Power, M.- The Audit Society - Second Thoughts, Ija2000_111

SUMMARY

This essay reviews the central arguments of TheAudit Society (Power, 1999) and re-considers thecauses and consequences of the audit explosion.In the UK during the 1980s there was a systemicgrowth of auditing activity which needed to beexplained. The causes of this audit explosion canbe found in three areas: the rise of the newpublic management; increased demands foraccountability and transparency; the rise ofquality assurance models of organisationalcontrol. It is argued that these hypothesisedcauses require further empirical support andthat more research is needed, particularly todemonstrate that the audit explosion is notsimply a UK phenomenon. The consequences ofthe audit explosion are also considered and therole of auditable performance measures inshaping individual and organisational activity isdiscussed. Again, it is concluded that greaterempirical analysis is required. The essay then

addresses the criticism that the book does notdene auditing sufciently clearly. It is arguedthat the vagueness of the word audit was animportant condition of possibility for the auditexplosion. Finally, the prospects for reconstruct-ing audit to avoid dysfunctional side effects areconsidered.

INTRODUCTION

Financial auditing is an inferential practice whichseeks to draw conclusions from a limited inspec-tion of documents, such as budgets and writtenrepresentations, in addition to reliance on oraltestimony and direct observation. There is anestablished tradition of investigating thisinference process experimentally within theframe of cognitive science, although very little, ifany, of this kind of auditing research is done inthe United Kingdom. In contrast to this experi-mental tradition, a smaller body of work hasfocused on the social and institutional support forthe elements of this inference process. Indeed, itis argued that the appropriate organizing modelfor understanding auditing is not that of theisolated act of practitioner judgement, but thatof collectively negotiated settlements (Power,

International Journal of AuditingInt. J. Audit. 4: 111-119 (2000)

Received January 1999ISSN 10906738 Revised October 1999Copyright 2000 Management Audit Ltd. Accepted December 1999

The Audit Society - Second Thoughts1

Michael PowerLondon School of Economics and Political Science, UK*

ABSTRACTThis essay reviews the central arguments of The Audit Society(Power, 1999) and re-considers the causes and consequences ofthe audit explosion. It is argued that many of the claimsrequire further empirical support and that more research isneeded, particularly to demonstrate that the audit explosion isnot simply a UK phenomenon. Although the word audit mayhave decreased in signicance since the book was written, thearguments can be applied to other forms of monitoringactivity.Key words: audit; audit society; audit explosion; auditability;internal control.

*Correspondence to: Department of Accounting andFinance, London School of Economics and Political Science,Houghton Street, London WC2A 2AE;email: [email protected]

112 M. Power

Copyright 2000 Management Audit Ltd. Int. J. Audit. 4: 111-119 (2000)

1995). From this point of view, the very possibil-ity of individual judgement is the product ofmany other factors, including training in institu-tionally accepted practices of evidencecollection, such as statistical sampling (Power,1992; Carpenter and Dirsmith, 1993).

This modest sociological project was given anew context and a dramatic shift of focus asauditing broadly understood began to play aprominent social and economic role during themid to late 1980s. For example, the word auditbegan to be used with increasing frequency bypoliticians, regulators and consultants in manydifferent fields: health and safety, medicine,education, intellectual property, environmentalmanagement as well as the more establishedcorporate nancial auditing area (Power, 1994).During some ICAEW funded research on theborder territory between environmental andnancial auditing (Power, 1997) I had a hunchthat something systematic was going on,something that could not be captured orexplained only by reference to financialauditing. This required a fresh view of nancialauditing as just one possibility in an evolvingassembly of techniques and procedures. Fromthis point of view, it made sense to explore theconnections and resemblances between nancialauditing and other forms of assessment, evalua-tion and inspection. In short, it was necessary tolook beyond the silo of financial auditingstudies at monitoring, checking and reporting intheir most general sense.

