Accounting logics reconfiguration of ERP (1)

Management Accounting Research 22 (2011) 181– 197

Contents lists available at ScienceDirect

Management Accounting Research

jo u rn al hom epage : w ww.elsevier.com/locate/mar

Accounting logics, reconfiguration of ERP systems and the emergence

of new accounting practices: A sociomaterial perspective

Erica L. Wagner a , Jodie Moll b , ∗ , Sue Newell c , d

a School of Business Administration, Portland State University, 631 SW Harrison Street, Portland, OR 97201, United States

b Manchester Business School, University of Manchester, Oxford Rd, Manchester M13 9PL, UK

c Department of Management, Bentley University, 175 Forest Street, Waltham, MA 02452, United States

d Warwick Business School, UK

Keywords: This paper extends our knowledge on how software-based accounting tools might work

Enterprise Resource Planning systems effectively within an organization. The empirical data that we focus on are events that

Sociomateriality University. We

Accounting logics unfolded following the introduction of a new ERP system at an Ivy League

negotiation process Reconfiguration describe a that occurred after roll-out that resulted in a reconfiguration

Legacy assets of the ERP to integrate some of the legacy functionalities that were familiar to organiza-

tional participants and which were considered by them to provide a more effective way to

manage their finances. Our contribution to the literature is not only to show the importance

of such post-roll-out modifications for creating a working information system, but also to

extend previous accounts of non-linear accounting change processes by emphasizing how

these modifications are dependent on the particular entanglement of users and technology

(the sociomaterial assemblage) rather than either features of the technology or the agency

of the humans involved. Moreover, our analysis of the case data suggests that management

accounting in particular may not be easily captured in ERP packages, even where the tech-

nology architectures are supposedly designed for a particular industry. The case data also

points to issues of affordability and the power of communities of practice as mediating the

extent to which these familiar accounting logics may become integrated within the ERP

system.

© 2011 Elsevier Ltd. All rights reserved.

1. Introduction reduce costly duplications of data ( Dillard and Yuthas, 2006; Scapens and Jazayeri, 2003 ), enable centrally stored

In the face of competitive global markets and constant information, making it easier to create different types of

innovations in technology and business models, changes financial reports ( Chapman and Chua, 2003 ) and improve

to accounting information systems are often implemented. fiduciary control ( Wagner and Newell, 2006 ). In particular, many organizations adopt accounting soft- Much of the literature in accounting has assumed ware packages (as variants of Enterprise Resource Planning such technology to be an ‘exogenous’ force of change for

[ERP] systems) to improve the transaction processing capa- accounting work routines ( Granlund and Malmi, 2002;

bilities ( Booth et al., 2000; Dechow and Mouritsen, 2005 ),

Rom and Rohde, 2006, 2007; Scapens and Jazayeri, 2003 ).

co-ordinate record keeping ( Chapman and Kihn, 2009 ),

From this view, the functionality of the system is set when

the adopting organization “flips the switches” of thousands

of embedded templates. These templates provide the orga-

∗ nization with a catalogue of standard Corresponding author. work practices –

E-mail addresses: elwagner@pdx.edu (E.L. Wagner), the so-called ‘best practice’ logic purportedly embedded

Jodie.moll@mbs.ac.uk (J. Moll), snewell@bentley.edu (S. Newell). within the product ( O’Leary, 2000; Wagner and Newell,

1044-5005/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.

doi: 10.1016/j.mar.2011.03.001

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197

2004; Wagner et al., 2006; Sia and Soh, 2007 ). Once the entanglement of users and technology (the sociomaterial

configuration options are selected this will supposedly

assemblage). In adopting this perspective, we seek to go

determine how accounting will be practiced. However,

beyond realist accounts that focus on the features of the

technology alone cannot force practice change, especially technology determining structures or social construction-

when the ‘best practice’ design logic is misaligned with the ist accounts that emphasize the agency of the humans

legacy practice logics. Where such incompatibility exists involved ( Leonardi and Barley, 2010 ). We identify how con-

resistance is often encountered ( Berente et al., 2007 ). These texts of use will differ in terms of the types and scope

kinds of problems are prominent in relation to the imple- of modification that are made in the post-roll-out phase

mentation of ERP systems because their integrated nature and thereby extend the literature on accounting change as

means that the work processes of different groups and non-linear and relational in nature (in particular, Andon

departments become more tightly coupled than was often et al., 2007; Dechow and Mouritsen, 2005; Quattrone and

the case in the legacy system environment when each Hopper, 2006 ).

