bundling management control innovations

7/29/2019 Bundling Management Control Innovations

1/32

Bundling management controlinnovations

A field study of organisational experimentingwith total quality management and the

balanced scorecard

Sven ModellThe University of Manchester, Manchester, UK

Abstract

Purpose The purpose of the paper is to examine how organisational experimenting with totalquality management (TQM) and the balanced scorecard affects the bundling of design characteristics

associated with these innovations in a Swedish central government agency.Design/methodology/approach The paper adopts an exploratory, longitudinal case studyapproach which builds on and extends the management fad and fashion literature.

Findings While both innovations encountered considerable implementation problems, theycontinued to exercise a lingering influence on the new performance management system emerging inthe focal organisation. The original adoption of the two innovations can be explained from atraditional management fashion perspective. However, the subsequent development of performancemanagement features a more complex mix of explanatory factors and highlights how the bundlingphenomenon is entangled with managerial learning processes. This resulted in a less linear trajectoryof change than that predicted by prior research on the notion of bundling.

Research implications The paper contributes to the literature on management fads and fashionsby refining its conception of the role of managers of adopting organisations and organisationaladoption, implementation and rejection of innovations.

Originality/value The paper constitutes a first attempt to examine in greater detail howorganisational experimenting with contemporary management control innovations affects the processof bundling in an individual organisation.

Keywords Balanced scorecard, Fashion industry, Management accounting, Total quality management

Paper type Research paper

1. IntroductionThe past two decades have seen a wave of innovations in management accounting andcontrol practices, notably epitomised by such notions as activity based costing/activitybased management and the balanced scorecard to mention but a few novel techniques.The adoption and diffusion of such techniques have been extensively examined in the

private (e.g. Bjrnenak, 1997; Brown et al., 2004; Gosselin, 1997; Malmi, 1999;Speckbacher et al., 2003) as well as the public sector (e.g. Chan, 2004; Jackson and

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0951-3574.htm

Previous versions of this paper were presented at the annual conference of the Accounting andFinance Association of Australia and New Zealand, the Gold Coast, and the 5th Asia PacificInterdisciplinary Research in Accounting Conference, Auckland. The author is grateful to PaulAndon, Christian Ax, Jane Baxter, Rob Chenhall, Anders Gronlund, Sue Llewellyn, HabibMahama, Hanne Nrreklit and Lee Parker for insightful comments. The research was funded bythe Swedish Research Council.

Bundling controlinnovations

59

Accounting, Auditing &

Accountability Journal

Vol. 22 No. 1, 2009

pp. 59-90

q Emerald Group Publishing Limited

0951-3574

DOI 10.1108/09513570910923015


2/32

Lapsley, 2003; Lapsley and Wright, 2004; Northcott and France, 2005). An importantsource of inspiration for much of this research has been the literature on managementfads and fashions (e.g. Abrahamson, 1991, 1996) suggesting that managerialinnovations are popularised as a set of conceptual components, which are more or less

successfully adopted by organisations. A closely related research topic is that of howconceptual components, or design characteristics associated with differentmanagement accounting and control innovations are bundled together asinnovations evolve (Ax and Bjrnenak, 2005, 2007; Bjrnenak and Olson, 1999). Thenotion of bundling implies that elements of different innovations are re-combinedinto new innovation packages in the process of adapting these to a particular context(Ax and Bjrnenak, 2007). It is thus a potentially important analytical category forunderstanding why and how fashionable innovations become implicated inorganisational control practices.

From a traditional fad/fashion perspective, bundling has primarily been describedas a supply-driven phenomenon as it is used by propagators of particular innovations(e.g. consultants, trade associations) to increase their attractiveness to potentialadopters (Ax and Bjrnenak, 2005, 2007). However, this perspective has recently beencriticised for neglecting the complexities of change associated with local adaptation ofinnovations in a particular organisational setting (Clark, 2004; Czarniawska and Sevon,2004; Jones and Dugdale, 2002; Mueller and Carter, 2005). The majority of empiricalresearch following the fad and fashion perspectives has focused on broad diffusion andadoption patterns and relies heavily on survey data (e.g. Bjrnenak, 1997; Lapsley andWright, 2004; Malmi, 1999) or secondary sources, such as references to particularinnovations in professional or trade publications (e.g. Abrahamson and Fairchild, 1999;Ax and Bjrnenak, 2005; Carson et al., 2000; Giroux, 2006; Mazza and Alvarez, 2000).Moreover, few of these studies have examined the connections between differentinnovations in any greater detail. Hence we have very limited knowledge of how

management accounting and control innovations interact as adopting organisationsimplement these and how this affects the need to re-conceptualise the bundlingphenomenon.

In the present paper, we address this gap in the management accounting literatureinspired by the fad/fashion perspective by viewing the bundling phenomenon as anoutcome of organisational experimenting with novel accounting and controltechniques over an extended period of time (Andon et al., 2007). Focusing on twoinnovations that have gained increasing popularity over the past two decades namely total quality management (TQM) and the balanced scorecard we examinehow these have affected the development of control practices in a Swedish centralgovernment agency. While drawing on the literature on management fads andfashions, we explicate how this needs to be extended to capture the complexity of the

bundling phenomenon.We start by outlining how the development of TQM and the balanced scorecard can

be understood from a fad/fashion perspective and elaborate on the critique informingour analysis of the notion of bundling. We then account for the research methodapplied. The empirical part of the paper starts with an overview of the supply ofmanagement innovations in Swedish central government before examining how suchinnovations have evolved in the case organisation. The concluding section summarisesour main findings and implications for future research.

AAAJ22,1

60


3/32

2. Conceptual frameworkThe literature on management fads and fashions seeks to explain diffusion processesbased on the assumption that organisational adoption of innovations is determined byimitation rather than rational choice (Newell et al., 2001). Drawing heavily on DiMaggio

and Powell (1983), Abrahamson (1991) introduced the fad and fashion perspectives ascomplements to the efficient choice perspective, suggesting that adoption results fromreasoned cost/benefit-centred decisions, and the forced selection perspective, accordingto which adoption is driven by the need to comply with regulatory powers. The fad andfashion perspectives differ, however, in that the former sees similar organisationswithin a particular group of adopters as the main source of imitation while paying littleattention to the origins of innovations. On the other hand, the fashion perspectiveemphasises the active involvement of fashion-setters, such as influential consultingfirms and other expert powers, typically located outside the group of adoptingorganisations in the creation of management fashions (see also Abrahamson, 1996).This has led a number of authors to deepen the analysis of the strategies applied bypromulgators of management fashions in disseminating innovations (e.g. Benders andvan Veen, 2001; Collins, 2000; Jackson, 2001; Kieser, 1997).

Against this backdrop, it is useful to start a discussion of how TQM and thebalanced scorecard may be bundled by examining the relationships between theseinnovations as they have been portrayed by key propagators. This leads to an interestin the rhetoric invoked as well as the constitutive design elements of these innovations(Ax and Bjrnenak, 2007).

TQM emerged as an increasingly fashionable management innovation in response tothe lacking competitiveness of US manufacturing industry in the 1980s and especiallythe perceived superiority of Japanese firms in delivering high-quality products inaccordance with customer demands while achieving operating efficiency (Giroux, 2006).Thus, the rhetoric surrounding the promulgation of TQM tends to be loaded with

references to the need for management practices fostering customer orientation andsatisfaction (Dean and Bowen, 1994; Mouritsen, 1997). The paramount role of thecustomer is also reflected in studies teasing out the core design characteristics of TQM asspecified by key propagators (Anderson et al., 1994; Hackman and Wageman, 1995).Based on the views of an expert panel, Anderson et al. (1994) condensed the 14 TQMprinciples prescribed by Deming (1986) into a smaller set of key concepts emphasisingthe role of visionary leadership for stimulating cooperation and learning as vehicles ofprocess management, continuous improvement and employee fulfilment with theultimate objective of enhancing customer satisfaction. Similarly, Hackman andWageman (1995) revisited a number of classical TQM writings and identified thenecessity of meeting customer needs as the primary motivation for TQM. Internalorganisational practices, such as continuous improvement efforts, cross-functional

teamwork and the development of process-orientated management techniques, wereportrayed as subsidiary to this overriding priority.

