Could South Australia’s proposed ban on political donations run into trouble in the courts?
21.11.2024
In March 2022, the Labor Party defeated the Liberal government in the South Australian election. Amongst the promises campaigned on by the victorious new Premier, the Hon Peter Malinauskas MP, was a commitment to ban political donations. It would be over two years before any such plan was announced, with the Attorney-General, the Hon Kyam Maher MLC, remarking in August 2022 that ‘[t]here are constitutional issues we need to look at when we do the drafting to make sure we overcome’. But on 12 June 2024, the Premier released a draft of the Electoral (Accountability and Integrity) Amendment Bill 2024 (Draft Bill or DB), with his government’s intention clear: ‘We want money out of politics’. At the time of writing, public consultation on the Bill has ended and we await its results.
The spectre of constitutional challenge has loomed over the government’s proposal. In addition to the Attorney-General’s comments, the Premier has admitted that ‘any legislation could be sorely tested’.
This post assesses the risk that such a challenge could succeed. As will be seen, though the Draft Bill pursues an undoubtedly groundbreaking agenda, it does so at the expense of non-incumbent parties and candidates. When regard is had to the requirements of the implied freedom of political communication—the constitutional doctrine under which any eventual legislation is most likely to be tested—and particularly the application of that doctrine in a series of High Court decisions concerning New South Wales’s electoral laws, it becomes clear that the Draft Bill risks being found invalid for unjustifiably burdening political communication. To understand that risk, however, we must first look to what the Draft Bill precisely proposes.
The Draft Bill
The Electoral Act 1985 (SA) (Electoral Act, Act or EA), which the Draft Bill seeks to amend, governs the administration of elections in South Australia as well as the funding and expenditure of parties, candidates, and MPs. For present purposes, three areas of regulation are relevant: donations, public funding, expenditure.
Donations
The crown jewel of the Draft Bill is the proposed s 130ZCE, falling within the proposed div 6A sought to be implemented into pt 13A of the Act. Whereas the current Act makes lengthy provision for reporting requirements regarding gifts made to parties and candidates without otherwise meaningfully limiting their quantum (EA ss 130ZE–130ZI), s 130ZCE would make it unlawful for a person to make or accept an ‘electoral donation’ to a party, MP, group, or candidate. ‘Electoral donation’ is defined as a ‘donation’ made to or for the benefit of a party, or to an MP or candidate solely or substantially for State electoral purposes or to fund ‘political expenditure’ (DB cl 28). ‘Political expenditure’ is defined as, generally, expenditure incurred in conducting polls and the like, generating political material, and publicly expressing views on candidates, parties, and political issues in the election (EA s 130A(1)). ‘Donation’ is relevantly defined as a gift, but does not include a party membership fee of $100 or less (indexed as at 2023, as with all other amounts referred to in the Draft Bill).
There is a caveat. This restriction is not applicable to a ‘designated participant’ (DB cl 28), which refers to parties formed after the last election and no later than eight months before the relevant election (‘newly registered political parties’) (DB cls 9 and 28), non-incumbent independent candidates (DB cls 3 and 28), and groups containing only non-incumbent independents (DB cls 3 and 28). These entities can receive electoral donations of up to $2,700 each and overall no more than their expenditure cap (DB cl 28).
Public funding
In order to compensate for the shortfall in available finance that parties and candidates will now face, the Draft Bill also alters the existing public funding scheme, which currently reimburses parties and ‘above-the-line’ Legislative Council candidate groups for campaign expenditure up to a maximum amount (EA pt 13A div 4). To be eligible for this funding, a candidate currently must either be elected, receive 2% of the total primary vote if they stood for the Legislative Council, or receive 4% of the total primary vote if they stood for the House of Assembly (EA s 130Q(1)). Similar requirements apply to Legislative Council candidate groups (EA s 130Q(2)). Under s 130P of the Act as-is, an eligible candidate (or group, as the case may be) is entitled to reimbursement after the election of up to $3 for each first preference vote they received if they were endorsed by a party with at least one incumbent MP or were an incumbent independent. Independent non-incumbent candidates or candidates endorsed by a party without an incumbent MP are entitled to up to $3.50 for each first preference vote that falls within 0% to 10% of the total primary vote and $3 for each first preference vote thereafter. The Draft Bill does not alter this area of the funding regime other than to raise the eligibility threshold for unsuccessful Legislative Council candidates and groups to 4% of the total primary vote (DB cl 15).