department/function had its own stand-alone IT system. In this paper then, we do not reject the potential use-

Research shows that often organizations only realize the fulness of an ERP system for management accounting but

incompatibility in practice logics once the configured sys- rather argue that the value of any information system, tem is rolled-out and users find that they can no longer whether it be integrated or standalone, can only be under-

carry out their legacy work practices and so begin to resist stood by focusing on how accounting, IT and its users ( Wagner et al., 2010; Leonardi and Barley, 2008 ). This leads are entangled to produce managerial accounting practice.

many organizations into a prolonged period of negotiation With this in mind we focus on two research questions:

and may result in substantial customizations to the ERP (1) what and how are accounting logics encoded within

( Wagner and Newell, 2006 ) or other adaptations, despite ERP systems? and (2) how, in the process of users inter-

this practice being discouraged by vendors and traditional acting with the ERP and retrieving information, do new systems development theories ( Boudreau and Robey, 2005; accounting practices emerge and does the system become

Berente et al., 2007, 2008; van Fenema et al., 2007 ). For

reconfigured to support them? These questions allow us to

instance, prior field studies ( Malmi, 2001; Granlund and consider how the logics underpinning practice are sensitive

Malmi, 2002 ) have demonstrated that as ERP implemen- to local circumstances and cannot be fully specified in any

tation projects unfold, and there is realization that the “strongly determinant way” ( Suchman, 2007 , p. 53) even

system is unable to cope with the accounting demands, though ERP-architectures inscribe, in their vanilla-state,

organizations are prone to developing separate spread-

particular ideals about practice, i.e., particular practice log-

sheet solutions or specialised software such as Cognos that ics.

can provide more flexible analysis of the accounts. These We explore the questions by examining the post- studies have shown that specialized software used to cap- implementation negotiations of a troubled ERP project in

ture the specifications required of advanced management an Ivy League university. The project was at risk of being

accounting techniques, such as the balanced scorecard and abandoned because of the practice logics that were initially

activity based costing, may over time be added to ERP-type configured into its grant accounting module. The ERP was

technologies. That ERP systems might require such modi- configured to support the financial accounting needs of the

fication has led several scholars to be critical of using them central administrative function composed of professional

for management accounting purposes arguing they are too accountants. However, the integrated nature of ERP meant

complex for this type of architecture (see Rom and Rohde, that these system architectures were not sufficiently flexi-

2007 , p. 50). They suggest that ERP systems be viewed as ble for enabling faculty to manage their research project

transactional management systems that are not designed budgets in their preferred ways. We thus identify how for strategic level management. Furthermore, stand-alone the ERP-system was customized to enable the university

software systems, which can provide a more user-friendly (hereafter ‘Ivy’) to accommodate the practices of both the

and flexible basis for analysis and reporting and more ad financial accounting centre and the faculty managing their

hoc management accounting practices, may provide a less project budgets.

risky and more cost effective alternative ( Rom and Rohde, The remainder of the paper is structured in 4 sec- 2006; Hyvönen, 2003 ). But, evidence from scholars such as tions. In Section 2 we discuss the key concepts which

Chapman and Kihn (2009) suggest that managers can be inform the sociomateriality lens, enabling us to “see satisfied with systems that possess high levels of integra- in the data” the process of negotiation that occurred tion even where they do not lead directly to improvements within the case of a university in the post-implementation

in performance. In keeping with the later cadre of research, phase of their ERP project, ultimately resulting in that we extend knowledge related to how ERP technologies are system drifting to eventually support some legacy-type made to work as systems ( Chapman, 2005 ) through a non- practices even though initially these were explicitly linear process of change ( Quattrone and Hopper, 2001 ). We excluded. Section 3 outlines our research methods. Sec-

provide an analysis of a case that shows how ERP systems tion 4 presents the case description and analyzes two can be reconfigured to incorporate legacy practices and sat- grant accounting practices and the negotiation process

isfy the flexibility (i.e. precision and frequency) demanded that ensued at go-live. In Section 5 we discuss the mean-

of managerial accounting. In doing this, we demonstrate ing and implications of our analysis for implementing not only the importance of post-roll-out modifications for enterprise-wide accounting functionality within organi-

creating a working information system, but also emphasize zations. The paper concludes with suggestions for future

how these modifications are dependent on the particular research.

