Management control systems design within its organizational

Management control systems design within itsorganizational context: findings from contingency-based

research and directions for the future

Robert H. ChenhallDepartment of Accounting and Finance, Monash University, Clayton, Victoria 3168, Australia

Abstract

Contingency-based research has a long tradition in the study of management control systems (MCS). Researchershave attempted to explain the e!ectiveness of MCS by examining designs that best suit the nature of the environment,technology, size, structure, strategy and national culture. In recent years, contingency-based research has maintainedits popularity with studies including these variables but redefining them in contemporary terms. This paper provides acritical review of findings from contingency-based studies over the past 20 years, deriving a series of propositionsrelating MCS to organizational context. The paper examines issues related to the purpose of MCS, the elements ofMCS, the meaning and measurement of contextual variables, and issues concerning theory development. A final sec-tion considers the possibility that contingency-based ideas could encompass insights from a variety of theories to helpunderstand MCS within its organizational context. # 2002 Elsevier Science Ltd. All rights reserved.

1. Introduction

The three purposes of this paper are to providea review of empirical, contingency-based researchas it has developed since the early 1980s, to criti-cally evaluate this work, and consider a variety oftheoretical foundations that may assist in devel-oping future research. The review is based, in themain, on research employing survey-based meth-ods that has been published in a broad selection ofaccounting and management journals.1 The review

is selective and illustrative of issues pertinent tothe development of a contingency-based frame-work for the design of MCS, and does not attempta comprehensive coverage of relevant researchThe paper is structured as follows. The next

section introduces the area of contingency-basedMCS research and provides an overview of find-ings over the past 20 years. The following ninesections review articles in terms of their contribu-tion to understanding topics considered withincontingency-based research. These are: the meaning

0361-3682/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.PI I : S0361-3682(01 )00027-7

Accounting, Organizations and Society 28 (2003) 127–168www.elsevier.com/locate/aos

E-mail address: [email protected] The journals include: Accounting, Organizations and Society; Accounting and Business Research; Accounting and Finance;

Accountability and Performance, Behavioral Research in Accounting; Journal of Accounting Research; Journal of Accounting andEconomics; Journal of Accounting and Public Policy; Journal of Cost Management; Journal of Financial Economics; Journal of Man-agement Accounting Research; Management Accounting Research, The Accounting Review. In addition, several articles are drawnfrom management journals: Academy of Management Journal, Academy of Management Review, Administrative Science Quarterly,Decision Science, Management Science and OMEGA.

mailto:[email protected]

of MCS, outcomes of MCS, and the contextualvariables of external environment, technology (tra-ditional and contemporary), organizational struc-ture, size, strategy and national culture. Eachsection comprises two parts: first, findings fromthe extant literature are presented and a series ofpropositions summarizing these findings areo!ered; and second, critical issues concerning eachvariable are examined with a view to identifyingareas that provide challenges for improvementand opportunities for future research. Followingthese sections, issues concerning theory develop-ment are examined. Finally, the potential role of avariety of theories in progressing understanding ofcontingency-based research in MCS is considered.

2. An organizational framework for contingency-based MCS research

The identification of contextual variablespotentially implicated in the design of e!ectiveMCS can be traced to the original structural con-tingency frameworks developed within organiza-tional theory. Theorists such as Burns and Stalker(1961), Perrow (1970), Thompson (1967), Lawr-ence and Lorsch (1967), and Galbraith (1973)focused on the impact of environment and tech-nology on organizational structure. Earlyaccounting researchers drew on this work toinvestigate the importance of environment, tech-nology, structure and size to the design of MCS.Reviews conducted 20 years ago by Waterhouseand Tiessen (1978) and Otley (1980) were able tostructure their commentaries by categorizing theearly research into these key variables.In considering MCS research since 1980, it is

apparent that these key variables have been con-firmed as descriptors of fundamental, generic ele-ments of context. Many recent studies, included inthis review, focus on contemporary aspects of theenvironment, technologies and structural arrange-ments. They draw on the original organizationaltheorists to develop arguments that help explainhow the e!ectiveness of MCS depends on the nat-ure of contemporary settings. Also, recentresearch has considered the relevance of additionalcontextual variables to the design of MCS. Per-

haps the most important new stream of literaturehas been that related to the role of strategy. Thishas been assimilated within the traditional orga-nizational model in ways that suggest importantlinks between strategy, the environment, technol-ogy, organizational structure and MCS (seeLangfield-Smith, 1997 for a review). The impor-tance of technology to MCS design has been enri-ched by research drawing on the manufacturingliterature (Hayes et al, 1988; Skinner, 1975), andthe work of economists such as Milgrom andRoberts (1990). Issues concerning the role of MCSwithin advanced manufacturing settings such asTotal Quality Management, Just-in-Time andFlexible Manufacturing have been explored (seeYoung & Selto, 1991 for a review). Researchershave gained new insights into the role of MCSwithin new structural arrangements, such asteams, by drawing on the human resource man-agement literature which investigates the dynamicsof teams including issues concerning performanceevaluation (Cohen, 1993; Katzenbach & Smith,1993). National culture has been identified as anelement of context following the development ofmulti-national operations in many organizations(see Harrison & McKinnon, 1999 for a review).In reviewing the past 20 years of contingency-

based research it is important to consider theextent to which progress has been made in devel-oping an empirical body of literature relatingMCS to elements of context. The conventional,functionalist contingency-based approach toresearch assumes that MCS are adopted to assistmanagers achieve some desired organizationaloutcomes or organizational goals. The appro-priate design(s) of MCS will be influenced by thecontext within which they operate. The followingnine sections consider: the meaning of MCS, theoutcomes of MCS and the key contextual variablesas they have evolved, historically, in the literature.First, the relationship between MCS and theexternal environment is considered. This is fol-lowed by technology (both traditional and con-temporary), structure and size. Next, strategy isexamined. Finally, the role of national culture inMCS design is reviewed. On the basis of theempirical findings, propositions are o!ered whichrelate contextual variables to the MCS. Assessing

128 R.H. Chenhall / Accounting, Organizations and Society 28 (2003) 127–168

these propositions requires considering short-comings in contingency-based research, identify-ing the extent to which progress has been made inaddressing these issues and noting opportunitiesfor improvements and future directions.2

3. The meaning of MCS

The terms management accounting (MA), man-agement accounting systems (MAS), managementcontrol systems (MCS), and organizational con-trols (OC) are sometimes used interchangeably.MA refers to a collection of practices such asbudgeting or product costing, while MAS refers tothe systematic use of MA to achieve some goal.MCS is a broader term that encompasses MASand also includes other controls such as personalor clan controls. OC is sometimes used to refer tocontrols built into activities and processes such asstatistical quality control, just-in-time manage-ment. The term MCS is used, in the main,throughout this paper.The definition of MCS has evolved over the

years from one focusing on the provision of moreformal, financially quantifiable information toassist managerial decision making to one thatembraces a much broader scope of information.This includes external information related to mar-kets, customers, competitors, non-financial infor-mation related to production processes, predictiveinformation and a broad array of decision supportmechanisms, and informal personal and socialcontrols. Conventionally, MCS are perceived aspassive tools providing information to assist man-agers. However, approaches following a socio-

logical orientation see MCS as more active,furnishing individuals with power to achieve theirown ends. Contingency-based research follows themore conventional view that perceives MCS as apassive tool designed to assist manager’s decisionmaking.Contingency-based research has focused on a

variety of aspects of MCS. These include dimen-sions of budgeting such as participation, impor-tance of meeting budgets, formality ofcommunications and systems sophistication, linksto reward systems (Bruns & Waterhouse, 1975;Merchant 1981), budget slack (Dunk, 1993; Mer-chant, 1985a; Van der Stede, 2000; see Dunk &Nouri, 1998 for a review), post completion audits(Chenhall & Morris, 1993; Smith, 1993) and var-iance analysis (Emsley, 2000). Examples of con-temporary innovations in MCS include ABC/ABM (Anderson & Young, 1999; Gosselin, 1997),non-financial performance measures (see Ittner &Larcker, 1998 for a review) and economic valueanalysis (Biddle, Bowen, & Wallace 1998).Researchers are likely investigating progress con-cerning balanced scorecards and target costing. Ata more general level, studies have consideredsophistication of controls (Khandwalla, 1972), reli-ance on accounting performance measures (RAPM)(Brownell, 1982; 1987; Hirst, 1981; Hopwood,1972, 1974; Imoisili, 1989; Otley, 1978; see Hart-mann, 2000 for a review), dimensions of informa-tion such as scope, timeliness and aggregations(Chenhall & Morris, 1986; Gordon & Narayanan,1984; Larcker, 1981), sophisticated capital bud-geting (Haka, 1987; Larcker, 1983), cost con-sciousness (Shields & Young, 1994), competitor-focused accounting (Guilding, 1999), strategicinteractive controls and diagnostic controls(Simons, 1995), information which is related toissues concerning customers, product design, time,cost, resources and profitability which is dis-tinguished on level of detail, updating frequencyand interactive use with operational personnel(Davila, 2000).

3.1. Critical evaluation

Overall, assessing findings from contingency-based research involves judging how the results

2 Since 1980 several commentators have provided critiquesof contingency research in management accounting based ontheir perceptions of shortcoming in prior studies (Chapman,1997; Covaleski, Dirsmith, & Samuel 1996; Fisher, 1995, 1998,Moores & Chenhall, 1994; Otley, 1980; Otley & Wilkinson,1988; and for a more general review of empirical research inMCS see Ittner & Larcker, 2000). In this paper, the main criti-cisms concerning variable definition and measurement are con-sidered within the critical evaluation of the contingencyvariables. Several authors note that contingency research hasnot considered interpretive and critical views of the world.These issues are examined in the final section of the paper.

R.H. Chenhall / Accounting, Organizations and Society 28 (2003) 127–168 129

accumulate to provide generalizable findingsconcerning MCS. As is common in many socialsciences, MCS researchers are faced with deci-sions on whether to build on an existing areaof study, such as the role of formal budgets,or identify emerging aspects of MCS, such asbalanced scorecards or target costing, and investi-gate the settings within which they may be mostbeneficial.Within the body of literature reviewed in this

paper there is a mixture of studies focused on sin-gle themes, and studies exploring unconnectedelements of MCS and context. Both types ofstudies are required. Studying the role of novelMCS practices within contemporary settings isnecessary to ensure that MCS research is relevant.Given that many dimensions of MCS and theircontexts change, novel studies will always berequired to address emerging issues (Atkinson,Balakrishnan, Booth, Cote, Groot, Malmi,Roberts, Uliana, & Wu, 1997). There is a pressingneed for studies into situations in which con-temporary MCS may be best suited. Research isjust starting to be published identifying con-tingencies surrounding the design and imple-mentation of ABC/ABM (Anderson & Young,1999; Krumwiede, 1998). However, there is verylittle published contingency work on balancedscorecards, target costing, life cycle costing, thebroad array of non-financial performance indic-tors including those related to assessing humanresource management initiatives. Examples of thelatter include measurement to guide and evaluatethe learning capabilities of the organization, mea-sures such as team maturity indexes and organi-zational climate surveys that attempt to assess thee!ectiveness of administrative innovations. Recentdevelopments in global operations, strategic riskmanagement (including real options), corporatesocial reporting (including triple bottom line),economic value analysis, 360 degree performanceevaluation, knowledge-based organizations, for-ensic accounting, intellectual capital, competitoraccounting and value chain are only just beginningto be understood by researchers. There is a needfor more research into service and not-for-profitorganizations as these entities become increasinglyimportant within most economies

Notwithstanding the importance of studyingcontrols that are relevant to contemporary set-tings, it is important to develop knowledge inways that ensure coherence in the study of ele-ments of MAS and contextual variables, and inthe findings of these studies. Such confidence canbe derived from replication studies which enhancethe validity and reliability of findings and therebyprovide a strong base to move forward by way ofmodel development (Lindsay, 1995). Commenta-tors have been critical that in most areas of MCSresearch, studies have not developed su"cient‘critical mass’ to confirm findings.In some areas of MCS that have attracted a

substantial research e!ort, such as RAPM, varia-tion in dimensions of variables across studies anddi!erent measures of the variables have inhibitedthe coherent accumulation of findings (Hartmann,2000; Kren & Liao, 1988). The way in whichstudies evolved within the area of RAPM helpsillustrate several di"culties in isolating the mean-ing and measurement of MCS variables. First, theprecise meaning of the concept of RAPM has beenconfused by lack of definition of what is account-ing and non-accounting and what is reliance(Hartmann, 2000). Given the ambiguity with theconcept, it is not surprising that researcherssought to gain clarification by modifying theirstudies as understanding of RAPM and itsmeasurement developed. Such refining of conceptsand measurement is common in other social sci-ences, such as psychology. It is unfortunate that itis not part of the MCS research tradition to spendmore time on developing robust measures of theelements of MCS, particularly when there isambiguity in the meaning of constructs. Forexample, it is not clear how balanced scorecardsshould be measured. It seems likely that the con-tent and implementation of balanced scorecardsvary widely between organizations. It would seemuseful to develop a valid measure of balancedscorecards that could then be used by researchersto explore its context. While such a valid measurewould enhance consistency between studies, a dif-ficulty exists in the dynamic nature of MCS prac-tices. MCS that are valid today may lose validityas they evolve through time. Certainly, because ofadvances in information technology (IT) software,


some types of balanced scorecards being employedtoday are more comprehensive and strategic innature than those being used 5 years ago. Simi-larly, the concept of RAPM and how it relates tobroader controls has changed since the early workin the 1970 and 1980s. Without accommodatingchanges in contemporary control systems, con-cepts and measures of MCS are unlikely toaddress pertinent, contemporary issues. Aresearch climate that encouraged the developmentof valid concepts and measures of MCS wouldhave to recognize the need for modification toincorporate the evolution of MCS.Participative budgeting has also been studied

widely. Unlike RAPM, participation in budgetshas almost universally been conceptualized andmeasured following Milani (1975). In some stud-ies, additional measures are employed to providesome validation of the primary measure (Brownell& McInnes, 1986). Other studies of budget relatedbehaviour have drawn on attitudes and satisfac-tion with budgets, as developed by Swieringa andMoncur (1975). There have been a considerablenumber of studies that have confirmed themeasurement of the generic MCS characteristicsof broad scope, timeliness, aggregation, and inte-gration. These studies have employed conceptsand measures developed by Chenhall and Morris(1986), sometimes with minor adjustments to suitthe particular setting, and appear to be robustacross a variety of settings. However, there hasbeen little replication or coherence in measure-ment development in studies examining MCSpractices of contemporary interest such as static-flexible budgets, non-financial performance mea-sures, activity-based accounting, competitor-focused accounting, and product developmentinformation. Similarly, while studies haveexplored important areas of MCS such as socialcontrols, personnel control, sophisticated inte-grative mechanisms, administrative controls,interpersonal controls, sophisticated controls,there has been very little replication.A further criticism related to the nature of

accounting controls within contingency-basedresearch is that these form only part of broadercontrol systems (Chapman, 1998; Merchant,1985b; Otley, 1980, 1994). Contingency-based

research has focused on specific elements ofaccounting controls, generic information dimen-sions of MCS, with a limited number of studiesexamining broader elements of control, such asclan and informal controls, or integrativemechanisms. A di"culty in studying specific ele-ments of MCS in isolation from other organiza-tional controls is the potential for serious modelunderspecification. Thus, if specific accountingcontrols are systematically linked with otherorganizational controls, studies that exclude or donot control for these elements within the researchmethod may report spurious findings. For exam-ple, a study focused only on formal budget sys-tems may argue that they are unsuitable inuncertain operating conditions as they includeincomplete information and lack flexibility. How-ever, evidence may indicate that successful orga-nizations rely extensively on formal budgets. Thisunexpected finding occurs as a consequence oflimiting the study to budgets without consideringbroader control and information networks. It maybe that successful organizations operating inuncertain conditions have formal budgets but theyare systematically combined with open and flexibleinformal communications between managers. Theformal budgets are useful in assisting planning andcurbing excessive innovation, while the informalcommunications provide broader information inflexible ways. Simons (1987, 1991,1995) showedthat formal budgets can provide interactivecontrols in uncertain conditions whereby thebudgets generate intelligence data to buildinternal pressure to break out of narrow searchroutines and encourage the emergence of newstrategic initiatives. Chapman (1998) also arguesthat in uncertain conditions e!ective organizationscan employ formal accounting but they shouldtake place within a situation that involves intenseverbal communication between organizationalgroups.A way of addressing these concerns is to identify

a variety of control taxonomies and consider howthey relate to various aspects of MCS. One suchtaxonomy involves classifying controls as rangingfrom mechanistic to organic. Mechanistic controlsrely on formal rules, standardized operating pro-cedures and routines. Organic systems are more


flexible, responsive, involve fewer rules and stan-dardized procedures and tend to be richer in data.3

Table 1 provides a grouping of elements of MCSand control types commonly found in research, interms of the organic or mechanistic nature ofcontrol.These taxonomies are useful for addressing

concerns of how MCS relate to broader controlsystems and can guide research into how parti-cular aspects of MCS are consistent with the con-trol ‘culture’ of organizations.Finally, it should be noted that there is a dis-

tinction between the adoption of MCS and theimplementation of the systems. Much can belearned about the success or otherwise of MCS byexamining how the control culture, organic ormechanistic, influences the processes of imple-mentation. This becomes particularly importantwhen studying the adoption of innovative MCSsuch as activity-based accounting and balancedscorecards which are closely linked to the organi-zation’s control culture (Anderson & Young,1999; Gosselin, 1997; Krumwiede, 1998; Shields,1995) and the extent of change in MCS, in general(Libby & Waterhouse, 1996).It seems clear that broader issues of control are

likely to have implications for research intounderstanding MCS design. There have beenadvances over the past 20 years in demonstratingthe importance of considering managementaccounting practices as aspects of MCS. Under-standing how specific aspects of managementaccounting relate to broader control concepts, as

outlined in Table 1, assists in researching thecomplementary or substitution e!ects of non-accounting controls. An important part of theresearch agenda is to understand how combina-tions of controls can be combined, to suit theparticular circumstances of the organization(Fisher, 1995). In studying broad controls, it isnecessary to be aware of the boundaries that someorganizations and accountants place around MASand MCS. Without such awareness, there can beconfusion as to what is formal accounting control,what is structural control, and what are personneland informal controls.

