Assurance on Sustainability Reporting: An Auditor's View.

In the current turbulent audit environment financial market regulators like the U.S. Securities and Exchange Commission (SEC), the World Bank,(1) the International Organization of Securities Commissions (IOSCO), as well as national governmental bodies, seem to be concerned that auditors are

becoming less interested in financial statement audits with the prospects for growth more pronounced in other assurance areas. Therefore, regulators may try to keep auditors exclusively in the "financial statement audit box" to circumvent concerns about auditor independence and an erosion of the effectiveness of financial audits (see Figure 1).

[ILLUSTRATION OMITTED]

In the long term, this is not a very attractive prospect for the audit profession, particularly if traditional financial statements lose their share within the "information for decision-making" market. Major developments in information technology and globalization are important drivers of market needs. It is within such an environment that the AICPA Special Committee on Assurance Services (1997) (hereafter SCAS 1997) advised that financial auditors try to break out of the traditional box. In this setting, searching for, developing, and getting experience with new assurance services is critical for the survival of the auditing profession.

According to the SCAS, the opportunities are multiplicative as the three-dimensional space expands, as illustrated in Figure 1, into new services, new customers, and new technologies. Whereas the cube reflecting the current financial statement audit is "protected," the space opened up by expanding assurance services, new customers, and new technologies is a "free for all" with many potential competitors.

The arrows in Figure 1 point to three core questions: "what" (service), "who" (customers), and "how" (technology). If the new service is a sustainability report, the new customers are stakeholders like consumers and nongovernmental organizations and the new technology is derived from financial statements audits conducted by multidisciplinary teams. The key question is: Who will be the preferred supplier?

In this paper, I will discuss "assurance on sustainability" as a new service area for an assurance engagement for financial auditors. Contrary to some of the new assurance services recognized by the SCAS, current verification of sustainability reporting results from existing and growing market demand.(2)

An important driver of this demand is the legal requirement to report on environmental issues in several countries. In Denmark and The Netherlands, special environmental reporting is legally required, while in Norway and Sweden, environmental issues have to be included in the annual report. In Australia, Canada, and the U.S., specific environmental output and related financial risks have to be reported (Kolk 2000, 363). Analyzing the international verification market of environmental reports, Kolk (2000, 367) concludes that financial auditors are verifying 56 percent of the reports.

The objective of an assurance engagement, as described in the recently issued IFAC Framework is central to this paper and reads as follows (IFAC 2000, para. 4):

   The objective of an assurance engagement is for a professional accountant
   to evaluate or measure a subject matter (data, systems and processes, or
   behavior) that is the responsibility of another party (management) against
   identified suitable criteria, and to express a conclusion that provides the
   intended user with a level of assurance about that subject matter.
   (emphasis added)

In the following sections, the different elements of this definition with respect to sustainability auditing will be singled out. The elements include the subject matter of a sustainability audit, criteria, audit objectives, audit procedures, and the reporting of a conclusion. To illustrate some practical implications, I refer to publicly available information within The Shell Report 2000. The Shell Report 2000 focuses on "sustainable development"(3) and includes a "verification statement" signed jointly by the financial auditors of KPMG and PricewaterhouseCoopers (PwC)(4) based upon work performed by multidisciplinary teams.

In this respect, the question of whether the financial auditor is the preferred supplier of assurance on sustainability is raised and the need for research is briefly discussed.

SUBJECT MATTER

The subject matter of an assurance engagement can comprise data, systems and processes, and behavior (IFAC 2000, para. 20). A sustainability report such as The Shell Report 2000 contains economic, environmental, and social performance data. It also accounts for (parts of) the accountability process and should provide an insight into the company's ethical behavior.

The IFAC recently formulated criteria regarding whether the subject matter is verifiable: "The subject matter of an assurance engagement is to be identifiable, capable of consistent evaluation or measurement against suitable criteria and in a form that can be subjected to procedures for gathering evidence to support that evaluation or measurement" (IFAC 2000, para. 18).

