This paper offers an economic explanation of an observed tendency to adopt and (re)design horizontal management control systems. To this end, the paper first describes generic vertical and horizontal management control templates: archetypes of management control. These archetypes draw on bureaucratic control mechanisms and some of them also on a market mechanism. Next, the observed tendency towards horizontalization is explained by making a differential transaction cost economics analysis of the movement towards horizontal archetypes of management control. This is not with the intention to generategeneral normative statements nor to generate a general positive economic theory of choices in management control systems design. Rather, the paper aims at acquiring a richer understanding of the economics of the horizontal direction into which management control systems are developing. In particular, it throws some light on the importance of various transactional and behavioural features for an efficient adoption and tailor-made design of management control systems in practice.
Inside organizations there is a tendency to concentrate the execution, and to
centralize the task control, of certain activities and services. Although this tendency
is not limited to facilitating and supporting activities, it is here that this tendency
most clearly manifests itself. Many supporting and facilitating activities like housing,
IT, cleaning services, maintenance and security tend to be separated from the
core activities of other organizational units and to be concentrated in so-called
‘shared service centres’. The tendency can also be seen in the context of accounting
activities, where some organizations are now considering or have actually realized the concentration and centralization of certain accounting activities into accounting
houses. Driven by the need for economies of scale and/or the need to create space
for flexibility, quality, speed and punctuality in the firm’s core business processes,
and supported by developments in information and communication technology the
execution and/or control of these activities is being shifted to a central level. In
some cases the centralization process is only a stage in a process that ends with
outsourcing the activities. Of course, outsourcing involves the transformation from a
relationship within the boundaries of an organization into a relationship in a market.
Processes of concentration and centralization encompass changes in the organization’s
structure as well as changes in management control relationships. In order to
enhance the efficiency and effectiveness of activities in shared service centres, management
control relationships will have to be (re)shaped: who will be influencing the
performance of the centre and what mechanisms and devices will be used to manage
performance from outside the unit? In many cases centralization keeps pace with
the adoption and implementation of horizontal management control systems: client–
supplier relationships. This is consistent with Otley’s observation that in many organizations,
systems of mutual accountability are emerging (Otley, 1994).
This paper concentrates, in particular, on management control change related to
centralization. Assuming that in practice economic motives are rather dominant
in starting a change process the paper will especially deal in some detail with
the economics of centralization and the economics of the movement towards the
adoption and implementation of horizontal management control relationships.
In order to set the scene, the following section contains a short case description. The
description underlines the practical relevance of the issue in this paper by providing
an insight into the relationships between centralization and change in management
control systems. It is based on a case study at Leiden University, conducted in 1995.
Documents were studied and several people within the University were interviewed.
One of the interviewees was the University’s controller, who was also head of the
task force charged with drawing up a proposal for a new management control
system for the facilities services. Drawing on the case study, section three contains
a number of generic management control archetypes. Horizontal archetypes of
management control are outlined in addition to a vertical one. In sections four
and five the movement towards centralization and towards horizontal management
control systems is explained and clarified by applying economic theory, especially
new-institutional organizational economics. This is done in order to generate a richer
understanding of the economics of centralizing the execution of facilities services and
of horizontalizing management control. The analysis may confirm the rationality of
the decisions made in practice, but it may also support some additional arguments
that are neglected by professional practitioners, or it may even lead to contradictory
conclusions. Finally, in section six some conclusions are drawn about theoretical
as well as practical impacts. Furthermore, the limitations of a new institutional
economics perspective are articulated.
Projects on centralization and concentration of facilitating activities often encompass
change in management control relationships. In this paper, choices of horizontal
management control systems were analysed using economic theory, especially
organizational economics. An interpretation of the main decisions in a project of
centralizing facilities services and of changing management control systems through
a TCE-lense provides an insight into the differential consequences of different
archetypes of management control in terms of production costs (economies of scale)
and transaction costs. Choices of archetypes can be linked to the characteristics of
the services involved, the extent of the user’s needs and the level of uncertainty and
complexity surrounding the performance of services.
The theoretical reasoning does not lead to findings which contradict the actions
of the practitioners in the case description. TCE-reasoning can lead a decision
maker, who has to decide upon the adoption of management control systems, to
choose internal centralization, as in the case of Leiden University. It does have
the potential, however, to add some elements to the reasoning of practitioners.
First, it emphasizes the link between the adoption and tailoring of (horizontal)
management control systems on the one hand, and the activities and transactions
they are expected to control on the other hand. Therefore, the theoretical reasoning
can lead practitioners to refine their arguments in choosing a management control
system, and to differentiate management control systems as the characteristics of the
services differ. Second (and closely connected with the first), the theoretical reasoning
brings the notion of transaction costs clearly into focus. In addition to economies of
scale, a potential impact on transaction costs can be important in choosing between
alternative archetypes of management control. Incorporating transactions costs into
the analysis implies explicit attention to behavioural risks connected with different
archetypes of management control.