4. Outcomes of MCS

Outcomes may be separated into issues relatedto the use or usefulness of the MCS, behaviouraland organizational outcomes. There is an impliedconnection between these outcomes. If the MCSare found to be useful then they are likely to beused and provide satisfaction to individuals, whothen presumably can approach their tasks withenhanced information. As a consequence, theseindividuals take improved decisions and betterachieve organizational goals. Clearly, there arebroad leaps in logic from useful MCS, toimproved job satisfaction and enhanced organiza-tional performance. Moreover, there is no com-pelling evidence to suggest that such links exist.Even within contingency-based research, the linkbetween enhanced organizational performanceand usefulness of some aspect of MCS may welldepend on the appropriateness of the useful MCSto the context of the organization.Considerations of interest to designers and

researchers of MCS have been the extent to whichthe systems provide information (Mia & Chenhall,1994), the degree of use (Abernethy & Guthrie,1994; Anderson & Young, 1999; Foster & Swen-son, 1997; Guilding, 1999), the usefulness of theinformation (Chenhall & Morris, 1986; Shields,1995) or the beneficial nature of the MCS (Chen-hall & Langfield-Smith, 1998b); the importance inmaking operational decisions (Bouwens & Aber-nethy, 2000); importance to product development(Davila, 2000), whether they are helpful to the

3 Several authors provide an elaboration of mechanistic andorganic control. Perrow (1970) distinguishes mechanistic fromorganic controls on the basis of manager’s discretion, power,and coordination within groups and interdependence betweengroups. Organic controls involve higher discretion and power,coordination by mutual adjustment and high interdependencebetween work groups. Ouchi (1977, 1979) identifies marketcontrols (prices), mechanistic formal bureaucratic controls(rules to control output of work and the behaviour of workers),and organic, informal clan controls (recruitment, traditions andceremonial control). Galbraith (1973) refers to mechanisticcontrols as rules, programmes and procedures, hierarchy, andgoal setting; and organic controls as creating slack resources,self-contained tasks, vertical information systems and lateralrelations.


organization (Guilding, 1999), and satisfactionwith the systems (Bruns & Waterhouse, 1975).Behavioural outcomes such as job satisfaction

have been important in human resource manage-ment. The provision of a work place environmentto enhance employee welfare or job satisfaction isseen by some as a worthwhile goal in its own right.Moreover, other things equal, it may be presumed

that individuals who are satisfied with their jobswill identify with organizational goals and workmore e!ectively. Interestingly, there have not beenmany MCS studies that have examined the e!ectsof MCS on job satisfaction (Banker, Potter, &Schroeder 1993; Brownell, 1982; Chenhall, 1986).A variety of studies have examined the e!ect ofMCS on job related tension or stress (Brownell &

Table 1Organic and mechanistic forms of MCS

More organic

Clan controls (Govindarajan & Fisher, 1990; Ouchi, 1980; control cultures and norms)Social controls (Merchant, 1985b, self and group controls), (Rockness & Shields, 1984; input controls—social controlsand budgets).Personnel controls (Merchant, 1985b, selection, training, culture, group rewards, resources), (Abernethy & Brownell, 1997,socialization and training)Sophisticated integrative mechanisms (Abernethy & Lillis, 1995; task forces, meetings, etc.)Prospect controls (Macintosh, 1994; focus on plans and the future, infrequent and general reporting)MCS that provide broad scope information, flexible aggregations and integrative information, and information provided in atimely way (Chenhall & Morris, 1986)Static/flexible budgets (Brownell & Merchant, 1990; flexibility of budgets to volume changes)Participative budgets (Shields & Shields, 1988; involvement of subordinates in setting budgets)Low reliance on accounting controls (Brownell, 1982, 1987; Hirst, 1981; use of more profit oriented controls or non-accounting)Budget slack (Dunk, 1993; Merchant, 1985a; excess resources over that needed to complete tasks e"ciently)Competitor-focused accounting (Guilding, 1999; competitor cost assessment, position monitoring and appraisal, strategiccosting and pricing)Strategic interactive controls (Simons, 1995; use of performance evaluation for strategic planning )Product development information (Davila, 2000; levels of detail, frequency of updating and pattern of usage for informationrelated to product cost and design, time related, customer related, resource inputs, profitability)

More mechanisticBudget constrained performance evaluation style (Hopwood, 1972; high emphasis on cost budgets)Budget control (Rockness & Shields, 1984)High reliance on accounting controls (Brownell, 1982,1987; Hirst, 1981; accounting for performance evaluation)High budget use (Bruns & Waterhouse, 1975; Merchant, 1981; importance, involvement, time spent on budgets)Narrow scope (Chenhall & Morris, 1986; financial, internal, historic)Sophisticated capital budgeting (Haka, 1987; Larcker, 1981; DCF, etc.)Sophisticated controls (Khandwalla, 1972; standard costing, incremental costing, statistical quality control, inventory control)Operating procedures, budgets and statistical reports (Macintosh & Daft, 1987).Administrative use of budgets (Hopwood, 1972; Merchant, 1981; importance of meeting budget, formality of communications,systems sophistication and participation)Inter personnel controls (Bruns & Waterhouse, 1975; lack of formal controls but centralization, lack of autonomy, pressureinducing actions by superiors)Output and results controls (Macintosh, 1994; Merchant, 1985b; outcomes or e!ectiveness)Behavior controls (Merchant, 1985a; Ouchi, 1979; Rockness & Shields, 1984; standardization, rules, formalization)Patriarchal control (Whitley, 1999; personal and informal, centralized control from the top)Action controls (Merchant, 1985b); process controls (manufacturing performance measures), (Chenhall, 1997; direct measuresof production processes)Diagnostic controls (Simons, 1995; use of control to provide feedback on operations)


Hirst, 1986; Hirst, 1983; Hopwood, 1972; Shields,Deng, & Kato, 2000). Unlike job satisfaction,stress appears to be more closely related to thenature of the MCS and is implicated in associa-tions with performance (Shields et al., 2000).Organizational outcomes in contingency-based

research have been dominated by self assessmentprocesses where individuals provide an indicationof their performance, or their organizational unit,across a range of potentially important managerialprocesses (e.g. Mahoney, Jerdee, & Carroll, 1963)or goals of the organization (Govindarajan, 1984).The issue of the validity of self-assessment isoften raised as a concern. Evidence suggests thata subordinate’s self-assessment correlates withobjective assessments (Bommer, Johnson, Rich,Podsako!, & Mackenzie, 1995; Venkatraman &Ramanujam, 1987) and with a superior’s sub-jective assessment (Furnham & Stringfield, 1994,Heneman, 1974; Riggo & Cole, 1992). Notwith-standing this evidence, it is always reassuringwhen a superior’s performance rating of therespondent is included in the study.Curiously, there has been little work relating

MCS change to share prices. Larcker (1983) foundthat firms adopting incentive performance plansexperienced an increase in capital investment anda positive security market reaction on disclosureof the plan to the market. Gordon and Smith(1992) reported that returns to investors werehigher for firms employing post completionreviews when matched with asymmetric informa-tion, capital intensity, capital expenditure andinsider ownership. Smith (1993) identified thatpositive returns were associated with post com-pletion reviews in abandonment decisions.McConnell and Muscarella (1985) report positiveassociations between announcements of increasesin capital investment plans, MCS and share pricemovement. However, Gordon and Silvester (1999)found no significant association between theinstallation of ABC and significant stock marketreaction. This poses the question as to whetherimproved understanding would follow fromstudying these associations within a variety oforganizational contexts? These studies do notemploy contingency-based approaches as theyexplore only the main e!ects between share price

movement and the adoption of elements of MCS.(Studies often examine industry e!ects and theimportance of capital expenditure). Progress inthis area may be limited due to the di"culties inextracting the e!ects of adopting di!erent MCSon share prices from other events that may beassociated with share price movements. With thenumerous possible events e!ecting share prices,control problems can become acute. Also, datacollection is complicated by the need to collectdata on the adoption and implementation of MCSby survey methods and to then match these withshare price changes. Also, perhaps the lack ofresearch in the area says something about the dif-ferent types of training between researchers infinance and management accounting.

4.1. Critical evaluation

Contingency-based studies have examined MCSas both dependent and independent variables. Toexamine fit between MCS and context, some com-mentators have claimed that the outcome variablesshould be some dimension of desired organizationalor managerial performance (Otley, 1980; Otley &Wilkinson, 1998). Good fit means enhanced per-formance, while poor fit implies diminished per-formance. While the ultimate goal of MCSresearch is to provide findings that assist managersachieve their goals or those of their organization,MCS research has continued to include dimen-sions of MCS, their use and usefulness, as theoutcome variable. Also, it is noteworthy that per-formance has been included as an independentvariable explaining some characteristics of MCS(see Langfield-Smith 1997, p. 226 for a discussion).While not explicit in most studies with MCS as

the outcome variable, it is implied that associa-tions between context and MCS reflect equili-brium conditions, or indicate optimal solutionsbecause of survival of the fittest conditions. Ifequilibrium is assumed, then studying perfor-mance is inappropriate as every firm has optimalperformance given its situation.Researchers studying MCS as the outcome

variables, also, note that this approach is justifiedby assuming that rational managers are unlikely toadopt or use MCS that do not assist in enhancing


performance. (An alternate view is that managersmay adopt MCS for institutional or political rea-sons that may be inconsistent with rational eco-nomic reasons.) Alternatively, some argue thatlinks between MCS, context and performance canbe tenuous as they involve many factors concern-ing the quality of managing the production pro-cesses (Birnberg et al, 1983; Kren & Liao, 1988).In general, if disequilibrium conditions areassumed, then it may be useful for contingency-based studies to first establish adoption and use ofMCS, then to examine how they are used toenhance decision quality and finally investigatelinks with organizational performance.Care is required when interpreting studies that

have outcome variables related to the characteristicsof MCS, such as ‘use’ and ‘usefulness’ of the systems.Individuals may be forced to use MCS systems, suchas budgets or DCF analysis in decision making,even though they find them of little use. Also,linking ‘use’, ‘usefulness’, ‘benefits’ or ‘satisfac-tion’ to organizational e!ectiveness is potentiallyproblematic. A particular MCS may be perceivedas not useful, rate low in satisfaction or benefitsbut organizational performance may be high dueto the supply of required information from othersources, either formal or informal. These issuescan be resolved by careful attention to theresearch question. It is quite legitimate to studythe adoption of systems and their use. However, itmay not be appropriate to claim that these out-comes are of value in improving organization per-formance. Similarly, the extent to which MCS areperceived as useful may not imply improved orga-nizational performance. If studying one aspect ofthe MCS in isolation from other sources of infor-mation, researchers should ensure that the studiedattribute is the sole source of the informationbeing studied. If an aspect of the MCS is beingconsidered within situations that include broaderinformation and controls, the potential influenceof these other controls should be included or con-trolled within the research design.In summary, despite the critique that con-

tingency-based studies should include organiza-tional performance as the dependent variable,studies still follow approaches with MCS as thedependent variable. Care in theory construction is

required in following either approach. Studies canprovide important insights into the extent ofadoption, use and usefulness of MCS, however, itshould not be assumed that the models necessarilylead to enhanced organizational performance.Similarly, if performance is the dependent variablethen compelling theory is required to show howthe combination of MCS and context enablemanagers to take more e!ective decisions thatenhance organizational performance.Given the assumption that research should

identify organizational performance as the criter-ion variable, a critical issue is what constitutesorganizational performance? Distinguishing o"-cial and operative goals would seem an essentialaspect of MCS research that includes considera-tion of goals, mainly as it flags that the issue oforganizational goals is far from unproblematic(Perrow, 1970). Investigating these goals requiresa dynamic approach that examines the goal for-mulation process. There are several issues thatbecome important. First, goal formulation orchange often involves the influence of new power-ful players, either within or outside the organiza-tion, who can dramatically change o"cial goals.MCS can act either as a tool to e!ect such changesor hinder their acceptance within the organization.For example, a new Chief Executive O"cer maystipulate that improved shareholder value is apriority. Consequently, performance measurementbased on Economic Value Analysis may be intro-duced in an attempt to align the actions of allemployees with the single objective of improvingeconomic value. Second, changes in the areas oforganizational e!ectiveness can redirect goals tothose areas of e!ectiveness. The unplanned dis-covery of a new technology that potentiallyincreases throughput can result in the adoption of‘timely’ delivery as a goal of the organization.Third, it is apparent that the measurement of

goals can have explicit e!ects on goal formulation,both intended and unintended. Goals may beselected or evolve as they can be measured readilyby the MCS. A preoccupation with formal, ‘hard’measures may direct attention to those measuresat the expense of the subtleties of the situation.For example, measuring aspects of customer oremployee satisfaction, the organizational culture


or intellectual capital often require more sub-jective assessments of progress and, as such, mayreceive less attention than activities subjected tohard measures such as production rejects orthroughput.Fourth, in addition to influencing types of goals,

MCS may a!ect goal achievement by establishingstandards or benchmarks for performance. Goalsthat are too hard may cause frustration and with-drawal, while standards that are too easy may notprovide su"cient challenge. Standards that areachievable but with su"cient stretch to provide achallenge are often recommended as ideal. How-ever, in today’s environment of intense competi-tion and global operations, requirements forsubstantial continuous improvement may meanthat di"cult standards based on continuousimprovements are required to survive. Perfor-mance measures can readily establish targets thatrequire continuous improvement.Fifth, recently many organizations have recog-

nized the need to satisfy multiple and potentiallycompeting goals. Mission statements identify therequirements to attract and maintain share-holders, employees, and customers and to do so inways that are socially acceptable. Accountantshave responded by refining triple bottom linereporting, environmental accounting, social cor-porate reporting and corporate sustainability(Epstein & Birchard, 2000).Sixth, aligning operative goals with o"cial goals

is an important aspect of strategic management.This is the essence of performance hierarchies andbalanced scorecards that attempt to capture theinteractive e!ects of o"cial goals associated withthe interests of shareholders, customers, the inter-nal processes and the potential for the organiza-tion to sustain itself by learning and innovation.Moreover, these types of performance manage-ment methods attempt to align strategy withoperations by translating o"cial goals into opera-tive goals and cascading the latter down throughthe organization. Of course, connections betweeno"cial and operative goals can be quite di!erentin similar organizations. Achieving shareholderwelfare might require organizations to follow dif-ferent operative goals concerning decisions onquality, cost, delivery and the like.

Linkages between MCS and organizationalgoals are quite explicit, as a primary function ofMCS is to measure progress towards achievingdesired organization ends. It is a useful exercisewhen evaluating characteristics of MCS used forreporting on goals to judge the extent to whichthey accommodate the following: consider multi-ple stakeholders; measure e"ciency, e!ectivenessand equity; capture financial and non-financialoutcomes; provide vertical links between strategyand operations and horizontal links across thevalue chain; provide information on how theorganization relates to its external environmentand its ability to adapt. Presumably, balancedscorecards or performance hierarchies provide amethodology to tackle many of these issues. Thecomplexity of achieving these expectations mayhelp to explain why many firms that attempt toadopt balanced scorecards have di"culty inimplementing them.

5. Contextual variables and MCS

Before examining the contextual variables, adistinction is noted between generic and specificdefinitions. When considering environment, spe-cific definitions refer to particular attributes suchas intense price competition from existing orpotential competitors, or the likelihood of achange in the availability of materials. Genericdefinitions attempt to capture the e!ects of specificattributes in a more generalized way. Genericdefinitions enable designers and researchers ofMCS to discuss the influence of contextual vari-ables without having to identify the particulardetails of individual organizations. Constructingtaxonomies of context and theories relating theseto the use of MCS and organizational outcomesbecomes more tractable. Clearly, to make pre-scriptive recommendations to a particular organi-zation it is necessary that the specific attributes ofthe environment be identified. Moving betweenthe generic and specific should not be problematicproviding the generic definitions are robust.Chapman (1997) provides a discussion of thetrade-o!s between simplicity, accuracy and gen-eralizability in variable definition


5.1. The external environment

The external environment is a powerful con-textual variable that is at the foundation of con-tingency-based research. Perhaps the most widelyresearched aspect of the environment is uncer-tainty. Early contingency research in organizationdesign focused on the e!ects of uncertainty onorganizational structure. Examples include Burnsand Stalker (1961), Galbraith (1973), Lawrenceand Lorsch (1967) and Perrow (1970). It isimportant to distinguish uncertainty from risk.Risk is concerned with situations in which prob-abilities can be attached to particular eventsoccurring, whereas uncertainty defines situationsin which probabilities cannot be attached and eventhe elements of the environment may not be pre-dictable. The importance of uncertainty as a fun-damental variable in MCS contingency-basedresearch has been stressed recently by Chapman(1997) and Hartmann (2000). Both reinterpretaspects of MCS research by examining the impactof environmental uncertainty.Uncertainty and risk do not provide a compre-

hensive description of the environment. Khand-walla (1977) provides a useful taxonomy ofenvironmental variables. These include turbulence(risky, unpredictable, fluctuating, ambiguous),hostility (stressful, dominating, restrictive), diver-sity (variety in products, inputs, customers), com-plexity (rapidly developing technologies). Otherelements of the environment that may generatepressure or provide opportunities include com-plexity and dynamism (Duncan, 1972), [simple-complex and static-dynamic (Waterhouse & Ties-sen, 1978)], controllable and uncontrollable(Ewusi-Mensah, 1981), ambiguity (Ouchi, 1979)or equivocality (Daft & Macintosh, 1981).

5.1.1. Findings: external environment and MCSIn accounting, uncertainty has been related to

the usefulness of broad scope information (Chen-hall & Morris, 1986; Chong & Chong, 1997; Gor-don & Narayanan, 1984; Gul & Chia, 1994) andtimely information (Chenhall & Morris, 1986);performance evaluation characterized by a moresubjective evaluation style (Govindarajan, 1984;Moores & Sharma, 1998); less reliance on incen-

tive-based pay (Bloom, 1998), non-accountingstyle of performance evaluation rather than eithera budget constrained or profit oriented style(Ross, 1995) and participative budgeting (Govin-darajan, 1986).4 Functional area, particularlyresearch and development (seen as facing higherenvironmental uncertainty compared to market-ing) combined with budgetary participation wasshown to enhance performance (Brownell, 1985).Some evidence suggests combinations of tradi-

tional budgetary controls and more interpersonaland flexible controls in conditions of environ-mental uncertainty. Ezzamel (1990) reported thathigh environmental uncertainty was associatedwith an emphasis on budgets for evaluation andrequired explanation of variances but also highparticipation and interpersonal interactionsbetween superiors and subordinates. Merchant(1990) found that environmental uncertainty waslinked to pressure to meet financial targets butthere was some flexibility by way of highermanipulation of information. In a study of fourcases, Chapman (1998) proposed that accountinghas a planning role to play in conditions ofuncertainty but there must be substantial interac-tions between accountants and other managers tocope with changing conditions as they unfold inunpredictable ways.Environmental hostility (di"culty) has been

associated with a strong emphasis on meetingbudgets (Otley, 1978). Hostility from intensecompetition has been related to a reliance on for-mal control (Imoisili, 1985) and sophisticatedaccounting, production and statistical control(Khandwalla, 1972). However, certain specific ele-ments of competitive position, such as strength ofmarket position and stages in product life cycleswere not associated with the importance of bud-gets or participation (Merchant, 1984). Also,environmental complexity (but only when derivedfrom suppliers and government), independent offunction, was associated with a reduced emphasison budgets (Brownell, 1985).From these illustrations it can be seen that a

consistent steam of research over the past 20 years

4 The theory used by Ross (1995) examines task uncertaintybut the study measures environmental uncertainty


has confirmed that uncertainty has been asso-ciated with a need for more open, externallyfocused, nonfinancial styles of MCS. However,hostile and turbulent conditions appear, in themain, to be best served by a reliance on formalcontrols and an emphasis on budgets. The ques-tion may be posed, what is the appropriate MCSfor organizations operating in conditions ofuncertainty, turbulence and hostility? The organi-zational design literature proposes that organiza-tions facing extreme pressure will initially tightencontrol as such pressure is likely to threaten short-term survival and then adopt more organic con-trols (Khandwalla, 1977). Little is known aboutthe appropriate design of MCS to assist in mana-ging complex and competing forces from theexternal environment. It would be useful to exam-ine how contemporary, interactive informationsystems can provide a blend of tight controls withthe opportunity to source more open, informaland subjective information. Certainly there is evi-dence that e!ective organizations combine tightcontrols with more open, informal and flexibleinformation and communication systems (Chap-man, 1998; Chenhall & Morris, 1995; Simons,1987).The following propositions summarize the

research findings relating MCS to the externalenvironment.Propositions concerning the external environment

and MCS:

The more uncertain the external environmentthe more open and externally focused the MCS.

