Communication, Corporate Annual Reports And Perception Engineering 9781845446093, 9780861767342

Citation preview

Accounting, Auditing & Accountability Journal

ISSN 0951-3574 Volume 15 Number 4 2002

Communication, corporate annual reports and perception engineering Guest Editor John K. Courtis Paper format Accounting, Auditing & Accountability Journal includes five issues in traditional paper format. The contents of this issue are detailed below.

Internet Online Publishing with Archive, Active Reference Linking, Emerald WIRE, Key Readings, Non-article Content, Institution-wide Licence, E-mail Alerting Service, and Usage Statistics. Access via the Emerald Web site: http://www.emeraldinsight.com/ft See p. 439 for full details of subscriber entitlements.

Access to Accounting, Auditing & Accountability Journal online ____________________________________ 439 Editorial advisory board ___________________________ 440 Abstracts and keywords ___________________________ 441 Preface John K. Courtis _________________________________________________

444

Economic man and disciplinary boundaries: a case study in corporate annual reports Robert White and Dallas Hanson___________________________________

450

Corporate annual reports: research perspectives used Patricia Stanton and John Stanton _________________________________

478

The information gap in annual reports Jill Hooks, David Coy and Howard Davey ____________________________

501

Developments in content analysis: a transitivity index and DICTION scores Robin Sydserff and Pauline Weetman _______________________________

523

Measurement distortion of graphs in corporate reports: an experimental study Vivien Beattie and Michael John Jones ______________________________

546

This issue is part of a comprehensive multiple access information service

CONTENTS

CONTENTS continued

Colour graphics and task complexity in multivariate decision making Stella So and Malcolm Smith ______________________________________

565

Communication and antithesis in corporate annual reports: a research note Jane Davison ___________________________________________________

594

Signaling gender diversity through annual report pictures: a research note on image management Richard A. Bernardi, David F. Bean and Kristen M. Weippert ___________

609

Accounting, Auditing & Accountability Journal online

AAAJ online

An advanced information resource for the entire organization Access via the Emerald Web site – http://www.emeraldinsight.com/ft Subscribers to this journal benefit from access to a fully searchable information resource stretching far beyond the current volume and issue. Accounting, Auditing & Accountability online is enhanced with a wealth of features to meet the need for fast, effortless, and instant access to the core body of knowledge. Furthermore, this user friendly electronic library may be networked throughout the subscribing organization to maximize the use and value of the subscription. This is augmented with advanced search facilities and ‘‘choice of access’’ through a variety of alternative journal gateways. Emerald online access includes: Institution-wide Licence Our liberal licence allows everyone within your institution to access the journals via the Internet, making your subscription more cost-effective. It has been designed to provide a comprehensive, simple system with, minimum adminstration. Access can be granted by IP range or username and password. ATHENS authentication is enabled.

Current Issue + Journal Archive Internet access via Emerald Fulltext to information as it becomes available online and to material from past volumes. Users can choose to browse the database for relevant articles or search using the powerful search engine provided.

Reference Linking Link directly from selected journal references cited by authors to abstracts of these papers. Where possible, access to the full text of references will be provided.

Active Reference Linking Direct links from the journal article references to abstracts of the most influential articles cited. Where possible, this link is to the full text of the article.

Emerald WIRE A fully searchable subject specific database, brought to you by Emerald Management Reviews, providing article abstracts from the world’s top management, computing and engineering journals. Directly relating to the journal subject area, the database is updated monthly and gives users details of how to obtain the full text of original articles.

Usage Statistics Online Journal Usage Statistics are now available. This feature allows Emerald Administrators to download their usage statistics with regard to their organization’s journal usage. Usage Statistics allow you to review the value of electronic disssemination of your journal subscriptions throughout your organization. They can also help determine the future trends for information within your organization. For further information go to: http://www. emeraldinsight.com/stats

Non-article Content The non-article content of our journals (product and materials information, industry trends, company news, conferences etc.) is available online, so that it can be accessed by users simultaneously. It is possible to browse or search by keyword, allowing users to keep up to date with everything that is happening in their specific subject areas.

Emerald Alert The table of contents e-mail alerting service will e-mail you the contents page of any chosen journal whenever the latest issue becomes available online. For further information please go to: http://www. emeraldinsight.com/alerts

Support Resources A comprehensive range of resources are available online that help users learn how to effectively use online information resources and that help information professionals market resources to their users. For further information please go to: http://emeraldinsight.com/support

Choice of Access Electronic access to this journal is available via a number of channels, enabling end users’ libraries to reach the content through their preferred delivery system. The Emerald Web site – http://emeraldinsight.com/ft – is the recommended means of electronic access as it provides fully searchable and high value-added access to the complete content of the journals. Refer to the next section for ‘‘how to access’’ via the Emerald Web site. Subscribers can also access and search the article content of this journal through the following journal gateway services: Ebsco Online http://www.ebsco.com/online Huber E-Journals http://e-journals.hanshuber.com/english Minerva http://www.minerva.at/ OCLC Firstsearch ‘‘Electronic Collections Online’’ http://www.uk.oclc.org/oclc/menu/eco.htm RoweCom’s ‘‘Information Quest’’ http://www.informationquest.com SilverPlatter http://www.silverplatter.com SwetsBlackwell’s ‘‘SwetsNetNavigator’’ http://www.swetsnetnavigator.nl

How to access this journal through Emerald Organizations must first register for online access (instructions provided at http://www.emeraldinsight.com/ register), after which the content is available to everyone within the organization’s domain. To access this journal’s content, simply log on either from the journal homepage or direct through the Emerald Web site.

Key Readings Abstracts of articles, relating to keywords chosen by the journal editor, are selected to provide readers with current awareness of interesting articles from other publications in the field. The abstracts are available online and are updated regularly upon publication of the latest issue.

Emerald Customer Support Services For customer service and technical help, contact: E-mail: [email protected] Telephone: (44) 1274 785278 Fax: (44) 1274 785204

439

AAAJ 15,4

440

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, p. 440. # MCB UP Limited, 0951-3574

EDITORIAL ADVISORY BOARD Margaret Abernethy University of Melbourne, Australia Carol Adams Monash University, Gippsland Campus, Australia Shahid Ansari California State University – Northridge, USA Ed Arrington University of North Carolina at Greensboro, USA Jan Bebbington University of Aberdeen, Scotland Niamh Brennan University College Dublin, Ireland Jane Broadbent Royal Holloway, University of London, UK Garry Carnegie Deakin University, Australia Chung Lai Hong Nanyang Technological University, Singapore David Cooper University of Alberta, Canada Craig Deegan RMIT University, Australia Mahmoud Ezzamel Cardiff University, UK Richard Fleischman John Carroll University, USA Tim Fogarty Case Western Reserve University, USA Warwick Funnell University of Wollongong, Australia Sonja Gallhofer Glasgow Caledonian University, Scotland Robert Gray University of Glasgow, Scotland Theresa Hammond Boston College, USA Jim Haslam Heriot Watt University, Scotland Trevor Hopper University of Manchester, UK Keith Hoskin University of Warwick, UK Christopher Humphrey University of Manchester, UK Linda Kirkham University of Manchester, UK

Katsuhiko Kokubu Kobe University, Japan Irvine Lapsley University of Edinburgh, Scotland Stewart Lawrence University of Waikato, New Zealand Tom Lee Edinburgh, Scotland Cheryl Lehman Hofstra University, USA Sue Llewellyn University of Edinburgh, Scotland Reg Mathews Charles Sturt University, Australia Kenneth Merchant University of Southern California, USA Markus Milne University of Otago, New Zealand Tom Mouck University of New Mexico, USA Christopher Napier University of Southampton, UK David Otley Lancaster University, UK David Owen Sheffield University Management School, UK June Pallot University of Canterbury, New Zealand Chris Poullaos University of Wollongong, Australia Vaughan Radcliffe Case Western Reserve University, USA Bob Scapens University of Manchester, UK Prem Sikka University of Essex, UK Tony Tinker City University of New York, USA Ken Trotman University of New South Wales, Australia Stuart Turley University of Manchester, UK Thomas Tyson St. John Fisher College, USA Stephen Walker University of Edinburgh, Scotland Paul Williams North Carolina State University, USA

Economic man and disciplinary boundaries: a case study in corporate annual reports Robert White and Dallas Hanson Keywords Company reports, Knowledge workers, Language, Economic systems, Communications This paper is an empirical response to two of Quattrone’s claims: first, that research in accounting is fragmented; and then that this follows from the blocking of communication by intra- and inter-disciplinary boundaries. Although we agree with much of Quattrone’s argument, and in particular with his problematising of ‘‘economic man’’, we draw an opposite conclusion. Rather than looking to a trans-disciplinary removal of boundaries, we use a survey of 30 years of research in corporate annual reports to defend narrowly disciplinary work. We make our case through discussing problems of intra- and interdisciplinary unity in research, the puzzle of the role of ‘‘economic man’’ in the study of annual reports, and the alternative to him in science and technology studies (STS). Our approach yields a better fit than Quattrone’s own solution with his aims of an evolutionary perspective that allows for historical shifts, and for a reflexivity that includes the inevitable entanglement of researchers in what they study. We conclude by noting that our approach is applicable to the study of corporate communication more generally.

Corporate annual reports: research perspectives used Patricia Stanton and John Stanton Keywords Company reports, Corporate image, Accountability, Marketing, Research Corporate annual reports are viewed through the lens of researchers of these documents. The aims are to obtain insight into how researchers view annual reports; to ascertain how the different ways of seeing the annual

report relate to each other; and to draw out the gaps in this diverse research in a continuing attempt to understand its role and purpose. Selective examination of a decade of corporate annual report research (1990-2000) reveals how researchers have sought to find visibility and meaning. Few studies address the document as a whole, in terms of the integration of the messages between the various parts of the report. Explanation of the changing structure and content of annual reports remains divided, largely because of the differing perspectives of researchers. They have revealed diversity in the ways of seeing the annual report and a tension in understanding its overall purpose and role.

The information gap in annual reports Jill Hooks, David Coy and Howard Davey Keywords Accountability, New Zealand, Electricity industry, Disclosure, Information Following radical restructuring of the electricity industry in New Zealand since 1987, the government adopted a ‘‘lighthanded’’ regulatory regime that used marketbased methods involving competition and transparent accountability. This accountability is in part discharged through the provision of information in the corporate annual report. To assess the quality of that communication, a disclosure index was developed and applied to the annual reports of the 33 electricity retail and distribution companies which comprise the entire industry in New Zealand. The index was developed using the ideas and opinions of 15 experts representing broad stakeholder groups. This paper compares the resulting scores for the extent and quality of each index item with the level of importance of those items as stated by the panel. Many items are not adequately disclosed, resulting in an information gap between stakeholders’ expectations and the disclosures provided by the electricity companies. This paper identifies the items and the detail about them needed to close that gap.

Abstracts and keywords

441

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, Abstracts and keywords. # MCB UP Limited, 0951-3574

AAAJ 15,4

442

Developments in content analysis: a transitivity index and DICTION scores Robin Sydserff and Pauline Weetman Keywords Accounting, Company reports, Foreign languages, Computer-based training, Research This paper responds to a call in the literature for methodological and empirical studies to advance research into accounting narratives, in the light of acknowledged areas of weakness and gaps in the accounting literature and with a view to investigating impression management. A general line of critique in the accounting literature points to a need to expand both the syntactic and thematic dimensions, with a particular focus on developing objective methods of analysis that allow computer-based measurement. The paper draws on the literature of managerial business communications, supported by that of applied linguistics, in bringing to accounting research a transitivity index and the application of diction analysis. Both have the potential to extend computer-based analysis of accounting narratives, subject to careful initial re searc h design and specification. The potential for a richer empirical analysis is demonstrated through an illustrative empirical application.

Measurement distortion of graphs in corporate reports: an experimental study Vivien Beattie and Michael John Jones Keywords Company reports, Financial accounting, Graphs, Measurement Graphs in corporate annual reports are a double-edged sword. While they offer the potential for improved communication of accounting information to users, the preparers of the annual reports can easily manipulate the graphs for their own interests. For over a decade, the empirical financial graphics literature has focused on examining company reporting practices. A particular concern has been measurement distortion, which violates a fundamental principle of graph construction. Unfortunately, it is not yet known whether observed levels of measurement distortion are likely to affect users’ perceptions of financial performance.

This study uses an experimental approach to address this issue. Pairs of graphs are shown to establish the level of difference that is just noticeable to graph readers. Six levels of ‘‘distortion’’ are investigated (5 percent, 10 percent, 20 percent, 30 percent, 40 percent and 50 percent). Results indicate that if financial graphs are to avoid distorting the perceptions of users, then no measurement distortions in excess of 10 percent should be allowed. Users with lower levels of financial understanding appear to be most at risk of being misled by distorted graphs. Further research will be necessary to investigate whether this impact upon perceptions subsequently affects users’ decisions in specific contexts.

Colour graphics and task complexity in multivariate decision making Stella So and Malcolm Smith Keywords Computer graphics, Information retrieval, Task analysis, Decision making, Gender Advancements in information technology and graphics software mean that colour graphics are an increasingly important part of the communication of business operations and corporate reporting. Unfortunately, the research literature on the effects of colour graphics on decision performance is sparse, and lends only limited and qualified support to the claims often made for colour coded graphics. There has been no research in the accounting environment of the impact of nonredundant colour graphics (i.e. those not complemented by numerical or pattern support) on decision-making performance. The existing literature suggests that gender, task complexity, field dependence and time constraints will all impact on the effectiveness of the use of colour, so this paper reports the results of a laboratory experiment designed to assess the interaction effects of non-redundant colour coding in bar charts with information complexity, and with gender. A multivariate bankruptcy prediction decision is the task environment. Non-redundant coding, rather than redundant coding, is used in this paper, to force subjects to use the actual colour coding in their decisions and in order to evaluate the effects of colour coding more

fully. The results suggest that proponents of colour graphics must qualify their claims. Colour graphics improve decision making, though their impact is significant only when information complexity is low, and then for female subjects only. Communication and antithesis in corporate annual reports: a research note Jane Davison Keywords Communications, Discipline, Company reports, Narratives, Visual aids The paper aims to identify one of the communication techniques which creative designers may use in the annual review/ annual report and accounts, described by Hopwood in 1996 as a ‘‘largely unresearched document’’. It offers a new dimension to add to existing work on graphs, accounting narratives, readability and visual images. Using analytical methods from within artistic disciplines, the paper examines the use of antithesis in structure, visual material and text of Reuters 2000 Annual Review and Report and Accounts. It is suggested that the framing and communicative power of such techniques may supplement the accounting

disclosures. Further possible applications and lines of enquiry are outlined.

Signaling gender diversity through annual report pictures: a research note on image management Richard A. Bernardi, David F. Bean and Kristen M. Weippert Keywords Gender, Equal opportunities, Directors, Company reports This research examines the differences in presentation of boards of directors in annual reports. Our sample consists of 472 corporations from the Fortune 500; 130 (342) of these corporations included (did not include) pictures of their boards of directors. The proportion of female directors was 11.0 percent for firms that did not include pictures of their boards and 14.5 percent for firms that included pictures of their boards in their annual reports. The difference in the gender mix of these two groups is significant (p ¼ 0:0002). This indicates that firms with a higher percentage of women on their boards signal this fact to stockholders, investors, and other constituents by including pictures of their boards in their annual reports.

Abstracts and keywords

443

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

AAAJ 15,4

444

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 444-449. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440540

Preface About the Guest Editor John K. Courtis is a Professor in the Department of Accountancy, City University of Hong Kong, where he has been for more than a decade. He was founding editor of Asia-Pacific Journal of Accounting, which he edited during its first six years. He is on the board of three journals and acts as reviewer for several others. While based at the University of Waterloo he was a National Examiner for the Certified General Accountants Canada, a University of Malaya resource person appointed by the Canadian International Development Agency, and a member of the Canadian Institute of Chartered Accountants Study Group on Information to be Included in the Annual Report to Shareholders. The Academies’ AustraliaChina Exchange awarded him a visiting fellowship. He is a Fellow of CPA Australia and a past Fulbright holder. He has published widely on aspects of annual report disclosure, and has concentrated in recent years on readability studies that introduce an Asian component. His teaching interest lies primarily in financial reporting theory.

Emerging sensitivity of accountants to the role of communication and perception engineering Communication is an important component of how accountants spend their time. In addition to the involvement of accountants with traditional accounting, they also spend considerable time in meetings with clients and colleagues, in writing reports and letters, in interacting on the phone and by e-mails, in reading myriad documents and in supervising others. For most accountants, the communication function in one form or another heavily dominates their time. Perhaps 80 percent of the average day-to-day routine of an accountant is spent on some aspect of communication, either as a preparer or as a user of information. Written and oral, formal and informal communication with clients and colleagues is not a trivial matter. Asking questions, seeking understanding, reading material, imparting knowledge and interpreting information occurs continuously throughout an accountant’s career. Ironically, however, the tenets of effective communication receive much less emphasis during the preparatory and early stages of development of a member of the accounting profession. Indeed, up until the last decade or so, one might have been excused for thinking that osmosis or mimicry were two principal means by which accountants learned how to effectively articulate their thoughts to colleagues, clients and users of accounting information. While this may somewhat overstate the case, the reality is that accountants do not exist simply as technicians to record and summarize. Book-keepers and computer and data processing operators can record routine transactions. Accountants differentiate themselves from technicians per se by being able to analyze and interpret sophisticated commercial and interactive information and by being able to communicate the meaning of this information to others. A major responsibility of an accountant in discharging professional duties is to ensure that the intended meaning of a message sent to others is capable of being received by them with the same understanding. Acting as a mere provider of information is not sufficient. Accountants also have a responsibility to ensure that what they provide is understandable, relevant, and internally consistent within a reporting

document, that it is presented unambiguously, and that it is presented so that key points will not be missed. It is not acceptable to users that accountants simply provide information and ignore whether that information is capable of being identified, understood and utilized. Accountants need to become as sensitive to the presentation, understandability and perception-engineering potential of information as they are to the preparation of its content. There are two sources of pressure on accountants to become more involved in communication issues. The first is that external users of accounting information are increasingly busy (or perceive themselves to be so) and desire ‘‘quick fixes’’ about informational items of interest. The second is that end users have a battery of available sources from which they can draw for items of information. With regard to the first of these influences, there is a growing awareness by preparers of accounting information that external users want more qualitative disclosures and ‘‘gimmickry’’ in the presentation of information so that they can reduce their search time. In response, narrative disclosures of financial and non-financial information are on the increase. Photographs, chart graphics, animation, colour, fonts and headings, and in all kinds of combinations are but some of the techniques used to arrest the attention of users, direct them towards specific informational items and make information gathering less onerous, if not somewhat pleasant. With regard to the second of these influences, users have access to multiple sources of information. Although credibility differs between these sources, busy users will gravitate towards those sources that present information quickly and clearly, with simple headings and fonts, and most probably which combine colour and graphics. As a result of both of these pressures it is incumbent on accountants to ensure that corporate releases are capable of being understood, and that they are not misleading. To assure themselves of this, accountants will therefore inevitably need to develop a greater sensitivity to the principles of effective communication. Whether accountants feel that their technical expertise should legitimately be extended to matters of presentation of corporate information is really not the issue. Accountants are already involved, will continue to be involved, and will increasingly take directive roles in what, how, when and why to communicate financial, non-financial, quantitative and non-quantitative information to external parties. The involvement of accountants in communication issues will continue to expand. Accountants are looked upon as credible interpreters of information. While marketers, economists, public relations personnel, actuaries and management can contribute to external communication, none of these other groups possess the same perception of credibility in the eyes of investors and others when it comes to accounting-based information. This perception extends to other forms of corporate information, even if not accounting-based. Accountants will either take the initiative in supervising the new forms of communication or others will persuade them to do so because there is no other group that adequately symbolizes the requisite independence and integrity.

Preface

445

AAAJ 15,4

446

Accountants can therefore take a proactive or reactive stance to contemporary issues in communication in accounting. They can take the initiative, or they can wait until directed by management and then contribute towards the form, location and expanded content of information. A present difficulty with taking the initiative is that accountants have no comparative advantage in communication issues, and indeed may even be less skilled in this regard than other groups, for example, marketers and public relations personnel. The background of accountants does not normally include exposure to content analysis of photographs, graphical construction techniques (except in a very rudimentary and casual way), animation devices to arrest attention, and the selection of appropriate colour applications. Nor does it include a study of which information is best located where in external reporting documents. Matters such as these arise more incidentally through on-the-job experience with clients. Moreover, accountants are relatively inexperienced with regard to Web site design, the formatting of information, and the application of colours, with sound, animation and interaction for users’ appeal, understanding and speed. Studies of the visual impact of information as well as impression management techniques are also matters not normally included in the preparation and experience of accountants. Facets of effective communication would involve a consideration of which background colours best highlight or mask information, colours that are culturally unacceptable, and colour combinations that arrest or divert reader attention. Likewise, communication extends to an understanding of graphical construction techniques that would likely cause users to misinterpret information or, alternatively, would present information in the most understandable manner, with the best user recall and be the most useful for decision making. Further facets would include a consideration of how different type fonts, headings, colour and photographs can arrest attention and modify perceptions. Other matters for consideration would involve how format and presentation techniques can best create a certain impression. Moreover, an understanding about the role of redundancy and how it can be used to reinforce important issues or divert attention to less important ones is also needed. These are all communication-based issues that have been emerging for at least the past decade or so, and accountants must acknowledge a responsibility to deal with them. They are within the bailiwick of accounting because it is accountants who are the underlying preparers of corporate information, and it is accountants who should therefore accept the responsibility for undertaking the communication. This is inevitable, or accountants will expose themselves to the vagaries and possible mischief of others, with potential for misunderstanding, misinterpretation, distortion, investor losses and litigation. The following anthology of six feature articles and two research notes, which appears in this ‘‘Communication, corporate annual reports and perception engineering’’ special issue of AAAJ, each emphasizes in its own way the legitimate role of communication within the domain of accounting. Today’s

audience of investors, clients, customers and others live a multi-faceted fastpaced existence. In the past, the onus was on the recipient to spend whatever time was needed to analyze communicated information, and to extract those messages that were perceived to be important for belief revisions and resource allocation decision-making. Today, however, users expect preparers to format and locate information in a manner that communicates deliberately and expeditiously. In other words, users of accounting (corporate) information expect to find what they are looking for more quickly and to be able to understand it more readily. This expectational shift has been met with increasing amounts of colour, animation, eye-catching headings, fonts, graphs and photographs in various sources such as annual reports, Internet Web sites, magazines, newspapers and other corporate releases. The eight papers contained herein are best viewed as four bracketed pairs of themes. The first pair emphasizes annual report research taxonomies; the second pair emphasizes annual report disclosure and measurement of narrative disclosures methodologies, the third pair emphasizes graphical presentation of information, and the final pair comprises research notes that emphasize the information content and signaling power of photographs. All articles, with the exception of one, focus directly on the annual report as the communication vehicle. The first two papers should be viewed as a set of taxonomies of prior annual report research. Whatever way one views the actual role of the annual report, it is normally the leading and most visible of corporate documents and it provides management with a unique opportunity to impress its readers. The first paper by White and Hanson uses ‘‘economic man’’ as a basis to rationalize and categorize much of the diverse and fragmented annual report research based on national, theoretical, methodological and political perspectives over the past 30 years. Patricia and John Stanton also analyze different research perspectives used in annual report studies published during 1990-2000 and identify substantive issues still remaining for investigation. They also make explicit that the ‘‘modern corporate annual report uses the tools of management, marketing and communication theory to construct a picture of the organisation’’. Impression management considers analysis of imagery, hedonistic bias and linguistic approaches. The marketing perspective incorporates annual reports into an overall strategy that sends target audiences a consistent message that promotes corporate goals. The political economy, organizational legitimacy and accountancy perspectives also add depth to the taxonomy. The second two papers should be viewed for their methodological contributions, extending investigative tools that have been in the literature for some time. For example, the use of a disclosure index to identify gaps in annual report informational items has been in the literature for more than a quarter of a century. Hooks, Coy and Davey introduce a contemporary weighted disclosure index of needed informational items to measure annual report accountability of electricity companies in New Zealand. Their methodology,

Preface

447

AAAJ 15,4

448

which with little modification can be replicated in any country and for any industry, identifies gaps in present disclosures that need to be remedied for effective communication to reach a wider stakeholder audience. The second methodological paper extends the work undertaken on the readability measurement of narrative disclosures within annual reports. This topic has been reported in the literature since 1952. Initially within the framework of formula-based readability studies that examine reading ease and understandability, more recent studies have focused on the use of narratives as a technique of impression management. Sydserff and Weetman develop a new transitivity index and the application of DICTION scoring to accounting narratives, as an operational complement to readability scores. Using a systemic approach to language study, DICTION is a computerized content analysis program that examines linguistic structures to detect persuasive and rhetorical narratology. A transitivity index is a measure of the number of passive constructions in the text and is linked to studies that are concerned with patterns of causal reasoning used to explain corporate performance. This study should be viewed as a catalyst for further linguistic-based studies in accounting communication. The remaining two feature articles are laboratory studies that focus on the role of graphical presentations of information. Graphs constitute one of the ‘‘quick fix’’ techniques users want for directed and purposeful information acquisition. While graphs can arrest reader attention and convey large amounts of information at a glance, hard evidence about the alleged benefits of graphs to users of accounting information is not yet part of the established literature. It is thought that graphical presentation aids in decision making, and it is thought that users can see through misleading graphical construction techniques, but the evidence is scant. Our understanding of the second of these issues is aided significantly through the first paper, where Beattie and Jones continue their industrious publication record about graphical use and abuse by examining the measurement distortion of graphs in annual reports. An experiment based on visual information processing theory revealed that users’ perceptions were affected by measurement distortions of approximately 10 percent and that no measurement distortion in excess of this threshold should be allowed. So and Smith extend graphical analysis into the role of colour, an issue in accounting communication that has been seriously neglected. Using a laboratory experiment, they assess the interaction effects of stand-alone coloured bar charts with information complexity and gender on decision performance. This is the only paper in the set that does not use the annual report as its vehicle of analysis, but was included because of its unique contribution to our awareness of the impact of colour within accounting communication, and because colour graphics form an integral part of financial reporting. They found evidence for the claim that colour graphics improves decision making, though its impact is significant only when information complexity is low, and then only for female subjects.

The final pair of papers comprises research notes that creatively employ annual report photographs to convey signals and perception engineering. Davison uses an annual report case study to apply analytical methods from artistic disciplines. Antithesis, a rhetorical device, is examined to show how it can influence and engender patterns of reading and thinking by annual report users. The research reveals that the messages communicated by creative design are frequently more compelling than those of the financial statements. This is important because lay and expert readers are looking beyond the accounting numbers for insights. The potential for perception engineering by annual report preparers through the use of antithesis is enormous. The second research note also focuses on photographs published within annual reports. Bernardi, Bean and Weippert use the question of gender diversity as the basis for examining 94 percent of the Fortune 500 companies with regard to how boards of directors are pictorially presented in annual reports. The point made by this research is that photographs signal information about the credibility, maturity, experience and vigour of directors in a visual way that cannot be captured within the traditional financial reporting system. In other words, photographs tell their own story and add information content. They possess a power and a viewpoint that is based on the agencies or individuals that construct them. Similarly, to Davison’s findings on antithesis, photographs can be used for perception engineering, and hence, this research topic opens up entirely new lines of enquiry into impression management. From these eight fresh contributions to the literature it can be seen that aspects of annual report communication possess exciting vistas that extend well beyond traditional financial reporting. Accountants should not dodge responsibility for information presentation issues. Accountants possess the requisite credibility, they are involved with diverse information selection and preparation, they are consulted about content consistency, and they contribute as part of a communication team to myriad corporate releases. The accountancy profession should more explicitly acknowledge its involvement in this role of communication and take constructive steps to understand the ramifications of narrative and visual techniques that contribute towards perception engineering. John K. Courtis Guest Editor

Preface

449

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

AAAJ 15,4

Economic man and disciplinary boundaries A case-study in corporate annual reports

450 Received April 2001 Revised October 2001, February 2002 Accepted March 2002

Robert White School of Sociology and Social Work, University of Tasmania, Hobart, Tasmania, and

Dallas Hanson School of Management, University of Tasmania, Hobart, Tasmania Keywords Company reports, Knowledge workers, Language, Economic systems, Communications Abstract This paper is an empirical response to two of Quattrone’s claims: first, that research in accounting is fragmented; and then that this follows from the blocking of communication by intra- and inter-disciplinary boundaries. Although we agree with much of Quattrone’s argument, and in particular with his problematising of ‘‘economic man’’, we draw an opposite conclusion. Rather than looking to a trans-disciplinary removal of boundaries, we use a survey of 30 years of research in corporate annual reports to defend narrowly disciplinary work. We make our case through discussing problems of intra- and inter-disciplinary unity in research, the puzzle of the role of ‘‘economic man’’ in the study of annual reports, and the alternative to him in science and technology studies (STS). Our approach yields a better fit than Quattrone’s own solution with his aims of an evolutionary perspective that allows for historical shifts, and for a reflexivity that includes the inevitable entanglement of researchers in what they study. We conclude by noting that our approach is applicable to the study of corporate communication more generally.

This paper is a response to Quattrone (2000). Where he described research in accounting as fragmented, ascribed this effect to intra- and inter-disciplinary divisions, and prescribed their trans-disciplinary removal, we make a case for narrowly disciplinary activity, and for the effectiveness of communication across disciplinary boundaries. While we agree with Quattrone’s (2000) attention to the methodological individualism of that ‘‘economic man’’ who strides through most of the social sciences, we reverse his argument by remodelling economic man and by showing how our model operates communicatively. We develop that approach through a survey of studies of corporate annual reports, and support its generalisability through the science and technology studies (STS) that Quattrone (2000) also invoked. Since the social sciences where organisations and corporations are studied are themselves organisational and corporate effects, the practice of research should be expected to repeat on a smaller scale the more general patterns in corporate and organisational activity. If academic disciplines are fractals in this way of the disciplined world studied in them, then studies of the effects of disciplinary Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 450-477. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440559

The authors are to grateful to Warren Sproule and Kathy Gibson for commenting on an earlier draft of the paper, and to the AAAJ’s editors and anonymous reviewers for their generous and helpful suggestions towards its improvement.

boundaries are at the same time studies of social life at large. That means that Economic man this study of disciplinary communication is applicable to communication more and disciplinary generally. boundaries After first narrowing the terms of our inquiry we begin with a crucial tension in Quattrone’s (2000) paper. When he queried methodological individualism, he also queried the unitary rationality attributed to economic 451 man as a calculating utility-maximiser. In assuming, however, that fragmentation in research is pathological and in looking to a trans-disciplinary perspective for a cure, he held out the promise of just such a unitary rationality, and then implicitly exemplified what he explicitly problematised. In the next section we focus on corporate annual reports, as a specific means of access to this general tension between unity and diversity. Following a note on the reports’ formal requirements, we find that study of them is indeed fragmented, not only along disciplinary lines, but on theoretical, methodological, political and national grounds as well. The field is still relatively unified, however, by the assumption of economic man. If economic man is problematic then so too is that apparent unity. We note in particular that students of the reports often shift from the individual agency assumed for economic man to a collective scale, while retaining their individualist assumptions. Since this raises difficulties in approaches to the authorship and readership of annual reports, findings in their study have to be splintered. Our survey, that is, moves from unity to fragmentation to unity and back to fragmentation. In our third section we turn to another point that Quattrone (2000) raised but did not develop, the usefulness of science and technology studies (STS) in addressing the problems in corporate and organisational study that he identified. After an introductory note on STS, we use it to work our way back through the problematic unity of economic man and through the problematic fragmentation of findings on corporate annual reports. This approach meets most of Quattrone’s (2000) aims while remaining open to the benefits of intraand inter-disciplinary work. On our assumption that research in the social sciences is a fractal of the social worlds studied in them, we conclude that attention to the crossing of disciplinary boundaries seen in the study of annual reports is a useful tool for understanding the crossing of social boundaries in corporate and organisational communication. Narrowing the inquiry Before we develop the case-study proper we need to make three limits in our study explicit. First, we should stress that even if we were so minded we are in no position to prescribe anything for research in accounting, for neither of us is trained in the discipline. Since both of us read journals of research in accounting with interest and pleasure, we were suspicious when we found Quattrone’s (2000) claim of non-communication in the field, but we remain conscious that we are reading and writing as outsiders. In fact, we are unsure of the disciplinary status of our case for strictly disciplinary activity. Since we are

AAAJ 15,4

452

aiming our study of research in accounting at a readership in accounting, our work is in a sense cross-disciplinary; since one of us has a doctorate in management and the other in sociology our work is inter-disciplinary in a narrow sense; and since we are developing a general case from the specific field of research in accounting our work is also in a sense trans-disciplinary. We can only hope that the logic of our argument both keeps those tensions in play and removes any implication that we are preaching from outside. As a reflection of our disciplinary limits, second, our survey of research into corporate annual reports is wide-ranging but not comprehensive. We have set aside, for example, an extensive literature on the technical use, validity and meaning of numerical accounts, even though this literature shows most clearly the working of economic man. While we find attractive the argument that accounting is a calculative technology whereby the moral is translated into the factual (Miller and O’Leary, 1993), the specialist reading of balance sheets is beyond our focus on the use of annual reports in corporate communication to a lay audience. Similarly, and as one of the paper’s reviewers noted, we have drawn far more of our material from journals concerned with the social side of accounting than from the more disciplinary journals of accounting research. That is consistent with our pre-given social and organisational/managerial focus. So we acknowledge that our sample of studies cannot be taken as representative of accounting as a discipline. Finally, we should note that throughout the paper we have adopted a convention that appears to belie our reflexive insistence that researchers of the social world are also in the social world. We would prefer to use the same language to describe productions of corporate knowledge, academic studies of those productions, studies of those studies, such as Quattrone’s and ours, and so on. This would be consistent with the avoidance in STS of ‘‘levels of analysis’’, and of the claim to a position above the fray implied in them, that we discuss later. Practically, however, it leads to confusion, and in particular with the words ‘‘analysis’’ and ‘‘analyst’’. These not only suggest the epistemology that is at issue, but also have technical meanings in the field with reference to ‘‘financial analysis’’. To avoid confusion, we use ‘‘analysis’’ only in that narrow sense, and use words such as ‘‘research’’ or ‘‘study’’ to describe all other work in the fields we cover. We stress that this distinction is a practical matter, and in no sense a claim to a privileged vantage-point in research. This narrowing of our terms still leaves us ample material to address the issues that Quattrone (2000) raised. We now turn to the details of his argument. Quattrone on intra-, inter- and trans-disciplinary research Quattrone started by claiming that ‘‘accounting research and knowledge have splintered into many divergent and incommensurable perspectives, each with their own epistemologies’’ (2000, p. 130). He attributed this effect to intradisciplinary treatments of accounting as a distinct field within a hierarchical ranking, ‘‘the encyclopaedia’’, and to researchers’ inter-disciplinary adoption of approaches that leave this ranking undisturbed. ‘‘The encyclopaedia’’ was

characterised by the classical model of ‘‘science’’ and by the methodological Economic man individualism of its translation into ‘‘social science’’. Quattrone (2000) and disciplinary diagnosed three problems in accounting research on that basis: first, boundaries fragmentation; second, a stress on either individuals or contexts in organisational study, and thus a typical privileging of single perspectives among the many available; and third, a lack of self-critique in studies of the 453 links between shifts in accounting theory and historical shifts in the world under study (Quattrone, 2000, p. 136-7). To redress these difficulties, he drew on the ‘‘equilibration’’ in Piaget’s genetic and constructivist epistemology to call for the removal of disciplinary boundaries. His ideal perspective would be trans-disciplinary (bringing a unity to the field), evolutionary (allowing for historical shifts), and reflexive (including the inevitable entanglement of researchers in what they study). We sympathise with much of this, and particularly with Quattrone’s (2000) constructivist focus on the issues of ‘‘methodological individualism’’ and ‘‘ontological gerrymandering’’. The first of these refers to the assumption of individual actors as units of study, and the second to appeals to the authority of one science or discipline to justify another; any warranting of the individualist economic man through an individualist epistemology shows the link between them. Aspects of Quattrone’s (2000) approach to them, however, call his conclusion into question. First, when he assumed a unified study that once prevailed but now is lost, he evoked the Garden of Eden or the tower of Babel rather more than the evolution that his argument required. Whether or not he saw the original sin in the attempt to build the hierarchical ‘‘encyclopaedia’’, when he depicted the fragmentation of intra- and inter-disciplinary work as pathological he in effect denied the process of speciation through environmental specialisation needed for his argument. He then preempted the tension he set between parts and the whole, or between fragmentation and unity. A similar effect appears in his looking to an epistemology, however nonindividualist, to ground his case for trans-disciplinarity. He did grant that epistemology ‘‘is not a ‘superior’ form of knowledge but just a different one’’ (Quattrone, 2000, p. 141). When he drew on epistemology to justify his argument, however, rather than on the accounting research he studied, he preempted his call for reflexivity through a version of ontological gerrymandering. He then restored what he had usefully put at issue: the hierarchical ranking given in epistemological distinctions between metatheoretical, theoretical and practical ‘‘levels of knowledge’’ (e.g. Quattrone, 2000, p. 146). These blocks in the evolutionary and reflexive moments of his argument recur in its third element, the call for a trans-disciplinary unity. When he read fragmentation in accounting research as pathological, he assumed, first, that a discipline should be unified, and then that it could be unified beyond the reflexive entanglements he identified. He thus implicitly restored the very dream of rational order epitomised in ‘‘the encyclopaedia’’.