Beyond that, however, two additions to the public funding scheme are notable. First, given that the current scheme operates only by way of reimbursement, the Draft Bill provides for the availability of advance funding for general elections as follows (DB cl 14):
Table 1
| Entity | Maximum advance funding |
|---|---|
| Newly registered political parties | $5,000 per candidate |
| Existing parties | 80% of the amount that would be payable under s 130P calculated by reference to first preference votes received at the last election |
| Incumbent independents | 80% of the amount that would be payable under s 130P calculated by reference to first preference votes received at the last election |
| Non-incumbent independents | $5,000 |
| Independent candidate groups for the Legislative Council with at least one incumbent candidate | 80% of the amount that would be payable under s 130P calculated by reference to first preference votes received at the last election |
| Independent candidate groups for the Legislative Council without an incumbent candidate | $5,000 |
Any such advance funding is deducted from what is payable after the election under s 130P (DB cl 14).
The second key addition to the funding scheme relates to ‘operational funding’ (DB cl 18). The Act as-is entitles parties (with at least one MP) to two payments per year of up to $7,000 if the party has less than five MPs, or up to $12,000 if the party has six or more MPs (EA s 130U). This funding currently cannot be used for political expenditure (EA s 130W).
By contrast, and in addition to a one-off payment to help parties and candidates comply with the new rules (DB cl 21), the Draft Bill entitles parties with at least one MP to two yearly operational funding payments of the lesser of $47,000 per MP or $700,000 (DB cl 20). It also extends operational funding to incumbent independents, who are entitled to two yearly payments of $15,000 (DB cl 20). Furthermore, and perhaps most significantly as regards funding, the restriction on the use of operational funding is heavily relaxed: parties can use up to 50% of their operational funding for political expenditure whilst independent MPs can use all of it (DB cl 23). None of this is extended to non-incumbents, however.
Expenditure
Finally, the Draft Bill seeks to amend the expenditure rules in pt 13A div 6. Most critically, it does away with the ‘opt-in’ character of the current scheme (which only applies to those who have sought public funding (EA s 130Y)) and instead applies the caps to all (DB cl 25). Beyond that, some dollar-value amounts have been reduced and all amounts have been re-indexed to a fixed point of the March 2023 quarter (DB cls 9 and 26), as opposed to the Act’s current reference of the March 2014 quarter (EA s 130A(8)), such that the caps have practically shrunk. For example, if an election were held in this financial year, a party whose expenditure cap is $500,000 could expend only that much under an amended Electoral Act whereas, under the Act today, it could expend $651,802.70.
The following table compares the optional caps under the current Act with the mandatory caps under the Draft Bill before indexation:
Table 2
| Entity | Cap under current Act (indexed as at March 2014) | Cap under Draft Bill (indexed as at March 2023) |
|---|---|---|
| Parties endorsing candidates for the Legislative Council only | $500,000 | $100,000 per candidate up to $500,000 |
| Parties endorsing candidates for the House of Assembly | $75,000 per electoral district in which a candidate is endorsed (less the amount that candidates expend in their own right—see below) plus $100,000 per any candidate endorsed for the Legislative Council up to $500,000 | Unchanged |
| Endorsed and independent candidates for the House of Assembly | $100,000 | Unchanged |
| Independent candidate groups for the Legislative Council | $500,000 | $100,000 per candidate up to $500,000 |
| Independent candidates for the Legislative Council | $125,000 | $100,000 |
An example further illustrates the discrepancy in the positions of parties and candidates under the Draft Bill when all of these proposed amendments are taken together. Consider the Labor Party, the Liberal Party, a hypothetical newly registered political party (New Party), a hypothetical incumbent independent in the House of Assembly (Independent (Incumbent) A), and a hypothetical non-incumbent independent candidate for the House of Assembly (Independent (Non-Incumbent) B). Let it be assumed in the scheduled 2026 State election that: (1) each party endorses a candidate in each House of Assembly seat and each Legislative Council seat up for re-election (as both Labor and the Liberals did at the last election); (2) that Independent (Incumbent) A received 10,000 first preference votes at the last election (which is approximately as many as independent MP Troy Bell received at the last election); (3) and that each entity has already received two instalments of operational funding, if eligible (two being the number of instalments that would be available if the reforms come into effect between March and September 2025). The following table sets out the expenditure caps and maximum public funding ahead of the election under the relevant formulae in the Draft Bill (funding under s 130P is excluded since this only reimburses parties and candidates after the election for money already spent):
