The Audit ExplosionThe word ‘audit’ is being used in the UK with growing frequency. In addition to financial audits, there are now environmental audits, value for money audits,management

The Audit Explosion

Michael Power

Open access. Some rights reserved.

As the publisher of this work, Demos has an open access policy which enables anyone to access our content electronically without charge.

We want to encourage the circulation of our work as widely as possible without affecting the ownership of the copyright, which remains with the copyright holder.

Users are welcome to download, save, perform or distribute this work electronically or in any other format, including in foreign language translation without written permission subject to the conditions set out in the Demos open access licence which you can read here.

Please read and consider the full licence. The following are some of the conditions imposed by the licence:

• Demos and the author(s) are credited;

• The Demos website address (www.demos.co.uk) is published together with a copy of this policy statement in a prominent position;

• The text is not altered and is used in full (the use of extracts under existing fair usage rights is not affected by this condition);

• The work is not resold;

• A copy of the work or link to its use online is sent to the address below for our archive.

By downloading publications, you are confirming that you have read and accepted the terms of the Demos open access licence.

Copyright Department Demos Elizabeth House 39 York Road London SE1 7NQ United Kingdom

[email protected]

You are welcome to ask for permission to use this work for purposes other than those covered by the Demos open access licence.

Demos gratefully acknowledges the work of Lawrence Lessig and Creative Commons which inspired our approach to copyright. The Demos circulation licence is adapted from the ‘attribution/no derivatives/non-commercial’ version of the Creative Commons licence.

To find out more about Creative Commons licences go to www.creativecommons.org

http://www.demos.co.uk/licence

http://www.demos.co.uk/licence

http://www.demos.co.uk

mailto:[email protected]

http://www.creativecommons.org

http://www.creativecommons.org

First published in 1994 by

Demos

9 Bridewell Place

London EC4V 6AP

Tel: 071 353 4479

Fax: 071 353 4481

© Demos 1996

All rights reserved

Paper No. 7

ISBN 1 898309 30 2

Cover illustration by Andrzej Krause

Printed in Great Britain by

White Dove Press

London

Typesetting by Bartle & Wade Associates

This page is covered by the Demos open access licence. Some rights reserved. Full details of licence conditions are available at www.demos.co.uk/openaccess

Introduction: General themes 1

Auditing and the shaping of accountability 9

Auditing and the rethinking of government 12

Audit as control of control 15

Auditing and the ideal of transparency 18

Auditing and regulatory failure 22

Making things auditable 25

Beyond audit 32

Summary 38

Notes 41

Acknowledgements 53

Demos iii

Contents


We must... decide whether our object in setting up the Guardian classis to make it as happy as we can, or whether happiness is a thing weshould look for in the community as whole.

Plato, The Republic

The people who produce this talk of change – professionals, politi-cians, administrators, committees, fund raisers, researchers and jour-nalists – are all mounting a complex sociodrama for each other andtheir respective publics. This takes the form of shamanism: a series ofconjuring tricks in which agencies are shuffled, new games invented,incantations recited, commissions, committees, laws, programmes andcampaigns announced. All this to give the impression that social prob-lems... are somehow not totally out of control. Promises and gesturescan be made, anxieties can vanish away or be exorcised, people can bereassured or mesmerised.

Stanley Cohen, Visions of Social Control

iv Demos

The audit explosion


The word ‘audit’ is being used in the UK with growing frequency. Inaddition to financial audits, there are now environmental audits, valuefor money audits, management audits, forensic audits, data audits, intel-lectual property audits, medical audits, teaching audits, technologyaudits, stress audits, democracy audits and many others besides.1 Moregenerally, the spread of audits and other quality assurance initiativesmeans that many individuals and organisations now find themselvessubject to audit for the first time and, notwithstanding protest andcomplaint, have come to think of themselves as auditees. Indeed there isa real sense in which 1990s Britain has become an ‘audit society’.2

What are we to make of this explosion of ‘audits’? What changes inthe style of government does it characterise? Is this a distinctive phasein the life of advanced industrial societies? More critically, how can apractice whose benefits are being privately questioned as never beforenevertheless come to occupy such an important role in public policy?Have alternatives to audit really become so unthinkable? Can we nolonger think of accountability without elaborately detailed policingmechanisms?

This essay explores these questions and goes on to ask whether theaudit explosion rests on firm intellectual and practical foundations orwhether it is as much a symptom of problems as their cure. It askswhether audits deliver what they promise in the form of greateraccountability, efficiency and quality or whether they in fact fuel the

Demos 1

Introduction: General themes


problems which they address by, for example, exacerbating distrust.Finally, it suggests a new agenda for balancing the aspiration forautonomy with external pressures for accountability.

The nature of the audit explosion is difficult to quantify but thereare a number of indicators. The establishment of the National AuditOffice and the Audit Commission in the early 1980s consolidated theaudit resources of central and local government respectively, and pro-vided an institutional focus for addressing the economy, efficiency andeffectiveness of publicly funded activities. Both these organisationshave expanded their work, particularly in value for money audit,bringing intensive scrutiny to many new areas such as the various ele-ments of the criminal justice system – police, forensic science, crownprosecution and probation services.3

Medical and teaching institutions are also set to become subject toextensive auditing regimes. Medical audits have acquired prominencein response to a recent government white paper4 and the newly estab-lished Higher Education Quality Council (HEQC) and its auditingactivities will become increasingly influential.5 In the field of qualityassurance more generally, the British Standards Institute has success-fully promoted BS5750, its standard for quality assurance, and hasused it to develop the BS7750 standard for environmental manage-ment systems. The European Commission has issued a regulation for avoluntary Eco-Management and Audit scheme which closely resem-bles BS7750. The likely take-up of these initiatives is still unclear but ithas been estimated that there is a $200 billion market to be covered byenvironmental consulting6 and the number of consulting organisa-tions in this area has risen dramatically in the last ten years.7 Therecent development of accreditation schemes for environmental audi-tors has provided a further stimulus.8

The major accounting firms grew very quickly during the 1980s.The proportion of university graduates entering traineeships withaccountancy firms peaked at over 10% in 1987 and is currently run-ning at about 8%.9 Of this number, a majority receive their primarytraining in financial audit work although relatively few remain in thisfield. One important dimension of the UK audit explosion is that

2 Demos

The audit explosion


unprecedentedly large numbers of young people are being trained andsocialised in the context of auditing.

In traditional financial audit the trends are less conspicuous. Whilethe number of statutory entities requiring audit has grown steadily, it isthe more intensive role of audit which is more notable. The statutoryfinancial audit of companies has become more highly regulated andcodified over the last twenty years. The Auditing Practices Committee,replaced by the Auditing Practices Board in 1991, was formed in 1976and has produced technical guidance on a wide range of issues.Developments have been evident in two particular fields: financialregulation and charities. In the case of both fields, particular statutoryinitiatives have extended the role of audit. The Financial Services Act1986 and the Banking Act 1987 have given auditors newly explicitresponsibilities for assessing internal controls and for communicatingwith regulators. Many have complained that the costs of these arrange-ments are out of proportion to their benefits.10 Charities have alsocome under renewed regulatory scrutiny. Supplementary provisions tothe Charities Act 1992, and accounting guidance specifically tailoredto the sector, reflect a determination to subject these organisations toincreased financial discipline via audit.11

In other areas, the audit explosion has taken different forms. Safetyand hazard audits in industry have grown naturally from health andsafety legislation. Data audits, which originated in the United Statesand are less prominent in the UK, have arisen from concerns aboutscientific fraud.12 UK public science will soon find itself subject tovalue for money, intellectual property and technology audits as gov-ernment seeks both to make science accountable to its funding publicsand to exploit its intellectual property base.13

Despite these developments, the audit explosion is only in part aquantitative story of human and financial resources committed toaudit and its extension into new fields. It also concerns a qualitativeshift: the spread of a distinct mentality of administrative control, a per-vasive logic which has a life over and above specific practices. One cru-cial aspect of this is that many more individuals and organisations arecoming to think of themselves as subjects of audit. To describe this

Demos 3



logic, this essay relies upon a more oblique and lateral approach to thephenomenon of audit than quantitative methods would permit.14

In what follows I will make eight more or less discrete arguments:First, that despite differences in context and meaning, there is a

common thread to the new uses of the word ‘audit’. Sceptics may doubtwhether the proliferating usage of a single word really signifies anysystematic relationship between the diverse contexts within which it isinvoked.After all, audit is hardly an unambiguous concept and it couldbe argued that the practices to which the label is attached are in factdiverse and that they are constituted by very different bodies of knowl-edge. For example, it is possible to distinguish audits on the basis oftheir relation to the auditee. Many audits, such as in medicine, are conceived primarily as internal reviews to improve decision-making.Some of the growth of audits has been of this kind, intended to sup-port rather than to discipline, and very different from ex post verifica-tions which have much more the character of a policing role and forwhich the independence of the auditor is crucial. The extent to whichaudits are oriented towards verification is therefore variable and manycommentators would wish to argue that value for money auditingplays an entirely different role.

But there are important linkages between the different contexts ofaudit. Forms of ‘self-audit’ rely upon bureaucratic procedures whichcan in principle be used for independent verification purposes, even incontexts such as medical audit.15 Indeed, checklists and protocols forapparently internal purposes often derive their authority from theirpotential use for external verification. Formal documents can be usedoutside their original context and in ways unanticipated by those whomay have designed them. In addition, the experience of other manage-ment areas suggests that even pre-decision reviews may have a ex postjustificatory function.16

Second, that audit is not just a series of (rather uninteresting) tech-nical practices. It must also be understood as an idea. It is usual, partic-ularly in official documents and text books, to conceive of audit onlyin terms of its technical and operational qualities. While this imagereinforces its reputation as a boring and parasitic practice, it disguises

4 Demos

The audit explosion


the importance of auditing as an idea. Audit has become central toways of talking about administrative control. The extension of audit-ing into different settings, such as hospitals, schools, water companies,laboratories, and industrial processes, is more than a natural and self-evidently technical response to problems of governance and accounta-bility. It has much to do with articulating values, with rationalising andreinforcing public images of control. The audit explosion is the explo-sion of an idea that is internal to the ways in which practitioners andpolicy makers make sense of what they are doing.17

Third, that the spread of audits and audit talk corresponds to a fun-damental shift in patterns of governance in advanced industrial soci-eties.As I have suggested above, the explosion of audit practices in newareas is, at least in the UK, not simply a quantitative intensification.It arises out of changing conceptions of administration and gover-nance.18 Accordingly, to understand this explosion we must dig deeperand look wider than preoccupations with technical and institutionalissues. I wish to suggest that audit has emerged at the boundarybetween the older traditional control structures of industrial societyand the demands of a society which is increasingly conscious of itsproduction of risks, in fields ranging from the environment, to medi-cine and finance.19 It is one of many features of a far-reaching transi-tion in the dominant forms of administration and control, both ingovernment and in business.

As such, audit is a way of reconciling contradictory forces: on theone hand the need to extend a traditional hierarchical command con-ception of control in order to maintain existing structures of author-ity; on the other the need to cope with the failure of this style ofcontrol, as it generates risks that are increasingly hard to specify andcontrol.