AAAJ 15,4

454

To address those tensions in Quattrone’s (2000) argument we take his points as empirical questions. Specifically, we assess his claim that transdisciplinarity is necessary on the basis that intra- and inter-disciplinary research strategies ‘‘have been unable to establish a dialogue between different perspectives in accounting research’’ (Quattrone, 2000, p. 147). That empirical move meets both his reflexive and evolutionary criteria, for it is drawn from practices in the field itself, and it rests on the assumption that in so far as modernity is evolutionary, so too are its effects in accounting research. We take our empirical case-material from one sub-subfield in accounting research, studies of ‘‘social’’ aspects of corporate annual reports. Our next step is to outline the genre’s formal requirements. Corporate annual reports: starting with a problematic unity Like Quattrone (2000), we start with unity. This unity, however, scarcely extends beyond a vague consensus that corporate annual reports comprise a relatively unified genre and that they are well worth studying. With researchers disagreeing over such basic issues as the reliability and credibility of the reports, the subfield is fuzzy from the start. The requirement that listed companies in capitalist economies present annually audited accounts to their shareholders and other investors imposes a degree of uniformity on corporate practice and allows a degree of uniformity in corporate study. While such statutory information is a staple of the large and growing industry of financial analysis and advice offered to professional and institutional investors and to the growing numbers of small shareholders, a recent shift towards continuous disclosure has made this use of the reports less important. In the interests of cost-saving and intelligibility, most regulators now permit companies to present abridged accounts. But the reports contain more than strictly financial data, and here too companies face statutory obligations. Under the typical Australian legislation, for example, each company must describe its operations, indicate any significant changes in its state of affairs, state its principal activities and major changes in them, detail specific and material events arising between the end of the financial year and publication of the report which may affect future operations, foreshadow other developments, and, finally, outline its compliance with regulation on the environment and corporate governance. Although the presentation and design of annual reports have drawn some attention (e.g. Graves et al., 1996; McKinstry, 1996; Preston et al., 1996), most of the research we examined was directed to their textual components, and to the scope for discretion in fulfilment of those formal obligations. The value of the textual components of annual reports as a focus of research has been widely noted and exploited. Given shifts in the provision and accessibility of information, companies may no longer use their reports, if they ever did, as a ‘‘key influencer’’ (Ind, 1997, p. 112). However, as ‘‘the most publicized and visible document produced by publicly owned companies’’ (Henriques and Sadorsky, 1999, p. 91), or as ‘‘management’s unique opportunity

to communicate directly with present and potential stockholders and creditors’’ Economic man (Tennyson et al., 1990, p. 391), an annual report has particular advantages for a and disciplinary corporation. It is: boundaries . . . the one communication medium to outside parties over which corporate management has complete editorial control. It is therefore not subject to the risk of journalistic interpretation and distortions possible through press reporting (Guthrie and Parker, 1989, p. 344).

This is also an advantage for researchers, for the more that managers control the reports the more the reports allow unobtrusive access to corporate activity. Written for purposes other than academic study, a report is ‘‘like a projective test taken inadvertently’’ (Bowman, 1984, p. 63). Besides conveying factual information about a firm, reports also ‘‘communicate implicit beliefs about the organization and its relationships with the surrounding world’’ (Fiol, 1989, p. 278). They are useful in comparative study, since statutory requirements mean that they ‘‘provide fairly comparable sets of data for a broad sample of corporations’’ (Bettman and Weitz, 1983, p. 165). They are also suited to longitudinal study, since the annual report is the only form of corporate disclosure ‘‘that is institutionalized and provided on a regular basis year after year’’ (Buhr, 1998, p. 169). It is easy to see why researchers have found their study so appealing, for no other medium offers the same blend of consistency, accessibility and wide applicability. No other medium yields the same access to corporate communication with lay audiences. While this gives a degree of unity to the field, however, that is about as far as agreement reaches. Researchers disagree even on the basic issue of the credibility of annual reports as sources of information on corporate activity. Some are optimistic. Bowman (1984), for example, took the discussion in them as ‘‘a reasonable surrogate of real activity’’ (1984, p. 64). Bettman and Weitz (1983) suggested that this was structurally enforced when they noted that writers of the letters to shareholders commonly included in annual reports faced pressures to be accurate: Although such letters are not formally audited, they are subject to a great deal of public scrutiny from stock analysts, shareholders and others. There could thus be severe consequences if obvious biases were shown in the causal reasoning presented (Bettman and Weitz, 1983, p. 171).

Courtis (1995) added an ethical element in holding that beyond compliance with regulatory and auditing requirements, a report is credible because ‘‘it reflects the integrity of management in communicating objectively and comprehensively’’ (Courtis, 1995, p. 4). Other writers are more sceptical. There is evidence, for example, that environmental disclosures in annual reports are at best weakly associated with other measures of environmental performance (e.g. Ingram and Frazier, 1980; Freedman and Wasley, 1990; Deegan and Rankin, 1996). Likewise, Steele (1982) noted the undue optimism of chairmen’s non-quantified forecasts, Tennyson and his colleagues (1990) concluded that predictions in reports are accurate only about 50 percent of the time (Tennyson

455

AAAJ 15,4

456

et al., 1990, p. 393), and Fiol (1995) found that there may be no correlation between the reports and internal company documents. Rather than take them at face-value, then, researchers have read the reports as ‘‘strategic illusions’’ (Salancik and Meindl, 1984), as ‘‘ideology’’ (Macintosh, 1990), as ‘‘numerology’’ (Lee, 1990), as ‘‘impression management’’ (Neu et al., 1998; White and Hanson, 2002), or as ‘‘rhetoric’’ (White and Hanson, 2000). Since the generic unity of corporate annual reports is problematic in these ways, Quattrone’s (2000) diagnosis of ‘‘fragmentation’’ seems immediately plausible. Any student of the reports could confirm Anderson and Epstein’s (1996, p. 13) claim that the ‘‘most important conclusion that can be drawn from a review of prior studies is the conflicting evidence they provide’’. We show in the next section that intra- and inter-disciplinary differences do produce conflict, and that sources of fragmentation also extend well beyond these effects. From problematic unity to fragmentation The relative uniformity in what is required of annual reports is no guarantee of uniformity in their study. Rather, we hold that evidence in the field has to be conflicting, and that this would still be so if the reports’ reliability and credibility were agreed. Consistent evidence requires consistent questions, consistent methods and consistent criteria of plausibility, and since that combination is found nowhere in the social sciences it would be rash to expect it here. We do not have space for systematic illustration of the resulting differences. While the details of conflict over particular findings would be relevant in a narrowly focused substantive study, the scope of our survey means that we are more interested in the sources of inconsistency than in the inconsistencies as such. Adding to the intra- and inter-disciplinary divisions that Quattrone stressed, we argue that the varieties of national contexts, topics, theories, methods and political stances in the study of annual reports all make conflicting evidence inevitable. Reports and national contexts Annual reports have been studied across a range of national contexts. To take only studies written in English, most of those we have already cited were of US corporations, but we can also note work from the UK (e.g. Dev, 1974; Lee and Tweedie, 1990; Bartlett and Chandler, 1997), Canada (e.g. Buhr, 1998), Hong Kong (Courtis, 1995), Finland (Niskala and Pretes, 1995), and Australia and New Zealand (e.g. Kabanoff and Holt, 1996; Palmer et al., 1997; Wilmshurst and Frost, 2000). While accounting standards and stockmarket listing rules are becoming increasingly global, enough national variation remains in what is required of annual reports to yield different emphases in these studies, even before cultural differences are included. Guthrie and Parker (1990) and Gamble et al. (1996) clearly showed this effect in their comparative surveys.

Reports and research focus Economic man Students of annual reports have used them to illuminate a range of substantive and disciplinary topics in the study of corporations. The reports have been examined for their boundaries use and reliability as sources of information for investors (e.g. Hines, 1982; Freedman and Jaggi, 1986; McConnell et al., 1986; Lee and Tweedie, 1990; Epstein and Freedman, 1994; Nagy and Obenberger, 1994; Anderson and 457 Epstein, 1996; Bartlett and Chandler, 1997). They are a focus in the study of ‘‘environmental responsibility’’ (e.g. Wiseman, 1982; Freedman and Wasley, 1990; Owen, 1992; Freedman, 1993; Gamble et al., 1995; Niskala and Pretes, 1995; Buhr, 1998; Henriques and Sadorsky, 1999), of ‘‘ethical investment’’ (e.g. Harte et al., 1991; Arnold and Hammond, 1994), and of ‘‘corporate social responsibility’’ more generally (e.g. Rockness and Williams, 1988; Guthrie and Parker, 1989; Patten, 1991; Wolfe, 1991; Gray et al., 1995; Milne and Adler, 1999). They have been used in studies of broad themes like corporate strategy (Bowman, 1978, 1984), organisational culture (Schein, 1984), managerial control, ideology and obfuscation (Neimark and Tinker, 1986; Macintosh, 1990; Courtis, 1998), constructions of class and gender (Tinker and Neimark, 1987), corporate reputation (Herremans et al., 1993), and corporate identity (Ind, 1997). They have also been used to study such specific topics as occupational safety and health (Chan, 1979), managerial claims of success or attributions of blame for failure (Bettman and Weitz, 1983; Staw et al., 1983; Salancik and Meindl, 1984), implicit organisational beliefs (Fiol, 1989), managers’ attentional patterns (D’Aveni and MacMillan, 1990; Abrahamson and Hambrick, 1997), the espoused values in organisations (Kabanoff and Holt, 1996), and downsizing (Palmer et al., 1997). In fact, it would be difficult to find an aspect of corporate functioning that has not been studied through annual reports. Reports and theory As that range of topics already suggests, students of annual reports have also brought to bear a range of the theoretical positions developed in the social sciences. Although Ullmann (1985) could talk with some justice of data in the field being in search of theory, much work is either explicitly theorised or couched in theory-testing terms. Parker (1982), for example, applied mass communication theory, and Hines (1982) studied an apparent anomaly between expectations under the ‘‘efficient markets hypothesis’’ and the results of shareholder surveys. Some writers have used agency theory (e.g. Abrahamson and Park, 1994), or its variant in the stakeholder theory that Ullmann (1985) invoked (e.g. Roberts, 1992; Henriques and Sadorsky, 1999). Others have turned to psycho-organisational theories to study ‘‘strategic illusions’’ (Salancik and Meindl, 1984), collective cognition (Abrahamson and Hambrick, 1997), or ‘‘impression management’’ (Neu et al., 1998). Others have tested theories of legitimacy (Guthrie and Parker, 1989; Patten, 1991; Campbell, 2000; Wilmshurst and Frost, 2000), or of the related legitimation (Palmer et al., 1997). Others again have adopted critical theory (Macintosh, 1990), or one or another

AAAJ 15,4

458

version of Marxist political economy more generally (e.g. Tinker et al., 1991; Gray et al., 1995). Reports and methods In using or testing that range of theories, students of annual reports have also used a range of social scientific methods. Content-analysis is the most common among our selected studies, but it includes many variants and leads to many developments or supplements (e.g. Wolfe, 1991; Jones and Shoemaker, 1994). Content-analysts have presented their results in graphical form (e.g. Gray et al., 1995), and have correlated them with secondary data (drawn from a range of sources), either simply (e.g. McConnell et al., 1986) or through multivariate or factor analysis (e.g. Roberts, 1992; Palmer et al., 1997). Some writers have used such other textual methods as semiotics (e.g. Fiol, 1989). The use of surveys and questionnaires is common, directed to both firms and individual shareholders (e.g. Lee and Tweedie, 1990; Harte et al., 1991; Epstein and Pava, 1993; Epstein and Freedman, 1994; Anderson and Epstein, 1996). Many writers have used historical records (e.g. Arnold and Hammond, 1994; Buhr, 1998). Some have interviewed managers (e.g. Fiol, 1989). Overall, most of the standard methods in the social sciences have been used to study annual reports. Reports and politics As also suggested by the range of theories we noted above, students of annual reports have adopted a range of political positions. In so far as they recommend the reports as sources of information, either for management or for individual investors (e.g. Bowman, 1984; Nagy and Obenberger, 1994), some seem to accept the processes of liberal-democratic capitalism. But despite the charge of ‘‘political quietism’’ levelled against much work in the field (Tinker et al., 1991), many writers do present explicitly ‘‘radical’’ accounts. The very framing of issues such as ‘‘corporate social responsibility’’ and ‘‘environmental responsibility’’ implies a politico-moral stance at odds with current arrangements. Reports and disciplines This critical tendency is most evident in studies of accounting, of auditing, and of corporate responses to socio-environmental issues. As Quattrone (2000) noted, however, intra-disciplinary divisions in accounting are deep, and these are matched by inter-disciplinary differences. A gradient of disciplinary emphasis appears in the studies that we studied. Critical advocacy in accounting (e.g. Macintosh, 1990; Owen, 1992) shades into the more detached studies in administrative theory (e.g. D’Aveni and MacMillan, 1990), in organisational theory (e.g. Abrahamson and Hambrick, 1997) and in management theory (e.g. Fiol, 1995), and these in turn shade into the more applied focus in financial analysis (e.g. Nagy and Obenberger, 1994), in public relations (e.g. Marino, 1995) or in policy-making (e.g. Walden and Schwartz, 1997). As Quattrone (2000) also stressed in his reflexive moment, disciplines are

enmeshed in the processes studied within them, and this one field displays the Economic man variety of entanglements that follow from the results in different disciplines and disciplinary meeting the needs of different publics. boundaries Given such disciplinary differences, and the range of national contexts, topics, theories, methods, and political emphases that they variously reflect, conflicting results are inevitable in the study of annual reports. It would be 459 vain to expect consistency. If the fragmentation that Quattrone (2000) found in accounting research at large is then evident in this one sub-field, however, the diagnosis of it as pathological is another question. Furthermore, the field is less divided than this first scanning of it suggests, for a degree of unity in disunity follows from the widespread adoption of rational and utility-maximising individuals as units of research. As Quattrone (2000) noted in his discussion of ‘‘methodological individualism’’, that unity is also problematic. We examine it in the next section through the rationality of economic man. From fragmentation to a problematic unity: using economic man The ‘‘economic man’’ who is axiomatic in neoclassical economics is a relatively unifying figure in the study of corporate annual reports, as he is in much of the modernist social science of the last two centuries. Many refinements of the model have been proposed, including emphases on forms of rationality that cannot be reduced to the strictly calculative. Critiques have been just as numerous, but although versions of these are prominent in the field, the individual rationality attributed to that utility-maximising individual is more common. In so far as the model is problematic, then, so too is the partial unity suggested by it. We identify and examine difficulties in accounts of the ‘‘agency’’ of the corporations publishing annual reports, difficulties that recur in the interrelated issues of how the authorship, audience and language of the reports are understood. Corporations as agents Whatever unity the study of corporate annual reports commands has a worm at its core: the routine treatment of corporations as ‘‘agents’’, or as purposive actors, belies the routine assumption of economic man and the methodological individualism it entails. This is an old problem, for writers have striven to distinguish between corporal and corporate selves ever since the concept of ‘‘legal person’’ was first applied to corporations in the thirteenth century (e.g. von Gierke, 1900; Coleman, 1990). Writers on annual reports have rarely joined these arguments, although it is generally accepted elsewhere that collective action is ‘‘paradoxical’’ given the assumption of that univalent rationality defining economic man (e.g. Olson, 1965). Even such an otherwise widely adopted approach as the ‘‘new institutionalism’’ is scarcely mentioned in the field, despite the promising focus on institutions as independent variables, on cultural arationality, and on structures that cannot be reduced to aggregates of individual actions (DiMaggio and Powell, 1991; Carruthers, 1995).

AAAJ 15,4

460

The issue is not entirely absent. In a study of the patterns of responsiveness to ‘‘risk’’ found in annual reports, for example, Bowman (1982) applied to corporations a model of ‘‘risk-seeking’’ that was derived from individual behaviour. He noted that speaking ‘‘of a company as an individual (i.e. a rational actor) is a kind of anthropomorphism that apparently is quite common in political science literature as well as in the literature of economic theory’’ (Bowman, 1982, p. 40, italics added). But while he was quite right that the usage was common, he did not address one of its effects. Since the model of economic man as rational actor rests on an assumption of a distinct human nature, the anthropomorphism belies the strictly rational analysis that the model allows. Writers on annual reports commonly gloss over this contradiction. Some do so by treating corporate outcomes as the sum of individual decisions. That was Abrahamson and Park’s (1994) approach, for example, when they used ‘‘agency theory’’ to study the accounting for ‘‘bad news’’ in annual reports. But as they implied when they described their results as managerial ‘‘concealment’’, their assumption left them no option but to take managers as either conspiratorial or self-deluding. Of course, groups of managers and individuals among them may well be scheming cabalists and/or self-deceivers, but that is still a limited set of choices. Ruling out from the start the possibility of emergent effects, or the chance that wholes may be more than the sum of their parts, Abrahamson and Park (1994) then in effect betrayed the rationality in their ‘‘agency theory’’ by begging the question of its adequacy. Another form of question-begging is more common. In an elision of scale, researchers often grant the salience of aggregate effects, but do so by treating corporations as individual agents, or as economic men writ large, while retaining an assumption of economic man as distinct. Guthrie and Parker (1989) typify this move, in their review of the legitimacy theory that they tested through annual reports: This theory is based upon the notion that business operates in society via a social contract where it agrees to perform various socially desired actions in return for approval of its objectives, other rewards and its ultimate survival. It therefore needs to disclose enough social information for society to assess whether it is a good corporate citizen . . . (Guthrie and Parker, 1989, p. 344).

The argument depends on ‘‘social contract’’ and ‘‘citizen’’ being read as more than metaphorical, but the liberal individualism implied in those concepts also requires the same assumptions of human nature as those underpinning economic man. That uneasy shift from an individual to a corporate scale is evident in most uses of psychological theories in the analysis of annual reports. It appears, for example, in the image of the ‘‘projective test’’ with which Bowman (1984) justified attention to the reports, it underpins uses of attribution theory (Bettman and Weitz, 1983; Staw et al., 1983; Salancik and Meindl, 1984), it recurs in studies of ‘‘attention’’ (D’Aveni and MacMillan, 1990) and of ‘‘cognition’’ (Abrahamson and Hambrick, 1997), and it can be implied in ‘‘impression management’’ (Neu et al., 1998; but see White and Hanson, 2002). Writers do not usually problematise that sliding across scale. Bettman and

Weitz (1983) came closer than most when they discussed their use of findings Economic man on self-serving attributions: and disciplinary Given the robustness of these findings at the level of individual performance, one would hypothesize the following pattern in explanations given for corporate performance: reasons internal to the organization will be cited for favorable performance outcomes and external factors will be noted for unfavorable outcomes . . . (Bettman and Weitz, 1983, p. 167).

Although they did note that studies of individuals may ‘‘provide insights’’ rather than be strictly applicable to corporations (Bettman and Weitz, 1983, p. 171), their hypotheses would be more plausible if they did not accompany the shift of scale with an assumption of the primacy of individual agents. Unless that assumption is questioned, whatever other theory is used turns into a version of agency theory, and economic man rides again. The issues of authorship, readership and language The conceptualisation of corporate agency has practical implications for the questions of the authorship and readership of annual reports. The issue of their authorship is especially acute, for there seems to be only anecdotal evidence on who writes them (e.g. Abrahamson and Hambrick, 1997, p. 519), and the access to the writers’ ‘‘intentions’’ required of agency theory is then problematic. Fiol (1995) in particular stressed the consequences: All studies using the letters to shareholders contained in annual corporate reports, including this one, have a common limitation. Researchers do not know the authors of these documents and do not know much about the conditions under which they were written (Fiol, 1995, p. 532).

Researchers typically differ in their allowance for this widely acknowledged difficulty. Bowman (1984), for example, held that: . . . although some people maintain that the prose in annual reports is written by public relations people, the truth is that the typical chief executive officer spends considerable time outlining the contents, sketching out much of it, and proofreading and changing most of it to his taste (Bowman, 1984, p. 63).

For him, then, and whatever the evidence for his ‘‘truth’’, the CEO was effectively the sole author. Salancik and Meindl (1984) similarly noted that while most firms employed public relations writers to produce letters to shareholders and annual reports, ‘‘what is written is generally taken as top management’s account. A president of a firm could not easily disclaim the contents of a letter he signed and published’’ (Salancik and Meindl, 1984, p. 243). Staw (1983) and his colleagues, however, gave a different emphasis: . . . the letters to shareholders . . . are signed by the president or chief executive officer of the firm, but it is arguable that they are organizational rather than individual communications. As noted recently in a Wall Street Journal article, ‘‘The shareholders’ letter bears only one or two signatures, but is generally a committee project. Public relations staffers or consultants, who often write the first draft, are aware that the copy has to be reviewed by the chief executive officer, the chief financial officer, the board of directors, and the legal department’’ (Straw et al., 1983, p. 585).

boundaries 461

AAAJ 15,4

462

On similar grounds, D’Aveni and MacMillan (1990) held that letters to shareholders are good indicators of major topics of organisational concern precisely because they are ‘‘the product of the inputs of many individuals’’ (D’Aveni and MacMillan, 1990, p. 640), and Abrahamson and Hambrick (1987) that if letters to shareholders are edited until they are acceptable to senior executives then they must reflect ‘‘some form of consensus among various managers in the upper echelons of organization’’ (Abrahamson and Hambrick, 1997, p. 519). With just what ‘‘organisational communication’’ means left uncertain, however, this potentially useful stress on collective writing does not sit easily with the sustained assumptions of individual agency. The readership of annual reports has been more widely studied, but it is as problematic as their authorship, and for similar reasons. Just as there is scant evidence on how the reports are written, so not enough is known about who reads them, about how readers understand their relation to writers, and about the outcome of their reading (e.g. Lee, 1990; Epstein and Pava, 1993; Anderson and Epstein, 1996). Again, researchers are typically divided on how to interpret what evidence is available. While the reports are required to be addressed formally to shareholders, for example, Ind criticised them for a ‘‘focus on a perceived narrow idea of audience (shareholders)’’ (Anderson and Epstein, 1997, p. 112), and while the managers of ‘‘ethical’’ funds do appear to respond to that focus (e.g. Rockness and Williams, 1988; Harte et al., 1991; Arnold and Hammond, 1994), individual investors may or may not (e.g. Epstein and Freedman, 1994; Nagy and Obenberger, 1994). There is a gap, for example, between what these investors do and how they would be expected to react to information in the reports under the hypothesis of an efficient market (Hines 1982), and Freedman and Jaggi (1986) found scant association between the environmental coverage there and individual investors’ decision making. Nor are these investors the only potential readers. Wolfe (1991) noted ‘‘considerable evidence that [the reports] target many different stakeholders’’ (Wolf, 1991, p. 288). Tennyson and his colleagues also stressed that broadly persuasive thrust when they took a report as ‘‘designed primarily to elicit response from its readers. This response may be to buy more stock, lend more money, refrain from selling currently held stock, or support management, etc.’’ (Tennyson et al., 1990, p. 391). Beyond shareholders, then, the readership may include financiers, consumers, suppliers, competitors, regulators, pressure groups, the press, the market, trade union officials, present and potential employees, and, vaguely, the community at large. As Tilt (1994) found in her study of environmental pressure groups, critics may be among the most avid readers. To study any one of those ‘‘groups’’ of readers is to re-encounter the puzzle of collective agency. This problem is especially acute with ‘‘the community’’. However corporate legitimacy is theorised, social consent is a precondition for both single companies and corporate activity in general. As Neu and his colleagues (1998) said, for example, ‘‘self-reporting of environmental information (usually in the annual report) pertaining to employee, community

and customer interactions often has the effect of maintaining not only firmEconomic man specific but also system-wide legitimacy’’ (Neu et al., 1998, p. 266). Yet how that and disciplinary effect is achieved is unclear. The reports’ writers might even face an impossible boundaries task in deciding on their audience. Bartlett and Chandler’s (1997) conclusion from their survey of the literature – ‘‘it is unlikely that general purpose annual reports are ever going to satisfy the widely differing information needs of a 463 large body of shareholders’’ (Bartlett and Chandler, 1997, p. 259) – is even more to the point if other potential readers are included. Overall, then, there is considerable uncertainty in the field, first over who constitutes the readership of annual reports, and then over the sense that these readers make of them. The conceptual difficulty of the agency of writers and readers and the practical difficulty of access to the writing of reports, to their writers’ intentions or to their readers’ interpretations and responses converge in the issue of how ‘‘language’’ is to be understood. Many researchers have noted the complexities here, for, unlike ‘‘collective agency’’, the variant meanings of language in use have hardly been neglected in the field. Claims such as that reports have an ‘‘essential and significant social meaning’’ (Macintosh, 1990, p. 168, emphasis added), or that ‘‘[o]ne of the tenets of effective communication is that the messages received by readers are interpreted in the same way as that intended by the sender’’ (Courtis, 1995, p. 4), are exceptions proving the rule. Writers far more usually take meanings and interpretations as diverse and arguable. Thus while the most common method in our sample of studies is some form of content-analysis, for its suitability in comparative and longitudinal study, those who use the method routinely note its limits (e.g. Wolfe, 1991; Jones and Shoemaker, 1994; Gray et al., 1995; Wilmshurst and Frost, 2000). Some writers use ‘‘triangulation’’ to allow for these limits, supporting content-analysis with historical records and interviews (e.g. Buhr, 1998). Others link their contentanalyses to the openness of metaphors (Palmer et al., 1997), or to the WhorfSapir hypothesis (Abrahamson and Hambrick, 1997). When researchers use approaches other than content-analysis, as in Neimark and Tinker’s (1986) constructionism, Fiol’s (1989) semiotics, or White and Hanson’s (2000) rhetoric, they similarly invoke arational effects beyond strictly rational study. The argument so far To recapitulate our trajectory so far, we first noted the relatively uniform constraints on the production of corporate annual reports, and then showed that, despite this generic unity, the results of studies of them are fragmented. Here we went beyond Quattrone’s (2000) stress on intra- and inter-disciplinary differences to argue that diversity also had to follow from the range of national, theoretical, methodological and political perspectives brought to the field. We next showed a relative uniformity within that disunity, in the common assumption of economic man. This leads to a conceptual inconsistency, however, in the sliding from the rational and individual agency of economic man to corporate action. Many writers have at least touched on the issue of collective agency when they stressed the difficulties that arise over the writing

AAAJ 15,4

464

and reading of annual reports, and many have also stressed the arational effects of language in those processes. The individual and rational economic man who enters the studies then emerges from them as a collective linguistic person, while insouciantly acting as if nothing has happened. So throughout the preceding three sections we have retraced Quattrone’s (2000) footsteps, on the smaller scale of the study of corporate annual reports. We agree with him on the limits of the methodological individualism of neoclassical economics that is commonly adopted in the field, we agree that reflexive attention to research is both necessary and fruitful, and we agree that research appears to be fragmented. But we disagree that this state is pathological, and that a new trans-disciplinary perspective is then required. To develop that argument we turn to another theme that he raised but did not develop, the usefulness of science and technology studies (STS) in the study of corporations and organisations. This yields one way of remodelling economic man while allowing for the effects of language. The disciplinary and the trans-disciplinary: the study of corporate annual reports and STS Quattrone had good reason to link disciplinary effects to his call for an evolutionary and reflexive perspective. Since the disciplines familiar now did emerge amid the evolution of modernity that is variously studied in them, he rightly argued that any disciplinary knowledge is always already entangled in more general processes. As we noted earlier, however, when he saw fragmentation in accounting research as pathological he implied the dream of clear ground somehow free of the reflexive tangles that he otherwise emphasised. That in turn meant that he did not make the most of the STS that he invoked briefly in ‘‘ontological gerrymandering’’ and in asides on Kuhn and Latour. This field is virtually constituted around responses to the inevitability of reflexive effects. We develop the link in this section. After an introductory note on STS and its fit with the study of corporate annual reports, we use it to work our way back through the alternating unities and disunities in the study of corporate annual reports. We go first to the problematic unity of economic man, then to the problematic disunity epitomised in disciplinary differences, and finally to a match between the generic unity of annual reports and the functioning of the social sciences. This yields a sense of the trans-disciplinary that depends on the value of intra- and inter-disciplinary boundaries rather than on their removal. Science and technology studies Since STS is familiar in organisational research it does not need detailed discussion here (e.g. Burrell and Morgan, 1979; Clegg, 1988; Czarniawska, 1997). Jasanoff et al.’s (1995) survey and Shapin’s (1995) brief review are still useful, and Biagioli (1999) gives a comprehensive coverage of work in the field. For the moment, we need only note that Quattrone (2000) was well justified in linking his reflexive project to it. Of course, it is neither the first nor only means

of access to the scope of inter-disciplinary activity. Many of the issues that we Economic man discuss, for example, are prefigured in accounts of ‘‘bounded rationality’’ (e.g. and disciplinary Simon, 1957, 1978). But the cross-disciplinary field of STS is one topical arena. boundaries Whatever unity it commands is a result of an empirical focus on science as practised, and writers in it routinely treat their own practice as subject in principle to the same constraints as those that they identify in other sciences. 465 Debates over the necessity for and problems in this reflexivity are all but constitutive of the field (e.g. Bloor, 1976; Woolgar, 1988a; Ashmore, 1989; Lynch, 2000). The refusal of analytical privilege that follows from acceptance of reflexivity means that to apply approaches in STS to accounting research need not entail the ‘‘ontological gerrymandering’’ that Quattrone (2000) took as one symptom of ‘‘the encyclopaedia’’. This rigging of the game refers to the common social scientific practice of: . . . making problematic the truth status of certain states of affairs selected for analysis and explanation, while backgrounding or minimizing the possibility that the same problems apply to assumptions upon which the analysis depends (Woolgar and Pawluch, 1985, p. 216).

