BRADFIELD (NSW, IND, 0.01)
Disputed Returns challenge to win by Nicolette Boele over Gisele Kapterian (Lib) by 26 votes
This is an ongoing thread to follow the challenge by Gisele Kapterian (Liberal) to Nicolette Boele's (IND) victory by 26 votes in the seat of Bradfield. This is the first CDR challenge by a candidate to be based on the count, rather than eligibility or alleged misconduct issues, since Fran Bailey's (Liberal) win in McEwen 2007 was challenged by now MP for the seat Rob Mitchell (ALP). On comments about the case available so far the case is likely to be very similar to McEwen in proceedings.
I previously covered the main part of the Bradfield postcount in a general teal seat postcount thread and the late postcount (distribution of preferences phase) and recount in a Bradfield specific thread. I also covered the Goldstein partial recount and count history on a separate thread that may be useful for comparisons.
Boele had been down for the count at times in the initial count but got back up again, mainly on a very strong batch of out of electorate prepolls followed by the very last batch of postals breaking strongly to her (not such an unusual thing). She went into the distribution of preferences 40 votes ahead but dropped back through the distribution to finish it 8 votes behind. This is not surprising because in the distribution the votes that get the most attention are votes for minor candidates, so the candidate more reliant on preferences is more likely to suffer if votes are found to be informal on further scrutiny. The recount has the opposite dynamic - at this stage the primary votes of the leaders have been less checked than the preferences, and so this can hurt the primary vote leader. Which it did. Boele started very slowly in the recount and was still three votes behind and running out of booths when there was a big shift to her in the Turramurra PPVC, which was the last big prepoll to be recounted. Boele also then made gains in other booths in the St Ives and Turramurra areas and finished 26 votes ahead.
Unlike the Goldstein postcount which saw several large corrections and errors before the result was eventually established, the Bradfield postcount and recount was about as smooth to these outside eyes as these things get. The corrections were generally small and on the scale of what is generally expected - changes of rulings on specific votes, very minor counting corrections and just one correction of slightly larger size (a 22 vote correction in Kapterian's favour). The 15-vote swing to Kapterian in Turramurra PPVC was on formality rulings, not a counting error.
The Liberals floated a possible challenge on June 8 and I noted that none of the claimed grounds were convincing. Claimed reasons for concern were that the distribution of preferences and the recount had different winners (not surprising for the reasons stated above), that the number of informal votes increased (not surprising because sequence errors can easily be missed in the earlier counting stages) and that the number of total votes increased (not surprising because the AEC does not know for sure at the start of the count exactly how many ballot papers there are and some slight movement in this total is normal).
There was not until now any public suggestion that the Liberals had issues with AEC interpretations of specific votes. While there will always be some lineball votes that the losing side might object to, there has also not been any suggestion that there are systematic errors. The Liberals will have to argue that there are a number of errors sufficient to overturn the margin so it will be interesting to see what those arguments are. If there were persistent patterns of suspect rulings I would expect these to have come to notice by now and the lack of such seems unpromising for their chances of getting enough votes overturned.
I have not yet seen the petition and will comment on it if/when available but the media reports so far indicate that is wholly about ballot paper interpretation; indeed Kapterian has stated that the petition does not seek a by-election (as could be the case if the Liberal Party was arguing voters were deprived of the ability to vote, or voted who should not have done.) The framing of the Liberal Party's decision to lodge the challenge is that this is about giving their candidate every chance by sending the reserved ballots to the "third umpire".
The process
The 2008 McEwen case was referred by the High Court to the Federal Court and decided by a single judge; I expect this one will be so too, as it is a fact and evidence heavy matter involving the interpretation of electoral law, and not a constitutional matter.
Assuming that is so, the court will examine the reserved ballots (about 800 that were challenged and decided on by the Electoral Officer for NSW during the recount). Following this the court can make the following decisions:
* The result stands.
* The result is reversed and Kapterian wins. In this case Boele would lose her seat immediately and be replaced by Kapterian.
* The election is void. In this case the seat is vacated and a by-election is held with a fresh nominations process; both Boele and Kapterian would presumably run again. However, this would only occur if at the end of the process the court ruled the correct result was a tie, or perhaps so close to a tie that after taking multiple voting into account a winner could not be decided. (The number of unexplained multiple markoffs in Bradfield is understood to be just two).
The court can also modify the margin. This happened in the McEwen case twice with the court initially amending the margin from 12 to 27 votes then later giving a supplementary ruling that changed it to 31.
There may be procedural legal argument but I would expect that at some point the judge will end up examining all the reserved ballots and producing a table listing the results of the re-examination.
The court is obliged to decide the case as quickly as it reasonably can. In 2008 the Court took just over four months to dismiss the petition from its lodging. This would take us to close to the end of the year. It may be that this case can be faster if there is less preliminary argument than in 2008.