While achievement of customer satisfaction or the fulfilment of customer needs aregenerally perceived as an overriding design priority, it is also possible to identify someconsensus about the management control techniques required for the realisation ofTQM. The development of an organisational climate conducive to cross-functionalcooperation and process-orientated management is typically considered to necessitatelateral, rather than hierarchically orientated control practices (see, e.g. Chenhall, 2008;


61


4/32

Munro and Hatherly, 1993). Popular conceptualisations of TQM emphasise the need formechanisms supporting lateral control, such as the empowerment of managers withcross-functional process responsibilities, team-based rewards and use of non-financialgoals and performance measures indicating how internal processes and performance

improvements contribute to customer satisfaction (see Hackman and Wageman, 1995;Johnson, 1992).

Similar to TQM, the origins of the balanced scorecard can also be traced to thegrowing concerns with the decline of US manufacturing industry in the 1980s, whichwas arguably exacerbated by the over-reliance on short-term, historically focused andpredominantly financial performance indicators (Johnson and Kaplan, 1987). However,while recognizing the importance of satisfying customers and achieving operatingefficiency the rhetoric of leading balanced scorecard propagators, such as Kaplan andNorton (1996), emphasises the alignment of performance management with theoverriding objective of maximising shareholder value (Ax and Bjrnenak, 2005).Another key element of the rhetoric surrounding the balanced scorecard is that suchlong-term, strategic goal alignment is achievable by managing the causal relationshipsbetween underlying performance drivers (Nrreklit, 2000, 2003). This is accomplishedby disaggregating the strategic vision and objectives of companies to a set of causallylinked indicators representing financial performance, customers views, internalbusiness processes and learning aspects stimulating long-term growth of the firm(Kaplan and Norton, 1996, 2001)[1].

In contrast to popular conceptualisations of TQM, emphasising the need for lateralcontrol, the balanced scorecard has a predominantly hierarchical orientation. This isnotably manifested by the suggestions that it constitutes a useful mechanism fordisaggregating organisation-wide objectives and targets into personal scorecards formanagers and employees at various hierarchical levels and that strategic goal alignmentis reinforced by cascading indicators up and down the organisational hierarchy

(Kaplan and Norton, 1996, 2001; see also Bukh and Malmi, 2005). While references toTQM in Kaplan and Nortons (1996, 2001) earlier works were rather tangential, morerecent advances (e.g. Kaplan and Norton, 2004) deal more extensively with it and portrayit as complementary to the balanced scorecard. However, this argument mainly rests onthe suggestion that indicators incorporated in the balanced scorecard may support theprocess improvement efforts embedded in TQM while the more fundamental challengesof reconciling hierarchical and lateral logics of control are largely eschewed (Chenhall,2008; Hackman and Wageman, 1995)[2]. Although the possibilities of linkingperformance evaluation and rewards to team performance are occasionally recognised(e.g. Kaplan and Norton, 2001), more critical assessments suggest that the balancedscorecard primarily reflects individualistic values pivoting on the notions of faircontracts and unequivocal accountability relationships with the individual as the key

organisational agent and object of control (Bourguignon et al., 2004; Nrreklit et al., 2006).Table I summarises this brief review of the relationships between TQM and the

balanced scorecard with particular reference to how overriding and subsidiary designpriorities, reflecting the fundamental motifs of these innovations, are linked to keymanagement control features. However, merely relying on popular conceptualisations offashionable management innovations is insufficient for understanding how these may bebundled. While an inspection of Table I suggests that TQM and the balanced scorecardrepresent largely incompatible logics of control, empirical research indicates that reliance

AAAJ22,1

62


5/32

on TQM may indeed be conducive to the adoption of balanced scorecard-likeperformance management systems (Hoque and Alam, 1999: Malmi, 2001). This cautionsagainst a conception of these innovations as monolithic wholesale solutions. Althoughthe communication between key propagators of TQM and the balanced scorecard seemsto have been rather limited (see also Chenhall, 2008), prior research shows that theseinnovations are both nebulous concepts, which may be subject to considerablere-construction in practice (see, e.g. Aidemark, 2001; Edenius and Hasselbladh, 2002;Lozeau et al., 2002; Modell, 2004; Mouritsen, 1997; Zbaracki, 1998).

This observation draws attention to the considerable ambiguity often inherent infashionable management innovations (Benders and van Veen, 2001; Giroux, 2006; Kieser,1997). Ax and Bjrnenak (2005, 2007) argued that the interpretative viability stemming

from this ambiguity constitutes a precondition for successful bundling of innovations.By utilising the scope for multifaceted interpretations, suppliers of innovations mayreconfigure these to make them compatible with a particular social setting or reduceconflicts with other innovations. Examples of this can be found in the dissemination ofthe balanced scorecard to public sector organisations and its subsequent transformationfrom a shareholder- to a stakeholder-focused model with less emphasis on financialperformance as an overriding design priority (Modell, 2004). Conversely, excessivestereotyping of the ideas underpinning various innovations can hamper their prospectsof influencing evolving management practices (Parker and Ritson, 2005).

Interpretative viability may be a valuable resource in the dissemination ofinnovations as it enables suppliers of management innovations to increase market size(Benders and van Veen, 2001). However, prior research into its role in the diffusion of

fashionable management innovations has been hampered by a lack of attention to thedynamics associated with the demand for such innovations (Giroux, 2006). Thislimitation can be traced to the tendency to focus on seminal fashion-setters as lonecreative geniuses (Clark, 2001, p. 1660), deliberately and skilfully developing novelmanagement techniques, while downplaying the active involvement of managers ofadopting organisations in shaping innovations (see also Clark, 2004). This suggests aninteresting paradox in the literature on management fads and fashions: inde-emphasising rational choice explanations of adoption it has tended to portray

TQM Balanced scorecard

Overriding design priority Customer satisfaction/fulfilmentof customer needs

Shareholder value/financialperformance

Subsidiary design priorities Continuous improvement Eliciting customer viewsProcess orientation Internal process improvementCooperation and learning Learning and growth

Key management controlfeatures

Lateral orientation: Hierarchical orientation:Cross-functional processresponsibilities

Strategic goal alignment

Team-based rewards Individualistic control andaccountability

Non-financial goals andperformance measures

Cascading of causally linkedfinancial and non-financialperformance measures

Table I.Core design

characteristics of popularconceptualisations of

TQM and the balancedscorecard


63


6/32

recipients of innovations, such as managers of adopting organisations as passive,unreflective actors while stressing the inherently intentional and rational manoeuvringof promulgators.

Consequently, the fad and fashion perspectives provide limited insights into the

complex interplay between the suppliers and adopters of innovations in the process ofbundling. In particular, the action repertoires of managers of adopting organisationsand their role in re-constructing and linking innovations remain under-researched. Wenow turn to examine these issues in greater detail.

3. Research methodThe following empirical study is set in the context of Swedish central government and,more specifically, in a central government agency the Swedish National Board ofStudent Aid (SNBSA) experimenting with TQM as well as the balanced scorecard.This case study is part of a larger research programme examining how recent centralgovernment reforms have influenced the development of performance management

and control practices of central government agencies (see Modell et al., 2007; Modelland Gronlund, 2007). This research programme entailed extensive interactions withsenior civil servants and managerial staff at different levels of a broad range of centralgovernment agencies. In combination with archival data sources and prior empiricalstudies, this research provides valuable insights into the context of central governmentreforms.

While being part of this broader, field-based data collection, the SNBSA wasselected for deeper analysis for the present study as it has experimented with multiplemanagement innovations over a relatively short time-span as well as adapting tochanges in the Governments reform agenda. This offers a promising opportunity toexamine the complexities associated with the bundling phenomenon in an individualorganisation. However, it should be noted that the importance of the concept of

bundling as a metaphor for understanding the process of change in this particular caseemerged inductively during the process of data collection and analysis and became theobject of deeper probing as the research progressed (Ahrens and Chapman, 2006).Overall, the case study was of an exploratory, non-interventionist nature (Lukka, 2005).