Debarred from such ‘‘backgrounding’’ by the logic of reflexivity, neither STS as a whole nor the sub-disciplines to which writers in it belong can be panaceas, and certainly not in strictly rational terms. Rather, the ‘‘methodological horrors’’ of reflexivity (Woolgar, 1988b) recall that ‘‘Cretan paradox’’ that St Paul identified when, in warning against ‘‘idle talkers and deceivers’’, he noted that a Cretan had said ‘‘Cretans are always liars’’. Not having St Paul’s solution of faith, writers in STS work with the difficulty through favouring empirical attention to the ‘‘how’’ of techno-scientific practice over the explanatory ‘‘why’’ that would imply a claim to privilege. In that sense, STS is a strategy in research rather than a gerrymandered solution (Lee and Hassard, 1999). Moreover, the domains of science and technology and organisational accounting are not as distinct as a first glance might suggest. Like science, the corporate activity at issue in accounting research in general and in the study of annual reports in particular entails the production and dissemination of authoritative knowledge; like corporate activity, scientific knowledge is an organisational effect (e.g. Whitley, 1984; Fuchs, 1992; Law, 1994; Callon, 1998). That is, there is an immediate fit between study of the study of scientific knowledge and study of the study of corporate knowledge. Points at issue in one field should be expected to be relevant in the other. One of these is the rationality of economic man. STS and economic man It is a routine finding in STS that in practice scientists do not follow the strictly rational rules in epistemological or methodological accounts of scientific man. Although scientists do of course report their results as formally rational, that is a conventional retroscription of what is always a ‘‘mangle of practice’’ (Pickering, 1995). Since ‘‘scientific man’’ is ‘‘economic man’’ in the laboratory,

AAAJ 15,4

466

these studies show alternatives or supplements to the latter’s rationality. When writers in STS hold that science is a communicative, practical and embodied activity, they typically do so by showing that conflicting rationalities are in tension. As Latour (1991a) put it, the networks of both nature-culture and scientific knowledge of them are ‘‘simultaneously real like nature, narrated like discourse, and collective like society’’ (Latour, 1991a, p. 6, italics removed), and each element here implies a distinct form of rationality in interaction with that in each of the other two. Minimally, the formal and individual rationality giving scientific man access to the ‘‘real’’ is in tension with both the collective rationality of political action and the collective arationality seen in responses to shared symbols. To stress that tension is not to deny the effectiveness of economic man. Instead, it implies the refusal of that model as axiomatic. To accept the tension between rationalities is to open the way to observation of the emergent and linguistically mediated construction of economic man (Callon, 1999, p. 193). This allows a revisiting of the problem we noted in the study of annual reports, that when writers assumed the individual and rational agency of economic man they could not, or at least did not, conceptualise collective or organisational action. They were hampered by the use of only one form of rationality, and also by the theory of a distinct human nature that economic man implies. Latour and his colleagues have particularly problematised that theory, in developing accounts of both nature/culture and knowledge of it that do not require an a priori privileging of the human. They took one startingpoint in the claim that texts and artefacts are as necessary as human actors to the social life constructed in and through science and technology. When Callon and Latour (1981) compared humans with other primates, for example, they noted that human sociality required transcendence of the somatic effects and face-to-face interaction more usually seen as defining it. While the action-at-adistance that enables human sociality does require humans to be disciplined into predictability, it cannot be limited to this effect. Rather, social evolution entailed the spread of techno-social networks that are ‘‘heterogeneously engineered’’ webs of ‘‘documents’’ and ‘‘devices’’ as well as of ‘‘docile bodies’’ (Law, 1986). Documents, texts, are ‘‘immutable mobiles’’ in the construction of networks, serving as delegates from one network to others (e.g. Latour, 1987, pp. 21-62). Devices, whether technological artifacts or organisational structures, stabilise some interactive possibilities and exclude others; they are society made durable (Latour, 1991b). In these terms, the human and the social are emergent effects of heterogeneous engineering, ‘‘agency’’ is one outcome among others, and the agentic self is itself a network. When corporal limits are transcended through textual/linguistic effects and through material artifacts, the self is as much a ‘‘hybrid collectif’’ as any other centre of calculation (Callon and Law, 1995, 1997). The line between individual and human subjects, on one hand, and collective and/or non-human subjects, on the other, disappears. While a

corporal self and a corporation differ in scale, they are then open to study in Economic man identical terms. and disciplinary That approach to agency redresses the common, almost routine, elision in boundaries the study of annual reports, from the assumption that only economic man can be an agent to the attribution of agency to corporations. We should repeat that this claim does not mean that we deny individual human agency. We do claim, 467 however, that the shift from the ontologically gerrymandering axiom of individual agency in the model of economic man to the treatment of agency as emergent along the lines of STS clears a logical flaw in the study of corporate annual reports. While it does not solve the problems in the field, since the issues of how all the collectivities involved in the writing and reading of annual reports achieve their effects remain as empirical questions, it does remove a block in approaches to the problems. In that case, it is even more rational than the assumption of rationality. It also opens the way to a fresh look at the collective action seen in disciplinary differences. Rationality and the disciplinary When we surveyed studies of corporate annual reports we did find the conflicting evidence expected from Quattrone’s (2000) diagnosis of ‘‘fragmentation’’, and we agreed that intra- and inter-disciplinary differences were sources of dispute. We doubted, however, that such conflicts are pathological, on the grounds that to posit a trans-disciplinary unity is to dream of a sanctuary beyond evolutionary effects and reflexive loops. If these loops are inevitable, and that is a key claim in STS, then allowance must be made for them in any approach to the disciplinary. Foucault’s (1975) account of ‘‘discipline’’ and Kuhn’s (1970) of ‘‘paradigms’’ are standard points of departure in studies of disciplinary knowledge. In STS, Law (1986), for example, drew on Foucault for his ‘‘docile bodies’’, and the same move is familiar in accounting research (e.g. Hoskin and Macve, 1986, 1993). The ‘‘disciplinary matrix’’ with which Kuhn (1974) replaced his ambiguous ‘‘paradigm’ is less familiar. It comprised ‘‘symbolic generalisations’’, ‘‘models’’ and ‘‘exemplars’’. Denoting routinely used and formally logical expressions, symbolic generalisations have dual effects; an equation like f ¼ ma both recursively defines its elements and implies acceptance of logico-mathematical argument. ‘‘Models’’ provide a group with preferred analogies, as in the effects of light being taken as either waves or particles. ‘‘Exemplars’’, finally, are lived as concrete problem-solutions that are learned in daily practice (Kuhn, 1974, pp. 297-8). The ‘‘disciplinary matrix’’ then entails the same tension between formal rationality and the cultural rationalities of language and of life as we found in the alternative to economic/scientific man in STS. As Geertz (1983a) said when he read Foucault and Kuhn in his own anthropological terms, study of those cultural rationalities entails an ethnographic ‘‘unpacking of performed meaning’’ or a ‘‘practical epistemology’’ (Geertz, 1983a, p. 29). Such a textual ethnography shows disciplines as ‘‘ways of being in the world, to invoke a Heideggerian formula, forms of life, to use a Wittgensteinian, or varieties of

AAAJ 15,4

468

noetic experience, to adapt a Jamesian’’ (Geertz, 1983b, p. 155). As with the tension Latour set between real, narrated and collective rationalities, the rationality of any discipline’s symbolic generalisations emerges in that disciplinary way of life. This entails the distinctive concrete problem-solutions that are lived as common sense and the distinctive analogical vocabularies that form the language of mundane interaction. This does make communication across disciplinary boundaries problematic. To understand a particular disciplinary knowledge requires an understanding of the language of that discipline, this requires familiarity with the ‘‘heroic myths’’ and ‘‘disciplinary heroes’’ of the discipline’s formation and development that suffuse its language (Becher, 1989), and this in turn requires a sustained and intensive disciplining that is difficult to repeat. In economic terms, the cost of joining a discipline is high. Furthermore, any member of a discipline could attest that this inter-disciplinary effect recurs on an intra-disciplinary scale; to join a discipline is always to be acculturated in a particular sub-discipline or speciality. Since contact across disciplinary and sub-disciplinary boundaries then must be incomplete, conflicting outcomes in the study of any issue are inevitable. Yet practitioners habitually make sense of differences that might appear bewildering to outsiders, and even be a source of contempt for the social sciences in general. The cross-citational webs typical of studies of corporate annual reports, for example, show that writers routinely use work resting on disciplinary assumptions other than their own, and that is before their national, political, theoretical and methodological differences are taken into account. The bibliographies in the studies we have cited all show these effects. In this practical epistemology, then, the writers refute Quattrone’s (2000) claim that they ‘‘have been unable to establish a dialogue between different perspectives in accounting research’’ (Quattrone, 2000, p. 147). To the contrary, their dialogue is unceasing. Those whom we have noted seem to act as bricoleurs who use whatever they can find, regardless of its disciplinary provenance, and translate it into the symbolic generalisations, analogical models and concrete problem-solutions that are familiar to their particular disciplinary audiences. This is neither specific to research in corporate annual reports nor a practice to criticise. There can be no sense-making without such bricolage. All researchers in the social sciences routinely and fruitfully go hunting and gathering beyond their disciplines proper. In these textually ethnographic terms, the explicitly disciplinary and subdisciplinary study of corporate annual reports is always already implicitly trans-disciplinary. On that basis we take our next step back, and link this messy trans-disciplinarity to the problematic unity of the corporate annual report as a genre. The trans-disciplinary and the world under study In his reading of accounting research, Quattrone (2000) noted that one outcome of the hierarchical ranking of knowledge in ‘‘the encyclopaedia’’ was the distinction between meta-theoretical, theoretical and practical ‘‘levels of

knowledge’’ (Quattrone, 2000, pp. 135-7). Both he and writers in STS Economic man emphasised that the reflexive entanglement of any study in what is studied and disciplinary makes these distinctions problematic. If the practice of study is necessarily boundaries enmeshed in this way, then it makes more sense to talk of ‘‘epistemic flattening’’ than to invoke epistemological hierarchies (Lynch and Bogen, 1997). That in turn means that the practice of research into corporate annual reports is 469 informative of corporate activity in general. The field as a whole, in all the multi-disciplinary variety that we have emphasised, is like a ‘‘trading zone’’ or entrepoˆt, where some common language is necessary (Galison, 1999; Lowy, 1992). Without that, researchers from one discipline could not use results from other disciplines, researchers of one issue to enrol findings from other issues, and researchers using one theory to deploy the outcomes of other theories. In this economic metaphor, economic man comes into his own, as the middle-man between importers and exporters. The language of strict rationality is then not so much a hierarchically superior universal as a pragmatic and epistemically flat lingua franca. In so far as it is necessary for intra- and inter-disciplinary exchanges, as typified in the study of annual reports, it is also necessarily a pidgin. Now, however flexible a pidgin is, it cannot convey all the complexities of the distinct languages it bridges. The same is true for inter-subdisciplinary and inter-disciplinary usage. The pidgin of rationality in the language of economic man cannot include the rich nuances that are coded into the dialects emergent in disciplinary traditions. Any translation into and out of the pidgin is bound to be limited, at best, and perhaps even an outright betrayal. As Callon (1986, p. 231) stressed, tradutore, tradittore. Again this means that while any researcher must have come across that process, judgement of the results is another matter. Few writers have the intimate knowledge of different disciplines that is required to assess the degree of betrayal. For example, we take the fact that the studies we have surveyed here survived the normal editing and reviewing processes to be published at all as evidence that the translations in them were acceptable to their respective readerships. But we are in no position to be judgemental about accounting on that basis, or of the other disciplines we have noted, even though we are confident that different translations would appear in other disciplinary contexts. Given our own disciplinary backgrounds, we can show the process, but must remain agnostic about its specific outcomes in other disciplines. The agnosticism that follows from our remodelling of economic man is preferable to the exclusions that accompany judgements from a particular disciplinary stance. There is no need for disciplinary cleansing. Abandonment of the traditions that give disciplines their richness would be too high a price to pay for cross-disciplinary work. This further implies that more can be gained from valuing disciplinary differences than from derogating them. The more distinct and complex that disciplines are, the richer the rationalised pidgin that can emerge from the selective, partial and opportunistic enrolment across their boundaries that we have shown here and that is so characteristic of the social

AAAJ 15,4

470

sciences. Our own partial survey of research into corporate annual reports is not atypical. Just as Granovetter (1973) saw a strength in ‘‘weak ties’’, we hold that one strength of inter- and intra-disciplinary boundaries is their inevitable permeability. On the reflexive logic and on the ‘‘epistemic flattening’’ that we have followed from STS and from Quattrone (2000), moreover, this effect should recur in the world studied in the social sciences. If the bricolage and the use of a rationalised pidgin that students of annual reports display in their own practice are fractals of organisational activity more generally, then the image of the entrepoˆt is more than a metaphor. Like disciplinary researchers, any corporate practitioner brings to the marketplace a knowledge that is shaped through particular analogical models and particular concrete problem solutions, and communicates there through particular symbolic generalisations that are necessarily stunted in translation. Their partners for exchange in the marketplace are so diverse that their use of the pidgin of economic man is more a symbol of their willingness to participate and a guarantee of at least a partial understanding than any yielding to imposed uniformity. Since corporate practitioners face more disparate readerships than do academic researchers, the effects that we have found in the pragmatically intra-, inter- and transdisciplinary study of corporate annual reports apply a fortiori to the production and dissemination of the reports and to the activity reported in them. Conclusion Since the study of corporate annual reports is a subfield both in accounting research and across the social sciences they are a useful focus for a study of disciplinary activity. Since the results of research into the reports are diverse and even contradictory, and since this appears to match Quattrone’s (2000) diagnosis of ‘‘fragmentation’’ in accounting research, we framed this partial survey of the field as an empirical response to his call for a constructivist, evolutionary, reflexive and trans-disciplinary perspective that would restore a lost unity. Our reading moved through alternating layers of unity and disunity. While annual reports do have the generic sameness ensured by their fulfilment of regulatory demands, the disciplinary differences brought to their study, and the political, theoretical, methodological and national differences as well, mean that evidence on them is as disparate as Quattrone (2000) suggested. A degree of uniformity is still evident, however, in the use of that rationalised and rationalising economic man whom Quattrone (2000) had treated under ‘‘methodological individualism’’. This in turn generates further fragmentation, in conceptual uncertainty. We pointed especially to the common assumption, on one hand, that only individual human subjects can be agents, and, on the other, to the common attribution of agency to the non-human and collective subjects of corporations. Thus far, then, we had agreed with Quattrone (2000), for the fragmentation in this one subfield of accounting research did match his diagnosis of the field as a whole. We disagreed, however, that this state was pathological, suspecting

that Quattrone (2000) belied his call for reflexivity by looking for safe ground Economic man beyond the complexity he had described. To develop that point, we introduced and disciplinary the science and technology studies that are cognate to studies of economicoboundaries organisational activity, and used them to work our way back through the layers of unity and disunity. STS yields, first, a sense of agency that clears the blockage of methodological individualism, and then a sense of rationalities in 471 tension that shows disciplinary differences as more an advantage than a hindrance. Once economic man is modelled in these terms, his rationalised language appears as a pidgin that allows members of different disciplines and sub-disciplines to communicate – which, contra-Quattrone (2000), they constantly and demonstrably do – while retaining the complexity of their distinct disciplinary usage. Since richer disciplinary language in turn enriches the pidgin, there are better grounds for strengthening disciplinary boundaries than for removing them. Finally, we held that the study of corporate annual reports is always already reflexive: researchers display in their own disciplinary practice what corporate practitioners display in their activities. Those executives and their delegates who prepare annual reports face the same difficulty as do members of disciplines in deriving rationalised justifications. They face an even more intense difficulty in communicating across corporate boundaries to diverse and fuzzily defined audiences. In that sense, our response to Quattrone (2000) in the form of a case-study of the intra-, inter- and trans-disciplinary study of corporate annual reports is also programmatic. We can generalise from our own experience as a lay audience to accounting research to corporate communication with lay audiences at large. We aim to use the strategy from STS that we have outlined here in our own research: specifically on the writing and reading of annual reports; and more generally on the emergence of corporate rationality. For the moment, we have shown one way to enrol the strengths of economic man and of disciplinary boundaries in an approach to the permanent puzzle of the rationalisation of corporate rationality that is at once constructivist, evolutionary and reflexive. References Abrahamson, E. and Hambrick, D.C. (1997), ‘‘Attentional homogeneity in industries: the effect of discretion’’, Journal of Organizational Behavior, Vol. 18, pp. 513-32. Abrahamson, E. and Park, C. (1994), ‘‘Concealment of negative organizational outcomes: an agency theory perspective’’, Academy of Management Journal, Vol. 37, pp. 1302-34. Anderson, R.H. and Epstein, M.J. (1996), The Usefulness of Corporate Annual Reports to Shareholders in Australia, New Zealand and the United States: An International Comparison, JAI Press, Greenwich, CT. Arnold, P. and Hammond, T. (1994), ‘‘The role of accounting in ideological conflict: lessons from the South African divestment movement’’, Accounting, Organizations and Society, Vol. 19, pp. 111-26. Ashmore, M. (1989), The Reflexive Thesis: Wrighting Sociology of Scientific Knowledge, University of Chicago Press, Chicago, IL. Bartlett, S.A. and Chandler, R.A. (1997), ‘‘The corporate report and the private shareholder: Lee and Tweedie 20 years on’’, British Accounting Review, Vol. 29, pp. 245-61.

AAAJ 15,4

472

Becher, T. (1989), Academic Tribes and Territories: Intellectual Enquiry and the Cultures of Disciplines, The Society for Research into Higher Education/Open University Press, Milton Keynes. Bettman, J.R. and Weitz, B.A. (1983), ‘‘Attributions in the board room: causal reasoning in corporate annual reports’’, Administrative Science Quarterly, Vol. 28, pp. 165-83. Biagioli, M. (Ed.) (1999), The Science Studies Reader, Routledge, New York, NY. Bloor, D. (1976/1991), Knowledge and Social Imagery, 2nd ed., University of Chicago Press, Chicago, IL. Bowman, E.H. (1978), ‘‘Strategy, annual reports, and alchemy’’, California Management Review, Vol. 20 No 3, pp. 64-71. Bowman, E.H. (1982), ‘‘Risk seeking by troubled firms’’, Sloan Management Review, Vol. 23 No. 4, pp. 33-42. Bowman, E.H. (1984), ‘‘Content analysis of annual reports for corporate strategy and risk’’, Interfaces, Vol. 14 No. 1, pp. 61-71. Buhr, N. (1998), ‘‘Environmental performance, legislation and annual report disclosure: the case of acid rain and Falconbridge’’, Accounting, Auditing & Accountability Journal, Vol. 11, pp. 163-90. Burrell, G. and Morgan, G. (1979), Sociological Paradigms and Organizational Analysis: Elements of the Sociology of Corporate Life, Heinemann, London. Callon, M. (1986), ‘‘Some elements of a sociology of translation: domestication of the scallops and the fishermen of St Brieuc Bay’’, in Law, J. (Ed.), Power, Action and Belief: A New Sociology of Knowledge?, Routledge & Kegan Paul, London. Callon, M. (Ed.) (1998), The Laws of the Markets, Blackwell, Oxford. Callon, M. (1999), ‘‘Actor-network theory – the market test’’, in Law, J. and Hassard, J. (Eds), Actor Network Theory and After, Blackwell, Oxford. Callon, M. and Latour, B. (1981), ‘‘Unscrewing the big leviathan: how actors macro-structure reality and how sociologists help them to do so’’, in Knorr-Cetina, K. and Cicourel, A. (Eds), Advances in Social Theory and Methodology: Toward an Integration of Micro- and Macrosociologies, Routledge & Kegan Paul, Boston, MA. Callon, M. and Law, J. (1995), ‘‘Agency and the hybrid collectif’’, South Atlantic Quarterly, Vol. 94 No. 2, pp. 481-507. Callon, M. and Law, J. (1997), ‘‘After the individual in society: lessons on collectivity from science, technology and society’’, Canadian Journal of Sociology, Vol. 22, pp. 165-82. Campbell, D.J. (2000), ‘‘Legitimacy theory or managerial reality construction? Corporate social disclosure in Marks and Spencer Plc corporate reports, 1969-1997’’, Accounting Forum, Vol. 24 No. 1, pp. 81-100. Carruthers, B.E. (1995), ‘‘Accounting, ambiguity, and the new institutionalism’’, Accounting, Organizations and Society, Vol. 20, pp. 313-28. Chan, J.L. (1979), ‘‘Corporate disclosure in occupational safety and health: some empirical evidence’’, Accounting, Organizations and Society, Vol. 4, pp. 273-81. Clegg, S.R. (1988), Frameworks of Power, Sage, London. Coleman, J.S. (1990), Foundations of Social Theory, The Belknap Press of Harvard University Press, Cambridge MA. Courtis, J.K. (1995), ‘‘Readability of annual reports: Western versus Asian evidence’’, Accounting, Auditing & Accountability Journal, Vol. 8 No. 2, pp. 4-17. Courtis, J.K. (1998), ‘‘Annual report readability variability: tests of the obfuscation hypothesis’’, Accounting, Auditing & Accountability Journal, Vol. 11, pp. 459-71.

Czarniawska, B. (1997), Narrating the Organization: Dramas of Institutional Identity, University of Chicago Press, Chicago, IL. D’Aveni, R.A. and MacMillan, I.C. (1990), ‘‘Crisis and the content of managerial communications: a study of the focus of attention of top managers in surviving and failing firms’’, Administrative Science Quarterly, Vol. 35, pp. 634-57. Deegan, C. and Rankin, M. (1996), ‘‘Do Australian companies report environmental news objectively? An analysis of environmental disclosures by firms prosecuted successfully by the Environmental Protection Authority’’, Accounting, Auditing & Accountability Journal, Vol. 9 No. 2, pp. 50-67. Dev, S. (1974), ‘‘Statements of company prospects’’, Accounting and Business Research, Vol. 4, pp. 270-4. DiMaggio, P.J. and Powell, W.W. (1991), ‘‘Introduction’’, in Powell, W.W. and DiMaggio, P.J. (Eds), The New Institutionalism in Organizational Analysis, University of Chicago Press, Chicago, IL. Epstein, M.J. and Freedman, M. (1994), ‘‘Social disclosure and the individual investor’’, Accounting, Auditing & Accountability Journal, Vol. l7 No. 4, pp. 94-109. Epstein, M.J. and Pava, M.L. (1993), The Shareholder’s use of Corporate Annual Reports, JAI Press, Greenwich, CT. Fiol, C.M. (1989), ‘‘A semiotic analysis of corporate language: organizational boundaries and joint venturing’’, Administrative Science Quarterly, Vol. 34, pp. 277-303. Fiol, C.M. (1995), ‘‘Corporate communications: comparing executives’ private and public statements’’, Academy of Management Journal, Vol. 38, pp. 522-36. Foucault, M. (1975/1977/1991), Discipline and Punish: The Birth of the Prison, Penguin, London. Freedman, M. (1993), ‘‘Accounting and the reporting of pollution information’’, in Lehman, C.R. (Ed.), Advances in Public Interest Accounting, vol. 5, JAI Press, Greenwich, CT. Freedman, M. and Jaggi, B. (1986), ‘‘An analysis of the impact of corporate pollution disclosures included in annual financial statements on investors’ decisions’’, in Neimark, M. (Ed.), Advances in Public Interest Accounting, Vol. 1, JAI Press, Greenwich, CT. Freedman, M. and Wasley, C. (1990), ‘‘The association between environmental performance and environmental disclosure in annual reports and 10Ks’’, in Neimark, M., Merino, B. and Tinker, T. (Eds), Advances in Public Interest Accounting, Vol. 3, JAI Press, Greenwich, CT. Fuchs, S. (1992), The Professional Quest for Truth: A Social Theory of Science and Knowledge, State University of New York Press, Albany, NY. Galison, P. (1999), ‘‘Trading zone: coordinating action and belief’’, in Biagioli, M. (Ed.), The Science Studies Reader, Routledge, New York and London. Gamble, G.O., Hsu, K., Jackson, C. and Tollerson, C.D. (1996), ‘‘Environmental disclosures in annual reports: an international perspective’’, International Journal of Accounting, Vol. 31, pp. 293-331. Gamble, G.O., Hsu, K., Kite, D. and Radtke, R. (1995), ‘‘Environmental disclosures in annual reports and 10Ks: an examination’’, Accounting Horizons, Vol. 9, No. 3, pp. 34-54. Geertz, C. (1983a/1993), ‘‘Blurred genres: the refiguration of social thought’’, in Local Knowledge: Further Essays in Interpretive Anthropology, Fontana, London. Geertz, C. (1983b/1993), ‘‘The way we think now: toward an ethnography of modern thought’’, Local Knowledge: Further Essays in Interpretive Anthropology, Fontana, London. Granovetter, M. (1973), ‘‘The strength of weak ties’’, American Journal of Sociology, Vol. 78, pp. 1360-80.

Economic man and disciplinary boundaries 473

AAAJ 15,4

Graves, O.F., Flesher, D.L. and Jordan, R.E. (1996), ‘‘Pictures and the bottom-line: the television epistemology of US annual reports’’, Accounting, Organizations and Society, Vol. 21, pp. 57-88. Gray, R., Kouhy, R. and Lavers, S. (1995), ‘‘Corporate social and environmental reporting: a review of the literature and a longitudinal study of UK disclosure’’, Accounting, Auditing & Accountability Journal, Vol. 8 No. 2, pp. 47-77.

474

Guthrie, J. and Parker, L.D. (1989), ‘‘Corporate social reporting: a rebuttal of legitimacy theory’’, Accounting and Business Research, Vol. 76, pp. 343-52. Guthrie, J. and Parker, L.D. (1990), ‘‘Corporate social disclosure practice: a comparative international analysis’’, in Neimark, M., Merino, B. and Tinker, T. (Eds), Advances in Public Interest Accounting, Vol. 3, JAI Press, Greenwich, CT. Harte, G., Lewis, L. and Owen, D. (1991), ‘‘Ethical investment and the corporate reporting function’’, Critical Perspectives on Accounting, Vol. 2, pp. 227-53. Henriques, I. and Sadorsky, P. (1999), ‘‘The relationship between environmental commitment and managerial perceptions of stakeholder importance’’, Academy of Management Journal, Vol. 42, No. 1, pp. 87-99. Herremans, I.M., Akathaporn, P. and McInnes, M. (1993), ‘‘An investigation of corporate social responsibility reputation and economic performance’’, Accounting, Organizations and Society, Vol. 18, pp. 587-604. Hines, R.D. (1982), ‘‘The usefulness of annual reports: the anomaly between the efficient markets hypothesis and shareholder surveys’’, Accounting and Business Research, Vol. 12, pp. 296-309. Hoskin, K.W. and Macve, R.H. (1986), ‘‘Accounting and the examination: a genealogy of disciplinary power’’, Accounting, Organizations and Society, Vol. 13 No. 1, pp. 37-73. Hoskin, K.W. and Macve, R.H. (1993), ‘‘Accounting as discipline: the overlooked supplement’’, in Messer-Davidow, E., Shumway, D.R. and Sylvan, D.J. (Eds), Knowledges: Historical and Critical Studies in Disciplinarity, University Press of Virginia, Charlottesville, VA. Ind, N. (1997), The Corporate Brand, Macmillan Business, London. Ingram, R.W. and Frazier, K.B. (1980), ‘‘Environmental performance and corporate disclosure’’, Journal of Accounting Research, Vol. 18, pp. 614-22. Jasanoff, S., Markle, G.A., Petersen, J.C. and Pinch, T. (Eds) (1995), Handbook of Science and Technology Studies, Sage, Thousand Oaks, CA. Jones, M.J. and Shoemaker, P.A. (1994), ‘‘Accounting narratives: a review of empirical studies of content and readability’’, Journal of Accounting Literature, Vol. 13, pp. 142-84. Kabanoff, B. and Holt, J. (1996), ‘‘Changes in the espoused values of Australian organizations 1986-1990’’, Journal of Organizational Behavior, Vol. 17, pp. 201-19. Kuhn, T.S. (1970/1994), The Structure of Scientific Revolutions, revised edition, University of Chicago Press, Chicago, IL. Kuhn, T.S. (1974/1977), ‘‘Second thoughts on paradigms’’ in The Essential Tension: Selected Studies in Scientific Tradition and Change, University of Chicago Press, Chicago, IL. Latour, B. (1987), Science in Action: How to Follow Scientists and Engineers Through Society, Harvard University Press, Cambridge, MA. Latour, B. (1991a/1993), We Have Never Been Modern, Harvard University Press, Cambridge, MA. Latour, B. (1991b), ‘‘Technology is society made durable’’, in Law, J. (Ed.), A Sociology of Monsters: Essays on Power, Technology and Domination, Routledge, London.

Law, J. (1986), ‘‘On the methods of long-distance control: vessels, navigation and the Portuguese route to India’’, in Law, J. (Ed.), Power, Action and Belief: A New Sociology of Knowledge?, Routledge & Kegan Paul, London. Law, J. (1994), Organizing Modernity, Blackwell, Oxford. Lee, N. and Hassard, J. (1999), ‘‘Organization unbound: actor-network theory, research strategy and institutional flexibility’’, Organization, Vol. 6, pp. 391-404. Lee, T.A. (1990), ‘‘Introduction: economic reality, faithful representation and accounting numerology’’, in Lee, T.A. and Tweedie, D.P. (Eds), Shareholder Use and Understanding of Financial Information, Garland, New York and London. Lee, T.A. and Tweedie, D.P. (Eds) (1990), Shareholder Use and Understanding of Financial Information, Garland, New York, NY. Lowy, I. (1992), ‘‘The strength of loose concepts – boundary concepts, federative experimental strategies and disciplinary growth: the case of immunology’’, History of Science, Vol. 30, pp. 371-96. Lynch, M. (2000), ‘‘Against reflexivity as an academic virtue and source of privileged knowledge’’, Theory, Culture and Society, Vol. 17 No. 3, pp. 26-54. Lynch, M. and Bogen, D. (1997), ‘‘Sociology’s asociological ‘core’: an examination of textbook sociology in light of the sociology of scientific knowledge’’, American Sociological Review, Vol. 62, pp. 481-93. Macintosh, N.B. (1990), ‘‘Annual reports in an ideological role: a critical theory analysis’’, in Cooper, D.J. and Hopper, T.M. (Eds), Critical Accounts, Macmillan, London. McConnell, D., Haslem, J.A. and Gibson, V.R. (1986), ‘‘The President’s letter to stockholders: a new look’’, Financial Analysts Journal, Vol. 42 No. 5, pp. 66-70. McKinstry, S. (1996), ‘‘Designing the annual reports of Burton PLC from 1930 to 1994’’, Accounting, Organizations and Society, Vol. 21, pp. 89-111. Marino, A.J. Jr (1995), ‘‘Separating your annual report from the herd’’, Public Relations Quarterly, Vol. 40 No. 2, pp. 44-7. Miller, P. and O’Leary, T. (1993), ‘‘Accounting experience and the politics of the product: economic citizenship and the modes of corporate governance’’, Accounting, Organizations and Society, Vol. 18, pp. 187-206. Milne, M.J. and Adler, R.W. (1999), ‘‘Exploring the reliability of social and environmental disclosures content analysis’’, Accounting, Auditing & Accountability Journal, Vol. 12, pp. 237-56. Nagy, R.A. and Obenberger, R.W. (1994), ‘‘Factors influencing individual investor behavior’’, Financial Analysts Journal, Vol. 50 No. 4, pp. 63-8. Neimark, M. and Tinker, T. (1986), ‘‘The social construction of management control systems’’, Accounting, Organizations and Society, Vol. 11, pp. 369-95. Neu, D., Warsame, H. and Pedwell, K. (1998), ‘‘Managing public impressions: environmental disclosures in annual reports’’, Accounting, Organizations and Society, Vol. 23, pp. 265-82. Niskala, M. and Pretes, M. (1995), ‘‘Environmental reporting in Finland: a note on the use of annual reports’’, Accounting, Organizations and Society, Vol. 20, pp. 457-66. Olson, M. (1965), The Logic of Collective Action, Harvard University Press, Cambridge, MA. Owen, D. (Ed.) (1992), Green reporting: Accountancy and the Challenge of the Nineties, Chapman & Hall, London. Palmer, I., Kabanoff, B. and Dunford, R. (1997), ‘‘Managerial accounts of downsizing’’, Journal of Organizational Behavior, Vol. 18, pp. 623-39. Parker, L.D. (1982), ‘‘Corporate annual reporting: a mass communication perspective’’, Accounting and Business Review, Vol. 12, pp. 279-86.

Economic man and disciplinary boundaries 475

AAAJ 15,4

476

Patten, D. M. (1991), ‘‘Exposure, legitimacy and social disclosure’’, Journal of Accounting and Public Policy, Vol. 10, pp. 297-308. Pickering, A. (1995), The Mangle of Practice: Time, Agency and Science, University of Chicago Press, Chicago, IL. Preston, A.M., Wright, C. and Young, J.J. (1996), ‘‘Imag[in]ing annual reports’’, Accounting, Organizations and Society, Vol. 21, pp. 113-37. Quattrone, P. (2000), ‘‘Constructivism and accounting research: towards a trans-disciplinary perspective’’, Accounting, Auditing & Accountability Journal, Vol. 13 No. 2, pp. 130-55. Roberts, R. (1992), ‘‘Determinants of corporate social responsibility disclosure: an application of stakeholder theory’’, Accounting, Organizations and Society, Vol. 17, pp. 595-612. Rockness, J. and Williams, P.F. (1988), ‘‘A descriptive study of social responsibility mutual funds’’, Accounting, Organizations and Society, Vol. 13, pp. 397-411. Salancik, G.R. and Meindl, J.R. (1984), ‘‘Corporate attributions as strategic illusions of management control’’, Administrative Science Quarterly, Vol. 29, pp. 238-54. Schein, E.H. (1984), ‘‘Coming to a new awareness of organizational culture’’, Sloan Management Review, Vol. 25 No. 1, pp. 3-16. Shapin, S. (1995), ‘‘Here and everywhere: sociology of scientific knowledge’’, Annual Review of Sociology, Vol. 21, pp. 289-321. Simon, H.A. (1957), Models of Man, Social and Rational: Mathematical Essays on Rational Human Behavior in a Social Setting, Wiley, New York, NY. Simon, H.A. (1978), ‘‘Rationality as process and as product of thought’’, American Economic Review, Vol. 68 No. 2, pp. 1-16. Staw, B.M., McKechnie, P.I. and Puffer, S.M. (1983), ‘‘The justification of organizational performance’’, Administrative Science Quarterly, Vol. 28, pp. 582-600. Steele, A. (1982), ‘‘The accuracy of chairmen’s non-quantified forecasts: an exploratory study’’, Accounting and Business Research, Vol. 12, pp. 215-30. Tennyson, B.M., Ingram, R.W. and Dugan, M.T. (1990), ‘‘Managing the information content of narrative disclosures in explaining bankruptcy’’, Journal of Business Finance & Accounting, Vol. 17, pp. 391-410. Tilt, C.A. (1994), ‘‘The influence of external pressure groups on corporate social disclosure: some empirical evidence’’, Accounting, Auditing & Accountability Journal, Vol. 7 No. 4, pp. 47-72. Tinker, T. and Neimark, M. (1987), ‘‘The role of annual reports in gender and class contradictions at General Motors: 1917-1976’’, Accounting, Organizations and Society, Vol. 12, pp. 71-88. Tinker, T., Lehman, C. and Neimark, M. (1991), ‘‘Falling down the hole in the middle of the road: political quietism in corporate social reporting’’, Accounting, Auditing & Accountability Journal, Vol. 4 No. 2, pp. 28-55. Ullmann, A.A. (1985), ‘‘Data in search of a theory: a critical examination of the relationships among social performance, social disclosure, and economic performance’’, Academy of Management Review, Vol. 10, pp. 540-57. von Gierke, O. (1900/1968), Political Theories of the Middle Ages, Cambridge University Press, Cambridge. Walden, W.D. and Schwartz, B.N. (1997), ‘‘Environmental disclosures and public policy pressures’’, Journal of Accounting and Public Policy, Vol. 16, pp. 125-54. White, R. and Hanson, D. (2000), ‘‘Rationality and rhetoric in the corporate world: the corporate annual report as an Aristotelian genre’’, Prometheus, Vol. 18, pp. 303-17. White, R. and Hanson, D. (2002), ‘‘Corporate self, corporate reputation and corporate annual reports: re-enrolling Goffman’’, Scandinavian Journal of Management, Vol. 18, pp. 285-301.

Whitley, R. (1984), The Intellectual and Social Organization of the Sciences, Clarendon Press, Oxford. Wilmshurst, T.D. and Frost, G.R. (2000), ‘‘Corporate environmental reporting: a test of legitimacy theory’’, Accounting, Auditing & Accountability Journal, Vol. 13 No. 1, pp. 10-26. Wiseman, J. (1982), ‘‘An evaluation of environmental disclosures made in corporate annual reports’’, Accounting, Organizations and Society, Vol. 7, pp. 53-63. Wolfe, R. (1991), ‘‘The use of content analysis to assess corporate responsibility’’, in Post, J.E. (Ed.), Research in Corporate Social Performance and Policy, Vol. 12, JAI Press, Greenwich, CT. Woolgar, S. (Ed.) (1988a), Knowledge and Reflexivity: New Frontiers in the Sociology of Knowledge, Sage, London. Woolgar, S. (1988b), Science: The Very Idea, Ellis Horwood and Tavistock, Chichester and London. Woolgar, S. and Pawluch, D. (1985), ‘‘Ontological gerrymandering: the anatomy of social problems explanations’’, Social Problems, Vol. 32, pp. 214-27.

In memoriam The editors and publisher of Accounting, Auditing & Accountability Journal regret to announce that one of the authors, Dr Robert (Bob) White, a lecturer in sociology and, in his spare time, an exceptionally well read philosopher and scientist, died before publication of this article. He will be sadly missed.

Economic man and disciplinary boundaries 477

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

AAAJ 15,4

Corporate annual reports: research perspectives used Patricia Stanton and John Stanton

478 Received June 2001 Revised January 2002 Accepted April 2002

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 478-500. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440568

Newcastle Business School, Faculty of Business and Law, University of Newcastle, Newcastle, Australia Keywords Company reports, Corporate image, Accountability, Marketing, Research Abstract Corporate annual reports are viewed through the lens of researchers of these documents. The aims are to obtain insight into how researchers view annual reports; to ascertain how the different ways of seeing the annual report relate to each other; and to draw out the gaps in this diverse research in a continuing attempt to understand its role and purpose. Selective examination of a decade of corporate annual report research (1990-2000) reveals how researchers have sought to find visibility and meaning. Few studies address the document as a whole, in terms of the integration of the messages between the various parts of the report. Explanation of the changing structure and content of annual reports remains divided, largely because of the differing perspectives of researchers. They have revealed diversity in the ways of seeing the annual report and a tension in understanding its overall purpose and role.