In an audit of financial statements, the auditor provides assurance as to whether the financial statements give a true and fair view (fairly present) by deploying the accounting framework (e.g., International Accounting Standards or other GAAP) to evaluate the preparation and presentation of the subject matter. Owing to accounting standards, financial data must be delivered in a form consistent with these criteria. However, there are as yet no generally accepted environmental and social accounting standards to consider when the assuror evaluates whether the environmental and social data satisfy these criteria. Therefore, in a particular engagement, the auditor has to specifically consider appropriate criteria at the onset to make sure the data presented in a sustainability report are auditable.

STAKEHOLDERS AS A FOCAL POINT

Compared to financial reporting, reporting for sustainability means a shift in paradigm, because sustainability reporting is not only a matter of disclosure, but also an integral element of the process of communication between the company and key stakeholders. It is an essential element in the communication, dialogue, learning, and the decision-making process (Zadek 1998, 1428) and provides an opportunity for stakeholders to see whether their concerns have been heard.

The most comprehensive standard for the accountability process is currently AccountAbility 1000 (AA1000 1999), which was developed by the Institute of Social and Ethical Accountability with the objective of improving accountability and overall performance of organizations by increasing the quality of sustainable reporting and verification. The process of sustainable reporting is governed by the principle of accountability and concerns the reflection of the aspirations and needs of all stakeholder groups, requiring consideration of powerless stakeholders including future generations and the environment. The stakeholder dialogue process is an important concept in sustainability reporting and verification, because the important norms and objectives for the defined (sustainability) issues are set by the individual company in consultation with its stakeholders. The model in Figure 2 shows the AA1000 process slightly adapted to show similarity with the financial audit process and the Sustainable Development Management Framework (SDMF) as described by Delfgaauw (2001).

[ILLUSTRATION OMITTED]

CRITERIA

Criteria are the standards or benchmarks used to evaluate or measure the subject matter of an assurance engagement. Criteria need to be suitable to enable reasonably consistent evaluation or measurement of the subject matter within the context of professional judgment (IFAC 2000, para. 23). Criteria can be either established or specifically developed (IFAC 2000, para. 51). Established criteria are those embodied in laws or regulations, or issued by recognized bodies of experts that follow due process. Specifically developed criteria are those identified for the purpose of the engagement and that are consistent with the engagement objectives.

The Shell Report 2000 (TSR 2000) does, for example, include management assertions referring to "specifically developed assertions" as well as to established criteria. Examples of the former are "improvement of Child Labor Conditions in Brazil" (TSR 2000, 23) referring to the awarded title of "Child Friendly Company," while an example of the latter is "Group security guidelines conform to the United Nations Basic Principles on the Use of Force and Firearms by Law Enforcement Officials" (TSR 2000, 26).

There are as yet no established criteria for sustainability reporting as a whole, so criteria have to be specifically developed for a number of issues. Criteria do exist for a few specific issues of social and environmental performance and for environmental reporting, but there still is a lot of discussion about the boundaries of environmental and social responsibilities. As long as there is no consensus on the meaning of the concept "sustainable development" and the implications for business, standards cannot become generally accepted. Examples of existing standards that may be considered in such an engagement are:

* SA8000 standards for social accountability toward employees,(5)

* International Labor Organization (ILO) conventions for social accountability toward employees,(6)

* World Business Council for Sustainable Development (WBCSD) social and eco-efficiency indicators,(7)

* European Union Eco-Management and Audit Scheme (EMAS) standard for environmental management systems,(8)

* ISO14001 standard for environmental management systems,(9)

* FEE Framework for Environmental Reporting,(10) and

* Global Reporting Initiative[TM] (GRI 2000) standards for sustainability reporting.

The GRI (2000) standards are currently the most comprehensive. GRI aims to develop generally accepted standards for sustainability reporting. However, developing such standards is premature given that the current level of understanding of the concept of sustainability does not allow agreement on standards yet.

