1 1 PUBLIC SECTOR AUDITING: A COMPARATIVE ANALYSIS OF THE ENABLING LEGISLATION OF THE AUDITORS-GENERAL OF AUSTRALIA # By Mr. Michael De Martinis* School of Accounting and Finance Victoria University of Technology P.O Box Melbourne, Australia and Professor Colin Clark Head, Public Sector Research Unit Victoria University of Technology P.O Box Melbourne, Australia April 2001 * Author for correspondence purposes. # The authors wish to thank Ruth Kiraka for her invaluable research assistance in collecting the data from the enabling legislation. This paper has also benefited from comments received on earlier drafts from Wayne Cameron, Auditor-General Victoria, John Parkinson (Convenor, Australasian Council of Auditors-General and Auditor-General, ACT), representatives from the other offices of the auditorsgeneral including: Ian Summers (Auditor-General, Northern Territory); Terry Hogan (the Audit Office of New South Wales); Barry Rowe (Office of the Auditor-General for Western Australia); Teresa Rafter (Queensland Audit Office); Janine McGuiness (Australian National Audit Office); Johann Jantz (Victorian Auditor-General s Office); Ian McGlen (South Australian Auditor-General s Department); and Dr. Arthur McHugh (Auditor-General, Tasmania). We acknowledge comments from Prof Haddon Storey, Public Sector Research Unit, Victoria University of Technology, Dr Bob Clift, participants at the European Accounting Association Conference Athens 2001, and also an anonymous reviewer. We are grateful for funding provided by the Public Sector Research Unit, Victoria University of Technology. Responsibility for any errors remains with the authors.
2 2 PUBLIC SECTOR AUDITING: A COMPARATIVE ANALYSIS OF THE ENABLING LEGISLATION OF THE AUDITORS-GENERAL OF AUSTRALIA ABSTRACT Supreme audit institutions, such as the auditors-general, are considered a crucial link in the accountability chain between parliament and the executive arm of government. In this context, two landmark reports commissioned by the Parliament of Australia on the Office of the Commonwealth (Federal) Auditor-General provide an appropriate legislative framework on which to base the role, responsibility, and scope of auditors-general (Joint Committee of Public Accounts (JCPA), 1989; 1996). The purpose of this study is to examine the enabling legislation of all nine auditors-general of Australia from a public sector accountability and independence perspective. Following on from JCPA (1989; 1996) and in particular English and Guthrie (2000), this study uses an accountability and independence framework to identify and compare the current legislation applicable to all nine auditors-general of Australia with regard to independence, autonomy, mandate, funding issues, and related parliamentary powers. This study finds that the enabling legislation of the Commonwealth and the Victorian auditors-general have markedly less differences from the suggested framework than the other seven auditors-general. Also, for all auditors-general examined, it highlights specific areas where provisions may be further strengthened through amendments to their enabling legislation. The framework and results of this study may be used to examine the extent to which the enabling legislation of supreme audit institutions in other jurisdictions adequately and appropriately deals with their powers, independence and autonomy, and parliamentary accountability mechanisms.
