges, see Quattrone (2000; see also Arrington and Puxty, 1991; Arrington and
Schweiker, 1992).
5
These elements of theory are illustrated in the summary maps by Luft and Shields (2003).
6
See Reiter and Williams (2002) and Mattessich (1984) for positive economic theories, Scapens
(1990) in the context of case studies informed by sociology and Merchant et al. (2003)
regarding research on organizational incentive systems. In addition, in the mainstream research
tradition the quality criteria for research are borrowed from natural sciences. This has led to
situations where, in particular, statistical generalizability has been used as ‘an amulet’ to
demonstrate contribution, while at the same time the economic significance of such statistically
significant results may have remained practically non-existent (see Lukka and Kasanen, 1995).
7
Many academics tend to believe that they have a major impact on practice through teaching. In
those universities where we have taught (internationally) it appears that the majority of students
are more interested in passing the exam and getting the degree than in the course contents: they
seem to forget amazingly quickly. Therefore, we are more pessimistic about the actual effec-
tiveness of this channel. Furthermore, some academics have stated that firms hire academics
as part-time consultants, which is supposed to prove that academics have a direct effect on
practice. The same regards publishing popularized pieces of research in practitioner-oriented
magazines and books. Our question in this regard is: why do we need to do this separately to
have an effect? Why cannot we offer research results that are applicable in practice in their
own right?
8
In their review, Luft and Shields (2003) considered only studies that found support for the
proposed hypothesis. Although probably a reasonable choice for the review purposes, it may
be seen to illustrate the interplay between theory and empirical data in current management
accounting research. If the data do not support the hypotheses, then it is not the fault of
improper (borrowed) theory, but inappropriate data or methods.
9
Many of the remarks we make here regarding the first and the third concern are tightly connected to
the prevailing tenure and elite systems as well as the university and education program rankings
 In Search of Management Accounting Theory 617

that are known to produce many undesired effects regarding research ethics and the accumulation
of scientific knowledge in general (Lee, 1997, 2003; Lee and Williams, 1999).
10
Ittner and Larcker (2002) echo this in their reply by saying ‘we did not intend this framework to
be a theory of managerial accounting practice, or a complete depiction of the many economic
and non-economic factors affecting managerial accounting choices and consequences.’
11
We do understand that management accounting has developed in line with financial accounting,
dominated by economic positivism and a small group of powerful US academics. The shift in
the 1960s into the ‘scientific mode’ explains where we stand at this point. But how long do we
have to continue this scientific agenda just to prove to some other academic fields that we really
do science?
12
We are suggesting here that, for example, VBM needs to be decomposed into its sub-constructs,
and we need to study all these sub-constructs simultaneously. We do not see much value of
asking organizations whether they use VBM and then regressing use to performance without
controlling what use means.
13
It should be noted that some of these questions might be answered by traditional research
approaches as well. However, providing descriptive answers to one or two of these questions
alone – i.e. without consideration of the other relevant questions put forward above – may
have no value in the end regarding the development of strategic performance measurement
systems in practice.
14
Lukka (2000, 2005) has recently enhanced a view emphasizing that the main purpose of the con-
structive research approach should be to produce theory contribution, the development of a new
innovative construction being only one result.
15
Particularly in the constructive research approach, testing the functionality of the created
construct in the spirit of pragmatism plays an important role in the research process. For the differ-
ent market test levels for constructs, see Labro and Tuomela (2003; cf. Kasanen et al., 1993).

References
Abernethy, M. A. and Chua, W. F. (1996) A field study of control system ‘redesign’: the impact of insti-
tutional processes on strategic choice, Contemporary Accounting Research, 13(2), pp. 569–606.
Ahrens, T. and Chapman, C. (2006) Doing qualitative field research in management accounting:
positioning data to contribute to theory, Accounting, Organizations and Society, 31(8), pp. 819–841.
Arrington, C. E. and Puxty, A. G. (1991) Accounting, interests, and rationality: a communicative
relation, Critical Perspectives on Accounting, 2(1), pp. 31–58.