The more hostile and turbulent the externalenvironment the greater the reliance on formalcontrols and an emphasis on traditional budgets.

Where MCS focused on tight financial controlsare used in uncertain external environmentsthey will be used together with an emphasis onflexible, interpersonal interactions.

5.1.2. Critical evaluationThe distinction between dimensions within the

external environment, such as uncertainty, hostilityand complexity are important to MCS design.More mechanistic, formal MCS tend to provide

incomplete information in uncertain conditionsand require rapid reformulation to cope with theunfolding unpredictability. However, in a complexsituation there is a need for more informationwithin the MCS but once designed the systemsshould be su"cient to assist in taking and imple-menting decisions. Clear specification of theenvironmental dimensions of interest is required,as di!erent theories are required to consider thee!ects of di!erent dimensions. There are richresearch opportunities to investigate appropriateMCS design for settings that are uncertain andalso hostile and complex.Interpreting studies that have examined the

influence of the external environment is compli-cated by the use of di!erent measures of the sameenvironmental construct. For example, Gordonand Narayanan’s (1984) study of the associationbetween perceived environmental uncertainty andmore broadly scoped MCS used a measure ofuncertainty that captured the intensity of compe-tition, the dynamic and unpredictable nature ofthe external environment, and elements of change.In studying the same type of MCS variables,Chenhall and Morris (1986) used a measure ofuncertainty which considered lack of informationon environmental factors, inability to assignprobabilities on how the environment will a!ectsuccess or failure, and not knowing the outcomeof decisions on how the organization would lose ifthe decision was incorrect. The measure used byGordon and Narayanan (1984) is more specificallyfocused on the external situation than Chenhalland Morris (1986) which has a composite ofexternal components and implications for internaldecisions. Even within the Gordon and Narayanan(1984) measure elements of unpredictability arecombined with di"culty. Tymond, Stout, andShaw (1998) provide a comprehensive review ofMCS research investigating the role of environ-mental uncertainty providing recommendationsthat the measures should involve top managers’perceptions of the external environment. Theapplication of a single valid and reliable measureof environmental uncertainty would assist in com-paring the results of studies examining uncertaintyand help build a coherent body of knowledge onthe e!ects of this variable of MCS design.


The environment will continue to be a centralelement of context in contingency-based research.The specific attributes of the environment arechanging and should be included in future studies.The external environment that most organizationsface includes increased social pressure on issuessuch as environmental ecology and the economicand social well being of employees and society.The implications for management and MCS ofglobal competition and operations are increasinglyimportant. As organizations become involved innetworks involving other entities such as jointventures and supplier and customer alliances, theboundaries between what is internal and externalbecome blurred and consequently the role of MCSwill likely change. Additionally, the way in whichthe environment exerts pressure should beexplored. Granlund and Lukka (1998) note thatpressure may come from economic causes, coer-cion from institutions, normative pressure derivedfrom appropriate social conduct, and the tendencyto mimic apparently successful practices.

5.2. Generic concepts of technology

Technology has many meanings in organizationbehavior. At a general level, technology refers tohow the organization’s work processes operate(the way tasks transform inputs into outputs) andincludes hardware (such as machines and tools),materials, people, software and knowledge. Threegeneric types of technology of importance to MCSdesign maybe identified from the organizationalliterature: complexity, task uncertainty and inter-dependence.5

Using these notions of technology, several keyattributes that may influence MCS design can bederived. First, organizations producing highlyspecialized, non-standard, di!erentiated productsare likely to employ complex unit/batch technolo-gies. These will tend to involve processes that havelow analyzability of processes and many excep-tions. Also, managers are likely to have imperfectknowledge of processes and low ability tomeasure outputs. A need for flexible responses tospecific customers increases interdependenciesacross the value chain involving reciprocal inter-actions with customers, suppliers and functionalunits such as marketing, production, purchasingand research and development. It might be expec-ted that these types of technologies would requirecontrols to encourage flexible responses, highlevels of open communication within the workforce and systems to manage the inter-dependencies. Traditional, mechanistic MCSbased on financial controls would not seem to suitthese circumstances.Second, organizations that produce standard,

undi!erentiated products employing capitalintensive, automated processes are likely toemploy mass production and process technolo-gies. These will involve highly analyzable pro-cesses and few exceptions. Knowledge ofprocesses and measures of output will be morereadily available. Interdependencies are moderatebeing sequential. This technology requires stan-dardized, administrative controls such as tradi-tional, formal financial MCS. A variant of thistechnology is where there are non-standard pro-ducts but the processes are well understood.Interdependencies with customers are likely to bereciprocal. This technology is typical of an orga-nization producing customized products butemploying reasonably automated processes. Con-trols are required that are flexible and that areconsistent with managing interdependencies. Areliance on traditional administrative controls,including financial MCS, is unlikely to providerequired flexibility and more open, informalcontrols will be more suitable to manage inter-dependencies. At the same time traditional, formalcontrols may assist in controlling processes thatare well understood.

5 Complexity derives from standardization of work, withlarge-batch and mass production (e.g. highly automated fac-tories), process and small-batch unit technologies representingincreasing levels of complexity (Woodward, 1965). Task uncer-tainty refers to variability in tasks and the analyzability ofmethods of performing the tasks with high variability andunanalyzable tasks inducing control di"culties and a need formore organic controls (Perrow, 1970). or by the knowledge oftransformation processes and predictability in measuring outputs(Ouchi, 1979). Interdependence increases the level of coordinationdi"culties, and has implications for control systems, as the inter-dependencies move from pooled (no direct relationship betweenadjacent processes), to sequential (one-way interdependencies), toreciprocal (two-way interdependencies) (Thompson, 1967).


5.2.1. Findings: standardized-automated processesand MCSTechnologies characterized by more (less) stan-

dardized and automated processes are served bymore (less) traditional formal MCS with highly(less) developed process controls (Khandwalla,1977); high (low) budget use (Merchant, 1984) andhigh (low) budgetary controls (Dunk, 1992). Highbudgetary slack provides a bu!er against low pre-dictability within the processes and is found less inmore predictable, automated processes with highworkflow integration (Merchant, 1985a). Alter-natively, slack will be positively related to lessautomated, less predictable job/batch type tech-nologies.

5.2.2. Task uncertainty and MCSTechnologies with high (low) task analyzability

are related to a high (low) reliance on standardoperating procedures, programs and plans (Daft& Macintosh, 1981); tasks high in di"culty andvariability are associated with a low reliance onaccounting performance measures (Hirst, 1983);knowledge of task transformations is associatedwith behavior control (but only limited supportwas found for relationships between measurabilityof output and control systems; Rockness &Shields, 1984); technologies with few (many)exceptions that are high (low) in analyzability areassociated with accounting (personnel) controls(Abernethy & Brownell, 1997). Mia and Chenhall(1994) demonstrated that marketing departmentsfaced more task uncertainty than productiondepartments and consequently used broad scopeinformation to enhance performance. Brownelland Dunk (1991) showed that there was a fitbetween conditions of low task di"culty, partici-pative budgeting and a high budget emphasis;while high task di"culty suited participation withor without a strong budget emphasis. Lau et al(1995) provided similar results, although theyfound that high participation and high task di"-culty provided a fit irrespective of budget empha-sis, while high participation and high budgetemphasis enhanced performance in low task di"-culty situations. Brownell and Merchant (1990)found that higher (lower) standardization of pro-ducts (high knowledge of input/output relations)

combined with flexible (static) budgets and low(high) participation to enhance performance.Brownell and Merchant’s (1990) findings that lowtask uncertainty combined with more flexiblebudgets is somewhat inconsistent with other find-ings linking high task uncertainty with moreinformal, open MCS.

5.2.3. Interdependence and MCSLow levels of interdependence have been linked

to budgets, operating procedures and statisticalreports; with statistical reports used for planningand informal coordination used in highly inter-dependent situations (Macintosh & Daft, 1987). Inlow interdependent public sector organizationsthere was an emphasis on budget analysis andmanagers’ influence on budgets but infrequentinteractions with superiors and little requiredexplanation from budgets (Williams, Macintosh,& Moore, 1990). In more complex situations(reciprocal interdependencies) there was lessemphasis on budgets and more frequent interac-tion between subordinates and superiors. High(low) interdependence was found to be associatedwith broad (narrow) scope MCS that focuses (lackof focus) on appropriate aggregations and inte-grative information (Chenhall & Morris, 1986).Strategies of customization were associated withhigh levels of interdependence with the latter cor-relating with the importance for operational deci-sions of the information characteristics ofintegration, aggregation and timeliness (Bouwensand Abernethy, 2000).6

Propositions concerning generic concepts of tech-nology and MCS:

The more technologies are characterized bystandardized and automated processes the moreformal the controls including a reliance on pro-

6 Di!erences in findings between Chenhall and Morris(1986) and Bouwens and Abernethy (2000) relate to the useful-ness of broad scope and timely MCS. Concerning broad scopeMCS, perhaps the interdependence considered in operatingdecisions, as studied by Bouwens and Abernethy (2000), relatesto internal considerations, and therefore, broad scope infor-mation which tends to be focused on external informationwould not be useful. There does not appear to be any obviousexplanation for di!ering results related to timely information.


cess control, and traditional budgets with lessbudgetary slack.

The more technologies are characterized byhigh levels of task uncertainty the more infor-mal the controls including: less reliance onstandard operating procedures, programmesand plans, accounting performance measures,behaviour controls; higher participation inbudgeting; more personal controls, clan con-trols, and usefulness of broad scope MCS.

The more technologies are characterized byhigh levels of interdependence the more infor-mal the controls including: fewer statisticaloperating procedures; more statistical planningreports and informal coordination; less empha-sis on budgets and more frequent interactionsbetween subordinates and superiors; greaterusefulness of aggregated and integrated MCS.

5.3. Contemporary technologies

Over the past 20 years MCS research has devel-oped to consider the role of advanced technologiessuch as Just in Time (JIT), Total Quality Man-agement (TQM) and Flexible Manufacturing(FM) as dimensions of context. To establishthe importance of these elements of technology,accounting researchers have dawn on theoriesfrom manufacturing developed by theorists suchHayes et al. (1988), Skinner (1975) and from eco-nomics such as notions of complimentarities asmodeled by Milgrom and Roberts (1990). Youngand Selto (1991) provide a review of new manu-facturing practices and some implications for per-formance measures and incentive schemes,arguing a need to consider technology changeswithin their organizational context.Notwithstanding the importance of manu-

facturing theories, understanding the appropriatefit between MCS and advanced technologies isassisted by reflecting on the basic, generic notionsof technology addressed above. Kalagnanam andLindsay (1999) argue that JIT is best suited toopen, informal, organic forms of controls. Theyclaim that organic systems can best manage theclose linkages or coupling within JIT that cancause variability (task uncertainty due to many

exceptions) between elements of production pro-cesses (interdependence). Organic systems are alsorequired to manage the need for flexible responsesto customers, which involves coordinating reci-procal interdependencies across the value chain.Finally, JIT implies continuous improvement thatis best served by commitment to change from theshop floor which is encouraged by organicsystems.Similar arguments may be made for implement-

ing innovative MCS in TQM and FM situations.These technologies have high variability and lowanalyzability. The low analyzability derives fromthe need to continually exploit potential com-plementarities between the various elements ofTQM practices (Chenhall, 1997). (In TQM situa-tions management may strive to develop processeswith high analyzability, but the need to con-tinually balance the way the technology deliverson competing priorities makes this task di"cult toanalyze). Also, TQM and FM involve the e!ectivemanagement of interdependencies within produc-tion processes including relationships with custo-mers, suppliers and other external parties.Controls are required to encourage managers andshop floor employees to focus on the critical ele-ments of variability within the TQM programmesand to provide e!ective links across the valuechain. This information is generated at both theprocess (cybernetic type controls such as statisticalprocess controls) and strategic levels (i.e. linkingprocesses to strategic outcomes). Continuousimprovement requires access to knowledge onworld’s best practice and systems to encourageinnovation. Appropriate control systems shouldbe open and informal, include broad scope infor-mation, benchmarking, and performance mea-sures that indicate links between strategy andoperations such as balanced scorecards and stra-tegic integrative controls.

5.3.1. Findings: advanced technologies and MCSIttner and Larcker (1995) demonstrated that

product focused TQM was linked to timely pro-blem solving information and flexible revisions toreward systems. They found for advanced (hol-istic) TQM, external benchmarking and theintegration of quality and strategic information


are important. Ittner and Larcker (1997) examinedthe association between quality programmes and avariety of strategic controls related to imple-mentation, internal and external monitoring.Links between quality and strategic controls werefound, with di!erences between countries. Also,sample-wide performance e!ects were restricted tocontrols concerning managers participation inapproving quality programmes and team for-mulation, with other associations contingent onindustry e!ects. Sim and Killough (1998) foundthat customer and quality performance was higherin TQM and JIT situations where there were cus-tomer and quality related performance goals andincentives compared to where fixed pay was used.Ittner, Larcker, Nagar, and Rajan (1999) reportedthat performance gains from supplier partnershippractices were associated with extensive use ofnon-price selection criteria, frequent meetings andinteractions with suppliers and supplier certifica-tion. These controls were not e!ective for armslength supplier relations. Kalagnanam and Lind-say (1999) found that organic MCS were asso-ciated with e!ective JIT systems.Some studies have examined the role of non-

financial performance measures in advanced tech-nologies. Banker et al. (1993) found that JIT,quality and teamwork were associated with theprovision of nonfinancial, quality and productivitymeasures to shopfloor employees. There is someevidence suggesting that relying on non-financialmeasures to evaluate managers in TQM situationsprovides interactive strategic control (Chenhall,1997). Mia (2000) found that the provision ofbroadly based MCS enhanced organizational per-formance in JIT settings. The broad MCS inclu-ded performance targets related to non-financialmanufacturing indicators, actual performance onthose targets, organizational financial indicatorsand industry and organizational trends on overallperformance. Customer focused manufacturing,together with advanced manufacturing technology(AMT), have been associated with non-financialmeasures (Perera, Harrison, & Poole 1997). It isnoteworthy that there is ambiguity in findingsrelated to the extent to which associations betweenusefulness of non-financial performance measuresand advanced technologies are related to

enhanced performance. For example, Chenhall(1997) found positive performance e!ects betweencombinations of non-financial measures andTQM, while Perera et al. (1997) did not. Oneexplanation for these di!ering findings is in the useof the performance measures. Chenhall (1997)related the measures to reward and compensationsystems, whereas Perera et al. (1997) did not makethis linkage. Perhaps the extent to which non-financial measures are used to evaluate and rewardmanagers may be important in understandinglinks between performance measures, advancedtechnologies and performance (Chenhall, 1997;c.f. Perera et al., 1997). This suggestion is con-sistent with Sim and Killough’s (1998) findingsthat incentive pay enhanced the positive e!ects ofTQM and JIT on customer and quality perfor-mance. Also, Larcker’s (1983) market-based studyfound that the combination of incentive schemesand capital investment was associated withimproved investor return. A recent study, report-ing a laboratory study, demonstrated the impor-tance of incentive schemes to enhance bothabsolute performance and rates of improvementby encouraging individuals to spend more timeon tasks and to use and analyze information(Sprinkle, 2000).Foster and Horngren (1988) found that flexible

manufacturing systems (FMS) were associatedwith performance measures focused on time,quality, operating e"ciency and flexibility. Therewas also a change in the costing methods (alloca-tions, treatment of costs as period and changes inthe components of direct costs). However, flexiblemanufacturing (FM) has been linked to a de-emphasis of e"ciency-based measures with con-trol derived from integrative liaison devices(Abernethy & Lillis, 1995). It is to be noted thatthere is a di!erence between FMS which are tech-nical systems such as computer assisted design andcomputer assisted manufacturing (CAD/CAM)and FM which is a generic notion of technologyemphasizing a strategy of flexible response andcustomization.More research is needed to explore whether

both focused formal controls at the operationallevel and more complex integrative devices cancoexist to assist control within TQM and FM


situations. Also, links between di!erent types ofcontrols for operational, managerial and strategicdecisions should be explored. For example, Chen-hall and Langfield-Smith (1998b) link perfor-mance with combinations of various traditionaland contemporary controls with a range of strate-gies and manufacturing practices.Propositions concerning advanced technologies

and MCS:

TQM is associated with broadly based MCSincluding timely, flexible, externally focusedinformation; close interactions betweenadvanced technologies and strategy; and non-financial performance measurement.

The extent to which combinations of advancedtechnologies and non-financial performancemeasures are associated with enhanced perfor-mance depends on the degree to which themeasures are used as part of reward and com-pensation schemes.

The advanced technologies of JIT and FMS areassociated with broadly based MCS such asinformal controls and greater use of non-financial performance measures.

FM is associated with the use of informal, inte-grative mechanisms.

Supplier partnership practices are associatedwith non-financial measures, informal meetingsand interactions across the value chain.