Developing generally accepted standards for sustainability reporting as a whole may be very complex because of the many different stakeholders and sustainability issues that are probably infinite in number. Interests of different stakeholders per issue may even conflict and lead to dilemmas to be managed by the company's board. One could think of the board's decision-making process to invest based upon the attractive calculated (financial) rate of return (shareholders) and consequences for human rights brought into discussion by nongovernmental organizations (NGOs). Ethical norms and reporting standards are, in that case, best negotiated between the stakeholder groups. In addition, the relevance of issues changes over time. As set out in AA1000, externally developed standards may not be suitable for every unique organization, and stakeholder dialogue remains important to discuss the relevance and completeness of reported issues and indicators.

However, there are also arguments in favor of generally accepted standards for sustainability reports. The advantages include comparability between different companies, facility of interpreting reported sustainability information, less confusion, and the facilitating of sustainability audits. Moreover, the relevance of financial issues also changes over time and differs between companies, and yet all companies have to report within the existing financial standards. Changes in relevant issues do not pose a threat to financial standards because they are adapted over time to new circumstances. However, when adaptations lag behind new developments (like the increasing relevance of human capital and other intangible assets), financial accounting standards produce financial reports that do not include all information relevant to the users.(11)

CHARACTERISTICS OF SUITABLE CRITERIA

As long as established criteria are not well developed, the auditor should assess whether any existing criteria are suitable to evaluate the subject matter. The auditor and those engaging him must agree on the criteria, although the auditor may also discuss the criteria to be used with the responsible party (the company) or the intended user (in sustainability reporting: the stakeholder) (IFAC 2000, para. 51). The decision as to whether the criteria are suitable involves considering whether the subject matter is capable of reasonably consistent evaluation using such criteria. Characteristics as to whether criteria are suitable include the following (IFAC 2000, para. 52):

* Relevance: relevant criteria contribute to conclusions that meet the objectives of the engagement, and have value in terms of improving the quality of the subject matter, or its content, so as to assist decision making by intended users;

* Reliability: reliable criteria result in reasonably consistent evaluation or measurement and, where relevant, presentation of the subject matter and conclusions when used in similar circumstances by similarly qualified professional accountants;

* Neutrality: neutral criteria are free from bias. Criteria are not neutral if they cause the practitioner's conclusion to mislead intended users;

* Understandability: understandable criteria are clear and comprehensive and are not subject to significantly different interpretation; and

* Completeness: complete criteria exist when all the criteria that could affect the conclusions are identified or developed, and used.

For a few subjects, criteria have been developed by authoritative institutes (e.g., the International Labor Organization, which developed the ILO code, including criteria for child labor, safety, and equality), but if existing criteria are not applicable or if no criteria exist for a certain subject, a company will need to develop applicable and verifiable criteria internally and in consultation with its relevant stakeholders.

The need for the auditor to assess the suitability of criteria highlights the necessity to verify the accountability process. The auditor must know and understand the stakeholder dialogue process by which the criteria are determined. Especially for determining the relevance, understandability, and completeness of specifically developed criteria, the auditor should know how stakeholders were involved in the development of those criteria. Irrespective of the application of established or specifically developed criteria, suitability characteristics of criteria should be similar to the verification objectives to be met by the auditor.

The lack of established standards also constitutes a threat to the role of the auditor. Without authoritative guidance on the audit of sustainability reports, and the associated degree of assurance, there may be an emergence of an expectations gap. For example, stakeholders may expect the auditor to assure that the sustainability report as a whole gives a fair view of the company's ethical behavior, while the auditor may not be capable or prepared to provide that level of assurance for the report as a whole.

Example: Child Labor

The following example of an assertion concerning child labor is meant to illustrate the difficulties in evaluating such an assertion vis-a-vis certain audit objectives. A few considerations are represented for the audit objectives of accuracy, completeness, relevance, neutrality, and understandability.

Subject matter: Child labor

Assertion: In Brazil, during the year 2000, the Company did not employ persons under the age of 14.

Criterion: ILO code(12)

Indicator: Worker's Age [is less than] 14 years

* Accuracy: If there were two children under the age of 14 employed, then would the assertion be accurate or not?