Data collection in the SNBSA took place between 2004 and 2007 and primarilycomprised semi-structured interviews, group discussions and a number of secondarydata sources. At the outset of the study, the organisation was in the midst of devising anew performance management system. Our initial contacts with the organisationindicated that a small number of headquarters staff had been instrumental in thesechanges and prior attempts to implement TQM- and balanced scorecard-inspiredcontrol techniques. Our first round of interviews (in October 2004) aimed at probinginto their experiences of these innovations and were relatively open-ended to allow

interviewees to elaborate freely on what they saw as the driving forces behind adoptionof various innovations, their benefits and the problems encountered duringimplementation. Similarly, in early 2005, the managers of three regional offices ofvarying size and geographical location were interviewed to examine the widerexperiences of the innovations concerned at the operating level. Later that year, we alsointerviewed the incoming Director-General and the government official in charge ofmonitoring the performance of the SNBSA to provide complementary perspectives onthe wider governance and reform context of the organization.

AAAJ22,1

64


7/32

A preliminary analysis of these interviews and documents provided by the SNBSA(e.g. internal planning documents, memos and presentation material describing itsexperiences of the innovations) revealed some interesting connections between theorganisations experimenting with different management techniques and its ongoing

development of strategic planning and performance management practices. Toexamine these connections in greater detail, additional interviews were conducted in2006 and 2007 with individuals extensively involved in past and ongoing changeinitiatives at various levels. These interviews were more focused to probe into issuesemerging as particularly critical in the earlier research phases (a sample of interviewquestions used in the later research phases is given in the Appendix). Hence weintroduced a longitudinal element in data collection allowing us to gradually sharpenour focus on issues of key interest and gain a deeper understanding of emergingcontrol practices. The distribution of interviewees across different interview phases isgiven in Table II. Following this iterative interview process, five of the 16 intervieweeswere interviewed on more than one occasion.

The interviews typically lasted between one and two hours. To stimulate an open

dialogue and avoid bias related to potentially controversial or sensitive issues werefrained from taping the interviews[3]. However, extensive notes were taken andtranscribed immediately after each interview. To compensate for the lack of taperecordings we adopted a systematic approach of respondent validation by returning ourtranscripts to the interviewees and asking for feedback. While this proved useful forstimulating further dialogue and probing deeper into certain issues (Bloor, 1978), we are

IntervieweeInterviewed

2004/2005Interviewed

2006Interviewed

2007 Total

Ministerial levelCivil servant in charge of performance monitoring 1

Headquarters staffDirector general 2Chief financial officer 1 1Chief financial officer 2

a

1Strategist 1 3Strategist 2 2Strategist 3 1Controller 1Staff specialist 1 1Staff specialist 2 1Staff specialist 3 1Staff specialist 4 1Manager, student grants and loans function 1

Office managersOffice manager 1 3Office manager 2 2Office manager 3 1Total 12 5 6 23

Note: a Chief Financial Officer 2 succeeded Chief Financial Officer 1 in 2005

Table II.Distribution of

interviewees acrossdifferent interview

phases


65


8/32

unable to use longer quotes from the interviews but mainly reproduce shorter statementsand expressions for illustrative purposes. The interview transcripts and additionalfeedback were analysed based on a grid technique (Yin, 1984) whereby data were codedand explanatory patterns were mapped across various parts of the organisation.

As a final means of validation and probing into emerging interpretations, weorganised feedback sessions, based on preliminary case study narratives and oralpresentations, with key informants on two occasions (April 2006 and October 2007).Especially the latter of these sessions, taking the form of a group discussion with sixheadquarters staff, proved valuable for shedding additional light on the process ofbundling and the unfolding implementation of the new performance managementsystem. The meaning of the bundling metaphor was presented to the participants, whowere subsequently encouraged to reflect on past and ongoing changes in light of this,while notes were taken and later transcribed. Such member checks constitute a crucialtechnique for assessing the credibility of researcher interpretations in qualitativeresearch (Lincoln and Guba, 1985) and helped enrich our analysis.

4. The supply of management innovations in Swedish central governmentSwedish central government agencies have been subject to an increasing element ofmanagement by objectives or managing for results over the past decades[4].Following budgetary reforms aimed at deregulation in the late 1980s, governmentcontrol of agencies was gradually re-orientated towards a heavier emphasis onperformance evaluation although the annual budgetary cycle and other planningactivities still have an important influence on the structuring of evaluation procedures.However, even though managing for results has been adopted as an officialgovernance and control mechanism throughout central government, several studiessuggest that it co-exists with a lingering element of detailed regulations of operations(e.g. Holmblad Brunsson, 2002; National Financial Management Authority, 1999a, b;

Sundstrom, 2003). Moreover, the meaning of managing for results remains vaguelydefined as it has evolved on an incremental basis rather than as a result of a single,clearly identifiable and sweeping reform initiative (Modell and Gronlund, 2007).

The gradual evolution of managing for results in the 1990s was paralleled by thediffusion of other management techniques among central government agencies. TQMand the balanced scorecard were two of the most salient innovations in this respect,displaying a varying degree of consistency with the ethos of managing for results.We briefly review the origins and development of these innovations in Swedish centralgovernment.

Total quality managementTQM-inspired practices began to diffuse across central government agencies on a

larger scale in the mid-1990s (Modell et al., 2007). A key source of inspiration in thisrespect was the Swedish Institute for Quality Insurance (SIQ). The SIQ is anon-governmental organisation dedicated to promoting modern quality managementpractices within the private as well as the public sectors. To this end, the organisationundertakes a range of promotional activities, including conferences, training, qualitycontests and the development of quality enhancement tools. The quality managementpractices emerging in several central government agencies since the mid-1990s havedrawn on these tools, especially the SIQ Model for Performance Excellence (Modell

AAAJ22,1

66


9/32

et al., 2007). This model was, in turn, heavily inspired by tools developed by prominentstandard-setters in the international TQM movement, such as the Malcolm BaldrigeAward in the USA and the European Foundation for Quality Management (EFQM).The model bears testimony to key TQM principles such as a heavy emphasis on

customer and process orientation, employee empowerment and continuousimprovement. These aspects are funnelled into a measurement index for assessingorganisational quality enhancement efforts, covering such areas as leadership,information and analysis, strategic planning, staff motivation and development,operating process orientation and results, and customer satisfaction. Customersatisfaction is currently weighted as the single most important assessment criterionaccounting for 30 per cent of the total index score (SIQ, 2006).

The SIQ has taken pains to portray the Model for Performance Excellence as a flexiblemanagement tool that is easily adaptable to a wide range of organisational settings andother management techniques. For example, the model is arguably compatible with otherinfluential quality standards, such as ISO 9001. Similarly, the model does not onlyemphasise quality aspects but also the achievement of productivity and financial resultsand stresses the importance of multidimensional performance measurement (SIQ, 2006).This flexibility is also mirrored by the considerable variations in the degree to whichTQM principles have permeated central government agencies. More broadly basedexaminations reveal that several agencies have implemented process-orientatedorganisational structures as a complement to hierarchical management controlpractices although few of them have conformed completely to TQM principles such asthose advocated by the SIQ (National Council for Quality and Development, 2003a).Deeper case studies suggest that the adoption of the models promoted by the SIQencountered considerable implementation problems and resistance but still left importantvestiges in some agencies, such as a growing emphasis on systematic quality control andcustomer orientation (Modell et al., 2007; Quist, 2003; Wiesel, 2006).

Whereas TQM-inspired practices started to diffuse across central governmentagencies without much government intervention, the notion of quality managementhas also been entangled with reforms emphasising the need for enhanced citizenorientation since the late 1990s. Following the Government Bill setting out this reformagenda (Swedish Parliament, 1997/1998) a new advisory agency the National Councilfor Quality and Development was formed in 1999. The agency initiated a search forsuitable quality management models for central government agencies but quicklyconcluded that it was preferable to continue developing models already emerging inpractice, such as the SIQ Model for Performance Excellence (Modell et al., 2007).

Following this initial search, the National Council for Quality and Development hasdocumented emerging quality management practices in central government agenciesand prescribed ways of further developing these in a series of reports (National Council

for Quality and Development, 2003a, b, 2004). Interestingly, however, emergingpractices are here re-cast to emphasise process orientation as the main metaphor ofchange. The agencys prescriptions are heavily influenced by the TQM literature andkey concepts such as customer orientation occupy an elevated position. Yet, centralgovernment agencies mainly seem to have approached quality management under thebanner of process orientation as a contrast to the hierarchical emphasis of themanaging for results system. The National Council for Quality and Development hasalso expended considerable efforts on explicating the importance of adopting a


67


10/32

process-orientated perspective to avoid alleged dysfunctions of hierarchical control,such as a lack of attention to customer needs (see National Council for Quality andDevelopment, 2004). However, little attention has been paid to the development ofperformance indicators to support quality enhancement.