Fourth, that the pervasive feature of the new wave of audits is thatthey work not on primary activities but rather on other systems ofcontrol. For example, recently proposed quality assurance mechanismsfor higher education require audits of the quality assurance systems ofhigher education institutions.20 This gives the audit a more remoteassurance role than is often understood by the publics which they are

Demos 5



intended to serve. It is in this sense that a gap between words and deedmay exist -commonly referred to in financial audit contexts as anexpectations gap. Audits are often not directly concerned with thequality of performance, whether environmental, educational or finan-cial, but rather with the systems in place to govern quality. This ‘polic-ing of policing’ distinguishes the audit explosion from an oldertradition of engineering-based quality control and its statisticallygrounded methods.

Fifth, that audits do not contribute automatically to organisationaltransparency. Despite the fact that audit talk is driven by demands forgreater transparency of organisational and individual action, thecapacity of audit to deliver this is problematic. Often the extension ofaudits can make organisations more obscure, and the audit processitself remains publicly invisible despite the commitment to makingorganisations transparent. It may be that the audit explosion signifies adisplacement of trust from one part of the economic system toanother; from operatives to auditors.

Sixth, that audits have the remarkable capacity of being invulnera-ble to their own failure. In recent years the primary concern of publicdebate has been with financial audit. Often regarded as a model forother forms of audit, it has been the subject of extensive critical com-mentary, set off in large part by the recent demise of the Bank ofCredit and Commerce International (BCCI) and the Maxwell empire.Rightly or wrongly, corporate collapse is always accompanied byscrutiny of the role of the auditors and, in some cases, litigation on thegrounds that they have performed their task negligently.21

One of the surprising features of these experiences is that they tendnot to call into question the role of audit itself. Instead, where audit hasfailed, the common response has been to call for more of it. Indeed, thegreat puzzle of financial audit is that it has never been a more power-ful and influential model of administrative control than now, whenmany commentators talk of an auditing crisis.22 Accordingly, I suggestthat the audit explosion shares an important character with all kinds ofpolicing: all have problematic criteria of success and are generally onlypublicly visible when they are seen to fail. But failure generally leads to

6 Demos

The audit explosion


a call for more policing and only rarely for a thorough analysis of whypolicing is failing.

Seventh, that audit is not passive but active. Not only does it shapethe activities which it controls in critical ways but it represents a veryparticular conception of accountability. Far from being passive, auditactively constructs the contexts in which it operates. The most influen-tial dimension of the audit explosion is the process by which environ-ments are made auditable, structured to conform to the need to bemonitored ex-post. Audits do not passively monitor auditee perform-ance but shape the standards of this performance in crucial ways, andpublic conceptions of the very problems for which it is the solution.

Eighth, that, notwithstanding the dominance of audits there areother ways of achieving accountability.

These arguments are intended to demonstrate the institutionalfoothold that audit now has in the public imagination while raisingserious questions about this impact. The central concern however isthat the audit explosion has made it difficult to think of alternatives toitself. But any society or organisation can use very different models ofcontrol and accountability, which can be summarised in the followinglists:

STYLE A STYLE B

Quantitative QualitativeSingle Measure Multiple MeasuresExternal Agencies Internal AgenciesLong Distance Methods Local MethodsLow Trust High TrustDiscipline AutonomyEx Post Control Real Time ControlPrivate Experts Public Dialogue

The audit explosion has involved an overwhelming priority for style Aas the solution to any problem (although value for money auditingcould be regarded as embodying both).23 Quantified, simplified, ex-post

Demos 7



forms of control by outsiders have increasingly displaced other typesof control. As a result of its institutional power, and its power as anidea, proponents of alternative styles have found it hard to gain anaudience.

This shift has brought a complex bundle of gains and losses’24 and itis not the intention of this essay to suggest that there have been nogains at all from the growth of audit. However, these gains are likely tobe most visible when used in conjunction with, rather than in opposi-tion to, elements of control style B. One example of this is when medical audits help practitioners reflect on clinical methods and man-agement as well as offering a mechanism for external evaluation. As inall things, the key is to achieve a balance and compromise. In this essaymy purpose is to offer a diagnosis which may assist in restoring a bal-ance that has been lost.

8 Demos

The audit explosion


Auditing seems as natural and as necessary as policing. One may argueabout its precise form, for example the balance between preventionand detection, but there is likely to be agreement that these things aredone because of the way society is.Without them there would be morefraud, deception, waste, error and poor administration.

Specialised academic stories have formalised these intuitions aboutthe need for audits: they will be demanded where there are relations ofaccountability between two parties together with a certain complexityor distance such that one, the ‘principal’, cannot easily and directly ver-ify the activities of the other, the ‘agent’.25 In recent economic literatureit is argued that it will be rational for the auditee to contract voluntar-ily to undergo an independent audit in order to make good offices visible. On this view audits are costly monitoring technologies whicharise naturally under conditions where agents expose principals to‘moral hazards’, because they may act against the principal’s interests,and where there are ‘information asymmetries’, because they knowmore than the principals. Audit is therefore a risk reduction practicewhich inhibits the deviant actions of agents. At the extreme thisimplies that ‘Four people performing a cooperative task, say loadingtrucks, find that the risk of any one of them slacking is such that theyhire a fifth to monitor their work’.26

Audits are needed when accountability can no longer be sustainedby informal relations of trust alone but must be formalised, made visible

Demos 9

Auditing and the shaping ofaccountability


and subject to independent validation. This story can be told in rela-tion to companies. When directors were generally regarded as trust-worthy and shareholders were perceived as largely ignorant ofbusiness matters, financial audit was very limited. Despite being ruf-fled by the occasional scandal, confidence in the expertise and honestyof directors seemed a sufficient guarantee of financial accountability.

This consensus began to dissolve in the 1930s largely as a conse-quence of the famous Royal Mail case and subsequent reforms to com-pany law.27 Over time financial audit began to assume an ever moreimportant regulatory function. The rise of the corporate economy,coupled with a statutory audit requirement, ensured its expandinginfluence and, by implication, that of its practitioners.

It has become commonplace to interpret these developments interms of an erosion of trust. The financial audit arose because the rela-tionship between management and increasingly distant providers offinance28 was becoming problematic. The practical problems were alsoreflected – and reinforced – both by abstract economic theories of thecorporation and by the organisation of Anglo-American capital mar-kets.29 It came to be accepted that actions could no longer be coordi-nated by trust and that instead independent ‘outsiders’ had to be used to restore that trust by providing ex post validations of auditee performance.

In banking and financial services auditors have in recent yearsassumed an increasingly visible regulatory role as informal controlstructures have been replaced. The demise of the Maxwell empire, andthe regulatory responses which it has stimulated, can be interpreted asyet a further episode in the cycle of mistrust ensuring that despite pub-lic criticism, financial audits will become more intensive and morecentral in relation to questions of corporate governance.30

I want to suggest an alternative to this common view which seesaudit as a response to problems of accountability that have originatedelsewhere. Instead I will argue that audit has spread as much becauseof its power as an idea, and that contrary to the assumptions of thestory of lost trust, its spread actually creates the very distrust it ismeant to address.

10 Demos

The audit explosion


It is important not to overstate this claim. It would certainly be farfetched to say that audit literally creates the pathologies for which it is theprescribed treatment, but I would not rule this out in particular cases.People may adapt their behaviour to reflect the fact that they are nottrusted and thereby confirm that they should not be trusted. For examplein higher education it has recently been suggested that academics havemanipulated examination results to conceal matters from funding coun-cil quality assessors, something which they probably would not havedone in the absence of an auditing process with funding implications.31

Moreover, in contrast to the popular image of audit as a derived andparasitic activity, audit shapes conceptions of accountability whichfavour audit as the solution.32 ‘Principals’ and ‘agent’, may seem to betheoretical categories, abstractions from practical realities, but in different guises such as ‘service provider’ and ‘customer’, they havebecome common in official ways of talking about accountability.

But rather than solving the problem of trust, these models ofaccountability simply displace it.33 If those engaged in everyday workare not trusted, then the locus of trust shifts to the experts involved inpolicing them, and to forms of documentary evidence or in manage-ment assurances about system integrity.34 Ultimately there is a ‘regressof mistrust’ in which the performances of auditors and inspectors arethemselves subjected to audit. Thus, Inland Revenue inspection hasbeen subject to value for money audit as have the police force. A some-what ironic parallel in finance has been the new regime for ‘auditingthe auditors’ under the requirements of the Eighth EuropeanCommunity Directive on company law.

These examples show that the audit explosion is not simply a prod-uct of the rise of specific audit specialists, such as the ‘Big Six’ firms ofaccountants. It has much to do with the momentum of audit as an ideaand as a system of knowledge.

Why then has the idea of audit, as a particular approach to account-ability issues, become so prominent in recent years? How has the logicof audit become so widely generalised? The answer lies largely in trans-formations in the role of government and conceptions of governance.

Demos 11

Auditing and the shaping of accountability


A ‘new public management’ has taken shape in the last twenty years,influenced greatly by images of private sector administration. Publicaccountability has been reframed in relation to concepts such as goaldefinition, efficient resource allocation, financial performance andcompetition.35 Of course, effectiveness of service delivery remains anideal but it is less prominent and forms one, often subordinate, compo-nent of managerial language in the public sector.

This realignment of public management styles and objectives has acomplex history which is beyond the scope of this analysis. However,for the purposes of this argument it is essential to understand that thereinvention of government36 is informed by two opposite tendencies.On the one hand, there are centrifugal pressures for the decentralisa-tion and devolution of services and for turning parts of governmentinto enterprises, whether through full privatisation or partial ‘enter-prization’. Deregulatory initiatives and the investment of regulatoryauthority in ‘private’ control agencies, such as the Securities andInvestments Board in the context of financial services, also reflect fun-damental transformations in the style of government, away fromdirect provision and towards oversight and rule-setting.37

On the other hand, there are equally powerful pressures to retain con-trol over functions that have been made autonomous. Financial over-sight of Next Steps and other government agencies is one component.

12 Demos

Auditing and therethinking ofgovernment


But as important is the intensive regulation of privatised utilities, andthe strong, if delegated, regulation of financial markets in the name ofconsumer protection and market integrity.

These competing pressures, to devolve on the one hand and to con-trol on the other, constitute a distinctive idea of government.Consistent with a liberal mission, the UK state is increasingly commit-ted not to interfere or engage in service provision directly; it seeks tofulfil its role by more indirect supervisory means. In many cases thestate has become regulator of last resort, operating indirectly throughnew forms of control (such as the independent regulator) which havethe appearance of being apolitical. The great attraction of audit andaccounting practices is that they appear to reconcile these centrifugaland centripetal forces better than the available alternatives.38 The con-sequence is a displacement in the terms of government discourse,from service-specific values of teaching, care and so on to moreabstract, financial and quantitative categories.

Why has this redesign of government happened? One plausibleexplanation is simply that the fiscal crisis of Western governmentswith generous welfare states has made much tighter financial disci-plines necessary. While it is for future economic historians to judge,there is no doubt a great deal of truth in this story. However, it is also alittle too lean and rational in its form. And it does not give us a feel forthe particular forms which disciplines have taken.

