The Victorian Audit Act Review

Submission

EXECUTIVE SUMMARY

• From November 1996 to April 1997, a Committee of Review undertook an inquiry into the Victorian Audit Act following the principles laid down in the Competition Principles Agreement.
• The Committee found that greater use of competition would be possible in the area of public audit.  To achieve this it recommended that all audits be put to tender.
• In order to meet the objectives of the audit legislation, the Committee suggested that the existing Victorian Auditor-General's Office should be split into two elements with different functions:
• a new entity-the Office of the Auditor-General-which would be responsible for the audit programme, methodology, practice and reporting to Parliament.  It would also control the tender process.
• a new government business enterprise-Audit Victoria-created out of part of the current Office, which would compete with private audit firms.  Such competition would take place on competitively neutral terms.
• The Committee also recommended that a tender panel should be created to determine which proposal would win a tender engagement.
• To ensure that the Auditor-General retained control of the process, certain reserve powers were granted to that officer.  These include the right of veto over a nominated winning tender, and the right to bypass the tender process in certain important circumstances.  The Auditor-General would report annually to the Public Accounts and Estimates Committee on all cases where these powers were exercised.
• The main benefits to Victoria from these proposals will come in three main forms:
• the quality of the audit work will improve with the more extensive use of competition;  and,
• the purchaser/provider split will improve the administration of the processes arising from a sharper definition of roles, with an Office of the Auditor-General which focuses more on strategic issues and an Audit Victoria with a commercial board concentrating on issues of practice and product quality;  and
• control of the audit will move closer to Parliament and further from the influence of the Executive of the day.
• Although there has been extensive public criticism of the Committee's findings, most of the attacks have been wide of the mark in terms of democratic accountability or have failed to note that the impetus for the Review sprang from the Competition Principles Agreement, to which all Australian States have committed themselves.  Other States will therefore be obliged to conduct similar reviews for much the same reasons in the future.

THE VICTORIAN AUDIT ACT REVIEW

The recent review of Victoria's Audit Act has drawn extensive criticism since its release in late April.  It has been attacked as undermining democratic government in Victoria, as being an attack on civil liberties, and as opening the way for potential government corruption.  Although these criticisms have been trumpeted far and wide in the media, much of the criticism has been based on wrong premises, is factually wrong, and is often misleading.

The review of the Audit Act was one of a series of reviews of legislation and regulation which flows from the pro-competitive environment established by governments following the Hilmer Report of 1993.  Under an agreed process, all governments will be required to review all legislation, including Audit Acts, over the next few years.  Victoria has decided to review its Audit Act early.  But the basic criteria and methodology chosen are those which other governments, too, will have to follow as part of their legislative review process.

THE COMPETITION PRINCIPLES AGREEMENT

The Competition Principles Agreement was signed at the April 1995 meeting of the Council of Australian Governments at which Prime Minister Keating and all State Premiers committed themselves to a fundamental change in the way government power is used in Australia.  In return for their agreement, and because many of the reforms favour the Commonwealth more than the States, the central government agreed to make competition payments of more than one billion dollars to the States over the next five years, in addition to other grants.

The Agreement was the culmination of a series of steps taken by the Premiers and Prime Ministers from the late 1980s.  A Special Premiers' Conference in 1990 set up the agenda;  and a Council of Australian Governments meeting in 1992 made the basic decision, in principle, to develop competitive and national markets.  The Meeting of Premiers commissioned the Hilmer Report, and then the Agreement, together with the changes to the trade practices legislation embodied in the Competition Policy Reform Act of 1995, established the legislative basis.

One important commitment made by the States in the Agreement was to review all legislation which, in any way, restricted competition.  The Competition Principles Agreement is quite specific.  At Principle 5 it states that

• Legislation should not restrict competition unless it can be demonstrated that:

• the benefits of the restriction to the community as a whole outweigh the costs;  and
• the objectives of the legislation can only be achieved by restricting competition

The test is most unusual because it requires that both conditions must be met before any restriction on competition can be sustained.  If any restriction on competition is not in the public interest it should not be maintained.  But, in the case of the review of the Audit Act, it also requires that if there is any way of achieving the objectives of the legislation which does not restrict competition, then the restriction should fall away.

The essential finding of the review committee was that there are competitive ways of achieving the objectives of the Act.  So, the Competition Principles Agreement requires the Government to use them, rather than retain restrictions on the use of competition.