The label for the hunch that something sys-tematic was going on over and above nancialauditing was the audit society. The conceptwas invented, like many ideas before it, in theLSE staff bar sometime in 1994; it was a kind ofmarker, a logical space for future work, ratherthan a well-dened concept with a clear point ofreference. The German sociologist Max Weberonce wrote that: an ingenious error is morefruitful for science than stupid accuracy(Quoted in Collins, 1992). It is for others to judgewhether, for all their aws, the arguments sup-porting the idea of the audit society are a usefulerror (See the critiques by Bowerman et al., 2000;Humphrey and Owen, 2000). It is clear that theideas of the audit society and of the auditexplosion (Power, 1994) require a great dealmore conceptual and empirical work. As theessays in this special volume of the InternationalJournal of Auditing suggest, we are still in search

of the audit society and the purpose of this essayis to revisit the central arguments and to pick upwhere they should be developed further.

To understand the audit society it is not suf-cient, although undoubtedly useful, simply toquantify the amount of auditing going on. It isalso important to understand the growth andcirculation of an idea of audit, a growth in whichaccountants have been powerful agents inselling their auditing capabilities but whichcannot solely be explained in this way. A practicelike nancial auditing is not simply a collectionof routines for collecting and evaluatingevidence. It is also very importantly an idea ormodel circulating in the institutional environ-ment, an idea whose fortunes practitionersthemselves cannot control, hence the expecta-tions gap for nancial auditing. This means thatauditing is not merely done in a technical sense,it is also used and appealed to, blamed, praised,regulated, reformed, written about and debated.The idea of audit and what it can or mightdeliver expresses the aspirational dimensions ofthe practice(s), dimensions which are not alwaysclosely linked to actual operational capacity. Inshort, the further one is from the detail, theeasier it is to talk of what auditing might, canand must do. Naturally, some of this talk takesplace in the profession itself, but it is not only aninternal affair.

Value for Money (VFM) auditing is a goodexample of the thesis; it was a powerful idea forpublic sector reform long before it approachedanything like a consistent and coherent practiceat the operational level. Indeed, the nuts andbolts of VFM auditing continue to be hotlydebated (e.g. whether effectiveness auditingextends to evaluation type work) and theconcept itself circulates freely despite consider-able operational variety (Pollitt et al., 1999). Sothe audit explosion was not only the explosionof concrete practices, it was also and necessarilythe growth and intensication of an idea aboutaudit in a number of different spheres. But why?

CAUSES

Although it is undoubtedly a simplication,three more or less discrete causes or pressuresgave rise to the audit explosion in the UK in thelate 1980s:

1. The cluster of ideas which constitute theNew Public Management (Hood, 1991)

The Audit Society - Second Thoughts 113


increased demands for financial and VFMauditing. The 1980s were characterised bynancial constraint and a commitment to orga-nizational and nancial reform in public sectorinstitutions. In this setting auditing and inspect-ing practices became highly valued andimportant tools of change. In different ways theNational Audit Office and the AuditCommission became prominent forces in gov-ernment, playing an evolving and complexconstitutional role. In the late 1990s thesepressures for change in the UK have not abated:for example, the Audit Commission willestablish a best value inspectorate (Thatcher,1999) and in health care standards of clinicalpractice will be enforced by the Commission forHealth Improvement (CHI). While the wordaudit may have decreased in importance, thedemand for monitoring has not. A freshexplosion may even be predicted on the back ofbest value demands.

2. The audit explosion was also driven byclosely related political demands on behalf ofcitizens, taxpayers, patients, pupils and othersfor greater accountability and transparency ofservice providing organizations. These demandsin the name of accountability were not onlyconned to the public sector; they have alsosurfaced in the private sector, most notably inthe developments in corporate governance.Indeed, private sector corporate governancethinking now provides a template for the organ-isation of public sector control (London StockExchange, 1998). As accountability and gover-nance issues span the traditional public-privatedivide, nancial auditing has been drawn into,and affected greatly by, these pressures. It is stillunclear how many of the proposals in the muchdiscussed Audit Agenda (APB, 1994), whichwas an exploration of the idea of financialauditing, will work out and the eld is beingpushed in many new directions in the name ofaccountability and assurance (AICPA). Forexample, internal auditing is acquiring a newprominence, particularly as a result of recentguidance in the UK on internal control (APB,1998; ICAEW, 1999; Power, 1999) and organisa-tions like the National Health Service areinvesting in risk management practices.