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197 183

2. A sociomateriality perspective of accounting 2.2. Relating notions of reconfiguration to accounting

practice

and ERP technology

2.1. Nature of sociomaterial practice In a packaged software product, the way the ERP tech-

nology is initially constituted is influenced by the particular

We adopt a practice theory perspective to yield insight configuration options selected plus any customizations into how accounting is made workable when it is entan- introduced. The accounting logics encoded in the software

gled with an ERP system. This approach is becoming

may be designed with certain problems in mind (i.e. costing

increasingly popular for its ability to offer new and inter- and performance measurement) ( Dechow and Mouritsen,

esting insights to accounting change (cf. Dechow and

2005 ) but they also involve inevitably vague categories (i.e.

Mouritsen, 2005; Jørgensen and Messner, 2010; Macintosh inventory) so that they can capture a broad range of orga-

and Quattrone, 2010; Pentland et al., 2010 ). Practice the- nizational activity ( Chapman and Chua, 2003 ). Trade-offs

ories focus on people’s everyday activities, analyzing how between organizational and technical concerns will also

they are produced and reproduced in a particular historical add to incomplete representations of the organization’s

and social context ( Levina and Vaast, 2005 ). This approach circumstances and this enables those downloading and

considers accounting as a thing that people do. This means interpreting accounting information produced by the ERP

it cannot be reduced to a set of routines that are enacted system to translate it and use it in different ways depend-

as designed across time and locations (cf. Scapens, 1994; ing on their intentions and adaptive abilities ( Bloomfield

Burns and Scapens, 2000 ). The focus on practice reflects

and Vurdubaki, 1997; Busco et al., 2007; Dechow et al.,

the idea that it is at the level of everyday activity that

2007; Dechow and Mouritsen, 2005 ). More bluntly put,

learning occurs as people and their technologies take on an ERP system cannot be understood to present informa-

particular shapes and meaning through situated use. Prac- tion that objectively reflects some reality, or as neutral

tices are considered to be in a constant state of change, technical solutions and tools ( Ciborra, 2002 , p. 68; in albeit some changes are very small. This means that there accounting see Hopwood, 1983, 1987 ). The designer 2 /users are always inconsistencies, even when people are suppos- viewpoints can be and are influenced in the process of

edly carrying out the same practice: ‘pursuing the same downloading, interpreting and using the data. It is this

thing necessarily produces something different’ ( Nicolini,

reflexivity between the ERP-system and the user that sug-

2007 , p. 894). gests that the design of an ERP-system with integrated There are many variants of practice theory 1 ; how- accounting functions is not something that stops when a

ever, here we analyze technology–human interaction

system goes live. Rather design is an on-going socioma-

from the sociomateriality practice perspective outlined

terial practice performed over time ( Suchman, 2007 , p.

by Orlikowski (2010) and Suchman (2007) . This perspec- 278).

tive does not assign agency either to persons or things

Previous studies of accounting and IT, borrowing from

but considers the social and technological to be ontologi- Ciborra (2002) , use the term ‘drift’ to describe the processes

cally inseparable, yet constitutively entangled ( Orlikowski,

by which these systems change ( Quattrone and Hopper,

2010; Orlikowski and Scott, 2008 ). It provides a way to 2001; Andon et al., 2007 ; see also Macintosh and Quattrone,

understand how meanings and materialities are inex-

2010 ). It has been argued that drift “resembles incom-

plete attempts at organizing” ( Quattrone and Hopper, 2001 , practice.

tricably related and influence the form of (accounting)

p. 428) and provides an explanation of the many ways

Importantly, this ‘relational ontology’ suggests that

in which IT systems can be altered through encounters

once one begins to look at an integrated information system with those using them such as sabotage, learning-by- like an ERP, despite it retaining an organizational identity, doing, or plain serendipity (see Ciborra, 2002 , p. 89 for multiple person/technology assemblages may be observed a more in-depth discussion). This concept enables one that help to explain how accounting can be interpreted and to understand how an organization might end up using

enacted differently by users even in the same organization legacy practices when new IT is introduced even where