Introduction A corporate annual report (CAR) can be viewed as a formal public document produced by public companies largely as a response to the mandatory corporate reporting requirements existing in most Western economies. Beyond this legal imperative, Hopwood (1996, p. 55) argues that CARs have become a highly sophisticated product of the corporate design environment, the main purpose of which is to proactively construct a particular visibility and meaning rather than revealing ‘‘what was there’’. This argument mirrors the sentiment of Hines (1988, p. 257) that, in communicating reality, reality is constructed: ‘‘We create a picture of an organization . . . and on the basis of that picture . . . people think and act. And by responding to that picture of reality, they make it so’’. Widely acknowledged as a means by which companies communicate with their various publics, the modern CAR uses the tools of management, marketing and communication theory to construct a picture of the organisation. Amongst researchers examining the role and purpose of the modern annual report, ways of seeing that picture differ. There is the information dispensing picture of the CAR providing management with a vehicle of communication with customers, shareholders, employees, suppliers, media, and government (Judd and Tims, 1991). There is the picture of the CAR communicating the ‘‘personality and philosophy of the firm’’ (Anderson and Imperia, 1992, p. 113). It is viewed also as a sophisticated marketing tool (Bekey, 1990) that can impart a particular organisational image for relevant publics (Neu et al., 1998). Following Hines (1988) and Hopwood (1996), the visibilities and meanings constructed by corporations in their annual reports are viewed through the lens of researchers of these documents. The objectives are to obtain insight into how researchers view annual reports through the frameworks they use to inform or

guide their analysis; to ascertain how the different frameworks relate to each Corporate annual other; and to draw out the gaps in this diverse research in the continuing reports: research attempt to understand the annual report’s role and purpose. perspectives The paper proceeds with a section outlining recent changes noted in the design and content of CARs. This is followed by the method adopted to examine the literature. The findings are presented in terms of the perspective 479 adopted. ‘‘Ways of seeing’’ are discussed. Implications drawn conclude the paper. Changes in design and content As formal communication documents, annual reports commonly comprise quantitative information, narratives, photographs and graphs. A certain order has evolved in which to disclose these and the required accounting information. CARs are commonly divided into two sections. The statutory required financial statements are usually assigned to a rear section, with a much larger up-front section containing mainly non-statutory matters. The division is emphasised by the use of different types of paper and colours (Marino, 1995). The instigation of content changes has been partly mandatory, partly voluntary. In the UK, Bartlett and Jones (1997, p. 61) observed that, between 1970 and 1990, total mandatory content increased rapidly as a result of changing demands from several regulatory bodies. However, voluntary disclosures, the amplification of data included in financial statements, expanded as the corporate report moved towards being a public relations document. The dominant theme of these voluntary disclosures was corporate social reporting (CSR), being largely concerned with a corporation’s interactions with the natural environment, employees, communities and customers. Disclosures are increasing, particularly among larger companies within environmentally sensitive industries (Deegan and Gordon, 1996). By theme, CSR disclosures appear reasonably consistent (Robertson and Nicholson, 1996) across all countries (Hackston and Milne, 1996; Roberts, 1991), with disclosures relating to human resources, environment and community, in that order, receiving most attention (Zeghal and Sadrudin, 1990). Since 1965, annual reports have grown in volume (Lee, 1994), although their size appears to have stabilised (Marino, 1995). Cover treatments focus on colour photographs (Marino, 1995). Narratives are giving way to pictorial forms, with an increasing emphasis on product-related matters designed to influence stakeholders. Design consultants, employed as image managers, create explicit images (e.g. company logos) and complement them with high-resolution colour, merged or fused pictures and narrative messages (Lee, 1994). Most reports carry a theme, commonly related to customers or employees. The front half is organised by line of business, with the CEO’s letter prominent (Marino, 1995). For larger companies, preparation of the annual report is increasingly the domain of external design agencies. In the UK, in 1999 only 6 percent of the FTSE 250 companies designed their annual reports in-house. This represents a fall of 43 percent from the number of companies designing their reports

AAAJ 15,4

480

in-house in 1995 (Valentine, 1999). Some agencies may undertake tasks spanning corporate identity, brochures, interiors, multimedia, as well as annual reports (Design Week, April 1999). For CARs, the design function can vary from ‘‘creating pretty picture books’’ to creating ‘‘a strategy for each annual report’s design and content by analyzing a company’s problems and what’s happening in that industry’’ (The CEO of Boller Coates and Neu, a USA award-winning design company, cited by Gatland, 1997). Bolton (1989, cited by Preston et al., 1996, p. 263) claims that the changes ‘‘threaten to overwhelm substantive discourse’’ by replacing it with non-contentious images and quotable statements. Even the growth of accounting narrative, although seen as a vital part of modern corporate reporting, is described by Jones (1996b, p. 42) as ‘‘unlikely to be impartial and . . . likely to be used by managements to give a more favourable impression of corporate performance than is warranted’’. Method The aim of this paper is to ascertain how researchers view corporate annual reports; to ascertain their ways of seeing them as well as what they saw. A researcher perspective is adopted because researchers are assumed to be more aware of the techniques employed in reports. Nevertheless, in interpreting CARs, researchers are perceiving or constructing a reality that attributes a particular role or purpose for annual reports. Examining the focus of researchers can show the diversity of constructions. Relationships between constructions and whether they further understanding of the role and purpose of CARs can be addressed. This paper seeks to capture contemporary views, by focusing on an interval, the last decade of the twentieth century. Over this decade, research into annual reports has been consistent, and has emanated from widely differing disciplines, including accounting, marketing and communications. Thus, this period is sufficient to bring out the richness and diversity of research. To locate such research, a keyword search of ‘‘annual reports’’ was made using both ABI Inform and the Accounting and Taxation Index of the period 1990 to 2000. All resulting references were reviewed for their relevance to the study, so that only those dealing with an empirical investigation of some aspect of a corporate annual report were included. Many of the results dealt with assertions about what an annual report should be. These, together with anonymous articles, are omitted from the study. While there may be research omitted, the diversity revealed by the keyword search should not be compromised. Given the wealth of research, a systematic approach is aided by classifying the perspectives of researchers in their viewing of the annual report. ‘‘Perspective’’ relates to the way researchers viewed the part(s) of the annual reports they examined to either the report as a whole, or to the context of the report, such as an overarching objective or schema. For example, researchers utilising a political economy perspective view reports as embedded in the complex relationships of society as a whole, whereas those utilising narrative

theory embed reports in less complex relationships between senders and Corporate annual receivers. Perspectives were derived inductively. Both researchers closely read reports: research each article and independently categorised it in the first instance by claims perspectives within the article to the use of a particular framework, such as legitimacy theory or narrative theory, for appraising the annual report. Where no claims to a particular framework were made, the article was assigned on the basis of 481 an apparent consistency with a particular perspective. Consistency was determined by the use of keywords or phrases associated with particular frameworks. Separating perspectives leads to the problem of boundaries. For example, as a communication instrument, an annual report may be examined from the viewpoint of communication effectiveness, or from the viewpoint of a marketing communication tool. The emphasis given in a study determines its placement. Corporate social responsibility disclosure is sometimes examined from a political economy perspective (Adams and Harte, 1998) but in most cases is treated as an element of the legitimation process (Campbell, 2000) so that it is not treated as a separate perspective. Studies examining annual reports from multiple perspectives (e.g. Adams and Harte, 1998) were allocated on their dominant perspective. The independently derived lists were compared. Where differences arose, an academic colleague was asked to read and assign the article. The residual category was used for those studies employing ‘‘tools’’ rather than a perspective, and for those studies which adopted a perspective, such as grounded theory, which was not used in any other examined study. Where possible, the results of ‘‘other’’ studies were used to enhance findings allocated to particular perspectives. The database search revealed 70 useable studies. Typically, researchers did not examine annual reports in their entirety, but concentrated on a particular aspect such as photographs or environmental information, wherever that aspect might be found with the report. The focus of the studies has been extremely broad, with no one area appearing to dominate the interest of cited authors. Numbers of studies identified in each perspective are given in Table I[1], with more detail of the subject of analysis and focus provided in Table II. Assigning studies where authors used particular frameworks that directed attention to the means of communication within annual reports was fraught with difficulties[2]. The trend noted earlier, to outsource annual report preparation to designers, is probably linked to the realisation that annual reports are being used to manage images and as marketing tools. Studies espousing a marketing perspective were differentiated from those focusing on Image management Marketing Organisational legitimacy Political economy Accountability Other

20 11 18 2 10 9

Table I. List of perspectives and numbers of studies

AAAJ 15,4

Perspective

Subject of analyis

Author(s)

Focus

Image management

Annual reports Annual reports Annual reports Narrative Narrative

Lee, 1994 Stocks, 1995 Buxton, 2000 Smith and Taffler, 1992 Subramanian et al., 1993

Narrative Narrative Narrative Narrative

Aerts, 1994 Jones and Shoemaker, 1994 Courtis, 1995 Clarke, 1997

Shape and structure Messages, design CSR, style Readability Readability, performance Verbal behaviour Review

Narrative Narrative Narrative

Thomas, 1997 Courtis, 1998 Jameson, 2000

Graphs Graphs Graphs Photographs Photographs

Beattie and Jones, 1992 Mather et al., 1996 Beattie and Jones, 1997 Anderson and Imperia, 1992 McKinstry, 1996

Photographs Photographs Photographs

Preston et al., 1996 Simpson, 1999 Douglis, 2000

Annual reports Marketing focus

Bekey, 1990 Droge et al., 1990

CEO letters CEO letters

Judd and Tims, 1991 Kohut and Segars, 1992

Annual reports Annual reports

Rezaee and Porter, 1993 Holiday, 1994

Annual reports

Annual reports

Herremans and Ryan, 1995 Marino, 1995 Leuthesser and Kohli, 1997 Brink, 1998

Annual reports

Mitchell, 1998

CSR disclosures

Guthrie and Parker, 1990 Patten, 1991

482

Marketing

Annual reports Mission statements

Legitimacy

CSR disclosures Table II. Studies of annual reports, 1990-2000

Readability Chairmen’s statements Linguistic structures Readability Choices, interpretation Use, abuse Australian CARs UK, US CARs Females Burton plc, 1930 to 1994 Specific images Images Uses Format, audience Business journalists’ comments Strategy, customers Strategy, communication Content, purpose Audiences, segmenting Marketing, firm value Common features Use, targets, Effectiveness Key messages, target audiences Brand building Cross country comparison, CSR Environmental, corporate performance (continued)

Perspective

Subject of analyis

Author(s)

Focus

CSR CSR CSR CSR CSR

Neu and Wright, 1992 Patten, 1992 Tilt, 1994 Gray et al., 1995 Hackston and Milne, 1996 Deegan and Gordon, 1996 Deegan and Rankin, 1996 Robertson and Nicholson, 1996 Patten, 1997 Adams et al., 1998

Stigma management Environmental Pressure groups Environmental NZ CSR

disclosures disclosures disclosures disclosures disclosures

CSR disclosures CSR disclosures CSR disclosures CSR disclosures CSR disclosures CSR disclosures

Environmental CARs v. internal communications Quantity and quality Cross country comparison, Europe Environmental, print media Environmental Environmental Australian graphs Environmental incidents Marks and Spencer Plc, 1969-1997

CSR disclosures CSR disclosures Graphs CSR disclosures CSR disclosures

Campbell, 2000

Political economy

Photographs CSR disclosures

Graves et al., 1996 Adams and Harte, 1998

Television rhetoric Equal opportunities information

Accountability

Environmental disclosures CSR disclosures

Harte and Owen, 1991 Ness and Mirza, 1991

Disclosures

Cooke, 1992

Governance

Keasey and Wright, 1993 Ormrod and Cleaver, 1993 Abrahamson and Park, 1994 Anderson and Epstein, 1996 Fekrat et al., 1996 Ramsay and Hoad, 1997 Adams and Hossain, 1998

‘‘Good’’ disclosers, water companies, UK Environmental, oil industry, agency theory Size, stock market listing, industry type Editorial

Annual report CSR disclosures Disclosure Voluntary disclosures

483

Environmental

Brown and Deegan, 1998 Buhr, 1998 Neu et al., 1998 Beattie and Jones, 1999 Deegan et al., 2000

Financial information Disclosure

Corporate annual reports: research perspectives

Contracts President’s letters Shareholder needs Environmental Governance Life insurance firms, NZ (continued)

Table II.

AAAJ 15,4

484

Table II.

Perspective

Subject of analyis

Author(s)

Focus

Other

Financial disclosure CSR disclosures

Gibbons et al., 1990

Grounded theory

Ethical investment CSR disclosures

Ze’ghal and Ahmed, 1990 Harte et al., 1991 Roberts, 1991

Annual reports MD&A disclosures Presentation CSR disclosures

Arfin, 1993 Pava and Epstein, 1993 Beattie et al., 1996 Gamble et al., 1996

Disclosures

Bartlett and Jones, 1997

Environmental, industry comparison Investment funds Environmental, employee-related Interviews Forecasts Prowse update Environmental, international comparison Bulmers, 1970-1990

image management because they embed annual reports in corporate strategy aimed at targeted audiences and messages, rather than as a wider public relations document. Each perspective is reported on in turn. The image management perspective Although the role or purpose of an annual report is not necessarily addressed by all studies included in this category, their authors utilise various tools to identify the ways in which parts of the front half of annual reports are used to convey particular messages. Studies early in the decade used simple techniques, such as counting the number of words, photographs or graphs included in various sections of an annual report. Later, reports are viewed as communication documents of symbols selected and ordered to give meaning to a story. Researchers then focus on how communication can be used to achieve particular meanings and outcomes. These outcomes echo, so that by the end of the decade, studies explicitly identify with an image or impression management perspective in which annual reports or parts thereof are used to create favourable impressions. Most studies (see Table II) focus on text, photographs or graphs, although several studies addressed the annual reports as a whole, a subdivision which will be utilised to examine their findings. Utilising a style viewpoint, Lee (1994) examined a sample of annual reports over a 20-year period, while Marino (1995), a designer, used a 10-year period. McKinstry (1996), in contrast, examined one corporation’s annual reports over a 64-year period. All observed that design consultants were employed as image managers; that a certain order evolved in which to disclose voluntary messages and required statutory information. These changes suggested that corporations participate in ‘‘consumer engineering’’ via the annual report (Lee, 1994). McKinstry (1996, p. 109) lamented that the changes were turning the annual report into ‘‘a corporate public relations tool’’. Less informed shareholders were seen as likely to succumb to a glossy overlay, which was

capable of overriding the numerical and other statutory messages relegated to Corporate annual the rear of annual reports. While these studies see the CAR in broad terms, the reports: research intent of the public relations exercise and how the content achieves that perspectives purpose remain unanswered. Analysis of text Narratives are seen as cultural artifacts that include texts, images, spectacles and events which ‘‘tell a story’’ (Baal, 1997, cited by Jameson, 2000). A story has two elements – the universe of all possible materials from which a story can be made and the materials actually selected. Selection means that what is omitted from a CAR is as significant as what is included in the narrative (Chatham, 1978, cited by Jameson 2000; Adams and Harte, 1998; Buhr, 1998). However, researchers have focused on the materials selected, examining various narrative sections of annual reports in detail. Although linguistic approaches to the study of business writing are still developing (Thomas, 1997), the effectiveness of a CAR as text depends on the relationship of three components: (1) the substance or content of an idea; (2) the activity of communicating the idea from the sender to the receiver (interpersonal component); and (3) the relevance to the context of the situation. Within these components, the choice of verb structures, thematic structures, subjects, context and cohesion, and condensations determine meaning (Thomas, 1997). Linguistic theory provides a range of language choices and constructions that report preparers can use to pursue their goals without misinformation or complex language. Narrative theories focus on different sources of meaning, the shaping of the narrative, the way the reader interacts with the text, and the multiplicity of narrative voices (different narrators) and genres (different forms of narration such as epistles, lessons, sermonettes, essays, question-and-answer dialogues). Despite their different foci, each suggests that the way a story is told, as well as what it says, matters. ‘‘Myriad stories’’ (Jameson, 2000, p. 10) may be abstracted from underlying materials, shaped into discourse and manifested in a particular medium. Structure is integral: meaning derives from the way the raw materials are selected, shaped and manifested. To maintain public image and to protect management from criticism, variations in linguistic constructions from good years to bad were found. CARs were carefully written, but language was used to guide the reader to a particular interpretation (Thomas, 1997). Applied to an examination of one corporation’s CEO letters over a five year, mixed performance period, linguistic analysis showed that the letters suggested and implied, but did not lie (Thomas, 1997). Language was used to blur distinctions about the causes of poor performance, often through objectification, and to position the company in terms of its priorities, presenting

485

AAAJ 15,4

486

the corporation in a positive light. Jameson (2000) analysed how narration was used in annual reports of 200 US mutual fund corporations by dividing them into two samples, top return and mixed return performance. Narratives within the mixed return reports used a non-linear structure, dramatized ideas through contrasting narrators, embedded a variety of sub-genres, and complemented verbal with visual discourse (McKinstry, 1996) in order to make readers active participants in constructing the story. As such, readers’ responses would depend on their intellectual understanding as well as their emotional interaction with the narrator’s personae and visual symbolism. Other investigations of narratives have concentrated on whether elements in the writing help or hinder reader comprehension and thus have examined readability, and by implication, understandability (see Table II). Narratives should be readable and understandable (Courtis, 1995; Jones, 1996a). Readability may be manipulated by management in order to present information in as favourable a light as possible (Jones, 1996b), an observation supported by studies of CSR disclosures (e.g. Deegan and Gordon, 1996; Deegan and Rankin, 1996) and of forecasts (Pava and Epstein, 1993). Acknowledging the shortcomings of the instruments used to measure readability, both Courtis (1995) and Jones (1996a) found that readability levels were too difficult for the ‘‘average reader’’ to understand. A link between readability and corporate performance was found: narratives of ‘‘good’’ performers were easier to read than those of ‘‘poor’’ performers (Subramanian et al., 1993; Smith and Taffler, 1992). Good performers used ‘‘stronger’’ writing in their reports than poorer performers (Subramanian et al., 1993). Although Thomas (1997) found that language was used to blur attributions, others (Clarke, 1997; Aerts, 1994) found attributions to be biased. Aerts (1994) calls it a ‘‘hedonic bias’’: a general tendency to attribute anything negative to external, environmental causes and to attribute favourable outcomes to internal dispositional factors. This is combined with an accounting bias: a tendency to explain negative performances in technical accounting terms (Aerts, 1994) or in convoluted language (Jones, 1996a), while positive performances are explained in strict cause-effect terminology whereby management’s responsibility for them becomes clear (Aerts, 1994). Excuses, justifications and apologies (defensive verbal forms) were seen to accommodate a negative action or effect such as poor performance; entitlements and enhancements (assertive verbal forms) were seen to claim positive characteristics such as good performances to try to maximise the desirable implications for the actor (Aerts, 1994). This bias is supported by interviews with those involved either in disclosure decisions or as outside consultants in relation to those decisions (Gibbons et al., 1990). These informants revealed that business affairs were arranged to produce desired disclosure content, together with management of the visibility of disclosure. This involved burying sensitive information in verbiage or highlighting good news, as well as correcting press stories. Performance indicators and emotive words were more or less mentioned in accordance with performance (Clarke, 1997).

Audiences give meaning to text. Phenomenologists argue that an audience is Corporate annual implicitly portrayed in the text and is actively involved in the temporal flow of reports: research reading, so that the audience is centrally positioned as a kind of co-author. perspectives Audience analysis involves understanding how the reader is depicted in, and interacts with, the text. Text has no independent meaning or reality, only that realised when a reader engages in the reading process. In this sense, narrative 487 in a CAR is a hyperstructure that engages readers as co-makers of a story (Jameson, 2000). Analysis of imagery The discourse texture as well as the underlying material and story affect how readers experience a text (Jameson, 2000). Narration as a means of achieving a desired interpretation was augmented by photography (Anderson and Imperia, 1992; Douglis, 2000) and graphs (Beattie and Jones, 1992, 1997, 1999). Because such images are important ways by which corporations seek to represent themselves to reading publics (Preston et al., 1996), the proposition is that the symbols can and are used to guide interpretation to particular outcomes. Photographs can be used to help personalise an impersonal reporting entity, but in doing so, do not reflect how people behave, but rather how one thinks they behave (Anderson and Imperia, 1992). They help to interpret, abstract and activate readers’ imaginations (Douglis, 2000) so that they do not faithfully represent or inform, but persuade and distract readers from other information in the report (Simpson, 1999). They persuade readers of the credibility of the reports (Graves et al., 1996). As well as what and how much is photographed, studies focus on elements such as orientation to the camera, head cant, gestures and smiles. Like photographs, graphs can be used to influence readers’ perceptions. Investigations of graphs have utilised graphical construction theory to identify practices and devices, such as measurement distortion, selectivity, slope parameters, colour and labelling, which ‘‘may impede effective communication’’ (Beattie and Jones, 1999, p. 65). Financial graphs were found to be another of the devices by which management enhanced perceptions of their performance (Beattie and Jones, 1992, 1997, 1999; Mather et al., 1996). Consistent with narrative disclosure studies, researchers concluded that financial graphs were frequently distorted in order to give a more favourable view of the corporation than was warranted, and that financial graphs were more likely to enhance good news, while minimising bad news (Beattie and Jones, 1999). The marketing perspective Based on their findings that perceptions of corporate or management performance were enhanced or that negative news was ‘‘deemphasised’’ (Subramanian et al., 1993), annual reports are asserted to be marketing tools (e.g. Subramanian et al., 1993; Anderson and Imperia, 1992). Within the marketing literature, a belief that all corporate communications should serve the marketing task means that annual reports should be used as part of an

AAAJ 15,4

488

integrated marketing communication (IMC) strategy. The focus is to enhance understanding of, association with, attitude and predisposition toward, a product, service or the corporation (Adcock et al., 1995). Described as ‘‘one of the most important communication trends of this decade’’ (Burnett and Moriarty, 1998, p. 24), IMC is a broad process encompassing everything from corporate image to the handling of customer inquiries. It has as its primary objective the unification of all the marketing communication tools (including CARs) a corporation uses to send target audiences a consistent message that promotes corporate goals. These tools also may be used to anticipate and to control unplanned messages so that IMC involves (Hartley and Pickton, 1999) three broad fields of managing consumer contact, marketing communications, and corporate communications. The need for integrated communication arises because resources spent on isolated activities such as corporate image building and brand building may be wasted and synergies lost in reaching target audiences. Annual reports are part of the corporate communications exercise, in turn integrated into the whole marketing communications strategy designed to achieve corporate objectives. As an instrument of marketing strategy, a CAR is part of the publicity function (involving public relations, the deliberate, planned and sustained effort to establish and maintain mutual understanding between an organisation and its publics) (Adcock et al., 1995, p. 264). As well, it contributes to other elements such as corporate identity and image. Further, the softness of boundaries between the three areas of IMC allows a CAR to contribute to brand public relations and customer management. The development of the annual report as a part of corporate marketing communications raises diverse issues. While there is exhortation to treat CARs as part of marketing communications, understanding the implications of this call seems lacking. Despite assertions, often from other disciplines, that a CAR is a marketing tool (e.g. Subramanian et al., 1993; Anderson and Imperia, 1992), no study examined the whole report as a marketing communication tool. Apart from Droge et al. (1990), the studies are piecemeal and fall short of establishing that the CAR is used as a part of an integrated marketing communication system. Droge et al. (1990) show the development of a marketing focus over time (1930 to 1950). Their work is novel by arguing the need to show consistency with marketing planning through the identification of target audiences, the determination of their needs and the development of messages that meet those needs. However, they seek to achieve this aim not through the analysis of annual reports, but through analysis and comment of business press journalists over this period. This vicarious approach reaches the conclusion that CARs are part of an overall marketing effort and not ‘‘isolated public relations documents’’ (Droge et al., 1990, p. 363). Enunciation of strategy was directly addressed. The CEO letter is used to express and build customer commitment in a small majority of a sample of business to business firms (Judd and Tims, 1991). CEO letters show differences

in the messages conveyed in that ‘‘consistent communication strategies are Corporate annual being followed based on favorable or unfavorable company performance’’ reports: research (Kohut and Segars, 1992, p. 17). Although mission statements can be used to perspectives establish corporate identity, express goals and strategies, and address specific stakeholder groups, Leuthesser and Kohli (1997) found that in a sample of 393 US corporations, only 16 percent contained explicit mission statements, with 489 customers being addressed most frequently. Bank annual reports also failed to adequately explain corporate strategy and decisions (Holliday, 1994). Linked to strategy and its implementation is the identification of market segments (groups of persons or organisations that are relatively homogeneous in their response to particular marketing messages). Identification of the various audiences of the CAR and ascertaining that they are targeted is a primary concern (Bekey 1990; Rezaee and Porter, 1993; Marino, 1995; Brink, 1998; Mitchell, 1998; Buxton, 2000) although there are many foci. Investors are acknowledged as only one of several possible audiences requiring a particular message. Droge et al. (1990) recognise 15 possible audiences. Bekey (1990) tries to show that CARs have become ‘‘sophisticated marketing tools’’ that cater to different audiences in different ways. In contrast, failure to identify audiences, their needs and whether they are met is criticised (Brink, 1998; Marino, 1995). Others see the ability of the CAR to reach target audiences as limited by the statutory requirements and thus urge the development of summary annual reports (Rezaee and Porter, 1993) or alternative media to reach particular stakeholders (Buxton, 2000). Messages addressed to these targeted audiences have received scant attention. The need for a CAR to ‘‘sell’’ a corporation as a brand to its multiple audiences (Mitchell, 1998) needs to be reconciled with the growing complexity of CARs (Rezaee et al., 1993) and threats to their communication effectiveness (Marino, 1995; Stocks, 1995). Despite growing complexity, more messages are urged. The value and importance of marketing is not reflected in CARs because of the dichotomy between financial and non-financial sections. More formal discussion of a few key marketing measures is argued (Herremans and Ryans, 1995). With so many targeted audiences and messages, to target successfully requires developments in style and design and/or alternative media such as the Internet (Buxton, 2000). Although researchers view a CAR as a marketing tool, their research so far has failed to confirm its strategic purpose; that is, to target particular audiences in a proactive manner. Their findings cast doubt on the ability to use it as a marketing tool. Failure to spell out strategy and its subsequent absence in other elements of the report (Herremans and Ryans, 1995), and to set clear measurable objectives on what the report is expected to achieve with its audiences (Brink, 1998) are current deficiencies. The requirement for a CAR to satisfy multiple target groups and still be tightly focussed in terms of strategy leads to tension. This is reflected in calls for either expanded CARs (Herremans and Ryans, 1995; Leuthesser and Kohli, 1997) or for smaller, tightly focussed documents (Marino, 1995; Bekey, 1990).

AAAJ 15,4

490

The political economy perspective Like marketing researchers, those adopting a political economy perspective view corporate disclosure as a proactive process. However, the purpose of the provision of information is viewed as much broader, being ‘‘designed to set and shape the agenda of debate and to mediate, suppress, mystify and transform social conflict’’ (Guthrie and Parker, 1989). As can be seen from Table II, a political economy perspective has been little used in the last decade, and then only in accounting journals. A political economy (of accounting) emphasizes the infrastructure or institutional environment that supports the existing system of corporate reporting, as well as the fundamental relations between class in society. According to Cooper and Sherer (1984), the perspective is characterized by three features. The first is the recognition of power and conflict in society, reflected in the distribution of income, wealth and power. Accounting reports operate for specific interests, but attention is drawn away from this by the classifications used in corporate annual reports. The second characteristic is the historical and institutional environment of the society. Adoption of a more emancipated view of human motivation and the role of accounting in society is the third characteristic. Political economy theory argues that dysfunctional structural relationships create a power elite that controls resources, resulting in power and conflict in society (Burchell et al., 1980), as well as an interplay between class and other forms of domination (Adams and Harte, 1998). CARs are seen to contribute to the maintenance of an ideology of oppression. As an ideological weapon, an annual report is viewed as a proactive tool by which corporate management influences and shapes what is important in society (Burchell et al., 1980). Using this perspective, CARs are used to ‘‘mediate, suppress, mystify and transform social conflict’’ (Tinker and Neimark, 1987, p. 72). Graves et al. (1996) contend that the pictures and gloss in US annual reports function rhetorically to assert not only the specific values and public relations agendas of individual companies, but the truth claims of the accounts themselves. They suggest that the pictures, gloss, and gratuitous information in contemporary annual reports are there to persuade the report reader of the truth claims of the accounts and thus to perpetuate the values that reside in them. Adopting the view that CARs can be influential in shaping what is important in society, Adams and Harte (1998) focused on the disclosure of equal opportunities information in annual reports, thus shedding light on women’s employment. However, they find that CARs are as important for what they omit as for what they disclose. Neimark (1992, p. 100) observes that ‘‘in preparing an annual report a company makes choices about the issues and social relationships that they consider sufficiently important or problematic to address publicly’’. The legitimacy perspective In contrast to both the marketing and the political economy perspectives, those adopting a legitimacy perspective (see Table II) see corporations as controlled

by community concerns (Brown and Deegan, 1998) and values. Because values Corporate annual change over time, corporations need to be responsive to changing values, so as reports: research to signal their value congruency with social values (Dowling and Pfeffer, 1975). perspectives Successful legitimation, according to Freidson (1986, cited by Neu and Wright, 1992, p. 647), depends on convincing both society at large and other institutional actors that a congruency of actions and values should and does 491 exist. Corporate management will react to public concern over corporate actions by increasing the level of corporate disclosures in annual reports if it perceives that their ‘‘legitimacy’’ is threatened by that public concern (Brown and Deegan, 1998). Organisational legitimacy research is published largely in accounting journals. Individual narrative passages are used to support the view that organisations seek legitimacy. Examination of these passages can be thematic or syntactic (Jones and Shoemaker, 1994). Within thematic studies the objective is to extract and analyse themes inherent within the message. The dominant theme in studies published in the 1990s is that of corporate social reporting: narrative disclosures are targeted to ascertain how a corporation constructs itself and its relationship with others (Gray et al., 1995). Environment disclosures were the most researched (see Table II), although women’s employment (Adams and Harte, 1998) and stigma management (Neu and Wright, 1992) were investigated. While disclosures are typically self-laudatory, with little or no negative disclosures being made by corporations (Deegan and Gordon, 1996; Hackston and Milne, 1996), disclosures did not match environmental performance (Fekrat et al., 1996). Disclosures appear to be included as corporate management responds to media attention (Brown and Deegan, 1998) or to successful prosecutions by environmental agencies (Deegan and Rankin, 1996). Environmental disclosures were linked to size of reporting corporation (Patten, 1991; Cooke, 1992; Hackston and Milne, 1996), to industry membership (Patten, 1991) especially of environmentally sensitive industries (Deegan and Gordon, 1996), to media attention in the form of newspaper articles (Brown and Deegan, 1998), and to pressure groups (Tilt, 1994). The studies largely confirmed what was previously known: that CSR disclosures, both voluntary and mandatory, and those relating to the environment are increasing (Gamble et al., 1996; Harte et al., 1991), particularly among larger companies. Differences between the ‘‘public’’ face and ‘‘private’’ face of social responsibility concerns resulted in stakeholder concerns more likely being addressed in internal communications than in a corporation’s annual report (Robertson and Nicholson, 1996). Through the lens of organisational legitimacy, researchers view the role of the CAR as a tool of corporate management whereby management signals its reactions to the concerns of particular stakeholders. The purpose is to communicate a congruency of actions and values with those of stakeholders seen as important to the legitimation process. In this decade, researchers have

AAAJ 15,4

492

looked narrowly at the legitimation process, being concerned largely with environmental issues. The accountability perspective Like the legitimacy perspective, accountability views corporations via their management as reacting to concerns of external parties. Accountability involves the monitoring, evaluation and control of organisational agents to ensure that they behave in the interests of shareholders and other stakeholders (Keasey and Wright, 1993). There are two interpretations of this concept. The narrow one deals with the relationship between the company and its shareholders so that the primary focus is on financial information within the annual report. The second interpretation deals with the relationship between a corporation and an audience wider than shareholders, so that its focus may be any disclosures within an annual report. Because accountability focuses upon the relationship between the corporation and users of its annual reports, information transmission between them depends upon the terms of that relationship (Owen et al., 1987). In its narrowest form, accountability is simply the requirement to report upon the extent to which a corporation has met its responsibilities to its owners. This rendering of responsibilities is viewed as a struggle with principals, not with affected publics, so that disclosures in annual reports are perceived to function in the interests of the principals (Gallhofer and Haslam, 1993). Management, by implication, is the weaker of the protagonists. Emphasis is on the law, and corporate compliance with the law (Tilt, 1994). As such, conveyors of accounts do not benefit from the accountability process (Gallhofer and Haslam, 1993). They are monitored, evaluated and controlled by organisational agents to ensure that they behave in the interests of shareholders and other stakeholders (Keasey and Wright, 1993). The value system underlying the approach is the acceptance of the prime importance of shareholders (Mathews, 1987), although stakeholders are acknowledged. Illustrating this focus, Anderson and Epstein (1996, p. 168) start from the innocuous proposition that annual reports should be ‘‘user driven’’. Their research design for finding what users want is based on a survey solely of shareholders’ and investors’ wants because ‘‘shareholders are the owners’’. Annual reports are commonly regarded as the key accountability mechanism (Lamond, 1995), with a focus on what is omitted from them; for example, contractual information (Ormrod and Cleaver, 1993), and criteria for board membership (Ramsay and Hoad, 1997). Board membership was found to influence concealment of adverse performances in presidents’ letters (Abrahamson and Park, 1994). Outside directors, large institutional investors, and accountants limit concealment, but small institutional investors and outside directors who are shareholders prompt it. Low disclosure is associated with subsequent selling of stock by top officers and outside directors. Abrahamson and Park (1994) claim that concealment by officers and its toleration by outside directors may be intentional.

Researchers examining CARs through an accountability perspective view Corporate annual an annual report as a document through which management reacts to the reports: research concerns of shareholders. However, CARs were found not to be a credible way perspectives to communicate with shareholders because of what they did not reveal (Anderson and Epstein, 1996). What was seen What was seen was gleaned from the front half of the reports rather than the whole report. What was seen in the front half was selective; for example, CEO letters or environmental narrative disclosures were examined only. The ways of seeing were varied. These were identified by claims of researchers that they were viewing CARs through a particular perspective. There is a tension in how researchers have sought to view CARs and thus the cross-section of findings point to multiple intentions and stories in their design and purpose. Depending on the perspective adopted, a CAR could either be a proactive document constructing and projecting a particular image, or merely a reactive document, responding to the concerns of particular groups (see Figure 1). Its targeted audiences could be wide – multiple publics – or targeted at specific groups such as shareholders. While not all perspectives can be placed in such a framework, for those in Figure 1, tensions in ways of viewing annual reports necessarily arise in relation to the elements of audience and purpose. Those tensions are exemplified in the increasingly popular perspective of image management. While many utilising this perspective view annual reports as pro-active documents used to develop and maintain particular corporate images, and to present information as favourably as possible, those examining reactions to crises such as bad news saw annual reports as reactive documents. As a result, images in annual reports were viewed as propaganda at one extreme (Simpson, 1999) to ‘‘tell[ing] it like it is’’ (Douglis, 2000) at the other. As this perspective develops, such tensions are likely to be resolved. However, the audience of image management was not conceived in targeted or narrow terms by any of the researchers.