3 3 1. INTRODUCTION Supreme audit institutions, such as auditors-general, are considered a crucial link in the accountability chain between parliament and the executive arm of government (and its use of public monies). However, relatively little research has examined the extent to which auditorsgeneral fulfil such accountability enhancing roles. Further, nor has a formal theory or model been developed that integrates prior government and public sector accountability literature. Whilst such a model would prescribe how accountability agencies, such as the auditor-general, should best fulfil their role in ensuring government and public sector accountability to parliament, the development of such a fully integrated theoretical model of government and public sector accountability is beyond the scope of this study. However, this study makes a significant contribution by considering an important aspect of the literature as related to the powers and accountability of the auditor-general in ensuring such government and public sector accountability. In contributing to the future development of an integrated theoretical model of government and public sector accountability, the purpose of this study is to examine the enabling legislation of all nine auditors-general of Australia from a public sector accountability and independence perspective. The basis of this examination stems from English and Guthrie (2000) and Joint Committee of Public Accounts (JCPA, 1989 and 1996), and involves identifying the extent to which the enabling legislation of each auditor-general contains provisions regarding a number of key suggested accountability and independence mechanisms. These relate to parliament s powers and oversight of the auditor-general, as well as on the extent to which the legislation addresses the auditor-general s powers, funding, mandate, and independence from direction by parliament. To help with the identification, English and Guthrie (2000) is adapted to provide a basis for further developing an accountability and independence framework that contains a number of key items relating to parliamentary accountability mechanisms and powers required by the auditor-general. The framework in this study extends English and Guthrie (2000) by presenting suggested legislative requirements that will provide for an appropriate level of parliamentary accountability and auditor-general power and independence. This framework allows for an identification of those issues not explicitly addressed in the enabling legislation of the auditors-general, as well as the extent of those items that are addressed in the enabling legislation. Further, the framework facilitates a comparison of the enabling legislation between all the auditors-general examined in this study. Overall, the closer the enabling legislative matches the framework, then the better the legislation in terms of the auditorgeneral having reasonably adequate and appropriate powers, mandate, funding, and independence from government. Whilst the focus of this study is to further develop an accountability and independence framework applicable to auditors-general that can be used to examine their enabling legislation with respect to the adequacy their powers, mandate, and independence, the framework can be applied in other accountability contexts. For example, the framework and results of this study may be used to examine the extent to which the enabling legislation of supreme audit institutions in other jurisdictions adequately deals with their powers, independence and autonomy, and parliamentary accountability mechanisms. Therefore, the framework has potential to make a significant contribution to the development of an overall theoretical model of government and public sector accountability.
4 4 Apart from this study s basic contribution towards the development of a formal theoretical model of government and public sector accountability, the study is further motivated by a recent parliamentary inquiry into a legislative framework for (its) officers of parliament. In April 2000, the Parliament of Victoria, Australia, through the Public Accounts and Estimates Committee (PAEC), requested submissions as part of its inquiry into a legislative framework for Victorian Statutory Officers of Parliament. The PAEC was to undertake an inquiry and report on: 1. An appropriate legislative framework for Victorian officers of Parliament such as the Ombudsman, the Auditor-General and other statutory office-holders, that would recognise the special position of statutory officers of the Parliament in terms of their relationship with the Victorian Parliament but which also ensures that their greater autonomy is accompanied by very clear accountability requirements; and 2. Developments in this area in other jurisdictions. In order to determine an appropriate legislative framework for Victorian Officers of Parliament, including the Auditor-General, this study compares the enabling legislation of all auditors-general of Australia in terms of specific differences in their powers, mandate, funding, and independence. Also compared are the accountability mechanisms available to parliament in terms of their appointment, tenure, and oversight. The examination includes the Commonwealth Auditor-General (C/W), the six State auditors-general: Victoria (VIC), New South Wales (NSW), South Australia (SA), Queensland (QLD), Western Australia (WA), and Tasmania (TAS), and the two Territory auditors-general: Australian Capital Territory (ACT) and Northern Territory (NT). The paper is structured as follows. The next section reviews the relevant public sector auditing and accountability literature. In particular, focus is on prior reports and studies related to the Commonwealth Auditor-General (Barrett, 1996b; English and Guthrie, 2000; JCPA, 1989; 1996; Taylor 1996b). Whilst Mulgan (1997) represents a significant contribution towards the development of a model of government and public sector accountability, English and Guthrie (2000) provide the basis of an accountability and independence framework that highlights key accountability mechanisms required for ensuring of the relationship between parliament and government. Section 3 then proceeds to suggest the legislative requirements for each item within the framework. The suggested legislative requirements are derived from a review of relevant professional and scholarly literature. Section 4 then uses this extended framework to identify where differences exist between the framework s suggested legislative requirements (as identified in Section 3) and the enabling legislation of the nine auditors-general. The final section presents a summary and discussion of the results as presented in the previous section. 2. LITERATURE REVIEW Literature on public sector accounting, auditing, and accountability is extensive (see, for example, English and Guthrie, 1991; Funnel and Cooper, 1998; Mulgan, 1997; Taylor, 1992 and Zifcak, 1997). In recent times the literature has acknowledged a model of accountability that comprises a number of complementary agencies, processes, and channels of accountability between the public and public servants (Mulgan, 1997). Figure 1, which is
5 5 sourced from Mulgan (1997), depicts the agencies of accountability to include Freedom of Information legislation, courts, ministers, parliament, and committees and officers of parliament including the ombudsman and the auditor-general. The four main processes of accountability comprise reporting or accounting, information-seeking or investigation, assessment or verification, and direction and control. The agencies can ensure the accountability of governments or public servants through a number of channels. Insert Figure 1 about here However, the literature also acknowledges that the effectiveness of these accountability agencies, particularly in context of the role of public sector audit, is dependent upon their having independence, autonomy, funding, and a comprehensive mandate (Barrett, 1996a and 1996b; De Martinis, Clark, and Roberts, 1998; English and Guthrie, 1991 and 2000; Funnell, 1996 and 1997; Taylor 1996a and b). In the context of this model, three issues surface that serve as a back-drop to this study. First, there is no reason to expect any one agency, channel, or institution to fulfil all functions of public accountability. Second, whilst there may be a perception that any channel may be considered ineffective or lacking power if it does not direct and control, Mulgan (1997: page 34) points out that focus should be on whether their scrutinising and auditing activities lead eventually to appropriate responses from government, provoked if necessary by other channels of accountability. Third, public sector audit is a necessary but not a sufficient agent for achieving at least the first three processes of accountability. Therefore, Mulgan (1997) argues that the key is to improve the scrutinising and auditing function of the auditor-general, as well as the other complementary channels including parliament and other officers such as the ombudsman. With regard to the auditor-general, the extent to which this office can adequately scrutinise and audit is at the heart of the continuing debate on the role of public sector audit in public sector accountability. In fact, this debate represents a significant part of this literature (see, for example, De Martinis, Clark, and Roberts, 1998; English and Guthrie, 1991 and 2000; Funnell and Cooper, 1998; Mulgan, 1997; Taylor, 1992; Walsh, 1995). Therefore, this literature provides the framework for examining the appropriateness and adequacy of current legislative frameworks of all independent officers of parliament such as auditors-general with regard to mandate, independence, autonomy, and funding issues. In addition, it also provides the basis for reporting on an appropriate legislative framework for Victorian Officers of Parliament that: (i) recognises the special position of Statutory Officers of the Parliament in terms of their relationship with the Victorian Parliament; and (ii) ensures that their greater autonomy is accompanied by very clear accountability requirements. An Accountability and Independence Framework Whilst Mulgan (1997) represents a significant contribution towards the development of an integrated model of government and public sector accountability, the model lacks prescription in terms of how the accountability channels and agencies would best fulfil their public accountability function. English and Guthrie (2000) and JCPA s (1989; 1996) significant contribution towards such a model of public sector accountability is the detailing of key accountability factors or mechanisms required by the auditor-general in terms of appropriate and adequate powers, mandate, funding, and independence from parliament. Thus providing