For there have also been other sources of the audit explosion.One set of influences has been the growing preoccupations withquality assurance which can be traced in part to concerns aboutindustrial competitiveness. Others include the breakdown of theconsensus behind the welfare state, public grievances againstexperts and professionals, and the rise of human-made risks innearly all areas of life.

In all these cases, the great attraction of the audit idea is its portabil-ity across such diverse contexts: public sector efficiency, corporategovernance, environmental management systems and so on. The wordsymbolises a cluster of values: independent validation, efficiency,rationality, visibility almost irrespective of the mechanics of the practice

Demos 13

Auditing and the rethinking of accountability


and, in the final analysis, the promise of control. All of these apparentvirtues have come together to make audit a central part of the ‘reinven-tion of government’. But audit is not just an idea, it is not just a story ofcontrol. What gets done in its name matters and it is this that we mustnow consider.

14 Demos

The audit explosion


One of the paradoxes of the audit explosion is that it does not corre-spond to more surveillance and more direct inspection.39 Instead,audits generally act indirectly upon systems of control rather thandirectly upon first order activities.

As organisations have grown in scale and complexity direct formsof inspection have become too expensive. Instead audits have becomeorganised around internal systems of control. The paradigm exampleis the systems audit which is the conventional model for financialaudits. Rather than examining large quantities of transactions, audi-tors focus on the control systems governing those transactions.40 Thisis equally true of the work of the European Court of Auditors which, ofnecessity, relies heavily on the work of national agencies such as theUK’s National Audit Office. Audit has thereby become the ‘control ofcontrol’,41 where what is being assured is the quality of control systemsrather than the quality of first order operations. In such a contextaccountability is discharged by demonstrating the existence of suchsystems of control, not by demonstrating good teaching, caring, manu-facturing or banking.

Financial auditing texts promote this kind of systems audit as one ofthe higher stages in the evolution of the practice. But the truth is thatthey are rationalising a shift away from direct contact with practiceswhich has been primarily driven by cost. The danger is that it is nowmore important to an organisation’s legitimacy that it is seen to be

Demos 15

Audit as control ofcontrol


audited than that there is any real substance to the audit. Even thefiercest critics have become caught up in this logic, as the public issuehas become the independence of auditors rather than their compe-tence or relevance.42 What these critics ignore is that even with strongguarantees of independence, systems based audits can easily become akind of ritual,43 concerned with process rather than substance, andgoverned by a ‘compliance mentality’ which draws organisations awayfrom their primary purposes.

But there is also another reason for the spread of systems auditsbeyond the economic unviability of real time inspection. As Day andKlein have argued in the context of the schools inspectorate, ‘inspec-tion... is about peer judgement by professionals reviewing the work oftheir fellow professionals’.44 The justification for audit is that (at leastfor government) trust and valuation has moved away from the profes-sionals, engineers or carers, so that even independent inspectors arenot deemed trustworthy, because they are embedded within the pro-fession. Instead only abstract systems of control can be deemed whollyindependent.

This very abstraction from first order detail has greatly helped theexplosion of audits in different fields. Detailed conceptions of qualitymay be very different if a supermarket is compared to a hospital. Butthe general principles of quality control systems for both can be madeto look very similar, enabling one to compare them at an abstract level.

This is a recent development. In the nineteenth century, such finan-cial audit guidance as existed was organised along industry-specificlines; the audit of railways was different from that of banks and so on.Over time this gave way to more abstract conceptions of the auditprocess despite the continuation of industry-specific guidance.45 Thesystems audit represents another stage in this evolution; by abstractingfrom local organisational diversity it has enabled audit to assume thestatus of an almost irresistible cultural logic.

This mobility and diversity of application of audit is its great attrac-tion; it can be invoked by marketeers and planners, entrepreneurs andregulators, consumers and producers, citizens and states. Indeed, ascitizens’ charters take hold audit will no doubt offer the prospect of

16 Demos

The audit explosion


realigning the relationship between patients and doctors, students andlecturers, passengers and transport operators.

But the institutional strength of audit also brings problems.Auditees develop creative strategies to cope with being audited. Inmany fields there is a sense that the tail of audit is increasingly wag-ging the dog of accountability and there are doubts about whetheraudits really empower the agents which they are intended to serve. It isto these questions that I now turn.

Demos 17

Audit as control of control


Accountability is so closely associated with ideas of transparency thatthe two concepts are often used interchangeably. Audits are usuallyjustified as enhancing the transparency of individual and corporateactions to those parties who have an interest in the nature and effectsof those actions. In other words, they are thought to shift power; fromprofessionals to the public, from experts to stakeholders.

However, if we look more closely at the nature of this transparencyand its democratising potential we soon find problems. How exactlydoes audit make things transparent? What are the mechanics of trans-parency?

To answer these questions we must distinguish two related issues:the transparency of the audit process and the transparency of auditfindings.

The closure of the Bank of Credit and Commerce International in1991 provides a good example. It stimulated considerable debate aboutthe nature of the financial audit function in banks. However, the offi-cial report by Justice Bingham on the collapse did not concern itselfwith the audit process, preferring instead to deal with the requirementfor auditors to report directly to bank supervisors.46 Its justificationwas that questions of technical process are matters for the expertsthemselves. Whatever the merits of this view it means that the auditprocess, the mechanism by which organisations are made transparent,is not an object of public policy or open to scrutiny.

18 Demos

Auditing and the ideal of transparency


Most commentators would agree that this opacity of process, epito-mised in the BCCI case, is at the heart of what is called the ‘expecta-tions gap’ in financial auditing: the difference between how financialauditors are perceived (responsible for the detection of fraud) and howthey see themselves (primarily responsible for forming a professionalopinion on the financial statements).47

Over the years there has been much debate on this issue within thefinancial auditing profession. However, there are really only two possi-ble solutions: either the users of audit opinions must be educated tohave appropriate (lower?) expectations or the audit product must bebrought into line with those expectations that do exist. In either case,the pressure is on those involved in financial audit to become morepublic about its objectives and process.

But throughout these debates on the audit ‘expectations gap’ therehas been little appreciation that it can provide a useful service to bothusers and auditors. This is not as perverse an idea as it sounds, and itapplies to all kinds of policing. Those doing the regulation, policing or auditing benefit from expectations which exceed what they candeliver, because these translate into higher fees and prestige.Meanwhile those in whose name the audits or policing is carried outbenefit from a sense of assurance, even if this is not firmly grounded.Many would rather not know that their vulnerability is greater than itseems.

One could go further and suggest that the audit explosion hasoccurred at least in part because of, rather than despite, expectationgaps about the nature of audit. Its very ambiguity has helped it servediverse needs,48 and its opacity has helped its expanding role in gov-ernment, serving the needs (and status) of the professionals involved,and comforting politicians and a wider public that things are undercontrol. Audit can be likened to a shiny black box on the surface ofwhich the aspirations of new regulatory programmes can be reflectedand made possible. From this point of view, it is actually undesirable tolook beneath the surface of audit practice into the box, to make theaudit process more publicly transparent. In an important sense regula-tors do not want to know what auditing really is.

Demos 19



Another telling example of non-transparency is to be found in thefield of environmental audits, where the general ambiguities of audithave been compounded by extensive concerns about what environ-mental audits really are. The widely accepted definition proposed bythe International Chamber of Commerce requires that environmentalaudits are primarily a ‘management tool’.49 This managerial under-standing of environmental audit has justified limiting their statutorystatus and their level of public disclosure. The justification for conceiv-ing environmental audit in this way is to preserve it as a market basedsolution to problems of environmental risk.

The EC Environmental Management and Audit (EMA) schemewhich was issued as a Regulation in 1992 is a voluntary scheme andrequires auditors (‘external verifiers’) to validate a limited environ-mental statement. Whatever benefits these environmental audits mayoffer in terms of improved management systems, cost savings, legalcompliance and so on, it is less clear how they contribute to theempowerment of external parties such as the public, local authorities,employees, shareholders or other firms. These uncertainties wereshown all too clearly by the problems involved in compiling a publicregister of contaminated land in the UK, and the government’s deci-sion to withdraw proposals for such a register just before they weredue to be introduced.

However, even extensive public disclosure in the name of trans-parency is an ambivalent phenomenon. Disclosure can have a pacify-ing effect on publics, can serve to convince them that something is orwill be done by someone and can ultimately deter inquiry rather thanencourage it. Disclosure can serve to amplify trust in the audit processrather than stimulate critical analysis of its results, since it often tendsto shift trust towards new audit institutions, such as accreditationarrangements. The effect is to deter rather than invite inquiry aboutthe auditee and the audit process. In other words trust in the fact thatan audit is done displaces public preoccupations with what is done andwhat is discovered.

As we have already seen, the audit explosion is characterised less byan opening up of organisations and more by the reinvestment of trust

20 Demos

The audit explosion


in new bodies of audit expertise and its legitimation through suchthings as accreditation and monitoring systems. For accountants, thishas taken the form of new disciplinary arrangements; for environmen-tal auditors it has generated a politics of accreditation for external ver-ifiers.50 All are designed to build confidence in the practitioners of anaudit function which remains hidden from view.51

The result is a paradox. The public image is that audits are con-ducted in the name of making visible the inner workings of organiza-tions. Corporate financial audits are formally intended to serve thegoal of shareholder control by linking the operations of corporateboardrooms to the decision-making calculus of distant financiers.Academic audits have as one of their goals the empowerment of ahitherto powerless student body. All promise external visibility ofinternal processes.52 Yet audit is itself an increasingly private and invis-ible expert activity. The alternative may be simpler:

‘So far, the emphasis of public policy has been to respond to com-plexity by setting up new institutions of accountability....this may, inturn, bring about excessive complexity in the machinery of accounta-bility and at the same time create dead ends. So, why not concentrateless on formal links or institutions and engage more in a civic dialogueto recreate at least something of the high visibility and directness ofthe face to face accountability..’53

Alternatives of this kind, involving direct accountability and activeinteraction, have been almost wholly ignored by those who see auditsas a universal panacea. But transparency alone does not empower, andparadoxically, may even serve to pacify and neutralise other possibleforms of accountability, such as those based on answerability. At theextreme, audits which have become tightly interwoven in regulatoryprogrammes can do more to promote obscurity than transparency.54

Demos 21



The public image of audit is constantly threatened by ‘real world’ prob-lems: the continuing visibility of ‘healthy’ companies which fail, theaccumulation of ecological risks, the decline of educational standards,the misuse of public revenues and so on.55

However, in all these case audit shares a characteristic with all polic-ing activities: it is usually only at times of perceived failure that auditsattract public attention. Successful auditing, even if this could be easilydetermined, is not very newsworthy. Indeed, one of the ironies of theaudit explosion is that, despite being part of a culture that is much moreconcerned with institutional (and personal) performance, it is nevereasy to see how effective audits are. Crime rates may fall because ofimproved policing or because of demographic changes. Equally, corpo-rate financial statements may be generally reliable because of goodauditing or because of good company accounting policy.56 In short,there is something profoundly elusive about the benefits of audit.57

The other side of the coin is that when things go wrong someonehas to take the blame. As informal methods of regulation have givenway to increasing formalisation, auditing has been periodically drawninto the blame allocation process. Legal judgments are the mostimportant part of that process but there are many others, not least ofwhich is critical press comment. Regulators often also blame eachother. The fraud related collapse of Barlow Clowes, the Levitt Groupand others attracted considerable adverse press comment but there

22 Demos

Auditing and regulatory failure


was also evidence of pressures between auditors and regulators withthe latter often seeking to blame and reform the former.58

Failures tend to lead to a combination of complex negotiations toallocate blame together with reforms of the audit process. Thus, in thecase of the collapse of BCCI and its aftermath the response hasfocused in part on the blameworthiness of auditors Price Waterhouse,and new codified guidance to auditors is to be produced; auditors willsoon have a statutory duty to report to regulators on certain mattersrather than, as prior to BCCI, a right to do so.59

But amongst these various battles to pass the buck, there is a strik-ing general trend; that audit almost always becomes more explicitlycodified in the wake of processes of blame allocation. In the wake ofscandals and frauds, audit failure is usually located as a particularproblem of fraudulent management or credulous auditors and isaddressed by extensive codification of new official audit guidance. Theprinciple seems to be that the general efficacy of audit must be pre-served even if the reputation of particular practitioners is not.