493

Figure 1. Perspectives

AAAJ 15,4

494

Researchers examining CARs from a political economy perspective take the widest view of what they intended to achieve. Almost all factors external to a corporation can impact on what is communicated within annual reports, so that all factors must be considered. Corporations are viewed as pro-active (Burchell et al., 1980; Guthrie and Parker, 1989), using annual reports as tools directed to a wide audience over which a corporation seeks to entrench its economic power. In contrast, the accountability perspective focuses on a corporation limited in action by the law. The discharge of accountability through CARs is examined in terms of a narrow audience, shareholders, whose right to an annual report is enshrined in law (Mathews, 1987; Anderson and Epstein, 1996). Reporting is viewed as a reaction to legal requirements. Examinations under this perspective focus on what is left out of the report as an integral part of interpreting the story (Adams and Harte, 1998; Jameson, 2000; Buhr, 1998). Organisational legitimacy sits in the same quadrant as accountability. Disclosure is directed towards influential stakeholders (Campbell, 2000). Groups addressed depend on the needs of legitimation. The approach is reactive in that a corporation is viewed as responsive to stakeholders, the media, and pressure groups (Brown and Deegan, 1998; Tilt, 1994). Viewing the CAR through this lens, researchers seek to establish how disclosures address particular concerns of particular groups in society from whom it seeks legitimation. Such disclosures are termed corporate social reporting. Researchers viewing CARs from a marketing perspective, similar to a legitimacy perspective, observe the capacity of the report to reach targeted audiences who can advance the interests of the corporation. Marketing communication is based on implementing clear strategic goals, a strategy that seeks to meet the corporation’s own objectives where the target audiences are the means to this end. As such, marketing is pro-active. The corporate annual report can be used to strengthen a corporation’s market position with respect to its products or brands or whole of corporation activities, or its management team. The aspect of meeting the demands of target groups creates confusion with legitimacy explanations. However, the role of the annual report in marketing as part of a pro-active integrated marketing communication process is guided by a strong strategic focus that distinguishes the two. While marketing communication also has much in common with image management, the former is distinguished by its targeted approach. Most researchers ascribe to a stakeholder perspective, although often not acknowledging its theoretical underpinning. Despite assertions that users are a wide cross-section of society, there is very little evidence of who actually uses them, although Harte et al. (1991) and Tilt (1994) found that ethical investment funds and environmental pressure groups are users. Arfin’s (1993) contradictory view that the annual report is a report to owners but that the best ones are conceived and used as corporate marketing tools represents this change in user focus. Similarly, Anderson and Epstein (1996, p. 168) identify only shareholders’ and investors’ wants. Nevertheless, they are inconsistent when they argue management should view the CAR as a vehicle to make

‘‘credible communications to the corporate shareholder, and other external Corporate annual users’’. reports: research The view that an annual report is a private document for owners is now not perspectives widely held. Jones (1996b) views publication of a CAR as an opportunity for corporations to convey appropriate messages both to shareholders and other interested user groups. His view of a need to address a wider audience is 495 similar to, but the sub-text different from, a more widely expressed belief that a CAR is selling a product and an image to customers, business prospects, the press, and employees (Anderson and Imperia, 1992, p. 114). The difference reflects a division in interpreting the purpose of the annual report, and consequently, in the advocacy of changes to it as a communication instrument. However, in spite of widspread assertion that it is becoming a marketing tool, studies supporting this view with strong evidence are lacking. Conclusion For the decade of research examined, researchers ‘‘constructed reality’’ (Hines, 1988) or ‘‘visbility and meaning’’ (Hopwood, 1996) by examining parts of the annual report or the tools used in constructing it. Few studies address the document as a whole, in terms of the integration of the messages between the various parts of the report. Herremans and Ryans’ (1995) is an exception. Explanation of the changing structure and content of annual reports remains divided, largely because of the differing perspectives of researchers. Not surprisingly, they constructed their own realities, selectively focusing on parts and elements that they perceived could give meaning. What they have revealed is diversity in the ways of seeing the annual report and a tension in understanding its overall purpose and role. These perspectives are difficult to reconcile. Whether an annual report is written from the perspective of seeking to reduce the effects of events perceived to be unfavourable to a corporation’s image, or as a proactive document seeking outcomes that advance the corporation’s or management’s objectives, reflects a division between the pursuit of legitimacy and corporate social responsibility on the one hand, and political economy, image management and marketing interpretations on the other. Accordingly, preparers presumably select and organise their material in terms of the kind of audience they seek to address. Although Parker (1982) has argued the need to identify the significance and requirements of audiences, little has been done. While marketing studies refer to target groups such as employees, customers and suppliers, and seek to show messages addressing such groups, few have appraised the suggested users or have suggested ways of monitoring the outcomes. This weakness is not confined to those adopting a marketing perspective. This deficiency impinges on the quality of accountability and governance communication provided in the modern annual report. If the credibility of information in the annual report underpins its corporate governance role (Keasey and Wright, 1993) then the implications of growing voluntary

AAAJ 15,4

496

reporting in relation to these tasks is in need of further research. If an annual report is part of a corporate communication strategy that pursues strategic objectives, such as strengthening the corporate image or brand, or seeks to strengthen other marketing objectives with particular stakeholders, then the story told may conflict with, or undermine the story on governance and accountability. In this story telling, largely conducted in the front half, but influencing the interpretation of the back half, communication studies suggest a complex arsenal of communication tools can be, and are increasingly, used. This arsenal is particularly used in the front part of the report. The selection and integration of narrative, language, photography, graphs and other symbols, creates what Jameson (2000) has aptly called a hyperstructure that engages the reader as a co-maker of a story. There are many stories in the one report. Multiple voices and multiple stories allow audiences to choose the story they want as well as supporting the targeting of various audiences. Notes 1. The authors wish to thank Linda English for her suggestion to include this section. 2. An anonymous reviewer pointed out the difficulties of using the term ‘‘communication’’ to identify these papers. We are indebted to that reviewer for the solution achieved. References Abrahamson, E. and Park, C. (1994), ‘‘Concealment of negative organizational outcomes: an agency theory perspective’’, Academy of Management Journal, Vol. 37 No. 5, October, pp. 1302-34. Adams, C.A. and Harte, G. (1998), ‘‘The changing portrayal of the employment of women in British banks’ and retail companies’ corporate annual reports’’, Accounting, Organizations and Society, Vol. 23 No. 8, pp. 781-812. Adams, C.A., Hill, W. and Roberts, C.B. (1998), ‘‘Corporate social reporting practices in Western Europe: legitimating corporate behaviour?’’, The British Accounting Review, Vol. 30 No. 1, March, pp. 1-21. Adams, M. and Hossain, M. (1998), ‘‘Managerial discretion and voluntary disclosure: empirical evidence from the New Zealand life insurance industry’’, Journal of Accounting and Public Policy, Vol. 17 No. 3, pp. 245-81. Adcock, D., Bradfield, R., Halborg, A. and Ross, C. (1995), Marketing Principles and Practice, 2nd ed., Pitman, London. Aerts, W. (1994), ‘‘On the use of accounting logic as an explanatory category in narrative accounting disclosures’’, Accounting, Organizations and Society, Vol. 19, No. 4/5, pp. 337-53. Anderson, C. and Imperia, G. (1992), ‘‘The corporate annual report: a photo analysis of male and female portrayals’’, The Journal of Business Communications, Vol. 22 No. 2, Spring, pp. 113-28. Anderson, R. and Epstein, M. (1996), The Usefulness of Corporate Annual Reports to Shareholders in Australia, New Zealand, and the United States: An International Comparison, JAI Press, Greenwich, CT. Arfin, F.N. (1993), Annual Reports that Pay their Way, Pitman Publishing, London. Bartlett, S. and Jones, M.J. (1997b), ‘‘Annual reporting disclosures 1970-90: an exemplification’’, Accounting, Business and Financial History, Vol. 7 No. 1, pp. 61-80.

Beattie, V. and Jones, M.J. (1992), ‘‘The use and abuse of graphs in annual reports: theoretical framework and empirical study’’, Accounting and Business Research, Vol. 22 No. 88, pp. 291-303. Beattie, V. and Jones, M.J. (1997a), ‘‘A comparative study of the use of financial graphs in the corporate annual reports of major US and UK companies’’, Journal of International Financial Management and Accounting, Vol. 8 No. 1. Beattie, V. and Jones, M.J. (1999), ‘‘Australian financial graphs: an empirical study’’, ABACUS, Vol. 35 No. 1, pp. 46-76. Beattie, V., Jones, M. and Mellett, H. (1996), ‘‘Annual reports: companies could do better’’, Management Accounting, July/August, pp. 52-4. Bekey, M. (1990), ‘‘Annual reports evolve into marketing tools’’, Financial Manager, Vol. 3 No. 1, January/February, pp. 50-60. Brink, A. (1998), ‘‘Oh no! I forgot to evaluate the annual report’’, Communication World, Vol. 16, December, pp. 24-7. Brown, N. and Deegan, C. (1998), ‘‘The public disclosure of environmental performance information – a dual test of media agenda setting theory and legitimacy theory’’, Accounting and Business Research, Vol. 29 No. 1, pp. 21-41. Buhr, N. (1998), ‘‘Environmental performance, legislation and annual report disclosure: the case of acid rain and Falconbridge’’, Accounting, Auditing & Accountability Journal, Vol. 11 No. 2, pp. 163-90. Burchell, S., Clubb, C., Hopwood, A. and Hughes, J. (1980) ‘‘The roles of accounting in organizations and society’’, Accounting Organizations and Society, Vol. 5 No. 1, pp. 5-27. Burnett, J. and Moriarty, S. (1998), Introduction to Marketing Communication: An Integrated Approach, Prentice-Hall, Upper Saddle River, NJ. Buxton, P. (2000), ‘‘Companies with a social conscience’’, Marketing, April, pp. 33-4. Campbell, D.J. (2000), ‘‘Legitimacy theory or managerial reality construction? Corporate social disclosure in Marks and Spencer Plc corporate reports, 1969-1997’’, Accounting Forum, Vol. 24 No. 1, March, pp. 80-100. Clarke, G. (1997), ‘‘Messages from CEOs: a content analysis approach’’, Corporate Communications: An International Journal, Vol. 2 No. 1, pp. 31-9. Cooke, T.E. (1992), ‘‘The impact of size, stock market listing and industry type on disclosure in the annual reports of Japanese listed corporations’’, Accounting and Business Research, Vol. 22 No. 87, pp. 229-37. Cooper, D.J. and Sherer, M.J. (1984), ‘‘The value of corporate accounting reports: arguments for a political economy of accounting’’, Accounting, Organizations and Society, Vol. 9 No. 3/4, pp. 207-32. Courtis, J.K. (1995), ‘‘Readability of annual reports: Western versus Asian evidence’’, Accounting, Auditing & Accountability Journal, Vol. 8 No. 2, pp. 4-17. Courtis, J.K. (1998), ‘‘Annual report readability: tests of the obfuscation hypothesis’’, Accounting, Auditing & Accountability Journal, Vol. 11 No. 4, p. 459. Deegan, C. and Gordon, B. (1996), ‘‘A study of the environmental disclosure practices of Australian corporations’’, Accounting and Business Research, Vol. 26 No. 3, pp. 187-99. Deegan, C. and Rankin, M. (1996), ‘‘Do Australian companies report environmental news objectively? An analysis of environmental disclosures by firms prosecuted successfully by the Environmental Protection Authority’’, Accounting, Auditing & Accountability Journal, Vol. 9 No. 2, pp. 50-67. Deegan, C., Ranking, M. and Voght, P. (2000), ‘‘Firms’ disclosure reactions to major social incidents: Australian evidence’’, Accounting Forum, Vol. 24 No. 1, March, pp. 101-28. Design Week (1999), ‘‘Wolf Olins leads ranking of top 20 UK design firms . . .’’, Design Week, Supplement, April, pp. 29-30.

Corporate annual reports: research perspectives 497

AAAJ 15,4

498

Douglis, P. (2000), ‘‘Photojournalism: telling it like it is’’, Communication World, February, Vol. 17, pp. 44-7. Dowling, J. and Pfeffer, J. (1975), ‘‘Organizational legitimacy: social values and organizational behaviour’’, Pacific Sociological Review, Vol. 18 No. 1, pp. 122-36. Droge, C., Germain, R. and Halstead, D. (1990), ‘‘A note on marketing and the corporate annual report: 1930-1950’’, Journal of the Academy of Marketing Science, Vol. 18 No. 4, pp. 355-64. Fekrat, M.A., Inclan, C. and Petroni, D. (1996), ‘‘Corporate environmental disclosures: competitive disclosure hypothesis using 1991 annual report data’’, The International Journal of Accounting, Vol. 31 No. 2, pp. 175-95. Gallhofer, S. and Haslam, J. (1993), ‘‘Approaching corporate accountability: fragments from the past’’, Accounting and Business Research, Vol. 23 No. 91A, pp. 320-30. Gamble, G.O., Hsu, K., Jackson, C. and Tollerson, C.D. (1996), ‘‘Environmental disclosures in annual reports: an international perpsective’’, The International Journal of Accounting, Vol. 31 No. 3, pp. 293-331. Gatland, L. (1997), ‘‘Boller Coates & Neu wins awards during recent annual report competition . . .’’, Crain’s Chicago Business, 13 October, p. 27. Gibbons, M., Richardson, A., and Waterhouse, J. (1990), ‘‘The management of corporate financial disclosure: opportunism, ritualism, policies, and processes’’, Journal of Accounting Research, Vol. 28 No. 1, Spring, pp. 121-43. Graves, O.F., Flesher, D.L. and Jordan, R.E. (1996), ‘‘Pictures and the bottom line: the television epistemology of US annual reports’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 57-88. Gray, R., Kouhy, R. and Lavers, S. (1995), ‘‘Corporate social environmental reporting. A review of the literature and a longitudinal study of UK disclosure’’, Accounting, Auditing & Accountability Journal, Vol. 8 No. 2, pp. 47-77. Guthrie, J. and Parker, L. (1989), ‘‘Corporate social reporting: a rebuttal of legitimacy theory’’, Accounting and Business Research, Vol. 19 No. 76, pp. 343-52. Guthrie, J. and Parker, L. (1990), ‘‘Corporate social disclosure practice: a rebuttal of legitimacy theory’’, Advances in Public Interest Accounting, Vol. 3, pp. 159-76. Hackston, D. and Milne, M.J. (1996), ‘‘Some determinants of social and environmental disclosures in New Zealand companies’’, Accounting, Auditing & Accountability Journal, Vol. 9 No. 1, pp. 77-108. Harte, G. and Owen, D. (1991), ‘‘Environmental disclosure in the annual reports of British companies: a research note’’, Accounting, Auditing & Accountability Journal, Vol. 4 No. 3, pp. 51-61. Harte, G., Lewis, L. and Owen, D. (1991), ‘‘Ethical investment and the corporate reporting function’’, Critical Perspectives on Accounting, Vol. 2, pp. 227-53. Hartley, B and Pickton, D. (1999), ‘‘Integrated marketing communications requires a new way of thinking’’, Journal of Marketing Communications, Vol. 5, pp. 97-106. Herremans, I. and Ryans, J. Jr (1995), ‘‘The case for better measurement and reporting of marketing performance’’, Business Horizons, Vol. 38 No. 5, September, pp. 51-60. Hines, R.D. (1988), ‘‘Financial accounting: in communicating reality, we construct reality’’, Accounting, Organizations and Society, Vol. 13 No. 3, pp. 251-61. Holliday, K. (1994), ‘‘Annual reports as marketing tools’’, Bank Marketing, Vol. 26, August, pp. 22-7. Hopwood, A.G. (1996), ‘‘Introduction’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 55-6. Jameson, D. (2000), ‘‘Telling the investment story: a narrative analysis of shareholder reports’’, Journal of Business Communication, Vol. 37 No. 1, January, pp. 7-38.

Jones, M.J. (1996a), ‘‘Readability of annual reports: Western versus Asian evidence – a comment to contextualize’’, Accounting, Auditing & Accountability Journal, Vol. 9 No. 2, pp. 86-91. Jones, M. (1996b), ‘‘Accounting narratives: an emerging trend’’, Management Accounting, April, pp. 41-2. Jones, M.J. and Shoemaker, P.A. (1994), ‘‘Accounting narratives: a review of empirical studies of content and readability’’, Journal of Accounting Literature, Vol. 13, pp. 142-84. Judd, V. and Tims, B. (1991), ‘‘How annual reports communicate a customer orientation’’, Industrial Marketing Management, Vol. 20 No. 4, pp. 353-60. Keasey, K. and Wright, M. (1993), ‘‘Issues in corporate accountability and governance: an editorial’’, Accounting and Business Research, Vol. 23 No. 91A, pp. 291-303. Kohut, G. and Segars, A. (1992), ‘‘The president’s letter to stockholders: an examination of corporate communication strategy’’, Journal of Business Communication, Vol. 29 No. 1, pp. 7-21. Lamond, D. (1995), ‘‘Dilemmas of measuring performance in a government trading enterprise: The State Bank of NSW’’, Australian Journal of Public Administration, Vol. 54 No. 2, pp. 262-72. Lee, T. (1994), ‘‘The changing form of the corporate annual report’’, The Accounting Historians Journal, Vol. 21 No. 1, June, pp. 215-32. Leuthesser, L. and Kohli, C. (1997), ‘‘Corporate identity: the role of mission statements’’, Business Horizons, Vol. 40, May/June, pp. 59-66. McKinstry, S. (1996), ‘‘Designing the annual reports of Burton PLC from 1930 to 1994’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 89-111. Marino, A. Jr (1995), ‘‘Separating your annual report from the herd’’, Public Relations Quarterly, Vol. 40 No. 2, Summer, pp. 44-7. Mather, P., Ramsay, A., and Serry, A. (1996), ‘‘The use and representational faithfulness of graphs in annual reports: Australian evidence’’, Australian Accounting Review, Vol. 6 No. 2. Mathews, M.R. (1987), ‘‘Social responsibility accounting disclosure and information content for shareholders: a comment on ‘researching the information content of social responsibility disclosures’’’, British Accounting Review, Vol. 19 No. 2, pp. 161-7. Mitchell, A. (1998), ‘‘Annual brand report’’, Management Today, June, pp. 110-11. Neimark, M. (1992), Hidden Dimensions of Annual Reports: Sixty Years of Social Conflict at General Motors, Markus Weiner, New York, NY. Ness, K.E. and Mirza, A.M. (1991), ‘‘Corporate social reporting: a note on a test of agency theory’’, The British Accounting Review, Vol. 23, pp. 211-17. Neu, D. and Wright, M. (1992), ‘‘Bank failures, stigma management and the accounting establishment’’, Accounting, Organizations and Society, Vol. 17 No. 7, pp. 645-65. Neu, D., Warsame, H. and Pedwell, K. (1998), ‘‘Managing public impressions: environmental disclosures in annual reports’’, Accounting, Organizations and Society, Vol. 23 No. 3, pp. 265-82. Ormrod, P. and Cleaver, K.C. (1993), ‘‘Financial reporting and corporate accountability’’, Accounting and Business Research, Vol. 23 No. 91A, pp. 431-9. Owen, D., Gray, R. and Maunders, K. (1987), ‘‘Researching the information content of social responsibility disclosures: a comment’’, British Accounting Review, Vol. 19 No. 2, pp. 169-75. Parker, L.D. (1982), ‘‘Corporate annual reporting: a mass communication perspective’’, Accounting and Business Research, Vol. 12 No. 48, pp. 279-86. Patten, D.M. (1991), ‘‘Exposure, legitimacy, and social disclosure’’, Journal of Accounting and Public Policy, Vol. 10 No. 4, pp. 297-308.

Corporate annual reports: research perspectives 499

AAAJ 15,4

500

Patten, D.M. (1992), ‘‘Intra-industry environmental disclosures in response to the Alsakan oil spill: a note on legitimacy theory’’, Accounting, Organizations and Society, Vol. 17 No. 5, Autumn, pp. 471-5. Patten, D.M. (1997), ‘‘Variability in social disclosure: a legitimacy-based analysis’’, Advances in Public Interest Accounting, Vol. 6, pp. 273-85. Pava, M.L. and Epstein, M.J. (1993), ‘‘How good is MD&A as an investment tool?’’, Journal of Accountancy, March, pp. 51-3. Preston, A.M., Wright, C. and Young, J.J. (1996), ‘‘Imag[in]ing annual reports’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 113-37. Ramsay, I.M. and Hoad, R. (1997), ‘‘Disclosure of corporate governance practices by Australian companies’’, Company and Securities Law Journal, Vol. 15, pp. 454-67. Rezaee, Z. and Porter, G. (1993), ‘‘Can the annual report be improved?’’, Review of Business, Vol. 15 No. 1, Summer/Fall, pp. 38-41. Roberts, C.B. (1991), ‘‘Environmental disclosures: a note on reporting practices in mainland Europe’’, Accounting, Auditing & Accountability Journal, Vol. 4 No. 3, pp. 62-71. Robertson, D.C. and Nicholson, N. (1996), ‘‘Expressions of corporate social responsibility in UK firms’’, Journal of Business Ethics, Vol. 15, October, pp. 1095-106. Rosenberger, J. (2000), ‘‘Not your typical annual report’’, Graphic Arts Monthly, Vol. 72 No. 5, May, p. 149. Simpson, L.L. (1999), The Annual Report: An Exercise in Ignorance?, Department of Accountancy and Business Law, Discussion Paper Series 197, Massey University, Turitea, December. Smith, M. and Taffler, R. (1992), ‘‘Readability and understandability: different measures of the textual complexity of accounting narrative’’, Accounting, Auditing & Accountability Journal, Vol. 5 No. 4, pp. 84-98. Stocks, D. (1995), ‘‘State of inspiration’’, Marketing, Vol. 24, August, p. 29. Subramanian, R., Insley, R. and Blackwell, R. (1993), ‘‘Performance and readability: a comparison of annual reports of profitable and unprofitable corporations’’, Journal of Business Communication, Vol. 30, pp. 50-61. Thomas, J. (1997), ‘‘Discourse in the marketplace: the making of meaning in annual reports’’, Journal of Business Communication, Vol. 34 No. 1, January, pp. 44-66. Tilt, C.A. (1994), ‘‘The influence of external pressure groups on corporate social disclosure. Some empirical evidence’’, Accounting, Auditing & Accountability Journal, Vol. 7 No. 4, pp. 47-72. Tinker, A. and Neimark, M. (1987), ‘‘Role of annual reports in gender and class contradictions at General Motors: 1917-1976’’, Accounting, Organizations and Society, Vol. 12 No. 1, pp. 71-88. Valentine, M. (1999), ‘‘Top design agency among FTSE 250 companies in 1999 is Pauffley’’, Design Week, 8 October, p. 4. Ze’ghal, D. and Sadrudin, A.A. (1990), ‘‘Comparison of social responsibility information disclosure media used by Canadian firms’’, Accounting, Accountability & Auditing Journal, Vol. 3 No. 1, pp. 38-53. Further reading Hines, R.D. (1982), ‘‘The usefulness of annual reports: the anomaly between the efficient markets hypothesis and shareholder surveys’’, Accounting and Business Research, Vol. 12 No. 48, pp. 296-309.

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

The information gap in annual reports

The information gap in annual reports

Jill Hooks School of Accountancy, Massey University, Auckland, New Zealand

David Coy and Howard Davey University of Waikato Management School, Hamilton, New Zealand Keywords Accountability, New Zealand, Electricity industry, Disclosure, Information

501 Received August 2001 Revised January 2002 Accepted April 2002

Abstract Following radical restructuring of the electricity industry in New Zealand since 1987, the government adopted a ‘‘light-handed’’ regulatory regime that used market-based methods involving competition and transparent accountability. This accountability is in part discharged through the provision of information in the corporate annual report. To assess the quality of that communication, a disclosure index was developed and applied to the annual reports of the 33 electricity retail and distribution companies which comprise the entire industry in New Zealand. The index was developed using the ideas and opinions of 15 experts representing broad stakeholder groups. This paper compares the resulting scores for the extent and quality of each index item with the level of importance of those items as stated by the panel. Many items are not adequately disclosed, resulting in an information gap between stakeholders’ expectations and the disclosures provided by the electricity companies. This paper identifies the items and the detail about them needed to close that gap.

Introduction Reform of the New Zealand electricity industry began in 1987. The objective was to make the industry more efficient by introducing market disciplines through competition. These reforms had a major impact on the retail and distribution sectors of the industry. Retail involves the wholesale purchase of electricity and its sale to consumers. Distribution (lines networks) involves the delivery of electricity from the transmission grid through local lines to consumers. Altogether, local networks provide electricity to 1.7m customers. Prior to the reforms, integrated retail and distribution entities were operated within a local body framework. These entities have now been corporatised and privatised. Most of the companies (70 percent) are owned by electricity consumers through trusts. The remainder are owned by local government (12 percent) or are public companies listed on the New Zealand Stock Exchange or a secondary market (18 percent). The companies are monitored by a light-handed regulatory regime, the cornerstone of which is the Electricity (Information Disclosure) Regulations 1994/1999. These regulations are enforced by the Ministry of Economic Development (MED). The authors acknowledge the extensive contributions made by the 15 members of the stakeholder panel whose expertise and opinions underpin this research. They thank the participants in presentations to the AAANZ Annual Conference, seminars at Massey University, Palmerston North and UNITEC, Auckland for helpful advice and suggestions. As always, final responsibility rests entirely with the authors.

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 501-522. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440577

AAAJ 15,4

502

In the absence of price control regulation, accountability through the disclosure of information is particularly important. Concern has been expressed about the accountability of electricity retail and distribution companies (Matthes, 1996; Macalister, 1997). This concern was compounded during the 1998 Auckland power crisis, when Mercury Energy’s inability to provide a continuous supply was perceived by many to be linked to reforms in the electricity industry, particularly in the areas of corporate governance and risk management[1]. A number of researchers (Winfield, 1978; Chang and Most, 1985; Boyne and Law, 1991) have noted the importance of the annual report as a vehicle for discharging accountability. Marston and Shrives (1991) concluded that the annual report is the most comprehensive document available to the public and is therefore the ‘‘main disclosure vehicle’’ (p. 196). Parker (1982) highlighted the importance of the annual report as a mass communication medium (p. 279). Although the annual report is not the only source of information about the performance of a company, it is considered to be an influential source because of its wide coverage and availability. Information communicated to stakeholders via the annual report is the focus of this research that is set within a framework of public accountability (Coy et al., 2001; Hooks et al., 2001). A public accountability paradigm acknowledges the rights of a broad group of stakeholders to information about company performance and activities. It is contended that accountability to those who are directly affected by a company’s performance is more important if the product is an essential commodity such as electricity. Accountability requires broadening the scope of disclosure beyond a financial focus to ensure that sufficient and meaningful qualitative information is also included in the corporate annual report. Coy and Pratt (1998) determined that the annual report has an important role in communicating and shaping the reality of the organisation in the public mind. How this reality is perceived depends in part on the extent and quality of information provided in the annual report. An evaluation of the extent and the quality of annual report disclosures is the subject of this paper. The purpose of the research is to identify stakeholders’ expectations of what should be disclosed in the corporate annual reports of New Zealand electricity retail and distribution companies from a public accountability perspective, and then to compare those expectations with the actual levels of disclosure in the 1998/99 annual reports of the 33 integrated retail and distribution companies comprising the population[2]. The second section of the paper describes the disclosure index developed for this purpose. The third section identifies the information gap by comparing disclosure quality with stakeholders’ expectations, and the fourth section provides a discussion and conclusion. The disclosure index An empirically developed disclosure index, known as the Electricity Annual Reports Index (or EARS Index), was used to assess the extent and the quality of

disclosure in the 1998/99 annual reports of the New Zealand electricity retail The information and distribution (or lines or network) companies. Disclosure indices have been gap in annual used as research tools by a number of researchers including Singhvi and Desai reports (1971), Baker and Haslem (1973), Buzby (1975), Barrett (1977), Benjamin and Stanga (1977), Firth (1978, 1979), McNally et al. (1982), Firer and Meth (1986), Robbins and Austin (1986), Chow and Wong-Boren (1987), Tong et al. (1990), 503 Adhikari and Tondkar (1992), Meek et al. (1995), Zarzeski (1996), Carson and Simnett (1997), Botosan (1997) and Craig and Diga (1998). The intention has often been to identify the motivation for the disclosure of voluntary items of information in corporate annual reports by testing relationships between various firm-specific variables drawn from agency theory, and voluntary disclosures. In most cases attention has been given to the number of disclosures (whether an item in a pre-prepared check list has been disclosed or not). Such items have been scored dichotomously (either 0 for non-disclosure, or 1 for disclosure). The index used in this research is more complex in that it also assesses the quality of the information disclosures. Other researchers (Barrett, 1977; Wiseman, 1982; Robbins and Austin, 1986; Wallace, 1988; Tong et al., 1990; Adhikari and Tondkar, 1992; Coy et al., 1993; Wallace and Naser, 1995; Botosan, 1997; Eng and Teo, 2000) assessed quality usually by allocating weightings for the importance of each item among the sub-elements related to its disclosure. The EARS index makes allowance for the relative importance of disclosure items and variations in the quality of individual disclosures in two ways. First, the relative importance of index items is brought into account using a system of weights. Weightings used by other researchers have generally been allocated by financial analysts on the basis that the indices have been developed from a decision-making perspective in which emphasis is generally on providing information to investors to support economic decisions (Cerf, 1961; Singhvi and Desai, 1971; Buzby, 1975; Benjamin and Stanga, 1977; Firth 1979; Firer and Meth, 1986; Robbins and Austin, 1986; Tong et al., 1990; Malone et al., 1993; Coy et al., 1993). For the EARS Index, with its broader accountability focus, the information items were selected and weighted by a panel of 15 stakeholders (auditor, lender, regulator, preparer, academic, environmentalist, electricity company employee, consumer, financial reporter, industry consultant, consumers’ institute representative, director, energy trusts executive, major electricity users group executive, financial analyst). The panel members were chosen because of their stakeholder position and their knowledge of what might be disclosed in corporate annual reports and the importance of those disclosures from an accountability perspective. Weightings for each information item were allocated using the following categories: . The item should not be disclosed. . Disclosure is of minor importance. . Disclosure is of intermediate importance.

AAAJ 15,4

504

. .

Disclosure is very important. Disclosure is essential.

The allocated weightings are a proxy of the information needs of the stakeholder panel. The weights were determined as the mean score of the 15 panellists’ opinions[3], where 0 ¼ ‘‘The item should not be disclosed’’ to 5 ¼ ‘‘Disclosure is essential’’[4,5]. The panel was involved with the selection and weighting of the items for the index, validation of the best-practice criteria, the evaluation and scoring process and with final discussions about the results. Second, the index makes allowance for differences in the quality of reporting of individual items. This is done by assessing how well each information item is reported compared with defined best practice criteria for that item. These criteria were developed from a number of sources, including the KPMG Model Annual Report 1999; The Annual Review of Corporate Reporting Practices 1995-1996 prepared by the Association for Investment Management and Research; Reporting Financial Performance: Current Developments and Future Directions, 1998 by the G4+1 group[6]; and the researchers’ knowledge developed partly from familiarity with extant reporting practice by the electricity industry. For example, the best practice criteria for the item ‘‘General development of the business’’ are: ‘‘identify major events within the past five years such as impact of industry reforms, mergers, disposition of assets, changes in mode of conducting business, frequency of price changes, number of customers, changes in number of customers’’. Disclosures that meet these criteria are awarded a score of 5; lesser scores are awarded pro rata for less detailed disclosures as judged by the assessors. A score of 4 is awarded if one section of the criteria is omitted from a report, and 3 is awarded if two sections are omitted or are incomplete, and so on. The EARS index comprises 67 information items arranged in eight categories including 13 mandatory items, 44 voluntary items, and ten that have elements of each. Cooke (1992) considered that, when the focus is disclosure to a wide range of users, an extensive list of items, both mandatory and voluntary, is appropriate. The disclosure index was developed within a framework of public accountability and each information item in the index was included because of its relevance to five tenets of accountability – compliance, efficiency, effectiveness, economy, and the future (Sutcliffe, 1985; Coy, 1995). The index was applied to the 1998/99 annual reports of all 33 retail and distribution companies which comprised the entire New Zealand industry at that time. The first author and another accountant evaluated each annual report independently. Where there were differences in scores awarded, an arbitration process followed until a consensus was reached for each item. The resulting scores were sent to the chief financial officers of each company for validation and are therefore considered to give a fair and reasonable indication of the level of disclosure in the reports[7].

Results The information The following tables compare the mean score[8] for extent and quality of gap in annual disclosure of each information item with its relative importance as rated by the reports stakeholder panel. The awarded scores range between 0 (no disclosure) and 5 (full disclosure). All 67 items in the EARS index may be disclosed with greater or less information content. 505 In Tables I-VIII, where there is a substantial difference between the mean score for a disclosure item for all 33 companies and its level of importance, the item is printed in italics, indicating the presence of an ‘‘information gap’’[9]. Items have been identified as mandatory or voluntary disclosures. Mandatory items are disclosures required by legally-backed New Zealand Financial Reporting Standards or the Companies Act (1993). Some disclosure items have both mandatory and voluntary elements, for example, in item 2.1 (Table II), Extent and quality of disclosure 1999 Mean score 0-5 1.0 1.1 1.2 1.3 1.4 1.5

Background about the company Brief narrative history of the company (v) Management’s objectives and strategies (v) Description of principal activities (v) Major contractual relationships (v) General development of business (v)

Level of importance

4.5 2.5 4.8 0.8 4.4

Intermediate Very important Table I. Intermediate Index item mean scores Very important compared with the Very important stakeholder panel’s Note: opinion about the level (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and of importance: voluntary elements. Items printed in italics indicate a substantial difference between the background about the mean score for the disclosure item and its level of importance (see note [9]) company

Extent and quality of disclosure 1999 Mean score 0-5 2.0 2.1

Level of importance

Information about management, major shareholders, related party transactions Names, backgrounds, affiliations, remuneration of 4.1 Very important directors (m/v) 2.2 Names, functions, remuneration of executive 3.6 Intermediate Table II. management (m/v) 2.3 Major owners of company’s stock and number of 4.4 Very important Index item mean scores compared with the shares owned (v) stakeholder panel’s 2.4 Number of shares owned by directors (v) 4.7 Very important 2.5 Related party transactions (m) 4.4 Very important opinion about the level of importance: Note: information about (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and management, major voluntary elements. Items printed in italics indicate a substantial difference between the shareholders, related mean score for the disclosure item and its level of importance (see note [9]) party transactions

AAAJ 15,4

Extent and quality of disclosure 1999 Mean score 0-5 3.0 3.1

506

3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10

Table III. Index item mean scores compared with the stakeholder panel’s opinion about the level of importance: assets

3.11 3.12 3.13 3.14 3.15

Assets Assets by segment: energy, contracting, network, generation (m/v) Details of measurement basis of assets (m) ODV valuation of network assets (v) Amount of asset revaluation each year (m) Fixed assets purchased and sold in current year (v) Fixed asset details (v) Investments (m/v) Goodwill (m) Capitalised interest (m) Location, nature, productive capacity of principal plant (v) % plant capacity utilized (v) Depreciation method (m) Capital expenditure planned (v) Current assets: debtors, inventories, cash (m) Asset management plan (v)

1.1

Very important

4.7 1.9 3.9 3.7

Essential Essential Very important Intermediate

3.6 4.3 2.7 2.3 2.2

Essential Very important Very important Very important Intermediate

1.7 4.0 1.6 4.8 1.9

Intermediate Essential Very important Very important Intermediate

Note: (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and voluntary elements. Items printed in italics indicate a substantial difference between the mean score for the disclosure item and its level of importance (see note [9])

Extent and quality of disclosure 1999 Mean score 0-5 4.0 4.1 Table IV. Index item mean scores compared with the stakeholder panel’s opinion about the level of importance: debt

Level of importance

4.2

Debt Total debt outstanding and debt repayment schedules (m) Current liabilities: creditors and provisions (m/v)

Level of importance

4.6

Very important

4.7

Very important

Note: (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and voluntary elements. Items printed in italics indicate a substantial difference between the mean score for the disclosure item and its level of importance (see note [9])

disclosure of directors’ remuneration is mandatory but disclosure of backgrounds and affiliations is voluntary. Background about the company All five items in this category ‘‘Background about the company’’ are voluntary disclosures (see Table I). Only two companies achieve best practice standard for the item ‘‘Management’s objectives and strategies’’ which is rated as ‘‘very

Extent and quality of disclosure 1999 Mean score 0-5 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12

Financial, operating and performance related data Breakdown of sales revenue by segment (m/v) Sales volume by segment (m/v) Breakdown of expenses by segment (m/v) Profit for each segment as above (m/v) Other earnings (m/v) Cost of electricity purchased, generated, distributed (v) Major elements of costs (v) Details of unusual or non-recurring items (m) Goodwill written off (m) Funding costs (m) Discussion of results for past year (v) Historical summary of operating and financial data (5 years) (v)

2.5 1.2 1.3 1.4 4.6 1.1

Level of importance

Essential Very important Very important Very important Very important Very important

The information gap in annual reports 507

3.3 4.4 3.7 4.7 3.7 2.7

Essential Essential Essential Table V. Very important Very important Index item mean scores compared with the Intermediate stakeholder panel’s opinion about the level Note: of importance: (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and financial, operating and voluntary elements. Items printed in italics indicate a substantial difference between the performancemean score for the disclosure item and its level of importance (see note [9]) related data

Extent and quality of disclosure 1999 Mean score 0-5 6.0 6.1 6.2 6.3

Forward-looking information Forecast of next years profits/earnings (v) Discussion on major factors influencing next year (v) Comparison of actual business performance to previously disclosed information (v)

Level of importance

0.5 2.3

Intermediate Very important

3.1

Very important

Note: (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and voluntary elements. Items printed in italics indicate a substantial difference between the mean score for the disclosure item and its level of importance (see note [9])

important’’ by the stakeholder panel. Furthermore, 16 (48 percent) of the companies make hardly any reference to the item at all. ‘‘Major contractual relationships’’ are also considered by the stakeholder panel to be ‘‘very important’’ and yet 18 (55 percent) of the companies give no information at all and a further 9 (27 percent) provide only a brief, indirect reference to a relationship. Overall, for this category, ‘‘Brief narrative history of the company’’, ‘‘Description of principal activities’’ and ‘‘General development of business’’ are well reported.

