The four NSW by-elections for Bega, Monaro, Strathfield and Willoughby are being conducted under rules where every voter is automatically being sent ballot papers in a postal vote pack.
You can find my guide to the by-elections at the ABC Elections site. Each page now includes candidate how-to-vote material.
In this post I will keep track of the number of pre-poll votes cast and the number of postal votes returned before polling day. As I explain inside this post, the rate of postal voting is certain to be very high given this automatic send out of postal vote packs. Voters can still vote pre-poll or vote on polling day, but many are certain to use the ballot papers sent to them.
Observation – In the final week of voting you would normally see a surge in pre-poll voting. That hasn’t happened with these by-elections. Pre-poll voting increased each day in week one of voting, but there has been no increase in the per day rate in week two. That voters were receiving postal vote packs last week has almost certainly caused some voters to use their postal vote rather than attend pre-polling. As polling day nears, the number of postals returned has begun to surge.
The rates of pre-poll and postal voting by Friday 11 February on the completion of pre-poll voting:
Bega – 29.4% of enrolled voters have cast a pre-poll and 21.3% returned a postal vote
Monaro – 23.1% of enrolled voters have cast a pre-poll and 15.3% returned a postal vote
Strathfield – 16.3% of enrolled voters have cast a pre-poll and 26.4% returned a postal vote
Willoughby – 9.6% of enrolled voters have cast a pre-poll and 28.1% returned a postal vote
Inside this post I breakdown the pre-polls and postals by electorate by day and compare them to the equivalent rates in 2019. I also explain the rules under which the election is being conducted, and also the changed counting procedures for the by-election.
Update on Counting Procedures – Postal votes envelopes will be processed in the week after the election, but there will be no counting of postal votes until Saturday 19 February.
It is important to stress that this all postal election is not available for the Federal election or the looming South Australian election. Both those elections will allow more voters to apply for postal votes, but full postal mail-outs aren’t an option for either election.
Comment on the Daily Rate of Pre-Poll Voting
The normal pattern of pre-poll voting is that it begins slowly and accelerates as polling day approached. The Chart below shows the data for the 2020 Eden-Monaro by-election. Eden-Monaro covers two of the state seats where there will be by-elections this Saturday, Bega and Monaro.
It is a very different pattern for the four by-elections. The chart below plots the by day rate of pre-polls taken in the four electorates.
There are two weeks of pre-polling compared to three for Eden-Monaro, but the difference in pattern is clear. Pre-polling per day for the by-elections increased through the first week but has stopped increasing in week two, though a slight uptick on the final day of pre-poll voting. Voters would have been receiving their postal vote packs at the end of last week, and it appears many voters have decided there is no rush to pre-poll.
| Pre-Poll Votes | Postals Returned | |||||
|---|---|---|---|---|---|---|
| Day | Daily | Total | Pct | Daily | Total | Pct |
| Mon 31 Jan | 1,102 | 1,102 | 1.9 | 0 | 0 | 0.0 |
| Tu 1 Feb | 1,340 | 2,442 | 4.1 | 0 | 0 | 0.0 |