( Quattrone and Hopper, 2001, 2006 ). It is for this reason particular legacy practices were purposefully ‘designed that Quattrone and Hopper (2006) refer to such systems as out’ of the IT architecture. For analytical purposes, this

being ‘heteromogeneous’. This occurs, they argue, because perspective suggests it is important to explore the ini- the logics underpinning accounting practice are sensitive tial design/configuration, who was involved/excluded, and

to local circumstances and cannot be fully specified even then to follow through to examine how the sociomaterial

though ERP-architectures inscribe, in their vanilla-state, assemblage is re-configured through a process of nego-

particular practice logics. In sociomateriality terms this tiation ( Ciborra, 2002; Orlikowski and Scott, 2008 ). The focus on how relations between humans and technologies mangling of the social and technical thus produces a dialec-

become enacted in practice is referred to as ‘performa-

tic that Pickering (1993) depicts as a series of resistances

tivity’, which helps us to understand how and why such and accommodations; here we explore this dialectic across

systems are prone to reconfiguration. time and across practice communities.

1 Ahrens (2009) suggests some of the more commonly used approaches

in accounting include governmentality, actor network theory and

2 In this paper we use the term designer since the project involved

accountability. designing a new ERP for the university market.

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197

2.3. Communities of practice, accounting logics and majority of the field work was conducted over the course

relationality of 18 months between June 1999 and August 2000 but a

series of follow-up interviews were conducted with a sub-

As Law and Singleton (2005) indicate – the nature of set of the Ivy population in 2002, 2005, and 2010 in order to

a thing depends on its nests of relationships (both those keep current with the state of the ERP and administrative

absent as well as those present). For example, with the work practices at the University.

implementation of an ERP system people may not be able to The 18-month intensive fieldwork phase corresponded

get the accounting information that they need or expected with the implementation and post-implementation of the

to be able to get. Thus, there can be an information asym- ERP project. The first author made four visits to the field

metry between people and machines limiting the scope of site where she was embedded within the ERP project envi-

interaction, and the capacity for accounting practice and ronment for eight weeks at a time (summer 99, winter

subsequent action ( Suchman, 2007 ). This view assumes 99, spring 00, and summer 00). During these intensive that ERP is an essential part of practice in organizations data collection periods the researcher attended meetings,

and by inscribing specific accounting logics sets limits on reviewed formal documentation, and conducted inter- forms of calculation. In Orlikowski’s (2005) terms these views. Years prior to conducting this study, the first author

material arrangements ‘scaffold’ social activity in particular was employed by Ivy as an accountant. This proved to

ways within practice communities. In the process of using be helpful for understanding Ivy University norms and the system, individuals and/or communities may end up communicating using local colloquialisms. The other two

questioning these logics. authors became involved only after the field study was

In the past, practice communities have been rela-

complete.

tively autonomous so that people/technology assemblages Recognizing the performative nature of some language,

have evolved in distinct ways, even if they are using the a narrative interview convention was adopted in order to

‘same’ technology. With an integrated ERP this indepen- avoid asking leading questions. 3 The narrative interview

dent evolution becomes more difficult and the multiple convention provides a time frame to structure the inter-

sociomaterial assemblages become more mutually consti- view [“Tell me what happened since we last met.”] and then

tutive. Thus, if one practice community decides that the encourages uninterrupted storytelling related to issues of

accounting logics inscribed within the ERP should change central importance to the interviewee ( Bauer, 1996 ). Rough

to better reflect the economic reality of the local entity transcripts of each interview were produced directly fol-

or its products or processes, they would need to impose lowing the day’s meetings and analyzed in terms of actors

this on others who may believe the existing practice to be and issues discussed. Recurrent themes were followed up

adequate. Unless those communities with differing prefer- in the next round of interviews in order to gather mul-

ences regarding what are the appropriate accounting logics tiple perspectives of the same situation and develop an can be decoupled, so that each could use their preferred understanding of the negotiations that were central in peo-

form of calculation, a period of negotiation would begin. ple’s minds at the time. This approach was also helpful Such decoupling is more difficult in an ERP environment, for reaching those actors who might have remained ‘silent

so the emergent sociomaterial assemblage within any

voices’ ( Star, 1991 ), because the field researcher was guided

particular practice community is restrained by other socio- not only by interviewee referrals, but also by contacting

material assemblages – one practice community cannot

allies and controversial agents whose names arose in the

evolve the system without it influencing others who may interviews. When a reference was made to a group, cause,

then resist attempts to modify the technology. Conflict is or action attributed to nonhumans such as the ERP itself,

generated because for some communities the co-evolving the researcher interviewed a delegate and reviewed tech-