Table VI. Index item mean scores compared with the stakeholder panel’s opinion about the level of importance: forwardlooking information

AAAJ 15,4

Extent and quality of disclosure 1999 Mean score 0-5 7.0

508

7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9

Performance measures Financial measures Debt to equity ratio (v) Funding cost cover (times interest covered) (v) Net tangible assets per share(v) EBIT/average net funds employed(v) Dividend per share (cents) (v) Return on total assets (v) Return on equity (v) Accounting rate of profit (v) Overhead costs per retail customer (v)

1.2 0.1 0.8 0.6 1.5 0.5 1.2 0.4 0.0

Intermediate Intermediate Intermediate Intermediate Intermediate Very important Very important Very important Intermediate

7.10 7.11

Pricing measures Average domestic power bill (v) Pricing information (v)

1.3 0.0

Intermediate Very important

0.7 2.5

Very important Very important

1.5

Very important

0.4

Very important

0.4 1.3

Very important Very important

0.0

Intermediate

0.3 1.0

Very important Very important

7.12 7.13 7.14 7.15 7.16 7.17 7.18 Table VII. Index item mean scores compared with the stakeholder panel’s opinion about the level of importance: performance measures

Level of importance

7.19 7.20

Efficiency measures Total number of interruptions (v) Average total duration of interruptions of supply per customer (v) Average number of interruptions of supply per customer: 5-10yrs (v) Number of faults per 100km of prescribed voltage line: 5-10yrs (v) Total costs costs per kilometre (v) Total costs per customer (v) Market measures Average consumption kwh per customer (5 years) (v) Number of customers by sector (v) Change in market share in major areas of activity (v)

Note: (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and voluntary elements. Items printed in italics indicate a substantial difference between the mean score for the disclosure item and its level of importance (see note [9])

Information about management As shown in Table II, four of the five items comprising the category ‘‘Information about management, major shareholders and related party transactions’’ are regarded as ‘‘very important’’ and the annual report disclosures meet the panel’s expectations. The items themselves are a mixture of mandatory and voluntary disclosures. Ten companies fail to give details of the voluntary item ‘‘names and functions of executive management’’. Although there are some shortcomings in respect of related party disclosures, most companies (82 percent) meet their mandatory reporting requirements in this

category. Two trust-held companies do not disclose ‘‘directors’ remuneration’’ The information and eight trust-held companies do not disclose ‘‘employee remuneration above gap in annual $100,000’’. The Companies Act S211 (3) permits non-disclosure of information reports about directors’ and employees’ remuneration if all shareholders agree that the annual report need not disclose this information. In the case of trust-held companies, the trustees are the shareholders and they had unanimously agreed 509 not to disclose this information. Viewed from a perspective of public accountability, this obscurity gives cause for concern. Assets As shown in Table III, the stakeholder panel rate four of the 15 items in the ‘‘assets’’ category as ‘‘essential’’, and a further seven as ‘‘very important’’. The ‘‘essential’’ items are: ‘‘details of the measurement basis of assets’’, ‘‘optimised deprival value (ODV)[10] valuation of network assets’’, ‘‘fixed asset details’’ and ‘‘depreciation method’’. For three of these, the quality of disclosures is good and stakeholders’ expectations are more or less met. However, disclosures about the ‘‘ODV valuation of network assets’’ fall a long way short. The Electricity (Information Disclosure) Regulations 1999 require electricity lines companies to report the ODV of their network assets to the MED. ODV details provided in the annual reports are sparse and none of the companies meets best practice disclosure requirements. In most cases, current ODV is given, but any previous ODV figures are omitted, making it impossible to study changes in valuations. Note that notwithstanding the stakeholder panel’s opinion that disclosure of this information is essential for annual reports, it is a mandatory disclosure only to the MED, and voluntary for annual reports. For four of the seven ‘‘very important’’ items in this category, stakeholders’ expectations are met or exceeded by the 1999 annual reports: ‘‘amount of asset revaluation each year’’, ‘‘investments’’, ‘‘goodwill’’ and ‘‘current assets’’. But expectations are not met in the case of the other three items; ‘‘assets by segment’’, ‘‘capitalised interest’’ and ‘‘capital expenditure planned’’. Extent and quality of disclosure 1999 Mean score 0-5 8.0 8.1 8.2 8.3 8.4 8.5

Other information Information on accounting methods used (m) Contingent liabilities (m) Forward contracts for committed purchases (v) Capital contributions by customers (v) Dividend distribution policy (v)

4.8 4.4 4.8 2.3 4.0

Level of importance

Essential Very important Intermediate Table VIII. Intermediate Index item mean scores Intermediate compared with the Note: stakeholder panel’s (m) ¼ mandatory disclosure; (v) ¼ voluntary disclosure; (m/v) ¼ disclosure has mandatory and opinion about the level voluntary elements. Items printed in italics indicate a substantial difference between the of importance: other mean score for the disclosure item and its level of importance (see note [9]) information

AAAJ 15,4

510

The mean score of 1.1 for the item ‘‘assets by segment’’ indicates a substantial information gap; 24 of the companies (73 percent) fail to disclose any information of this type. The year ended 1999 is the final reporting year when companies have integrated energy, generation and network businesses. The Electricity Reform Act 1998 required electricity companies to divest either their retail business or their lines business. In subsequent years they will be either generation and/or retail companies, or distribution (lines/network) companies. In future, a breakdown between contracting, network (gas and electricity) and other businesses (for lines companies) or between generation and energy (for retail companies) will be important. Conventionally, segmental reporting requires the reporting of operations that are in different industries or countries to be disclosed separately. In this research, it is contended that best practice disclosure would require segmentation that relates to the way in which the business is organised. Some companies own discrete networks and separate disclosure of the value of each of these is important for accountability purposes. None of the companies with discrete networks show these as segments. This makes it impossible for interested stakeholders to discern the resources allocated to the different activities, and hence to make any reasonable judgements about the level of profitability, pricing or the extent of cross-subsidies. ‘‘Capitalised interest’’ (mean score 2.3) and ‘‘capital expenditure planned’’ (1.6) also fall well short of stakeholders’ expectations, even after making allowance for those companies that do not hold any capitalised interest[11]. Similarly, little information is disclosed on capital expenditure planned for the year compared to actual capital expenditure. Most of the disclosures related to this item were for capital commitments for the next financial year. Other shortcomings in the disclosure of assets in the 1999 annual reports are also noted. Only 17 (53 percent) of the companies provide a clear statement of the opening balance of the asset revaluation reserve, current revaluations and the closing balance. Sometimes asset revaluations are included in the company accounts, other times they are not[12]. As a result of the 1998/99 asset revaluations, network asset values increased significantly (for example, Eastland Energy by 115.7 percent; Westpower 54 percent; Counties Power and Otago Power 38 percent)[13]. There is no consistent reporting of this change. Eastland Energy’s annual report shows the distribution system at cost. The chief executive’s report refers to the $32m increase in value based on the ODV of network assets but this information is not disclosed as a footnote to the financial statements. Westpower recognises the $15.3m increase in value and reports it in the statement of movements in equity. Counties Power gives no value at all for the distribution system and Otago Power reports the distribution system at cost and there is no mention of ODV. Disclosure of details of ‘‘fixed assets purchased and sold’’ is rated as of ‘‘intermediate importance’’. Current disclosure is generally limited to totals in the statement of cash flows with disclosure of a total gain or loss on sales shown in the statement of financial performance. However, the best-practice

benchmark for full disclosure requires details of individual sales or purchases. The information Only one company provides this information. gap in annual Best practice disclosure requires the reporting of asset cost, amount at reports valuation, accumulated depreciation and details of the useful lives of the assets, but companies generally fail to include the historical cost of assets that have been revalued or to disclose the estimated life of assets. This latter aspect is 511 particularly important in the case of distribution assets (networks) as it is possible to extend the useful life in order to reduce the depreciation allocation. As would be expected, ‘‘depreciation methods’’ are described in the statement of accounting policies but many companies report depreciation over such an excessive range of years as to make the information impossible to use (for example, 10-80 years). It appears that some companies subsume ‘‘goodwill’’ into a separate item, identifiable intangible assets, in their 1999 reports. In terms of the benchmark, ‘‘goodwill’’ is required to be shown separately at cost less accumulated amortisation. Where goodwill is disposed of on the sale of part of the company, the opening balance should be stated, less the amount related to the sale of part of the company, less amortisation. ‘‘Location, nature and productive capacity of plant’’ has become more important as companies diversify and own networks or generation plants in various parts of the country. Only two companies give clear maps to show the location of their plant/networks. A total of 22 (66 percent) companies do not disclose information about the ‘‘percentage of plant capacity utilized’’, and some electricity company chief financial officers (CFOs) comment that this is sensitive information[14]. However, nine (28 percent) companies did show this information clearly in graphical form. Although these two items are regarded as only of intermediate importance, they assist in establishing an overall view of the company and its operations. ‘‘Asset management plan’’ details were very sparse with 18 (55 percent) having either very little or no information related to the maintenance of assets, particularly network assets. This is a key aspect of the Electricity (Information Disclosure) Regulations 1999 but the information provided for that forum is detailed, complex, extremely lengthy, and difficult to understand. Companies could therefore usefully (for stakeholders) provide a one-page summary in their annual report of the detailed plan submitted to the MED. This should include a statement that asset management plans which meet the guidelines developed by the ministry are in place for every one of the company’s significant systems; a statement that the maintenance and replacement required by these plans during the current year has been completed; and a list of any deferred maintenance or deferred capital replacements (Institute of Chartered Accountants of New Zealand (ICANZ), 1998). In summary, there are some substantial shortcomings in the reporting of assets by the electricity companies, with the largest information gaps relating to the items ‘‘ODV valuation of network assets’’, ‘‘assets by segment’’ and ‘‘capital expenditure planned’’. Although these disclosures are not mandatory

AAAJ 15,4

512

for annual reports, nevertheless, the stakeholder panel did regard them all as being either ‘‘very important’’ or ‘‘essential’’ disclosures from an accountability perspective. Debt Both items in the ‘‘debt’’ category are mandatory disclosures and are well disclosed and clearly meet stakeholders’ expectations (see Table IV). Financial, operating and performance-related data The aim of this section of the EARS index is to evaluate the transparency of performance reporting. A significant part of this category relates to segmental information, with best practice disclosure requiring segmentation according to the way in which the firm is organised and managed. The segments to be reported are based on organisational units for which financial records would normally be maintained for management purposes. Although this is not currently GAAP in New Zealand, it is considered best practice (AIMR, 1997). In addition, S44 (5)(d) of the Energy Companies Act 1992 requires consolidated financial statements to include an operating statement in respect of each significant activity. This information must be available to shareholders. This means that the information is not generally publicly available as the reporting requirement for trust-owned companies is to the trustees. Although retail and lines businesses are now separated, companies may operate both natural monopoly (lines) activities and any other competitive activities other than electricity retailing or generation, for example, communication networks. Segmental information, therefore, is required to assist in identifying whether the natural monopoly component of lines businesses has excessive costs or profits. The majority of companies state that they operate in only one industry. However, some operate discrete networks in different regions of New Zealand. Companies also have network (in some cases electricity and gas) and contracting businesses, or retail and generation businesses. As shown in Table V, of the 12 items in the ‘‘financial, operating and performance related data’’ category, six are rated as ‘‘essential’’ or ‘‘very important’’ and have a substantial information gap. The two ‘‘essential’’ items are ‘‘breakdown of sales revenue by segment’’ (mean score 2.5) and ‘‘major elements of cost’’ (3.3). There is little segmentation of sales revenue, sales volume, expenses and profit in the electricity company annual reports. Only five of the 32 companies give a full segmental breakdown, 12 provide no breakdown of sales revenue and 20 do not disclose breakdown of expenses by segment. Few companies receive the maximum score of 5 for ‘‘major elements of costs’’ due to their failure to disclose marketing costs, salaries, restructuring, and asset maintenance costs. With the exception of restructuring costs, these are required by the Electricity (Information Disclosure) Regulations 1999 so their

inclusion in the annual report is appropriate for best-practice reporting and is The information considered ‘‘essential’’ by the stakeholder panel. gap in annual ‘‘Cost of electricity purchased or generated’’ is generally not disclosed by 25 reports companies (79 percent) and has a mean score of 1.1. Thus annual report users are unable to calculate gross margin, an essential item for accountability purposes and considered ‘‘very important’’ by the panel. The standard of 513 ‘‘discussion of results for the past year’’ is good and 12 (36 percent) meet bestpractice disclosure requirements. However, 16 (48 percent) receive a score of 3 marks or less, indicating room for improvement in this area. The related item, ‘‘historical summary of operating and financial data’’ is not as well reported and only seven companies meet best-practice disclosure. Some companies provide no summary at all (21 percent). The stakeholder panel rate this item as of only intermediate importance. However, the inclusion of a statement of trends in annual reports is considered important in terms of best-practice disclosure[15]. It is acknowledged that most of the best practice criteria used as a benchmark for this category relate to voluntary disclosures. However, as noted by Olsson (1980), legislative requirements of annual reporting represent only a part of the informational needs of stakeholders. Significant areas for improvement of disclosure in respect of this category relate to: . segmental information which reflects the organisational structure of the company, e.g. networks (split gas and electricity) contracting, other; . cost of electricity purchased / distributed; . costs of salaries, asset maintenance, marketing, and restructuring; . a five-year historical summary of operating and financial data including financial and efficiency measures. Nine of the items in this category are voluntary or part-voluntary disclosures and the three mandatory items ‘‘unusual items’’, ‘‘goodwill written off’’ and ‘‘funding costs’’ are well disclosed. Forward-looking information Results for the three items comprising ‘‘Forward-looking information’’ are shown in Table VI. Information related to the future is necessarily speculative. Opinions differ on the relevance of disclosing this information. However, one of the elements identified within an accountability framework is equity (fairness). Lev (1988) was concerned about reducing information asymmetry in order to achieve equity. In his opinion disclosure of earnings’ forecasts should be mandated. The New Zealand Statement of Concepts (cl. 3.8) supports the disclosure of predictive results for operations, financial position and cash flows (ICANZ, 1993). Only one company included a ‘‘forecast of next year’s profits/ earnings’’ and few companies included any worthwhile discussion on ‘‘major factors influencing next year’’. Most companies do not even make a general comment on the expected direction of change for the coming year’s earnings,

AAAJ 15,4

514

changes in legislation or market conditions. However, Y2K checks and concerns are often carefully described. It is contended that information about probable future development that is based on well-founded expectations is a significant component of an accountability regime. For accountability to be effective, the actual performance of the organisation needs to be compared with its objectives and the variance between the two needs to be communicated to stakeholders. The New Zealand Statement of Concepts (cl.3.5) states that accountability requires that reports identify financial or non-financial objectives and targets and measure actual achievements against those objectives and targets. In addition, the Energy Companies Act 1992 (S39(2)(e)) requires the board to report, in a statement of corporate intent, the performance targets and other measures by which performance has been judged in relation to objectives. Therefore, this information should be available for incorporation into the more comprehensive document, the corporate annual report. Only 16 (48 percent) of the companies compare actual results to targets according to best practice disclosure. However, eight (24 percent) give no information in this respect at all. All companies should include information about long-term financial goals. These should also be included as part of the highlighted key ratios. None of the ‘‘Forward looking information’’ items identified by the stakeholder panel and included in the EARS index is a mandatory disclosure for electricity company annual reports. Performance measures The panel identified 20 items relating to performance measures that should be disclosed in annual reports, all of which are voluntary (see Table VII). None was regarded as ‘‘essential’’, 12 are rated as ‘‘very important’’ and eight as of ‘‘intermediate importance’’. The performance measures are classified into four groups: (1) financial (nine items); (2) pricing (two items); (3) efficiency (six items); and (4) market (three items). Overall, the disclosures about performance measures in the 1999 annual reports are the weakest of all eight categories in the index and have the most substantial information gap between stakeholders’ expectations and information currently being provided. The highest score is attributed to ‘‘average total duration of interruptions of supply per customer’’ that gains 2.5 (out of the maximum possible of 5), but 17 out of the 20 items score only between 0.0 and 1.3. Disclosures about almost all performance measures are poor. A few companies disclose common financial measures such as debt to equity ratio, dividends per share, return on equity, but these are seldom given for a five-year

period to enable readers to examine trends. It was possible for report users to The information calculate some of the measures themselves, at least for the current year, but this gap in annual should not be necessary. reports Consumers are interested to know how much electricity costs, how often the power goes off and how quickly supply is restored. ‘‘Pricing measures’’ and ‘‘change in market share’’ are applicable to retail activities. These measures are 515 significant as part of a total accountability package for this ‘‘competitive’’ sector of the industry, which is largely exempt from the accountability requirements of the Electricity (Information Disclosure) Regulations 1999. The two pricing measures: ‘‘average domestic power bill’’ and ‘‘pricing information’’ are considered by the panel to be of intermediate importance, and very important respectively. Efficiency measures relate to the performance of distribution (lines) companies and are regarded as very important. Some companies provided best practice levels of disclosure about ‘‘average total duration of interruptions of supply’’ or ‘‘average number of interruptions of supply per customer’’ but other efficiency measures or market measures are not provided. On their own performance measures mean little. To have meaning they must be compared with performance measures of other firms in the industry or with the same firm over time. This is currently not possible from the information disclosed in the annual reports of the electricity retail and distribution companies. Other information The two items regarded as most important in the ‘‘other information’’ category are mandatory disclosures and are well reported by the companies (see Table VIII). These are ‘‘information on accounting methods used’’ (essential, score 4.8) and ‘‘contingent liabilities’’ (very important, 4.4). ‘‘capital contributions by customers’’ is generally treated as income but not disclosed separately, hence the low scores for this item. Most companies fail to state an accounting policy for this item. As accounting practice can differ from inclusion of capital contributions as part of income to a deduction from the asset to which the contribution relates, disclosure of the accounting policy is important. Discussion and conclusion The New Zealand Government chose to regulate the restructured electricity industry with a ‘‘light-handed’’ approach using yardstick monitoring. Accountability through information disclosure is the cornerstone of the regulatory regime. Electricity distribution companies must provide specified data on their performance to the MED. The aim of the disclosures is to increase transparency and to provide a means for interested parties to detect anticompetitive behaviour. However, concern has been expressed at the lack of truly comparable data against which electricity power companies can be assessed (Matthes, 1996). The chief executive of the Consumers’ Institute stated

AAAJ 15,4

516

that ‘‘only big businesses have the clout to tear the information apart and make sense of it all’’ (Macalister, 1997, p. 36). The MED (1999) reported that the performance of line owners in meeting the information disclosure requirements for the 1999 year had been, almost without exception, highly unsatisfactory. United Networks expressed a concern that ‘‘the current regulated disclosures are very remote from consumers’’ (United Networks, 2000, p. 4). In addition, the Ministerial Inquiry into the Electricity Industry (June, 2000) stated that, despite the disclosure regime which has been in place since 1994, the information that is available on the performance of distribution companies is poor. There is, therefore, a continuing demand for information about electricity company activities in order to discharge their accountability requirements. The aim of this research is to enhance the usefulness of the annual report as a vehicle for discharging accountability obligations and as a mechanism for communicating information to users. The intention is to ensure that the quality of information provided does not differ significantly across electricity companies. The informational side of being accountable to the general public can be discharged in many ways including the use of press releases, the reporting of meetings with analysts and employees, and, more commonly nowadays, using corporate Websites, in addition to the annual report. The corporate annual report is the most comprehensive of the communication channels and has the potential to make information easily and routinely available in a single document. Lang and Lundholm (1993) argue that measurement of disclosure levels in corporate annual reports provides a surrogate indicator of a company’s general level of accountability. Information disclosure is not without cost (see Craswell and Taylor, 1992; McKinnon and Dalimunthe, 1993). The cost-benefit relationship of providing additional information is an important consideration but it is thought that provision of voluntary annual report disclosures has net benefits in that stakeholders’ information needs are met and accountability requirements are discharged. This research acknowledges the combined roles of the regulator (MED) and electricity companies in what is intended to be a self-regulatory environment that emphasises accountability. Accountability also focuses upon the relationship between the corporation and users of its annual reports. In a narrow sense, accountability relates to the requirement to report to owners. This research emphasises the public interest aspect of reporting (Cooper and Sherer, 1984) and the need to provide relevant and meaningful information to stakeholders. To be worthwhile, information about company performance has to meet the specific needs of different stakeholders. Their information needs are a focus of this paper. Stakeholders’ expectations of annual report disclosures were understood in the context of the organisation and the industry itself and acknowledge the dominant position of the sections of the industry that have monopoly characteristics.

The industry-specific EARS disclosure index developed for this study The information includes voluntary and mandatory disclosures. Voluntary disclosure in the gap in annual annual reports of New Zealand companies has increased (Ryan, 1990, 1993). reports However, Tables I-VIII show that many items of voluntary information, which stakeholders believe to be very important or even essential, are not being adequately disclosed. These tables show the nature and size of the information 517 gap. Many of the annual reports present limited amounts of information. There is therefore an opportunity for expanding the extent and the quality of disclosure in the annual reports of electricity retail and distribution companies. The main areas where companies should focus their attention in order to improve the quality of their disclosures in line with best practice reporting and the needs of stakeholders are: . management objectives and strategies (very important); . major contractual relationships (very important); . segmental information: assets, revenue, expenses, profit (very important/essential); . optimised deprival value of assets (essential); . capital expenditure planned (actual v. budget) (very important); . capitalised interest (very important); . cost of electricity purchased/generated (retail companies) (very important); . major elements of costs: marketing, salaries, restructuring, asset maintenance (essential); . forecast of next year’s profits/earnings (intermediate); . discussion on major factors influencing next year (very important); and . all performance measures: financial, pricing, efficiency, market (intermediate/very important). Only one of the items listed above is a mandatory disclosure for electricity annual reports, ‘‘capitalised interest’’. It is not surprising that most of the items representing the ‘‘information gap’’ are of a voluntary nature. The information expectations of stakeholders, are, after all, likely to focus on having access to information they do not have rather than information currently available. The value of the study lies in identifying the information items, currently not available or under-reported in annual reports, that serious stakeholders would like to see. Parliamentarians and other regulators may use this knowledge when modifying accountability regimes in future. Report preparers in companies wishing to be more accountable for their own sake, rather than merely meeting the minimum statutory and other regulatory obligations, receive guidance on changes they may make voluntarily. This study identifies information gaps in the annual reports of the New Zealand electricity retail and distribution companies. Although the EARS

AAAJ 15,4

518

index is necessarily industry and country specific, we offer the approach used here as a generic method that may be applied to other industries and in other countries. In summary, this involved four discrete steps (1) Establishment of a stakeholder panel to identify the items to include in the index and their relative importance. (2) Construction of the index. (3) Application of the index to the annual reports. (4) Feedback of the results to the panel and report preparers to validate the findings. The results, and especially the information gaps that have been identified, should assist report preparers, accounting regulators and legislators in improving reporting in future to the extent that it may come closer to meeting public expectations. Radebaugh and Gray (1997) note the increased interest in corporate accountability and consider that it reflects a demand by a wide audience, including the general public, for information. Tornqvist (1999) identified this increased demand as related in part to forward-looking information and segmental reporting. These two aspects as well as the others listed above go to the core of being properly accountable. Raffournier (1995) noted that studies on annual report disclosure have not tested the relationship between extent of disclosure and ownership diffusion. There is therefore an opportunity to further explore the relationship between different company structures (public, trust-owned, and council-owned companies) and the level of disclosure. This research has involved listening to the information needs of potential users of electricity company annual reports. Their concern is to make electricity retail and distribution companies and their activities more transparent by providing sufficient, understandable and easily accessible financial and nonfinancial information that enables stakeholders to fairly assess performance and progress, including strategies for setting prices. A prime way of achieving this is through more thorough and comprehensive reporting in the corporate annual report. This research provides electricity retail and distribution companies’ report preparers with clear indications on the kinds of information and disclosures needed to reduce the information gaps in future. Notes 1. Inquiry into the Auckland power supply failure, Ministry of Commerce, 15 May 1998. 2. For most of the 1998/99 year the electricity companies were integrated retail and distribution companies. The Electricity Industry Reform Act (1998) required these companies to split and become either line companies or retail companies. Only three companies in the analysis chose to become retailers and sold their line networks. This change in activity was taken into account in scoring the annual reports. 3. Panellists’ opinions were captured initially using a postal questionnaire, and then followed up with individual interviews that sought explanations and clarifications of their opinions.

4. The highest weight of an EARS index item is 3.90 (measurement basis of assets; accounting methods used) and the lowest weight is 1.90 (brief narrative history of the company). 5. In keeping with the public accountability perspective adopted in this research, equal weight was attached to the opinions of all panel members. Other researchers may give more weight to the views of some groups who they deem to be more important, e.g. researchers adopting a decision making perspective may give more weight to the views of those who are expected to exert greater influence on the buying and selling of stocks and shares, such as financial analysts. 6. G4+1 is made up of representatives from Australia, New Zealand, UK, USA and the International Accounting Standards Committee. The group’s views on performance reporting in the future are explained in Johnston and Robbins (1992). 7. Full details of the EARS index are published in Hooks et al. (2001). 8. Mean scores have been calculated and reported here. Technically, the index scores are ordinal rather than ratio data and the median is the appropriate indicator of mid-points. However, other researchers have used the mean rather than the median to summarise similar Likert-type data (Buzby, 1975; Dinius and Rogow, 1988). The differences between the means and the medians in this data set are not material. 9. A ‘‘substantial’’ difference is determined as follows: an item classified as ‘‘essential’’ with a mean score for all 33 reports of < 3:5; ‘‘very important’’ < 2:5; and ‘‘intermediate importance’’ < 0:5. 10. Optimised deprival value is the lesser of optimised depreciated replacement cost or economic value on a discounted cash flow basis (such being the greater of value in use or disposal value.) It stems from the premise that the value of an asset to an entity is the amount that the entity would lose if deprived of that asset. 11. In calculating company disclosure index scores and index item means, index items that do not apply to a company are ignored. 12. In order to recognise the value of the asset to the entity, increments in value are credited directly to an asset revaluation reserve for that class of asset. This reserve forms part of the equity of the firm. 13. Report to the Minister of Energy, Inquiry into the Electricity Industry, June 2000 pp. 15-16. 14. Opinions were collected from electricity company chief financial officers via feedback received following the dissemination of the EARS index results to them. 15. KPMG Model Annual Report 1999; AIMR Annual Review of Corporate Reporting Practices (1996). References Adhikari, A. and Tondkar, R. (1992), ‘‘Environmental factors influencing accounting disclosure requirements of global stock exchanges’’, Journal of International Financial Management & Accounting, Vol. 4 No. 2, pp. 75-105. Association for Investment Management and Research (AIMR) (1997), Corporate Information Committee Report, Financial Analysts Federation, New York, NY. Baker, H.K. and Haslem, J.A. (1973), ‘‘Informational needs of individual investors’’, The Journal of Accountancy, November, pp. 64-9. Barrett, M.E. (1977), ‘‘The extent of disclosure in annual reports of large companies in seven countries’’, The International Journal of Accounting, Spring, pp. 1-25. Benjamin, J.J. and Stanga, K.G. (1977), ‘‘Differences in disclosure needs of major users of financial statements’’, Accounting and Business Research, Summer, pp. 187-92.

The information gap in annual reports 519

AAAJ 15,4

520

Botosan, C.A. (1997), ‘‘Disclosure level and the cost of equity capital’’, The Accounting Review, Vol. 72 No. 3, pp. 323-49. Boyne, G. and Law, J. (1991), ‘‘Accountability and local authority annual reports: the case of Welsh District Councils’’, Financial Accountability & Management, Vol. 7 No. 4, pp. 179-94. Buzby, S.L. (1975), ‘‘Company size, listed versus unlisted stocks, and the extent of financial disclosure’’, Journal of Accounting Research, Spring, pp. 17-35. Carson, E. and Simnett, R. (1997), ‘‘Voluntary disclosure of corporate governance information’’, working paper series, School of Accounting, University of New South Wales, Sydney. Cerf, A.R. (1961), Corporate Reporting and Investment Decisions, The University of California Press, Berkley, CA. Chang, L. and Most, K. (1985), The Perceived Usefulness of Financial Statements for Investors’ Decisions, Florida International University Press, Miami, FL. Chow, C.W. and Wong-Boren, A. (1987), ‘‘Voluntary financial disclosure by Mexican corporations’’, The Accounting Review, Vol. LXII No. 3, pp. 533-41. Cooke, T.E. (1992), ‘‘The impact of size, stock market listing and industry type on disclosure in the annual reports of Japanese listed corporations’’, Accounting and Business Research, Vol. 22 No. 87, pp. 229-37. Cooper, D.J. and Sherer, M.J. (1984), ‘‘The value of corporate accounting reports: arguments for a political economy of accounting’’, Accounting, Organizations and Society, Vol. 9 No. 3/4, pp. 207-32. Coy, D.V. (1995), ‘‘A public accountability index for annual reporting by NZ universities’’, doctoral thesis, University of Waikato, Hamilton. Coy, D. and Pratt, M. (1998), ‘‘An insight into accountability and politics in universities: a case study’’, Accounting, Auditing & Accountability Journal, Vol. 11 No. 5, pp. 540-61. Coy, D., Fischer, M. and Gordon, T. (2001), ‘‘Public accountability: a new paradigm for college and university annual reports’’, Critical Perspectives on Accounting, Vol. 12 No. 1, pp. 1-31. Coy, D., Tower, G. and Dixon, K. (1993), ‘‘Quantifying the quality of tertiary education annual reports’’, Accounting and Finance, Vol. 33 No. 2, pp. 121-9. Craig, R. and Diga, J. (1998), ‘‘Corporate accounting disclosure in ASEAN’’, Journal of International Financial Management and Accounting, Vol. 9 No. 3, pp. 246-74. Craswell, A.T. and Taylor, S.L. (1992), ‘‘Discretionary disclosure of reserves by oil and gas companies: an economic analysis’’, Journal of Business Finance and Accounting, January, pp. 295-308. Dinius, S.H. and Rogow, R.B. (1988), ‘‘Application of the Delphi method in identifying characteristics big eight firms seek in entry level accountants’’, Journal of Accounting Education, Vol. 6, pp. 83-101. Eng, L.L. and Teo, H.K. (2000), ‘‘The relation between annual report disclosures, analysts’ earnings forecasts and analyst following: evidence from Singapore’’, Pacific Accounting Review, Vol. 11 No. 2, pp. 219-39. Firer, C. and Meth, G. (1986), ‘‘Information disclosure in annual reports in South Africa’’, Omega International Journal of Management Science, Vol. 14 No. 5, pp. 373-82. Firth, M. (1978), ‘‘A study of the consensus of the perceived importance of disclosure of individual items in corporate reports’’, The International Journal of Accounting, Vol. 14 No. 1, pp. 57-70. Firth, M. (1979), ‘‘The impact of size, stock market listing, and auditors on voluntary disclosure in corporate annual reports’’, Accounting and Business Research, Autumn, pp. 273-80. Hooks, J., Coy, D. and Davey, H. (2001), ‘‘Information disclosure in the annual reports of New Zealand electricity retail and distribution companies: preliminary findings’’, Department of

Accounting, Working Paper Series, No. 70, University of Waikato Management School, Waikato. Hooks, J., Coy, D. and Davey, H. (2001), ‘‘The annual reports of New Zealand electricity companies: assessing quality’’, Pacific Accounting Review, Vol. 13 No. 2, pp. 35-69. Institute of Chartered Accountants of New Zealand (ICANZ) (1993), Statement of Concepts for General Purpose Financial Reporting, Wellington. Institute of Chartered Accountants of New Zealand (ICANZ) (1998), Submission to the Ministry of Commerce on Proposed Amendments to the Electricity Information Disclosure Regulations 1994, October, Wellington. Johnston, L.T., and Robbins, W.A. (1992), Reporting Financial Performance: Current Developments and Future Directions, FASB, Norwalk, CT. KPMG (1999), Model Annual Report: Diverse Group Ltd, KPMG Technical Department, Auckland. Lang, M. and Lundholm, R. (1993), ‘‘Cross-sectional determinants of analysts’ ratings of corporate disclosure’’, Journal of Accounting Research, Vol. 31 No. 2, pp. 246-71. Lev, B. (1988), ‘‘Toward a theory of equitable and efficient accounting policy’’, The Accounting Review, Vol. LXIII No. 1, pp. 1-22. Macalister, P. (1997), ‘‘Power failure: why isn’t competition working in the electricity sector?’’, Management, June, pp. 33-41. McKinnon, J.L. and Dalimunthe, L. (1993), ‘‘Voluntary disclosure of segment information by Australian diversified companies’’, Accounting and Finance, Vol. 23 No.1, pp. 33-50. McNally, G., Eng L.H. and Hasseldine, C.R. (1982), ‘‘Corporate financial reporting in New Zealand: an analysis of user preferences, corporate characteristics and disclosure practices for discretionary information’’, Accounting and Business Research, Winter, pp. 11-20. Malone, D., Fries, C. and Jones, T. (1993), ‘‘An empirical investigation of the extent of corporate financial disclosure in the oil and gas industry’’, Journal of Accounting, Auditing & Finance, Vol. 8 No. 3, pp. 249-73. Marston, C.L. and Shrives, P.J. (1991), ‘‘The use of disclosure indices in accounting research: a review article’’, British Accounting Review, Vol. 23, pp. 195-210. Matthes, R. (1996), ‘‘Making electricity line companies accountable’’, New Zealand Manufacturer, November, pp. 10-11. Meek, G.K., Roberts, C.B. and Gray, S.J. (1995), ‘‘Factors influencing voluntary annual report disclosures by US, UK and continental European multinational corporations’’, Journal of International Business Studies,Vol. 26 No. 3, pp. 555-72. Ministry of Economic Development (1999), Newsletter, No. 16, December, Wellington. Olsson, R.C. (1980), ‘‘The changing concept of annual reports’’, The Chartered Accountant in Australia, July, pp. 29-38. Parker, L.D. (1982), ‘‘Corporate annual reporting: a mass communication perspective’’, Accounting and Business Research, Autumn, pp. 279-86. Radebaugh, L.H. and Gray, S.J. (1997), International Accounting and Multinational Enterprises, John Wiley, New York, NY. Raffournier, B. (1995), ‘‘The determinants of voluntary financial disclosure by Swiss listed companies’’, The European Accounting Review, Vol. 4 No. 2, pp. 261-80. Robbins, W.A. and Austin, K.R. (1986), ‘‘Disclosure quality in government financial reports: an assessment of the appropriateness of a compound measure’’, Journal of Accounting Research, Vol. 24 No. 2, pp. 413-21. Ryan, J.B. (1990), New Zealand Company Financial Reporting, University of Auckland, Auckland.

The information gap in annual reports 521

AAAJ 15,4

522

Ryan, J.B. (1993), New Zealand Company Financial Reporting, University of Auckland, Auckland. Singhvi, S.S. and Desai, H.B. (1971), ‘‘An empirical analysis of the quality of corporate disclosure’’, The Accounting Review, January, pp. 129-38. Sutcliffe, P. (1985), Financial Reporting in the Public Sector – A Framework for Analysis and Identification of Issues, Australian Accounting Research Foundation (AARF), Melbourne. Tong, T.L., Kidam, Z.A. and Wah, C.P. (1990), ‘‘Information needs of users and voluntary disclosure practices of Malaysian listed companies’’, The Malaysian Accountant, April, pp. 2-7. Tornqvist, U. (1999), ‘‘An empirical study of accountability: delegation of responsibility and external disclosure in some Swedish companies’’, The European Accounting Review, Vol. 8 No. 1, pp. 139-56. Wallace, R.S.O. (1988), ‘‘Corporate financial reporting in Nigeria’’, Accounting and Business Research, Vol. 18 No. 72, pp. 352-62. Wallace, R.S.O. and Naser, K. (1995), ‘‘Firm specific determinants of the comprehensiveness of mandatory disclosure in the corporate annual reports of firms listed on the stock exchange of Hong Kong’’, Journal of Accounting and Public Policy, Vol. 14, Winter, pp. 311-68. Winfield, R. (1978), ‘‘An investigation into private shareholder usage of financial statements in New Zealand’’, Accounting Education, Vol. 18, pp. 93-101. Wiseman, J. (1982), ‘‘An evaluation of environmental disclosures made in corporate annual reports’’, Accounting, Organisations and Society, Vol. 7 No. 1, pp. 53-63. Zarzeski, M.T. (1996), ‘‘Spontaneous harmonization effects of culture and market forces on accounting disclosure practices’’, Accounting Horizons, Vol. 10 No. 1, pp. 18-37. Further reading Buzby, S.L. (1974), ‘‘Selected items of information and their disclosure in annual reports’’, The Accounting Review, Vol. XLIX No. 3, pp. 423-35. Cooke, T.E. (1993), ‘‘Disclosure in Japanese corporate reports’’, Journal of Business Finance & Accounting, Vol. 20 No. 4, pp. 521-35. UnitedNetworks (2000), Submission to the Ministerial Inquiry into the New Zealand Electricity Industry, UnitedNetworks, Auckland.