The National Council for Quality and Development has also cautioned against thelack of inter-organisational process orientation stemming from the Governmentsemphasis on the objectives of single agencies and has called for enhanced collaborationbetween agencies to better serve citizen and user needs (National Council for Qualityand Development, 2004). However, the Government has been reluctant to endorse theappeals for far-reaching process orientation. It has rather insisted that agencies shouldconform to the hierarchical managing for results system although it has recentlyemphasised the need to strengthen the systems focus on citizen- or user-orientatedindicators such as customer satisfaction (Modell et al., 2007).

The balanced scorecard

Similar to the emergence of TQM-inspired practices, the balanced scorecard has diffusedacross central government agencies without much direct government intervention butthrough extensive involvement of private-sector consultants and other authoritativepromulgators. As early as 1996, the National Audit Office issued a report describing howthe balanced scorecard might be used to support managing for results within centralgovernment agencies (National Audit Office, 1996). The use of the balanced scorecard byvarious agencies has since been repeatedly documented by the National FinancialManagement Authority (2000, 2006)[5]. A few common themes emerge from thesedocuments and prior empirical research in central government agencies.

First, the balanced scorecard is portrayed as largely compatible with the ethos ofmanaging for results. While some agencies have adopted the idea of linking thebalanced scorecard to strategic visions and objectives, others have aligned it to

politically established objectives embedded in annual appropriation directives. Mostagencies also disaggregate the objectives, targets and performance indicators to lowerhierarchical levels although some only use the tool at an organisation-wide level andmake limited use of it for management control. However, few, if any agencies use thebalanced scorecard for distributing performance-based rewards (see also Carmona andGronlund, 2003; Wiesel, 2006).

Second, similar to many private sector firms in Sweden (see Ax and Bjrnenak,2005) most agencies have complemented the traditional categorisation of performanceindicators into four perspectives with an explicit personnel or human resourceperspective and have transformed the balanced scorecard to encompass a widerstakeholder focus (see also Carmona and Gronlund, 2003). While the National FinancialManagement Authority recognises this as an important divergence from the original

balanced scorecard concept, it endorses it as a development compatible with theSwedish tradition of emphasising employee interests more strongly than in the USA(National Financial Management Authority, 2000).

Finally, the notion of linking performance measures into causal chains of leadingand lagging indicators seems to be poorly understood. Although the National FinancialManagement Authority emphasises the integration of various perspectives into acoherent and holistic view of organisational performance, none of the agencies beingdescribed have made any great strides towards this end. The balanced scorecard is

AAAJ22,1

68


11/32

primarily used as a means of categorising and linking objectives, targets andperformance indicators within each perspective but without much consideration ofhow various perspectives interrelate. The possibilities of continuously tracing causalrelationships between various perspectives are also reduced by the lack of coordinated

timing of the compilation of performance information in some agencies. Indicatorswithin some perspectives are only reported at long intervals (e.g. annually) whileothers are subject to more frequent reporting (e.g. monthly).

SummaryTo summarise, both TQM and the balanced scorecard have been promoted by amixture of private and public sector promulgators but only involving limitedintervention by the Government. Government-sponsored promotion of theseinnovations has largely pivoted on the documentation and dissemination ofemerging practices in central government agencies. Similar to the more generaldevelopment of TQM and the balanced scorecard there appears to have been verylimited communication between key promulgators of these innovations and littlebundling on the supply side (Chenhall, 2008). Taken together, this has resulted in afragmented supply of management innovations. Novel management techniques,especially those inspired by TQM, are neither wholly compatible nor tightlycoordinated with the managing for results model forming the backbone ofgovernment control of agencies. However, the promotion of novel managementtechniques has involved some deviations from standard TQM and balanced scorecardprescriptions. We now turn to examine how such supply conditions have affected thetransformation and bundling of management control innovations in the SNBSA.

5. The case of the Swedish National Board of Student AidContextual background

The SNBSA administers a range of state-subsidised financial support schemes gearedtowards secondary and tertiary education students and is held accountable to theMinistry of Education and Science. A substantial part of its operations entailsadministration of government grants and loans to university students andadministration of re-payment schemes. The SNBSA was formed in 1964 andfunctioned as a supervisory body for a large number of autonomous regional studentaid boards administering financial support schemes to university students. In 1992,these organisations were merged into a unified central government agency. However,the regional organisation was long retained as a strong local presence was consideredimportant for serving students effectively. It is only recently that the number ofregional offices has been reduced following the introduction of new operating practicesand demands for enhanced efficiency (see the following).

In 2005, the SNBSA distributed 23 billion SEK in the form of various supportschemes and administered outstanding loans of 175 billion SEK[6]. Its operatingbudget is typically covered by state grants and fees charged to users of its services (e.g.administrative charges, penalty charges for delayed re-payments) in roughly equalproportions. In 2005, the total operating budget amounted to 728 million SEK and theagency employed around 1,000 staff. The organisation serves around two millionindividuals and has an important impact on the daily lives of a relatively largeproportion of the population.


69


12/32

The overriding objectives of the SNBSA, as spelled out by the Government, pivot onits role in furthering enrolment in higher education such that economic, geographic andsocial differences in the student population are evened out. However, very littleinformation about how the agency contributes to meeting these wider educational

objectives is available in its external reporting and management control systems. Bycontrast, similar to many other central government agencies the SNBSA is subject tovery detailed targets and reporting requirements covering a broad range of short-termoperating aspects as well as financial performance.

This short-term emphasis is also strongly reflected in the agencys internal controlpractices. Regional office managers have budgetary responsibilities primarilycomprising salaries and rental charges. However, budgetary control was longsubordinate to more detailed and frequent control of operating aspects. Theorganisation exercises very tight control of short-term service aspects such as leadtimes in processing loan applications and accessibility in terms of response times fortelephone, mail and electronic mail. Interviewees argued that these performance

aspects had always been of central concern within the SNBSA. For example, a memberof headquarters staff with long experience from the agency characterised theorganisations control practices as follows:

It is a very inward-looking perspective with a focus on lead times and responding to phonecalls. It has been the same thing for the past 30 years.

Achievement of short-term targets for such performance aspects is critical to maintainservice levels and constitutes a considerable challenge for the organisation duringcertain parts of the year. Operations are highly cyclical, with a peak workload at thestart of every university semester. Although the SNBSA takes considerable stepstowards meeting such challenges through employment of temporary staff andre-allocation of resources, it has repeatedly been exposed to public criticism and

negative media attention for having excessively long lead times and poor accessibility.Complaints are also addressed to the Ministry of Education and Science. This has bredpolitical concerns and has occasionally resulted in a strained relationship between theagency and the Minister of Education. Hence continuous monitoring of lead times andaccessibility indicators has high priority within the Ministry. According to thegovernment official in charge of performance monitoring, the Minister of Educationtakes a direct interest in these matters. In times of escalating public criticism, as wasthe case during the recent re-structuring of the organisation, the Ministry has alsoimposed increasingly detailed controls requiring monthly reporting of lead times andaccessibility indicators from the agency.

The short-term emphasis of traditional control practices was reinforced by the

leadership style of the senior management team heading the SNBSA between the late1970s and 1999. The former Director-General was characterised as a charismaticperson with a preference for direct intervention in day-to-day matters, but not alwaysconcerned with the long-term consequences for the organisation. Control practiceswere described as ad-hoc and situated in the back pocket of the DeputyDirector-General but involving little dialogue or feedback. One interviewee describedmanagement control practices as a matter of:

. . . receiving long letters [from the Deputy Director-General] and then you got a budget.