Here the problem of materiality surfaces. Information is material if its omission or misstatement could influence decisions taken on the basis of the report. In financial accounting, materiality can usually be determined from the perspective of a relatively similar group of stakeholders. In social accounting, defining levels of materiality and reliability becomes more difficult. Not only are the needs of stakeholder groups diverse, but also it should be the stakeholders themselves who determine the level at which information becomes material. In addition, people and money cannot obviously be treated in the same way. A wrongly treated human being can never be "immaterial."(13)

* Completeness: If there has been child labor in Bangladesh that was not reported, then would the assertion be complete or not?

There may be reasons of time or financial constraints that force the company to decide not to include all stakeholders, operations, or locations in a cycle of the accountability process. In that case, the selection criteria and plans for future inclusion should be communicated.

* Relevance: If child labor in Bangladesh were a "hot issue," then would the assertion be relevant or not?

Information is relevant when it has the ability to influence the decisions of users. The great diversity of stakeholders as users of sustainability reports means that an auditor cannot possibly judge relevance in isolation from stakeholders.

* Neutrality: If the company said they made every effort to prevent child labor, then would the assertion about Brazil still be neutral? If child labor in Bangladesh existed and was not reported, then would the assertion about Brazil be neutral?

In both cases, the neutrality criterion may not be satisfied. The information is not free from bias if the company stresses its behavior as good behavior if child labor problems exist in Bangladesh.

Note that besides being an audit objective, neutrality is in fact a condition for verification. The phrase "the company made every effort to" is by definition not verifiable. In the second case, the company only reports on the location where it performed well. The exclusion of other locations may mislead the users of the report and is therefore not neutral if it is not explicitly communicated (see "completeness").

* Understandability: If the assertion was "the company did not employ very many young persons," then would it still be understandable?

This assertion is not clear because the adjective "very many" is subjective and the adverb "young" can be interpreted many in different ways.

One practical remark: As long as a generally accepted reporting framework is lacking, "neutrality" of the report has to be stressed (toward the auditee). Companies reporting on performance are inclined to report that they showed good behavior. That is understandable, as they put effort into the process, but it is certainly not neutral, not free from bias, and therefore not verifiable.

KEY PERFORMANCE INDICATORS AS BASIS FOR DEVELOPING STANDARDS

An important question is whether it is possible to develop generally accepted standards for sustainability reporting as a whole. As this may be too complex because of many different stakeholders and issues, stakeholder dialogue is essential in developing criteria and indicators. Shell sees key performance indicators (KPIs), which imply a selection of issues, as the logical basis for setting targets, driving continuous improvement across its operations, and for developing standards of reporting and verification. Shell is committed to create together with outside stakeholders a number of KPIs reflecting performance in the three dimensions of sustainability (Delfgaauw 2001).

As part of the stakeholder dialogue, Shell explored the following dialogues as basis for selecting KPIs:

1. Postal services or via web site

2. KPI interviews with selected stakeholder groups (external)

3. Shell People Survey (internal)

4. Dialogue embedded in internal processes and culture (internal)

From a verification perspective, the auditor should be engaged at identifying which stakeholders have been included (stakeholder-mapping techniques) and identifying issues and related indicators (AA1000 1999). In the Shell case, verifiers reviewed important dialogue processes and internal control systems. Based upon the outcome of the dialogues Shell identified the following KPI categories (TSR 2000):

* Economic: economic performance and wealth creation;

* Environmental management of environmental impacts and potential impact on climate change;

* Social respect for people (including human and employee rights) and

* Governance and values: integrity and management (stakeholder perception of quality engagement).

Criteria for screening KPIs are listed in Exhibit 1 (TSR 2000, 32).

EXHIBIT 1

Criteria for Screening Key Performance Indicators

Criteria for screening a KPI

* Under Shell's control or high degree of influence

* Can drive the business toward a clear target

* Relevant to internal/external audience

* More than a measure of compliance

* Related to critical activities

* Benchmarkable

* Verifiable

* Meaningful at corporate level

* Builds on existing data streams

Consistent application of KPIs (currently being specifically developed) by lines of industries, over a period of time, could lead to KPIs being transformed into generally accepted standards for sustainability reporting. One could think, for example, of the development of KPIs as "contributions to human rights organizations recognized by the UN expressed in U.S.$ per year by companies in the oil and refinery industry" or KPIs as "employee satisfaction per country expressed on a 5-point Likert scale and employee absenteeism per year per country," both in the clothing manufacturing industry.