6 6 the basis of a framework that can be used to examine the enabling legislation of auditorsgeneral with respect to these factors. It was noted above that auditors-general are considered a crucial link in the accountability chain between parliament and the executive arm of government (and its use of public monies). In this context, landmark reports by the JCPA on the Office of the Commonwealth (Federal) Auditor-General (now known as the Australian National Audit Office) such as Guarding the Independence of the Auditor-General (JCPA, 1996) and The Auditor-General: Ally of the People and Parliament (JCPA, 1989), have a common theme regarding the appropriate legislative framework on which to base the role, responsibility and scope of auditors-general. Recommendations therein focus on a legislative framework that ensures government accountability through auditors-general that are not only independent, but are seen to be independent. For example, JCPA (1996: page 7) states that the independence of the auditorgeneral is fundamental to public accountability, and that it is necessary to put beyond doubt the auditor-general s freedom from direction by the executive or the Parliament in the performance of audit functions. In the earlier inquiry, the JCPA (1989) noted the need to strengthen or at least maintain traditional accountability of federal government organisations and officials to parliament through public sector audits that emphasise compliance, probity, and financial processes. Of particular significance is JCPA (1989). The report details a framework that identifies four accountability mechanisms available to parliament related to audit mandate, independence, and funding. These are: scope of audit in the public sector; powers of parliament in relation to audit; funding as determined by parliament; and oversight. It also identifies the necessary powers required by auditors-general to conduct audits. These include wide mandate, independence from direction by the executive arm of government, and sufficient funding as recommended by parliament (not the executive). Table 1, sourced from English and Guthrie (2000), depicts this framework and is discussed in more detail in the next section. Insert Table 1 about here Based on JCPA (1989; 1996), English and Guthrie (2000) examine the role, powers, and autonomy of the Commonwealth Auditor-General in context of the accountability mechanisms available to parliament, as per Table 1. The two-part framework describes (i) the accountability mechanisms available to parliament related to audit scope, auditor-general appointment, tabling of reports, funding, and oversight, and (ii) the powers required by auditors-general to conduct audits related to independence, mandate and funding source. This framework also serves as the framework for this study s comparative analyses. In the first part of the framework, the four accountability mechanism available to parliament relate to: (i) scope of audit in the public sector; (ii) powers of parliament in relation to audit; (iii) funding determined by parliament; and (iv) oversight of the auditor-general. On the scope of audit, there should be no impediment to the auditor-general conducting financial statement and performance audits of all government entities. On the powers of parliament, it should have the power to request audits, have all audit reports tabled in the parliament, and have a parliamentary audit committee appoint and oversee the auditor-general and his/her office. On funding of the auditor-general, this should be determined by parliament via input from parliamentary audit committee. On oversight, an independent audit of the auditorgeneral s office should be conducted regularly.
7 7 In the second part of the framework, the powers or autonomy required by auditors-general to conduct audits relate to three issues: (i) the mandate to perform audits; (ii) independence from direction by the executive arm of government; and (iii) funding. On scope and mandate, auditors-general should perform financial statement and performance audits of all government agencies, authorities, companies, and individuals. On independence, specific issues that should be enshrined in law include: the auditor-general being: free of direction/control from the executive arm of government; reporting to parliament; having wide information gathering powers, appointed by the parliament, and appointed as an officer of parliament; and able to determine the terms and condition of employment of staff. On funding level, this should be sufficient to enable the auditor-general to exercise effective mandate, and be recommended by parliamentary audit committee (not treasury). To facilitate the comparison between each auditors -general enabling legislation, the framework as contained in Table 1 is restructured such that the main legislative requirements necessary for each accountability mechanism and power/autonomy issue are separated. That is, Table 2 details the suggested legislative requirements for each of the three main accountability mechanisms, while Table 3 details the suggested legislative requirements for the two main power/autonomy issues. Both these tables form the