The High Court has published its reasons for dismissing an application challenging the Australian Electoral Commission’s practice of publishing information about indicative two-candidate preferred counts for divisions of the House of Representatives. The Electoral Act 1918 (Cth) requires the scrutiny of votes in divisions to include an Indicative TCP Count. This count takes place after first preference votes are counted, and it is a ‘count … that, in the opinion of the Australian Electoral Officer, will best provide an indication of the candidate most likely to be elected for the Division’. The process involves the AEC revealing one or two candidates selected by the Commission as the TCP Candidates (the two most likely to be elected), and the progressive results of the indicative counts leading to the final count are called the TCP Information.

The plaintiffs were candidates for the United Australia Party for House and Senate seats in the 18 May 2019 election. In April 2019, the plaintiffs filed an application for a constitutional writ challenging the AEC’s practice of releasing Indicative TCP counts for divisions where polls had closed and the counts had begun while other the polls for other divisions still remained open. The plaintiffs contended that the AEC could not do so for two reasons: first, publishing that information before the close of all polls was not authorised by the Electoral Act, or in the alternative, publishing TCP counts will polls were open in other seats would ‘impermissibly distort the voting system’ and ‘compromise the representative nature of a future Parliament’ contrary to the constitutional requirement for direct and popular choice in ss 7 and 24 of the Constitution.

On 7 May 2019, after the conclusion of the Full Court hearing, the High Court dismissed the application. On 14 August 2019 the Court published its reasons for that decision, which was unanimous.

The joint judges (Kiefel CJ, Bell, Keane, Nettle, Gordon and Edelman JJ) held that publishing TCP information while other polls were still open is authorised by s 7(3) of the Electoral Act. Their Honours also rejected the factual foundations for the application, in particular rejecting the applicants’ claims that the TCP information gave the AEC’s ‘imprimatur’ to one candidate or outcome, and that publishing TCP information affected the requirements of direct and popular choice in the Constitution (at [6]). The joint judges noted three difficulties with the plaintiffs’ case (at [7]). First, there were no facts showing that publishing TCP information affected the decisions of voters in Divisions that were still open. Second, the plaintiffs did not contend that the AEC could not publish the results of first preference counts after a division had closed but before all other divisions had closed. Third, the Indicative TCP Count predicted the candidate most likely to be elected, and was not an expression of the AEC’s view about a desirable or undesirable outcome.

After laying out the constitutional and Electoral Act provisions on elections, the AEC’s role and s 7(3), which provides that the AEC ‘may do all things necessary or convenient to be done for or in connection with the performance of its functions’ (at [8]ff), the joint judges described the AEC’s established practice on TCP counts (at [16]ff) and their entrenchment in the Electoral Act (at [27]ff). Turning then to the factual foundation for the plaintiffs’ challenge, the joint judges noted four issues. First, the plaintiffs had not demonstrated any effect that Indicative TCP Counts had on electoral choices in other divisions: the articles they submitted on the ‘bandwagon effect’ of possible influences on later voters by earlier results in other districts were not studies of Australia but rather of the US, France and Denmark: those systems differ markedly from Australia’s, the studies did not examine any actions similar to the publication of TCP counts, and indeed one study suggested that no effects on turnout was identifiable in parliamentary (as opposed to presidential) elections (at [31]–[36]). Second, the plaintiffs’ claim that voting ‘may be affected’ by the release of TCP counts was not just unfounded, but the plaintiffs had also conceded in the agreed facts that a range of other possible vote influencing factors existed (from opinion polls to exit polls to how-to-vote cards, among others), and the plaintiffs did not clarify how, if at all, the publication of TCP counts might interact with these factors (at [37]). Third, there was no factual basis for the plaintiffs’ claim that the AEC’s processes for identifying the TCP Candidates were inaccurate or misleading (at [38]–[39]). Fourth, the AEC’s publication of TCP Candidates did not involve the AEC giving some ‘imprimatur’ or endorsement to any candidate or outcome, but rather simply involved a prediction about the likely outcome of a vote count (at [40]–[41]).

Having rejected the factual basis of the plaintiffs’ statutory challenge to the Indicative TCP Count process, the joint judges turned to the argument that publishing TCP Information prior to the close of all polls is not authorised by the Electoral Act. Noting that the Indicative TCP Count  is ‘a predictive statutory exercise that will best provide an indication of the candidate most likely to be elected’, the joint judges held that this was authorised by the broad powers in s 7(3) that give the AEC the power to do all things necessary or convenient to be done for or in connection with its functions of promoting public awareness of elections and publish material on election results (at [43]–[44]). Nothing in the statutory framework limits the AEC’s powers to publish the TCP Information once polls close in a division, and it is open to the Commission to decide that publishing it in this way is necessary and convenient for promoting public awareness of elections and publishing material related to results (at [47]ff). Given the AEC holds that power, the question of whether it is preferable for the power to be exercised or the manner of that exercise is not a matter for the High Court, and while questions around the limits of the power could arise in future, they do not arise here (at [50]–[51]).

Turning finally to the constitutional challenge, the joint judges reiterated that the points on the lack of a factual basis for the plaintiffs’ claims was sufficient to reject their points on the constitutionality of the AEC’s practice (at [52]). There is no factual basis for the distortion contrary to ss 7 and 24 claimed (at [53]), and indeed the parties agreed that there was no practicable means, before the May 2019 election, of assessing the extent or likelihood of the effect, if any, on voters’ electoral choices if they knew the identity of TCP Candidates (at [54]). Finally, again, the publication of the Indicative TCP Count and the TCP Information did not constitute the AEC’s endorsement of a candidate, and no constitutional question of whether a government agency could support particular candidates does not arise (at [55]).

Gageler J agreed with the orders and much of the reasoning of the joint judges, and added his own specific reasons for rejecting the plaintiffs’ contentions. Gageler J noted that the plaintiffs’ arguments that publishing the TCP Information prior to the national close of polls was beyond the power in s 7(3) involved a number of propositions that he thought sound: that the power, if it existed, could only be found in s 7(3); that it could not be used to publish information that would not be ‘necessary or convenient to be done for or in connection with the performance of its functions’; and that a publication whose timing led it to find favour, or create the appearance of favouring, one candidate would be beyond power (at [59]). For Gageler J, the purpose of publishing the Indicative TCP Count as required by s 274 was to provide an early indication of likely election results, and to achieve that purpose the information had to be published before the actual result was known (at [60]). The basis of that power was the express conferral in s 7(3) and the functions of the AEC noted in s 7(1) (at [61]ff). Publishing the TCP Information was needed to give efficacy to the Indicative TCP Count, and thus it was supported by s 7(3) as a thing necessary to be done by the Commission in connection with the performance of its functions (at [63]). In Gageler J’s view, the AEC had the power to publish the TCP Information, regardless of whether that publication related to the specific functions in s 7(1) of promoting public awareness of election matters or publishing material related to the AEC’s functions (at [63]). While the timing needed to be before the final count, the AEC does not have unconstrained discretion to decide on the timing of the earliest publication of the TCP Information (at [64]). While things ‘necessary or convenient’ allow considerable latitude, the AEC could not publish the information at time that would be likely to have favoured, or create the appearance of favouring, particular candidates (at [65]–[67]). Here, however, that the publication might have had that effect on the plaintiffs was neither self-evident nor supported by the facts or the academic studies raised by the plaintiffs (at [67]). The plaintiffs’ defect here, for Gageler J, was not in the legal argument but in the lack of a factual foundation (at [68]).