AUDIT PROCEDURES

Evidence gathering techniques include the traditional financial audit methods:(14)

* Inquiry: consists of seeking information of knowledgeable persons inside or outside the entity;

* Observation: consists of looking at a process or procedure being performed by others;

* Inspection: consists of examining records, documents, or tangible assets;

* Computation: consists of checking the arithmetical accuracy of source documents and accounting records or performing independent calculations;

* Confirmation: consists of response to an inquiry to corroborate information contained in the accounting records; and

* Analytical procedures: consists of the analysis of significant ratios and trends including the resulting investigation of fluctuations and relationships that are inconsistent with other relevant information or deviate from predictable amounts.

Additional sources, however, may be necessary to gather sufficient and competent evidence during an audit of a Sustainability Report. Examples might include surveys of focus groups, expert commentary such as independent market research agencies, and stakeholder panels to advise the auditor (AA1000 1999, 45).

Examples of verification work performed are summarized in the assuror's report as follows (TSR 2000, 27):

We assessed the statements and data marked with the colored symbol in the social, economic, and managing our business sections by:

* Reviewing management processes and supporting evidence to assess the appropriateness and robustness of these processes and the fairness of the account Shell gives in this Report.

* Assessing the conformity of the updated Shell Group security guidelines with UN Standards relevant to the use of force.

* Interviewing Shell people and, in the case of the development of the Bribery and Corruption Primer, external experts to confirm their involvement.

* Observing, reviewing, and sample testing the KPI development process to assess the accuracy and consistency of the processes used to synthesize stakeholder input contained in records of stakeholder dialogue, agreed as accurate records by both parties.

* Testing data presented in relation to "Tell Shell" and the Shell General Business Principles (SGBP) letters to assess accuracy.

* Reviewing evidence to assess the completeness of the distribution of the Bribery and Corruption Primer to Country Chairmen and the receipt of SGBP letters from Country Chairmen.

* Assessing evidence to support management assertions regarding Shell's participation in external initiatives.

Challenging the traditional financial statement auditor is the statement by Shell that "conventional verification methods alone do not provide the assurance needed. We (Shell) are committed to working with others to challenge traditional thinking and make progress in this area" (TSR 2000, 4). This view emphasizes the need to answer another important question: whether the auditor's level of understanding of social and environmental issues is high enough to evaluate these types of management assertions. If this is not the case, joint responsibility and reporting by an auditor and one or more experts is indispensable.

REPORTING A CONCLUSION

This section will discuss the content and design of a conclusion on a sustainability audit.

According to the IFAC, the auditor expresses a conclusion that provides a level of assurance as to whether the subject matter conforms in all material respects with the identified suitable criteria (IFAC 2000, para. 26).

The auditor's report should be designed to meet the needs of stakeholders. Stakeholders need guidance on the meaning of and value added by the assurance engagement to prevent unrealistic expectations. Therefore, the report should be "long form" rather than "short form" (standardized format) and should clearly set out the scope of the engagement, assurance objectives, work undertaken, and conclusions. The composition of the audit team should also be explained, outlining the professional qualifications of each team member (AA1000 1999).

The financial statement auditor's report states an opinion as to whether the accounts are "true and fair" or "fairly presented" within the scope and constraints outlined. As already explained, it is currently nearly impossible, or at least not cost-effective, to provide a high level of quality assurance on the sustainability report as a whole. The opinion may therefore include different levels of assurance for different parts of the report. For each part of the report, suitable wording of the auditor's opinion should be found. Different audit objectives may require expressions other than "true and fair view" or "fairly presented." To illustrate, the Report from the Verifiers in The Shell Report 2000 (p. 5) in Exhibit 2 is an interesting example of reporting reasonable level of assurance with regard to different subject matters.