3. The third related cluster of pressures for anaudit explosion have come about from the rise ofquality assurance practices and related transfor-mations in regulatory style. The origins of these

pressures are complex and diverse, but a simpleand plausible story can be told with two parts.

The first part of the story concerns theemergence of quality assurance from an indus-trial production context to become a universalschema (e.g. ISO 9000). This all purposestructure is represented in gure 1.

Quality assurance programmes require thatorganisations and their sub-units establishobjectives, design performance measures toreect those objectives, monitor actual perform-ance and then feed the results of this monitoringback for management attention. Qualityauditing works both as the specic monitoringand reporting part of the system and also,importantly, as the verication of the systemstructure as a whole i.e. the integrity of the entireloop of self-observation which this proceduralstructure represents. This epitomises control ofcontrol (Power, 1999, p.66).

The second part of the explanation is a paralleltransformation in regulatory style, which ismoving away from a command and controlmode of operation. The intention is to regulatetarget organizations indirectly from below.Audit in its various forms becomes an importantpossible solution to the problem of regulatorycompliance, it denes a space in which regulato-ry compliance can be negotiated andconstructed (Hutter, 1997). Regulatory systemsrely increasingly upon control of control i.e. theaudit of self-control arrangements which areincreasingly characterised in terms of risk man-agement. The audit of these internal controls orself-checking arrangements is a growingindustry, an internal control explosion(Maijoor, 2000) which borrows its structure from

OBJECTIVES

PERFORMANCE MEASUREMENT

MONITORING AND AUDIT

REPORTING AND FEEDBACK

Figure 1: A General Model of Self-Auditingand Control

114 M. Power


the quality assurance model in gure 1. Perhapsnowhere is it more apparent than in the nancialservices industry where regulation is explicitlydesigned to rely on internal and externalchecking functions (Goodhart et al., 1998).Whether regulators are relying on in-house riskmanagement models for derivatives, or systemsfor teaching quality (Gray and Berry, 2000), thegeneral model in gure 1 is the same. Delegationand internalization of control are coupled toaudit processes which seek to reconcile locallearning and improvement mechanisms todemands for performance and regulatory com-pliance.

These then are three overlapping pressures forincreased auditing, evaluation and monitoringactivity. But does this story make too little ofaccountants and of Thatcherism in the 1980s?Certainly the large accounting rms have beeninuential agents of change, and the positionand role of accountants within economic andpolitical life in the UK as compared to othercountries has undoubtedly been a decisivefactor in shaping the rise of auditing in thiscontext. However, that this is an incompletepicture is suggested by the fact that nancialaudit, following the Eighth Directive, is itselfsubject to the same style of auditing, through theoperations of the Joint Monitoring Unit (JMU),as many hospitals, schools and other organiza-tions. One can hardly talk of a conspiracy, arational one at least, when the conspirators aresubject to the very same changes they areimposing on everyone else. It is interesting thatmany small UK accounting practitioners havecomplained about the audit of their auditingpractices in exactly the same terms that havebeen levelled against quality and medicalauditing i.e. it is too bureaucratic and toofocused on form and process.

That The Audit Society is not simply about theconsequences of Thatcherism is clear from theintensication of the culture of checking underNew Labour in the UK. The welfare state isincreasingly being displaced by the regulatorystate, and instruments of audit and inspectionare becoming more central to the operationalbase of government. However, it remains to beseen how well this argument travels to othercountries and systems. The analysis wasintended to be generic, and one can hypothesisethat an audit explosion would follow thediffusion of the new public management.

Further comparative research is needed to testthis view.

CONSEQUENCES

It has been suggested that society is experiment-ing on itself (Beck, 1992). Certainly, there hasbeen little planning or testing of new auditingpractices. Auditing has been introduced as anagency of organizational change without ameasured consideration of benets and possibledysfunctional effects. Although some assess-ment in terms of compliance costs has beenintroduced, audit and related ideas of monitor-ing continue to be understood uncritically; theycan be rapidly disseminated without developingan understanding of what may be at stake. Inthis respect it is plausible to suggest that theaudit explosion is fundamentally an ideological-ly driven system for disciplining and controllingdoctors, teachers, university lecturers and so on,and not an instrument of genuine accountability.