AAAJ22,1

70


13/32

This leadership style was associated with a culture that one office manager describedas bohemian. She also characterised the Deputy Director-General, in particular, as aperson who:

. . .

absorbed every novelty, although he may not have had a vision of what to use it for.Consequently, a number of new technical solutions to facilitate payment procedureswere introduced on the initiative of senior management. As explicated in the following,this willingness to experiment has also been manifest in the adoption of novelmanagement techniques. Figure 1 illustrates the approximate timing of the changesunfolding since the late 1990s.

Experimenting with total quality managementStarting in 1998, the SNBSA has made a number of attempts to implementprocess-orientated work practices, inspired by TQM principles. As a first step, anumber of core operating processes were identified based on specific services provided to

different categories of beneficiaries. While these processes were largely based on theexisting organisation of operations (e.g. payment and repayment procedures) these werecross-functional in that employees located in regional offices generally split their timebetween different processes. A new customer service process, pivoting on managementof customer contacts via telephone, was also identified to focus attention to this critical areaof operations. So called process owners, with a responsibility to develop these lateralprocesses, were appointed and plans were drawn up to establish a central committee toensure that these individuals received sufficient support and resources. However, theprocess owners were not granted formal budgetary responsibilities, which remainedwith the regional office managers. Hence the new process-orientated responsibility areascame to overlap with the hierarchical structure in place for management control.

This early experimenting with process orientation was heavily inspired by existing,

TQM-based standards. An important source of inspiration was the recommendationsissued by the SIQ. Documentary evidence testifies to the increasing service demands ofusers and the public criticism of the SNBSA as a general impetus for the need for amore customer-focused approach to operations. However, interviewees rather pointedto the existence of a small group of employees firmly convinced of the benefits ofprocess orientation as a more direct driving force behind adoption and implementation.Implementation also relied heavily on external consultants. One member ofheadquarters staff thus characterised the adoption decision as follows:

It was a consultancy idea and a bit of a fashion thing but there was a problem of explainingthe value added by process orientation.

Figure 1.Timing of the

experimenting withmanagement control

innovations in the SNBSA


71


14/32

The Director-General supported the initiative. However, similar to the group ofemployees driving the implementation efforts, his endorsement was partly attributedto a striving for external recognition and legitimization. For example, one officemanager suggested that:

The whole thing was based on SIQs model . . . and then it was also possible to win qualityawards and that was something [the former Director-General] liked[7].

The view of process orientation as predominantly influenced by externalstandard-setters was corroborated in our follow-up interviews and feedbacksessions. One key informant suggested that the process orientation efforts were notreally driven by top management but rather by consultants.

Although the work with process orientation continued after the appointment of anew Director-General in 1999 and was gradually extended to headquarters functions,he grew increasingly wary of re-organising operations. In a feedback session, a keyinformant recalled that:

I remember how we discussed how to divide the roles between process owners and officemanagers. I think the idea was to allocate more resources and authority to the process owners. . . but then the office managers asked what should we do?. This led [the Director-General],who had inherited the work with process orientation, to pull the brake.

Hence the implementation of process-orientated control practices lost some of itsmomentum. While process orientation was still seen as a tool for improving operationsits role in allocating resources and responsibilities was de-emphasised. Consequently,several interviewees claimed that the role of the process owners became increasinglydiffuse. The process orientation efforts at the headquarters level also gradually groundto a halt. In early 2002, a memo summarising an internal evaluation of the experiencesof process orientation identified the following main problems:

(1) Unclear distribution of tasks and roles between various parts of theorganisation and a feeling that process orientation was implemented in atop-down manner without much consultation between headquarters andregional offices.

(2) Unclear articulation of the responsibilities of the process owners and failure tointegrate these into management teams in charge of day-to-day, operating-leveldecisions.

(3) Lack of integration of local development initiatives linked to process orientationand failure to adopt a more holistic view of these efforts.

(4) Attitude problems among staff hindering a more customer-orientated approachto service delivery.

While our informants confirmed many of these observations, they also suggested thatthe work with process orientation assumed a very technical guise as a result of theneed to adapt process descriptions to fit existing information systems supportingpayment and re-payment routines. For example, in one of our feedback sessions, oneinformant recalled:

One problem [with process orientation] was that there were only technicians involved. Thiswas reinforced by the problems of dealing with the millennium bug. It all became a matter oftechnology.

AAAJ22,1

72


15/32

Another informant confirmed this:

Process orientation was equated with IT. There was no focus on customers and we werestuck with the IT systems that had been built up

While the introduction of process orientation was officially lauded as a means ofenhancing the level of customer orientation, the extensive process mapping beingundertaken was thus characterised as having a very introvert focus. Yet, someinterviewees indicated that customer orientation gradually became more accepted thanthe work with process orientation. One reason for this, as an informant with longexperience from the SNBSA explained, was that the notion of customers was not reallynew to the organisation but was introduced by the former Director-General already in the1980s to emphasise the need for flexibility. While this was said to be rather controversialat the time, he suggested that it had supported the more systematic customer orientationefforts embedded in TQM-inspired practices. A concrete manifestation of this systematicapproach was the clearer customer segmentation initiated as part of the processorientation efforts. For example, the organisation now refers to customers with special

needs (e.g. beneficiaries with significant problems of re-paying loans) and have takensteps to provide more targeted services to these. Extensive efforts have also been madeto train staff in customer management.

There were also indications of the enhanced emphasis on customer orientationimpinging on the meaning of performance. The work on process orientation did notinvolve any far-reaching attempts to devise new performance indicators to clarify theresponsibilities of the process owners. However, more aggregated customersatisfaction indexes were introduced to visualise the striving towards enhancedcustomer orientation. As illustrated in the following, such indicators have also formed anintegral part of the performance management systems evolving in more recent years.

Experimenting with the balanced scorecardIn late 2000, the SNBSA started to develop a balanced scorecard-inspired performancemanagement system, which was introduced more broadly in conjunction with theannual budgeting and planning process. According to one interviewee assuming a keyrole in this respect, the initial interest in this model originated from the newDirector-General who wanted to try something new and engaged consultants. In aninternal memo dated around the same time, the Director-General motivated theintroduction of the system as follows:

To manage our task effectively and contribute to achievement of the educational policyobjectives and simultaneously accommodate the demands of other stakeholders, the visionand overall objectives of the SNBSA have been further specified in comparison with last

operating year through the development of the balanced goal card. In the card, there are fourfocus areas of utmost importance for operations with objectives, strategies, critical successfactors and measures for each (excerpt from internal memo, September 20, 2000).

In contrast to existing control practices, the innovation was portrayed as a broadlybased and more goal-directed performance management system, which would be:

. . . both short-term and long-term in nature. With the help of this strategic goal card control ofoperating activities is linked to the long-term vision and the strategies of the SNBSA (excerptfrom internal memo, September 20, 2000).


73


16/32

Similarly, promotional material used for internal dissemination of the innovationprimarily justifies its use with reference to general shortcomings of traditional controlsystems being emphasised in the balanced scorecard literature, such as the provision ofmisleading information for decision-making, a lack of non-financial and more

externally focused performance measures and risks of sub-optimisation (Kaplan andNorton, 1996), rather than more specific, contextualised descriptions of the challengesfacing the SNBSA. Hence the justifications for adopting the new system resonated withmuch of the rhetoric surrounding the balanced scorecard (Nrreklit, 2003). However,the change in terminology, with the notion of scorecard being replaced by goalcard, was conditioned by the overly obtrusive connotations of the former term, or asone interviewee explained:

We were not allowed to say scorecard, because then it seemed too much like control.

Similar to the work on process orientation, the implementation of the goal cards wasdriven by a small group of enthusiasts under the influence of a range of externalsources. One of the leading propagators of the balanced scorecard in the SNBSA

acknowledged that she first encountered the concept while attending an executivemanagement class at a nearby business school and described herself as totallyconverted to the idea. Several interviewees also emphasised the extensiveinvolvement of consultants in introducing the concept more broadly in theorganisation and developing it to fit the SNBSA. However, even though oneinterviewee suggested that the consultants gave the system an SNBSA touch, a moredominant feeling was that it only entailed limited adjustments in comparison with theoriginal balanced scorecard model (which seemed well-known to most interviewees).For example, one member of headquarters staff described the system graduallyevolving as the balanced scorecard straight off the shelf. Another intervieweereferred to one of the publications on the topic by the National Financial ManagementAuthority as the Bible for how to devise such systems in the organisation. The goalcards subsequently emerging also show considerable affinity to the balancedscorecard concept (see Figure 2).