EXHIBIT 2

Report from the Verifiers

To: Royal Dutch Petroleum Company & The "Shell" Transport and Trading
Company, p.l.c.

We have been asked to verify the reliability of selected performance
data and statements. We have done so and marked these sections with the
different coloured symbols illustrated below within The Shell Report
2000 of the Royal Dutch/Shell Group of Companies. The preparation of
The Shell Report is the responsibility of management. Our
responsibility is to express an opinion on the reliability of the data
and statements indicated, based on the verification work referred to
below.

In our opinion:

* the data on financial performance marked with the purple symbol are
properly derived from the audited Financial Statements of the Royal
Dutch/Shell Group of Companies for each of the ten years ended 31
December 1999

* the health, safety and environmental (HSE) statements and graphs,
together with the explanatory information, performance data tables
and notes in the Annex, properly reflect the performance of the
reporting entities for each of the HSE parameters marked with the red
symbol

* the statements and data marked with the yellow symbol relating to the
systems and processes Shell has put in place to manage social
performance are supported by appropriate underlying evidence and
present a balanced view.

Basis of opinion

There are no generally accepted international standards for the
reporting or verification of environmental performance data or of
processes to manage social performance. We have adopted a verification
approach that reflects emerging best practice, using a framework based
on the principles underpinning international standards on financial
auditing and reporting. Therefore, we planned and carried out our work
to obtain reasonable, rather than absolute, assurance on the
reliability of the performance data and statements tested. We believe
that our work provides a reasonable basis for our opinion.

Verification work performed

In planning and conducting our work, we included environmental and
social experts within our team. The work carried out is described in
the Report:

* Financial (........)

* Environmental, including health and safety (.........)

* Social (..........)

In addition, we examined the draft Report to confirm the consistency of
the information reported with the findings of our work.

Considerations and limitations

It is important to read the HSE statements and graphs in the context of
the explanatory information and notes in the Annex and the notes to HSE
graphs and performance tables. HSE data are subject to many more
inherent limitations than financial data given both their nature and
the methods used for determining, calculating or estimating such data.

We did not carry out any work on data reported in respect of future
projections and targets.

It is also important to note that the financial data reported are not
sufficient to ensure a thorough understanding of the financial results
and the financial position of the Group.

KPMG Accountants NV
The Hague

PricewaterhouseCoopers
London

From a content point of view, the Shell verifiers issued an integrated report, embracing the economic, social, and environmental dimensions. The report starts by issuing conclusions for the specific dimensions, provides an explanation of the verification work undertaken, and explains considerations and limitations. Unique is the comprehensive explanation provided by Shell--in its role as auditee--as to why transparency is important and verification is necessary: "Beyond assuring accuracy and reliability, verification increases stakeholder confidence that what is being reported is a fair picture of performance. It also improves our ability to monitor and manage our business" (TSR 2000, 4-5).

Also unique for an auditors' report, the verification work carried out is described ("message from the verifiers") in the notes of the economic, social, and environmental parts of the report. To break with the traditional layout, colored symbols (purple for economic, red for environmental, and orange for social) point to the text in the report that has been verified to meet "the needs of stakeholders."

SOME CONCLUSIONS

Based upon the first experiences with sustainability reporting and auditing, some preliminary conclusions can be drawn. First, the lack of established criteria for sustainability reporting is currently an obstacle in sustainability audits. Extensive knowledge and experience has to be formulated before a consistent and all-inclusive framework for sustainability reporting can be developed. As key performance indicators are being developed for internal management (to drive continuous improvement) and external reporting, data will gain importance. However, sustainability includes a very important prospective element, so it could be that systems and processes are a more important subject matter for stakeholders to be verified than retrospective behavior and data.

Furthermore, the reporting and auditing of the nonfinancial systems and processes play an important part in accounting for ethical behavior, while the auditor's assessment of the suitability of specifically developed criteria creates the need for verification of the stakeholder engagement and dialogue process. Ethical behavior and accountability is by definition a process of continuous improvement, because knowledge and experience are built during the process. The process of deliberation and decision making should be made transparent. Consequences of decisions to be made by the company will frequently--and inherently--be conflicting between the different dimensions of sustainability (profit vs. human rights). Therefore providing insight into ethical behavior should always include the accountability process in addition to important performance data. In other words, companies should report explicitly about the manner in which dilemmas are being solved.