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

Developments in content analysis: a transitivity index and DICTION scores Robin Sydserff Heriot-Watt University, Edinburgh, UK, and

Pauline Weetman University of Strathclyde, Glasgow, UK

Developments in content analysis

523 Received November 2001 Revised March 2002 Accepted April 2002

Keywords Accounting, Company reports, Foreign languages, Computer-based training, Research Abstract This paper responds to a call in the literature for methodological and empirical studies to advance research into accounting narratives, in the light of acknowledged areas of weakness and gaps in the accounting literature and with a view to investigating impression management. A general line of critique in the accounting literature points to a need to expand both the syntactic and thematic dimensions, with a particular focus on developing objective methods of analysis that allow computer-based measurement. The paper draws on the literature of managerial business communications, supported by that of applied linguistics, in bringing to accounting research a transitivity index and the application of DICTION analysis. Both have the potential to extend computer-based analysis of accounting narratives, subject to careful initial research design and specification. The potential for a richer empirical analysis is demonstrated through an illustrative empirical application.

Introduction This paper is motivated by the need to reflect the increasing importance of accounting narrative as a means of communicating financial information. It responds to calls for methodological and empirical research to advance the literature (Jones and Shoemaker, 1994) and for using research from other fields to find ways to machine-code the precision of natural language and any bias contained in it (Core, 2001). The contribution of this paper is primarily methodological, in extending the range of analytical methods available to researchers investigating accounting narratives. Following Sydserff and Weetman (1999), an interdisciplinary approach is advocated in bringing to the accounting research community methods which have a sound theoretical basis in the specialist literature of The authors thank particularly the Editor of this special issue, Professor John K. Courtis, and two anonymous reviewers for helpful, constructive criticism. The authors are grateful for comments on an earlier version of this paper from participants at the Financial Reporting and Business Communications Conference, Cardiff Business School, 2000. The authors are also grateful to Professor Ken Hyland, English Department, City University of Hong Kong, Dr Karen Corrigan, Department of English Literary and Linguistic Studies, University of Newcastle, Mr Hamish Buchan, Deutsche Bank and Chairman of the Association of Investment Trust Companies’ Working Party on the non-statutory part of the Report and Accounts of an Investment Trust Company, Mrs Anne Sydserff and Mr James Caw for their input to the first author’s doctoral research, on which this paper is based.

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 523-545. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440586

AAAJ 15,4

524

linguistics[1]. Moreover, this paper brings to the accounting research community an expanded awareness of the managerial business communications literature, which has emerged as an interface between the accounting literature concerned with communication and the applied linguistics literature[2]. The paper acknowledges the desire for objective computerised methods while also adding a note of caution on the problems of automating the analytical process. The methods are developed in light of the recognised assessment criteria for methodological development; in particular, validity, objectivity and reliability (Jones and Shoemaker, 1994), and have a particular orientation towards the investigating of impression management, in recognition of the increasing importance in the literature of issues associated with impression management in accounting narratives. The research question addressed in this paper asks: are there text-focused methods in the managerial business communications literature which can be identified as offering potential for the accounting researcher investigating impression management, such that these methods: satisfy the recognised assessment criteria for methodological development, and are capable of being used in empirical applications investigating impression management? The second section reviews prior research to establish the need for extending text-focused methods in accounting and to establish a basis on which applications may be imported from other disciplines. The third and fourth sections describe the transitivity index and DICTION scoring to show that they satisfy tests of suitability for application to accounting narratives and are particularly apposite for investigating impression management. The fifth section applies these methods in an illustrative empirical application and the sixth section concludes. Prior research Studies of narrative reporting that focus on the textual characteristics may be categorised as investigating syntactic and thematic structure. Syntactic analysis focuses on the structural organisation of texts, while thematic analysis is concerned with the information content. This paper makes a contribution to analysing both syntactic and thematic structure. Previous analysis of syntactic structure in the accounting literature is based mainly on readability scores. For example, a number of studies have specifically sought to investigate impression management through readability including Adelberg (1979), Courtis (1986), Jones (1988), Baker and Kare (1992), Kohut and Segars (1992), Smith and Taffler (1992), Subramanian et al. (1993), Courtis (1995, 1998) and Clatworthy and Jones (2001). The central focus of these studies is the relationship between readability and performance, measured primarily by profitability. Courtis (1998) and Clatworthy and Jones (2001) bring a further dimension to the literature, readability variability. Smith and Taffler (2000) categorise thematic analysis into ‘‘form-oriented’’ analysis, which involves routine counting of words, and ‘‘meaning-oriented’’ analysis, which focuses on the underlying themes in the texts under

investigation (p. 627). A typical focus of form-oriented analysis would be Developments in keyword variables. For example, a number of studies focus on ‘‘positive’’/ content analysis ‘‘negative’’ keywords (e.g. Hildebrandt and Snyder, 1981; Abrahamson and Park, 1994; Abrahamson and Amir, 1996). Typically, form-oriented analysis will rely on some form of objective, computerised analysis of texts based on a compendium or taxonomy of keywords. There is, however, a degree of 525 subjectivity involved in constructing such a taxonomy and in identifying words as, for example, positive and negative. A particular focus of ‘‘meaningoriented’’ analysis are those studies investigating patterns of causal reasoning and attribution (e.g. Bettman and Weitz, 1983; Ingram and Frazier, 1983; Staw et al., 1983; Salancik and Meindl, 1984; Clapham and Schwenk, 1991; Aerts, 1994, 2001). A number of studies embrace both form-oriented and meaningoriented approaches to thematic analysis: Smith and Taffler (2000), for example, integrate keyword analysis and thematic analysis in investigating whether a firm’s discretionary narrative disclosures measure its financial risk of bankruptcy. Readability formulas originated in the assessment of children’s writing. Their applicability to technical texts such as annual reports is questionable (Jones and Shoemaker, 1994, pp. 164-5). However, while validity is problematic, there are compelling reasons for the prevalence of readability formulas. They are inexpensive to use, objective and reliable, and can be helpful in detecting certain obvious classes of error such as excessive sentence length (Schriver, 1989, p. 244). Courtis (1998) acknowledges the limitations of readability formulas but argues for their continued use in the absence of alternatives. He proposes variability of readability scores as a possible variation to identify impression management (see also, Clatworthy and Jones, 2001). Alternatives to readability are relatively rare. Kohut and Segars (1992) used a mixture of manual and computerised coding in their investigation of readability. The manual coding was at a relatively simplistic level; for example, word counts, number of sentences. Adelberg (1983) devised an accounting syntactic complexity formula as a new instrument for predicting the readability of selected accounting communications. The method, however, was only outlined in preliminary form, intended for directional use for those engaged in accounting writing, or for those engaged in critiquing accounting writing. Sydserff and Weetman (1999) offered a texture index with a strong basis in applied linguistics but heavily reliant on manual analysis. While offering in-depth understanding and codification of the structure of a narrative, their approach has limited application where large-scale statistical analysis is required. Their primary contribution was to lay the ground for an interdisciplinary approach to developing methods for use in accounting applications. The challenge in finding an operational complement to readability scores is to use methods that are, or could be, computerised (Core, 2001). They must also satisfy conditions of face validity, in measuring what they purport to measure, by demonstrating the objectivity of the coding method and correct

AAAJ 15,4

526

measurement specification (Jones and Shoemaker, 1994, p. 162). Reliability in matters such as inter-coder reliability is also important (Milne and Adler, 1999). However, there will always need to be a balance between the objective benefits of simplicity, automation and a reduction of judgemental input and the subjective value of a refined and sophisticated level of analysis (Smith and Taffler, 2000). Recent developments in the managerial business communications literature offer the potential for the development of alternatives to readability formulas[3]. Thomas (1997) analysed linguistic structures used by companies in a series of president’s letters to shareholders over a five-year period of declining performance. The particular linguistic structures investigated were transitivity (verb type and verb form), thematic structure, context, cohesion and condensations. Overall, the study found that in both ‘‘good news’’ and ‘‘bad news’’ annual reports, the use of particular linguistic structures was associated with management’s intention to maintain its public image and to protect itself from criticism. In particular, ‘‘as the news becomes more negative, linguistic structures suggest a factual, ‘objective’ situation caused by circumstances not attributable to any persons who might otherwise be thought responsible’’ (p. 47). This paper builds on Thomas’ approach in developing a comprehensive transitivity index for use in accounting applications[4]. In addition to the transitivity index, Ober et al.’s (1999) limited accounting application of DICTION analysis offers significant potential for further development. DICTION (Hart, 2000a) is a computerised content analysis program that examines a text for verbal tone across five variables. The accounting application seen in Ober et al. (1999) was limited to only one variable. This paper demonstrates the possibilities of DICTION for accounting applications across the full range of variables. As a computerised approach, DICTION shares a number of the attractions of readability formulas, principally its ease of use, objectivity and reliability. Moreover, a stronger case for DICTION in terms of validity can be made when compared, for example, with readability formulas. With its focus on verbal tone, DICTION can be categorised as a form-oriented approach to thematic content analysis (Smith and Taffler, 2000). Both the transitivity index and DICTION analysis have their theoretical basis in what is referred to as the systemic approach to language study. The focus of the systemic approach (see e.g. Halliday, 1976, 1978, 1985; Lemke, 1989; Fawcett and Halliday, 1978; Butler, 1985) is how linguistic structures are exploited in strategic narrative construction. This focus on strategic narrative construction, what might otherwise be termed persuasive and rhetorical narratology, renders these methods particularly attractive to an accounting research community concerned increasingly with investigating impression management. The term ‘‘impression management’’ refers to the process by which individuals attempt to control the impressions others form of them (Leary and Kowalski, 1990, p. 34). As an extension of the earnings management literature, the term ‘‘impression management’’ has come to be associated in the

accounting literature with the wider documentary context of the annual report Developments in (Hopwood, 1996) and, in particular, the use of disclosure methods such as content analysis graphs (e.g. Steinbart, 1989; Beattie and Jones, 1992a,b, 1993; 1994a,b; Mather et al., 1996; Beattie and Jones, 1996, 1997; Courtis, 1997; Beattie and Jones, 1998, 1999, 2000a,b, 2001) and pictorial illustrations (e.g. Graves et al., 1996; McKinstry, 1996; Preston et al., 1996; Preston and Young, 2000) as well as 527 narrative reporting. The investigation of impression management in accounting narratives embraces, inter alia, readability studies, studies of causal reasoning and attribution, and differentiation studies. Examples of studies of readability, causal reasoning and attribution have been cited earlier in this section. Differentiation studies focus on searching for systematic differences between the narratives of ‘‘good performers’’ and ‘‘poor performers’’ as evidence of impression management (e.g. Frazier et al., 1984; McConnell et al., 1986; Swales, 1988; Tennyson et al., 1990; Kohut and Segars, 1992; Smith and Taffler, 1995, 2000; Smith, 1998). The empirical application reported in this paper uses such ‘‘tests of differentiation’’. The transitivity index The transitivity index is a measure of the number of passive constructions in a text. It is developed from Thomas’ (1997) model for the investigation of linguistic structure through analysis of transitivity, which goes some way towards redressing the lack of emphasis on the syntactic dimension in previous work on accounting narratives. As indicated above, the theoretical basis of this approach can be found in what is referred to as the systemic approach to language study. Consider the following sentences by way of illustration: . The company experienced a downturn in the performance of the portfolio. . A downturn in the performance of the portfolio was experienced by the company. . A downturn in the performance of the portfolio was experienced. The first sentence uses the active voice, while the second and third sentences use the passive voice. In the third sentence the use of the passive voice permits the omission of the ‘‘agent phrase’’ (‘‘by the company’’). There are therefore two stages of depersonalisation that result from the use of the passive voice. The first is the movement of the ‘‘agent phrase’’ from the start to the end of the sentence, while the second stage is the omission of the agent phrase. Whichever strategy is employed, the agent behind the action is downplayed. In this regard, Thomas (1997) notes that active voice and active verbs promote the idea of a company that is moving forward, that is progressive, aggressive and successful in the marketplace. Use of the passive voice is reserved for those occasions when the writer finds it advantageous to be distanced from the message (Thomas, 1997, pp. 52-3). Thomas found a

AAAJ 15,4

528

predictable increase in passive constructions as the years passed and profits decreased (p. 53). As the news becomes more negative, linguistic structures suggest a factual, objective situation caused by circumstances not attributable to any persons who might otherwise be thought responsible (p. 47). Passive constructions give a text a veneer of objectivity, neutrality, scientific ‘‘truth’’ or ‘‘fact’’ (Carter et al., 1997, p. 224). A useful link can be made with those studies which investigate patterns of causal reasoning and attribution (e.g. Bettman and Weitz, 1983; Ingram and Frazier, 1983; Staw et al., 1983; Salancik and Meindl, 1984; Clapham and Schwenk, 1991; Aerts, 1994, 2001). Broadly, these studies are concerned with the patterns of causal reasoning used to explain corporate performance. Researchers hypothesise the following pattern in explanations given for corporate performance (Bettman and Weitz (1983) are typical); reasons internal to the organisation will be cited for favourable performance outcomes and external factors will be noted for unfavourable outcomes. The existence of such self-serving attributions may indicate a distortion of causal reasoning about corporate performance (p. 167). The discussion above indicates that passive voice may be used in a similar way as a particular linguistic method of disassociation with bad news disclosures. The transitivity index used in this paper embraces a manual approach to analysis. We were initially attracted to using the number of passive sentences reported in readability statistics, such as those found in commonly-used word processing packages. We shared the view of Core (2001) that there was potential for objectivity and reliability associated with such a computerised content analysis, together with its ease of use. However, on carrying out a comparative manual analysis we found that the computerised method of a widely-used package fails to code sentences sufficiently accurately to be relied on as a research tool for precise quantification[6]. The primary purpose of the word processing package is to provide a yardstick for assessment and direct the writer to areas that require clarification or improvement; it is not designed as a precision tool for analytical research. Our manual analysis is based on the ‘‘t-unit’’ (text-unit), defined as one independent clause with all subordinate clauses attached to it. Generally, this equates to a sentence as written, but there are instances where two units of narrative which satisfy this definition are linked by a conjunction. The evaluation of narrative in t-units is a common practice in the linguistics literature (see Allard and Ulatowska, 1991; Cameron et al., 1995; Witte and Faigley, 1981). In this paper a t-unit is categorised as passive if it contains a passive construction. It is acknowledged that categorising on this basis necessitates some loss of detail. For example, a t-unit may contain more than one passive construction or a mixture of active and passive constructions. The potential loss of detail is, however, mitigated by having the t-unit as the basic unit of analysis, rather than the sentence as written.

Passive constructions can be categorised as passive verbs and passive Developments in verbals. The first category, the passive verb, is by far the most common. content analysis Passive verbs are created by combining a form of the verb ‘‘to be’’ with the past participle of the main verb. This structure finds expression in various tenses. Table I illustrates. As indicated in Table I, a t-unit in the passive voice will not always include 529 an agent of the action. Further, only transitive verbs (verbs which take objects) can be transformed into passive constructions. Finally, active sentences containing certain verbs cannot be transformed into passive structures. The most important of these is the verb ‘‘to have’’. The second category of passive construction is passive verbals (Table II). Verbals are words or phrases that seem to carry the idea of action or being but do not function as a true verb. They are sometimes called ‘‘non finite’’ (unfinished or incomplete) verbs. There are three types of verbals which take on features of the passive voice: infinitive phrases in the passive voice, passive gerunds and passive participles. Based on these categories and definitions, the rules for classification are set out in Table III. While the use of passive constructions gives the text a veneer of objectivity or neutrality, and can be used by writers as a linguistic mechanism to disassociate themselves from the text, some caution must be observed in assuming that the use of passive constructions is a deliberate strategy of impression management employed by a writer reporting ‘‘bad news’’. For example, as a genre, scientific writing is characterised by the use of passive voice (Swales, 1990). It may be that the writer of the text is influenced by that particular mode of writing through educational or vocational experience. Also, a text of any significant length is likely to employ both active and passive constructions as an expected or normal mode of writing (Carter et al., 1997, pp. 223-5). If the analysis is applied with careful regard to grammatical rules for passive constructions, face validity is satisfied because the transitivity index will measure what it purports to measure. The link between transitivity and the attribution literature in accounting strengthens face validity. Although the coding is manual, it is a sentence-based coding instrument with detailed decision rules which is likely to lead to a high degree of inter-coder reliability (Milne and Adler, 1999). DICTION scores DICTION analysis (Hart, 2000a) is a computerised content analysis programme that examines a text for verbal tone. Verbal tone is measured in terms of five master variables: ‘‘certainty’’, ‘‘optimism’’, ‘‘activity’’, ‘‘realism’’ and ‘‘commonality’’. The master variables are constructed on the basis of 31 individual dictionary scores and four calculated variables (Hart, 2000a). The dictionaries were constructed from the analysis of more than 20,000 texts, which yield a total word corpus, taking the dictionary scores together, in excess

The portfolio(s)

The portfolio(s) The portfolio(s) The portfolio(s)

The portfolio(s) The portfolio(s)

The portfolio(s) The portfolio(s)

Present tense

Past tenses Perfect Imperfect Pluperfect

Future tenses Future Future perfect

Conditional tenses Conditional Conditional perfect

Table I. Passive verbs

Subject

will be will have been

have been were/were being had been

are/are being

Auxiliary (verb ‘‘to be’’) Plural

managed managed

managed managed managed

managed

Past participle

would be/should be would be/should be managed would have been/should have been would have been/should have been managed

will be will have been

has been was/was being had been

is/is being

Singular

530

Tense

by [name] by [name]

by [name] by [name]

by [name] by [name] by [name]

by [name]

Agent phrase (can be omitted)

AAAJ 15,4

1. Infinitive phrases in the passive voice An infinitive phrase in the passive voice is formed by combining the infinitive form of the verb ‘‘to be’’ (either in the present (‘‘to be’’) or the past (‘‘to have been’’)) with a past participle. Infinitive phrases in the passive voice can occur in different positions in the sentence, e.g. Present To be affected by adverse market conditions is a risk that must be accepted. The trust continues to be affected by adverse market conditions. Past To have been affected by adverse market conditions was a risk that should have been anticipated. The trust’s performance appeared to have been affected by adverse market conditions.

Developments in content analysis

531

2. Passive gerunds A gerund is a noun formed from a verb. Passive gerunds take two forms: ‘‘being’’ followed by a past participle (the present) and ‘‘having been’’ followed by a past participle (the past). Passive gerunds can occur in different positions in the sentence, e.g. Present Being affected by the adverse market conditions, the trust is changing its investment policy. The trust, being affected by the adverse market conditions, is changing its investment policy. Past Having been affected by adverse market conditions, the trust experienced a downturn in performance. The trust, having been affected by adverse market conditions, experienced a downturn in performance. Note: the passive gerund always refers to the subject of the main clause. In the examples above, the subject is the trust which is being or having been affected 3. Passive participles The passive gerund forms described above are often used without the auxiliaries (‘‘being’’ and ‘‘having been’’), leaving only the past participle. These are referred to as passive participles. Passive participles can occur in different positions in the sentence, e.g. Present: [Being] affected by the adverse market conditions, the trust is changing its investment policy. The trust, [being] affected by the adverse market conditions, is changing its investment policy. Past [Having been] affected by adverse market conditions, the trust experienced a downturn in performance. The trust, [having been] affected by adverse market conditions, experienced a downturn in performance.

of 10,000. None of the search items is duplicated in the lists, allowing the researcher to gain a rich understanding of the particular text being analysed. Certainty is defined as ‘‘language indicating resoluteness, inflexibility, completeness and a tendency to speak ex cathedra’’ (Hart, 2000a, p. 32). Optimism is defined as ‘‘language endorsing some person, group, concept or event or highlighting their positive entailments’’ (Hart, 2000a, p. 34). Activity is defined as ‘‘language featuring movement, change, the implementation of ideas and the avoidance of inertia’’ (Hart, 2000a, p. 34). Realism is defined as ‘‘language describing tangible, immediate, recognizable matters that affect

Table II. Passive verbals

AAAJ 15,4

Rules for classification are established by the following question: Does the ‘‘t-unit’’ contain: A B

a passive verb a passive verbal

If the answer to either or both of these is ‘‘yes’’, then the ‘‘t-unit’’ is categorised as passive.

532 Table III. Transitivity: rules for classification

Note: In a number of instances a passive verb will be immediately followed by a passive verbal in a relationship of mutual dependence. Typically, this occurs with a passive verb and an infinitive phrase in the passive voice (passive verbal). In such instances, the passive verb and the passive verbal should be treated as one passive construction

people’s everyday lives’’ (Hart, 2000a, p. 35). Commonality is defined as ‘‘language highlighting the agreed-upon values of a group and rejects idiosyncratic modes of engagement’’ (Hart, 2000a, p. 37). Like the transitivity index, DICTION falls within the scope of systemic linguistics, rendering the approach attractive to accounting researchers investigating impression management. Moving beyond the more general systemic approach, the specific dimension of theoretical linguistics which provides the basis for DICTION is linguistic semantics (see Frawley, 1992; Lyons, 1995; Saeed, 1997). As a method of analysing semantic content, DICTION is well established in the applied linguistics literature. Empirical applications include Hart (1997b, 2000b), Hart and Gourgey (1998) and Hart and Jarvis (1997). Finally, the validity of DICTION as a computerised content analysis programme has been attested by independent research (e.g. Frey et al., 1991; Morris, 1994; West, 2001). An accounting application of DICTION analysis is seen in Ober et al. (1999), but limited to the ‘‘certainty’’ variable. They found no significant difference in the use of certainty in the narratives of ‘‘good performers’’ when compared to ‘‘poor performers’’. This paper advocates the exploitation of the full range of analysis offered by DICTION and also suggests some refinements with regard to the analysis of ‘‘certainty’’ in Ober et al. (1999). Subsequent to their work, based on DICTION 4.0 (Hart, 1997a), which was the first version of DICTION to be made commercially available, a new version, DICTION 5.0 (Hart, 2000a) incorporates a number of refinements, in particular the provision of comparative data. It is described here and used in the illustrative empirical application. In terms of the accounting literature, DICTION analysis probably finds its closest parallel in the form-oriented approach to thematic content analysis employed by Smith and Taffler (2000). They constructed a content analysis dictionary comprising 168 different keywords using the Oxford Concordance Program. Keywords were then categorised on the basis of a four-factor cognitive structure, based on Houghton’s (1988) work on the measurement of connotative meaning in accounting. Smith and Taffler (2000) found that the form-based approach, which embraces the benefits of simplicity, automation

and a reduction of judgemental input, yielded results that were similar to a Developments in more sophisticated thematic based subjective methodology. content analysis As a form-oriented approach, DICTION offers considerable potential for the accounting researcher. It is simple to use, it is automated, and yet it embraces a considerable degree of sophistication. The dictionaries have been constructed by experts in linguistics. With a total word corpus in excess of 10,000, 533 DICTION is considerably more comprehensive than existing form-oriented approaches in the accounting literature. The basic unit of analysis in DICTION is a 500-word norm. The programme default is to generate one set of scores for the entire text (regardless of length) that allows comparison with texts of different lengths and with normative values, which are reported on the basis of a 500-word norm. Where the 500word standard is the basic default for DICTION, a number of specific options are available to the researcher, depending on whether the text is less than 500 words or greater than 500 words. Where a particular text is less than 500 words, DICTION offers two options. The default option is to make corrective counts, thereby standardising it to a 500-word basis. The second option reports raw scores. When a text is more than 500 words, the researcher has three options. The default options is to generate a 500-word equivalent score. This is done by averaging its 500-word units together. For that part of the text which does not correspond to a 500-word unit, an extrapolation to a 500-word ‘‘equivalent’’ allows that section then to be included in the averaging process. DICTION also allows the researcher to analyse the first 500 words only of a text. Finally, the text can be segmented into 500-word units and each processed separately. This allow the researcher to investigate variability, for example, by comparing the first 500 words of a given text with the middle 500 words and the final 500 words. A distinctive feature of DICTION 5.0 is the availability of a number of normative values for comparative purposes. One of these is ‘‘corporate financial reports’’, which consists of a sample of 48 annual financial reports from Fortune 500 companies. The texts analysed comprise the president’s letter and sections of the management discussion and analysis (MD&A). Use of such comparatives would require some caution, because the sample relates to US companies and to relatively large companies. The empirical section of this paper does not rely on the DICTION comparatives but they could be of interest in other types of accounting analysis, particularly where benchmarking against US practice is a feature. DICTION also allows the researcher to customise the word-lists that give rise to the existing variables and also to construct new word-lists and, therefore, new variables. Apart from the ‘‘certainty’’ score, all of the master variables are reported in the illustrative application in this paper without adjustment. Following Ober et al. (1999), an adjusted ‘‘certainty’’ formula is used, where the two subtractive variables ‘‘numerical terms’’ and ‘‘self-reference’’ were deleted from the original formula. ‘‘Numerical terms’’ was deleted on the basis that ‘‘such specificity in

AAAJ 15,4

534

business communications is commonly considered to contribute to, rather than detract from certainty’’ (Ober et al., 1999, p. 288). ‘‘Self-reference’’ was deleted on the basis that such a style of writing ‘‘promoted acceptability of responsibility for one’s writing’’ (p. 288). As a feature of DICTION, the potential for adaptability of the general model to specific situations of application is seen as a particular strength. This is similar to the general-specific character of the texture index (Sydserff and Weetman, 1999). In relation to validity and reliability, the objectivity of DICTION analysis is a particular strength. Its automated nature, both for coding and quantification, renders it attractive as a research instrument. The overview presented in this section is indicative of strength in face validity. In particular, the specific theoretical basis of the approach in linguistic semantics, the fact that the approach is well established in the applied linguistics literature and the independent attestation of the approach all point to strength in face validity. Data The narratives analysed in this empirical application are the chairman’s statement and manager’s report of a particular sub-category of the investment trust industry sector – ‘‘Smaller companies: UK’’. Investment trust companies are particularly apposite as a data source for a methodologically focused application because they give intra-industry comparability, comparability of accounting narratives and the availability of externally determined comparative performance statistics. The industry regulator, the Association of Investment Trust Companies (AITC), defines sub-sectors within which the degree of intra-industry comparability is enhanced. The particular sub-category investigated here is ‘‘Smaller companies: UK’’, whose investment objective is that at least 80 percent by value of the investment trust company’s portfolio will be invested in the shares of UK registered companies and 50 percent by value of the portfolio invested in the shares of smaller and medium sized companies (AITC, Monthly Information Service). This is the largest single sub-category, incorporating 27 trusts. The common objective gives a focus to the narrative of chairman’s statement and management report that avoids the complexities which may frustrate analysis of larger narratives such as the management discussion and analysis of US companies or the operating and financial review of UK companies. The comparative performance statistics used in this study are those produced by the AITC. For each month-end, the AITC’s monthly information service includes detailed comparative performance statistics for each subcategory. For each of the 27 trusts in the sector, a written request was made in April 2000 for the latest annual report. After a number of follow-up enquiries, 26 trusts responded with a copy of their annual report. The period of coverage in terms of the accounting year-ends of the annual reports received was from 5 April 1999 to 31 March 2000.

The 26 investment trusts were split into ‘‘good performers’’ and ‘‘poor Developments in performers’’ on the basis of net asset value (NAV) total return. This is the content analysis industry norm for performance comparison of investment trusts, used both by the trusts themselves, in monitoring their performance against a designated benchmark index, and by external sources, which provide comparative performance statistics for particular sectors. Two performance ranks are 535 constructed. ‘‘Rank 1’’ is based on NAV total return over one year. ‘‘Rank 2’’ is based on an amalgam of NAV total return over one year, three years and five years. For each performance rank the top ten and bottom ten trusts were categorised as good performers and poor performers respectively. The shortterm/longer-term performance distinction yields different sets of good performers and poor performers, depending on which performance measure is used (Sydserff, 2001). The use of two performance ranks covering a short-term and a longer-term perspective was based on the advice of a leading industry expert who is a member of the AITC. Illustrative empirical application This illustrative empirical application is based on testing one of the hypotheses typical of the impression management literature. We have chosen the comparison of good and bad performers so that we can focus on indicating that the new metrics proposed do differentiate as relative measures, in a different way from the established readability scores, without requiring to consider any interpretation of absolute values. We leave the interpretation of absolute values to subsequent research. Our illustrative hypothesis is as follows: Ho. There is no significant difference in [Flesch readability score] or [transitivity index] or [DICTION scores] for the chairman’s statement/ manager’s report of ‘‘good performers’’ and ‘‘poor performers’’. HA. There is a significant difference in [Flesch readability score] or [transitivity index] or [DICTION scores] for the chairman’s statement/ manager’s report of ‘‘good performers’’ and ‘‘poor performers’’. We have used the chairman’s statement as a document found typically in most annual reports, and the manager’s report as a more technical report on performance which, for this industry, has an element of the wider operating and financial review or the management discussion and analysis. For all of the trusts included in the sets of good performers and poor performers, both the chairman’s statement and the manager’s report were scored for transitivity and DICTION. Along with the transitivity scores based on manual analysis, computerised transitivity scores and Flesh readability scores based on the readability statistics in Microsoft Word (version 7.0) are reported. Apart from the adjusted ‘‘certainty’’ variable, all of the other DICTION master variables are extracted directly from reports generated by the DICTION software. ‘‘Certainty’’ scores were computed using the component variables scores generated by DICTION. The default 500-word equivalent standardising text

AAAJ 15,4

536

option is used to control for variation in input text length. The specific content analysis procedures for the manual analysis of transitivity were as follows. Two independent coders with expertise in linguistic analysis scored the texts for transitivity using the rules for application detailed in Tables I to III. The texts were cross-checked and all discrepancies resolved. Finally, the coding was verified by the author who designed the coding method. The median scores for all measures are summarised in Table IV. A MannWhitney test for differences of medians was applied to all scores. The higher readability score in the manager’s report of good performers, when compared to poor performers based on the longer-term performance measures, supports the obfuscation hypothesis, namely that management is not neutral in its presentation of narrative information and will seek to obfuscate failures or bad news disclosures (Courtis, 1998, p. 466). However, no significant differences were found in the chairman’s statement or in the manager’s report where performance is distinguished by a short-term measure. The lower transitivity scores for good performers support an expectation that the narratives of poor performers are characterised by a more objective, detached style, indicative of management desire to distance itself from the message communicated. On the basis of these results, it appears that verbal voice is a particular linguistic dimension that is exploited by management as an impression management strategy in accounting narratives. While there is similarity in the significance levels of the differences in results from the computerised analysis and the manual analysis, one must be cautious of concluding that the computerised model is a reliable proxy for the manual approach. For example, it is clear from the median scores reported that the manual approach is more comprehensive in its capture of passive constructions. When the narratives were manually coded to determine the transitivity scores, it was noted that a number of passive constructions were simply omitted from the computerised analysis. Moreover, these omissions were not systematic. At this stage we can only suggest that the computerised version may function as a useful yardstick or benchmark for the relative passive character of a text, but reliance should not be placed in absolute values. No significant differences were found between good performers and poor performers for the certainty score. Similarly, Ober et al. (1999) found no significant difference between good performers and poor performers for the use of ‘‘certainty’’ in Management’s discussion and analysis in US corporate reports. They suggested (p. 292) that management will ‘‘tell it like it is, no matter whether profits have increased or decreased’’. Equally, however, it may be that a marked tone for certainty is in itself a self-serving strategy, adopted by poor performers to imitate good performers. Results on ‘‘optimism’’ are mixed. It might be expected that poor performers will mirror good performers in their expression of ‘‘optimism’’. The absence of differentiation for the chairman’s statement (using the longer-term performance measure) and for the manager’s report (both performance measures) suggest that such a strategy is being followed. These findings are

Good performers

Poor performers

Significance

Short-term performance ranking Flesch readability score [CS] Flesch readability score [MR]

32.90 34.15

32.85 35.80

ns ns

Longer-term performance ranking Flesch readability score [CS] Flesch readability score [MR]

36.70 38.50

31.60 35.00*

ns p = 0.0932

27.65 23.25

37.35** 33.30**

p = 0.0263 p = 0.0168

17.0 14.5

24.00* 25.0**

p = 0.0607 p = 0.0063

27.6 22.4

33.90** 28.60

p = 0.0454 ns

15.0 13.0

22.00* 19.00

p = 0.0604 ns

Short-term performance ranking Score for ‘‘certainty’’ [CS] Score for ‘‘certainty’’ [MR]

49.59 46.84

48.09 47.09

ns ns

Longer-term performance ranking Score for ‘‘certainty’’ [CS] Score for ‘‘certainty’’ [MR]

49.82 47.42

47.30 47.09

ns ns

Short-term performance ranking Score for ‘‘optimism’’ [CS] Score for ‘‘optimism’’ [MR]

54.08 51.75

50.54* 52.09

p = 0.0832 ns

Longer-term performance ranking Score for ‘‘optimism’’ [CS] Score for ‘‘optimism’’ [MR]

53.09 51.23

51.45 51.90

ns ns

Short-term performance ranking Score for ‘‘activity’’ [CS] Score for ‘‘activity’’ [MR]

48.35 48.32

49.16 50.74*

ns p = 0.0710

Longer-term performance ranking Score for ‘‘activity’’ [CS] Score for ‘‘activity’’ [MR]

47.71 48.21

48.33 49.10*

ns p = 0.0636

Short-term performance ranking Score for ‘‘realism’’ [CS] Score for ‘‘realism’’ [MR]

46.38 46.03

46.42 47.66

ns ns

Longer-term performance ranking Score for ‘‘realism’’ [CS] Score for ‘‘realism’’ [MR]

46.59 46.69

46.24 47.77

Short-term performance ranking Manual analysis Transitivity score [CS] Transitivity score [MR] Computerised analysis Transitivity score [CS] Transitivity score [MR] Longer-term performance ranking Manual analysis Transitivity score [CS] Transitivity score [MR] Computerised analysis Transitivity score [CS] Transitivity score [MR]

ns ns (continued)

Developments in content analysis

537

Table IV. Summary of median scores for good and poor performers in a set of investment trust companies

AAAJ 15,4

538

Table IV.

Good performers

Poor performers

Significance

Short-term performance ranking Score for ‘‘commonality’’ [CS] Score for ‘‘commonality’’ [MR]

51.39 46.85

51.72 49.84**

ns p = 0.0359

Longer-term performance ranking Score for ‘‘commonality’’ [CS] Score for ‘‘commonality’’ [MR]

51.60 47.46

52.82 49.76

ns ns

Note: Mann-Whitney test for medians: ** ¼ significant at less than 5 per cent; * ¼ significant at less than 10 percent. CS ¼ Chairman’s statement; MR ¼ Manager’s report

qualified, however, by evidence of a more emphatic tone for ‘‘optimism’’ in the chairman’s statement of ‘‘good performers’’ where the short-term performance measure is applied. A more emphatic tone for ‘‘activity’’ was found in the manager’s report of poor performers on both performance measures. No significant differences were found for the chairman’s statement. It could be said that the semantic feature ‘‘activity’’ is indicative of a company that is forward-looking, progressive and self-determining, controlling its own success. On this basis, it might be surmised that the creation of an emphatic tone for ‘‘activity’’ would be an impression management strategy adopted by poor performers in their narratives to mirror good performers. No differentiation is observed in ‘‘realism’’. A particular strategy of management in reporting poor performance might be obfuscation through weakening the semantic content for ‘‘realism’’. The empirical evidence in this study, however, suggests that such a strategy is not pursued. The results for ‘‘commonality’’ scores provide some interesting insights as to aspects of management intention in constructing accounting narratives. It could be argued that, in the absence of impression management strategies, good performers would exhibit a less marked tone for ‘‘commonality’’ than poor performers, consistent with their desire to set themselves apart from the group, emphasising their diversity and exceptional performance. This inference is borne out to some extent by the lower ‘‘commonality’’ score in the manager’s report of good performers (using a short-term performance measure). This evidence is countered by the absence of differentiation for the chairman’s statement (both performance measures) and for the manager’s report (using a longer-term performance measure). This may be evidence of an impression management strategy on the part of poor performers, to emphasise group identity so as to avoid isolation as a poor performer. Again, this is an aspect of the strategy associated with the variables ‘‘certainty’’, ‘‘optimism’’ and ‘‘activity’’, whereby the narratives of poor performers will mirror the semantic tone that characterises the narratives of good performers.