Arrington, C. E. and Schweiker, W. (1992) The rhetoric and rationality of accounting research,
Accounting, Organizations and Society, 17(6), pp. 511–533.
Baiman, S. and Demski, J. S. (1980) Economically optimal performance evaluation and control
systems, Journal of Accounting Research, 18(suppl), pp. 184–220.
Baxter, J. and Chua, W. F. (2003) Alternative management accounting research – whence and
whither, Accounting, Organization and Society, 28(2/3), pp. 97–126.
Bazerman, M. H. (2005) Response: conducting influential research – the need for prescriptive
implications, Academy of Management Review, 30(1), pp. 25–31.
Boland, L. A. and Gordon, I. M. (1992) Criticizing positive accounting theory, Contemporary
Accounting Research, 9(1), pp. 142–170.
Chapman, C. S. (1997) Reflections on a contingent view of accounting, Accounting, Organization and
Society, 22(2), pp. 189–205.
Chenhall, R. (2003) Management control systems design within its organizational context: findings
from contingency-based research and directions for the future, Accounting, Organization and
Society, 28(2/3), pp. 127–168.
Chenhall, R. and Morris, D. (1986) The impact of structure, environment, and interdependence on
the perceived usefulness of management accounting systems, The Accounting Review, 61(1),
pp. 16–35.
618 T. Malmi & M. Granlund

Chong, V. K. and Chong, K. M. (1997) Strategic choices, environmental uncertainty and SBU
performance: a note on the intervening role of management accounting systems, Accounting
and Business Research, 27(4), pp. 268–276.
Chua, F. W. (1986) Radical developments in accounting thought, The Accounting Review, 61(4),
pp. 601–632.
CIMA (1996) Official Terminology (London: CIMA).
Covaleski, M. A. and Dirsmith, M. W. (1988) The use of budgeting symbols in the political arena: an
historically informed field study, Accounting, Organizations and Society, 13(1), pp. 1–24.
Demski, J. S., Dopuch, N., Lev, B., Ronen, J., Searfoss, G. and Sunder, S. (1991) A Statement on the
State of Academic Accounting. Statement to the Research Director of the American Accounting
Association.
Flyvbjerg, B. (2001) Making Social Science Matter – Why Social Inquiry Fails and How It Can
Succeed Again (Cambridge: Cambridge University Press).
Gordon, L. A. and Narayanan, V. K. (1984) Management accounting systems, perceived environmental
uncertainty and organizational structure: an empirical investigation, Accounting, Organization and
Society, 9(1), pp. 33–47.
Gul, F. A. and Chia, Y. M. (1994) The effects of management accounting systems, perceived environ-
mental uncertainty and decentralization on managerial performance, Accounting, Organization
and Society, 19(4/5), pp. 413–426.
Hambrick, D. C. (1994) What if the academy actually mattered? (1993 Presidential Address),
Academy of Management Review, 19(1), pp. 11–16.
Hinings, C. R and Greenwood, R. (2002) Disconnects and consequences in organization theory?
Administrative Science Quarterly, 47(3), pp. 411–421.
Hopwood, A. (2002) ‘If only there were simple solutions, but there aren’t’: some reflections on
Zimmerman’s critique of empirical management accounting research, European Accounting
Review, 11(4), pp. 777–785.
Humphrey, C. and Scapens, R. (1996) Theories and case studies of organizational accounting practices:
limitation or liberation? Accounting, Auditing and Accountability Journal, 9(4), pp. 86–106.
Inanga, E. L. and Schneider, W. B. (2005) The failure of accounting research to improve accounting
practice: a problem of theory and lack of communication, Critical Perspective on Accounting,
16(3), pp. 227–248.
Ittner, C. D. and Larcker, D. F. (2001) Assessing empirical research in managerial accounting:
a value-based management perspective, Journal of Accounting and Economics, 32(1-3),
pp. 349–410.
Ittner, C. D. and Larcker, D. F. (2002) Empirical managerial accounting research: are we just
describing management consulting practice, European Accounting Review, 11(4), pp. 787–794.