The import of the Agreement is not widely understood.  It creates a presumption in favour of competition, at all times, in all contexts.  Only if competition is not possible can a government impose a legislative structure which restricts competition;  and then, only if the restriction is also in the public interest.  This unusual structure, quite deliberately, follows on from the Hilmer Report.

Many commentators on the Audit Act review have obviously not read the Agreement or the Hilmer Report, or have misunderstood those documents.  The misunderstanding is clear when people argue that the mode of operation of the Auditor-General should not be changed since it is working quite well.  But the fact that something is effective in its current form is not an issue in a competition review.  The review committee is required to consider whether there are competitive alternatives which can achieve the objectives of the legislation.

This point is, again, relevant to those who point to earlier positive reviews of the operation of the Victorian Auditor-General's Office.  What has changed since those reviews is that the governments of Australia have agreed to give precedence to competition, and to prefer competitive structures over restrictive ones.

THE AUDIT ACT 1994

The Audit Act 1994 is a slight document, at just 26 sections.  There is no statement of objectives either in the Act or provided in the Second Reading Speech.  It does have a statement of purpose "to provide for the office of the Auditor-General and the audit of public accounts".  But no hint is given of what is expected to be achieved by the office or the audit.  Since the Competition Principles Agreement requires an evaluation of whether the objectives of the Act can be met by competitive means, the statement of objectives is important.

The review committee read and consulted widely before coming to the view that the implied objective was

To provide for an Auditor-General:

• to report on the use of public funds and other resources to enable Parliament, the Auditor-General's principal client, to assure itself of the appropriate (legal, efficient, and effective) use of those resources, and
• to facilitate an external audit service to authorities and their responsible Ministers.

In order to achieve its objectives, the Act then goes on to set out how Auditors-General should be appointed, dismissed and funded.  Here it is important to notice that the appointment is made by the Governor-in-Council and on terms decided by the Governor-in-Council, but dismissal requires the assent of Parliament.  In practice this means that appointment and conditions are determined by the Premier of the day.  The basic funding for the Office is then allocated as part of the budget of the Department of Premier and Cabinet.  While many people have assumed that the Auditor-General is an officer of the Parliament, or is formally at a distance from Executive government, that is not the case.  Both appointment and financing flow very directly from the Premier and the Department of Premier and Cabinet.  The review committee felt that this was a weakness of the current institutional design and suggested that the position be moved further from the Executive and closer to the Parliament.

The activity of public-sector auditing is then specified by a series of sections of the Act.  Section 8 requires that financial statements of authorities be audited annually and Section 9 that reports based on such audits be sent to the authority concerned, to the portfolio Minister, and to the Minister for Finance.  Two features stand out:  the Auditor-General does not make financial reports directly to Parliament, the line of responsibility is to the Ministers;  and secondly, the possibility that the work be done by persons "authorised by the Auditor-General" is allowed for.  Again we have the line of reporting being to members of Executive government rather than to the Parliament directly.

Section 15 gives wide powers to the Auditor-General to report to both Houses of Parliament on any audit undertaken and Section 16 allows the Auditor-General to undertake audits to determine whether authorities are achieving their objectives effectively, economically, efficiently and in compliance with legislation.  Reporting is again to the Parliament and funding is from the Parliament.  This differs from financial audits where authorities pay an amount to Consolidated Revenue for the audit while the audit activity is budgeted from the allocation of the Department of Premier and Cabinet.

Section 13 is important.  It requires that all audits follow the guidelines set down by the Australian Society of Certified Practising Accountants and the Institute of Chartered Accountants in Australia.  That is, all audits, whether undertaken by the Office or by private auditors, must use exactly the same professional guidelines.

The key points from this review of the Act are that

• in organisational terms, the audit function is currently far closer to the Executive arm of government than to the Parliament;
• this is despite our view that the objectives of the Act set Parliament as the primary client of the audit function;
• the audit serves more than one objective, viz, servicing the informational needs of Parliament and also facilitating the external audit function of government authorities;  and
• public-sector auditing follows the same professional standards whether they are undertaken by public agencies or private firms.

The fact that we saw Parliament as the primary client of the audit led us to the view that models which suggested that the authorities should choose their own auditor were inappropriate.  A large number of submissions had argued against allowing agencies to make audit appointments and we felt that they were correct in that view.