These are complex processes but the tendencies are clear. Successivesequences of failure involve the use of audits as a restorer of comfort,each time in a more intensive form, and each time apparently betterimmunised against failure, since every failure is particular and everysolution general. And in all this there is a good deal of cosmetics.Indeed, technically detailed realignments of audit practice sometimesresemble the elaboration of epicycles to preserve Ptolomaic cosmol-ogy, an impressive effect of form rather than content.

Similar doubts apply to the validity of environmental audit as amanagement tool which functions to reconcile commercial advantageand the production of goods to environmental concerns about theproduction of bads.60 This reconciliation is epitomised by the conceptof BATNEEC, Best Available Technology Not Entailing ExcessiveCost. So far, environmental auditing does not yet have its BCCI orMaxwell and so we cannot point to a regulatory failure. However, weshould be wary, since environmental audit may come to assume aninstitutional importance similar to that of financial audit and isalready making alternatives to itself unthinkable.

Demos 23

Auditing and regulatory failure


We should be sceptical above all because audit embodies a notalways justifiable optimism about the possibility of achieving controlover processes which have become ever more complex, such asEuropean Union fraud, industrial pollution and so on. Rather thansolving problems, it may also insulate us from that complexity, despitethe occasional scandal, in order to preserve faith in its efficacy. Thatfaith depends in turn on matters of organization or process alwaysbeing auditable. It is to this dimension of the audit explosion that Inow turn.

24 Demos

The audit explosion


The public image of auditing as a dull but worthy practice implies thatit is neutral. Official views emphasise the collection of evidence of anappropriate quality and quantity.61 If unambiguous standards of audi-tee performance can be established, then audit simply verifies compli-ance with such standards. Where there is deviation from standardsthen auditors report and advise. From this point of view, the standardsof auditee performance are independent of the audit process.

Yet the opposite is often the case. Audits do as much to constructdefinitions of quality and performance as to monitor them. For exam-ple, financial accountability demands financial audit as a control overthe ‘quality’ of the accounts provided by the auditee and the auditorachieves this by the periodic and independent ex post inspection ofbooks, records and control systems. However, most financial auditorswould admit, and take some pride in, their often decisive influence onthe standards of performance to which the auditee is subjected. Theyare far from being neutral relays in an exogenously given accountabilitysystem. Not only do these standards of financially conceived perform-ance make the accountability of corporations visible in highly specificways; they also constitute a ‘financial rationality’ which has been influ-ential in recent transformations in public sector audit practice.

Public sector audits officially enable government bodies, via theoperations of the UK NAO and the Audit Commission, to observethemselves critically and objectively in the service of initiatives in

Demos 25

Making things auditable


public financial management and policy accountability. There hasbeen much debate about the policy neutrality of Value For Money(VFM) auditing and the role of the NAO more generally, and it isbecoming widely recognised that VFM audit, which lies at the heart ofthese changes, is much more than a monitoring ‘technique’. It isintended both to evaluate and to shape the performance of the auditeein three dimensions: economy, efficiency and effectiveness.

However, it is well known that in many service sectors the notion ofeffectiveness is not easily calibrated and requires a range of financialand non-financial measures. It has been suggested that VFM priori-tises that which can be measured and audited in economic terms –efficiency and economy – over that which is more ambiguous andlocal – effectiveness.62

Performance in the sense of effectiveness is difficult to assess wherepractitioners are unaccustomed to thinking in such terms. For exam-ple, nurses, policemen and women, prison officers, research workersand others, have traditionally used highly localised standards of qual-ity control. In all these cases, too, concepts of success and failure areproblematic: rising levels of ill-health, crime and prison disorder havemany causes and cannot simply be regarded as ‘failures’ of the servicein question. For all its proclaimed sensitivity towards context, VFMdemands that effectiveness be quantifiable. It does this by standardis-ing measures of effectiveness (on the one hand) and/or by reducingeffectiveness to standardisable measures of economy and efficiency.Either way, there is a necessary drift towards ‘managing by numbers’which enables a drift towards centralised forms of control and the dis-placement of concerns about good policy by concerns about goodmanagement.63

Where the measurement and attribution of outputs from a serviceare ambiguous, or the preserve of the service expert, there is a ten-dency to concentrate upon inputs. For example, in the case of childcare it may be that social workers themselves are unable to agree aboutwhether fostering or residential care is most effective in nurturing thebalanced development of children. In this case, it is natural to focus onunambiguous measures of input, primarily cost. It follows that efficiency

26 Demos

The audit explosion


in this context may come to be seen in terms of cost saving for existinglevels of service provision rather than an improved relationshipbetween inputs and outputs, which in industrial contexts representsproductivity.64 In this way, like other forms of audit, VFM involves adisplacement from first order experts, such as teachers, social workers,police and so on to second order experts, such as accountants andmanagers.

These displacements do not necessarily reflect public distributionsof trust. Indeed it is an irony that audits of this kind bring a shift fromprofessions the public trusts more-such as doctors, police and teach-ers, to a profession the public trusts less (the accountants) at the insti-gation of a profession the public trusts least (the politicians).

Undoubtedly, there are benefits from disturbing closed professionalcultures; local forms of peer review may simply foster a tyranny ofexpertise.VFM may be genuinely democratising, opening up perform-ance to wider scrutiny. Measures of economy, efficiency and effective-ness may be arbitrary but if they are widely accepted they may also beimpartial and public.65 However, against the possible benefits of a newimpersonal objectivity in public service must be weighed the transfor-mations in the service itself. It is a difficult balance.

The concept of VFM has also become increasingly linked to that ofquality despite their different institutional origins. Recent initiatives inacademic audit for both teaching and research provide another inter-esting example of the complex logic of audit process because here toothere is currently widespread debate about what constitutes quality.66

There has been widespread concern that the Research AssessmentExercises conducted by the higher education authorities in the UKhave contributed to a climate in which quantity takes precendenceover the quality of publications67 because of the mechanisms ofresearch output measurement which enable an audit of research per-formance to be conducted. The significant economic consequences of such audits inevitably generate new incentive structures forresearchers. The problem is not that review and assessment methodsare wrong per se but that the very technologies of audit may paradoxi-cally achieve the opposite of their intended effect.

Demos 27



One other element of this debate is the role of users – specifically inthis case the weight to be attached to student questionnaires as meas-ures of teaching quality. Many lecturers favour questionnaires andwere using them as a form of self-audit prior to the recent initiatives.Others are opposed on principle. Whatever the intrinsic merits, ques-tionnaires make teaching ‘auditable’, and thus easier to regulate in thename of quality.68 In other words what is at stake here are not only thedeclared objectives, the improvement of teaching quality, but also astyle of control over teaching. The ideal of ‘auditability’ gives commit-ments to quality and effectiveness operational meaning.

For verificatory activity always requires that there is something ver-ifiable. In many instances, these types of information are required bylegislation; for example, company law requirements for financialreporting and the maintenance of proper books and records, or otherarrangements such as the pronouncements of organisations such as the British Standards Institute (BSI) and the Higher EducationFunding Council which emphasise the role of control systems.69

It was noted in the introduction that audit is generally a form of con-trol of control.What is subject to inspection is the auditee’s own systemfor self monitoring rather than the real practices of the auditee.70 Whatis audited is whether there is a system which embodies standards andthe standards of performance themselves are shaped by the need to beauditable. In this way, the existence of a system is more significant foraudit purposes than what the system is; audit becomes a formal ‘loop’by which the system observes itself.71 While there is a sense in whichthis immunises the auditee from the audit process, the necessity of hav-ing an ‘auditable system’ nevertheless impacts upon real time practicesand has obvious resource implications for auditees. Auditable systemsrequire subjects to represent themselves primarily as auditees.Undoubtedly, new games will emerge around these systems-drivenaudit processes. Forms of ‘non-compliance compliance’ and creativestrategies of submitting to audit processes will frustrate the mission ofaudit.72 At the same time organisations may be encumbered with struc-tures of auditability embodying performance measures which increas-ingly do not correspond to the first order reality of practitioners’ work.

28 Demos

The audit explosion


The construction of auditability around auditable systems is perhaps most evident in the British Standards Institute initiatives onquality assurance (BS5750) and environmental management systems(BS7750). Both standards, which are highly abstract as compared toother BSI standards, emphasise the development of management con-trol systems. The standards are the core of an accreditation processwhich has become a new product and is sold to companies who wishto gain a marketing advantage from being seen to comply.

These BSI standards have focused managerial attention in manybeneficial directions but it is also true that their value as productsresides to a great extent in their promise of auditability qua accred-itability.

The critical question is whether the ideal of auditability implicit in these initiatives may in fact eclipse the development of first orderstandards themselves. The assurance process may simply become tooimportant. BS7750 has been developed with a view to requiring com-panies to articulate their own benchmarks of environmental perform-ance against which they can improve in subsequent years. Hence,BS7750 articulates the structure of an environmental managementsystem but not the standards of performance themselves. While this islikely to concentrate managerial energies in positive directions thereare also dangers as audit begins to take on a life of its own increasinglydecoupled from the processes and events which it is intended toaddress.

The most recent instance of this tendency is the draft guidance forexternal verifiers under the EMA scheme,73 which is symptomatic ofthe way auditing creates webs of increasingly dense procedure andcompliance checklist of suitably elaborate detail.74 The ideal checklistreflects the relevant performance of the auditee and ensures the com-pleteness and visibility of audit work. The checklist determines the rel-evant facts about the auditee and insulates the particular auditor fromneeding to worry about the background to the list, as well as providinghim or her with a defence against critics, especially in a court of law.