5.3.2. Critical evaluationThe three generic concepts of technology that

have been used in MCS research (complexity, taskuncertainty and interdependence) are separateconstructs but there are some common themesconcerning uncertainty. It seems likely that con-version of inputs into outputs within less complex,mass production technologies is more program-mable and predictable than in job or batch styledtechnologies servicing customized products. Highlevels of predictability are associated with thethroughput of process technologies but not for themanagement of breakdowns and other exceptions.The construct of task uncertainty concerns lack ofinformation and is a combination of variability or

lack of knowledge about alternatives and uncer-tainty about how to analyze the variations, ormeasure outputs, that occur during the conversionof materials into output. Higher levels of inter-dependence, where the work of one sub-unit iscomplicated by having to rely on another, raisesthe possibility of more uncertainty derived fromlack of control over the supplying sub-unit.The importance of uncertainty as an aspect of

both environment and technology has lead tosome ambiguity between environmental and tech-nological uncertainty in MCS research. Forexample, Hirst (1983) argued that accounting per-formance measures would be inappropriate inconditions of environmental uncertainty but mea-sured uncertainty with a composite measure com-prising both elements of task and environmentaluncertainty arguing that the concepts are measur-ing the same thing. Ross (1995) theorizes e!ectsbetween task uncertainty and performancemeasures but uses measures of environmentaluncertainty. Clarification of links between envir-onmental and technological uncertainty isrequired to isolate potentially di!erent e!ects ofthese variables on MCS design. For example,external uncertainty implies a lack of informationwhich makes it di"cult to plan types of productsand services, levels of output and create con-tingency plans. Also, it makes evaluation di"cultas demand may change in ways beyond the con-trol of managers. This suggests that more flexible,interactive MCS are required to encourage learn-ing and adaptation and evaluate managers on thebasis of more subjective measures or againstadjustable criteria dependent on changing circum-stances. The uncertainty associated with technol-ogy is, in part, derived from the environment withthe technology being responsive to the uncertaintyassociated with markets and product require-ments. Thus, technology may respond to environ-mental uncertainty by becoming more flexible orby employing JIT techniques. The appropriateMCS design is likely to be more flexible andorganic. However, uncertainty, also, is causeddirectly within the technical processes, indepen-dently from environmental conditions. This maybe derived from a search for improvements inproduct design and cost e"ciencies and is likely to


increase concern with managing uncertainty andcomplexity associated with the production pro-cesses. These conditions may prompt the adoptionof planning and evaluation systems such as activ-ity-based accounting, non-financial manufacturingperformance measures and supplier networks.Despite the links between environmental and

task uncertainty, where possible researchersshould draw on work that has tried to resolveissues related to the validity and reliability ofmeasures concerning these contextual variables.An example of this is Brownell and Dunk (1991)who sought to reconcile findings related to the roleof task uncertainty in the study of budgetary rela-ted behaviours. Studies by Brownell and Hirst(1986) and Hirst (1983) used a measure of taskuncertainty that aggregated the separate dimen-sions of task di"culty (analyzability) and varia-bility (number of exceptions). Brownell and Dunk(1991) argued that such a composite measure isinappropriate as it mixes up the potential e!ects ofdi"culty and analyzability. They found that taskdi"culty and not task variability moderated thee!ects of budget behaviours on performance.The area of contemporary manufacturing prac-

tices, such as JIT, TQM, FM and AMT, has pro-vided many opportunities for contingency-basedresearch (Young & Selto, 1991). Ideas from eco-nomics concerning complementarities are likely toprove useful in modelling the way multiple aspectsof manufacturing can be combined optimally(Milgrom & Roberts, 1990).7 Developing anunderstanding of best manufacturing practicesand the way in which manufacturing aligns withor provides the impetus for strategy would seem tobe a necessary step in ensuring that MCS designmaintains relevance to the technical core of orga-nizations. Closer cooperation between MCSresearchers and manufacturing technology expertsand industrial engineers would be fruitful. Theimportance of advances in IT cannot be under-

estimated (Chapman & Chua, 2000). The adop-tion of interactive IT systems, such as SAP R/3,often trigger the adoption of particular perfor-mance and costing systems.As the average life span of products decrease,

consideration of the life cycle of products hasbecome a concern in manufacturing. Short pro-duct life cycles place demands for new productinitiatives and alter cost structures. Also, decreas-ing life cycles increase operating risk and requireincreased capital investment. Understanding howMCS innovations, such as target costing, canassist management within these settings will likelybecome increasingly important.There has been little work that has investigated

how MCS are best suited to di!erent stages in thegrowth of firms. Important topics are the role ofmore formal systems at the stages of new firmformation, early growth, maturity and decline.Questions arise concerning the requirements ofMCS at these di!erent stages, particularly the extentto whichMCS can assist in the transition from earlygrowth to more mature stages. Moores and Yuen(2001) provide an examination of issues concerningdi!erent aspects of MCS that are important for dif-ferent stages of the growth cycle of firms.Finally, it is noteworthy, that most contingency-

based MCS research has involved large, manu-facturing organizations. There have been somestudies in the hospital and hospitality sectors but,on the whole, there has been little research inves-tigating the service and government sectors. Someexamples include studies within government agen-cies (Williams et al., 1990), hospitals (Abernethy &Brownell, 1999), research and development(Shields & Young, 1994) and marketing depart-ments (Foster & Gupta, 1994). The growth inimportance of service industries such as hospitalityand tourism, and the introduction of managerialapproaches to public sector management providemany opportunities for future research.

5.4. Organizational structure

Organizational structure is about the formalspecification of di!erent roles for organizationalmembers, or tasks for groups, to ensure that theactivities of the organization are carried out.

7 It is important to note di!erences between theories basedon contingent compared to complementary relationships. Con-tingent relationships consider the design of controllable vari-ables, e.g. budgets, in response to exogenous variables, e.g. theenvironment. Complementary relationships involve the co-design of multi controllable variables, e.g. aspects of manu-facturing.


Structural arrangements influence the e"ciency ofwork, the motivation of individuals, informationflows and control systems and can help shape thefuture of the organization.There have been various definitions of organi-

zational structure. An important distinction is thedi!erence between the outcomes of structure andthe structural mechanisms. Lawrence and Lorsch(1967) refer to structure, generically, as the way inwhich the organization is di!erentiated and inte-grated. Di!erentiation is concerned with theextent to which sub-unit managers act as quasientrepreneurs, while integration is defined as theextent to which the subunits act in ways that areconsistent with organizational goals. The mechan-isms to achieve di!erentiation involve decentraliz-ing authority, while integration involves rules,operating procedures, committees and the like.Pugh, Hickson, and Turner (1968) and Pugh,Hickson, Hinings, & Turner (1969) empiricallyidentified examples of structural mechanisms thathave been used commonly in contingency-basedresearch, including centralization, standardization,formalization and configuration.Burns and Stalker (1961) discuss structure, gen-

erically, in terms of mechanistic and organicapproaches. The means to achieve these forms ofstructure involve mechanisms such as rules, pro-cedures and openness of communications anddecision processes. Perrow (1970) identified struc-ture in terms of bureaucratic and non-bureau-cratic approaches. Designers of MCS have beenconcerned with formulating MCS to be consistentwith the intent of organizational structure. Con-sequently, it is useful to consider the extent towhich MCS are mechanistic or organic, or towhich they di!erentiate or integrate.8

The choice of structure in organizational con-tingency research has focused on the appropriatestructure to fit the levels of uncertainty in theenvironment (Burns & Stalker, 1961; Drazin &Van de Ven, 1985; Galbriath, 1973; Lawrence &Lorsch, 1967), strategy (Chandler, 1962) and theorganization’s technology (Galbraith, 1973; Per-row, 1970; Thompson, 1967; Woodward, 1965).

Generally, it is believed that more organic struc-tures are suited to uncertain environments. How-ever, it should be noted that Lawrence and Lorsch(1967) identified a need for higher levels of di!er-entiation to cope with diverse and uncertainenvironments and that this causes potential inte-gration problems which require sophisticated liai-son mechanisms (integrative personnel, meetings),rather than rules and procedures. This type ofresponse is something of a hybrid between mechan-istic (for di!erentiation) and organic types ofstructure (for integration) to manage uncertainty.A large body of literature suggests that strate-

gies characterized by diversification require di!er-entiated, divisional structures (Chandler, 1962;Chenhall, 1979; Dyas & Thanheiser, 1976; Shan-non, 1973). Also, it may be argued that once par-ticular structures are in place then decisions will beinfluenced by the opportunities a!orded managersfrom authority granted to them and, perhaps, thepolitical interests of those individuals. Thus,strategy might follow structure (Donaldson, 1987).Often the structural arrangements have importantimplications for information flows that may shapeor bias the future directions of the organization(Bower, 1970).In the prior section extensive links between

technology and types of controls were drawn. It is,also, noteworthy that early studies of organiza-tional design identified important links betweentechnology and structure. Particularly, earlyresearch found that changing to more e"cienttechnologies did not necessarily lead to enhancede!ectiveness accepted. Implementing the techni-cally e"cient technologies required reformulating(Trist & Bamforth, 1951) roles and structures withdynamic negative e!ects on the way individualsrelated to the new technologies, and consequently,a deterioration in performance. It was apparentthat socio-technical approaches were required toensure improved organizational performance.These early observations are important to manyrecent structural innovations such as work-basedteams that attempt to harness developments intechnology with the e"cient blending of appro-priate skills and the motivating force of teamwork.When evaluating contingency relationships

between MCS and structure, elements of environ-

8 Table 1 considers the way in which elements of MCS canbe grouped as mechanistic or organic.


ment, technology and strategy are likely to beimplicated in the relationships and, as such, muchcan be gained by considering them at the same time.

5.4.1. Findings: organizational structure and MCS

Large firms with sophisticated technologies thatare decentralized have been associated with astrong emphasis on formal MCS (Bruns &Waterhouse, 1975); and large, diverse, decen-tralized firms used more administrative controls(importance placed on budgets, sophisticatedbudgets, formal patterns of communications andparticipation in budgets) (Merchant, 1981).Decentralized organizations were identified asperceiving aggregated and integrated informationas useful (Chenhall & Morris, 1986). Managerialperformance was associated with the interactionbetween decentralization and each of the char-acteristics of broad scope, integrated, aggregatedand timely information received/obtained fromMCS (Chia, 1995).There is some evidence linking MAS to functional

di!erentiation. Functional di!erentiation (more resp-onsibility over areas of manufacturing) was linked toformality of budgetary processes (Merchant, 1984).Hayes (1977) found that the importance of evaluatingfactors related to internal operations, external condi-tions and interdependencies depended on thefunctional nature of departments. In productiondepartments, overall e!ectiveness was associated withfactors related to the performance of internal factors.For marketing, performance of factors related to theexternal operating conditions and interdependencieswere most important. Functional di!erentiation hasbeen linked to environmental uncertainty to demon-strate how research and development units, com-pared to marketing, are better suited to participativebudgeting (Brownell, 1985).Mia andChenhall (1994)found that marketing, compared to production,involves higher task uncertainty that explained whymarketing managers used broad scope informationmore e!ectively than those in production. Concern-ing particular functional decisions, Foster and Gupta(1994) identified that improvements in MCS wouldbe valued for pricing decisions, customer mix, salesforce/promotions and product mix. Costing infor-mation was perceived as useful for decisions con-

cerning products and customers. There was adi!erence between potential and actual use of MCSin the area of marketing.Budgetary participation has been studied exten-

sively and associated with a wide variety of con-textual elements (see Shields & Shields, 1988, for areview). Structural contingencies linked to e!ec-tive participative budgeting have included func-tional di!erentiation, specifically research anddevelopment compared to marketing (Brownell,1985); leadership style employing high comparedto low budget emphasis (Brownell, 1982); a con-sideration rather than initiating style of leader-ship (Brownell, 1983); as well as the findings,mentioned above, related to decentralization(Merchant, 1981). As noted, the theories usedto examine functional di!erentiation relied on linksto external environmental uncertainty, rather thanstructure, per se (Brownell, 1985).The ways in which MCS combine with elements

of organizational structure to provide di!erentia-tion and integration within contemporary organi-zational structures provide many opportunities forworthwhile research. Particularly, there are fewstudies that have considered the fit betweenorganic structures and MCS. Organizational the-ory would suggest a need for flexible, open infor-mation systems rather than tight budgetarysystems. Gordon and Narayanan (1984) foundthat organic structures were best served by broadscope and future oriented information. Someresearchers have found that more organic, beha-viourally oriented implementation is required toensure the success of activity-based accounting(Foster & Swenson, 1997; Shields, 1995). Gosselin(1997) found that activity-based costing is adoptedand implemented in organizations with moremechanistic structures. Particularly, mechanisticstructures (vertical di!erentiation or bureaucraticdecision processes) facilitates adoption of activity-based costing (an administrative innovation) andcentralization and formalization were associatedwith implementing activity-based costing. Organicstructures were more suited to activity analysisand activity-cost analysis (technical innovation).Presumably, organizations proceeding from activ-ity analysis to activity-based costing would requireelements of organic and mechanistic structures to


carry them through the stages of activity analysisto activity-based costing.An important element of contemporary struc-

tures is teams. As yet there are few studies thathave considered the role of MCS within teambased structures. Young and Selto (1993) pre-dicted that teamwork and problem solving abil-ities of shop floor employees would be associatedwith high performance related to JIT outcomes.Their study in a single organization did not revealthese associations due, in part, to an inability ofworkers to address process problems and poorimplementation of JIT-compatible managementcontrols. Scott and Tiessen (1999) reported thatteam based structures were associated with hightask complexity and that team performance wasassociated with use of comprehensive performancemeasures (financial and non-financial), formulatedparticipatively and used for compensation. In anexperimental study, Drake, Haka, and Raven-scroft (1999) found that in team structures theinteraction between ABC (c.f. volume basedaccounting) and rewards based on group incen-tives (c.f. assessment of individuals compared toother workers) was associated with cooperativeinnovations, lower costs and higher profits.Propositions concerning organizational structure

and MCS:

Large organizations with sophisticated technol-ogies and high diversity that have more decen-tralized structures are associated with moreformal, traditional MCS (e.g. budgets, formalcommunications).

Research and development departments com-pared to marketing departments, which facehigher levels of task uncertainty, are associatedwith participative budgeting; and marketingcompared to production departments, whichface higher levels of external environmentaluncertainty, are associated with more open,informal MCS.

The structural characteristics of functional dif-ferentiation based on research and develop-ment compared to marketing, leadership stylecharacterized by a consideration compared toinitiating style, and higher levels of decen-

tralization are associated with participativebudgeting.

Decentralization is associated with the MCScharacteristics of aggregation and integration.

Team based structures are associated with par-ticipation and comprehensive performancemeasures used for compensation.

Organic organizational structures are associatedwith perceptions that future orientated MCSare more useful, and with the e!ective imple-mentation of activity analysis and activity-costanalysis.

5.4.2. Critical evaluation

Structural mechanisms have been conceived ofas involving di!erentiation and integration(Lawrence & Lorsch, 1967). Concerning di!er-entiation, conventional thinking in managementaccounting proposes that decentralization shouldbe combined with profit center responsibilityaccounting systems. To achieve integration simplemechanisms, such as operating procedures andformal budgets, have been recommended. It is ofinterest to observe the extent to which these recom-mendations appear somewhat inconsistent with thesuggestions of Lawrence and Lorsch (1967) thathighly di!erentiated organizations should employcomplex liaison mechanisms to achieve integra-tion. Closer inspection of empirical findings sug-gests that comprehensive and formal mechanisticcontrols might be only one aspect of coordinativee!orts in di!erentiated organizations. Khandwalla(1972, 1977) found that large decentralized com-panies employed sophisticated controls but alsoutilized high levels of participation and humanrelations approaches to coordinate activities. Cer-tainly, participation in budgeting has been linkedto decentralized organizations. Merchant (1981)found participation was one aspect of adminis-trative controls. Gul, Tsui, Fong, and Kwok(1995) found an association between decentraliza-tion and participative budgeting. How the partici-pation of individuals in formal budgets might linkto more organic forms of control is an interestingarea for further research. The role of budgets


within organizations that have developed struc-tures based on delayering, developing teams andempowering employees should be investigated.Galbraith (1973, p. 145) alludes to the need tofocus on the process of decision making and con-flict resolution in situations in which there isambiguity and conflict between the various struc-tural units and roles within organizations. (SeeChenhall & Langfield-Smith, 1998a, for a study ofthe role of management accounting in firms devel-oping change programmes focused on teams).Care should be employed in selecting measure-

ment instruments related to structure. Structurehas been measured in terms of decentralization ofauthority (Bruns & Waterhouse, 1975; Chia, 1995;Chenhall & Morris, 1986; Gul et al., 1994; Libby& Waterhouse, 1996; Merchant, 1981), structuringof activities (Bruns & Waterhouse, 1975), inter-dependence (Chenhall & Morris, 1986; Macintosh& Daft, 1987) and organic-mechanistic orienta-tions (Gordon & Narayanan, 1984). Measures ofdecentralization, structuring of activities andinterdependence have relied, in the main, on thosedeveloped by the Aston school (Pugh et al., 1968,1969). The organic-mechanistic nature of structurehas been derived from Khandwalla (1977). TheAston measures have been subjected to consider-able scrutiny and empirical testing for validity andreliability in the organizational literature. The useof more novel measures, such as those related toteam-based structures, will require considerationof work that has developed these measures(Cohen, Ledford, & Spreitzer, 1996).As with other elements of context, in con-

temporary settings, structure remains an impor-tant factor in understanding MCS design. Manyargue that adjustments to structure are required toensure employee commitment to organizationalgoals related to continuous improvement (Kat-zenbach & Smith, 1993). Structural innovations,such as delayering, flat structures, networking,process orientations and team-based work groupsconcern the removal of barriers between organi-zational activities. Such seamless organizationalstructures appear to be inconsistent with tradi-tional profit centres and responsibility accounting,yet many organizations maintain these hier-archical structures. Empowerment and teamwork

have replaced participation as the appropriateconcept for understanding the e!orts of manyorganizations to gain employee involvement.Team-based structures, either as permanent work-based teams or special purpose teams, are wide-spread. Issues of coordination, performance evalua-tion and reward systems in team-based organizationsare important research areas. Much can be learnedfrom linking MCS research agendas with work ofhuman resource management researchers.

5.5. Size

Growth in size has enabled firms to improve e"-ciency, providing opportunities for specializationand the division of labour. Large organizationstend to have more power in controlling theiroperating environment, and when employing large-scale mass production techniques have reducedtask uncertainty. However, as an organizationbecomes larger the need for managers to handlegreater quantities of information increases to apoint where they have to institute controls such asrules, documentation, specialization of roles andfunctions, extended hierarchies and greater decen-tralization down hierarchical structures (Child &Mansfield, 1972). Contemporary large organiza-tions often develop close associations with suppli-ers and customers, which blur the boundariesbetween organizations, thereby increasing furtherthe size of the entity. Size has also provided orga-nizations with the resources to expand into globaloperations, sometimes by way of mergers, take-overs, licensing or other collaborative arrangements.These developments create additional administrativeconcerns due to increased levels of complexitywithin the production processes and with mana-ging interdependencies with global partners.

5.5.1. Findings: size and MCSFew MCS studies have explicitly considered size

as a contextual variable. In the main, studies haveexamined relatively large organizations, usuallyjustifying this as it is large firms that tend to adoptthe type of practices incorporated within moreformal MCS.Studies that have examined size have considered

its e!ects together with other elements of context


such as technology, product diversity and haveexamined an array of controls. Khandwalla (1972,1977) found that large firms were more diversifiedin product lines, employed mass production tech-niques, were more divisionalized and made greateruse of sophisticated controls and environmentalinformation gathering such as forecasting and mar-ket research. The papers by Bruns and Waterhouse(1975) and Merchant (1981), discussed earlier interms of organizational structure, provide evi-dence related to size. Bruns and Waterhouse(1975) identified two forms of control associatedwith size: administrative with large firms and per-sonal with small firms. Administrative controlcomprised more sophisticated technologies, for-malized operating procedures, high levels of spe-cialists and work related rules. Managers perceivedthat employees had high levels of control and hadhigh levels of participation in setting standardsand spent more time in budgeting. They perceivedbudgets as limiting innovation and flexibility instructuring organizations. Interpersonal controlinvolved centralized decision making, individualssaw themselves as having more interaction onbudget related matters, not having their methodsof reaching budgets accepted and being requiredto explain budget variances. Individuals weresatisfied with their superior–subordinate relation-ships. Merchant’s (1981) study also considered sizeas an aspect of a multiple variable approach.Large, diverse firms were more decentralized, usedsophisticated budgets in a participative way andemployed more formal communications.Propositions concerning size and MCS

Large organizations are associated with morediversified operations, formalization of proce-dures and specialization of functions.

Large organizations are associated with moredivisionalized organizational structures.

Large size is associated with an emphasis onand participation in budgets and sophisticatedcontrols.

5.5.2. Critical evaluationMost contingency-based MCS research has

studied larger organizations but has not considered

size variation within larger entities. This is unfor-tunate as there is evidence from early organiza-tional contingency studies that the relationshipbetween size and administrative arrangementssuch as specialization, formalization and the ver-tical span increases with size but at a decliningrate. Thus, while it is reasonable to assume thatlarge firms employ formal MCS, it is possible thatdi!erent types of controls will be appropriatewithin these large firms, depending on size.The role of MCS in smaller or medium sized

entities has received little attention in the con-tingency-based MCS literature (see Reid & Smith,2000), nor has the role of MCS in firms that changesize due to rapid internal growth, take-over or mer-ger been explored. It seems likely that the role offormal and interpersonal controls would di!erdepending on size and rate of change in size. Manyopportunities for MCS, contingency-based researchare likely to be found in the area of small andmedium sized business (see for example articles inthe Journal of Small Business Management and theInternational Small Business Journal).An impact of technological change and struc-

tural reform has been to reduce the number ofemployees, both shop-floor employees and thenumber of middle level managers. In as much asthe number of employees is associated with coor-dination and control issues, reduced size, due tothe substitution of capital for labour, will haveimplications for MCS. For example, the combina-tion of process controls to monitor machines andinformal controls for evaluating people will likelybecome more important where there are feweremployees operating and managing capital inten-sive technologies.Concerning measurement, there are several ways

of estimating size including profits, sales volume,assets, share valuation and employees. The use offinancial measures can make comparisons betweenorganizations di"cult as di!erent accountingtreatments can be found between firms. Mostcontingency-based MCS studies have defined andmeasured size as the number of employees. Num-bers of employees has been found to correlate withnet assets (Pugh et al., 1968, 1969).It is possible that the precise measure of size

could be important depending on the element of


context and dimensions of MCS being studied. Ifthe theory is considering the e!ectiveness of bud-gets to coordinate individuals activities thenemployees is appropriate. If, however, the study isexamining the e!ects of environment on the e!ec-tiveness of customer focused accounting then salesand assets might be more appropriate as thesemeasures capture market power that can lead tobarriers to entry or industry concentration.Khandwalla (1972) argues that forecast sales arethe best indicator or how size may relate to plan-ning, budgeting and structural modification.