The goal cards gradually evolved over the coming years. However, thedevelopment of the system primarily involved compilation of existing performanceindicators into the various focus areas rather than a more open-ended search forsuitable indicators linked to a coherent analysis of the organisations strategy. Forexample, customer satisfaction was identified as an overriding performance criterionwithin the customer perspective and linked to the customer satisfaction indexesemerging from the TQM-inspired change efforts (see Figure 2). However, thedevelopment of the goal cards was not tightly coordinated with the parallel efforts toimplement process-orientated work practices, or as one of the former process owners

recalled:Process orientation ran parallel to the development of the goal cards . . . We described andmodelled the processes. But it ran as two separate tracks and we wondered what [the leadingpropagator of the balanced scorecard in the SNBSA] was really up to. We didnt think therewas much to learn from the goal cards.

Neither was any analysis undertaken of the link between customer satisfaction and theover-riding political objectives of the SNBSA pivoting on its effects on enrolment inhigher education. Similar to many other Swedish organisations in the public as well as

AAAJ22,1

74


17/32

the private sectors, the goal cards also encompassed a clearly defined employeeperspective (Ax and Bjrnenak, 2005; National Financial Management Authority, 2000,2006). However, the organisation relied on the information in its existing employeedevelopment plans to assess achievement of the objectives established within thisperspective.

The organisation-wide goal card depicted in Figure 2 was also transformed into

office-specific goal cards, featuring a stronger emphasis on short-term targets andperformance indicators. However, these goal cards mainly included measures whichhad long been of key concern in the organisation, such as lead times and accessibilityindicators, without much analysis of how they contributed to achievement of moreoverriding objectives. Thus, one member of headquarters staff summarised the effortsto compile performance indicators as follows:

We found a lot of measures, but it turned out to have a very operational focus. We were notable to capture the more long-term measures.

Figure 2.The organisation-wide

goal card of the SNBSAin 2003


75


18/32

Other informants referred to the goal cards as a sorting or packaging mechanism,arguing that little thought was given to how various performance dimensions wereinterlinked. One member of headquarters staff also suggested that such attempts tobring order to management control practices rather resulted in a lack of overview and

information overload:

My point of departure is that controlling the SNBSA is fairly simple, but there is a tendency

for more and more demands to be placed on the organisation. So the overall objects become

complex and this was one reason why the goal card became too unwieldy.

Although the general idea of balancing short- and long-term performance measures

and linking these into chains of leading and lagging indicators was emphasised inpromotional material accompanying the goal cards, little of this materialised in theactual development of the system. At the outset, the idea of balance was described as amatter of:

. . . achieving balance between the various focus areas being established and also balance

between the various levels of the organisation (excerpt from internal memo datedSeptember 20, 2000).

In the subsequent development of the system, however, the emphasis on the goalcards for disaggregating performance measurement across various hierarchical levelsgained the upper hand, while little effort was expended on analysing or mapping

causal relationships between various focus areas. Office managers were expected touse the goal cards for systematically evaluating how individual employeescontributed to meet the overall vision and objectives of the SNBSA. However, far from

all office managers seem to have made whole-hearted efforts to this end, with onearguing that he preferred ongoing, personal contacts with subordinates and anotherseeing the goal cards as a bureaucratic tool and never fully apprehending it. Even

where more concerted efforts were made to use the goal cards at the individual levelthe opinion was that this did not result in tighter control.

The goal cards also encountered a range of additional implementation problems.A widespread concern among headquarters staff as well as office managers was thatthe latter had not been sufficiently involved in the development and implementation ofthe system, which progressed in a top-down manner. Another frequently voicedperception was that the purpose behind the goal cards was unclear and that

relatively little emphasis was placed on regular use of the system for performanceevaluation based on quantifiable criteria linked to operating actions. For example, onemember of headquarters staff argued that:

From my perspective [the goal cards] became too theoretical and something that ran

parallel to operations. The link to everyday operations was poor.

Similarly, the Chief Financial Officer explained that budgetary control evolved on a

largely parallel basis and was tightened to encompass more frequent feedback tooffice managers but not linked to other performance evaluation practices. She went onby saying that:

The problem within the SNBSA is that control of operations and financial control have never

been linked together.

AAAJ22,1

76


19/32

This observation was not only attributed to the evolution of the goal cards, but alsoto the old culture of the organisation where things got done by means of informalcommunication and decision-making. As explicated in the following, however, thestriving for enhanced formality has been accentuated in recent years as a result of the

ensuing development of performance management.

Towards a new performance management systemDuring this period of experimenting with novel management techniques, the financialsituation of the SNBSA deteriorated considerably. Costs increased steadily as a resultof several factors such as growing operating volumes and investments in computersystems to deal with the millennium bug. In addition, a centralised call-centre functionwas built up to strengthen telephone services and improve accessibility between 2001and 2003. However, this initially led to additional cost increases, which were onlypartially compensated for through government funding.

In 2003, a major re-structuring of the organisation was initiated to cut operating

costs. A review of headquarters functions was initiated and the number of regionaloffices was reduced from 24 to 13. Workloads were also gradually redistributed suchthat the administration of certain types of services was concentrated to a smallernumber of offices. Consequently, the number of staff was reduced by about 200 toaround 1,000. These changes caused considerable upheaval and were accompanied bya period of deteriorating service levels, public criticism and negative media attention.Several interviewees recalled these changes as highly stressful. Managerial attentionwas also absorbed to the detriment of the development of novel control practices. Forexample, one office manager explained that the implementation of process orientationand the goal cards stopped dead as a result of the managerial efforts expended onorganisational re-structuring. Similarly, the Chief Financial Officer described 2003 as alost year as far as the development of control practices was concerned. The view thatthe work with process orientation and the goal cards came to an abrupt end aroundthis time as a result of lacking managerial attention was confirmed in a feedbacksession and is mirrored by the following exchange between two of the participants:

Top management didnt know what to do with process orientation. There were no clearobjectives for it.

No, there wasnt enough space in their hard disks.

Coming out of this period of re-organisation, however, the SNBSA re-energised theefforts to develop the organisations control practices. In 2004, a small team ofstrategists, several of who had previously been extensively involved in the work onprocess orientation and the goal cards, was formed to assist the Director-General in

this matter. The task of this group was to undertake a comprehensive review of theorganisations strategy and produce a coherent, long-term plan for the development ofthe SNBSA.

An overriding concern in the ensuing development of the long-term, strategic planwas to better align short-term improvement efforts to the objectives of the SNBSA thanhad been the case in the previous experimenting with process orientation and the goalcards. However, some key elements of these initiatives were retained. The vision ofthe SNBSA as a progressive and active service agency was further specified into


77


20/32

three broadly defined objectives similar to those found in the organisation-wide goalcards, namely:

(1) Enjoying full citizen confidence.

(2) Conducting operations in an efficient, lawful and democratic manner.(3) Being an attractive workplace.

Vestiges of the balanced scorecard logic are also evident in the new performancemanagement system, called the target map, being developed to guide theorganisation towards this vision. The original versions of the target map weregraphical representations linking overriding objectives to more specific functionalstrategies, which were then further specified into targets over a period of three years.However, the targets for each functional strategy were not identical across the years,but were to be updated as targets for the previous year were achieved. In addition, thetargets reflect different, causally linked performance aspects believed to contribute to

the achievement of long-term objectives. Targets for the current year were translatedinto more specific, short-term action plans, which were also updated as certain actionsand projects were completed. One office manager thus characterised the target map asa living document rather than a static control mechanism.

The target map was more widely introduced to office managers in the early half of2005 with an invitation to suggest changes to the model. Concomitantly, the systemgradually started to be used for performance evaluation three times per year as anintegral part of the annual planning and budgeting process. The development workcontinued after the appointment of a new Director-General in 2005 who fully supportedthis initiative but also initiated some simplifications of the design of the target map,which had now grown into very extensive graphic exhibits (covering four large papersheets). A number of short-term targets were removed and some broad areas of prioritywere added to further specify the strategies emanating from the overriding objectives.The resultant target map, included in the strategic plan for the SNBSA for 2006-2009,is provided in Figure 3.