But what should be the knowledge basis of audit? Some would saythat legal knowledge is crucial. However, the auditor has always been

Demos 29



burdened with the need to understand the regulatory environment ofthe audited entity. In my view the key to auditing is the ‘inferentiallogic’ (or lack of it) which enables an auditor to form conclusionsabout the whole on the basis of examining and testing some of itsparts. At various periods in its history, financial audit practice hasattempted to establish linkages between a non-inferential proceduralknowledge base and ‘higher’, abstract bodies of inferential knowl-edge.75 However, the latter have been invoked as much to confer scien-tific credibility upon the practice as to provide instrumental guidance.Hopes were expressed that statistical sampling could place auditing ona scientific footing and the importance of sampling technologies to ascientific image of audit cannot be underestimated. Indeed, the sym-bolic importance of statistical sampling is relatively unaffected by itsactual applications. As a way of talking, writing and justifying, sam-pling makes possible rational representations of the audit process, andthereby of the auditee.76

The general point is that the system of auditing knowledge isincreasingly self-referential. It models organisations for its own pur-poses and impacts to varying degrees upon their first-order opera-tions. Ironically, one area where this is particularly apparent is in thecontext of the regulatory arrangements for financial auditors in theUK which have come about as a consequence of the EC EighthDirective on Company Law. This imposed new requirements for theregulation of financial auditors, including that they should be licensedand subjected to compliance visits by monitors.77 These complianceaudits are decoupled from underlying organisational processes butnevertheless force the auditee to take on some of the shapes requiredby the audit process, if only because resources are being spent on com-pliance.78

The policy question is whether it has all gone too far. Are auditingmechanisms of control themselves out of control? Is there a price to bepaid for a logic of auditability? Are there real benefits to service qualityand effectiveness which override the local doubts of threatened practi-tioners? Or is the language of quality and VFM an elaborate rhetoricfor cost reduction in the face of a public sector borrowing crisis? There

30 Demos

The audit explosion


is no doubt that VFM has been instrumental in the reorganization ofpublic life around audit and accounting processes which make organi-sations ‘selectively visible’.79 And there can be no doubt about themomentum which quality assurance initiatives now have.

In a real sense power has shifted from social workers, doctors, andother public sector workers towards auditors and quality insuranceinspectors because the latter increasingly control the terms of publicdiscourse. Moreover their power is such that there is now almost noway that reservations about audit can be articulated without appearingto defend privileges and secrecy.

However, the logic of auditability is far from being entirely mono-lithic and public concerns about too much reviewing, about preoccu-pations with systems rather than performance and about the cost ofaudit in relation to its claimed efficacy are becoming more apparent.So what is to be done?

Demos 31



The analysis developed above is intended to be critical but withoutbeing wholly dismissive. It does not point to a conspiracy of the vestedinterests of accounting practitioners since the audit explosion is visiblein areas where they have little influence. Rather I have attempted tosupply a diagnosis of the position in which we find ourselves. Before wecan make prescriptions about how audit practices may be re-designed,we need to understand the economic and political conditions underwhich a practice, which might ordinarily be regarded as supplementaryto other forms of administration, has become so central.

We seem to have lost an ability to be publicly sceptical about thefashion for audit and quality assurance; they appear as ‘natural’ solu-tions to the problems we face. And yet, just as other fashions havecome and gone as the basis for management thinking, the audit explo-sion is also likely to be a passing phase.

The seeds of a change may be there. There is certainly dissatisfac-tion with the primacy of audit which is becoming more vocal. Firstly,the cost of audit in relation to its claimed benefits, particular in thefields of financial audit and quality assurance arrangements, is beingquestioned in the light of its potential impact upon competitiveness.Secondly, there may be difficulties with the public legitimacy ofaccountants, particularly in the public sector. A recent survey of publicopinion suggests, as noted earlier, that accountants are less unpopularthan politicians but much less popular than the police, teachers and

32 Demos

Beyond audit


doctors.80 Thirdly, there are concerns about organisational governanceand the efficacy of audit arrangements in relation to them. In otherwords, the syndromes of regulatory failure may not always reaffirmthe mission of audit; for example, some would now argue that capitalmarket stability might be better served by increasing the internal representation of institutional stakeholders rather than by reliance on external forms of measurement and oversight.

But these grievances do not yet form part of an institutionallyacceptable language. The first difficulty in creating such a language isthat we need to generate a scepticism about our current preoccupa-tions with ideals of ‘performance’ and ‘quality’ and the technologiesthrough which they are made operational while recognising that sucha scepticism has dangers; namely that it risks ignoring the benefits ofaudit and providing an apologia for outmoded forms of expert privi-lege, insulating actors from all forms of responsibility and therebythreatening accountability itself.

The second difficulty is that if policy recommendations are to beconstructed on the back of the diagnosis which I have offered thenthey will need to find their primary conceptual resources in some veryunfashionable areas. For example, concepts of trust and autonomy willneed to be partially rehabilitated into managerial languages in someway. In the mid 1990s we have reached a position where, in the pursuitof performance measurement, anxieties have been fuelled thatthreaten to destroy the commitment of individuals to their organisa-tions to such an extent that this may undermine performance.Insecurity among senior executives calls into question the very idea of‘organisational loyalty’.81 The audit explosion is not the only culprithere but it is one of a range of administrative instruments whichthreaten to be as much part of the problem as of the cure.

It should be clear that the policy issue is not simply a question ofreducing the quantity of auditing. It is rather to design organisations insuch a way that they better manage themselves and their relationshipsto outsiders so that audit is not automatically seen as a ‘natural’ solu-tion. If one dominant institutional logic is to be replaced, then new onesmust fill what would otherwise be a vacuum. My proposals can only be

Demos 33

Beyond audit


suggestive but if audit is as much a norm as a technical practice then itfollows that one needs new vocabularies as well as new practices.

Firstly, a new respect for local specificity needs to be generated and,by the same token, a suspicion of control instruments which standardisefrom a distance. This suspicion already exists but it requires institutionalmechanisms to give it voice and to make public the potential of auditpractices to create systems which serve the audit process and little else.We may need to give up certain myths of long distance regulation.82

Accordingly, the concept of accountability needs to be both loosenedand tightened. It needs to be loosened as regards the levels of ritualisticdetail through which auditability is conducted by remote agencies ofcontrol. The benefits of these highly elaborated audit procedures areincreasingly out of line with the burden of costs which they imposeupon the auditee.83 However, this loosening must also be supplementedby performance concepts which seek to restore, at the extreme, face toface forms of peer group accountability and structures which cutthrough layers of process and make organisational dialogue possible.Some might see this as an appeal to self-regulation and the restorationof expert privilege but it is more an argument that corporate commu-nity is a necessary precondition for corporate governance. The tensionbetween consensus and discipline is a constant difficulty as recent crit-icism about the style of HEFC academic audit reports indicates.84

Whereas recent initiatives in corporate governance seem to havemagnified the importance of audit, it must be possible to conceive ofgovernance outside this framework. For this we will need to take seri-ously new institutional spaces in which stakeholders of every varietycan assert their claims as ‘principals’. Rather than placing faith in auditcommittee mechanisms, a greater variety of possibilities for represen-tative forums needs to be considered. Public discussion about the ben-efits of the Cadbury proposals has gravitated towards the costs andbenefits of non-executive directors; the efficacy of committees of thiskind is more or less assumed.85

Secondly, we need to generate a healthy scepticism about officialmodes of disclosure and audit and to experiment with non-(financial)accounting and non-audit based methods linking stakeholders and

34 Demos

The audit explosion


enterprises. Again, new languages are needed. Performance measure-ment will need to be supplemented by qualitative concepts such asfacilitation. In addition we will require new ways of defining theboundaries of organisations since audit arises from the externalisationof insiders (more traditionally described as the separation of owner-ship from control) and the self-fulfilling image of the corporation as anetwork of contracts.

Where audit and the forms of disclosure which arise out of it rein-force the externality of outside stakeholders, we may need organisa-tional innovations through which they can be brought back in throughrights of access and inquiry. Accounting and auditing conceptions ofthe ‘user’ of financial statements are generally very abstract. Whileaccounting and information system re-design will continue to play animportant role, so too may forms of organisational participation forreal as opposed to mythical users of financial information. Whereasaudited disclosure tends to deter inquiry and encourage trust in theaudit process, other mechanisms may be capable of encouraging andsustaining inquiry. Forms of internalisation of stakeholders may alsobe information-efficient to the extent that formal audit and account-ing would become redundant. In addition, institutionalised avenuesfor whistleblowing, a practice from which the Inland Revenue hasbenefited, could provide another alternative to the audit model.86

In the light of these suggestions the ambivalence of Citizens’Charters must be noted. Such charters embody a mixture of rights toknow and rights to choose. However, it is likely that the right to choosewill not have much substance if the choices in question run counter tothe need to curtail public expenditure. If so, the operational weight ofcitizens’ charters will fall upon the right to know or, in other words, thetransparency of service organisations. In this respect it must be askedwhether they really herald a new era of popular governance driven byparticipation and dialogue or whether they will effectively become an‘Auditors’ Charter’, a symptom of the failure of democracy and empow-erment rather than its cure.87

In addition, there needs to be greater sensitivity to the manner inwhich instruments of verification can transform the contexts in which

Demos 35

Beyond audit


they are applied and can bias accountability in unintended directions.Decisions need to be taken about whether this is desirable or not and,in place of faith, some empirical knowledge needs to be acquired aboutthe behavioural effects of auditing in different fields. For the auditexplosion has been driven less by an empirically grounded under-standing of the productive benefits of audit and more by a pervasivebelief, almost ideological in form, in the need for the discipline whichit provides. It is ironic that in many instances audit practices are them-selves insulated from the market forces and ideals of deregulationwhich they serve.

In the end, looking beyond audit requires that we also rethink whatit is that makes organisations work. Western fascination for Japaneseorganisational structures seems to have been unable to affect the auditmentality.88 Japanese structures depend much more on ‘style B’ typesof control: horizontal rather than vertical, trust-creating, qualitativerather than quantitative.

Like them, we would benefit from having less respect for abstractforms of portable knowledge and more respect for non-standard andtacit kinds of knowledge which are complex and close to their prod-ucts. Here I would make a personal plea on behalf of the complexity ofuniversity teaching. The tide of consumer enfranchisement mayempower students in one sense but it may also impoverish them in thelonger run by cultivating an aversion to difficulty, ambiguity and critique unless it is carefully managed. Courses will increasingly bedesigned primarily with student evaluations and other audits in mindsuch that teachers will avoid risk and therefore innovation. Withoutwishing to invoke a mythic golden age, an important relationshipbetween teachers and students may nevertheless be damaged.89

In looking beyond audit an entire regulatory apparatus will need tobe redesigned with the question,‘What are organisations for?’ firmly inits sights. The interim report of the RSA inquiry into ‘Tomorrow’sCompany’ points to the need for an ‘inclusive approach’ which is local-ist in regulatory spirit and clearly has implications for auditing.90 If weend up using more of the Style A forms of audit described in section 1,they will need to be light and modest compliance exercises which do

36 Demos

The audit explosion


not generate more expectations than they merit and whose benefitscan be clearly articulated. Although this will undoubtedly challengeexisting structures of professional status, we will not be sending thenation’s brightest and best to work in an abstract ‘economy of compli-ance’. Instead we will be placing them closer to the productiveprocesses – in services, manufacturing and the public sector – in alltheir diversity. In this way we will surely be in a better position to cre-ate quality rather than just to police it.