ERP may enhance the sociomaterial relationship (as when nical documentation ( Pouloudi and Whitley, 2000 ). After

an accountant in the central office gets data from all dis- each phase of fieldwork, the first author “re-transcribed”

tributed departments via the ERP in one format that makes the rough transcripts in order to communicate the “spo-

his/her job of reconciliation easier) while in other cases ken features of discourse” such as tone, mood and pace of

it may distract from the relationship (as when the faculty the narration ( Riessman, 1993 ), and to fill in all missing

member can no longer get the information that s/he feels narrative.

is needed from the ERP to manage a research budget). The analysis in this paper represents a re-examination

The sociomaterial practice perspective that we describe of the original data focused on specifically contributing to

in this section is designed to provide a more inclusive view the accounting literature related to the entanglement of

of how accounting practices emerge from the complex accounting logics and ERP software. The data was first ana-

and shifting assemblages between the accounting calcula- lyzed with the intention of understanding the logics of Ivy’s

tions, the IT architecture and the people in the organization accounting practices. Next, we undertook a systematic and

( Orlikowski, 2010 ). We next describe the methodology that careful reading of the transcripts using the three theoret-

allows us to see these shifting assemblages. ical constructs introduced in section two: sociomaterial

3. Research methods

The 3 empirical sections of this paper Follow-up interviews in 2002, informed by a 2005 are and 2010 were semi-structured in that several specific questions were asked by the field researcher and

longitudinal case study (138 interviews with 53 employ- interviewees were also free to tell stories about the issues that were most

ees) (see Table 1 ) between June 1999 and April 2010. The prominent to them about recent developments.

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197 185

Table 1

Interviews conducted at Ivy League University.

Participants Community of Summer 99 Winter 99 Spring 00 Summer 00 Summer 02 Summer 05 Spring 10

practice

6 Central

3 1 5 3 2 2 2 administrative leader

9 Project managers 8 5 6 6 1 1 0

17 Project team 17 12 11 11 1 0 0 members

32 Total central 28 18 22 20 4 3 2 administration and project team

12 End user 6 5 12 5 0 0 0

9 Faculty 0 4 6 3 0 0 0

21 Total faculty and 6 9 18 8 0 0 0 faculty support

staff

53 Total 34 27 40 28 4 3 2

practice, reconfiguration, and relationality. This enabled Internal Audit department who work in collaboration with

us to identify several episodes of negotiation from which external auditors. Consequently, audit compliance is a con-

the entanglement of accounting logics and IT architectures cern to Central leadership and bears significance on any

could be observed to produce a change in the accounting decisions regarding modernization of Ivy’s administrative

practices that were used within Ivy. We focus on these in systems. Second, the mismanagement of grant funds can

the empirical sections of the paper to understand the vari- become a liability for the institution. As such, Central lead-

ances between the accounting logics inscribed in the ERP ership is motivated to decrease inconsistencies in practice

system and those that had shaped the accounting practices across faculty grants.

that had been improvised by staff. We were thereby able to From the academic perspective, when a faculty mem-

empirically examine the modification of both the software ber is awarded funds from an external grant agency there

and work practices which has not received much attention are certain reporting requirements tied to the use of the

in the accounting literature. monies. Funding bodies typically make awards based on

the initial monetary requests in the grant application

4. The case: findings and analysis where detailed categories of expenses submitted in the

application sum up to the total grant award. Principal

In this section, we first describe the organizational

Investigators (PIs) on grants usually have multiple pro- context. Next, we consider the accounting logics underpin- posals and awards ongoing at any given time meaning ning the legacy practice of commitment accounting and that their scientific activities are ‘project driven’ but the

its replacement practice time-phased budgeting. Follow-

University general ledger and chart of accounts do not sup-

ing this we examine the negotiation process that occurred port project-oriented reporting. Faculty are concerned with

leading to the emergence of new accounting practices and scientific endeavours that are enabled through the grant

a reconfigured ERP system. funding. This often means that a PI will spend grant funds

based on how much grant money remains without ever

4.1. Organizational context writing a formal plan for how the funds should be spent

over time. The various practices faculty employed in such

“Ivy” University is one of the eight Ivy League institu- reporting activities lead to inconsistencies in Ivy’s overall

tions located in the Northeast region of the United States accounting practices.