Ittner, C. D., Larcker, D. F. and Meyer, M. W. (2003) Subjectivity and the weighting of
performance measures: evidence from a Balanced Scorecard, The Accounting Review, 78(2),
pp. 725–758.
James, W. (1955) Pragmatism and Four Essays from the Meaning of Truth (The New American
Library).
Jönsson, S. and Lukka, K. (2007) There and back again: doing interventionist research in management
accounting, in: A. Hopwood, M. Shields and C. Chapman (Eds) The Handbook of Management
Accounting, Vol. 1, pp. 373–397 (Amsterdam: Elsevier).
Kaplan, R. S. (1998) Creating new management practice through innovation action research, Journal
of Management Accounting Research, 10, pp. 89–118.
Kaplan, R. S. and Anderson, S. R. (2004). Time driven activity based costing, Harvard Business
Review, November, pp. 131–138.
Kaplan, R. S. and Norton, D. P. (2001) Transforming the Balanced Scorecard from performance
measurement to strategic management: part I, Accounting Horizons, 15(1), pp. 87–104.
Kasanen, E., Lukka, K. and Siitonen, A. (1993) The constructive approach in management accounting
research, Journal of Management Accounting Research, 5, pp. 243–264.
 In Search of Management Accounting Theory 619

Labro, E. and Tuomela, T.-S. (2003) On bringing more action into management accounting research:
process considerations based on two constructive case studies, European Accounting Review,
12(3), pp. 409–442.
Lee, T. A. (1989) Education, practice and research in accounting: gaps, close loops, bridges and magic
accounting, Accounting and Business Research, 19(75), pp. 237–253.
Lee, T. A. (1997) The editorial gatekeepers of the accounting academy, Accounting, Auditing and
Accountability Journal, 10(1), pp. 11–30.
Lee, T. A. (1999) Anatomy of a professional elite: the executive committee of the American
Accounting Association, Critical Perspectives on Accounting, 10(2), pp. 247–264.
Lee, T. A. (2003) Accounting and auditing research in the United States, in: C. Humphrey and B. Lee
(Eds) The Real Life Guide to Accounting Research: A Behind-the-Scenes View of Using
Qualitative Research Methods (Oxford: Elsevier).
Lee, T. A. and Williams, P. W. (1999) Accounting from the inside: legitimizing the accounting
academic elite, Critical Perspectives on Accounting, 10(6), pp. 867–895.
Llewelyn, S. (2003) What counts as ‘theory’ in qualitative management and accounting research,
Accounting, Auditing and Accountability Journal, 16(4), pp. 662–708.
Luft, J. and Shields, M. D. (2002) Zimmerman’s contentious conjectures: describing the present and
prescribing the future of empirical management accounting research, European Accounting
Review, 11(4), pp. 795–803.
Luft, J. and Shields, M. D. (2003) Mapping management accounting: graphics and guidelines for theory-
consistent empirical research, Accounting, Organization and Society, 28(2/3), pp. 169–249.
Lukka, K. (2000) The key issues of applying the constructive approach to field research, in: T. Reponen
(Ed.) Management Expertise for the New Millennium (Publications of the Turku School of
Economics, Series A-1).
Lukka, K. (2005) Approaches to case research in management accounting: the nature of empirical
intervention and theory linkage, in: S. Jönsson and J. Mouritsen (Eds) Accounting in Scandinavia:
The Northern Lights (Liber and Copenhagen Business School Press).
Lukka, K. and Granlund, M. (2002) The fragmented communication structure within the accounting
academia: the case of activity-based costing research genres, Accounting, Organization and
Society, 27(1/2), pp. 165–190.
Lukka, K. and Kasanen, E. (1995) The problem of generalizability: anecdotes and evidence in
accounting research, Accounting, Auditing and Accountability Journal, 8(1), pp. 71–90.
Lukka, K. and Mouritsen, J. (2002) Homogenity or heterogeneity of research in management
accounting, European Accounting Review, 11(4), pp. 805–811.
