Normative accounting literature has treated technology
as an ‘absent presence’, neglecting to consider the materiality
that these types of architectures provide in the practice
of accounting. When materiality is taken into account, a
unidirectional relationship is theorized whereby the focus
has been to understand the impact of ERP-systems on
accounting through either privileging IT in explaining the
form of accounting, or the reverse: emphasizing management
accounting’s impact on IT development/modification.
The sociomaterial perspective demonstrates how an ERP
can be enacted in different ways as it connects with practices
of different communities of users (Law and Singleton,
2005). Nevertheless, this differential enactment has limits
created by the material configuration. In Ivy, the ‘best
practice’ was defined by professional accountants from the
central administration who encoded accounting logics that
suited their particular interests related to institutional risk.
In doing this, the needs of the distributed faculty who had
not been professionally trained to manage their project
budgets and moreover, had different interests compared
to the central accountants (i.e., to spend all their budget in
the interests of their research projects vs. to reduce institutional
risk) were ignored. This led to post-implementation
negotiations, and our analysis shows how an accounting
system can be made functional as different communities
resist and accommodate the ‘best practices’ (Pickering,
1993) until eventually a reconfigured system is created that
includes valued functionality that allows different communities
to accommodate their needs.
Demonstrating this recursive relationship between
accounting practices, technologies and its users contributes
to the recent call in the management accounting literature
for further longitudinal studies of accounting that examine
in-depth the bidirectional nature of the relationship
between management accounting and ERP technologies
(see Rom and Rohde, 2007; Luft and Shields, 2003). The
detailed Ivy case contributes to our understanding of
how working accounting information systems are created
around an ERP. We have shown that accounting is not an
object in its own right but is better seen as a set of practices
that are scaffolded by material objects (e.g., ERP) that are
used by diverse communities to perform different tasks.
Quattrone and Hopper (2006) thus describe ERP as being
‘heteromogeneous’ in the sense that they appear to be
homogeneous precisely because they are actually heterogeneous,
providing a space whereby users can experiment
with them and create different sociomaterial assemblages.
However, given the integrated nature of an ERP, the
freedom to experiment through practice may be more limited
than is suggested by this concept. Performability is
restricted by the ‘best practice’ logic that has been initially
configured in the ERP. In Ivy this set the stage for negotiations
which lead to reconfigurations that accommodated
diverse practices, with these diverse practices and interests
exposed rather than concealed by the ERP. In accommodating
these diverse interests the ERP was reconfigured
in such a way that the legacy management accounting
practice was reinstated. At face value this seems to reinforce
the findings by Granlund and Malmi (2002) that ERPs
have little impact on management accounting practice and
answer their question as to whether this minimal impact
was because practice change lags implementation (they
looked only at companies who had very recently introduced
an ERP) in the negative. However, this obscures the