that are grouped together for their academic excellence and In this climate, Ivy embarked on an ERP project to selective private education. The governance structure of replace all administrative systems and create an ‘inte- Ivy is relatively decentralized with administration taking grated operating platform for the 21st century’. The place in the disparate academic and support departments choice of an enterprise computing model was expected to

and then consolidated and reconciled centrally in order to increase fiduciary control and decrease audit risk while

report financial status and meet regulatory concerns. funnelling more administrative responsibilities from the

The lifeblood of a research university is the receipt central administration to academic and support depart-

and effective use of research grant monies. Ivy has over

ments. The introduction of an ERP is an example of a

4000 different grants for which they have fiduciary respon- situation where Ivy leadership anticipated change around a

sibility. Grants and contracts revenue account for close ‘best practice’ ideal. Theoretically, however, we understand

to 30% of the University’s total operating budget. Within that even though an ERP-architecture would inscribe a par-

Ivy, grant activity is ‘rolled-up’ and reported alongside

ticular grant accounting logic, actual accounting practice

all other University funds for two main reasons. First,

will be sensitive to local circumstances ( Suchman, 2007 ). improper management of grant dollars can potentially Ivy contracted with a single software vendor (“Vision”)

increase the University’s risk position. Grant related activ- to adapt their government-ERP system by developing two

ities are closely scrutinized by legal counsel and Ivy’s

new modules uniquely needed for the higher education

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197

context (including managing projects associated with grant money to make it look like it used to . . . You don’t like

awards) that would be first implemented in Ivy. Vision

it? You’re out of the consensus picture. If you are more

was expected to provide a robust financial system that inclined to accept the changes and deal with them, then

would integrate accounting and budget functionality with you are in the narrow universe of people we will work to

all other administrative functions. This strong financial have consensus with.” [Technical team leader, summer

functionality drove vendor selection: 2000]

“I heavily leaned in [the] direction of wanting to go with Academic constituencies 4 and Central administrators

the strongest financial system. I thought that the largest all recognized the importance of moving away from dis-

pay-off from the project, when you really looked at it, crete silos of activity to an integrated and more transparent

ultimately would be in better financial data and the abil- accounting practice to manage institutional risk, comply

ity to do more interesting things on the clinical and with regulatory bodies, avoid litigious hazards, and act grants management side . . . [these] things seemed to me as competent fiduciaries. The Director of Financial Plan-

would drive the process to create a robust business envi- ning, a spokesperson for Ivy’s institutional perspective, ronment.” [VP for Finance and Administration, summer stressed the importance of accountability and control for

Ivy’s future:

Ivy conducted system-modernization and business-

“ . . . Certainly the motivation for having this more high

process redesign efforts within the key functional areas of powered enterprise software is that the place has financial management, human resources and payroll, and become more complex and we need better data. We need

grants and contracts administration. The financial manage- to make better decisions based on data . . . it’s a recog- ment components included general accounting, financial nition of the need to do that because we’re running a

planning and reporting, purchasing and accounts payable. huge financial behemoth. We’re the highest graded Because of the importance attached to the management financial institution in the state. All these . . . for-profit

of project funds, the capability of the system to monitor the companies aren’t rated as highly as we are. We’re ‘Triple

present state of accounts in relation to past expectations A’: ‘Triple A’ in both Moody’s and Standard & Poors. I

and future plans were key criteria in designing the sys-

mean – you know – it’s extraordinary. We have a billion

tem. It was also rationalized that an integrated approach dollars of debt. That’s a billion dollars of bond holders out

would simplify future upgrades and promised to mini-

there that are, you know, trusting us with their money. So

mize the need for expensive migration efforts later on.

it’s extraordinary, . . . ” [summer, 2000]

Institution-wide all-fund budgeting would also be possible

without time-consuming and error-prone manual recon- This financial management vision spearheaded the ERP

ciliations. Such motivations are not uncommon especially project. Ivy

needed to institute new practices that would

when seeking to retrieve accounting-based information in increase their control of financial transactions and address accountability through official statements of record. an integrated manner to create different types of reports Shift-

( Chapman and Chua, 2003 ). Ivy’s academic communi-

ing accounting practice was also seen to be important for managing future operating activities through iterative

evolved practices that were distinct, would now become cycles of planning, budgeting, review, and reporting. This led designers of the original ERP to purposefully design-out, more interdependent. This approach was expected to

ties, that had historically been relatively autonomous and

“professionalize the workforce” the by shifting types of the budgeting practice logic of ‘commitment accounting’ and design-in what questions and analyses that faculty and their support staff they considered should be the new practice logic, that would be mandated were asking with regards to their financial position: as the institutional standard, of ‘time-phasing’. From the practice perspec-