basis of the comparative analyses as discussed in Section 4 of this paper and appearing in the first row of Tables 4 to 8. In the next section, for each of the key items appearing in the Tables 2 and 3, prior literature is used to support the suggested legislative requirements. Insert Table 2 about here Insert Table 3 about here 3. SUGGESTED LEGISLATIVE REQUIREMENTS This section discusses the suggested legislative requirements in detail. The basis for the suggested legislative requirements stems from a review of the literature, in particular from recommendations contained in the landmark reports on the Australian National Audit Office (JCPA, 1989 and 1996) and from English and Guthrie (2000). For example, some recommendations resulted in the introduction of a legislative framework that promoted the Commonwealth Auditor-General s independence of the executive arm of government, and strengthened the role of the parliament in ensuring that audit functions are adequately resourced. They also lead to the Commonwealth Auditor-General being an Independent Officer of the Parliament. As such, most of the following discussion owes much to the prior work as contained in JCPA (1989 and 1996). Note, however that when the enabling legislation of the Commonwealth Auditor-General is compared with this study s suggested legislative requirements, differences do exist. Accountability Mechanisms Available to Parliament
8 8 A. Powers of Parliament in Relation to Audit Table 1 shows that a key accountability mechanism to parliament is its power over the appointment and oversight of the auditor-general and his or her office. Therefore, in the context of ensuring such powers, auditor-general independence (as distinct to independence of the public sector audit office (Funnell, 1996)), requires that the appointment, tenure, and removal be determined by parliament itself, and not by the executive arm of government. The same applies regarding the determination of remuneration. Further, parliament should have the power to request audits and have all audit reports tabled in the parliament. At the same time, the auditor-general should not be under any direction or control from, or be impeded by, the executive, nor lack discretion in determining the type of audit and auditee. Although JCPA (1996: page 59) states that auditor independence can be enshrined in legislation, both the professional standards (eg. AUP 32) and academic literature maintain that it should be enshrined in legislation (Barrett, 1996a and 1996b; English and Guthrie, 1991; Funnell, 1997; and Taylor, 1996a and 1996b). However, Funnell (1996) notes that given the importance accorded to independence for public sector auditors, relatively little discussion exists of how interpretations of public sector audit independence have been constructed and sustained. Further, he adds that no concerted attempt has been made to identify the groups which have played the greatest roles in the construction of meaning (Funnel, 1994). An interesting point made by Funnell (1996) relates to the apparent confusion encouraged by the executive between independence of the auditor-general as a person, with that of his or her office. Funnell (1996: page 110) states that to accomplish this the executive has stressed personal characteristics of the auditor-general (i.e. honesty, unbiased, objective) and relevant provisions of audit legislation as both indications and guarantees of the independence of public sector audit, as opposed to the independence of the auditor-general. Until recently, this approach has been effective in sustaining an image of an independent public sector audit function but without placing the auditor-general beyond the reach of the executive. In placing the auditor-general beyond executive reach, Barrett (1996a: page 138) states that one cannot effectively deal with the issue of independence unless one considers the framework that creates the Auditor-General s role as a whole. For example, the appointment and dismissal processes, term and conditions of appointment, the nature of the audit office and its relationship to the auditor-general and the audit mandate, incorporating power to access information including cabinet documents all impact on the extent of independence which exists in practice. In light of the above, and given that the governor or governor-incouncil act on recommendations of the executive government, it would be preferable to provide a more substantial role for the parliament, rather than the governor or governor-incouncil, with regard to the auditor-general s appointment, tenure, and remuneration. In summary, and as shown in Table 2, Column A, the suggested legislative requirements as related to enshrining adequate power to parliament in relation to audit are as follows. The auditor-general should be appointed by parliament or a parliamentary committee. The term of office of the auditor-general should be for a period of at least seven years and be eligible for re-appointment. An independent remuneration tribunal should determine remuneration of the auditor-general, and payment to the auditor-general should be from the Consolidated Fund. Removal of the auditor-general should only be by parliament. The process for appointment of an acting auditor-general should be the same as the process for appointing the auditor-general.