Demos 37

Beyond audit


1. There has been an explosion of audits in many different fields: medi-cine, science, education, technology, environment, intellectual propertyto name but a few. Audit has assumed the status of an all purpose solu-tion to problems of administrative control. Despite concerns about itscosts, the benefits of audit are assumed by its proponents rather thanproven. Even though the efficacy of financial audit is currently beingdoubted as never before, particularly in the wake of corporate scandals,auditing remains powerful. Indeed, alternative mechanisms of long-distance control have become almost unthinkable.

Audit is an emerging principle of social organization which may bereaching its most extreme form in late twentieth-century Britain. It isas important as an ideal – a set of ideas or logics – as a practice. Manyaudit practices have grown because of changes in public sector man-agement and newly prominent ideals of quality, governance andaccountability. And yet paradoxically, while audit technologies havecontributed to managerial concerns about ‘performance’, the perform-ance of audit itself is far from being unambiguous and free from pub-lic dispute. But the spread of audits constitutes a major shift in power:from the public to the professional, and from teachers, engineers andmanagers to overseers.

2.Audits are not simply answers to problems of accountability. Theyalso shape the contexts in which they are demanded in importantways. Submission to audit has become such a benchmark of institutional

38 Demos

Summary


legitimacy that resistance and complaint look like attempts to preserveabuses of privilege and secrecy. The audit explosion has its origins invery recent transformations of government which have sought todevolve many of its functions while retaining regulatory oversight.

3. Audit is linked to ideals of organisational transparency andaccountability. Yet audits are themselves often very specialised andopaque to a wider public. Audits may provide comfort to stakeholderswho are remote from day to day practices but, in doing so, they oftendeter substantive inquiry which would empower stakeholders. Auditarrangements can bring an end to dialogue inside and outside organi-sations, rather than helping it. In so far as audit is directed at evokinggood feelings about organisational practices it may also become a newform of image management rather than a basis for substantive analy-sis. In this sense, the fact of audit is becoming more important than thehow of audit.

4. Audits are usually publicly visible when they fail. Their benefitsare often ambivalent and a source of controversy. Audit reconstitutesitself in a syndrome of regulatory failure: it emerges from crises insti-tutionally secure despite processes of blame allocation within the reg-ulatory world. Problems are defined as particular, while generalsolutions are usually offered.With each crisis, audits become more for-malised and intensive.

5. Audits are not passive practices but strongly influence the environ-ments in which they operate. Instead of involving direct observation,audit is largely an indirect form of ‘control of control’ which acts on sys-tems whose role is to provide observable traces. In a number of areasthis results in a preoccupation with the auditable process rather than thesubstance of activities. This in turn burdens the auditee with the need toinvest in mechanisms of compliance, a fact which has produced a con-sistent stream of compliant. Concepts of performance and quality are indanger of being defined largely in terms of conformity to auditableprocess. Indeed, the construction of auditable environments has necessi-tated record-keeping demands which only serve the audit process.

6. Auditing is a peculiar form of alchemy which, in making auditeesauditable, produces regulatory comfort. It is conservative not in the

Demos 39

Summary


sense of a conspiracy of vested interests, but in the sense of being a sys-tem of knowledge which filters and appropriates the unforeseen. In theprocess of constructing subjects as responsible auditees local struc-tures of trust are displaced and potentially distorted. Any reduction inaudit intensity, and the possibility of forms of organisation in whichgroups and individuals are given autonomy, is literally unthinkablewithout a new institutional language which does not so much rejectaudit as assign it to its proper place.

In looking beyond audit we shall need to recognise that a certainstyle of accountability, which values independent scrutiny, is one valueamong others. Audit displaces trust from first-order to second-orderverificatory activities. We may need to rehabilitate trust at the level offirst-order performance, change the organisational conditions underwhich audit appears to emerge naturally and even give up on the ritu-alistic details in which accountability is discharged by audit. In doingso we need to reposition audit as a local and facilitative practice, ratherthan one that is remote and disciplinary, so as to enable rather thaninhibit public dialogue. External forms of audit will need to be moremodestly conceived. This will require a broad shift in control philoso-phy: from long distance, low trust, quantitative, disciplinary and ex-post forms of verification by private experts to local, high trust,qualitative, enabling, real time forms of dialogue with peers. In thisway we may eventually be in a position to devote more resources tocreating quality rather than just to policing it.

40 Demos

The audit explosion


1 It has also been brought to myattention that the terms ‘audit’ and‘auditor’ have for many years playeda central role in the ideas of theChurch of Scientology.

2 This essay is based in part upon.‘The Audit Society’ forthcoming inAnthony Hopwood and Peter Miller(Eds) Accounting as Social andInstitutional Practice, CambridgeUniversity Press, Cambridge,1994.

3 See Carol Jones. ‘Auditing CriminalJustice’ British Journal ofCriminology Vol 33. No.2, Spring1993, pp.187-202.

4 See. Working for patients, HMSOLondon, 1989.

5 The HEFC has advertised for anumber of posts in early 1994,including quality auditors on a part-time basis. The job description isinteresting because managementexperience in relation to educationis preferred but is not regarded asessential.

6 See ‘UK prospects in the boomingglobal environmental market’ ENDSReport 212 (September 1992).

7 See ‘Environmental consultants rideout recession’ ENDS Report 213(October 1992).

8 See ‘NACCB to be accreditationbody for BS7750 and EC Eco-Audit’ENDS Report (November 1993)pp.36-7.

9 See. University Statistics 1991-92,HMSO. 1993, pp.64-5.

10 For evidence of the burden on smallcompanies, see ‘Unloved expense’The Financial Times February 21993.

11 Of nearly 13,5000 charitiesregistered with the CharityCommission, only 11% submittedaccounts to them in the early 1990s.See ‘Act of charity brings audit dayof reckoning’ The Financial Times19 November 1992.

12 A new journal entitledAccountability in Research: Policiesand Quality Assurance, waslaunched in 1989.

13 See, for example the white paper.Realising our potential: A strategy forscience, engineering and technology,HMSO. London 1993. The newpublic management of science is

Demos 41

Notes


discussed in Brad Sherman.‘Governing Science: Patents andPublic Sector Research’, Science inContext, forthcoming. Vol 7, no.3,1994.

14 In this sense I am drawn to theclaim that ‘The canon of exhaustionof evidence is a peculiar one: itseems tied to an increasingminiaturization of focus, so that themore we “know” about a subject, themore details we know.Anaesthetization of the intellect isthe inevitable product of this formof proof. Richard Sennett. The Fallof Public Man, Faber and Faber,London 1986, p.43.

15 See Adrian Gain and JonathanRosenhead, ‘Problem Structuringfor Medical Quality Assurance’ LSEWorking Papers in OperationalResearch (November 1993). Thisempirical study reports on variousways to generate consensus andshared objectives around the auditprocess. However, in the design ofaudit arrangements the tensionsbetween clinical and managerialresource-based judgements provedthe most intractable, with clinicianshaving considerable anxiety aboutthe coopting of audit processes fordisciplinary purposes.

16 For an exploration of this see NilsBrunsson, ‘Ideas and Action:Justification and Hypocrisy asAlternatives to Control’, Accounting,Organizations and Society (1993)Vol 18, No.6, pp.489-506

17 These comments draw heavily fromStanley Cohen, Visions of SocialControl, Polity Press, Cambridge1985, pp.155-160.

18 It would be easy to regard the auditexplosion as extending

rationalisation in Weber’s sense.Certainly it creates its ownbureaucratic machinery and can beregarded as a form ofrationalisation. But itsinstrumentality is problematic andoften obscure. One therefore has apuzzle which escapes Weber’sframework: widespread investmentin a practice with ambiguousfunctional credentials.

19 See Ulrich Beck, Risk Society Sage,London, 1992.

20 See ‘The Special IntelligenceAgency’ The Times HigherSeptember 24 1993 pp.16-17. Thisarticle contains a cartoon whichepitomises the essential structure ofthe audit society. It depicts a largequality assurance inspectorchecking over the shoulder ofanother smaller official who is inturn checking the work of anothereven smaller individual whoseidentity as lecturer or student is notmade clear. The quality assuranceinspector is remote from the firstorder activity. His or her role is thatof control of control.

21 Critics, such as the Labour PartyMP Austen Mitchell, regardindependence as the underlyingproblem. They argue that financialauditors are too close to theinterests of management whocontrol their remuneration.Accordingly the objectivity ofauditors is systematically impairedand can only be improved byinstitutional change. These criticssuggest that a new body, such as ageneral audit council, could providethe necessary oversight andeffective discipline. Others see theproblem in terms of the

42 Demos

The audit explosion


expectations which audit createsamong consumers and the ‘gap’between them and auditors’ ownconception of their mission. Initialresponses to this problem have beenconceived in terms of educatingconsumers of audit services to havethe ‘correct’ expectations. Now thereis an, albeit cautious, mood for adifferent response-to bring theaudit product into line with theseexpectations (See ChristopherHumphrey, Peter Moizer and Stuart Turley, The AuditExpectations Gap in the UnitedKingdom Institute of CharteredAccountants in England and Wales,London 1991). While practitionerscan be precise about the costs ofaudit, its benefits in terms ofproviding assurance to other parties about the quality offinancial statements haveconsistently defied precision,despite attempts to integratestatistical techniques into auditprogrammes.

22 This crisis has different faces.Firstly, financial audit practitionersare facing a mature marketexperiencing competition with avengeance. Secondly, they are alsofacing a growth in litigation (not yetevident in other fields of auditing)which is increasing the risk theyface (See ‘Auditors turn cold as legalclaims hot up’ The financial TimesOctober 11 1993). Thirdly, they areexperiencing an erosion ofreputation as the consuming publicbegins to doubt the value of thisstatutory product. Under thesecircumstances it is to be expectedthat audit firms would seek todiversify into markets for other

services, including audits in otherareas.

23 I would not necessarily concedethat ‘democracy’ and‘communications audits’ areexceptions to my claim since themechanics of audit, even if designedwith style B in mind, has thepotential to drift towards style A.See David Beetham, AuditingDemocracy in Britain, Charter 88Trust Publications, London 1992; A.Booth, Communications Audit: AGuide for Managers, Gower, London1988.

24 See ‘Auditing the Accountants’ ThePolitical Quarterly Vol.64. No.3,1993, p.270.

25 David Flint, Philosophy andPrinciples of Auditing, MacmillanEducation, London 1988. Flintargues that the ‘primary conditionfor an audit is that there is arelationship of accountability or asituation of public accountability’(p.23). On this view accountabilityrelations are logically prior to audits.

26 See Charles Perrow,‘EconomicTheories of Organization’ in SharonZukin and Paul DiMaggio (eds)Structures of Capital: The SocialOrganization of the Economy,Cambridge University Press,Cambridge 1990, p.123. Perrow isgenerally critical of the abstractionfrom institutional context whichsustains this form of analysis. Healso argues that the workingassumptions of agency theoryreflect a capitalist form ofproduction which ‘started becausefour workers could not trust eachother’ (p.123). Indeed, one of thedangers of agency theory forPerrow is that people increasingly

Demos 43

Notes


think in its terms. Perrow alsosuggests that the monitor becomesdominant in relation to the fourworkers. This suggests that theperformance of those who aremonitored must be registered insuch a way as to make monitoringpossible. I develop this themefurther in section 7 of theintroduction.