In addition, currently the auditor can only provide quality assurance--or judge relevance and completeness--at the level of individual assertions.(15) This is because there are too many different assertions and too many different stakeholders to provide a high level of quality assurance on the report as a whole.

Finally, some thoughts about the position of the financial auditor in the verification of sustainability market. Major competitors that are expected to enter this market are social or environmental experts, like sociologists, ethicists, and physicists. In addition, representatives of nongovernmental organizations can be expected to be involved in verification of sustainability reports. Especially in the environmental dimension, specialists issuing verification statements are available.(16) However, financial auditors do possess some relative advantages, such as the existence of large multidisciplinary firms performing audits in accordance with their worldwide firm's comprehensive audit approaches. Also, financial auditors get significant support from professional bodies of accountants. Last, financial auditors can rely on their reputation (expertise of verification as a process), have experience in cooperating with other experts, and are known to be independent. The latter attribute stresses the need to focus on verification and not on consultancy. However, to verify sustainability reports, multidisciplinary cooperation seems to be indispensable. Finally, it can be said that financial auditors have the necessary skills in reviewing information systems, verification of data, and in the reporting of information to those outside the organization. In the words of Bob Elliott (Chairing the SCAS, 1997), "we have many of the needed competencies."

Need for Research

Sustainability reporting and verification are at a very early stage of development. Thus research, experimentation, and dialogue are necessary to stimulate continuous improvement.

An inquiry into stakeholders' expectations of verification of sustainability reporting could detect an expectations gap. Do stakeholders want independent third parties to verify the report and should (local or direct) stakeholders be involved in the verification? Should reporting and verification be legally required? What level of assurance do stakeholders expect and is that level achievable? While levels of assurance do differ among different parts of a sustainability report, research should investigate how different levels of assurance can be made clear in the auditor's report in order to prevent the occurrence of an expectations gap. Most importantly, the convenience of established standards for sustainability reporting should be investigated, since standards should be tailored to stakeholders' information needs.

Assuming established standards are useful, research into the effectiveness of standards like the GRI sustainability reporting standards is opportune (GRI 2000). Furthermore, the role of the auditor in relation to the stakeholder dialogue should be studied. An interesting research issue is what is the appropriate degree of involvement in the stakeholder dialogue to verify that it meets certain norms (formulated in a process standard). Finally, the methodology to assess the suitability of specifically developed criteria in relation to the stakeholder dialogue needs study to assure an effective provision of assurance on sustainability.

This paper is based on a presentation, entitled "Assurance on Sustainability: A Challenge for the Financial Auditor," at a plenary session of the 6th International Symposium on Audit Research, Maastricht, The Netherlands, July 7, 2000. The author thanks Steven Maijoor, Ted Mock, and the editor for helpful comments and suggestions. This article refers to publicly available information about The Shell Report 2000. Verification of this report is being discussed from an outsider's viewpoint.

(1) The profession has been criticized by the World Bank, particularly in the context of the SE Asian economic crisis, for alleged failure by auditors to ensure that financial statements properly comply with the accounting framework.

(2) Examples of sustainability verification statements of British Airways, ENI, ING, Novartis, Norsk Hydro, Shell, Statoil, Volkswagen, Body Shop, and Novo Nordisk, can be found on the Internet.

(3) For further details see Delfgaauw (2001).

(4) KPMG and PwC are appointed jointly as financial auditors of the Royal Dutch/Shell Group of Companies. One of the consequences is that both firms are jointly and severally responsible (and can be held liable) for the audit. In addition, Royal Dutch/Shell can make use of specific expertise from both firms.