5.6. Strategy

Strategy is somewhat di!erent from other con-tingency variables. In a sense it is not an elementof context, rather it is the means whereby man-agers can influence the nature of the externalenvironment, the technologies of the organization,the structural arrangements and the control cul-ture and the MCS. The role of strategy is impor-tant as it addresses the criticism that contingency-based research assumes that an organization’sMCS is determined by context and that managersare captured by their operating situation.Recently, MCS research has recognized that

managers have ‘strategic choice’ whereby they canposition their organizations in particular environ-ments. Thus, if the current product range is toouncertain, reformulating product strategy into amarket that is more predictable may remove thepressure from the environment. It may, also, limitpotential opportunities and therefore require theorganization to examine its attitudes to the trade-o! between potential returns and acceptable riskand uncertainty. Notwithstanding the strategicdirection selected by the organization, con-tingency-based research predicts that certain typesof MCS will be more suited to particular strate-gies. The powerful influence of strategy is evi-denced by the popular use of terms such asstrategies of TQM, the strategic imperative of anempowered workforce and strategic managementaccounting. Langfield-Smith (1997) provides asummary of research into MCS and strategy.Several generic taxonomies have been developed

including entrepreneurial-conservative (Miller &

Friesen, 1982); prospecters-analysers-defenders(Miles & Snow, 1978); build-hold-harvest (Gupta& Govindarajan, 1984); and product di!erentia-tion-cost leadership (Porter, 1980). Evidence fromthe strategy-organizational design research sug-gests that strategies characterized by a con-servative orientation, defenders, harvest and costleadership are best served by centralized controlsystems, specialized and formalized work, simpleco-ordination mechanisms and attention directingto problem areas (Miles & Snow, 1978; Miller &Friesen, 1982; Porter, 1980). Strategies character-ized by an entrepreneurial orientation, pro-spectors, build and product di!erentiation arelinked to lack of standardized procedures, decen-tralized and results oriented evaluation, flexiblestructures and processes, complex co-ordinationof overlapping project teams, and attentiondirecting to curb excess innovation. Simons (1994)argues that four dimensions of MCS link to strat-egy: belief systems to communicate and reinforcebasic values and missions, boundary systems toestablish limits and rules to be respected, diag-nostic controls to monitor outcomes and correctdeviations, and interactive controls to enable topmanagers to personally involve themselves withsubordinates and operations with a view to forcingdialogue and learning.

5.6.1. Findings: strategy and MCSFrom MCS research, evidence suggests links

between strategy and cost control and to formalityof performance evaluation. The studies arefocused on strategy at the strategic business unitlevel, rather than corporate or functional levels.Most of the studies explore the associationbetween MCS and strategic typologies. Con-servatives, defenders and cost leadership strategiesfind cost control and specific operating goals andbudgets more appropriate than entrepreneurs,prospectors and product di!erentiation strategies(Chenhall & Morris, 1995; Dent, 1990; Simons,1987). Simons (1991) found that entities with littlesense of urgency about creating a vision, did notemploy interactive controls. These generalizationsare fairly simplistic. Merchant (1990) found noassociation between di!erent growth strategiesand pressure to meet financial targets. Simons


(1987) demonstrated that tight controls wereapparent in more entrepreneurial strategies, per-haps to balance excessive innovation and to helplearning in uncertain environments. Chenhall andMorris (1995) found that tight control was sui-table for conservative strategies; however, tightcontrol was also found in entrepreneurial situa-tions but, importantly, operating together withorganic decision styles and communications.Again, the apparent paradox can be explained bythe need for organic systems to encourage innova-tion and tight control to curb excessive innovation.Concerning performance measurement, build

compared to harvest strategies, which involvelow specialization and di"culty in measuringoutcomes, suit more subjective and long-termcontrols e.g. (Govindarajan & Gupta, 1985).Product di!erentiation was associated with alow emphasis on meeting budgets (Govindarajan,1988), and when combined with high servicesharing was associated with a reliance on moresubjective behavioural control (Govindarajan &Fisher, 1990). A study by Abernethy and Brownell(1999) found that hospitals undergoing strategicchange (a more prospector type of strategy)used budgets interactively, focusing on dialogue,communication and learning (more organic stylesof control). Van der Stede (2000) showed thatproduct di!erentiation strategies were associatedwith less rigid budgetary control, which in turn,was associated with increased budgetary slack,although there were no direct e!ects betweenstrategy and slack.Evidence on the usefulness of more broad scope

planning information for prospector companiesand for those following build compared to harveststrategies was found by Guilding (1999). In thisstudy, the scope of the information related tocompetitor-focused accounting which incorpo-rated competitor cost assessment, competitiveposition monitoring, competitor appraisal basedon published financial statements, strategic costingand strategic pricing. Bouwens and Abernethy(2000) found that the level of importance tooperational decision making of more integrated,aggregated and timely information was correlatedwith customization strategies. While associationswith broad scope information were not found, the

study focused on importance for ‘operational’decisions, which presumably excluded decisionsconcerning markets and customer requirementswhich are more likely to involve broad scopeinformation.

Propositions concerning strategy and MCS:

Strategies characterized by conservatism,defender orientations and cost leadership aremore associated with formal, traditional MCSfocused on cost control, specific operating goalsand budgets and rigid budget controls, thanentrepreneurial, build and product di!erentia-tion strategies.

Concerning product di!erentiation, competitorfocused strategies are associated with broadscope MCS for planning purposes, and custo-mization strategies are associated with aggre-gated, integrated and timely MCS foroperational decisions.

Entrepreneurial strategies are associate withboth formal, traditional MCS and organicdecision making and communications.

Strategies characterized by defender and har-vest orientations and following cost leadershipare associated with formal performancemeasurement systems including objective bud-get performance targets, compared to moreprospector strategies which require informal,open MCS characterized by more subjectivelong term controls and interactive use of bud-gets focused on informal communications.

5.6.2. Critical evaluationIdeally, the role of strategy is dynamic involving

managers in continually assessing the way combi-nations of environmental conditions, technologiesand structures enhance performance. MCS has thepotential to aid managers in this process byassisting them in formulating strategy related tomarkets and products, required technologies andappropriate structures. MCS can then be impli-cated in the implementation and monitoring ofstrategies, providing feedback for learning andinformation to be used interactively to formulatestrategy. Few studies in MCS have investigated


these issues (see Simons, 1987, 1991, 1994), rathermost have been restricted to identifying MCS thatare appropriate for di!erent strategic archetypes.While there are some common elements in thesedi!erent strategic archetypes, there are significantdi!erences, consequently care is needed in devel-oping theory that is specific to the archetypesemployed in the study. For example, Fisher andGovindarajan (1993) develop theory to examinestrategy and alternative controls based on the dif-ferent needs derived from combinations of strate-gic mission, using concepts of build, hold andharvest, and competitive strategy, using productdi!erentiation and low cost taxonomies.The extent to which these archetypes, which

were developed in the 1970s and 1980s, maintaintheir relevance to contemporary settings is question-able (Kotha & Vadlamani, 1995; Miller & Roth,1994; Shortell & Zajack, 1990). Strategies arebeing complicated by the need for most organiza-tions to be both low cost producers and to providecustomers with high quality, timely and reliabledelivery. More meaningful associations betweenstrategy, environment and internal operations maybecome apparent if specific elements of strategicpriorities are investigated. Relevant research isavailable based on contemporary strategic prio-rities (Miller, DeMeyer, & Nakane, 1992) and hasbeen applied in management accounting research(Chenhall & Langfield-Smith, 1998b).The role of MCS in organizations involved in

strategic change is an important topic. Someinsights can be gained by examining movementacross dimensions of archetypes , such as achange from harvest to prospector orientations(Abernethy & Brownell, 1999), However, greaterunderstanding of how MCS is implicated in stra-tegic change is possible by considering theoriesthat relate to di!erences between incremental,synthetic and discontinuous change (Tushman &Naylor, 1986), the role of strategic intent (Hamel& Prahlad, 1989) and strategic resources (Amit &Schoemaker, 1993), the di!erence between inten-ded and emerging strategies (Mintzberg, 1994),styles of management that encourage change(Kanter, 1982), the impediments to change of anyformal resource allocation process (Quinn, 1985),and the way MCS can be used to manage both

evolutionary and revolutionary change (Simons,1994).There have been concerns with the measurement

of strategy. Measures used to study strategy havebeen criticized as mixing up elements of theenvironment with organizational attributes (thusstudies of strategy and environment would beinvalid). Measures tend not to relate to competitorswhich makes comparisons across industrygroups problematic. Managers have di"cultyrelating to descriptions used to capture generictypologies such as build, harvest and prospect.(see Langfield-Smith, 1997, p. 227 for discussionof strategy measures). Strategy research shouldconsider work that has attempted to validatestrategy measures such as Dess and Davis (1984),Kotha et al. (1995), Miller and Friesen (1986),Miller and Roth (1994) and Shortell and Zajac(1990).

5.7. Culture

The relationship between the design of MCSand national culture represents an extension ofcontingency-based research from its organiza-tional foundations into more sociological con-cerns. The basic proposition is that di!erentcountries possess particular cultural character-istics. This predisposes individuals from withinthese cultures to respond in distinctive ways toMCS. Culture has becomes important in thedesign of MCS, over the past 20 years, as manycompanies have developed multi-national opera-tions. These companies face the issue of whetherto transfer their domestic MCS overseas, or rede-sign their systems to fit the cultural characteristicsof the o!shore entities. Compared to studies ofother contextual variables, research into culturehas been limited and is somewhat exploratory.There is a plethora of meanings of culture.

However, Kaplan (1965) claims there is consensusamong anthropologists that culture is composedof patterned and interrelated traditions, which aretransmitted over time and space by non-biologicalmechanisms based on man’s uniquely developedlinguistic and non-linguistic symbolizing cap-abilities. Culture can be described by inherenttraits such as knowledge, belief, art, morals, law,


custom, and other capabilities and habits acquiredby man as a member of society (Seymour-Smith,1986). However, often culture is conceptualized asa set of characteristics isolated to suit the metho-dological and scientific needs of the research com-munity. The most widely used characteristics weredeveloped by Hofstede (1984) who described thecultural values as power distance (acceptance ofunequal distribution of power), individualism vs.collectivism (placing self-interest ahead of thegroup), uncertainty avoidance (preference to avoiduncertainty and rely on rules and structures),masculinity vs. femininity (achievement, assertive-ness and material success vs. modesty and pref-erence for quality of life) and, subsequently,confucian dynamism (status, respect for tradition,protecting one’s face; Hofstede & Bond, 1988).Virtually all MCS contingency-based studies haveused these values to study the influence of culture.

5.7.1. Findings: culture and MCSContingency-based research in MCS has exam-

ined associations between cultural dimensions andelements of structure such as standardization,decentralization, and control system character-istics such as formality on controls, reliance ofaccounting performance measures and budgetaryparticipation. Overall, the research has providedmixed results as to whether culture does havee!ects across aspects of MCS. There are few areaswhere consensus can be drawn. This is becausestudies have examined di!erent combinations ofcultural dimensions and have considered aspectsof MCS in di!erent ways. As a consequence thereis little overlap between studies to enable themesto be drawn or comparisons made and general-izations developed. The following are examples ofstudies that have examined accounting controls.Harrison (1992) demonstrated that di!erencesbetween Singapore and Australian managers didnot moderate the relationship between budgetemphasis in evaluation and either job related ten-sion or job satisfaction. However, the relationshipbetween reliance on accounting performance mea-sures and low job related tension and high jobsatisfaction was stronger for Singapore managers,the explanation being that these managers had lowindividualism and high power distance compared

to Australian managers (Harrison, 1993). O’Con-ner (1995) argued that the low power distancefound in western parent companies would dom-inate over high power distance found in their localSingapore subsidiaries, thereby enhancing thee!ectiveness of the parent MCS. Using thesearguments he found that the relationship betweenrole ambiguity and superior/subordinate relation-ships (perceptions of competence and trustworthi-ness) and both participation in budgeting and inperformance evaluation were stronger in foreignsubsidiaries than local Singapore entities. Mer-chant, Chow, and Wu (1995) studied Taiwaneseand US firms and found that culture was notimportant in explaining use or e!ectiveness of thedegree of subjectivity in profit center manager’sperformance evaluation. Only use of long-termincentives was more important in Taiwanese firms.Several studies have considered broader aspects

of MCS with less equivocal results. Snodgrass andGrant (1986) found that Japanese, compared toUS, companies experience less explicit controlsand more implicit controls in monitoring, evalua-tion and rewarding. Ueno and Wu (1993) alsofound di!erences between Japanese and US man-agers on MCS characteristics. They theoreticallylinked individualism with US managers andfound, empirically, that they used more formalcommunications, built slack, used controllabilityin budgeting and long-term horizons for perfor-mance evaluation. Uncertainty avoidance waslinked to Japanese firms to explain a preferencefor broad time horizons and structured budgetaryprocesses. These associations were not supportedempirically leading to the conclusion that indivi-dualism is the dominant predictor of MCS. Vance,McClaine, Boje and Stage (1992) studied formalityof controls, team development, appraisal systems,intrinsic/extrinsic rewards and frequency of feed-back in Indonesian, Malaysian, Thai and USfirms. Significant di!erences were found betweenUS and Asian firms but also between the di!erentAsian firms. This study is distinctive as it usedboth Hofstede’s dimensions plus other concepts ofculture drawn from anthropology. Finally, studiesusing experimental methods have failed to supportexpected e!ects and have revealed ambiguousfindings (Chow et al., 1991; Chow et al., 1994)


Given the exploratory nature of research exam-ining culture and the lack of consensus on find-ings, only a general proposition relating culture toMCS is presented.Proposition concerning culture and MCS:

National culture is associated with the design ofMCS.

5.7.2. Critical evaluationThe dominant notion of culture employed in

MCS-culture research has been the Hofestede(1984) values. However, several criticisms can bemade as to how this approach to defining andmeasuring culture has been employed (Harrison &McKinnon, 1999). First, it assumes that the dif-ferent values have the same intensity within a cul-ture. If this is not the case, then some value maybe more dominant than others and have a promi-nent e!ect. Second, some studies do not considerall of Hofstede’s values. It is possible that omittedvalues may have e!ects that are relevant to thestudy. Third, most studies assume that countriesdi!er on values and proceed to test for di!erencesbetween countries without directly assessing cul-tural values. However, countries’ cultures maybechanging due to education and globalization. It is,therefore, important to check that the assumedvalues of a country are still apparent in con-temporary studies. Fourth, while Hofstede’svalues provide a convenient tool for research, itdoes represent a restricted view of culture. Itsexclusive use has prohibited development ofunderstanding how behaviour is influenced by thefundamental traits that influence how individualsthink, feel and respond. More subtle notions ofculture involving myths and ritual, language andnarrative are not considered. It seems likely thattheories and methods drawn from anthropologyand sociology are more suited to understandinghow these subtle factors combine to influence howindividuals respond to MCS.While national culture has been studied exten-

sively, it seems likely that other variables such asmarkets and technologies may interact with cul-tures in systematic ways to e!ect MCS design. Forexample, the adoption of certain types ofadvanced technologies appears to work most

e!ectively if attributes of collectivism are appar-ent. This combination of technology and culturemay suggest that certain types of performancemeasures, suited to the technology but consistentwith collectivism, would be more appropriate. Onevariable that o!ers promise in the study of cultureis organizational culture (Martin, 1992). It is pos-sible that a strong organizational culture maydominate national culture in the work situation.Little work has been completed in the area oforganizational culture and MCS design. As withnational culture, the meaning of organizationalculture and its study are well served by the appli-cation of the research paradigms and methodsfrom sociology and anthropology.

5.8. Continuing relevance of traditional elementsof context

Insights into the present-day context of MCScan be gained by reflecting on the issues drawnfrom traditional contingency-based work. Theenvironment will become more uncertain, hostileand complex as a result of contemporary pres-sures. There will be a need for organizations todevelop increased environmental responsibility.Technologies will be found to have varyingdegrees of complexity, uncertainty and inter-dependencies that promote control issues. Struc-tures will be employed that assist in developingmore organic ways to communicate, but also pro-vide enhanced di!erentiation to motivate andposition individuals close to the business opera-tions. Additionally, structures will be sought thatempower individuals, with the purpose of provid-ing a healthy and fulfilling work environmentwhile better equipping the organization to achievebest practices. The challenges to coordinationderived from size will increasingly become impor-tant as organizations enlarge due to developingclose relationships with suppliers and customersand engaging in global operations by directexpansion, acquisition and merger. Notions ofstrategy are likely to be redefined and it will benecessary for MCS researchers to keep abreast ofstrategy commentators who reflect on the rele-vance of concepts developed by earlier writers.Culture will increase in relevance as firms continue


to develop multi-national operations and willlikely best be researched by conceiving culture inricher terms than the value systems of Hofestede.