To analyse the consequences of auditing it isnecessary to focus on the development of whatis audited i.e. the performance measures andother forms of accounting which provide anauditable front stage for an organization (seegure 1). There is a developed research literaturewhich draws attention to the role of accountinginnovations, particularly in managementaccounting, in constructing organisational life(Miller 1994). This is not just a matter of whathas been called creative accounting i.e. thegames that exist to circumvent the intentionsbehind rules and regulations. Rather, accountingpractices, and performance measurementsystems more generally, create and support awindow on organizational life, one which isoften demanded by outside agencies, and whichmakes various kinds of internal and externalintervention possible.

The hunch behind The Audit Society is that thedesign of accounting reports, and of the per-formance measures by which organizations canbe judged, is greatly inuenced by the impera-tive of making them auditable, and that thishas much to do with agendas for control of theseorganisations. It follows that many auditprocesses are not neutral acts of verication butactively shape the design and interpretation ofauditable performance. Audit agencies, such asthe UK Audit Commission, are active in shapingperformance measures which enable audit and



inspection. Indeed, Bowerman et al (2000) arguethat it is audit which is really a by-product of amore fundamental performance measurementexplosion. They have evidence that many per-formance indicators are produced but are notaudited, and that there is more of an audit messthan a coherent audit society. However, justbecause a performance measure is not in factaudited does not mean that it was not designedwith potential auditability in mind. In the endthe argument is an empirical one.

Empirical work also needs to focus on thegrowing population of auditees, ie on the indi-viduals who have experienced an intensicationof checking and evaluation of what they do. Weare beginning to see how different games ofcreative compliance are being played aroundthe audit process, games which both frustrateofcial intentions and which also lead to dys-functional behaviour. Auditing can create newinterests at the expense of others. Academics inthe UK do not need to be told about some of thepitfalls of teaching and research evaluationexercises (Various, 1999). In some cases elaboratebureaucratic systems (reflecting the figure 1template) are created to manage an externalevaluation process. Whatever the performancemeasure (emissions levels, protability, exami-nation results, research quality, effectiveness ofmedical care etc), the audit process concentratesprimarily, but not always exclusively, on theloop itself, on the integrity of the structure ofself-observation and control which is theinternal control system.

The audit explosion refers to the rise of thissystems structure and its role in supporting reg-ulation and managerial control. However, gure1 and the system loop are only a blueprint. Thereis an extensive literature which questionswhether organizations do or could ever functionaccording to the formal self-transparency andvigilance required of gure 1 and its real worldinstantiations (such as ISO 9000 and its variants)(See Strathern, 1997). The formal managementcontrol system functions primarily to make acertain style of auditing possible; it buffers theauditor from an increasingly complex evidencebase, is cheaper than extensive attention toactual organizational process, and permits theaudit to provide more or less comforting signalsto regulators and politicians.

In The Audit Society I argue that institution-alised pressures exist for audit and inspection

systems to produce comfort and reassurance,rather than critique. If this is true and auditingsystems are primarily about reafrming order,then it will be interesting to see how auditableoutcome based performance measurement pro-gresses in the face of system decay, especiallygiven political rhetorics of zero tolerance forpoor performance. We can expect to see acuteproblems for anxious managers who, much liketheir former Soviet counterparts, will need con-siderable creativity to manage auditableperformance favorably in the face of objectivedecline. New pressure points, in the form ofcontrol agents and audit departments, will becreated at great cost to manage these contradic-tory forces. In the case of the UK researchassessment exercise, the explosion of journals ashomes for the research output of academics iswell known, but it is far from clear that thesubstance of research has really improved(Various, 1999). Forcing individuals whose com-parative advantages lie elsewhere to produceresearch, or feel that they should, is hardly anoptimal use of resources. In this way a systemwith the legitimate objective of makingacademics accountable for the quality ofresearch, and which is designed to measure andevaluate research quality, ends up affecting theeld adversely, even to the extent of creating asystemically sub-optimal transfer market.