9 9 There should be no requirement to have an annual work plan prepared by the auditor-general and submitted to parliament or other officers and committees. Parliament, through the relevant committee, should have the power to request audits, and all audit reports of the auditor-general should be tabled in the parliament. B. Funding of the Office In the previous section, discussion centred on providing parliamentary power related to the auditor-general s independence, including powers to appoint, remove, and determine remuneration. In this section, focus shifts to addressing the separate but equally important issue of public sector audit office independence, or in other words on the issue of source and level of funding for the auditor-general s office (Funnell, 1996). Barrett (1996a: page 138) argues that dependence for funding does potentially, if not actually, impact on the independence issue. In this connection, there have been questions about control by the executive through the estimates process in negotiations with Finance or Treasury agencies. Funnell (1996: page 118) states that the greatest threat to the operation of an independent public sector audit office in Australia, which the executive has not sought to highlight, has been the auditor-general s dependence on the executive for the resources necessary to meet the audit mandate. Therefore, in order to achieve both an independent auditor-general and an independent public sector audit office, parliament, not the executive, should have the necessary powers to determine the auditor-general s employment conditions (for example, appointment, removal, and remuneration), and the level of funding. Table 2 Column B suggests that from both the accountability viewpoint and auditor-general power and autonomy requirements, funding (level) should be determined by parliament (via input from a parliamentary audit committee), and not Treasury. Further, the office of the auditor-general should have sufficient funds to enable the auditor-general to exercise effective mandate. This would enable the auditor-general to fulfil a comprehensive and extensive mandate. Further, the auditor-general should not need to recover costs from auditees for financial statement audits. C. Oversight of Auditors-General Figure 1 shows a model of accountability that comprises a number of complementary agencies, processes, and channels of accountability between the public and public servants (Mulgan, 1997). Public sector audit is a necessary but not a sufficient agent for ensuring public accountability. While agencies such as the auditor-general serve as vehicles of public accountability, the accountability debate also includes the extent to which the agencies themselves should be accountable to parliament. Phrases such as watching the watch-dog imply that the accountability agencies also should be accountable to parliament. Therefore, in a complete model of public accountability, parliament should have powers to appoint, remove and remunerate accountability agencies such as the auditor-general, as well as have powers of oversight. Further, parliament should have the power to appoint an independent auditor of the auditor-general, and that the independent auditor should report to a parliamentary audit committee (Table 1). In summary, and as shown in Table 2, Column C, the suggested legislative requirements as related to enshrining adequate oversight of the auditor-general are as follows. There should be an independent auditor appointed by parliament to conduct a financial statement audit of
10 10 the auditor-general, and report to parliament. An independent auditor should conduct a performance audit of the auditor-general s office at least once every three years. Powers required by Auditors-General to Conduct Audits D. Mandate of Auditors-General to Perform Audits It was previously mentioned that in order to ensure public sector audit office independence, parliament, not the executive, should have the necessary powers to determine the level of adequate funding. This is because funding has a bearing on the extent to which the office can carry out its mandate, whether comprehensive or not. Barrett (1996a: page 139) states that the nature and extent of the mandate have an important bearing on the office because it effectively determines the resource requirement including the professional and technical abilities likely to be demanded. Therefore, adequate funding of the auditor-general s office and achievement of a comprehensive mandate are not mutually exclusive. That is, without adequate funding, the latter cannot be effectively achieved, and similarly, a wide and comprehensive mandate necessitates adequate funding. Auditors-general should have powers to carry out a comprehensive audit mandate. This includes conducting financial statement, compliance/regularity, and performance (or valuefor-money ) audits of government agencies, authorities, companies, and individuals (Table 1). This power then allows the auditor-general and his office to effectively carry out an important accountability mechanism on behalf of parliament, such that he or she should have no impediment to conduct financial and performance audits of all government entities. In summary, and as shown in Table 3, Column A, the suggested legislative requirements as related to enshrining adequate mandate and scope of the auditor-general to perform audits are as follows. The auditor-general should have a comprehensive mandate to conduct financial statement and performance audits of authorities, agencies, departments, statutory bodies, companies, and individuals, and other bodies funded by government. E. Independence from Direction by Parliament or its Committees Table 1 lists a number of requirements related to ensuring that auditors-general have independence and autonomy from direction by the executive arm of government. Most of these issues have been previously discussed in context of the framework behind Table 1. For example, it was recommended that both the independence of the auditor-general and his or her office should be enshrined in law. That is, the appointment, removal, and remuneration of the auditor-general, as well as the funding of his or her office are determined by parliament, and that such funding be sufficient for effective conduct of a comprehensive mandate. Further, it was noted that parliamentary oversight of the auditor-general is necessary so that as an agency of accountability, he or she also is accountable to parliament (that is, the notion of watching the watch-dog ). In summary, Table 3, Column B, details the suggested legislative requirements as related to enshrining adequate independence from direction by parliament or its committees are as follows. The auditor-general should be independent from direction by parliament or its committees, free of direction or control from the executive of government, parliamentary committees, or individuals, and be an independent officer of the parliament. The auditorgeneral should have discretion to determine the type of audit and auditee, and have wide