6. Issues related to theory development

There are various forms of theoretical fit thathave been used to classify contingency-basedresearch in MCS: selection, interaction and sys-tems (Drazin & Van de Ven, 1985). Selectionstudies examine the way contextual factors arerelated to aspects of MCS with no attempt toassess whether this association is linked to perfor-mance (Chenhall & Morris, 1986; Merchant,1985a). Interaction approaches include studiesthat examine how organizational context moder-ates the relationship between MCS and organiza-tional performance (Brownell, 1983, 1985; Dunk,1993; Govindarajan & Gupta, 1985). Systemsmodels consider the way in which multiple aspectsof controls systems and dimensions of contextcombine in a variety of ways to enhance perfor-mance (Chenhall & Langfield-Smith, 1998b; Selto,Renner, & Young 1995). Luft and Shields (2000)provide a refinement to classify and discuss the-ories employed in contingency-based MCSresearch. This involves considering the structuralrelations between variables, the nature of thecausality between the variables and the levels ofanalysis.9

6.1. Structural relationships between variables

There are several forms of structural relation-ships. Linear additive models examine the asso-ciation of various elements of MCS with outcomes(Hayes, 1977; Merchant, 1981), or of context withMCS (Anderson & Young, 1999; Bruns & Water-house, 1975; Guilding, 1999; Merchant, 1984).These theories provide separate arguments foreach variable acting in isolation with no attentionto the explicit relationship between the explana-

tory variables. The inclusion of variables is justi-fied as they assist in improving the prediction of theoutcome criteria. The dominant form of analysishas been simple correlations or linear regression.Interaction models are used where the nature or

strength of a relationship between MCS and anoutcome criterion will depend on the influence ofparticular aspects of context (Brownell, 1982,1983, 1985; Davila, 2000). Interaction variablemodels have been the dominant forms in con-tingency-based research. For linear interactionmodels, moderated regression analysis or analysisof variance is appropriate. Hartmann and Moers(1999) provide an extensive review of the short-comings of interaction or moderated regressionmodels as applied to budgetary research over thepast 20 years.Intervening models involve the specification of

causal paths between MCS, context and outcomes(Shields et al., 2000; Van der Stede, 2000). Studiesmay identify the antecedents to MCS, or they maydemonstrate how the relationship between MCSand outcomes are explained by intervening vari-ables. It is often important to decompose theassociation between MCS and outcomes intoindirect e!ects operating through the interveningvariable and direct e!ect which captures allremaining e!ects influencing the associationbetween MCS and outcomes. Initially, studiesexamining intervening models used a combinationof linear regression and simple correlations toidentify paths between variables and then usedthese paths to decompose correlations of interestinto direct and indirect e!ects (Chenhall & Brow-nell, 1988; Shields & Young, 1993). More recently,powerful structural equation models, such as EQS,LISTREL, AMOS and PLS, have been employedwhich enable latent variables to be constructedfrom multi-item questionnaires and to identify,simultaneously, statistical significance with multi-ple dependent variables (Anderson & Young,1999; Shields et al., 2000). It is possible to combinemoderating variables within an intervening modelby examining the extent to which a variable mod-erates the e!ects on one or more of the paths.(Scott & Tiessen, 1999). Given the recent criticismdirected towards moderating variable modelsthere is a danger that researchers will try and force

9 See also, Briers and Hirst (1990) and Fisher (1995, 1998)for discussions of theory development within MCS contingencyresearch, Ittner and Larcker (2000) for issues related to MCSresearch in general.


arguments about interaction e!ects into interven-ing variable models.A fourth form of modelling involves systems

approaches which also describe fit but do so bytesting multiple fits simultaneously, involving awider variety of dimensions of context and MCS.Variation in performance stems from variations inoverall systemic fit, with multiple, equally e!ectivealternatives being possible. Techniques to test sys-tems models include the use of Euclidean distance(Selto et al., 1995) and cluster analysis (Chenhall& Langfield-Smith, 1998b). These approaches areless rigorous than regression and require manydecisions in terms of the type of analysis, andgiven the complexity of the relationships betweenvariables, interpretation and theory building canbe di"cult. They do, however, provide a way ofaddressing the criticism that contingency-basedresearch provides only a partial understanding ofMCS and its context. For exploratory research,Ittner and Larcker (2000) note the potential ofrecursive partitioning to split samples into asequence of sub-groups thereby generating a tree-like structure that describes a nesting of indepen-dent variables (Ittner et al., 1999).It is noteworthy that to gain acceptable statis-

tical power in more complex models large samplesizes are required. Thus, the relatively small sam-ples in some contingency-based studies limits thestatistical power of the results. Moreover, it isunusual for contingency-based studies to reportstatistical power.

6.2. Causality

Concerning causality, contingency-based MCSresearch has, in the main, been survey based and thistends to limit the scope of the studies to considersituations involving unidirectional relationships (MCSdetermines outcomes) or bi-directional relationships(MCS determines outcomes which then determinesMCS). Most of the MCS research implicitly assumesunidirectional relationships. If the relationships arebi-directional, then it is possible that they are simul-taneously determined representing a situation inequilibrium, or they are related cyclically whereMCSdetermines outcomes, then outcomes determineMCS, followed by MCS e!ecting outcomes and so

on. Given the existence of cyclical relationships,the predictions from contingency-based theorymay di!er depending on which stage of the cycle isbeing proposed. Moreover, given that most con-tingency-based research has used cross-sectionalsurvey methods, the results are relevant to onlyone stage of the cycle.

6.3. Levels of analysis

The issue of levels of analysis is important totheory construction within contingency-basedresearch. Care is required in maintaining con-sistency between the theory, the unit or level ofanalysis and the source of measurement. Considerexamining the usefulness of budgets to evaluatesub-unit performance. Budget usefulness is con-sidered to depend on environmental uncertaintyand managers’ experience with budgets. The use-fulness of budgets may be considered as a sub-unitvariable and the appropriate concept of environ-ment is one that applies to the particular sub-units, such as uncertainty with sub-unit productsor suppliers. The assumption is that all managerswithin the sub-unit will be expected to respond inthe same way to the environmental uncertainty.Any di!erences at the individual level that maypotentially e!ect budget usefulness are noise.However, if individuals within the sub-unit areexpected to respond di!erently because of di!er-ent experience with budgets the issue arises ofwhat is the appropriate level of analysis? The use-fulness of budgets and environmental uncertaintyare sub-unit variables and experience with budgetsis individual level. If an individual level is adoptedthen the usefulness of budgets at the sub-unit leveland the environmental uncertainty facing the sub-unit are inappropriate as the uncertainty is assumedto be the same for all individuals within the sub-unit. If the theory includes both sub-unit andindividual levels, then the sub-unit level of analysiscan be preserved by splitting the existing sub-unitsinto new sub-units based on di!erent degrees ofthe individual level variable. For example, newsub-units would be created which capture the fourcombinations of high/low uncertainty and high/low experience with budgeting. More generally,the appropriate model for this is an interaction


model. For a comprehensive discussion of theseissues and an evaluation of an extensive list ofMCS studies, see Luft and Shields (2000).

7. Alternate theories and contingency-basedresearch

The term contingency means that something istrue only under specified conditions. As such thereis no ‘contingency theory’, rather a variety of the-ories may be used to explain and predict the con-ditions under which particular MCS will be foundor where they will be associated with enhancedperformance. Contingency-based research has itsfoundations in organizational theory, which con-siders contextual variables only at the organiza-tional level. There is a viewpoint that advances incontingency-based research will be best served bydeveloping and refining theory within its organi-zational core. Certainly, the concepts and ideasfrom organizational theory continue to provide acoherent and rich foundation to examine tradi-tional and new MCS within contemporary set-tings. Much can be gained in understanding theimplications of contemporary elements of envir-onment, technology and structure to the designand implementation of MCS by considering theinsights provided by early organizational theories.For example, Chapman (1997) examined the roleof uncertainty in MCS design by reflecting onGalbraith’s (1973) theories relating uncertainty tothe supply and demand for information. Kalag-nanam and Lindsay (1999) develop theory on theimportance of organic controls for JIT situationsby employing ideas from Woodward (1965).Given the fairly obvious proposition that most

events and the outcomes of those events are likely todepend on the contextual settings, an important issueis whether future contingency-based frameworks canbe advanced by integrating insights from alternatetheoretical perspectives into organizational adapta-tion and functioning. Theories from economicsand psychology, as well as organizational theories,have much to say about the adoption and imple-mentation of MCS. These theories follow a func-tionalist approach that considers the utility ofMCS in achieving purposeful outcomes.

Theories from economics, such as agency the-ory, have in the main, considered the role ofincentive schemes to gain from employees oragents commitment to organizational goals thatare prescribed by principals. Agents are assumedto be self-serving and opportunistic (see Baiman,1982, 1990 for reviews of agency theory related toMCS research). Most studies have employed ana-lytic research techniques. A number of studiesemploying agency theory have used surveymethod to study organizational slack (Dunk,1993), responsibility accounting (Baiman,Larcker, & Rajan, 1995), performance measures(Bushman, Indjejikian, & Smith, 1995) and parti-cipative budgeting (Shields & Young, 1993).Shields (1997) provides a review of various typesof MCS research, including studies that haveemployed agency theory.Agency theories have been criticized for not

considering the context in which principals andagents contract and for not investigating thetrade-o!s with other elements of control systems.(Merchant & Simons, 1986; Shields, 1997). Theseideas may be developed by considering self-servingbehaviour as a moderating variable in the rela-tionship between incentives and performance, withmore organizationally focused attitudes being analternative requiring di!erent forms of incentiveschemes (Davis, 1997a, 1997b). Concerning therole of non-financial considerations, Luft (1997)argues that agency theory relationships may besupported empirically but the inclusion of factorsimportant to agents, such as ethical and fairnessconsiderations, may moderate these findings.Chenhall and Langfield-Smith (2001) argued thatincentive schemes based on agency relationshipsare most e!ective when the implementation of theincentive schemes account for issues related toorganizational justice.Population-ecology theory asserts that fit is

attained by a process of Darwinian natural selec-tion working through births and deaths in thepopulation of organizations (Hannan & Freeman,1989). Organizations are selected for survivalwhich have appropriate adaptive mechanisms andthose that do not fail. The analysis is done at theaggregate population level, without explicitly con-sidering how individual organizations adapt.


While population-ecology has been criticized as itdoes not consider individual organization adapta-tion, it does examine issues concerned with thebirth and death of organizations, areas that areneglected by contingency researchers. Population-ecology and contingency-based research might bedeveloped by examining the preconditions that areassociated with those organizations selected forbirth and those associated with mortality. Forexample, environments rich in opportunities maybe associated with new start-up firms, or certaininteractions between strategies, internal structuresand control systems might be associated with thosepopulations experiencing higher levels of mortality.The area of psychology has relevance to under-

standing MCS and has provided the basis forsome research over the past 20 years. Thisresearch has attempted to identify if individualcharacteristics such as personality or cognitivestyle a!ect the way individuals react and respondto di!erent aspects of MCS. For example, studieshave found that the e!ectiveness of budgetaryparticipation is moderated by an individual’s locusof control (Brownell, 1981), or the levels ofauthoritarianism of superiors and subordinates(Chenhall, 1986). It is possible that personalityfactors may be important moderators in the rela-tionship between conventional organizationalcontextual variables and the usefulness of MCS.For example, Hartmann (2000) argues that therelationship between the acceptance of RAPMand environmental uncertainty may be moderatedby an individual’s tolerance for ambiguity withlow tolerance individuals more readily acceptingRAPM in conditions of uncertainty as it helpsreduce ambiguity. Individual cognitive style hasbeen associated with a proclivity for individuals touse di!erent forms of information, such as oppor-tunity cost (Chenhall & Morris, 1991). It has beenshown that MCS success is likely to depend on theextent to which individuals have organizationalcommitment (Nouri & Parker, 1998), the generationof high levels of trust between employees and man-agers (Ross, 1994), or whether organizational justiceis achieved in implementing MCS (Libby, 1999).Concern with individual attributes can usefully

be combined with organizational context byexamining the compatibility between individuals

and their work situation. This has been referred toas person-environment fit (Deci, 1980) and per-son-organization fit (Kristof, 1996). Theseapproaches assert that environmental or organi-zational factors provide explanations of behaviourbased on observable events but that considerationof individuals can enhance predictions as theybring a unique interpretation to the situation.Often, person-environment fit examines the extentto which individuals demand for financial, physi-cal and psychological resources, as well as taskrelated opportunities, fits with the supply of theseattributes from the organization. Alternatively, fitis seen as the extent to which the individual’sabilities fit the organization’s requirements forcontributions. Shields et al. (2000) draw on per-son-environment fit to argue that stress may bederived from di!erences between performancedemands of a task and the individual’s perfor-mance capabilities. Participation in standard set-ting was shown to decrease stress by increasingindividual’s feeling of control. Fisher (1996) foundthat the usefulness ofMCS could be determined fromconsidering individuals’ locus of control and thelevels of uncertainty in the environment. Contrary toexpectations, individuals with an external locus ofcontrol found broad scope and timely informationmore useful when they perceived the environment asuncertain, compared to those with an internallocus of control. Govindarajan (1988) and Gupta& Govindarajan (1984) provide additional MCSexamples of how individual personality can becombined with organizational level variables.It seems likely that personality, cognitive style

and issues associated with commitment, trust andorganizational justice could help explain the wayindividuals react to information in di!erent con-textual settings, and as such can be included read-ily within contingency-based frameworks. Whencombining di!erent levels of analysis, care isrequired in theory development and method toensure that combinations of individual and orga-nizational variables are theoretically and empiri-cally legitimate.Another area that draws on concerns with the

way managers take decisions is behavioral econom-ics. This approach emphasizes what actually hap-pens, rather than the logical conditions necessary


for things to happen, to generate a strongdescriptive base for economic research. A largebody of research, originally associated with theCarnegie school (Cyert & March, 1963; March &Simon, 1958), but also explicit in the psychologyof economic decision making (Kahneman &Tversky, 1984; Katona, 1951), has suggested thatindividuals have cognitive limitations that influ-ence decision making. Factors such as limitedinformation processing capacity, selective percep-tion, satisficing rather than optimizing and boun-ded rationality all help explain why individualsbehave in ways that may be inconsistent with pre-dictions based on assumptions of rational eco-nomic decision making.Important challenges to understanding the way

managers approach resource allocation decisionsinclude ideas of muddling through by Braybrookeand Lindblom (1970). They argue that rather thanusing formal, analytical, rational-comprehensiveplanning, managers use seat of the pants judgmentto muddle through. Cohen, March, and Olsen’s(1972) garbage can model of behaviour suggeststhat mangers have a repertoire of problemresponses. Managers recognize problems whenthey match situations in which they have devel-oped solutions. A di"culty with these observa-tions for functionalist contingency-based researchis that there is little that is prescriptive in terms ofdesigning MCS. However, these types of issues areimportant to understand, as they may provide thediagnostics for why the design of MCS, whichappears to fit context, still do not generate e!ec-tive organizational performance.The work of Williamson (1985, 1986) focused

on information problems and how managers takedecisions. A major contribution of relevance toorganizational control concerns identifying whenthe performance of the firm is influenced by itsorganizational structure. Williamson examines theissue of when transactions are better completedwithin firms and when they are best executed bymarkets. Issues of divisional structures, profitcenters and transfer pricing have been informed bythese theories (Colbert & Spicer, 1995; Spicer,1988; Spicer & Ballow, 1983; Swieringa & Water-house, 1982). Importantly, Williamsons’s workrecognizes that there is no obvious single optimal

method for internal organization. At any one timethe appropriate structures and controls willdepend on product portfolios or the extent of ver-tical integration. Gilad, Kaish, and Loeb (1988)provide a brief overview of the development andcontribution of behavioural economics.A criticism of contingency-based research is that

it has relied on traditional, functionalist theoriesand has not applied more interpretive and criticalviews. Alternative approaches, derived fromsociology literature, have been used in MCSresearch to provide this interpretive and criticalfocus. In the main, these approaches have rejectedthe assumptions upon which functionalist con-tingency research is based. A strength of ‘alter-native’ approaches is that they show the potentialconflict between individuals and groups and howMCS may be implicated in these struggles. Forexample MCS are not assumed to lead, necessa-rily, to enhanced e!ectiveness, rather they are usedfor political and power purposes by groups withinthe organization or within society at large, and arenot associated with the welfare of the organiza-tion. These themes are attractive to researchapproaches that are radical or socially critical.Baxter and Chua (2000) provide a review of thevarious streams of sociology that have been usedin management accounting research.10

The generation of propositions concerning novelrelationships, processes and their contextual set-ting are often best identified and elaborated byusing case study methods. Case studies are verypowerful for identifying research problems and indeveloping and generalizing theory (Covaleski etal., 1996). [See Baxter & Chua (1998) for a sum-

10 MCS are not seen as passive mechanisms to be used bymanagers to assist in optimizing resource allocation. Rather,they may be used to legitimate particular power relationshipswithin organizations or enable groups within society to main-tain their command over resources or political direction. MCSmay be motivated by mimicry and compliance rather than aneed for enhanced e"ciency. Managers may espouse intent fore"ciency but respond to MCS in ways constrained by boundedrationality, limited information processing capacity, selectiveperception and satisficing rather than optimizing behaviour.MCS may be instrumental in limiting progress because it inhi-bits innovative thought or it may have a role in assisting in theadoption of change by providing the basis to control the newinitiatives.


mary and synthesis of this form of research,Atkinson & Sha"r, (1998) for a discussion of thecase study method of research in MCS, andYoung & Selto, (1993) for di"culties in caseresearch]. However, restricting research to caseslimits the possibilities for causal inference andgeneralizability to broader populations. It is di"-cult to make progress in understanding the moresubtle insights derived from alternate approacheswithout attempting to identify general patterns ofcausation (Donaldson, 1985). Much can be gainedby combining case evidence with surveys withincontingency-based frameworks. For example,Young and Selto (1993), Shields and Young(1993), Kalagnanam and Murray Lindsay (1999)and Davila (2000) present site visits or case studiesas part of problem identification and theory con-struction to propose relationships between MCSand contextual variables that are then tested byuse of survey method.An important issue is whether ‘alternate’ theories

of MCS research can be combined with traditional,functionalist models. While these paradigms havedi!erent theoretical and philosophical bases, someresearchers have used contingency-based ideas todevelop convergence between these approaches.Many of the insights concerning the role of insti-tutions within society on the adoption of MCS canbe combined readily with contingency concepts(Gieger & Ittner, 1996; Scott, 1987). Also, the wayin which power is implicated in the adoption and useof MCS to e!ect resource distribution or inducechange can be examined within contingency-basedapproaches (Bari! & Galbraith, 1978; Hage,1980). Moreover, understanding of the influenceof power and politics may be illuminated by con-sidering theories related to environmental, techni-cal and structural context. A contingency-basedapproach attempts to map variables and demon-strate potential relationships between these vari-ables, which may include power and politics, andindicate potential links with outcomes.Caution must be directed at any approach pro-

viding some unification between functionalist and‘alternate’ approaches. Literature examining MCSfrom various organizational, economic and psy-chological perspectives assume that the study ofMCS is conducted within situations that can be

well specified and understood. The search is forgeneralizable findings, unique situations are seenas anomalies and are important only as they helpunderstand how to move towards well structured,ordered solutions. Sociological approaches use avariety of theories to understand organizationalsettings that are often so ill structured that reg-ularities cannot be meaningfully represented.Some commentators claim that di!erent theorieso!er fundamentally di!erent insights into the nat-ure of MCS and should not be blended but keptseparate providing alternative ways of under-standing the multiple roles of MCS in organiza-tions. Any attempt at amalgamation is unlikely toattain a true synthesis as one theory inevitablesubsumes others (Covaleski, Dirsmith, & Samuel,1996; Dirsmith, Covaleski, & McAllister, 1985).However, a proliferation of theoretical alter-natives, without an integrative framework, can beconfusing to both managers and students andmuch is lost in fragmentation across many uncon-nected streams of research. Some contingency-based researchers see a challenge in providing anintegrating framework, that combines structure andprocess, to assist managers, students and researchersfind a path through the many diverse paradigmsused to study MCS (see Donaldson, 1995, for anattempt to integrate a variety of theories usingstructural contingency frameworks as the unifyingtheme). Also, attempts to assimilate ideas fromalternative theories could generate constructivedebate on competing organizational ends, the roleof di!erent groups within organizations and sta-keholders, and a variety of values and purposesassociated with MCS including the implications ofalternatives to traditional rational economicvalues, and the role of di!erent elements of orga-nizational context (Jonsson & Macintosh, 1997).