Teaching and medicine are in a similarposition; newly established systems of auditingmay have damaged local cultures of rst orderpractice. Certain activities, which are valuedlocally, are not represented by ofcial systems orare lost in some general concept of goodwill.Ofcial auditable indicators, in the form of, say,exam results and waiting times, have strangeeffects on the incentive structure of individualsin organisations.2 If auditing processes getdecoupled from core organizational activities,these effects may be minimal and the auditprocess becomes an expensive but harmlessritual, which is important for external legitima-cy. Where performance measures and systemsdeveloped for audit do eventually force changesin organizational habits, these effects need to besystematically documented and fed back intothe design process. In short, there is a need formore research on the effects of making agents inorganisations accountable in terms of auditablemeasures of performance.

116 M. Power


The audit society reflects the growinginuence of auditors themselves as they moveinto areas of wider influence and serviceprovision. In particular, auditors are de factomajor interpreters of political mandates;accountability is concretely realised in the auditprocess. While there is considerable discussionabout the constitutional position of the NAOand Audit Commission (White andHollingsworth, 1999) ideals of policy neutralitycontrast with operational realities where juris-dictions are blurred and where the distinctionbetween audit and policy is unclear. The UKAudit Commission has come to be an explicitagent of change by promoting the systemswhich make auditing possible (Henkel, 1991;Thatcher, 1999)

Organizations must be changed internally tobe audited, and in many cases this may bedesirable. It is not denied that it may be a goodthing to require teachers to think seriously abouttheir objectives (Gray and Berry, 2000) or to makethe private world of hospital consultants moretransparent. Nor does the idea of the auditsociety assume a homogenous process of trans-formation across all elds. The key issue is todraw attention to the fact that audits do notoperate neutrally and have effects on the auditee.The audit explosion has been insensitive towhether these intended changes have been suc-cessful, and what unintended changes may haveresulted. But to say all this is to leave much of theempirical work to be done in exploring the con-sequences of the growth of auditing and views ofthese consequences will undoubtedly differ, asthe essays in this volume demonstrate.

CRITIQUE

Because the concept of the audit society wasinvented in advance of detailed empirical work,there remain many unresolved issues andproblems. Two will be addressed below.

1. It has been argued that the use of the wordaudit is not signicant. Auditing practices arevaried, and evaluation and inspection aredifferent again. (Bowerman et al., 2000;Humphrey and Owen, 2000). A great deal oftime and effort has been invested in deningand distinguishing meanings of audit, especiallyin emergent areas such as environmental andclinical auditing. There are also ongoing discus-sions about the nature of VFM auditing and the

difference between performance or effectivenessauditing and nancial auditing (Pollitt et al1999). Add to this discussions about evaluationsand assessments which do not have a veriableassertion, claimed as one of dening characteris-tics of audit, as their object and it begins to seemas if the audit society concept has no clearreference point.

It is difficult to deny the reality of thisdiversity, especially when this is reected in textbooks and ofcial accounts. Despite a reasonableattempt (Power, 1999, chapter 6), the relationbetween auditing, evaluation and inspectionremains far from clear. However, two pointsmust be made. First, despite all these attempts atdenition and distinction, the vague idea ofaudit has been an important reference point inpolicy; this vagueness can be observed andanalysed. From this point of view, The AuditSociety is as interested in understanding thepower of the word audit as it is in dening it.The idea of audit (and auditors) consists ofgeneral and highly idealised elements, and isappealed to and used in a wide variety of policycontexts. The rise of auditing is also as muchabout the cultural and economic authoritygranted to people who call themselves auditorsas it is about what exactly these people do.Indeed, we know that the people we callauditors (and inspectors) actually do manydifferent things.

To call something an audit places it within aeld of social and economic relations whichwould be different if it was called an assess-ment or evaluation. Labelling activities asaudits makes it possible for them to acquire theidealised characteristics of audit over time. Forexample, the label clinical auditing was used torename idiosyncratic research activity. Once thiscollection of ad hoc practices came to be calledauditing, it also became possible for it toacquire a new public accountability role. Inshort, against critics of The Audit Society, theargument is that vagueness is an essential partof the phenomenon. Once practices can beappropriated by the discourse of audit they canbegin to acquire similarities and shared opera-tional templates. In the audit society denitionsof auditing are dependent variables.