Malmi, T. (1999) Activity-based costing diffusion across organizations: an exploratory empirical
analysis of Finnish firms, Accounting, Organization and Society, 24(8), pp. 649–672.
Malmi, T. and Brown, D. (2008) Management control systems as a package – challenges, opportunities
and research directions, Management Accounting Research, 19(4), pp. 287–301.
Malmi, T., Jarvinen, P. and Lillrank, P. (2004) A collaborative approach for managing project cost of
poor quality, European Accounting Review, 13(2), pp. 293–317.
Martin, J. D. and Petty, J. W. (2000) Value Based Management: The Corporate Response to the
Shareholder Revolution (Harvard Business School Press).
Mattessich, R. V. (1984) Modern Accounting Research: History, Survey, and Guide (Vancouver:
Canadian Certified General Accountants’ Research Foundation).
Mattessich, R. V. (1995a) Critique of Accounting: Examination of the Foundations and Normative
Structure of an Applied Science (London: Quorum Books).
Mattessich, R. V. (1995b) Conditional-normative accounting methodology: incorporating value-
judgments and means-end relations of an applied science, Accounting, Organizations and
Society, 20(4), pp. 259–284.
Merchant, K. A., Van der Stede, W. A. and Zheng, L. (2003) Disciplinary constraints on the advance-
ment of knowledge: the case of organizational incentive systems, Accounting, Organization and
Society, 28(2/3), pp. 251–286.
620 T. Malmi & M. Granlund

Modell, S. (2005) Triangulation between case study and survey methods in management
accounting research: an assessment of validity implications, Management Accounting Research,
16(2), pp. 231–254.
Morin, R. A. and Jarrell, S. L. (2001) Driving Shareholder Value: Value-Building Techniques for
Creating Shareholder Wealth (New York: McGraw-Hill).
Nagel, E. (1968) The Structure of Science (Harcourt: Brace and World Inc).
Oldroyd, D. (1986) The Arch of Knowledge. An Introductory Study of the History of the Philosophy of
Science (Kensington: NSWU Press).
Otley, D. (1980) The contingency theory of management accounting: achievement and prognosis,
Accounting, Organization and Society, 5(4), pp. 575–607.
Otley, D. (1999) Performance management: a framework for management control systems research,
Management Accounting Research, 10(4), pp. 363–382.
Quattrone, P. (2000) Constructivism and accounting research: towards a trans-disciplinary perspec-
tive, Accounting, Auditing and Accountability Journal, 13(2), pp. 662–708.
Reiter, S. A. and Williams, P. F. (2002) The structure and progressivity of accounting research: the
crisis in the academy revisited, Accounting, Organization and Society, 27(6), pp. 575–607.
Salmon, W. C. (1989) Four Decades of Scientific Explanation (University of Minnesota).
Scapens, R. W. (1990) Researching management accounting practice: the role of case study methods,
British Accounting Review, 22(3), pp. 259–281.
Sutton, R. and Staw, B. (1995) What theory is not, Administrative Science Quarterly, 40(3),
pp. 371–384.
Swieringa, R. J. (1998) Accounting research and policy making, Accounting and Finance, 38(1),
pp. 29–49.
Tuomela, T.-S. (2005) The interplay of different levers of control: a case study of introducing a new
performance measurement system, Management Accounting Research, 16(3), pp. 293–320.
Watts, R. L. and Zimmerman, J. L. (1986) Positive Accounting Theory (Englewood-Cliffs, NJ:
Prentice-Hall).
Weick, K. E. (1989) Theory construction as disciplined imagination, Academy of Management
Review, 14(4), pp. 516–531.
Whetten, D. A. (1989) What constitutes a theoretical contribution? Academy of Management Review,
14(4), pp. 490–495.
Woolgar, S. (1988) Science: The Very Idea (Routledge).
Zimmerman, J. L. (2001) Conjectures regarding empirical managerial accounting research, Journal of
Accounting and Economics, 32(1– 3), pp. 411–427.