8. Conclusion

Contingency-based research has approached thestudy of MCS assuming that managers act with anintent to adapt their organizations to changes incontingencies in order to attain fit and enhancedperformance. There is a considerable body of lit-erature, which while not without imperfections in


method, has provided a basis for generalized pro-positions between elements of MCS and context.The basic framework and potential strength of themethod provide a basis to persist with con-tingency-based research to uncover generalizablefindings that can enhance desired organizationaloutcomes. To maintain the relevance of MCScontingency-based research, scholars will need tofocus their attention on contemporary dimensionsof MCS, context and organizational and socialoutcomes. Notwithstanding the need to studyissues of contemporary relevance much can begained by reflecting on the work of original orga-nizational theorists and more recent thinking inareas such as strategy, organizational and culturalchange, manufacturing, information technologyand human resource management. Other approa-ches based in economics and psychology canreadily be included within contingency-based fra-meworks. While founded in non-functionalistapproaches to studying MCS, insights drawn from‘alternate’ theories, also, can assist in elaboratingthe traditional contingency-based model. More-over, contingency-based research can provide anordered way to integrate thinking about thesociological processes e!ecting MCS in action,perhaps combining these insights with conven-tional elements of contingency-based models.Such a research agenda involves many issues con-cerning theory development and model construc-tion that provide challenges for researchers.

Acknowledgements

My particular thanks to Mike Shields, ChrisChapman, John Waterhouse and anonymousreviewers for their advice and suggestions. Also, Igratefully acknowledge valuable commentsreceived from participants at the AOS 25th Anni-versary Conference.

References

Abernethy, M. A., & Brownell, P. (1997). Management controlsystems in research and development organizations: the roleof accounting, behavior and personnel controls. Accounting,Organizations and Society, 22(3/4), 233–248.

Abernethy, M. A., & Brownell, P. (1999). The role of budgetsin organizations facing strategic change: an exploratorystudy. Accounting, Organizations and Society, 24(3), 189–204.

Abernethy, M. A., & Guthrie, C. H. (1994). An empiricalassessment of the fit between strategy and management infor-mation systems design. Accounting and Finance, 33, 49–66.

Abernethy, M. A., & Lillis, A. (1995). The impact of manu-facturing flexibility on management control system design.Accounting Organizations and Society, 20(4), 241–258.

Amit, R., & Schoemaker, P. J. H. (1993). Strategic assets andorganizational rent. Strategic Management Journal, 14, 33–46.

Anderson, S.W., &Young,M. S. (1999). The impact of contextualand process factors on the evaluation of activity-based costingsystems. Accounting Organizations and Society, 24(7), 525–559.

Atkinson, A., & Sha"r, W. (1998). Standards for field researchin management accounting. Journal of ManagementAccounting Research, 10, 40–68.

Atkinson, A. A., Balakrishnan, R, Booth, P, Cote, J. M.,Groot, T, Malmi, T., Roberts, H., Uliana, E., & Wu, A.(1997). New directions in management accounting research.Journal of Management Accounting Research, 9, 79–108.

Baiman, S. (1982). Agency research in management accounting:a survey. Journal of Accounting Research, 2, 154–213.

Baiman, S. (1990). Agency research in managerial accounting: asecond look. Accounting Organizations and Society, 15(4), 1–39.

Baiman, S., Larcker, D. F., & Rajan, M. V. (1995). Organiza-tional design for business units. Journal of AccountingResearch, 33(2), 205–229.

Banker, R. D, Potter, G., & Schroeder, R. G. (1993). Reportingmanufacturing performance measures to workers: anempirical investigation. Journal of Management AccountingResearch, 3, 34–55.

Bari!, M. L., & Galbriath, J. R. (1978). Intraorganizationalpower considerations for designing information systems.Accounting, Organizations and Society, 3(1), 15–27.

Baxter, J. A., & Chua, W. F. (1998). Doing field research:practice and meta-theory in counterpoint. Journal of Man-agement Accounting Research, 10, 69–87.

Baxter, J. A., & Chua, W. F. (2000). The other managementaccounting-whence and whither, 25th Anniversary Conference,Accounting, Organizations and Society. Oxford: Pergamon.

Biddle, G. C., Bowen, R. M., & Wallace, J. S. (1998). DoesEVA beat earnings? Evidence on the association with stockreturns and financial values. Journal of Accounting and Eco-nomics, 24, 301–336.

Birnberg, J. G., Turopolec, L., & Young, S. M. (1983). Theorganizational context of accounting. Accounting Organiza-tions and Society, 111–129.

Bloom, M. (1998). Relationship among risk, incentive pay andorganizational performance. Academy of Management Jour-nal, 41(3), 283–297.

Braybrooke, D., & Lindblom, C. (1970). A strategy of decision.New York: Free Press.

Bommer, W., Johnson, J., Rich, G., Podsako!, P., & Mack-enzie, S. (1995). On the interchangeability of objective andsubjective measures of employee performance: a meta-analy-sis. Personnel Psychology, 48, 587–605.


Bouwens, J., & Abernethy, M. A. (2000). The consequences ofcustomization on management accounting system design.Accounting, Organizations and Society, 25(3), 221–259.

Bower, J. (1970). Managing the resource allocation process.Cambridge: Harvard University Press.

Briers, M., & Hirst, M. (1990). The role of budgetary informa-tion in performance evaluation. Accounting, Organizationsand Society, 15(4), 373–398.

Brownell, P. (1981). Participation in budgeting, locus of controland organizational performance. The Accounting Review,October, 844–860.

Brownell, P. (1982). The role of accounting data in performanceevaluation, budgetary participation, and organizationale!ectiveness. Journal of Accounting Research, 20, 12–27.

Brownell, P. (1983). Leadership style, budgetary participationand managerial behaviour. Accounting, Organizations andSociety, 8, 307–321.

Brownell, P. (1985). Budgetary systems and the control offunctionally di!erentiated organizational activities. Journalof Accounting Research, 23, 502–512.

Brownell, P. (1987). The role of accounting information,environment and management control in multi-nationalorganisations. Accounting and Finance, 27, 1–16.

Brownell, P., & Dunk, A. S. (1991). Task uncertainty and itsinteraction with budgetary participation and budget empha-sis; some methodological issues and empirical investigation.Accounting, Organizations and Society, 8, 693–703.

Brownell, P., & Hirst, M. K. (1986). Reliance on accountinginformation, budget participation, and task uncertainty: testof a three way interaction. Journal of Accounting Research.Autumn, 241–249.

Brownell, P., & McInnes, P. (1986). Budgetary participation,motivation and management performance. The AccountingReview, October, 587–600.

Brownell, P., & Merchant, K. (1990). The budgetary and per-formance influences of product standardisation and manu-facturing process automation. Journal of AccountingResearch, 28(2), 388–397.

Bruns Jr., W. J., & Waterhouse, J. H. (1975). Budgetary controland organizational structure. Journal of AccountingResearch, Autumn, 177–203.

Burns, T., & Stalker, G. (1961). The management of innovation.London: Tavistock.

Bushman, R. M., Indjejikian, R. J., & Smith, A. (1995).Aggregate performance measures in business unit managercompensation: the role of intrafirm interdependencies. Jour-nal of Accounting Research, 33(Supplement), 101–129.

Chandler, A. D. (1962). Strategy and structure—chapters in thehistory of American industrial enterprise. Cambridge: MIT Press.

Chapman, C. S. (1997). Reflections on a contingent view ofaccounting. Accounting, Organizations and Society, 22, 189–205.

Chapman, C. S. (1998). Accountants in organizational networks.Accounting, Organizations and Society, 23(8), 737–766.

Chapman, C. S., & Chua, W. F. (2000). Technology, organiza-tional form and accounting, 25th anniversary conference:Accounting, Organizations and Society. Oxford: Pergamon.

Chenhall, R. H. (1979). Some elements of organizational con-trol in Australian divisionalised firms. Australian Journal ofManagement, Supplement, 4, 1–36.

Chenhall, R. H. (1986). Authoritarianism and participativebudgeting: a dyadic analysis. The Accounting Review, 2, 263–272.

Chenhall, R. H. (1997). Reliance on manufacturing perfor-mance measures, total quality management and organiza-tional performance. Management Accounting Research, 8,187–206.

Chenhall, R. H., & Brownell, P. (1988). The e!ect of partici-pation on job satisfaction and performance: role ambiguityas an intervening variable. Accounting, Organizations andSociety, 13(3), 225–233.

Chenhall, R. H., & Morris, D. (1986). The impact of structure,environment and interdependencies on the perceived useful-ness of management accounting systems. Accounting Review,61, 16–35.

Chenhall, R. H., & Morris, D. (1991). The e!ect of cognitivestyle and sponsorship bias on the treatment of opportunitycosts in resource allocation decisions. Accounting, Organiza-tions and Society, 16(1), 27–46.

Chenhall, R. H., & Morris, D. (1993). The role of post com-pletion audits, managerial learning, environmental uncer-tainty and performance. Behavioural Research in Accounting,5, 170–186.

Chenhall, R. H., & Morris, D. (1995). Organic decision andcommunication processes and management accounting sys-tems in entrepreneurial and conservative business organiza-tions. Omega, International Journal of Management Science,23(5), 485–497.

Chenhall, R. H., & Langfield-Smith, K. (1998a). Factors influ-encing the role of management accounting in the develop-ment of performance measures within organisational changeprograms. Management Accounting Research, 9, 361–386.

Chenhall, R. H., & Langfield-Smith, K. (1998b). The relation-ship between strategic priorities, management techniques andmanagement accounting: an empirical investigation using asystems approach. Accounting, Organizations and Society,23(3), 243–264.

Chenhall, R. H. and Langfield-Smith, K. (2001). The role ofemployee incentive pay in sustaining strategic change,Working paper.

Chia, Y. (1995). Decentralization, management accounting(MCS) information characteristics and their interactione!ects on managerial performance: a Singapore study. Jour-nal of Business Finance and Accounting, 22(6), 811–830.

Child, J., & Mansfield, R. (1972). Technology, size and orga-nizational structure. Sociology, 6, 369–393.

Chong, V., & Chong, K. (1997). Strategic choices, environ-mental uncertainty and SBU performance: a note on theintervening role of management accounting systems.Accounting and Business Research, 27(4), 268–276.

Chow, C. W., Shields, M. D., & Chan, Y. K. (1991). The e!ectsof management controls and national culture on manu-facturing performance: an experimental investigation.Accounting, Organizations and Society, 16, 209–226.


Chow, C. W., Kato, Y., & Shields, M. D. (1994). Nationalculture and the preference for management controls: anexploratory study of the firm-labor market interface.Accounting, Organizations and Society, 19, 381–400.

Cohen, M. D., March, J. C., & Olson, J. P. (1972). A garbagecan model of organizational choice. Administrative ScienceQuarterly, 17, 1–25.

Cohen, S. G. (1993). New approaches to teams and teamwork.In J. Galbraith, & E. E. Lawler III, et al. (Eds.), Organizingfor the future—the new logic for managing complex organiza-tions (pp. 194–226). San Francisco: Jossey-Bass.

Cohen, S. G., Ledford Jr., G. E., & Spreitzer, G. M. (1996). Apredictive model of self-managing work team e!ectiveness.Human Relations, 49(5), 643–665.

Colbert, G. J., & Spicer, B. H. (1995). A multi-case investiga-tion of a theory of the transfer pricing problem. Accounting,Organizations and Society, 20(6), 423–456.

Covaleski, M. A., Dirsmith, M. W., & Samuel, S. (1996).Managerial accounting research: the contributions of orga-nizational and sociological theories. Journal of ManagementAccounting Research, 8, 1–35.

Cyert, R. M., & March, J. G. (1963). A behavioural theory ofthe firm. Englewood Cli!s, NJ: Prentice-Hall.

Daft, R. L., & Macintosh, N. J. (1981). A tentative explorationinto the amount and equivocality of information processingin organisational work units. Administrative Science Quar-terly, 207–244.

Davila, T. (2000). An empirical study on the drivers ofmanagement control systems’ design in new productdevelopment. Accounting, Organizations and Society, 25,383–409.

Davis, J. H. (1997a). Davis, Schoorman, and Donaldson reply:the distinctiveness of agency theory and stewardship theory.The Academy of Management Review, 22(3), 611–613.

Davis, J. H. (1997b). Towards a stewardship theory of man-agement. The Academy of Management Review, 22(1), 20–48.

Deci, E. D. (1980). The psychology of self-determination. Lex-ington, US: Lexington Books, D.C. Heath and Company.

Dent, J. F. (1990). Strategy, organization and control: somepossibilities for accounting research. Accounting, Organiza-tions and Society, 15, 3–25.

Dess, G. D., & Davis, P. S. (1984). Porter’s (1980) genericstrategies as determinants of strategic group membership andorganizational performance. Academy of Management Stud-ies, 27(3), 467–488.

Dirsmith, M. M., Covaleski, M., & McAllister, J. (1985). Ofparadigms and metaphors in auditing thought. Con-temporary Accounting Research, 2, 46–68.

Donaldson, L. (1985). In defence of organizational theory: areply to the critics. Cambridge, UK: Cambridge UniversityPress.

Donaldson, L. (1987). Strategy and structural adjustment toregain fit and performance: in defence of contingency theory.The Journal of Management Studies, 24(1), 1–24.

Donaldson, L. (1995). American anti-management theories oforganization: critique of paradigm proliferation. Cambridge,UK: Cambridge University Press.

Drake, A. R., Haka, S. F., & Ravenscroft, S. P. (1999). Costsystem and incentive structure e!ects on innovation, e"-ciency and profitability in teams. The Accounting Review,74(3), 323–345.

Drazin, R., & Van de Ven, A. H. (1985). Alternative forms offit in contingency theory. Administrative Science Quarterly,30, 514–539.

Duncan, R. (1972). Characteristics of organizational environ-ments and perceived environmental uncertainty. Adminis-trative Science Quarterly, September, 313–327.

Dunk, A. S. (1992). Reliance on budgetary control, manu-facturing process automation and production sub-unit per-formance: a research note. Accounting, Organizations andSociety, 17(3/4), 185–239.

Dunk, A. (1993). The e!ects of budget emphasis and informa-tion asymmetry on the relation between budgetary partici-pation and slack. The Accounting Review, 2, 400–410.

Dunk, A., & Nouri, H. (1998). Antecedents of budgetary slack:a literature review and synthesis. Journal of Accounting Lit-erature, 17, 72–96.

Dyas, G. P., & Thanheiser, H. T. (1976). The emerging Eur-opean enterprise—strategy and structure in French and Ger-man industry. London: Macmillian.

Epstein, M. J., & Birchard, B. (2000). Counting what counts:turning corporate accountability to competitive advantage.New York: Perseus Books.

Emsley, D. (2000). Variance analysis and performance: twoempirical studies. Accounting, Organizations and Society, 1,33–47.

Ewusi-Mensah, K. (1981). The external organizational envir-onment and its impact on managerial information systems.Accounting, Organizations and Society, 6(4), 310–316.

Ezzamel, M. (1990). The impact of environmental uncertainty,managerial autonomy and size on budget characteristics.Management Accounting Research, 1, 181–197.

Fisher, C. (1996). The impact of perceived environmentaluncertainty and individual di!erences on management infor-mation requirements. Accounting, Organizations and Society,361–369.

Fisher, J. G. (1995). Contingency-based research on manage-ment control systems: categorization by level of complexity.Journal of Accounting Literature, 14, 24–53.

Fisher, J. G. (1998). Contingency theory, management controlsystems and firm outcomes: past results and future directions.Behavioural Research in Accounting, 10(Supplement), 47–64.

Fisher, J. G., & Govindarajan, V. (1993). Incentive compensationdesign, strategic business unit mission, and competitive strat-egy. Journal of Management Accounting Research, 5, 129–144.

Foster, G., & Gupta, M. (1994). Marketing, cost managementand management accounting. Journal of ManagementAccounting Research, 6, 43–77.

Foster, G., & Horngren, C. (1988). Flexible manufacturingsystems: cost management and cost accounting implications.Journal of Cost Management, Fall, 16–24.

Foster, G., & Swenson, D. W. (1997). Measuring the success ofactivity-based cost management and its determinants. Jour-nal of Management Accounting Research, 9, 107–139.


Furnham, A., & Stringfield, P. (1994). Congruence of self andsubordinate ratings of managerial practices as a correlate ofsuperior evaluation. Journal of Occupational and Organiza-tional Psychology, 67, 57–67.

Galbraith, J. (1973). Designing complex organizations. USA:Addison Wesley Publishing Company.

Gieger, D., & Ittner, C. (1996). The influence of funding source andlegislative requirements on government cost accounting prac-tices. Accounting, Organizations and Society, 21(6), 549–567.

Gilad, B., Kaish, S., & Loeb, P. D. (1988). From economicbehavior to behavioral economics: the behavioral uprising ineconomics. In P. Earl (Ed.), Behavioural economics, Vol. II(pp. 437–458). Edward Elgar Publishing Limited.

Gordon, L. A., & Narayanan, V. K. (1984). Managementaccounting systems, perceived environmental uncertaintyand organization structure: an empirical investigation.Accounting, Organizations and Society, 1, 33–47.

Gordon, L. A., & Smith, K. J. (1992). Postauditing capitalexpenditures and firm performance: the role of asymmetricinformation. Accounting, Organizations and Society, 17(8),741–757.

Gordon, L. A., & Silvester, K. J. (1999). Stock market reac-tions to activity-based costing adoption. Journal of Account-ing and Public Policy, 18(3), 229–251.

Gosselin, M. (1997). The e!ects of strategy and organizationalstructure on the adoption and implementation of activity-based costing. Accounting, Organizations and Society, 22(2),105–122.

Govindarajan, V. (1984). Appropriateness of accounting datain performance evaluation: an empirical examination ofenvironmental uncertainty as an intervening variable.Accounting, Organizations and Society, 9, 125–135.

Govindarajan, V. (1986). Impact of participation in the bud-getary process on managerial attitudes and performance:universalistic and contingency perspectives. Decision Sci-ences, 17, 496–516.

Govindarajan, V. (1988). A contingency approach to strategyimplementation at the business-unit level: integratingadministrative mechanisms with strategy. Academy of Man-agement Journal, 31, 828–853.

Govindarajan, V., & Fisher, J. (1990). Strategy, control systemsand resource sharing: e!ects on business-unit performance.Academy of Management Journal, 33, 259–285.

Govindarajan, V., & Gupta, A. K. (1985). Linking controlsystems to business unit strategy: impact on performance.Accounting, Organizations and Society, 51–66.

Granlund, M., & Lukka, K. (1998). It’s a small world of man-agement accounting practices. Journal of ManagementAccounting Research, 10, 153–179.

Guilding, C. (1999). Competitor-focused accounting: anexploratory note. Accounting, Organizations and Society,24(7), 583–595.

Gul, F., & Chia, Y. (1994). The e!ects of managementaccounting systems, perceived environmental uncertaintyand decentalization on managerial performance: a test of athree-way interaction. Accounting, Organizations and Society,19(4/5), 413–426.

Gul, F., Tsui, J. S. L., Fong, S. C. C., & Kwok, H. Y. L. (1995).Decentralization as a moderating factor in the budgetaryparticipation-performance relationship: some Hong Kongevidence. Accounting and Business Research, 25(98), 107–113.

Gupta, A. K., & Govindarajan, V. (1984). Business unit strat-egy, managerial characteristics, and business unit e!ective-ness at strategy implementation. Academy of ManagementJournal, 25–41.

Hage, J. (1980). Theories of organization: form, process andtransformation. Lexington, MA: Lexington Books.

Haka, S. (1987). Capital budgeting techniques and firm specificcontingencies: a correlational analysis. Accounting, Organi-zations, and Society, 12(1), 31–48.

Hamel, G., & Prahlad, C. K. (1989). Strategic intent. HarvardBusiness Review, May–June, 63–76.

Hannan, M. T., & Freeman, J. (1989). Organizational ecology.Cambridge: Harvard University Press.

Harrison, G. L. (1992). The cross-cultural generalizability of therelation between participation, budget emphasis and job-rela-ted attitudes. Accounting, Organizations, and Society, 17, 1–15.