Overall, the results from tests of differentiation between good performers Developments in and poor performers are mixed. While providing some evidence in support of content analysis impression management, in particular with regard to transitivity, ‘‘optimism’’ and ‘‘activity’’, the absence of differentiation across a number of dimensions investigated is suggestive of the view that management is even-handed in its presentation of narrative information. While equating the absence of 539 differentiation with the absence of strategic impression management is appropriate for a number of textual dimensions, in some instances the absence of differentiation may be indicative of impression management, particularly in relation to ‘‘certainty’’, ‘‘optimism’’ and ‘‘activity’’. It could be argued that the managers of the poor performers will use impression management to make their narrative resemble as closely as possible the verbal tone of the good performers. Conclusion This paper has shown that the transitivity index and the use of DICTION scores, both previously used in the managerial business communications literature, offer potentially useful alternatives for the accounting researcher investigating impression management. The transitivity index balances, but does not supplant, the use of readability scores and allows readability to become part of a portfolio of syntactic text measures rather than an isolated indicator. The DICTION score adds to a portfolio of methods within the general category of form-oriented thematic content analysis. Both satisfy the essential requirement of face validity and are capable of being relatively reliable in terms of consistency between coders. The illustrative application has shown that both techniques are capable of being used in an empirical application in accounting that can contribute to the understanding of impression management within one business sector. The application to investment trust narratives gives a controlled comparison of narratives related to good performance and poor performance and shows that a more informed understanding begins to emerge of how impression management may cause some aspect of texts to differ while others are aligned in similar wording, possibly to obfuscate any difference in performance. Limitations lie in this being a first presentation of data with no prior evidence for comparison. Methodological limitations lie in the difficulty of finding any single text measurement technique that could address all the significant weaknesses and gaps identified in prior literature. We have shown that the development of a transitivity index that requires comprehensive coverage of passive constructions may need more sophisticated computerbased analysis than is found in conventional word processing packages. A degree of caution must be observed when applying the DICTION methodology (without adaptation to word-lists) to accounting narratives beyond the US context. The skills of the researcher in designing the research instrument are not yet replaceable by off-the shelf computer software, but the possibilities for codification and computer-based analysis are increasing.

AAAJ 15,4

540

Despite this note of caution regarding rushing to the computer software, we conclude that the analysis and empirical illustration offered in this paper is sufficient to provide first indication that the transitivity index and DICTION analysis do merit further exploration in accounting studies. Notes 1. This interdisciplinary focus recognises a shift in the discipline of linguistics from a formalist to a functionalist paradigm with the emergence of the field of applied linguistics or discourse analysis. Researchers in this area of linguistics contend that narrative discourse analysis has an important role to play in the wider communities of social scientific research, hence the increasing development of relevant interdisciplinary models of language, frameworks which turn the insights of linguists into comprehensive and usable forms (Fairclough, 1995, p. 210). The texture index developed for accounting applications by Sydserff and Weetman (1999), together with the transitivity index and DICTION analysis developed in this paper, are examples of such usable analytical models. For a fuller discussion of the nature of such an interdisciplinary approach see Sydserff and Weetman (1999). The particular paradigm pursued in this paper is to develop approaches which give rise to quantitative measurement, approaches which can be regarded as operational complements to readability formulas. There is also potential, although beyond the scope of this paper, to exploit these quantitative approaches alongside more qualitative approaches in applied linguistic analysis, such as critical discourse analysis. Gallhofer et al.’s (2001) detailed exegesis and application of critical discourse analysis in an accounting context indicates the potential for such an approach to function as an all-embracing context for a holistic evaluation. 2. The methods developed in this paper have as their basis studies published in The Journal of Business Communication. Although using the methods in limited accounting applications, the studies are written expressly from an applied linguistics/managerial perspective, rather than an accounting perspective and require considerable development to meet the specific requirements of accounting researchers. The texture index (Sydserff and Weetman, 1999) was drawn directly from the applied linguistics literature and had not previously been used in an accounting application. 3. It is not claimed that the methods developed in this paper are the only methods in the managerial business communications literature offering potential to accounting researchers. One such approach would be metadiscourse analysis (Hyland, 1998). This study, building on an extensive body of prior research, compared the chairman’s statement and directors’ report of 137 high- and medium-performing Hong Kong companies in order to show how companies use the chairman’s letter to influence readers and project a positive corporate image. The particular dimension of linguistic structure investigated was metadiscourse, a term that Hyland takes to refer to those aspects of text structure which are employed in strategic narrative construction. Metadiscourse is a critical element of persuasive discourse as it seeks to influence how readers understand propositional information (pp. 225-6). There are two dimensions to metadiscourse: textual metadiscourse and interpersonal metadiscourse. The former is concerned with the construction of the text through the use of linking devices, essential to the readability of the text (p. 228), while the latter is concerned with the tenor of the discourse. On the basis of a higher level of metadiscourse observed for the chairman’s statement when compared with the directors’ report, the study concluded that the chairman’s statement is used strategically in order to direct readers as to how they should understand and appraise the subject matter (p. 224). Hyland’s approach is not developed further in this paper for a number of reasons. First, those dimensions embraced by textual metadiscourse are developed more fully in the texture index (Sydserff and Weetman, 1999). Second, there is a considerable degree of overlap between the categories of interpersonal metadiscourse and the semantic variables

investigated using DICTION analysis. Third, the principal orientation of this paper is to develop and apply new text-focused methods of evaluation. Metadiscourse analysis is already at an advanced stage of development and has been used in an in-depth empirical accounting application. The methods developed in this paper are proposed as being of particular relevance, both in terms of their potential for adaptability to meet the requirements of accounting researchers, and in terms of their complementarity with existing methods. 4. Exploiting a particular dimension of Thomas’ (1997) approach is not problematic. The approach does not claim inclusivity in terms of the dimensions captured and advocates no form of cross-summation across the different dimensions. The linguistic dimensions ‘‘context’’, ‘‘cohesion’’ and ‘‘condensations’’ are embraced more fully in the texture index (Sydserff and Weetman, 1999). The potential for developing methods in relation to verb choice (the first dimension of transitivity) and thematic structure is considered a matter for further research. 5. This categorisation reflects what are only broad groupings and themes in what can be regarded as an emerging literature. In this regard, it is difficult at this stage to indicate precisely what is embraced by the term ‘‘impression management’’ and to synthesise an impression management theory that gives rise to clear predictions regarding empirical observations. A contrast can be made with the earnings management literature where a more robust theory has been developed. The problematic of synthesising a theory of impression management can be illustrated with regard to the investigation of differentiation between the narratives of ‘‘good performers’’ and ‘‘poor performers’’. While it might be predicted that higher levels of readability and a greater prevalence of external attributions in the narratives of ‘‘poor performers’’ (i.e. the presence of differentiation) are realisations of managerial intention to manage impressions, equally, it might be surmised that the absence of differentiation in relation to aspects of thematic content are realisations of impression management strategies whereby ‘‘poor performers’’ provide signals which imitate ‘‘good performers’’. In other words, different variables may function differently vis-a`-vis differentiation as evidence of impression management. This complexity is mirrored in the methods developed in this study, where both the presence and absence of differentiation are associated with impression management. Looking to the future, it may be that in developing a robust theory of impression management, such inherent complexity must be accommodated, where predictions and testable hypotheses will be developed, for example, in relation to the particular narrative dimension that is the focus of attention. 6. See Sydserff (2001) for further discussion. 7. See www.scolari.co.uk or www.scolari.com for more information. References Abrahamson, E. and Amir, E. (1996), ‘‘The information content of the president’s letter to shareholders’’, Journal of Business Finance and Accounting, Vol. 23 No. 8, pp. 1157-82. Abrahamson, E. and Park, C. (1994), ‘‘Concealment of organizational outcomes: an agency theory perspective’’, Academy of Management Journal, Vol. 37 No. 5, pp. 1302-34. Adelberg, A.H. (1979), ‘‘Narrative disclosures contained in financial reports: means of communication or manipulation’’, Accounting and Business Research, Vol. 9 No. 35, pp. 179-89. Adelberg, A.H. (1983), ‘‘The accounting syntactic complexity formula: a new instrument for predicting the readability of selected accounting communications’’, Accounting and Business Research, Vol. 13 No. 51, pp. 163-75.

Developments in content analysis

541

AAAJ 15,4

542

Aerts, W. (1994), ‘‘On the use of accounting logic as an explanatory category in narrative accounting disclosures’’, Accounting, Organizations and Society, Vol. 19 Nos. 4/5, pp. 337-53. Aerts, W. (2001), ‘‘Inertia in the attributional content of annual accounting narratives’’, European Accounting Review, Vol. 10 No. 1, pp. 3-32. Allard, L. and Ulatowska, H.K. (1991), ‘‘Cohesion in written narrative and procedural discourse of fifth-grade children’’, Linguistics and Education, Vol. 3, pp. 63-79. Baker, H.E. and Kare, D.D. (1992), ‘‘Relationship between annual report readability and corporate financial performance’’, Management Research, Vol. 15, pp. 1-4. Beattie, V.A. and Jones, M.J. (1992a), The Communication of Information Using Graphs in Corporate Annual Reports, ACCA Research Report, No. 29. Beattie, V.A. and Jones, M.J. (1992b), ‘‘The use and abuse of graphs in annual reports: a theoretical framework and empirical study’’, Accounting and Business Research, Vol. 22 No. 88, Autumn, pp. 291-303. Beattie, V.A. and Jones, M.J. (1993), ‘‘Effect of graphical presentations on insights into a company’s financial position – an innovative educational approach to communicating financial information in financial reporting: a comment’’, Accounting Education, Vol. 2 No. 4, pp. 303-9. Beattie, V.A. and Jones, M.J. (1994a), ‘‘An empirical study of graphical format choices in charity annual reports’’, Financial Accountability and Management, Vol. 10 No. 3, August, pp. 215-36. Beattie, V.A. and Jones, M.J. (1994b), ‘‘Information design and manipulation: the case of financial graphs in corporate annual report’’, Information Design Journal, Vol. 7 No. 3, pp. 211-6. Beattie, V.A. and Jones, M.J. (1996), Financial Graphs in Corporate Annual Reports: A Review of Practice in Six Countries, Institute of Chartered Accountants in England and Wales, London. Beattie, V.A. and Jones, M.J. (1997), ‘‘A comparative study of the use of financial graphs in the corporate annual reports of major US and UK companies’’, Journal of International Financial Management and Accounting, Vol. 8 No. 1, pp. 33-68. Beattie, V.A. and Jones, M.J. (1998), Graphical Reporting Choices: Communication or Manipulation?, Certified Research Report 56, Certified Accountant Publications, London. Beattie, V.A. and Jones, M.J. (1999), ‘‘Australian financial graphs: an empirical study’’, Abacus, Vol. 35 No. 1, pp. 46-76. Beattie, V.A. and Jones, M.J. (2000a), ‘‘Changing graph use in corporate annual reports: a time series analysis’’, Contemporary Accounting Research, Vol. 17 No. 2, pp. 213-26. Beattie, V.A. and Jones, M.J. (2000b), ‘‘Impression management: the case of inter-country financial graphs’’, Journal of International Accounting, Auditing and Taxation, Vol. 9 No. 2, pp. 159-83. Beattie, V.A. and Jones, M.J. (2001), ‘‘International financial graphs: a six country comparison’’, The International Journal of Accounting, Vol. 36 No. 2, pp. 195-222. Bettman, J.R. and Weitz, B.A. (1983), ‘‘Attributions in the board room: causal reasoning in corporate annual reports’’, Administrative Science Quarterly, Vol. 28, pp. 165-83. Butler, C. (1985), Systematic Linguistics: Theory and Applications, Batsford Academic and Educational, London. Cameron, C.A., Lee, K., Webster, S., Munro, K., Hunt, A.K. and Linton, M.J. (1995), ‘‘Text cohesion in children’s narrative writing’’, Applied Psycholinguistics, Vol. 16, pp. 257-69. Carter, R., Goddard, A., Reah, D., Sanger, K. and Bowring, M. (1997), Working with Texts: A Core Book for Language Analysis, Routledge, London.

Clapham, S.E. and Schwenk, C.R. (1991), ‘‘Self-serving attributions, managerial cognition, and company performance’’, Strategic Management Journal, Vol. 12, pp. 219-29. Clatworthy, M. and Jones, M.J. (2001), ‘‘The effect of thematic structure on the variability of annual report readability’’, Accounting, Auditing & Accountability Journal, Vol. 14 No. 3, pp. 311-26. Core, J.E. (2001), ‘‘A review of the empirical disclosure literature: discussion’’, Journal of Accounting and Economics, Vol. 31, pp. 441-56. Courtis, J.K. (1986), ‘‘An investigation into annual report readability and corporate risk-return relationships’’, Accounting and Business Research, Vol. 16 No. 64, pp. 285-94. Courtis, J.K. (1995), ‘‘Readability of annual reports: Western versus Asian evidence’’, Accounting, Auditing & Accountability Journal, Vol. 8 No. 2, pp. 4-17. Courtis, J.K. (1997), ‘‘Corporate annual report graphical communication in Hong Kong: effective or misleading?’’, The Journal of Business Communication, Vol. 34 No. 3, pp. 269-88. Courtis, J.K. (1998), ‘‘Annual report readability variability: tests of the obfuscation hypothesis’’, Accounting, Auditing & Accountability Journal, Vol. 11 No. 4, pp. 459-71. Fairclough, N. (1995), Critical Discourse Analysis, Longman, London. Fawcett, R. and Halliday, M. (1978), New Developments in Systematic Linguistics, Batsford, London. Frawley, W. (1992), Linguistic Semantics, Erlbaum, Hillsdale, NJ. Frazier, K.B., Ingram, R.W. and Tennyson, B.M. (1984), ‘‘A methodology for the analysis of narrative accounting disclosures’’, Journal of Accounting Research, Vol. 22, pp. 318-31. Frey L.R., Botan, C.H., Friedman, C.H. and Kreps, G.L. (1991), Investigating Communication: An Introduction to Research Methods, Prentice Hall, Englewood Cliffs, NJ. Gallhofer, S., Haslam, J. and Roper, J. (2001), ‘‘Applying critical discourse analysis: struggles over takeovers legislation in New Zealand’’, Advances in Public Interest Accounting: Advances in Accountability: Regulation, Research, Gender and Justice, Vol. 8, pp. 121-55. Graves, O.F., Flesher, D.L. and Jordan, R.E. (1996), ‘‘Pictures and the bottom line: the television epistemology of US annual reports’’, Accounting, Organizations and Society, Vol. 21, pp. 57-88. Halliday, M.A.K. (1976), System and Function in Language, Oxford University Press, London. Halliday, M.A.K. (1978), Language as Social Semiotic, Edward Arnold, London. Halliday, M.A.K. (1985), An Introduction to Functional Grammar, Edward Arnold, London. Hart, R.P. (1997a), DICTION 4.0: The Text Analysis Program CD-ROM and User’s Manual, Sage Publications Software, London. Hart, R.P. (1997b), ‘‘Rhetoric, hope, and American politics’’, in Trent, J. (Ed.), At the Helm in Speech Communication, Allyn and Bacon, Boston, MA. Hart, R.P. (2000a), DICTION 5.0: The Text Analysis Program, Scolari/Sage, Thousand Oaks, CA. Hart, R.P. (2000b), Campaign Talk: Why Elections are Good for Us, Princeton University Press, Princeton, NJ. Hart, R.P. and Gourgey, H. (1998), ‘‘Accepting the political mantle: stylistic considerations’’, Political Communication, special CD-ROM edition. Hart, R.P. and Jarvis, S.E. (1997), ‘‘Political debate: forms, styles and media’’, American Behavioural Sciences, Vol. 40, pp. 1085-122. Hildebrandt, H.H. and Snyder, R. (1981), ‘‘The pollyanna hypothesis in business writing: initial results, suggestions for research’’, The Journal of Business Communication, Vol. 18, pp. 5-15.

Developments in content analysis

543

AAAJ 15,4

544

Hopwood, A.G. (1996), ‘‘Making visible and the construction of visibilities: shifting agendas in the design of the corporate report: introduction’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 55-6. Houghton, K.A. (1988), ‘‘The measurement of meaning in accounting: a critical analysis of the principal evidence’’, Accounting, Organizations and Society, Vol. 13 No. 3, pp. 263-80. Hyland (1998), ‘‘Exploring corporate rhetoric: metadiscourse in the CEO’s letter’’, The Journal of Business Communication, Vol. 35 No. 2, pp. 224-5. Ingram, R.W. and Frazier, F.B. (1983), ‘‘Narrative disclosures in annual reports’’, Journal of Business Research, March, pp. 49-60. Jones, M.J. (1988), ‘‘A longitudinal study of the readability of the chairman’s narratives in the corporate reports of a UK company’’, Accounting and Business Research, Vol. 18 No. 72, pp. 297-305. Jones, M.J. and Shoemaker, P.A. (1994), ‘‘Accounting narratives: a review of empirical studies of content and readability’’, Journal of Accounting Literature, Vol. 13, pp. 142-84. Kohut, G.F. and Segars, A.H. (1992), ‘‘The president’s letter to stockholders: an examination of corporate communication strategy’’, The Journal of Business Communication, Vol. 29, pp. 7-21. Leary, M.R. and Kowalski, R.M. (1990), ‘‘Impression management: a literature review and two component model’’, Psychological Bulletin, Vol. 107, pp. 34-47. Lemke, J. (1989), ‘‘Intertextuality and text semantics’’, in Gregory, M. and Fries, P. (Eds), Discourse in Society: Functional Perspectives, Ablex, Norwood, NJ, pp. 85-114. Lyons, J. (1995), Linguistic Semantics: An Introduction, Cambridge University Press, Cambridge. McConnell, D., Haslem, J.A. and Gibson, V.R. (1986), ‘‘The president’s letter to stockholders: a new look’’, Financial Analysts Journal, Vol. 42, pp. 66-70. McKinstry, S. (1996), ‘‘Designing the annual reports of Burton plc from 1930 to 1994’’, Accounting, Organizations and Society, Vol. 21 No. 1, pp. 89-111. Mather, P., Ramsay, A. and Serry, A. (1996), ‘‘The use and representational faithfulness of graphs in annual reports: Australian evidence’’, Australian Accounting Review, Vol. 6, pp. 56-63. Milne, M.J. and Adler, R.W. (1999), ‘‘Exploring the reliability of social and environmental disclosures content analysis’’, Accounting, Auditing & Accountability Journal, Vol. 12 No. 2, pp. 237-56. Morris (1994), ‘‘Computerized content analysis in management research: a demonstration of advantages and limitations’’, Journal of Management, Vol. 20 No. 4, pp. 903-31. Ober, S., Zhao, J.J., Davis, R. and Alexander, M.W. (1999), ‘‘Telling it like it is: the use of certainty in public business discourse’’, The Journal of Business Communication, Vol. 36 No. 3, pp. 280-300. Preston, A.M. and Young, J.J. (2000), ‘‘Constructing the global corporation and corporate constructions of the global: a picture essay’’, Accounting, Organizations and Society, Vol. 25, pp. 427-49. Preston, A.M., Wright, C. and Young, J.J. (1996), ‘‘Imag[in]ing annual reports’’, Accounting, Organizations and Society, Vol. 21, pp. 113-37. Saeed, J.I. (1997), Semantics, Blackwell Publishers, Oxford. Salancik, G.R. and Meindl, J.R. (1984), ‘‘Corporate attributions as strategic allusions of management control’’, Administrative Science Quarterly, Vol. 29, pp. 238-54. Schriver, K.A. (1989), ‘‘Evaluating text quality: the continuum from text-focused to readerfocused methods’’, IEEE Transactions on Professional Communication, Vol. 32 No. 4, pp. 238-55.

Smith, M. (1998), ‘‘Conflicting messages in annual reports’’, Accountability and Performance, Vol. 4 No. 2, pp. 43-60. Smith, M. and Taffler, R. (1992), ‘‘The chairman’s statement and corporate financial performance’’, Accounting and Finance, Vol. 32, pp. 75-90. Smith, M. and Taffler, R. (1995), ‘‘The incremental effect of narrative accounting information in corporate annual reports’’, Journal of Business Finance & Accounting, Vol. 22 No. 8, pp. 1195-210. Smith, M. and Taffler, R. (2000), ‘‘The chairman’s statement: a content analysis of discretionary narrative disclosures’’, Accounting, Auditing & Accountability Journal, Vol. 13 No. 5, pp. 624-46. Staw, B.M., McKechnie, P.I. and Puffer, S.M. (1983), ‘‘The justification of organisational performance’’, Administrative Science Quarterly, Vol. 28, pp. 582-600. Steinbart, P.J. (1989), ‘‘The auditor’s responsibility for the accuracy of graphs in annual reports: some evidence of the need for additional guidance’’, Accounting Horizons, September, pp. 60-70. Swales, G.S. (1988), ‘‘Another look at the president’s letter to stockholders’’, Financial Analysts Journal, March-April, pp. 71-3. Swales, J. (1990), Genre Analysis: English in Academic and Research Settings, Cambridge University Press, Cambridge. Sydserff, R. (2001), ‘‘The development and application of text-focused methods for evaluating accounting narratives, with a view to investigating impression management’’, unpublished PhD thesis, Heriot-Watt University, Edinburgh. Sydserff, R. and Weetman, P. (1999), ‘‘Methodological themes: a texture index for evaluating accounting narratives: an alternative to readability formulas’’, Accounting, Auditing & Accountability Journal, Vol. 12 No. 4, pp. 459-88. Tennyson, B.M., Ingram, R.W. and Dugan, M.T. (1990), ‘‘Assessing the information content of narrative disclosures in explaining bankruptcy’’, Journal of Business Finance and Accounting, Summer, pp. 391-410. Thomas, J. (1997), ‘‘Discourse in the marketplace: the making of meaning in annual reports’’, The Journal of Business Communication, Vol. 34 No. 1, pp. 47-66. West (2001), ‘‘The future of computer content analysis: trends, unexplored lands, and speculations’’, in West, M.D. (Ed.), Theory, Method, and Practice in Computer Content Analysis, Ablex Publishing, Westport, CT and London, pp. 159-76. Witte, S.P. and Faigley, L. (1981), ‘‘Coherence, cohesion and writing quality’’, College Composition and Communication, Vol. 32, pp. 189-204. Further reading Hart, R.P. (2001), ‘‘Redeveloping DICTION: theoretical considerations’’, in West, M.D. (Ed.), Theory, Method, and Practice in Computer Content Analysis, Ablex Publishing, Westport, Connecticut and London, pp. 43-60. Subramanian, R., Insley, R.G. and Blackwell, R.D. (1993), ‘‘Performance and readability: a comparison of annual reports of profitable and unprofitable corporations’’, The Journal of Business Communication, Vol. 30, pp. 49-61.

Developments in content analysis

545

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0951-3574.htm

AAAJ 15,4

546 Received October 2001 Revised November 2001 Accepted April 2002

Measurement distortion of graphs in corporate reports: an experimental study Vivien Beattie Department of Accounting, Finance and Law, University of Stirling, Stirling, UK, and

Michael John Jones Cardiff Business School, Cardiff, UK Keywords Company reports, Financial accounting, Graphs, Measurement Abstract Graphs in corporate annual reports are a double-edged sword. While they offer the potential for improved communication of accounting information to users, the preparers of the annual reports can easily manipulate the graphs for their own interests. For over a decade, the empirical financial graphics literature has focused on examining company reporting practices. A particular concern has been measurement distortion, which violates a fundamental principle of graph construction. Unfortunately, it is not yet known whether observed levels of measurement distortion are likely to affect users’ perceptions of financial performance. This study uses an experimental approach to address this issue. Pairs of graphs are shown to establish the level of difference that is just noticeable to graph readers. Six levels of ‘‘distortion’’ are investigated (5 per cent, 10 per cent, 20 per cent, 30 per cent, 40 per cent and 50 per cent). Results indicate that if financial graphs are to avoid distorting the perceptions of users, then no measurement distortions in excess of 10 per cent should be allowed. Users with lower levels of financial understanding appear to be most at risk of being misled by distorted graphs. Further research will be necessary to investigate whether this impact upon perceptions subsequently affects users’ decisions in specific contexts.

Accounting, Auditing & Accountability Journal, Vol. 15 No. 4, 2002, pp. 546-564. # MCB UP Limited, 0951-3574 DOI 10.1108/09513570210440595

Introduction For over 200 years, graphs have been used in many technical and everyday contexts to communicate information effectively. Voluntary presentation graphics are increasingly used in the corporate annual reports of large companies in many countries (see, for example, Beattie and Jones, 2001). This increase in usage can be attributed largely to the changing role of the corporate report – from a formal, statutory document for shareholders to a major advertising and public relations document, serving multiple purposes and multiple audiences (Hanson, 1989; Squiers, 1989; Lee, 1994; Hopwood, 1996). The communication advantages of graphs are well-established and are fourfold. First, graphs attract our attention, especially if their visual saliency is increased by the use of colour (Leivian, 1980). Such visual representations become ‘‘graphical sound bites’’ (Henry, 1995, p. 35). Second, because graphs rely on spatial, rather than linguistic intelligence, we can use our dominant visual sense to ‘‘see’’ the data in a direct and immediate way. This facilitates comparisons and the identification of patterns, trends and anomalies (Korol, 1986; Harris, 1996, p. 164). Third, the data can be readily retreived (Wainer,

1992). Fourth, in the specific context of corporate annual reports, graphs Measurement provide oases of colour and interest that enliven the presentation of distortion of information. graphs in reports Unfortunately, it is also well-established that the preparers of corporate reports have incentives to manipulate the content of these reports, or at least to manage the impression conveyed by them. This behaviour is generally aimed 547 at creating a more favourable view of the company’s performance than is warranted. For studies that discuss the impact of these incentives on accounting disclosure choices, see Watts and Zimmerman (1986) (positive accounting theory); Tweedie and Whittington (1990) and Revsine (1991) (the ‘‘selective financial misrepresentation’’ hypothesis); Murphy and Zimmerman (1993) (the ‘‘cover-up’’ hypothesis); Lewellen et al. (1996) (self-serving behaviour); and Preston et al. (1996, p. 119) (impression management). This manipulation can take a number of forms, including biasing (the selection of favourable information items) and focusing (the enhancement of degradation of aspects of the information set) (Birnberg et al., 1983). In recent years, regulators have paid increasing attention to those aspects of the annual report package that lie outside the audited financial statements, recognising the importance of ‘‘financial communication rather than mere financial reporting’’ (FRC, 1995, p. 23, 1999). These aspects include narratives, graphs and photographs. Regulators have also shown increasing concern regarding the potential for this discretionary material to be manipulated (this concern follows naturally from a period of intense concern regarding the manipulation of the financial statement numbers themselves (e.g. Levitt, 1998)). For example, in the UK, the Department of Trade and Industry’s (DTI) review of company law has proposed that the operating and financial review should be ‘‘reviewed’’ by auditors (DTI, 2000, pp. 189-90), a proposal that goes far beyond the current UK auditing requirements in relation to information in documents containing audited financial statements (APB, 1999). Moreover, the UK’s Accounting Standards Board (ASB) has issued a discussion paper that recognises that graphs are a powerful medium of communication and makes five recommendations regarding the use of graphs in annual reports. These recommendations cover selectivity in the graphs shown, selectivity in the length of time series shown, measurement distortion, the need for simple twodimensional formats and the need for related commentary to be located adjacent to the graphs (ASB, 2000, pp. 28-9). Similarly, the sustainability reporting guidelines issued by the Global Reporting Initiative (GRI, 2000, pp. 8-9) comment on the value of graphs in reports, but note the importance of neutrality in presentation. Beattie and Jones (1992a, p. 1) identify three forms of graphical infidelity. Selectivity relates to bias regarding the choice of variables graphed. Measurement distortion occurs where the physical representation of the numbers on the graph is not directly proportionate to the underlying numbers. Finally, presentational enhancement arises where the design of the graph in

AAAJ 15,4

548

some way enhances or degrades certain aspects of the information set, for example because of the use of three-dimensional forms or because the final year’s results are ‘‘highlighted’’ in a brighter colour. Measurement distortion is the topic of this particular paper. Prior empirical studies of financial graphs have found that material measurement distortions (usually defined by the prior literature as distortions in excess of 5 per cent) occur in a significant number of graphs and that these distortions generally give a more favourable view of the companies’ performance than is warranted. Unfortunately, although the incidence of measurement distortions has been empirically demonstrated on many occasions (see, for example, Steinbart, 1989; Mather et al., 2000), these studies have been conducted without any insight into what level of measurement distortion is generally sufficient to influence a user’s perception of a company’s performance. As a result, the practical significance of the findings from empirical studies of company practice remains unclear. The purpose of the present study is to address this issue. For the first time, we provide experimental evidence to ascertain what level of measurement distortion is sufficient to affect the perceptions of users. Our results not only frame the prior financial graphics research, they also provide much-needed guidance for future research in this area. The remainder of this paper is structured as follows. Section two reviews the theoretical and empirical literature relevant to measurement distortion and formally presents the research question and hypothesis. Section three outlines the experimental method adopted. The results are then presented in section four. A discussion section then uses our results to contextualise the prior research; it also discusses the limitations of the study and identifies directions for further research. The final section concludes. Prior literature The volume of research into financial graphs has increased rapidly since Steinbart’s seminal paper on US data, published in 1989. Researchers in Australia, the UK, the USA and Hong Kong have investigated the use and abuse of graphs in annual reports and prospectuses in both national and crossnational studies. They have documented instances of selectivity, measurement distortion and presentational enhancement. A key aspect of most of these studies has been the empirical documentation of measurement distortion. At least nine studies (see Table I) have investigated this issue in detail. Studies of measurement distortion in graphs have all used a graph discrepancy index. This index is a way of measuring the misrepresentation of the underlying numerical data when they are graphically portrayed. The graph discrepancy index used in financial accounting studies originates from Tufte (1983). Tufte is a famous graphical researcher who devised a ‘‘lie factor’’. In the accounting literature, Taylor and Anderson (1986) adapted Tufte’s lie factor as follows:

319 Fortune 500 annual reports for 1986

1990 annual reports of 85 US and 91 UK leading companies 74 companies covering 12 countries

USA

UK

Australia

USA/UK

12 countries

Australia

Australia

Steinbart (1989)

Beattie and Jones (1992b)

Mather et al. (1996)

Beattie and Jones (1997)

Frownfelter and Fulkerson (1998)

Beattie and Jones (1999)

Mather et al. (2000)

484 IPO prospectuses issued pre 1994

1991 annual reports of 89 leading listed companies

1991 and 1992 annual reports for (i)143 top listed companies and (ii) 44 not-forprofit entities

1989 annual reports of 240 large companies

423 graphs from 50 Fortune 500 annual reports for 1977 and 1978

USA

Johnson et al. (1980)

Companies studied

Country

Study

30% (>|5%|) (i) 13.3% (>|10%|) (ii) 51% (>|5%|)

US: 24% (>|5%|) UK: 24% (>|5%|)

Mean ¼ +10.7% (across 4 KFVs) (i) Mean ¼ +16.4% (across 4 KFVs) (ii) Mean ¼ +105.6% (across max. of 3 graphs per entity) US ¼ +16% UK ¼ +7% (across 4 KFVs)

+86% (across 200 non-KFVs) +2% (across 4 KFVs)

+3.5% (adjusted to take into account favourable and unfavourable trends) (across 4 KFVs)

Non-KFVs: 63% KFVs: 48% (>|5%|)

34% (>|5%|)

Overall 68% (>|5%|)

26% (>|10%|)

Mean ¼ +11% (across 3 KFVs)

Mean absolute GDI of 136%

Not given

Incidence of material measurement distortion

Not given

Mean level of measurement distortiona,b

KFVs: unfavourable distortion significantly more likely than favourable distortion (continued)

Not significant

Significant

Not given

(i) Significant (where performance measured as change in variable over period graphed) (ii) Not applicable

Not significant

Significant

Not given

Favourable/unfavourable; association with corporate performance

Measurement distortion of graphs in reports 549

Table I. Key features and findings of studies of measurement distortion in corporate reports

1991 or 1992 annual reports of 50 of the top 100 companies in each country

Companies studied Australia: +1% France: +36% Germany: –13% The Netherlands: +3% UK: +86% USA: +30%

Not given

Not given

Favourable/unfavourable; association with corporate performance

Notes: a KFV ¼ key financial variable, typically sales, profits, earnings per share and dividends per share b GDI ¼ graph discrepancy index. c Courtis (1997) reports on a study of the 1994-95 annual reports of 327 companies listed on the Hong Kong Stock Exchange, 35 per cent of which include graphs. Of these graphs, 52 per cent were classified as ‘‘misleading’’, in that they violated one of five graph construction guidelines (one of which related to measurement distortion). No specific figures are reported in the paper, however

6 countries: Australia, France, Germany, The Netherlands, UK and USA

Beattie and Jones (2000)

Table I.

Country

Incidence of material measurement distortion

550

Study

Mean level of measurement distortiona,b

AAAJ 15,4

Graph discrepancy index ¼ ½ða=bÞ
1  100% where a ¼ percentage change ðin cmsÞ depicted in graph; i:e: height of last column
height of first column  100% height of first column b ¼ percentage change in data: If, for example, a company’s sales rise from £5m to £10m over a five year period, and this is portrayed in a column graph with the height of the column in year 1 being 5cm and the height of the year 5 column being 10.5cm, then the graph discrepancy index is þ10%: GDI
½ð110=100Þ
1  100% where a ¼ ½ð10:5
5Þ=5  100% ¼ 110 b ¼ ½ð10
5Þ=5  100% ¼ 100 In the absence of measurement distortion, the value of the index is zero. Positive (negative) values indicate the per centage by which the trend in the data is exaggerated (understated) by the graph. Non-zero values can arise from either specific features of graph design (such as a non-zero or a broken axis) or from inaccurate draughtsmanship. To determine whether the distortion is favourable (i.e. flattering) or unfavourable, the nature of the variable and the direction of the trend line must be taken into account. The four most commonly graphed financial variables are sales, profits, earnings per share and dividends per share (Beattie and Jones, 1992a,b). These variables are often termed key financial variables (KFVs). In the case of these variables, higher values are seen as ‘‘better’’ than lower values. Consequently, the exaggeration of an upward trend and the understatement of a declining trend both give a more favourable impression of a company’s performance. In annual reports, it has in recent years been more usual to encounter upward, rather than downward, trends in KFVs. There is, however, no empirical evidence regarding what constitutes a material distortion. Some authors have, however, speculated. For example, Tufte (1983, p. 57) argues that distortions in excess of 5 per cent indicate ‘‘substantial distortion, far beyond minor inaccuracies in plotting’’. In the accounting literature, many studies systematically document the level of measurement distortion found in corporate annual reports and prospectuses, most studies dealing with corporate annual reports. A few studies also

Measurement distortion of graphs in reports 551

AAAJ 15,4

552

investigate the features of graph design (such as a non-zero axis) that give rise to the measurement distortion. Typically, these studies distinguish between favourable and unfavourable measurement distortion and investigate the association between measurement distortion and measures of company performance. The majority of studies are single-country. However, a few more recent studies undertake cross-national comparisons. The key features and findings of these studies in relation to measurement distortion are summarised in Table I. Table I demonstrates a widespread incidence of material distortion. Using a 5 per cent cutoff, the percentage of distorted graphs ranges from 24 per cent (Beattie and Jones, 1997) to 68 per cent (Frownfelter and Fulkerson, 1998). The mean level of measurement distortion found ranges from –13 per cent (in Germany: Beattie and Jones, 2000) to +86 per cent (in the UK: Beattie and Jones, 2000). A few studies also report the frequency distribution of graph discrepancy scores. For example, Beattie and Jones (1992b, p. 300: not shown in table) report that 70 per cent of graphs show no material absolute distortion (i.e.