“By making a decision to go with [Vision] finan-

tive, we expect negotiations to ensue because previously

cials senior management either consciously or semi- autonomous practice communities are coupled and no

consciously – I think it was for former – was making it longer able to use their preferred form of calculation for

impossible for [Ivy] to continue doing business in frag- grants management. Indeed, this turned out to be a con-

mented silos. Like it or not, you’ve got to work with a tentious issue for the University that resulted in academics

new way of accounting. It’s integrated – it’s slower, it’s refusing to work with the financial management module.

a pain in the ass . . . and the blue-haired ladies who used In this way the Ivy study supports existing accounting lit-

to do it the old way for years decide it’s absolutely ter- erature that found ERP systems to encroach on vital but

rible – they don’t want to do it ‘cause its not Ivy’s way. idiosyncratic local knowledge leading to the creation of

But implementation is about setting up an environment. work-arounds ( Chapman and Chua, 2003; Quattrone and

You make a set of decisions – a set of changes at the top Hopper, 2006; Hyvönen et al., 2006, 2008; Granlund, 2007 ). that force change regardless of whether it’s consensus or Over a short period of time the sociomaterial assem-

not – ‘cause you say – you change their ability to do it any blage was upset and the accounting practices began to

other way. You just can’t do grant accounting like you

used to do grant accounting! Nothing you do is going to

change that – no amount of moaning is going to change 4 The academic constituency included faculty, their support staff (FSS)

it . . . They can fire [the VP], they can fire [the controller], and academic managers. FSS completed administrative work on behalf

they can fire me, but nothing’s going to change the fun- of faculty and were their first point of contact. FSS were overseen by

damental you can’t go back and you can’t spend enough academic managers who were responsible for programmatic and admin- istrative functions within a department or area.

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197 187

shift from those that had been pre-defined and configured. had to probe the PI about their scientific activities to get

In Ciborra’s (2002) terms it began to drift. Long standing them to identify, classify, and earmark as many dollars as

legacy assets were bolted-onto the ERP system in order

possible from the current instalment of the grant. While

to create a working information system that was used these conversations were held with all PIs the granularity

by both academic and administrative communities. The and frequency of these discussions was a locally negotiated

notion that even when doing the ‘same thing’ differences outcome. Individual temperaments, available time, depart-

in practice are inevitable ( Nicolini, 2007 ) foreshadows

mental norms, professional working relationships, and the

the sociomaterial negotiations involved. We identify the stage and nature of research, all influenced the extent to

post-implementation phase as one of re-figuring the grant which future expenses were identified, entered through

accounting practices and as one that was reflective of the the DAS interface, and classified as commitments that were

way in which the users and technology were constitutively expected to occur during a specific grant period. The ad hoc

entangled. The relationality between different fields-of-

nature of grant accounting makes it evident why a practice

practice at Ivy helps us to understand how a working theory lens is so important because it allows us to see how

information system was created that involved socioma- small variations in process are inherent to conducting work

terial accommodations that included the support of some – even when it is supposedly the same practice, in this case

legacy-type practices (commitment accounting) alongside monitoring grant dollars. The project management process

the system-supported ‘best practices’ (time-phased bud-

links the PI back to the funding body reminding her of what

geting). she said she would do to achieve her academic goals and

This organizational context sets the stage for investigat- while each PI has reporting obligations, how she gets to this

ing people’s everyday activities and analyzing how these stage will vary ( Nicolini, 2007 ).

activities are produced and reproduced in a particular his- PIs viewed the pending commitments as money that

torical and social context ( Levina and Vaast, 2005 ). Each had been spent therefore it was expected that the balance

construct, along with the negotiation process explaining of the grant reflected this. This amount was recorded in

how new accounting practices emerged, is described in a PI report. This PI report was different from the finan- Table 2 .

cial statement of record aka “the Greens” which were colour-coded financial statements reporting only the actual