Table 3
| Labor Party | Liberal Party | New Party | Independent (Incumbent) A | Independent (Non-Incumbent) B | |
|---|---|---|---|---|---|
| Available advance funding | $1,290,104.80 | $1,205,083.20 | $265,000 | $24,000 | $5,000 |
| Operational funding usable for political expenditure | $700,000 | $700,000 | $0.00 | $30,000 | $0.00 |
| Total pre-election public funding | $1,990,104.80 | $1,905,083.20 | $265,000 | $54,000 | $5,000 |
| Expenditure cap | $4,002,500 | $4,002,500 | $4,002,500 | $100,000 | $100,000 |
| Shortfall | $2,012,395.20 | $2,097,416.80 | $3,737,500 | $46,000 | $95,000 |
As seen here, incumbents are at a significant advantage—for example, the Labor Party would be entitled to at most $1,990,104.80 in funding whereas the New Party would be entitled to at most $265,000. The fact that non-incumbents may still receive donations under the Draft Bill is of little consolation, given that incumbents need only comply with the statutory requirements and apply for public funding whereas non-incumbents will have to undertake the laborious process of hopefully courting private donations. What is evident is that there is a wide gulf between the funding available to incumbents and that available to non-incumbents who are otherwise in the same position.
Constitutional vulnerability
The implied freedom of political communication
The Draft Bill proposes an ambitious agenda. However, with that ambition comes risk in the form of a potential challenge to any amended Act’s constitutional validity, most likely in the form of a challenge invoking the implied freedom of political communication.
The implied freedom is a restriction upon the legislative powers of the Commonwealth, States, and Territories which stops them from creating unjustified burdens upon the ‘free flow of communication’ on government and political matters (Unions NSW v New South Wales (2013) 252 CLR 530 at [27] (Unions (No 1))). The existence of the implied freedom, first recognised in two 1992 decisions, is premised on the notion that legislatures cannot frustrate the operation of the system of responsible and representative government prescribed by the Constitution, and that free political communication is ‘indispensable’ to that system (Australian Capital Television Pty Ltd v Commonwealth at 138 (Mason CJ) (ACTV)).
Since Brown v Tasmania (Brown), the relevant test has been that a law which burdens the freedom (Step 1) must pursue a legitimate end (Step 2) and be reasonably appropriate and adapted to do so in a manner compatible with the system of representative and responsible government (Step 3). Step 3 is to be undertaken by successively asking whether the law is suitable, necessary, and adequate in the balance (‘proportionality testing’ or ‘structured proportionality’).
As a side note, it is worth observing that it is not certain that a majority of the High Court will employ structured proportionality if a constitutional challenge to the reforms comes before it. With the successive retirements of French CJ, Kiefel CJ, Bell J, and Keane J, the original quartet of Justices who endorsed structured proportionality in McCloy is gone. The new Chief Justice, Gageler CJ, prefers an alternate test of ‘calibrated scrutiny’ and has been a vocal critic of structured proportionality, as has Gordon J, the Court’s current senior puisne Justice. Though Edelman, Gleeson, and Steward JJ have each endorsed and applied structured proportionality in implied freedom cases, such as LibertyWorks Inc v Commonwealth (LibertyWorks), only Edelman J has defended the practice at meaningful length (see, eg, LibertyWorks at [199]–[202]), and Steward J’s application of the practice has been subject to his Honour’s view that the implied freedom does not even exist (LibertyWorks at [249]). Neither of the Court’s two newest Justices, Jagot and Beech-Jones JJ, have had the opportunity to express a view either way on the bench, leaving structured proportionality’s future in Australian jurisprudence uncertain. All that said, structured proportionality remains applicable doctrine for the time being and, as Professor Stone has observed, it is not evident in any event that a different formulation of the ‘reasonably appropriate and adapted’ limb would necessarily lead to cases being decided differently in practice (at 148).