Mitchell v Bailey (2008 McEwen case)
The McEwen main judgement is well worth a read as background to this case for those interested; it is likely to be referred to frequently. Many votes had been ruled informal where there was a reasonable interpretation that allowed them to be ruled formal. For instance a ballot paper contains the numbers 1,2,3,4,6,7,8 and a figure that could plausibly in isolation be the letter S or the number 5. Intuitively it is overwhelmingly likely the voter intended to write a 5 and happens to write their 5 in a way that could also look like an S. Largely as a result of such issues the Court changed 141 ballots from informal to formal and only twelve from formal to informal. The number of votes for Mitchell that were fished out of the informal pile easily exceeded the margin, but the judge did not only examine the votes Labor objected to but examined all the reserved votes and found that Bailey had been more disadvantaged by incorrect formality calls than Mitchell.
(It is not clear from the judgement text what became of the infamous "V8 Supercar" vote on which the voter according to Labor's petition had numbered all the boxes, crossed out the names of the candidates and replaced them with the names of motor racing drivers.)
The rulings made in the McEwen case are very well known and are reinforced in AEC practice so I would expect that the chance of blatant errors here is a lot lower and that the chance of a margin shift even of the size of that in McEwen isn't high. But we will see.
Updates will be added as the case proceeds and a link will remain in the sidebar in the Upcoming and Recent Elections section.
Update 16 July
The ABC has reported some details of the petition, which I have not yet seen myself.
"The petition claims the electoral officer wrongly rejected at least 56 ballots which favoured Ms Kapterian.
This includes 22 ballots where the officer concluded certain numbers were not distinguishable from other numbers, and 34 ballots where numbers were deemed illegible."
Distinguishability depends on whether the Electoral Officer can confidently conclude that of two similar numbers, for instance, one is a 1 and one is a 7 and not the other way round. As concerns illegibility, for instance if a ballot is 1,2,(mysterious squiggle),4,5,6,7,8 it is not enough to assume that the mysterious squiggle is a 3 just for the sake of rendering the ballot formal. The mysterious squiggle must reasonably resemble a 3. Kapterian also claims 93 ballots favouring Boele were accepted that should have been rejected based on similar arguments (she alleges 49 with duplicate numbers and 44 cases with unclear numbering). It's highly unlikely that the AEC would have both been too lax on one candidate and too harsh on another.
"She argues a further two ballots favouring Ms Boele were admitted despite "having upon it a mark or writing … by which the voter could be identified.""
This depends on what the marks or writing are. If they are initials (for instance where a voter crosses a number out, rewrites it and puts their initials to confirm the change) then the McEwen case has plenty of precedent regarding this. The mere presence of initials does not identify the voter as there are likely to be many voters with any given combination. Something like a name and address may be deemed to identify a voter.
17 July
Anne Twomey's video here is a good watch. She mentions that in the McEwen case, although the candidates didn't object to all the reserved ballots between them in the case, the court nonetheless had to review all the ballots and invited submissions on six that neither side had objected to. The reason for this was the court needs to determine whether the result could have been different after making necessary corrections. Twomey also explains the term "illegal practices" that may be confusing in reading the McEwen case.
20 July
The petition is here. (Link fixed now 31/7, somehow nobody complained)
31 July
Documents (pay per view) have started appearing in the High Court portal, including directions submissions.
8 August
The petition - including the matter of what to do if the margin ends up being two votes or less (within the margin of possible double voting impacts) - has been
kicked down to the Federal Court as expected. A Federal Court docket has opened.
22 August
The
Guardian reports and the
court page confirms the Kapterian and Boele teams are being given 21 hours each to examine the 792 reserved ballots based on which they can make submissions. There will be a further day to examine them in mid-September after submissions are known. Further submissions will then be filed by 22 September, and then a document confirming agreements or otherwise between the parties will be filed by 25 Sep. A hearing will occur on 2 October.
25 September - It's Over, Prematurely!
Today we've suddenly had the news that Gisele Kapterian has
abandoned the challenge, meaning Boele has now conclusively won and will serve a full term.
I have to say I'm a little nonplussed about this turn of events. The fact that there was a challenge forced the AEC to issue an unsatisfactory 2PP estimate for this seat with various minor flow on effects for things such as national preference flows and polling flow estimates. In theory the Liberals' scrutineering could have been good enough to know they weren't in a winning position but somehow this was not the case until we got this far into the review of the disputed ballots. But by stopping now we miss a chance to get any fresh rulings to confirm that the AEC is correctly following the precedents of McEwen, or any new information on what the margin of litigation in such cases is. Nonetheless, it's understandable that once the Liberals realised they had no hope they would throw in the towel rather than keep everyone hanging even longer.