The analysis was by no means intended to generate general normative statements
or, alternatively, a general positive theory on choices of management control
archetypes. The intention was just to find out, whether a review through the lenses
of TCE would lead to a confirmation of the arguments and line of reasoning of
the practitioners, to a possible refinement of the arguments, or to contradictory
conclusions.
Of course this analysis has its limitations. As Burns and Scapens state, being
fundamentally rooted in neo-classical economic theory, transaction cost economics
has difficulty in analysing processes of change (Burns and Scapens, 2000, p. 5).
Although TCE has the potential to offer an economic rationale for the direction of
the change towards horizontal archetypes of management control systems (or the
outcomes of processes of adopting and (re)designing management control systems),
it does not have the potential to clearly reveal and depict the processes inside the
organization, when moving from one stage to another. Old institutional economics is
better equipped for this (Scapens, 1994; Burns and Scapens, 2000).
TCE furthermore starts from the premise that, for a (boundedly) rational decision
maker, there is a deliberate choice regarding organizational forms. Applied to the
research topic in this paper, this implies that several alternative management
control archetypes are consciously weighed ‘ex ante’ and that the decision maker
tries to take an optimal decision. This contradicts notions in new institutional
146 Ed G. J. Vosselman
sociology (Powell and DiMaggio, 1991) stating that there is non-choice behaviour
in organizations. Whereas through the lenses of transaction cost economics, the
adoption of a management control archetype is viewed as the product of purposive
choice behaviour based on technical-rational considerations, sociological variants of
institutional theory view observed archetypes as the consequences of non-choice
behaviour in an embedded institutional environment. This non-choice behaviour is
captured by the notion of isomorphism. This means, that organizations just come to
imitate each other’s archetypes of management control because they are in the same
environments or organizational fields.
TCE does not pay any attention to elements from outside the organization that influence
changes in management control systems inside the organization. Roberts and
Greenwood (1998) in this regard suggest, that although people making decisions on
organizational forms seem to be engaged in processes of purposive and deliberate
choice, the process is institutionally constrained. Purposive, efficiency-seeking
behaviour guides the decision maker’s motives to evaluate the existing archetype of
management control and to search for possible alternative archetypes, because they
are forced to do so by competitive forces in the market or by the institutional pressure
of powerful actors. But during the process there will be preconscious institutional constraints.
These constraints (Roberts and Greenwood, 1998, p. 359) imply that the complete
set of known archetypes may contain several that are not assessed by decision
makers, because they do not match the ‘taken-for-granted’ values and beliefs about
appropriateness in that organizational field or in those organizational contexts. To
put it differently: the set of ‘legitimated’ alternative archetypes is smaller than the full
set of known archetypes. Furthermore, decision makers tend to emulate successful
archetypes in the organizational field (mimetic isomorphism). So, organizations may
have a bias towards archetypes with a high level of cognitive legitimacy. Decision
makers tend to assess the efficiency of different archetypes of management control
against the overall performance of observed organizations in their field and to imitate
the actions of organizations that are perceived as being successful. Furthermore, the
evaluation of the selected set of control archetypes will be dependent on normative
requirements, i.e. the existing norms and values, within the organizational field.
The decision to adopt a certain archetype of management control will also be
made in the presence of postconscious institutional constraints. These constraints consist
of coercive pressures or normative pressures to select from the subset of designs
those designs currently either endorsed by or adopted by legitimated institutional
leaders, consultants or other influential professionals. If decision makers do not act
within these constraints, they run the risk of losing legitimacy, which may outweigh
any expected efficiency gain. As such, ‘it is possible that postconscious institutional
constraints may forestall the adoption of an otherwise more efficient organizational
design’ (Roberts and Greenwood, 1998, p. 361).
Yet, it may be expected that purposive efficiency seeking choice behaviour will
continue to be present in organizations. Professional practitioners like controllers
with the responsibility for the economic rationality of the organization will have to
try and take some rational decisions on the adoption and (re)design of management
control systems. The concluding model (Table 1) in this paper may be of use to them
in deciding on the desired outcomes.
The model can be refined by using it as a basis for future explanatory case
studies providing explanations of the adoption and design of management control
Towards Horizontal Archetypes of Management Control 147
systems. Or it can also be used in a constructive approach, as advocated by Kasanen
et al. (1993), the model serving as an ‘ex ante theory’ in designing and adopting
management control systems in various organizations. These designs might be a first
step in generating ‘ex post theory’. By means of reflecting on the actual functioning
of designed and implemented management control systems in organizations, an
economic theory on the adoption of management control systems can be developed.