Despite the problems encountered in the implementation of the goal cards, severalinformants emphasised the continuity between this change initiative and thedevelopment of the target map. For example, in a feedback session a controllerextensively involved in the design of the target map explained that:

When we started with the target map, the organisation had burnt itself badly on the balancedscorecard. But we also saw the merits of building on this. We surely had the balancedscorecard in mind when we started drawing the target map.

Other interviewees explained that the main similarity between the balanced scorecardand the target map was the ambition to adopt a holistic view of the relationshipsbetween the three overriding objectives in the latter. Some of them saw the targetmap as just another way of packaging, or organising managing for results in theSNBSA. The link between the balanced scorecard and the target map was alsoemphasised by one of the strategists, heavily involved in the organisations priorexperimenting with various management techniques:

The new system is also a balanced scorecard, but nobody has realised that.

AAAJ22,1

78


21/32

In a follow-up interview, she reiterated this view, saying that she saw the target mapas something in between TQM and the balanced scorecard and emphasised the closelink between short-term action plans and targets as an important means of stimulatingcontinuous improvement. However, with the exception of the heightened emphasis oncustomer orientation, informants were generally unable to elaborate on the linkages

between the target map and the previous, TQM-inspired work with processorientation when queried about this topic. Yet, one of them explained that customerorientation was:

. . . the objective with highest priority. Everything we do is linked to that.

Notably, customer-orientated performance indicators introduced as a result of theTQM-inspired change efforts, such as customer satisfaction indexes, also occupy aprominent position as a key target in the target map. Since the appointment of the new

Figure 3.Simplified depiction of the

target map of theSNBSA in 2006


79


22/32

Director-General in 2005, concerted efforts have been made to improve customersatisfaction, especially by focusing on key underlying performance drivers such astelephone response times. Significant improvements were also reported in suchperformance aspects in 2006. Office managers explained that an important means to this

end was more focused utilisation of staff than had previously been the case, especiallyduring peak seasons. In 2006, greater efforts were also initiated to measure satisfactionwith specific service aspects (e.g. telephone service, e-mail correspondence) moreaccurately rather than mainly relying on aggregate customer satisfaction indexes.

Several informants ascribed this enhanced emphasis on more concrete,customer-orientated performance aspects to personal, hands-on initiatives by thenew Director-General. Whereas her predecessor was generally seen as having apenchant for formulating ambitious visions and delegating decision making, a memberof headquarters staff characterised the new Director-General as:

. . . a very practically orientated leader. Targets and controls are more grounded in practicenow.

Similarly, in a follow-up interview an office manager suggested that:

Controls have been better structured over the past year . . . Above all, there is more workshopthan talk today. It is largely an effect of [the new the Director-General] as she intervenes moredirectly and makes things happen.

In our interviews, the new Director-General also placed considerable emphasis ondeveloping customer-orientated management practices while echoing the concernswith the overly technical emphasis of the earlier experimenting with processorientation. When asked to elaborate on the evolution of the notion of customerorientation she explained that:

The difference from the earlier attempts with process orientation in the SNBSA is that it is

less of a matter of drawing process maps, which scared a lot of people, than talking aboutprocesses and taking this as an overall approach. The focus is now on who the customer is.Processes have to add value to the customer.

It is worth noting that the Director-General previously headed one of the leadingpropagators of TQM-inspired practices among Swedish central government agencies(the National Council for Quality and Development). However, rather than insisting onfull-scale implementation of process-orientated control practices, a more cautious andselective approach has been adopted. A member of headquarters staff elaborated onthis as follows:

Our current process thinking is not supposed to result in any re-organisation . . . What wehave now is a half-way solution and it is more a matter of describing processes. People never

understood the meaning of processes. It has calmed down with [the new Director-General]and it has been a process of maturation.

In our interviews, the Director-General also disassociated herself from the earlierexperimenting with process orientation in the SNBSA with reference to the widerexperiences of such practices in Swedish central government:

I have no plans to re-organise the SNBSA as they tried to do earlier but I rather try to instilprocess thinking as a general approach. No-one has succeeded in creating a totallyprocess-orientated organisation. Everyone has abandoned it.

AAAJ22,1

80


23/32

The unfolding development of control practices thus features some important vestigesof the earlier experimenting with TQM- and balanced scorecard-inspired techniques,although a rather selective approach has been taken in integrating elements of theseinto the new performance management system. In contrast to these earlier change

efforts, however, our informants emphasised that the subsequent development of thetarget map was an entirely internal project unfolding without much explicit influencefrom external sources. The involvement of external consultants was minimal and thestrategists emphasised that the ambition had been to start from scratch. Ratherthan compiling existing performance measures into the target map, a moreopen-ended approach was adopted starting with the identification of causally relatedobjectives and targets and then initiating a search for suitable indicators linked tothese. One of the strategists emphasised the advantages of this approach, sayingthat:

We didnt have any help from consultants which was good, because from them we only gotthe model [i.e. the balanced scorecard].

The initial justifications for the enhanced emphasis on strategic planning anddevelopment of the target map also made much less reference to fashionablemanagement techniques, but largely pivoted on the need to revitalise the organisationafter a period of radical and stressful change. Promotional material used to introducethe strategic planning activities to employees emphasised the need for a fresh startand setting the organisation on course for 2008 after having gone through one of thebiggest changes ever in the history of the SNBSA. Similarly, interviewees stressed thenature of the initiative as a response to the crisis that the SNBSA had recentlyexperienced. This was especially emphasised by the office managers and wasgenerally appreciated as it shifted their attention from the daily grind. For example,one of them stated that the target map had implied a born again experience for her.

Another important impetus behind the strategic planning efforts was the striving tolink internal control practices more closely to political priorities embedded in theemerging, citizen-orientated reform agenda. A member of headquarters staff describedthis as follows:

The focus on the political objectives has become clearer through our work on the strategy.Earlier, we didnt really know where the SNBSA was going.

Planning documents produced by the strategists also placed considerable emphasison citizen orientation. The language of citizen orientation was here used largelyinterchangeably with that of improved customer service. For example, the need forimproved customer service was justified as follows:

The transformation from exercising a purely administrative role to becoming a widely

appreciated, service-orientated agency that effectively meets the needs of the new citizenwill bring demands for change in the SNBSA. The demands imply that development ofoperations and the concomitant development of services need to be increasinglycitizen-driven to meet requests for simple and constantly available, but also individualisedand qualified service (excerpt from internal planning document, August 26, 2004).

Office managers felt more involved in the work on the target map than the previousefforts to implement process orientation and the goal cards and argued thatimplementation had enhanced their understanding of the system. In particular, the new


81


24/32

visualisation of the linkages between short-term action plans and various targets wasappreciated as it clarified the impact of specific efforts on the long-term strategy of theorganisation. An important reason for this enhanced understanding was that seniormanagement had adopted a more tightly coordinated approach, emphasising the need

for broad involvement rather than delegating work to a small group of highlycommitted but isolated individuals. Whereas such an approach was adopted from theoutset, with the Director-General expending considerable efforts on communicating themeaning of the target map to office managers and other managerial staff, hissuccessor has continued along this path. In an interview, she underlined this as follows:

. . . implementation depends on the degree of management commitment and then I mean thewhole management team . . . Earlier, we have had a few highly committed individuals but thenew control techniques havent been thoroughly anchored in different areas of operation.

At the same time, the importance of senior management owning the new strategywas emphasised by several informants. Several interviewees also suggested thatimplementation of the target map had progressed in a largely top-down manner. One

of the strategists found this paradoxical, arguing that:

. . . there is more centralised control today, but it is also more accepted. Earlier on, there was alot of internal talk about how centralised the SNBSA was, but now nobody reacts against it.

While reflecting the same, over-riding priorities the development of the target mapbetween 2006 and 2007 entailed additional simplifications such that it largely ignoresthe links between various targets over the years. A controller explained the rationalefor this:

Earlier, we had a perspective of five years but now we only concentrate on one year at thetime because we dont know very much about the future anyway.