(5) Social Accountability 8000 (SA8000) has been developed under the auspices of the Council on Economic Priorities Accreditation Agency (CEPAA) to promote socially responsible production. The CEPAA developed SA8000 in 1998 in accordance with 12 International Labor Organization conventions and UN human right treaties. CEPAA accredits independent audit firms to monitor conformity to the SA8000 Standard (http://www.cepaa.org).

(6) International Labor Organization, C138, Minimum Age Convention, 1973, Art. 2 paragraph 4.

(7) WBCSD is a coalition of some 140 international companies united by a shared commitment to sustainable development, i.e., environmental protection, social equity, and economic growth (http://www.wbcsd.ch).

(8) Eco-Management and Audit Scheme was adopted by the European Council on June 29, 1993 and allows voluntary participation by companies in the industrial sector. The aim of the scheme is to promote continuous environmental performance improvements of industrial activities by committing sites to evaluate and improve their own environmental performance (http:www.europa.eu.int/comm/environment/emas).

(9) ISO 14001, published in 1996, offers a common, harmonized approach for organizations to achieve and demonstrate sound environmental performance. The main objective of ISO 14001 is to help organizations manage the environmental aspects and impacts of their operations while working toward continual improvement (http://www.bvqina.com/ iso14000.html). Further information about the International Organization for Standardization (ISO) can be found on the web site http://www.iso.ch.

(10) FEE discussion paper toward a generally accepted framework for environmental reporting, Brussels, European Federation of Accountants, Environmental Task Force, 1999.

(11) Webster (2000, 216-224).

(12) International Labor Organization, C138 Minimum Age Convention 1973, Art 2 paragraph 4, specifies a minimum age of 14 years for employees and workers in countries whose economy and education facilities are insufficiently developed.

(13) Since it is difficult to verify the reliability of a sustainability report as a whole, assessment of materiality for the Report as a whole is not useful. Materiality levels can be set at the level of individual assertions such as human rights. Because of the complexity of issues and lack of established standards, it might be more effective to discuss "materiality" of detected errors and deviations (ex post), than to set materiality levels ex ante to plan the extensiveness of testing.

(14) Note that the "Techniques to obtain audit evidence" as mentioned by AA1000 (1999, 44) are similar to (financial) types of audit evidence (Knechel, 1998, 25-27).

(15) Srivastava and Mock (2000).

(16) Aspinwall, Arthur D. Little, ERM, and DNV are some examples (Kolk 2000, 367).

REFERENCES

AccountAbility; Institute of Social and Ethical Accountability. 1999. Accountability 1000 (AA1000). London, U.K.: Framework.

American Institute of Certified Public Accountants (AICPA), Special Committee on Assurance Services (SCAS). 1997. The Report of the Special Committee on Assurance Services.

Delfgaauw, T. 2001. Reporting on sustainable development: A preparer's view. AUDITING: A Journal of Theory & Practice (Supplement): 67-74.

Global Reporting Initiative[TM] (GRI). 2000. Sustainability reporting guidelines on economic, environmental, and social performance. Boston, MA. Available at www.globalreporting.org.

Haan de, Leonie. 2000. Strategic analysis for sustainability audits. Unpublished paper, University of Amsterdam.

International Federation of Accountants (IFAC). 2000. International Standard on Assurance Engagements. New York, NY.

Knechel, W. R. 1998. Auditing. Cincinnati, OH: South-Western College Publishing.

Kolk, A., 2000. Verificatie van milieuverslagen. (Verification of Environmental Reports.) Maandblad voor Accountancy en Bedrijfseconomie (September): 363-375.

Srivastava R., and T. J. Mock. 2000. Evidential reasoning for WebTrust assurance services. Journal of Management Information Systems 16 (Winter): 11-32.

The Shell Report (TSR) 2000. 2000. How do we stand? People, Planet & Profits. Available at: http://www.shell.com.

Webster, A. M. 2000. New math for a new era. Interview with Baruch Lev. Fast Company January-February): 216-224.

Zadek, S. 1998. Balancing performance, ethics, and accountability. Journal of Business Ethics 17: 1421-1441.

Philip Wallage is a Professor at the University of Amsterdam and Partner of KPMG Accountants, Amsterdam.