2. A second unresolved problem is the relationbetween auditing, organizational democracyand transparency. Being made to be accountable,being required to account and having this



account audited, are not the same as being mademore transparent or publicly accessible(Bowerman et al., 2000). One might argue thatsome audit processes discharge a certain style ofaccountability precisely to avoid transparency. Inother words, that an audit is done can be moreimportant than what is done and to whom anyreport is made; being audited per se is a badge oflegitimacy. Indeed, what seems to be importantis less the disclosure to stakeholders promisedby the rhetorics of accountability, which lit thefuse for the audit explosion, and more theprivate discipline of information gathering andcontrol which the audit process imposes.

A related puzzle or paradox of the auditsociety concerns the nature and periodicity ofaudit and inspection reports and the extent towhich such reports support deliberativeprocesses. Where audit reports are not designedto continue stakeholder dialogue then it isnecessary to trust the auditors and their inde-pendence becomes an important benchmark oftrust. Quality kitemarks and financial auditreports are like this. They are intended simply toindicate the quality of systems, products orservices. In contrast National Audit Ofce VFMaudit reports are in narrative form. And socialaudits, which seem messy and diverse, mayworry much less about independent auditreporting and more about stakeholder involve-ment (Cotton et al. 2000). Here the hypothesis isintriguing: greater stakeholder involvement inauditing, of whatever kind, may alleviateanxieties about auditor independence and maysupport less standardised reporting.

In The Audit Explosion I suggested that theaudit society is a less trusting society. However,it was pointed out that audit may in certain cir-cumstances be essential to restore trust betweenparties and to support institutions, such ascapital markets (Hatherly, 1995). This needsfurther investigation since many teachers anddoctors claim that evaluation processes haveachieved precisely the opposite by erodinginformal goodwill and by making individualsdevelop new incentives around crude perform-ance measures. There is some work on theseeffects and more is needed, but it is unlikely thatissues of trust in the audit society are as self-evident as The Audit Society suggests.

Research is also needed to address andevaluate the implications of new auditing andinspection institutions for the audit society

thesis. For example, the National Institute forClinical Excellence (NICE) will set clinicalstandards and the Commission for HealthImprovement (CHI) will enforce them andoversee local clinical auditing activity. A superregulator for OFTEL, OFSTED and other regula-tors has been proposed. Where do theseprocesses of control of control end? I havesuggested that the audit society threatens aninnite regress to the nth auditor as furtherlayers of regulatory inuence are created. Butsuch a theoretical regress must in practice stopsomewhere, and further work is needed todocument and explain the patterns of control ofcontrol in different elds.

CONCLUSION: RECONSTRUCTINGAUDIT

The primary critical intention of The AuditSociety remains relevant despite its weaknessesas an empirical set of arguments; the book issimply a plea for greater understanding of theconsequences of checking and monitoring forindustries, organizations and individuals. Someforms of audit and inspection need to beradically redesigned with greater sensitivity totheir consequences and without slavishadherence to performance measures which servethe audit process and little else. But there is nogeneral critical argument here, and each areaneeds to be looked at on its own merits. Theprospects of a light, self directed audit process,which harnesses productive learning and self-help to regulatory compliance, is an attractiveideal. Defensive auditing is the antithesis of thisideal.

An important and relatively under-researcheddevelopment is the rise of the internal auditfunction. This was suggested in an ICAS report(McInnes, 1993) and is now more evident asauditing is being redesigned with a risk man-agement emphasis (Selim and McNamee 1999a;1999b). Depending on how one looks at the phe-nomenon, there has been an internal controlexplosion (Maijoor, 2000) or an audit implosion(Power, 2000). In The Audit Society it was arguedthat the internal control system was becomingthe primary auditable object. Recent develop-ments in corporate governance and thecorresponding increase in signicance of theinternal auditor conrm this view and may inturn lead to a systematic re-balancing of the

118 M. Power


relation between internal and external checking(ICAEW, 1999). Once society overcomes itsconcerns about formal auditor independence, itwill become clear that expertise, operational inde-pendence and proximity to real time cultures ofcontrol are desirable. The Institute of InternalAuditors in the UK and USA sense the profes-sional opportunities and external financialauditors will face increasing competition.