27 See Dick Edwards, A History ofFinancial Accounting, London:Routledge, 1989, chapter 12.

28 Despite the fact that traditionalhistories of financial audit describeits emergence in terms of theseparation of ownership(principals) and control (agents),the demands of capital markets andthe varied forms of regulating theenterprise, its precise role has neverbeen uncontested. Questions ofauditor liability and shifting publicexpectations, usually in the light ofperceived audit failure, have beenmore or less continuous pressuresfor modification and change. In theUK, there has recently beenrenewed concern with the mannerin which financial audit servesshareholders who are remote fromthe audit process. The CadburyReport on Corporate Governancecan be interpreted in part as anattempt to revitalise theaccountability relationship uponwhich financial audit depends bydevising mechanisms to makeshareholders internally visible. Inthis way, the category of ‘principal’must be constantly reinforced if it isnot to become merely formal.

29 See Michael Jensen and WilliamMeckling, ‘Theory of the Firm:Managerial Behaviour, Agency

Costs and Ownership Structure’Journal of Financial Economics,1976, pp.305-60. This famous essayhas influenced not only generationsof economic theorists but has alsofiltered into organisationalthinking. Economics has thisdistinctive property that its theoriescan become true by virtue of theworld changing to conform to thetheory. Assumptions which areinitially abstract from the worldoften become the basis for changingit.

30 See Coopers & Lybrand, ExecutiveBriefing: Corporate Governance andthe Role of the Auditor, June 1992,for evidence that financial auditorsare hesitant about this extension oftheir role and responsibilities. Seealso ‘Pieces of Cadbury still awaitconsumption’ The Financial TimesFebruary 3 1994.

31 See ‘The parts assessors can’t reach’The Times Higher EducationalSupplement, February 4 1994.

32 I have argued that this isparticularly the case in emergingcontexts of auditing. See MichaelPower, ‘Auditing and EnvironmentalExpertise: Between Protest andProfessionalisation’, Accounting,Auditing and Accountability Journal,1991, Vol.4, No.3, pp.37–62. In thecontext of environmental auditpractices, the audit is developingprior to the institutionalisation ofan accountability framework. Auditis shaping this framework indistinctive ways by, for example,emphasizing the role of audit as a‘management tool’. In this way thenegotiability of the category ofprincipal is preempted andnormalized. Despite claims that the

44 Demos

The audit explosion


relevant ‘principals’ arecommunities, future generations oreven nature as such, environmentalaudit has emerged as a practice witha strong managerial flavour. Thiscontrasts with earlier forms of‘social’ audit which sought tocontest and protest againstcorporate environmental effects.

33 See Peter Armstrong, ‘Contradictionand Social Dynamics in theCapitalist Agency Relationship’Accounting, Organizations andSociety, 1991, pp.1-25. Like Perrow,Armstrong suggests that the‘visualisation of organisations assystems of contracts’ is central tothe cogency of agency theories oforganisation. However, contractsnever substitute entirely for trustbut only displace it. Hence thereliance on monitors such asauditors to restore trust is apotential regress. Armstrong arguesthat this displacement of trust isprimarily responsible for the rise ofthe monitor at the expense ofoperational management in BritishIndustry. Once set in motion, ‘Anaccounting managerial culturecreates both a supply and a demandfor additional accounting functionswhich therefore tend toexpand...’(p.19).

34 The Auditing Practices Board iscurrently revising the structure ofauditing standards. These threeareas will be covered by three newStatements of Audit Standard(SAS): 300 ‘Audit RiskAssessment’:440 ‘ManagementRepresentations’:520 ‘Using thework of an expert’.

35 See Christopher Hood, ‘A PublicManagement for All Seasons’, Public

Administration, 1991, pp.3-19.These changes cannot simply belaid at the door of Thatcherismsince they have parallels throughoutdeveloped economies. Certainlythere are supply side peculiarities inthe UK and the role of the largeaccountancy firms is one factor. AsChristopher Lasch suggests,professions invent many of theneeds which they claim to satisfy byplaying upon public fears and bycreating structures of dependency.See The Culture of Narcis ism,Warner Books, New York, 1979p.385.

36 See D. Osborne & T. GaeblerReinventing Government, AddisonWesley, Reading, Massachusetts1992.

37 See Theo Mars, ‘Public SectorOrganization: Where Next?’ IDSBulletin (1992) vol. 23, No.4, pp.18-30.

38 The necessity of such areconciliation is a peculiarity of‘liberal’ forms of government whichmust somehow intervene with ‘non-interventionist’ instruments.Accounting and audit appeal toliberal ideologies because, asapparently apolitical practices, theycome to be recognised as necessaryby social agents. See Nik Rose andPeter Miller ‘Political Power beyondthe State: Problematics ofGovernment’ British Journal ofSociology (1992, pp.173-205.)

39 I would exclude tax inspectionsystems from the audit explosiondespite the existence of PAYE andVAT audits. These practices have alonger history than the recentchanges which I discuss. Taxinspection is primarily a form of

Demos 45

Notes


direct monitoring for compliancewith (the relatively unambiguous)rules which it serves. While it iscertainly a disciplinary technology(See Alistair Preston. ‘The TaxmanCometh: Some Observations on theInterrelationship betweenAccounting and Inland RevenuePractice’. Accounting, Organizationsand Society, 1989, pp.389-413), ithas its own history. Furthermore,Inland Revenue practices havethemselves become subject to theeffects of the audit explosion in thepublic sector as inspection practicesare required to demonstrate their‘value for money’.

40 The tendency of monitoringpractices to shift towards a moreindirect ‘control of control’ isinternally paradoxical. It expressesboth a recognition of the economicand epistemic impossibility ofdirect observational control at thesame time as it reaffirms theappearance of control via themaintenance of systems integrity.This paradox is overcome by anepistemic shift in the concept ofauditability. As first-order activitiesbecome increasingly unauditablewithin the existing portfolio oftechniques, the audit shifts its focustowards second-order ‘systems ofcontrol’ which necessarily focusupon process.

41 This development of audit awayfrom real-time forms of inspectionconforms to certain postmodernmotifs regarding ‘loss of reference’.The primary reference point forauditing has become the evidentialtraces inscribed by systems ofcontrol which are increasinglyformal in their structure and

operation. The focus on systemsmeans that audits may have less todo with control in a traditionaldirect sense and more to do withformalising the allocation ofresponsibility for systems integrityand process. For more on theconstruction of responsibility seeNils Brunsson. ‘Deciding forResponsibility and Legitimation:Alternative Interpretations ofOrganizational Decision Making’,Accounting, Organizations andSociety 1990, pp.47-59, and PeterMiller. ‘Accounting and Objectivity:The Invention of Calculable Selvesand Calculable Spaces’. Anuals ofScholarship, 1992, pp.61-86.

42 In ‘The Politics of FinancialAuditing’, The Political Quarterly,1993. Vol 63. No.3. pp.272-284. Iargue that responses to apparentfailures in financial auditing are asmuch about avoiding inquiry intothe technical efficacy of audit ingeneral as they are about findingparties to blame. The institutionalrole of audit is thereby preserved.

43 See John Meyer and Brian Rowan,‘Institutionalised Organizations;Formal Structure as Myth andCeremony’, American Journal ofSociology, 1977, pp.340-363.

44 Patricia Day & Rudolf Klein,Accountabilities: Five Public Services,Tavistock, London 1987. p.174.

45 See Rose and Miller op cit.46 See Inquiry into the Supervision of

the Bank of Credit and CommerceInternational, HMSO, London,1992.

47 See Christopher Humphrey et. al.op cit.

48 Any practice must have ‘enoughlogic for the needs of practical

46 Demos

The audit explosion


behaviour, neither too much-since acertain vagueness is oftenindispensable, especially innegotiations-nor too little, since lifewould then become impossible’,Pierre Bourdieu, ‘From Rules toStrategies’, in Other Words: Essaystowards a Reflexive Sociologytranslated by Matthew Adamson,Polity Press, Cambridge 1990, p.73.Andrew Abbot makes a similarpoint when he remarks thatprofessional knowledge must be‘abstract enough to survive smallmarket shifts but not so abstract asto prevent monopoly’ See TheSystem of Professions: An Essay onthe Division of Expert LabourUniversity of Chicago Press,Chicago 1988, p.324.

49 International chamber ofCommerce, Effective EnvironmentalAuditing, ICC, Paris, 1991.

50 The recent European Communityregulation relating toenvironmental managementsystems and audit distinguishesbetween ‘environmental audit’ as aninternal affair and ‘externalverification’ as a more independentform of scrutiny to establish thatthe conditions for registration aresatisfied. The latter is vital to thecredibility of the scheme and thereare lessons to be learned from thequality assurance initiatives whichprovide the general model forenvironmental management.See ‘Need a quality certificate? Ask Tom, Dick or Harry’The Financial Times March 8 1993; ‘Standards Institute Defends Quality ManagementBenchmark’ The Financial TimesJuly 14 1993.

51 For example, recent concerns in theUK with the wording of financialaudit reports, the only aspect ofaudit work which is publicly visibleas a truncated and coded report,have been conducted in the name ofimproving communication. SeeAuditing Practices Board, Proposalsfor an Expanded Audit Report, APB,London, October 1991. Theproblem is particularly pertinent inthe context of the auditor’s opinionon whether an enterprise is a goingconcern. See Auditing PracticesBoard, SAS 130 Exposure Draft-theGoing Concern basis in FinancialStatements, London, APB,December 1993.

52 See Auditing Practices Board. TheFuture Development of Auditing:A Paper to Promote Debate, APB,London, November 1992; CadburyCommittee, The Financial Aspects ofCorporate Governance, London1992. Both these documents can beread as attempts to rethink audit inorder to reconnect distantstakeholders to the inner workingsof organisations. Audit is beingshaped as the technology ofcorporate governance. However, ithas been argued that thesedocuments nevertheless reflect anaccounting culture. See JudithFreedman, ‘Accountants andCorporate Governance: Filling aLegal Vacuum’ The PoliticalQuarterly, 1993, pp.285-97.

53 Day and Klein op cit p.249.54 These phrases are borrowed from

Jurgen Habermas, ‘The NewObscurity: The Crisis of the WelfareState and the Exhaustion of UtopianEnergies’, in Jurgen Habermas,The New Conservatism: Cultural

Demos 47

Notes


Criticism and the Historian’s Debateedited and translated by ShierryWeber Nicholsen, MIT Press,Cambridge MA, 1992, pp.48-70, andGianni Vattimo. The TransparentSociety, Polity Press, Cambridge,1992. Habermas’s diagnosis of thepresent impasse of the welfare stateis instructive. Ideals of solidarityhave given way to the regulatorylogic of money and administrativepower. In this sense the rise of auditis indicative of the exhaustion ofutopian energies. Audits ensureaccountability to individuals as‘clients’ rather than citizens and it isno accident that the audit explosionhas accompanied the displacementof older languages with that ofmarkets, missions and management.These abstract concepts aremutually reinforcing.

55 See ‘Tidal wave of dirty water’The Financial Times February 51994 which examines fallingstandards in the UK public sector.

