The process of awarding local and regional grants has been a regular target of criticism by auditors-general dating back at least to the "sports rorts" affair of 1993-94 and the "regional rorts" program of 2003-07. The recent audit report on the Award of Funding under the Regional Jobs and Investment Packages (Report 12, 2019-20) is the latest addition to a long and distinguished series. The Regional Jobs and Investment Packages is a competitive grants program established by the Coalition government as a result of a 2016 election commitment. It is intended to drive economic growth and create jobs in 10 regions across five states by co-funding local projects.

The key issue with such schemes has always been the tension between the official criteria for awarding grants, typically based on local need and the public interest, and the politicians' desire to use government funds for partisan purposes, to reward supporters and attract voters in marginal seats.

Auditors-general and the Australian National Audit Office have tended to take an uncompromising position, insisting that grants should always be awarded on public interest grounds as spelled out in the appropriate rules and guidelines, leaving no explicit room for considerations of political advantage. Departmental officials are required to administer the whole process, receiving the applications, assessing their relative merits, and recommending accordingly to the minister. While ministers have the final say, they too are expected to take note of departmental advice and to decide in terms of the public interest.

How much discretion ministers have to override advice from officials has varied over the years. In the lead-up to the "rorts" scandals, ministers from both Labor and the Coalition were able to play fast and loose with the official criteria with little heed paid to departmental recommendations. The result, as the Auditor-General carefully documented, was a marked political imbalance in terms of grants awarded, with electorates held by government parties receiving an unjustifiably disproportionate share of grants. The grants programs were clearly exposed as blatant exercises in pork-barrelling.

Another approach, taken by a Coalition minister in the Howard government, was to reject a number of departmental recommendations but then require his officials to change their recommendations to be in line with his own preferences, thus giving the false appearance that the final decisions had the department's approval. As the Auditor-General commented at the time, such compromising of due process was "below the standard expected of Australian government agencies in administering grant programmes".

These days, there is greater formal recognition that ministers have a legitimate role in making the final decisions on discretionary grants. This gives them the right, as in most other policy matters, to go against the recommendations of their officials, which, after all, are only advice, not directions. That is, ministers may reach their own judgments about issues such as local needs and the public interest. However, in exercising this right, they are required to exercise due process and be accountable. In the words of the Department of Finance's Commonwealth Grants Rules and Guidelines, ministers, like public servants, must act in accordance with what is "efficient, effective, economical and ethical". In practice, if ministers depart from official advice, they are expected to do so openly and explicitly and also within the parameters of the program's objectives and criteria. They are also specifically required to report annually on all cases where they have decided to approve a grant which the relevant official has recommended against. This allows a degree of political input into final decisions because concepts such as local needs and the public interest are open to differing interpretations. But the accountability built into the framework is designed to deter the more extreme partisan outcomes.

In the case of the Regional Jobs and Investment Packages, initial vetting of applications was the responsibility of the Department of Infrastructure. (Much of the assessment was actually contracted out to the Department of Industry, which in turn engaged a further sub-contractor, a series of arrangements which the ANAO finds to have been very sloppily handled). The department prepared written advice assessing the eligibility of each application and scoring applications in terms of four weighted merit criteria (relevance to a local investment plan, economic benefit to the region, value for money, and capacity to carry out the project). This advice was then considered by a ministerial panel of four ministers who made the final decisions.

The ANAO notes a "high" level of disagreement between the officials' advice and the ministers' decisions. The ministers rejected 28 per cent of applications recommended by the officials and approved 17 per cent of applications rejected by officials. Is this divergence a cause for concern? Where the ministerial panels departed from the advice, they recorded the criteria which they were applying differently (usually in relation to economic benefit and value for money). But the panels did not specify a new numerical score for changed assessments. According to the ANAO, this adversely affected accountability because there was no "clear line of sight" between the department's assessments, the panel's subsequent adjustments and the funding decisions.

Whether the refusal to re-score assessments makes any substantial difference is open to question. No doubt, if they had been so instructed, the ministers' staffers could have attached some numbers calculated to justify the ministers' decisions. But, as often in such exercises, these numbers would have ended up as ex-post rationalisations for decisions made on more general grounds. Listing the general criterion or criteria on which the independent decisions were based is probably the most practicable level of accountability that can be expected.

Such accountability requirements may discourage blatant pork-barrelling but they do not guarantee that ministers have not distorted the process for partisan purposes. On this question, the ANAO's most effective evidence has always been in terms of outcomes. Has the government favoured its own seats over those of the opposition? In relation to the RJIP scheme, Mike Kelly, Labor member for Eden-Monaro, had raised specific concerns that his electorate had received less favourable treatment than the neighbouring Coalition-held seat of Gilmore on the NSW south coast.

The ANAO examined recommendations overturned by the ministerial panel according to which party held the electorates from which the applications originated. It discovered that three-quarters of the rejections came from Coalition-held seats and only one quarter from Labor-held seats. Similarly, approvals for non-recommended applications were distributed 60/40 per cent in favour of Coalition seats. That is, ministers, not surprisingly, took a more active interest in applications from areas where the local members were party colleagues.

However, the ANAO found no clear evidence that this disproportionate level of interest had led to outcomes favouring Coalition-held electorates. Overall, the proportion of successful applications by electorates according to party almost exactly mirrored the proportion of original applications for each type of electorate.

In relation to the two south coast electorates, similarly, the proportion of successful applications closely followed the proportion of original applications, though the total amount of funds granted reflected a 7 per cent increase for Gilmore. In these electorates, the proportion of panel decisions differing from departmental recommendations was about average for the scheme as a whole, though the interventions in Gilmore were more favourable to applications that had been recommended against. The figures could suggest some possible bias in favour of the Coalition in Gilmore, though the ANAO does not make that inference. Instead it refers to the smallness of the sample size and enters a null result on the general issue of bias.

So long as ministers have the final say in the distribution of funds for local grants, some risk of unduly partisan bias is inevitable. The alternative of leaving decisions totally in the hands of non-elected officials or independent appointees would lack the broader perspective that elected politicians can bring. Moreover, our system of representative democracy, in which parties compete in promising specific benefits to local voters, assumes that the victorious party will have some capacity to deliver on its commitments.

The present system, under the Finance guidelines, guarantees a sensible level of due process and accountability, particularly if the process is known to be subject to the hawk-eyed scrutiny of the ANAO. The scandalous pork-barrelling of the past is now much harder to achieve. In the current case, Infrastructure's experiment with outsourcing some aspects of the administration to Industry and beyond has been exposed as a major failure. But in terms of political bias the outcomes appear reasonable.