Harrison, G. L. (1993). Reliance on accounting performancemeasures in superior evaluation style—the influence ofnational culture and personality. Accounting, Organizations,and Society, 18, 319–339.

Harrison, G. L., & McKinnon, J. (1999). Cross culturalresearch in management accounting systems design: a reviewof the current state. Accounting, Organizations, and Society,24(5/6), 483–506.

Hartmann, F., & Moers, F. (199). Testing contingencyhypotheses in budgetary research: an evaluation of the mod-erated regression analysis. Accounting, Organizations andSociety, 24(4), 291–315.

Hartmann, F. (2000). The appropriateness of RAPM: towardsthe further development of theory. Accounting, Organizationsand Society, 25(4-5), 451–482.

Hayes, D. (1977). The contingency theory of managerialaccounting. The Accounting Review, 1, 23–39.

Hayes, R., Wheelwright, S. C., & Clark, K. B. (1988). Restoringour competitive edge: competing through manufacturing. NewYork: Free Press.

Heneman, H. (1974). Comparison of self-report and superior-rating of managerial performance. Journal of Applied Psy-chology, 59, 638–642.

Hirst, M. K. (1981). Accounting information and the evalua-tion of subordinate performance: a situational approach. TheAccounting Review, 56, 771–784.

Hirst, M. K. (1983). Reliance on accounting performancemeasures, task uncertainty and dysfunctional behaviour.Journal of Accounting Research, 21(2), 596–605.

Hofstede, G. H. (1984). The cultural relativity of the quality oflife concept. Academy of Management Review, 27, 389–398.

Hofstede, G. H., & Bond, M. H. (1988). The Confucius con-nection: from cultural roots to economic growth. Organiza-tional Dynamics, 16, 5–21.

Hopwood, A. (1972). An empirical study of the role ofaccounting data in performance evaluation. Journal ofAccounting Research, Supplement, 156–182.


Hopwood, A. (1974). Leadership climate and the use ofaccounting data in performance evaluation. The AccountingReview, July, 485–495.

Imoisili, O. A. (1985). Task complexity, budget style of evalu-ating performance and managerial stress: an empirical inves-tigation. Unpublished dissertation, Graduate School ofBusiness, University of Pittsburgh.

Imoisoli, O. A. (1989). The role of budget data in the evalua-tion of managerial performance. Accounting, Organizations,and Society, 14, 325–335.

Ittner, C. D., & Larcker, D. F. (1995). Total quality manage-ment and the choice of information and reward systems.Journal of Accounting Research, 33(Supplement), 1–34.

Ittner, C. D., & Larcker, D. F. (1997). Quality strategy, strate-gic control systems, and organizational performance.Accounting, Organizations, and Society, 22(3/4), 295–314.

Ittner, C. D., & Larcker, D. L. (1998). Innovations in perfor-mance measurement, trends and research implications. Jour-nal of Management Accounting Research, 10, 205–238.

Ittner, C. D., & Larcker, D. L. (2000). Assessing empiricalresearch in management accounting: a value-based manage-ment perspective, forthcoming in. Journal of Accounting andEconomics.

Ittner, C. D., Larcker, D. F., Nagar, V., & Rajan, M. V. (1999).Supplier selection, monitoring practices and firm performance.Journal of Accounting and Public Policy, 18(3), 253–281.

Jonsson, S., & Macintosh, N. B. (1997). Cats, rats, and ears:making the case for ethnographic accounting research.Accounting, Organizations, and Society, 22(3/4), 367–386.

Kahneman, D., & Tversky, A. (1984). Choices, values andframes. American Psychologist, 39(4), 1–10.

Kalagnanam, S. S., & Lindsay, R. M. (1999). The use oforganic models of control in JIT firms: generalizing Wood-ward’s findings to modern manufacturing practices.Accounting, Organizations and Society, 24(1), 1–30.

Kanter, R. M. (1982). The middle manager as innovator. Har-vard Business Review, July-August, 95–105.

Katona, G. (1951). Psychological analysis of economic beha-viour. New York: McGraw Hill.

Kaplan, D. (1965). The superorganic: science or metaphysics.American Anthropologist, 67, 958–976.

Katzenbach, J. R., & Smith, D. K. (1993). The wisdom ofteams: creating the high performance organization. Cam-bridge: Harvard Business School Press.

Khandwalla, P. (1972). The e!ects of di!erent types of compe-tition on the use of management controls. Journal ofAccounting Research, Autumn, 275–285.

Khandwalla, P. (1977). Design of organizations. New York:Harcourt Brace Jovanovich.

Kotha, S., & Vadlamani, B. L. (1995). Assessing generic stra-tegies: an empirical investigation of two competing typolo-gies in discrete manufacturing industries. StrategicManagement Journal, 16, 75–83.

Kren, L., & Liao, W. M. (1988). The role of accounting infor-mation in the control of organizations: a review of the evi-dence. Journal of Accounting Literature, 7, 280–309.

Kristof, A. L. (1996). Person-organization fit: an interactive

review of its conceptualizations, measurement, and implica-tions. Personnel Psychology, 49, 1–49.

Krumwiede, K. R. (1998). The implementation stages of activ-ity-based costing and the impact of contextual and organi-zational factors. Journal of Management AccountingResearch, 10, 240–277.

Langfield-Smith, K. (1997). Management control systems andstrategy: a critical review. Accounting, Organizations andSociety, 22(2), 207–232.

Larcker, D. F. (1981). The importance of selected informationcharacteristics and strategic capital budgeting decisions. TheAccounting Review, July, 519–538.

Larcker, D. F. (1983). The association between performanceplan adoption and corporate capital investment. Journal ofAccounting and Economics, 5, 3–30.

Lau, C. M., Low, L. C., & Eggleton, I. R. C. (1995). Theimpact of reliance on accounting performance measures onjob-related tension and managerial performance, additionalevidence. Accounting, Organizations and Society, 20, 359–381.

Lawrence, P., & Lorsch, J. (1967). Organization and environ-ment. Homewood, Ill: Irwin.

Libby, T., & Waterhouse, J. H. (1996). Predicting change inmanagement accounting systems. Journal of ManagementAccounting Research, 8, 137–150.

Libby, T. (1999). The influence of voice and explanation onperformance in a participative situation. Accounting, Orga-nizations and Society, 24(2), 125–137.

Lindsay, R. M. (1995). Reconsider the status of tests of sig-nificance: an alternative criterion of adequacy. Accounting,Organizations and Society, 20(1), 35–53.

Luft, J. L. (1997). Fairness, ethics and the e!ect of managementaccounting on transaction costs. Journal of ManagementAccounting Research, 9, 200–216.

Luft, J. L., & Shields, M. D. (2000). Mapping managementaccounting: making structural models from theory-basedempirical research, 25th anniversary conference: Accounting,Organizations and Society. Oxford: Pergamon.

Macintosh, N. (1994). Management accounting and control sys-tems: an organizational and behavioral approach. Chichester,UK: John Wiley.

Macintosh, N., & Daft, R. L. (1987). Management controlsystems and departmental interdependencies: an empiricalstudy. Accounting, Organizations and Society, 23–48.

Mahoney, T. A., Jerdee, T. H., & Carroll, S. J. (1963). Devel-opment of managerial performance: a research approach. Cin-cinnati: South Western Publishing.

March, J. G., & Simon, H. A. (1958). Organisations. NewYork: John Wiley.

Martin, J. (1992). Cultures in organizations: three perspectives.Oxford: Oxford University Press.

McConnell, J. J., & Muscarella, C. J. (1985). Corporate capitalexpenditure decisions and the market value of the firm.Journal of Financial Economics, 14, 399–422.

Merchant, K. (1981). The design of the corporate budgetingsystem: influences on managerial behavior and performance.The Accounting Review, 4, 813–829.


Merchant, K. (1984). Influences on departmental budgeting: anempirical examination of a contingency model. Accounting,Organizations and Society, 9(4), 291–307.

Merchant, K. (1985a). Budgeting and the propensity to createbudgetary slack. Accounting, Organizations and Society,10(2), 201–210.

Merchant, K. (1985b). Control in business organisations. Bos-ton: Pitman.

Merchant, K. (1990). The e!ects of financial controls on datamanipulation and management myopia. Accounting, Organi-zations and Society, 15, 297–313.

Merchant, K., Chow, C. W., & Wu, A. (1995). Measurement,evaluation and reward of profit center managers: a crosssectional field study. Accounting, Organizations and Society,20, 619–638.

Merchant, K., & Simons, R. (1986). Research and control incomplex organizations: an overview. Journal of AccountingLiterature, 5, 184–201.

Mia, L. (2000). Just-in-time manufacturing, managementaccounting systems and profitability. Accounting and Busi-ness Research, 30(2), 137–151.

Mia, L., & Chenhall, R. H. (1994). The usefulness of manage-ment accounting systems, functional di!erentiation andmanagerial e!ectiveness. Accounting, Organizations andSociety, 19(1), 1–13.

Milani, K. (1975). Budget setting, performance and attitudes.The Accounting Review, 50, 539–558.

Miles, R. W., & Snow, C. C. (1978). Organizational strategy,structure and process. New York: McGraw Hill.

Milgrom, P., & Roberts, J. (1990). The economics of modernmanufacturing technology, strategy and organisation. TheAmerican Economic Review, June, 311–328.

Miller, D., & Friesen, P. H. (1982). Innovation in conservativeand entrepreneurial firms: two models of strategic momen-tum. Strategic Management Journal, 1–25.

Miller, D., & Friesen, P. H. (1986). Porter’s (1980) genericstrategies and performance: an empirical examination withAmerican data: part 1: testing Porter. Organizational Studies,7(1), 37–55.

Miller, J. M., DeMeyer, A., & Nakane, J. (1992). Benchmarkingglobal competition-understanding international suppliers, cus-tomers and competitors. Homewood, Ill: Irwin.

Miller, J. M., & Roth, A. V. (1994). A taxonomy of manu-facturing strategies. Management Science, 40(3), 285–304.

Mintzberg, H. (1994). The rise and fall of strategic planning.New York: The Free Press.

Moores, K., & Chenhall, R. H. (1994). Framework and MCSevidence. In K. Moores, & P. Booth (Eds.), Strategic man-agement accounting (pp. 12–26). New York: Wiley.

Moores, K., & Sharma, D. (1998). The influence of environ-mental uncertainty on performance evaluation style andmanagerial performance. Accountability and Performance,4(2), 1–16.

Moores, K., & Yuen, S. (2001). Management accountingsystems and organizational configuration: a life-cycle per-spective. Accounting, Organizations and Society, 26, 351–389.

Nouri, H., & Parker, R. J. (1998). The relationship betweenbudget participation and job performance: the roles of bud-get adequacy and organizational commitment. AccountingOrganizations and Society, 23, 467–483.

O’Conner, N. (1995). The influence of organizational cultureon the usefulness of budget participation by Singaporean-Chinese managers. Accounting Organizations and Society, 20,383–403.

Otley, D. T. (1978). Budget use and managerial performance.Journal of Accounting Research, 16(1), 122–149.

Otley, D. T. (1980). The contingency theory of managementaccounting: achievement and prognosis. Accounting Organi-zations and Society, 4, 413–428.

Otley, D. T. (1994). Management control in contemporaryorganizations: a wider perspective. Management AccountingResearch, 5, 289–299.

Otley, D. T., & Wilkinson, C. (1988). Organizational beha-viour: strategy, structure, environment and technology. InK. R. Ferris (Ed.), Behavioural accounting research: a criticalanalysis (pp. 147–170). New York: Century Publications Co.

Ouchi, W. (1977). The relationship between organisationalstructure and organisational control. Administrative ScienceQuarterly, 22, 95–113.

Ouchi, W. (1979). A conceptual framework for the design oforganizational control mechanisms. Management Science,25(9), 833–848.

Ouchi, W. (1980). Markets, bureaucracies and clans. Adminis-trative Science Quarterly, 25, 129–141.

Perera, S., Harrison, G., & Poole, M. (1997). Customer focusedmanufacturing strategy and the use of operations based non-financial performance measures: a research note. Accounting,Organizations and Society, 22(6), 557–572.

Perrow, C. (1970). Organizational analysis: a sociological view.California: Wadsworth Publishing Company.

Porter, M. (1980). Competitive strategy. New York: The FreePress.

Pugh, D., Hickson, D., Hinings, C., & Turner, C. (1968).Dimensions of organisational structure. AdministrativeScience Quarterly, 13, 65–105.

Pugh, D., Hickson, D., Hinings, C., & Turner, C. (1969). Thecontext of organizational structures. Administrative ScienceQuarterly, 14, 91–114.

Quinn, J. B. (1985). Managing innovation: controlling chaos.Harvard Business Review, May-June, 73–84.

Reid, G. C., & Smith, J. B. (2000). The impact of contingencieson managerial accounting systems development. Manage-ment Accounting Research, 11, 427–450.

Riggo, R., & Cole, E. (1992). Agreement between subordinateand superior ratings of supervisory performance and e!ectson self and subordinate job satisfaction. Journal of Occupa-tional and Organizational Psychology, 65, 151–158.

Rockness, H., & Shields, M. (1984). Organizational controlsystems in research and development. Accounting, Organiza-tions and Society, 9, 165–177.

Ross, A. (1994). Trust as a moderator of the e!ect of perfor-mance evaluation style on job related tension: a researchnote. Accounting, Organizations and Society, 19, 629–635.


Ross, A. (1995). Job related tension, budget emphasis anduncertainty. Management Accounting Research, 6, 1–11.

Scott, W. (1987). The adolescence of institutional theory.Administrative Science Quarterly, 32, 493–511.

Scott, T. W., & Tiessen, P. (1999). Performance measurementand managerial teams. Accounting, Organizations andSociety, 24(3), 107–125.

Selto, F. H., Renner, C. J., & Young, S. M. (1995). Assessingthe organizational fit of a just-in-time manufacturing system:testing selection, interaction and systems models of con-tingency theory. Accounting, Organizations and Society, 20,665–684.

Seymour-Smith, C. (1986). Macmillan dictionary of anthro-pology. London: Macmilllan Press Lmt.

Shannon, A. D. (1973). The strategy and structure of BritishEnterprise. Boston: Harvard University Press.

Shields, M. D. (1995). An empirical analysis of firms’ imple-mentation experiences with activity-based costing. Journal ofManagement Accounting Research, 7, 148–166.

Shields, M. D. (1997). Research in management accounting byNorth Americans in the 1990s. Journal of ManagementAccounting Research, 9, 3–61.

Shields, M. D., Deng, F. J., & Kato, Y. (2000). The design ande!ects of control systems: tests of direct and indirect-e!ectsmodels. Accounting, Organizations and Society, 25(2), 185–202.

Shields, J. F., & Shields, M. D. (1988). Antecedents of partici-pative budgeting. Accounting, Organizations and Society,23(1), 49–76.

Shields, M. D., & Young, S. M. (1993). Antecedents and con-sequences of participative budgeting: evidence on the e!ectsof asymmetrical information. Journal of ManagementAccounting Research, 5, 265–280.

Shortell, S. M., & Zajac, E. J. (1990). Perceptual and archivalmeasures of Miles and Snow’s strategic types: a compre-hensive assessment of reliability and validity. Academy ofManagement Journal, 33, 817–832.

Sim, K. L., & Killough, L. N. (1998). The performance e!ectsof complementarities between manufacturing practices andmanagement accounting systems. Journal of ManagementAccounting Research, 10, 325–346.

Simons, R. (1987). Accounting control systems and businessstrategy: an empirical analysis. Accounting, Organizationsand Society, 12, 357–374.

Simons, R. (1991). Strategic orientation and top managementattention to control systems. Strategic Management Journal,12, 49–62.

Simons, R. (1994). How new top managers use control systemsas levers of strategic renewal. Strategic Management Journal,15, 169–189.

Simons, R. (1995). Levers of control. Boston: Harvard Uni-versity Press.

Skinner, W. (1975). Manufacturing: the formidable weapon.New York: John Wiley.

Smith, K. J. (1993). Investment monitoring systems, abandon-ment of capital assets and firm performance. Journal ofManagement Accounting Research, 5, 281–299.

Snodgrass, C., & Grant, J. H. (1986). Cultural influences onstrategic planning and control systems. Advances in StrategicManagement, 4, 205–228.

Spicer, B. H. (1988). Towards an organizational theory oftransfer pricing process. Accounting, Organizations andSociety, 149–165.

Spicer, B. H., & Ballow, V. (1983). Management accountingsystems and the economics of internal organization.Accounting, Organizations and Society, 73–96.

Sprinkle, G. B. (2000). The e!ect of incentive contracts on learn-ing and performance. The Accounting Review, 75(3), 299–326.

Swieringa, R. J., & Moncur, R. H. (1975). Some e!ects of par-ticipative budgeting on managerial behaviour. New York:National Association of Accountants.

Swieringa, R. J., & Waterhouse, J. H. (1982). Organizationalviews of transfer pricing. Accounting, Organizations andSociety, 149–165.

Thompson, J. D. (1967). Organizations in action. New York:McGraw Hill.

Trist, E. L., & Bamforth, E. K. (1951). Some social and psy-chological consequences of the long-wall method of coal-getting. Human Relations, 4(1), 3–38.

Tushman, M., & Nadler, D. (1986). Organizing for innovation.California Management Review, Spring, 74–92.

Tymond Jr., W. G., Stout, D. E., & Shaw, K. N. (1998). Criticalanalysis and recommendations regarding the role of perceivedenvironmental uncertainty in behavioural accountingresearch. Behavioural Research in Accounting, 10, 23–46.

Ueno, S., & Wu, A. (1993). The comparative influence of cul-ture on budget control practices in the United States andJapan. International Journal of Accounting, 28, 17–39.

Vance, C. M., McClaine, S. R., Boje, D. M., & Stage, D.(1992). An examination of the transferability of traditionalperformance appraisal principles across cultural boundaries.Management International Review, 32, 313–326.

Van der Stede, W. A. (2000). The relationship between twoconsequences of budgetary controls: budgetary slack crea-tion and managerial short-term orientation. Accounting,Organizations and Society, 25(6), 609–622.

Venkatraman, N., & Ramanujam, V. (1987). Measurement ofbusiness economic performance: an examination of methodconvergence. Journal of Management, 13, 109–122.

Waterhouse, J., & Tiessen, P. (1978). A contingency frameworkfor management accounting systems research. Accounting,Organizations and Society, 3(1), 65–76.

Whitley, R. (1999). Firms, institutions andmanagement control:the comparative analysis of coordination and control systems.Accounting. Organizations and Society, 28(5/6), 507–524.

Williams, J. J., Macintosh, N. B., & Moore, J. C. (1990). Budget-related behavior in public sector organizations: some empiricalevidence.Accounting. Organizations and Society, 15(3), 221–246.

Williamson, O. E. (1985). The Economic institution of capital-ism. NewYork: Free Press.

Williamson, O. E. (1986). The economic organization: firms,markets and policy controls. Brighton: Wheatsheaf.

Woodward, J. (1965). Industrial organization: theory and prac-tice. London: Oxford University Press.


Young, S. M., & Selto, F. H. (1991). New manufacturingpractices and cost management: a review of the literature anddirections for future research. Journal of Accounting Litera-ture, 10, 265–298.

Young, S. M., & Selto, F. H. (1993). Explaining cross-sectionalworkgroup performance di!erences in a JIT facility: a criticalappraisal of a field-based study. Journal of ManagementAccounting Research, Fall, 300–326.


Documents

Management control systems design within its organizational