ARE THERE COMPETITIVE ALTERNATIVES?

The Competition Principles Agreement requires that the question be asked as to whether the objectives of the Act can be met without restricting competition.  In an attempt to answer it, the review committee considered a range of alternatives both within Australia and internationally but ultimately came down to two basic models which needed to be considered.

In the Northern Territory, all audits are put out by contract.  The Auditor-General there acts as the agent for Parliament by directing the collection of information but does not do any of the audit work at all, relying on competition to undertake the task.  This is a clear demonstration that the audit function can be performed without restricting competition in the way that Victoria's Act does.

The second example is provided by New Zealand.  In that country a large part of the audit work is put out to competitive tender with a government business enterprise, Audit New Zealand, competing with private-sector providers.  Again this example demonstrates that it is perfectly practicable to use competition for public audit.

In developing its structure, New Zealand made a number of decisions as to how it should be implemented which influenced the thinking of the review panel.

The first was that they split the functions into two along a purchaser and provider model.  A government business enterprise, Audit New Zealand, was established to undertake auditing in direct competition with private auditing firms -- and it has been quite successful in this task.  The task of controlling the audit process, the purchaser in this model, was given to a different agency, the Office of the Auditor-General.  This purchaser/provider split is a normal way in which competition has been introduced across a wide range of public administration, with generally excellent results, and that has been the experience in public auditing in New Zealand.  Victorians know the system well from local government applications and its use is becoming widespread.  As an example, the Commonwealth Attorney-General recently split his Department into a purchasing agency managed by the Solicitor-General and a government business enterprise which offers legal services in competition with private legal firms.  This separation then of Audit Victoria from the Office of the Auditor-General is quite standard and should not be particularly controversial.

New Zealand however chose to retain the Auditor-General as the head of Audit New Zealand in what he describes as a "shareholder role".  We did not recommend that this be followed in Victoria for reasons discussed below.  An interesting feature of the New Zealand model which we did recommend for adoption was the use of a two-envelope system for tenders.  This describes a process where tenders are originally evaluated on the basis of the quality of the proposal and only secondarily on price.  This was supported in the belief that high-quality auditing was more important than any cost savings.  The third important aspect of their system which again we supported was the use of a tender panel to make the judgement of which was the preferred tender.  We recommended that a representative of the Office and two independent persons form the panel.  The Premier announced subsequently that he felt that the independent persons should be chosen by the Public Accounts and Estimates Committee of the Parliament.

CAN THE COMPETITIVE MODEL BE USED TO SATISFY THE OBJECTIVES OF THE ACT?

Remember that we considered that the objectives of the Act were to allow Parliament to be well informed and to provide for an external audit for public authorities.  The Northern Territory and New Zealand examples demonstrated that it was perfectly possible for public auditing to be provided in a competitive setting.

What remained for the committee was to find an organisational and institutional framework such that Parliament could be sure that it was getting information of the type that it wanted and that the external audit function was performed appropriately.  Of these, the latter was the simplest.  There is little doubt that private-sector auditors are able to undertake financial audits to the same standard as those provided by public-sector auditors.  Both groups work to the same professional standards, staff move back and forward between the two sectors, and the task is well known and relatively uncontroversial.  With performance audits we judged that there were a large number of entities capable of providing the necessary skills to undertake such work.  The extension of competition in public auditing does not seem to be impracticable;  the skills are available, and there is a vigorous market in operation for work of that sort.

Trying to find the right balance between the Executive, the Parliament and the independence of the Auditor-General was a more complicated task.  The basic thrust of our recommendations is to move the Auditor further from the Executive and closer to the Parliament while maintaining the important independence features of the position.

We have proposed that the Auditor-General become an independent Officer of the Parliament.  The word "independent" is intended to emphasise that the officer is not subject to the direction of the Speaker of the Assembly or President of the Council in the performance of his or her duties.  The term "Officer of the Parliament" is intended to symbolise the primary responsibility of the Auditor-General to the Parliament.  We have proposed as well that the Office of the Auditor-General be included within the budget of the Parliament rather than that of the Department of Premier and Cabinet.  This is intended to provide a sharper separation between the Executive and the Office and so that the Parliament can make allocation decisions about the amount of auditing it wishes to see undertaken rather than this being decided in the Premier's Department.  We have also recommended much closer liaison between the Public Accounts and Estimates Committee and the Office in the specification of audits.