56 In the ‘Politics of FinancialAuditing’ I have formalised theproblem of how the success of auditcan be conceived. The problem isthat it is impossible to disentanglesuccessful auditing from successfulcompany management andfinancial reporting. In the same wayit is impossible to distinguishsharply between successful policingand other reasons for changingpatterns of crime. Hence there is acertain asymmetry which underliesthe politics of failure. Corporatefailure will often give rise to publicquestions about financial auditfailure. However, corporate successwill not usually yield public

approbation for the role of audit. Itis in this sense that, as a matter ofpublic perception, audits always fail.Only greater publicity about theprocess of audit could overcomethis asymnetry.

57 An exception would be the financialsavings claimed on behalf of valuefor money audits, althoughquestions could always be raisedabout whether these were adequate.AC258 See Michael Power, ‘Auditingand the Politics of Control in theUK Financial Services Sector’ in J.McCahery, S. Picciotto and C. Scott(Eds) Corporate Control andAccountability, Clarendon Press,Oxford 1993, pp.187-202.

59 See Auditing Practices Board,Proposed Statement of AuditingStandards-Exposure Draft 620 TheAuditors Right and Duty to Reportto Regulators in the FinancialServices Sector, APB. London July1993.

60 See also Ulrich Beck, ‘FromIndustrial Society to Risk Society:Questions of Survival. SocialStructure and EcologicalEnlightenment’, Theory, Culture andSociety 1992, pp.97-123.

61 Early characterisations of the auditprocess consisted in the accountableparty defending his actions inperson to a relevant audience. Theauditee was often visible throughoutthis process of defence but it wasthe aural as much as theobservational intimacy which wasthe relevant parameter. Derivedfrom the latin ‘audire’, early auditstook the form of ‘hearings’ betweenauditor and auditee (See PeterWolnizer, Auditing as IndependentAuthentication. Sydney University

48 Demos

The audit explosion


Press, Sydney, 1987 pp.35-39).Today, as modern financial auditpractice has largely disengaged itselffrom these quasi-judicial origins,other bodies of knowledge havebeen invoked to shape the practice.Audit has emerged as a ‘proto-scientific’ have not entirely diedaway. Thus, while law hascontributed the institutional contextwithin which the demand forfinancial audits is shaped, theknowledge base of audit and theclaims to expertise of itspractitioners have becomedisengaged from their origins as aform of ‘hearing’ (although vestigesof these origins are to be found inthe European Court of Auditors).Audit has emerged from the law toclaim its own expertise and itsattestation functions have come todevelop an institutional momentumof their own.

62 Brendan McSweeney, ‘Accountingfor the Audit Commission’. PoliticalQuarterly Spring 1988 pp.28-43;also Day and Klein op cit.

63 See Carol Jones, ‘Auditing criminaljustice’ op cit.

64 It must be said that VFMpractitioners are aware of this issue-it does not happen behind theirbacks. See Peter Kimmance. ‘TheWidening Scope of LocalGovernment Audit and Private SectorParticipation’ in A. Hopwood and C.Tomkins, Issues in Public SectorAccounting Phillip Allen, 1984,Oxford, pp.243-4.

65 See Ted Porter, ‘Quantification andthe Accounting Ideal in Science’Social Studies of Science (1992) Vol.22, pp.633-52. Porter argues that the

objectivity of accounting does notlie in its capability for accuraterepresentation so much as its abilityto provide a public and impartialmedium.

66 Numerous articles have beenpublished on this subjectthroughout 1992 and 1993 in theTimes Higher Education Supplement.

67 For example, see ‘Poor research:ranking blamed’ The Times HigherEducational Supplement December4 1993.

68 This has become embodied in thestudents’ charter in which studentsare reconceived as having consumerrights. Such a programme dependscrucially upon audit.

69 The UK Higher Education FundingCouncil has been exploringmethods of making academicresearch expenditure moreaccountable and, in turn, auditable.Time sheets for academics havebeen proposed as a basis of makingacademic research auditable (SeeHEFC, Accountability for ResearchFunds, March 1993). In addition,pilot teaching quality audits havebeen conducted One academicdepartment was judged to be‘satisfactory’ rather than ‘excellent’largely because claims for excellencewere not supported by sufficientevidence. This is indicative of theaudit explosion and the shift fromthe substance of quality to theprocess of quality assurance hasgenerated complaint: ‘VCs rejectquality red tape’ The Times Higher Educational SupplementJanuary 22 1993; ‘Concern atpointless quality rules’ The TimesHigher Educational SupplementApril 9 1993.

Demos 49

Notes


70 In this respect audit is consistentwith Poster’s concept of the‘superpanoption’ which constitutesits own system of self-referencewhile exerting a disciplining effecton the subject. Poster seems toattribute a disciplinary power todata bases as such whereas I wouldwish to implicate them within thespecific practices of audit andsurveillance which they makepossible. See Mark Poster, The Modeof information, Polity Press,Cambridge, 1990, chapter 3.

71 Audit may be a narcissistic practicewhich feeds off its ownrepresentations. It is a simulacrumin Baudrillard’s sense. See JeanBaudrillard, Simulations, translatedby Paul Foss et. al., Semiotext, NewYork, 1983.

72 See note 31 above.73 See CEC DG XI/UK Department of

the Environment, Research into theDevelopment of Codes of Practice forAccredited Environmental Verifierswithin the Framework of theProposed Eco-management andAudit Regulation. Judge Institute ofManagement Studies, CambridgeJune 1993.

74 This image of audit knowledgewould be disputed by financialaudit practitioners and there ismuch discussion on this point.Some argue that checklists imposetoo much structure on the auditprocess and thereby inhibitprofessional judgement. However,practitioners will generally resistcomplete codification of theirexpertise in order to preserve itselite aura. Hence, debates aboutstructure and judgement in theaudit process have less of a purely

technical character and have muchto do with maintaining images ofexpertise.

75 Andrew Abbott, op cit.76 Michael Power, ‘From Common

Sense to Expertise: Reflections onthe Pre-history of Audit Sampling’,Accounting, Organizations andSociety, 1992 pp.37-62; BrianCarpenter and Mark Dirsmith,‘Sampling and the Abstraction ofKnowledge in the AuditingProfession: An ExtendedInstitutional Theory Perspective’,Accounting, Organizations andSociety 1993 pp.41-63.

77 See Stella Fearnley and MichaelPage ‘Audit Regulation. One YearOn’ Accountancy, January 1993 p.59.They observe that auditpractitioners subjected to audit inthis way have responded bystandardising their ‘official’ auditapproach in order to demonstratetechnical competence. By doing sothey tended to ‘equate better fileswith better audits’. This is aconsequence of the norm ofauditability where even the ‘audit ofaudit’ requires that it beproceduralised. Results of the mostrecent audit of the auditors wereavailable in early 1994 and can beinterpreted both as failure andsuccess. Failure because there seemsto be a problem of audit quality forsmaller practices, success becausethe critical stance of the monitoringagencies demonstrates their ownindependence.

78 It is important to emphasise thatthere is nothing conspiratorialabout this. It is not, directly at least,a question of cartels andprofessional interests. Audit, as the

50 Demos

The audit explosion


‘self observation of the economicsystem’ (Niklas Luhmann, EcologicalCommunication translated by J.Bednarz Polity Press, Cambridge,1989) is a system of knowledge inits own right. It functions by virtueof an auditing logic which demandsauditable auditees as the conditionof possibility of its own functioning.The auditing system of knowledge‘productively misunderstands’(Gunther Teubner, The Two Facesof Janus: Rethinking legal PluralismCardozo Law Review March 1992,Vol 13. No 5. pp. 1443-62) theauditee in order to make itauditable. In turn, auditees arepotentially alien disturbances to thissystem which reacts by renderingthem familiar and auditable. But, ifauditing is an autopoietic, self-sustaining practice it is itself thedisturbing environment for auditeeswho may adapt their practices inthe name of ideals of verifiability,calculability and responsible control(Peter Miller and Ted O’Leary,Accounting and the Constructionof the Governable Person.Accounting, Organizations andSociety 1987, pp.235-66). The keyempirical question is how great isthis submission to the audit process.

79 See Anthony Hopwood, ‘Accountingand the Pursuit of Efficiency’, inAnthony Hopwood and CyrilTomkins, Issues in Public SectorAccounting Phillip Allan, Oxford1984. pp. 167-187. The overall effectis the displacement of non-accounting skills in public sectormanagement.

80 See The Henley Centre Planning forSocial Change Survey 1992.

81 See ‘City warned of big increase instaff turnover’ The Financial TimesFebruary 11 1994.

82 See William McInnes (ed) Auditinginto the Twenty First Century,Institute of Chartered Accountantsof Scotland, Edinburgh 1993. Thisdocument argues for a shift inemphasis from external to internalforms of audit. The danger in theseinnovative proposals is that if theinternal audit function has beenconstructed as a mirror image ofthe external audit then theproblems raised in this essay will bedisplaced but not solved.

83 This has always been the complaintby small companies. It is alsobecoming more evident in highereducation. See, for example.‘Universities balk at review teamcosts’ The Times Higher EducationalSupplement September 17 1993.

84 See ‘Council rejects quality advice’The Times Higher EducationalSupplement February 18 1994. TheHigher Education Quality Councilhas resisted advice frommanagement consultants that itsaudit reports should be moreexplicitly critical. The consultants,Coopers & Lybrand, also suggest a move towards ‘licensed auditors’ with system-wideresponsibilities analogous tofinancial auditors.

85 This is ironic in the light of post-Metallgesellschaft deliberationsabout the merits of UK auditcommittee arrangements over theGerman supervisory boardstructure. When the Cadburyproposals were first mooted in theUK, the German system wasregarded by many as a preferable

Demos 51

Notes


model. See ‘Metallgesellschaft board“not at fault”’ The Financial TimesFebruary 8 1994.

86 An interesting possibility isprovided by Public Concern atWork, an independent centreresourced by the Joseph RowntreeCharitable Trust. Its aim is tofacilitate structures which enableemployees to register concernsabout the conduct and standards ofbusiness.

87 Similar reservations about suchcharters have been expressed in‘Auditing the Accountants’ ThePolitical Quarterly Vol 64, No 3.p.270.

88 In the context of the United States,Johnson and Kaplan have argued

that auditing has seriously inhibited the evolution ofmanagement accounting systemswith damaging effects oncompetitiveness. See ThomasJohnson and Robert Kaplan.Relevance Lost-The Rise and Fall of Management Accounting.Harvard Business School,Harvard, 1987.

89 See Martin Trow,‘The Business ofLearning’ The Times HigherEducational Supplement October 81993.

90 See The Royal Society for theEncouragement of Arts.Manufactures & Commerce,Tomorrow’s Company, RSA. London1994.

52 Demos

The audit explosion


I am very grateful to the following people for commenting upon ear-lier drafts of this paper: Andrew Barry, Nils Brunsson, AnthonyHopwood, John Law, John Meyer, Peter Miller, Brendan McSweeney,Geoff Mulgan, Christopher Napier, Keith Robson, Nikolas Rose, BradSherman, Claire Sinnott, Christopher Swinson, Lucy Zedner. Manyothers have provided invaluable comments at seminars and work-shops.

Demos 53

Acknowledgements


Documents

The Audit ExplosionThe word ‘audit’ is being used in the UK with growing frequency. In addition to financial audits, there are now environmental audits, value for money audits,management