Those matters aside, let us turn to assess the proposed scheme through the Brown analysis.
Step 1—Burden
The proposed expenditure limits would certainly impose a burden upon political communication for the purposes of Step 1 of the Brown analysis. However, this is not because of the inability of potential donors to express their political views through the medium of donations. McCloy is instructive here: there it was argued by the plaintiff, a wealthy property developer, that a legislative cap on donations by an individual or other entity of $5,000 burdened his ability to ‘build and assert political power’, which he argued was an aspect of the implied freedom (McCloy at [25]). This was unanimously rejected. French CJ, Kiefel, Bell, and Keane JJ held that no part of the freedom involves a right for individuals to make substantial donations to parties and candidates per se, and that such restrictions are in fact conducive to the equal distribution of political power (McCloy at [28]). Referring to Unions (No 1), their Honours made it clear—in reasoning later applied in Unions NSW v New South Wales (2019) 264 CLR 595 (Unions (No 2)) and Unions NSW v New South Wales (2023) 277 CLR 627 (Unions (No 3))—that the vice of donation caps is in their second-order effect upon the wherewithal of parties and candidates to fund public communication (McCloy at [24]). As six of the Justices noted (McCloy at [35] (French CJ, Kiefel, Bell and Keane JJ), [181] (Gageler J), [344] (Gordon J)), this position is to be contrasted with that of the United States Supreme Court in the landmark Citizens United v Federal Election Commission, 558 US 310 (2010), wherein it was held that ‘an independent expenditure is political speech’ (at 360 (Kennedy J for the Court)).
As such, the burden to be identified is that upon the practical and relative ability of parties and candidates to engage in political communication under the proposed scheme. As a proposal which seeks to limit the degree to which parties and candidates can fund their political communication, the Draft Bill clearly does burden political communication. That is not all, though: as Table 3 indicates, beyond the general burden of the expenditure caps, the Draft Bill imposes a further burden upon non-incumbents by restricting the amount of pre-election funding they can receive compared to incumbents. This differential burden will factor significantly into Step 3 of the Brown analysis.
Step 2—Legitimate end
It may be assumed, as a majority of the High Court did in Unions (No 2) (at [35]–[38], [146]), that the provisions implementing the scheme pursue the same legitimate end as the Act overall. Here, that would relevantly be the ‘[enhancement of] public confidence in the electoral process’ (EA s 130B(a)), given that the Draft Bill’s Explanatory Guide cites public concern about the influence of wealthy donors upon politicians’ decision-making (at 11).
Step 3—Reasonably appropriate and adapted
Under prevailing doctrine, the subsequent question of whether the statutory scheme is reasonably appropriate and adapted to achieve that purpose is to be assessed first by asking whether it is suitable (that is, actually capable of achieving the statutory purpose). The Draft Bill likely survives this stage of analysis, given that the High Court held in McCloy that blanket political donation caps under NSW law were not only compatible with our system of representative and responsible government, but enhanced it by ensuring the influence of wealthy donors did not impede ‘the practical enjoyment of popular sovereignty’ (at [45]–[47]). That reasoning applies a fortiori to the Draft Bill’s more severe restrictions.
But the Draft Bill likely runs into trouble at the latter two stages of proportionality testing, being the questions of whether the provisions implementing the scheme are necessary and adequate in the balance. Whether the proposed scheme is necessary, in the sense that ‘there is no obvious and compelling alternative, reasonably practicable means of achieving the same purpose which has a less restrictive effect on the freedom’ (McCloy at [2]) is likely to be a complex, fact-sensitive question with which the Court will need to grapple. That said, it would be an uphill battle for South Australia to argue that it cannot improve confidence in the electoral system without causing the party of government to have roughly 7.5 times as much available pre-election public funding as a non-incumbent party with the same number of candidates and subject to the same expenditure cap (see Table 3). Indeed, when asked about the Draft Bill’s favouring of large parties, the Premier’s only defence was ‘but that’s the sort of discrepancy that exists now’ before acknowledging that this is a matter that may need to be addressed in the consultation process.