The independence of the Auditor-General was reinforced by a number of recommendations.  The Auditor-General was made "responsible for overall and specific audit priorities, audit methodology, overseeing the conduct of public sector audits and reporting to Parliament" (recommendation 6.8.1(a)).  In simple terms, this means that the Auditor-General can decide what audits need to be done.  He or she would also be able to establish probity and other rules to prevent conflict of interest where potential auditors were involved in other business activities with the party being audited.  In addition, the audit process would be subject to ongoing review by the Office.  This would involve interaction between the Office and the auditing entity, evaluation of material in process, revisions of the scope of the audit if that were appropriate, and the briefing and debriefing of auditors.  Finally, and fundamentally to the process, the Auditor-General would make a report to the Parliament.

In addition to these process issues, the Auditor-General would have a number of reserve powers.  One is that he or she would have the power to reject the recommendation of the tender panel and award the tender to another party.  We can envisage a situation where the Auditor-General believed that there was a conflict of interest which made it inappropriate to award the tender to a particular entity and provided strong public-interest grounds to veto the proposed tenderer.  A second reserve power involves cases where the Auditor-General believes that it is in appropriate to put a piece of work to tender at all.  The committee judged that "rarely and in cases of extreme public interest" the office bearer might make a judgement that this was the preferred course of action.  He or she would be able to make such a judgement under the proposed model.

In either of these situations, where the competitive process was bypassed, we have recommended that the Auditor-General be required to report on such cases to the Public Accounts and Estimates Committee of the Parliament.  This is not the case of asking for permission but reporting annually on the decisions which have been taken over the course of the year.

Our reason for not following the New Zealand example by making the Auditor-General head of both the Office of the Auditor-General and Audit Victoria was because we thought it important that the Auditor-General have this final right of veto over nominated contractors.  It violates any rules of competitive neutrality to have the Auditor-General as head of a party submitting tender bids and also being able to veto the selected tenderer.  The choice then is either that the Auditor should not be head of Audit Victoria and should have the veto power, or that the Auditor-General should head Audit Victoria and not have the veto power.  In our judgement the former was preferable.

WHAT ARE THE PRINCIPAL BENEFITS OF THE PROPOSED MODEL?

We believe that the proposed model will enhance the quality of the public-sector audit in Victoria.  This improvement will result from three key factors:

(i) The purchaser/provider model has been shown in a wide variety of environments to produce significantly improved outcomes.  These arise in part from the purchasing agency, the Office of the Auditor-General in this case, being able to concentrate on the management tasks of making strategic judgements about what audits were best done, where the emphasis should be put, what were the best methodologies and so on.  The separation of the two functions requires much clearer definition of the tasks to be performed and this too has been a source of quality improvement.  The providing agencies (here Audit Victoria) also benefit from being more focused on the practical and operational issues.  They are removed from the strategic issues so that their objectives become clearer and management is able to concentrate resources more tightly on their specific tasks.

(ii) The quality of financial auditing should improve as the result of neutral and transparent competition between public and private providers.  This is a requirement of the Competition Principles Agreement and was included in that Agreement following strong recommendations and extensive argumentation in the Hilmer Report and long-standing support from COAG.

(iii) Performance auditing at present is not contracted out, and is not contracted out in the New Zealand case (although their system of public-sector management makes direct comparison difficult).  In our view the work of performance auditing, of evaluating whether agencies are achieving their objectives effectively, efficiently and economically, is likely to be the source of greatest gains.  It is very difficult for the Victorian Auditor-General's Office to review the wide diversity of public institutions and do a good job of each.  The skills required to evaluate prisons, universities, trains, hospitals, public utilities, and mainstream government Departments do not reside in any single institution.  We felt that there are a wide range of relevant skills in other entities which should be allowed to compete for such work, and that such competition will improve the quality of the work done.

In our view the gains can be achieved without compromising either the independence of the Auditor-General or the primacy of Parliament.  The model we have suggested gives the Auditor-General, supported by a skilled and professional Office of perhaps thirty staff, control over the audit process.  He or she decides on audit priorities, determines audit methodology, can veto the nominated tenderer, interacts with the winning tenderer in the conduct of the work, evaluates the audit reports and reports to the Parliament.  In addition, the Auditor-General can decide to bypass the tender process completely if the matter concerned is sufficiently serious.