4.2. Sociomaterial practice: the logics of commitment income and expenses that hit the general ledger in that

accounting and time phased budgeting financial cycle. From time to time, administrators recon-

ciled the Green reports against the DAS system and the

Long before the conceptualization of ERP moderniza- PI Report to make sure the commitment accounting sys-

tion at Ivy, persons and things were figured together in tem was pulling the correct numbers from the mainframe

a particular way in order to manage scientific activities financial system. Together these material arrangements funded through grant monies. This sociomaterial config- scaffold commitment accounting grant management prac-

uration was formally known as commitment accounting tices and in turn, the viewpoints of users were influenced

where people and their technologies took on particular in the process of downloading, interpreting, and using this

shapes and meaning through situated use. Faculty and their accounting data in a reflexive manner. The PIs had “learned

Support Staff (FSS) were largely responsible for adjusting to see” in a particular way that was supported by the socio-

all discrepancies between what the PI actually spent in

material arrangements of commitment accounting. Each PI

a particular budgetary category and what had been bud- was involved in the practice of creating a “close enough

geted. This reconciliation process involved handwritten

match” of actual research activities to the artefact of the

notes/emails to administrators from faculty about their

financial system and its statement of record. Goodwin calls

spending activities, printed reports from the Mainframe

this process of learning the meta-data of work practice

financial system, PI Reports from the Distributed Account- part of developing a professional vision (within Suchman,

ing System (DAS), and MS Excel. The DAS was written in 2009 ) – in this instance the vision of being a scientific prin-

Focus programming language by faculty and staff in the cipal investigator. The nature of grant accounting, thus, late 1970s and had a very basic interface allowing users depended on the nest of relations between the people and

to enter and download data from the mainframe financial things involved in the accounting practice.

chart of accounts, match them with detailed categories that For central administrators, this form of reporting was

the faculty decided and then reconcile that balance with seen to be indicative of the poor financial planning prac-

what faculty was committed to spend. tices of faculty. This was made no more obvious to them

PIs told administrators how much money to set aside than when they considered the spending patterns of the

for project activities. They referred to these obligations PIs who were in the habit of going on sprees as the grant

as ‘commitments’. Administrators classified these com- completion dates began to close in. The PIs wanted to try

mitments into one of two conceptual categories: pending and reach a zero balance to ensure that funding would not

actuals and future plans. The former represent expenses

be returned to the funding body. Around the time that Ivy

that had been incurred but for reasons of timing, had

had decided to replace its legacy systems this practice had

not yet ‘hit’ Ivy’s general ledger and were excluded from become so problematic that central accounting had begun

that cycle’s mainframe financial reports; whereas commit- to ask questions about the large accounting transfers that

ting for future plans took into account activities that PIs were recorded late in the life of externally funded grants.

expected to occur at a later time, but before the completion The central accountants were concerned that auditors of the grant. To budget for these categories administrators would see inconsistencies (e.g. spending $4000 a month

E.L. Wagner et al. / Management Accounting Research 22 (2011) 181– 197

Table 2

Key constructs for explaining how new accounting practices emerge.

Theoretical construct Illustration

Sociomaterial Practice – technical agencies are tied together with Many local grant management practices existed that were

those involved in the practice; an ERP sets up a particular accommodated by the commitment accounting sociomaterial

sociomaterial configuration. configuration but were undermined by the new time-phased

budgeting configuration.

Reconfiguration – focuses on how persons and things might be figured Eventually the project team responded to the demands of faculty and

together differently when resistance is encountered to a new worked with them to re-configure the ERP to accommodate faculty

sociomaterial arrangement following ERP implementation. ways of working and set up various support structures.

Relationality – multiple practice communities exist across the field of During design, the ERP-project team configured the ERP to meet their

accounting within an organization; different communities will have needs. Following implementation this met with faculty resistance;

more or less power to influence the sociomaterial practices faculty were powerful voices within this environment and so were

surrounding an ERP configuration/reconfiguration and this can able to force others to listen to their needs and interests and so effect

change over time. the reconfiguration.

for 10 months and the last 60 days charged $85,000) in the sum of each category at the institutional level. The capac-

spending process making the transfers suspect in terms of ity for action was quite different from the sociomaterial

reflecting actual scientific practice. Next, we consider the assemblage of commitment accounting. Time-phased bud-

proposed integration of management accounting activities geting begins when faculty and staff meet at the beginning

through time-phased budgeting. of the fiscal year for an extended financial budgeting ses-

As Suchman notes (2009) “professions constitute those sion. This provides the opportunity for all projects to be