The audit implosion thesis is consistent with amore responsive regulatory philosophy (Ayresand Braithwaite 1992) and there is a potentialboth for the worst excesses of the audit society tobe realized and for something more relevant,effective and sensitive to be created. As auditprocesses are re-designed to reect risk manage-ment ideals, the distinction between audit andconsulting becomes increasingly blurred(Jeppeson 1998), and this may not be to the tasteof regulators like the SEC. But a form of antici-patory and integrated risk management,overseen by the auditor/risk specialist mayrepresent the best opportunity to align publicpolicy and corporate objectives. The motif ofcontrol of control is likely to remain relevantand useful in characterising a regulatory systemwith a greater accent on internal self-inspection.This is especially evident from recent develop-ments in nancial services regulation (Goodhartet al., 1998)

Finally, the audit society can be understoodas a label for a loss of condence in the centralsteering institutions of society, particularlypolitics. So it may be that a loss of faith in intel-lectual, political and economic leadership hasled to the creation of industries of checkingwhich satisfy a demand for signals of order. Inthe UK auditing and inspection will be set towork in the name of best value and joined-upgovernment, but we may be forced to under-stand auditing as part of a general language ofdecline which attempts to bridge the wideninggulf between plans and achievements. Onemight even see auditing as an elaborate form ofconfession and periodic purication of organiza-tional order. The interpretations can becomeever more exotic but the critical point is clear: agreat deal of audit activity has had little to dowith efciency, and few large companies believethe stories of value added that nancial auditorspromote. The audit explosion is not simply afunctional response to complexity and toincreasing risk in different areas of society. And

yet equally there is a need for more sensitive andmore efficient technologies of checking andassurance because of the manufactured dangersthat we now face. The essays in this special issueof the International Journal of Auditing have madean important start in providing a more system-atic understanding of the consequences of audit.

NOTES1. This paper was originally presented at the

Seventh National Auditing Conference, CraneldSchool of Management, Cranfield University,March 21-2 1997 and at the ESRC / CIMA NewPublic Sector workshop, London School ofEconomics and Political Science, April 28-9, 1997.

2. In the private sector it might be assumed thatauditable performance measures are fully inte-grated into the core of corporate life. However,here there have also been extensive discussionsabout the limitations of nancial indicators andthere is currently much interest in the reporting ofnon-financial information (ICAEW, 1999).Experimentation in the UK with the Operatingand Financial Review is an attempt to pluralisecorporate reporting, but the doubtful auditabilityof qualitative information is regarded as a seriousconstraint on developments.

REFERENCESAICPA Assurance Services: New opportunities for CPAs,

http://www.aicpa.org/assurance.Auditing Practices Board (1998) Providing Assurance

on Internal Control, London: Auditing PracticesBoard.

Auditing Practices Board (1994) The Audit Agenda,London: Auditing Practices Board.

Ayres, I. and Braithwaite, J. (1992) ResponsiveRegulation: Transcending the Deregulation Debate,Oxford: Oxford University Press.

Beck, U. (1992) The Risk Society, London: Sage.Bowerman, M. Raby, H. and Humphrey, C. (2000) In

Search of the Audit Society: Some Evidence fromHealth Care, Police and Schools. InternationalJournal of Auditing 4(1), pp. 71-100.

Collins, R. (1992) Webers lost Theory of Capitalism: ASystematization. in Granovetter, M. and Swedburg,R. (eds.) The Sociology of Economic Life, Boulder, Co:Westview Press, pp. 85-110.

Cotton, P., Fraser, I. and Hill, W.Y. (2000) The SocialAudit Agenda - Primary Health Care in aStakeholder Society. International Journal of Auditing4(1), pp. 3-28.

Carpenter, B. and Dirsmith, M. (1993) Sampling andthe Abstraction of Knowledge in the AuditingProfession: An Extended Institutional TheoryPerspective. Accounting, Organizations and Society,18(1), pp 41-63.