However, our follow-up interviews in 2006 and 2007 suggest that the target map had

brought few novel performance indicators into use at the operating level, but that itprimarily functioned as a means of structuring strategic planning. Those in charge ofdeveloping such indicators explained this with reference to the technical difficulties indevising measures reflecting several key targets, such as the coherence of rules andcosts per customer. Some of the strategists also voiced concerns regarding the limitedefforts of office managers to communicate the meaning of the strategy and the targetmap to lower-level employees.

6. Concluding discussionThis study constitutes a first attempt to examine in greater detail how contemporarymanagement control innovations are bundled in an individual organisation withparticular reference to TQM and the balanced scorecard. Our findings provide aninteresting contrast to prior research on this topic in that we find little evidence ofdeliberate bundling on the supply side (Ax and Bjrnenak, 2005, 2007; Bjrnenak andOlson, 1999). We rather illustrate how the process of bundling is part of a less lineartrajectory of change involving considerable experimenting by an adopter over anextended period of time. While the implementation of process orientation and the goalcards largely progressed as two independent projects, the enhanced emphasis oncustomer orientation and use of customer satisfaction as a key performance indicatorprovided a point of connection between these innovations. Such performance aspects

AAAJ22,1

82


25/32

have also received heightened attention in the subsequent transformation of the goalcards into the target map while other vestiges of TQM were graduallyde-emphasised. Yet, an element of process thinking still permeates the discourseinvoked by senior management.

In this gradual process of bundling elements of TQM and the balanced scorecard,there is evidence of an increasingly reflective and reasoned managerial posture. Whileproblematic, the earlier experiences of aligning objectives, strategies, targets andshort-term actions, and adopting a multidimensional and holistic view of performancethrough the goal cards were re-mobilised in the design of the target map. However,this occurred as part of a more careful analysis of the political objectives and strategyof the SNBSA and was followed by a series of simplifications to increase the clarity ofthe system. Similarly, vestiges of TQM, such as customer-orientated performanceindicators, were gradually refined and made more operational while being unbundledfrom other elements of TQM proving more problematic in the context of the SNBSA(e.g. process-orientated responsibility structures and resource allocation). This partlyoccurred against the backdrop of the wider, negative experiences of process-orientatedmanagement practices in Swedish central government (Modell et al., 2007). In addition,senior management seems to have taken stock of the experiences of the earlier,fragmented implementation efforts by adopting a more coordinated approach aimed atbroader involvement.

These findings negate the view of managers of adopting organisations asunreflective actors blindly enacting readily packaged solutions, which permeates muchof the literature on management fads and fashions (Clark, 2001, 2004; Giroux, 2006).The traditional fashion perspective is clearly valid in explaining the initial adoption ofTQM- and balanced scorecard-inspired techniques, given the considerable influencefrom authoritative sources outside Swedish central government (e.g. the SIQ, privatesector consultants) and the limited presence of efficient choice and forced selection

rationales (Abrahamson, 1991; Malmi, 1999). However, the unfolding development ofthe target map reveals a more complex mix of rationales behind managerial action.Although the initial, fashion-driven experimenting with TQM and the balancedscorecard has shaped the experiences of key organisational actors, they subsequentlyappear to have adopted a more rational posture in an environment where littleprescriptive advice for how to bundle such innovations is available. This position wasrather underpinned by the enhanced emphasis on goal-directed, long-range planning inan attempt to revitalise the organisation after a stressful period of radicalre-structuring and to better align control practices with political objectives and theGovernments emerging reform agenda. In this process, management has alsodemonstrated heightened attention to the risk of novel control practices becomingoverly stereotyped (Parker and Ritson, 2005) and has made considerable

context-specific adjustments along the way.These findings suggest two inter-related refinements to the fashion perspective in

advancing research on the bundling phenomenon. First, the notion of bundling mayneed to be re-conceptualised to illuminate how it is entangled with managerial andorganisational learning processes. While fashionable innovations provide importantinputs to such processes, our findings suggest that learning entails important elementsof serendipity as well as rationality as organisations experiment with variousinnovations. While the emerging literature on the bundling of management accounting


83


26/32

and control innovations has ignored such processes (Ax and Bjrnenak, 2005, 2007;Bjrnenak and Olson, 1999) it may be extended through integration withorganisational learning theories recognising that bundling involves some balancingbetween goal-directed behaviour and less malleable experimenting (see, e.g. Crossan

et al., 1999; March, 1991). Successful adjustments of innovations to a particularorganisational context may require selective incorporation of design elements,emerging as particularly useful during the process of implementation but beingmediated by repositories of organisational experience. However, to accomplish moregoal-directed change managers also need to ensure that such design elements are inline with the overall purpose of the new innovation package gradually evolving.

Second, recognising that experimenting with fashionable management innovationsentails important learning aspects, we need to abandon the view of such innovations asrelatively transient phenomena characterised by limited stability and impact onorganisational practices over time (Abrahamson, 1996; Abrahamson and Fairchild,1999; Collins, 2000; Jackson, 2001). Our findings confirm prior research suggesting thatfashion-driven adoption of management control innovations is associated withconsiderable implementation problems (Modell, 2004; Northcott and France, 2005).However, categorising the largely failed implementation of TQM and the balancedscorecard in the SNBSA as organisational rejection of these innovations(Abrahamson, 1991) would be an over-simplification given their lingering influenceon the new performance management system evolving in the organisation. Extendedanalyses of organisational histories of experimenting with different managementinnovations and the residues it leaves in organisations are required to fully understandthe reasons for such unintended consequences of change (Andon et al., 2007;Hopwood, 1987). However, such analyses are still rare in management accountingresearch informed by the fad and fashion perspectives.

While longitudinal field studies would be a preferred research strategy to

accomplish these refinements, we also believe that our observations have someimplications for the predominantly cross-sectional, survey-based literature on theadoption of management control innovations. This literature has paid scant attentionto the complexity of implementation processes and has concentrated on individualinnovations (Modell, 2007). To capture how prior organisational experiences ofmanagement control innovations affect the implementation of a particular technique,survey research may be enriched by qualitative inquiries into these experiences in afew, carefully selected organisations. A mixed methods approach would thus seem apromising avenue for examining the adoption and implementation of managementcontrol innovations and how this relates to the bundling phenomenon (Modell, 2005).

At a more general level, this paper also contributes to the recent debate on whethermanagement accounting research should concentrate on largely fashion-driven

innovations (Ittner and Larcker, 2001, 2002 vs Zimmerman, 2001). Our findingsunderscore that such innovations might play an important role in the continuousshaping of management control practices even though they may prove problematicfrom the perspective of implementation. We thus concur with Ittner and Larcker (2002)in that the role of fashionable management control innovations is certainly worthy ofserious academic research. However, more attention needs to be paid to the bundling ofsuch innovations than has hitherto been the case. By focusing on how organisationsmake selective use of the constituent elements of innovations over time rather than

AAAJ22,1

84


27/32

being pre-occupied with managerial fashions per se, we may be better positioned tocapture the more fundamental role of such innovations for the purpose of managementcontrol (Weick, 2001).

Notes

1. See Abernethy et al. (2005), Bukh and Malmi (2005) and Nrreklit (2000) for critical analysesof the nature the causal linkages embedded in the balanced scorecard.

2. A similar neglect of the potential conflicts between lateral and hierarchical logics of controlis discernible in Hoques (2003) discussion of the links between TQM and the balancedscorecard, which mainly addresses measurement-centred issues associated with combiningthese innovations. However, Chenhall (2008) observed that the balanced scorecardincorporates few elements associated with the lateral orientation of techniques such as TQM.

3. The choice not to tape interviews was partly conditioned by the tense political climatesurrounding the SNBSA at the outset of the study. According to some informants, thissituation was exacerbated by the strained relationship between the Director-General and the

Minister of Higher Education at the time.4. In Swedish central government parlance, management by objectives and managing for

results are typically used interchangeably. However, the latter term is more commonlyaccepted and we therefore use this consistently throughout this paper.

5. The National Financial Management Authority was formed as an advisory agency in 1998as the consultative branch of the National Audit Office was separated from state auditing.

6. 1 SEK approximately 0.11 Euro.

7. The SIQ arranges annual contests modelled on the prestigious Malcolm Baldrige Award inthe USA.

References

Abernethy, M.A., Horne, M., Lillis, A.M., Malina, M.A. and Selto, F.H. (2005), A multi-methodapproach

bundling management control innovations

Documents