Similar considerations apply if the statutory scheme makes it to the final assessment of adequacy in the balance, ‘a criterion requiring a value judgment, consistently within the limits of the judicial function, describing the balance between the importance of the purpose served by the restrictive measure and the extent of the restriction it imposes on the freedom’ (McCloy at [2]). Permeating the High Court’s decisions on NSW’s electoral laws in McCloy and Unions (No 2) is the importance of a relatively level field of public debate. It was that matter which led to the general donation caps in McCloy being found valid (at [45]–[47]) and which largely led to a discriminatory cap on political expenditure by trade unions being found invalid in Unions (No 2) (at [39]–[40], [101]). True though it is that the importance of equality in political voice weighs in favour of the validity of the proposed scheme to the extent that it places wealthy donors on an even level with electors of lesser means. But this cuts both ways: the Draft Bill equates electors with each other at the expense of equality between incumbent and non-incumbent candidates and parties. Indeed, as Professor Twomey points out, an attempt to apportion shares in political messaging according to parties’ and candidates’ existing positions in the legislature, at clear detriment to non-incumbents, was the very kind of law which was struck down in the foundational case of ACTV. Mason CJ said in that case, regarding a legislative attempt to restrict non-incumbents and independents to only 10% of available political broadcasting time (at 146 (emphasis added)), that:
The replacement regime, which rests substantially on the provisions relating to the grant of free time, is weighted in favour of the established political parties represented in the legislature immediately before the election and the candidates of those parties; it discriminates against new and independent candidates. By limiting their access to a maximum of 10 per cent of the free time available for allocation, Pt IID denies them meaningful access on a non-discriminatory basis…. The consequence is that the severe restriction of freedom of communication plainly fails to preserve or enhance fair access to the mode of communication which is the subject of the restriction. The replacement regime, though it reduces the expenses of political campaigning and the risks of trivialisation of political debate, does not introduce a ‘level playing field’. It is discriminatory in the respects already mentioned. In this respect I do not accept that, because absolute equality in the sharing of free time is unattainable, the inequalities inherent in the regime introduced by Pt IIID are justified or legitimate.
It is not clear on what basis the discriminatory scheme in ACTV could be meaningfully distinguished from that under the Draft Bill, particularly given that the High Court held in McCloy and the Unions cases that restrictions on campaign resources are as apt to restrict political communication as are direct, immediate restrictions on communication. On this basis, it is not at all obvious that in any balancing exercise the High Court would find the scheme’s exchange of equality between candidates for equality between electors to be adequate.
Conclusion
The proposed reforms would change the landscape of South Australian electoral politics. They would ban the making of political donations to incumbent parties and candidates, but not non-incumbents. At the same time, they would open the State’s coffers to those incumbents much more generously than to non-incumbents. In doing so, the proposed reforms run the real risk of being struck down. At a time when trust in liberal democratic institutions is faltering globally, there is an obvious need to build and earn that trust, as South Australia seeks to do—but there is an equal need to do it right.
With the outcome of the Draft Bill’s consultation process still to come, there is still much time for the lopsided favouring of incumbents under the public funding scheme to be addressed. Whether this happens will no doubt require deep consideration of the High Court’s attitude to electoral finance laws to date, and of the public purse’s ability to fund a truly level playing field. If the outcome, however, is that non-incumbents will indeed have to campaign with a comparatively meagre war chest, the Malinauskas government’s long-awaited reforms could well meet their end in Courtroom 1 at Parkes Place.
Sebastian Long graduated from the University of Sydney with a Bachelor of Arts and Bachelor of Laws (Hons II). He is currently undertaking a sabbatical before commencing as a graduate at HWL Ebsworth Lawyers in 2025.
Suggested citation: Sebastian Long, ‘Could South Australia’s proposed ban on political donations run into trouble in the courts?’ (21 November 2024) <https://www.auspublaw.org/blog/2024/11/could-south-australias-proposed-ban-on-political-donations-run-into-trouble-in-the-courts>