Goodhart, C., Hartmann, P., Llewellyn, D., Rojas-Suarez, L. and Weisbrod, D. (1998) FinancialRegulation: Why, How and Whither Now? London:Routledge.

Gray, I and Berry, A. (2000) Some Things inModeration: A Case Study of Internal Audit.International Journal of Auditing 4(1), pp. 51-69.

Hatherly, D. (1995) Review of The Audit Explosion.Accounting and Business Research 24(96): 350-1.

Henkel, M. (1991) Government, Evaluation and Change,London: Jessica Kingsley.

Hood, C. (1991) A Public Management for AllSeasons. Public Administration 69(1), pp.3-19.

Humphrey, C. and Owen, D. (2000) Debating thePower of Audit. International Journal of Auditing4(1), pp. 29-50.

Hutter, B. (1997) Compliance: Regulation andEnvironment, Oxford: Clarendon Press.

ICAEW (1999) Internal Control: Guidance for Directors ofListed Companies Incorporated in the United Kingdom,London: Institute of Chartered Accountants inEngland and Wales.

ICAS (1993) Auditing into the Twenty First Century,Institute of Chartered Accountants in Scotland.

Jeppesen, K.K. (1998) Reinventing Auditing,Redening Consulting and Independence. EuropeanAccounting Review 7(3), pp. 517-539.

London Stock Exchange (1998) The Combined Code,London: LSE.

Maijoor, S. (2000) The Internal Control Explosion.International Journal of Auditing 4(1), pp. 101-109.

McInnes, W. (1993) Auditing into the Twenty FirstCentury, Edinburgh: Institute of ChartertedAccountants in Scotland.

Miller, P. (1994) Accounting as Social and InstitutionalPractice: An Introduction. In Hopwood, A. andMiller, P. (eds) Accounting as Social and InstitutionalPractice. Cambridge: Cambridge University Press,pp. 1-39.

Pollitt, C., Girre, X., Lonsdale, J., Mul, R., Summa, H.and Waerness, M. (1999) Performance or Compliance?Performance Audit and Public Management in FiveCountries, Oxford: Oxford University Press.

Power. M. (1992) From Common Sense to Expertise:Reections on the Pre-History of Audit Sampling.Accounting, Organizations and Society 17(1), pp.37-62.

Power, M. (1994) The Audit Explosion, London:DEMOS.

Power, M. (1995) Auditing, Expertise and theSociology of Technique. Critical Perspectives onAccounting 6, pp.317-339.

Power, M. (1997) Expertise and the Construction ofRelevance: Accountants and Environmental Audit.Accounting, Organizations and Society, 22(2), pp.123-146.

Power, M. (1999) The Audit Society: Rituals ofVerication, Oxford: Oxford University Press.

Power, M. (2000) The Audit Implosion: Regulating Riskfrom the Inside (London: ICAEW)

Selim, G and McNamee, D. (1999a) Risk Managementand Internal Auditing: What are the EssentialBuilding Blocks for a Successful Paradigm Change?International Journal of Auditing 3(2), pp. 147-155.

Selim, G. and McNamee, D. (1999b) The RiskManagement and Internal Auditing Relationship:Developing and Validating a Model. InternationalJournal of Auditing 3(3), pp. 159-174.

Strathern, M. (1997) Improved Ratings: Audit in theBritish University System. European Review 5,pp.305-21.

Thatcher, M. (1999) Riding Audits Big Bang. PublicFinance March, pp.18-21.

Various (1999) Knowledge for What? The IntellectualConsequences of the Research Assessment Exercise.History of the Human Sciences 12(4), pp.111-146.

White, F. and Hollingsworth, K. (1999) Audit,Accountability and Government, Oxford: ClarendonPress.

AUTHOR PROFILE

Michael Power is P.D. Leake Professor ofAccounting at the London School of Economicsand Political Science. He is also the Director ofthe Centre for the Analysis of Risk andRegulation and a former fellow of the Instituteof Advanced Study, Berlin.

Documents

Power, M.- The Audit Society - Second Thoughts, Ija2000_111