The changes suggested move control of the audit closer to the Parliament and further from the influence of the Executive than is currently the case.  The budget responsibility is shifted from the Department of Premier and Cabinet and to the Parliament.  The main recommended reporting line is direct to the Parliament and not directly to the Ministers as has been the case.  The closeness of the Auditor-General to the Parliament is signalled by proposing that the position be that of an independent Officer of the Parliament.  The role of the Public Accounts and Estimates Committee is also enhanced.  In our view, each of these changes is consistent with the Westminster political tradition and can only serve to strengthen our democracy.

CRITICISMS OF THE REVIEW

"We have ignored the public interest"

In our view, the public interest will be served through the clear separation of the roles of purchaser and of provider;  through transparent, fair and neutral competition for audit engagements;  through an enhancement of the role of Parliament;  and through the preservation of the essential elements of the independence of the Auditor-General.

Already 75 per cent of financial audits are contracted out, so it is obvious that audits can be managed by the Office of the Auditor-General under contract.  Our recommendations extend this principle so that fair competition between public and private providers will be the norm.  Competition is likely to reduce the direct costs of the audit but to increase the indirect costs involved in managing contracts etc.  The other costs, such as the compliance costs of parties audited, are likely to fall.  This will occur as a result of improvements in the skills of the parties undertaking audits.  Equally, those entities which currently have the Victorian Auditor-General's Office as their external auditor but pay for a second external audit should not have to undergo that duplication, with consequent savings to the taxpayer.

"There will be a loss of information and reduction in the knowledge base"

Almost all implementations of the purchaser/provider split model have been subject to this criticism but there is very little support for it in practice.  Separating the purchasing entity from the providing agency clearly creates the possibility that information, which is necessary for the purchaser to make the decisions it wants to, may not flow to it from the providing entity.  Of course the potential problem is one which emerges in any decision-making structure:  how do we get the necessary information to the people who make the important decisions?  That is, it is not really a problem related to the purchaser/provider split but a standard problem whenever decisions need to be made.  All decision-makers need to ensure the appropriate flow of information.

In public auditing in Victoria, some 75 per cent of financial audits are currently contracted out.  This means that it has been necessary for the Victorian Auditor-General's Office to develop procedures to ensure that there is an appropriate flow of information into the Office from the entities which undertake the audits.  The model we have proposed is based on the use of exactly the same mechanisms to achieve exactly the same results.  That is, the purchaser/provider split model provides no new issues in relation to the flow of information.

It has also been argued that the complete separation of purchaser and provider means that the staff in the Office will lose valuable knowledge about auditing.  This supposes that senior auditors, who have spent a lifetime in the profession and who continue to manage others in the audit task, will forget how to do the work.  This is hardly credible.

"We failed to understand the constitutional issues"

The Audit Act is not part of the Constitution of Victoria, and the Constitution itself is simply another Act of the Victorian Parliament, subject to change by straightforward legislative means.

The term "constitution" can also be interpreted more broadly as comprising the set of conventions and practices which guide our political processes.  We believe that our arguments are consistent with the imperatives of these broader constitutional norms.  By shifting budgetary responsibility from the Department of Premier and Cabinet to the Parliament, we have moved the Office from the domain of the Executive and towards the Parliament.  We have similarly shifted the reporting responsibility towards the Parliament, and we have proposed that the Auditor-General become an independent Officer of the Parliament confirming symbolically the shift of locus.  The role of the Public Accounts and Estimates Committee of the Parliament is also enhanced and we expect that its interaction with the Office of the Auditor-General would deepen over time.

"The Auditor-General should not have been reviewed under competition policy"

Our review was not a review of the Auditor-General or of the Office, but rather of the Audit Act.  Under the Competition Principles Agreement all Australian State Governments are required to review all Acts of Parliament to evaluate whether they contain restrictions on competition.  These reviews need to be undertaken under guidelines consistent with the Agreement.  We understand that the NSW review is currently in train.

CONCLUSION

The Victorian review of the Audit Act is one of the early reviews required under competition policy.  It raises important issues of methodology, practice and politics.  In addition, it again demonstrates that implementing reform is hard work.  The outcome is likely to influence much of the competition reform agenda.