This obvious power imbalance is a potential threat to the integrity of democratic government, which governments worldwide have sought to constrain through various forms of monitoring and regulation. Along with codes of conduct and whistleblower protection, regulation of lobbying has become one of the devices used to counter diminishing trust in government and growing fears of corruption.

The Organisation for Economic Cooperation and Development, for example, has produced three significant reports on the topic (Lobbyists, Governments and Public Trust, 2009, 2012 and 2014). As the Auditor-General's report spells out, in typically cautious but unambiguous language, Australia's performance is well below the best international standard, both in policy design and in implementation.

Australia's lobbyists register is part of the lobbying code of conduct introduced by the Rudd government in 2008 as one of Labor senator John Faulkner's suite of policies designed to increase government integrity. The code's stated purpose is "to promote trust in the integrity of government processes" and ensure that "contact between lobbyists and government representatives is conducted in accordance with public expectations of transparency, integrity and honesty". It requires lobbyists to register, giving the name of their business and its owners, the names of those employed to carry out the lobbyist's functions and the names of clients for whom the lobbyist is acting.

Former ministers or parliamentary secretaries are barred for 18 months from lobbying on any matter they dealt with while in office. Senior members of the Australian Public Service and the Defence Force face a similar ban but only for 12 months. Responsibility for administering the register, which is published online, lies with the secretary of the Department of the Prime Minister and Cabinet. Government representatives, in turn, are obliged to make sure that any lobbyists they deal with are listed on the register.

The code includes general obligations on lobbyists to behave openly and honestly with government representatives and not to exaggerate or misrepresent the nature of their access to government. Lobbyists also need to keep their lobbying activities separate from any activities on behalf of a political party.

From the beginning, the register's scope and status were limited. First, as is often pointed out, the register and the rest of the code apply only to third-party lobbyists; that is, lobbyists who act on behalf of paying clients. Unlike codes in some other jurisdictions, the code does not try to cover the many in-house lobbyists employed by large, wealthy organisations to further their interests with governments. The original rationale for this restriction, we are told, was that in the case of employees of major companies or peak bodies, the very nature of the lobbyists' employment would indicate to ministers and officials whose interests they were representing.

Auditor-General Grant Hehir makes no direct comment on the restriction and the reasons given. To do so would be beyond his remit, which is to evaluate the implementation of existing policy. But by drawing it to our attention, he invites an immediate response. If the restriction is justified by the ease of identifying in-house lobbyists, the purpose's only purpose must be to inform government representatives who the lobbyist is acting for. But what about the other, high-sounding exhortations to honesty and integrity, not to mention the ban on revolving-door appointments? These should surely apply to all lobbyists, not just third-party lobbyists.

Admittedly, third-party lobbyists are easier to define and identify than in-house lobbyists, making for easier administration of any register. But experience elsewhere shows that some approximate definitions are possible; for instance, in terms of the employee's conditions of appointment or the proportion of time spent on lobbying. If the regulation of lobbyists is intended to restrain the excessive power of vested interests, it needs to cover all people professionally engaged in the activity.

A second limit is that the code and register exist simply as administrative orders and are not enacted in any legislation. Again, many other jurisdictions have followed a legislative approach, which, as the Auditor-General points out, allows for mandatory requirements and standards and can authorise regulators to impose penalties for non-compliance. The current regime's purely voluntary nature is a serious weakness.

Third, when PM&C was given responsibility for administering the code and register in 2008, the then prime minister decided that the department's role in monitoring compliance would not be "significant". We can infer strong resistance at the time from PM&C to taking on a new supervisory function. This objection must have been overcome by an assurance, recorded in writing, that they wouldn't have to do much! Herein lies the cause of many of the problems that follow.

While the audit report's underlying purpose may be to draw attention to an important but neglected aspect of government integrity, the official objective is to evaluate PM&C's implementation of the current policy. Here the report found that administration of the register generally meets the limited expectations adopted in 2008. The department effectively processes applications from lobbyists and regularly updates the register, and reminds listed lobbyists of their obligations under the code. However, the department often fails to meet its time frames for entering new information. It could also do more to evaluate its information's accuracy.

The department does little to secure lobbyists' compliance with the code, in keeping with its original brief. In theory, checking for compliance is delegated to individual government representatives dealing with lobbyists. But the department has not developed any strategy to raise government representatives' awareness of their monitoring obligations.

Finally, and most significantly, the department is criticised for failing to assess whether the register and the code are achieving the policy objectives established in the code. The Auditor-General clearly expects departments to assess the effect of the policies they administer and, in classic audit-speak, to "identify key learnings to inform advice to government on potential refinements to current policy settings". In other words, it isn't good enough for PM&C to continue administering an unsatisfactory policy without advising government how it could be improved.

In his response to the draft report, the secretary of PM&C, Martin Parkinson, accepts some of the findings, including the need to develop an evaluation framework. He stresses the original decision that the code and register are an "administrative initiative not a regulatory regime", which explains the department's approach to implementation. He also argues that some of the recommendations are better suited to a "legislatively based regime that regulates all lobbyists, i.e. not just those employed by third parties". In this way, under the cover of defending the department's performance of its current brief, Parkinson may be quietly flagging his support for a more robust structure, grounded in legislation and with a broader remit.

Through this report, both the Auditor-General and the PM&C secretary, while remaining scrupulously loyal to current government policy, have pointed the way towards more effective regulation of those engaged in lobbying. Two key improvements would be legislative enactment and a broader definition of lobbyists. To this should be added the need for an independent regulator to oversee compliance with the register and the code. Locating this monitoring function within a department under ministerial direction, such as PM&C, will never work well. As with the Finance Department's former administration of politicians' work-related expenses, departmental control is bound to lead to routine box-ticking and timid enforcement. The role needs to be given to an independent integrity agency, such as the Ombudsman (or an eventual anti-corruption agency), or to a separately designated commissioner.

The Auditor-General has placed this issue firmly on the policy agenda. It's now up to politicians to consider whether to pursue it further. There is a strong public-interest argument in favour of more extensive regulation. Greater transparency surrounding who is lobbying government on behalf of whom should aid public scrutiny of government decision-making and reduce the level of hidden influence-peddling. A code of conduct should remind lobbyists to respect basic democratic norms. Limiting the speed of the revolving door would boost government integrity.

But the case should not be oversold. International experience suggests that even the strictest codes are rarely enforced. Moreover, transparency and codes alone will not significantly reduce the unequal power of wealthy interests. The United States, for example, has led the world in registering and monitoring lobbyists yet its democracy is in thrall to wealthy special interests.

By all means, turn up the floodlights on the playing field so we can get a better view of the game and discourage foul play. But don't expect much levelling of the field itself.

Richard Mulgan is an emeritus professor at the Australian National University's Crawford school of public policy. richard.mulgan@anu.edu.au