| Wed 2 Feb | 2,016 | 4,458 | 7.6 | 0 | 0 | 0.0 |
| Th 3 Feb | 2,077 | 6,535 | 11.1 | 0 | 0 | 0.0 |
| Fri 4 Feb | 2,132 | 8,667 | 14.7 | 0 | 0 | 0.0 |
| Sat 5 Feb | 1,272 | 9,939 | 16.9 | 0 | 0 | 0.0 |
| Mon 7 Feb | 1,381 | 11,320 | 19.2 | 4 | 4 | 0.0 |
| Tu 8 Feb | 1,329 | 12,649 | 21.5 | 15 | 19 | 0.0 |
| Wed 9 Feb | 1,230 | 13,879 | 23.6 | 2,897 | 2,916 | 5.0 |
| Th 10 Feb | 1,309 | 15,188 | 25.8 | 6,526 | 9,442 | 16.0 |
| Fri 11 Feb | 2,154 | 17,342 | 29.4 | 3,115 | 12,557 | 21.3 |
2019 Totals
19,516 pre-poll votes taken representing 33.6% of enrolment
1,444 postal votes and 1,620 iVotes together representing 5.3% of enrolment
| Pre-Poll Votes | Postals Returned | |||||
|---|---|---|---|---|---|---|
| Day | Daily | Total | Pct | Daily | Total | Pct |
| Mon 31 Jan | 495 | 495 | 0.9 | 0 | 0 | 0.0 |
| Tu 1 Feb | 755 | 1,250 | 2.2 | 0 | 0 | 0.0 |
| Wed 2 Feb | 996 | 2,246 | 3.9 | 0 | 0 | 0.0 |
| Th 3 Feb | 1,248 | 3,494 | 6.0 | 0 | 0 | 0.0 |
| Fri 4 Feb | 1,359 | 4,853 | 8.4 | 0 | 0 | 0.0 |
| Sat 5 Feb | 1,298 | 6,151 | 10.6 | 0 | 0 | 0.0 |
| Mon 7 Feb | 1,389 | 7,540 | 13.0 | 0 | 0 | 0.0 |
| Tu 8 Feb | 1,305 | 8,845 | 15.2 | 0 | 0 | 0.0 |
| Wed 9 Feb | 1,220 | 10,065 | 17.3 | 1,832 | 1,832 | 3.2 |
| Th 10 Feb | 1,319 | 11,384 | 19.6 | 3,363 | 5,195 | 8.9 |
| Fri 11 Feb | 2,040 | 13,424 | 23.1 | 3,672 | 8,867 | 15.3 |
2019 Totals
15,641 pre-poll votes taken representing 26.9% of enrolment
1,675 postal votes and 2,926 iVotes together representing 7.9% of enrolment
| Pre-Poll Votes | Postals Returned | |||||
|---|---|---|---|---|---|---|
| Day | Daily | Total | Pct | Daily | Total | Pct |
| Mon 31 Jan | 333 | 333 | 0.6 | 0 | 0 | 0.0 |
| Tu 1 Feb | 418 | 751 | 1.4 | 0 | 0 | 0.0 |
| Wed 2 Feb | 651 | 1,402 | 2.5 | 0 | 0 | 0.0 |
| Th 3 Feb | 948 | 2,350 | 4.3 | 0 | 0 | 0.0 |
| Fri 4 Feb | 935 | 3,285 | 5.9 | 0 | 0 | 0.0 |
| Sat 5 Feb | 1,200 | 4,485 | 8.1 | 0 | 0 | 0.0 |
| Mon 7 Feb | 857 | 5,342 | 9.7 | 82 | 82 | 0.1 |
| Tu 8 Feb | 824 | 6,166 | 11.2 | 648 | 730 | 1.3 |
| Wed 9 Feb | 758 | 6,924 | 12.5 | 2,159 | 2,889 | 5.2 |
| Th 10 Feb | 841 | 7,765 | 14.0 | 7,877 | 10,766 | 19.5 |
| Fri 11 Feb | 1,228 | 8,993 | 16.3 | 3,848 | 14,614 | 26.4 |
2019 Totals
6,660 pre-poll votes taken representing 11.5% of enrolment
1,224 postal votes and 3,567 iVotes together representing 8.2% of enrolment
| Pre-Poll Votes | Postals Returned | |||||
|---|---|---|---|---|---|---|
| Day | Daily | Total | Pct | Daily | Total | Pct |
| Mon 31 Jan | 209 | 209 | 0.4 | 0 | 0 | 0.0 |
| Tu 1 Feb | 249 | 458 | 0.8 | 0 | 0 | 0.0 |
| Wed 2 Feb | 306 | 764 | 1.4 | 0 | 0 | 0.0 |
| Th 3 Feb | 518 | 1,282 | 2.3 | 0 | 0 | 0.0 |
| Fri 4 Feb | 518 | 1,800 | 3.3 | 0 | 0 | 0.0 |
| Sat 5 Feb | 623 | 2,423 | 4.4 | 0 | 0 | 0.0 |
| Mon 7 Feb | 548 | 2,971 | 5.4 | 1 | 1 | 0.0 |
| Tu 8 Feb | 527 | 3,498 | 6.4 | 51 | 52 | 0.1 |
| Wed 9 Feb | 551 | 4,049 | 7.4 | 2,993 | 3,045 | 5.5 |
| Th 10 Feb | 535 | 4,584 | 8.3 | 8,957 | 12,002 | 21.8 |
| Fri 11 Feb | 690 | 5,274 | 9.6 | 3,479 | 15,481 | 28.1 |
2019 Totals
5,205 pre-poll votes taken representing 9.0% of enrolment
2,269 postal votes and 3,616 iVotes together representing 10.1% of enrolment
Election Rules
At the start of Covid, the NSW Electoral Act was amended allowing by-elections to be conducted based on regulations that could suspend parts of the Electoral Act and apply altered rules. These regulations could be used if the electoral district was declared to be impacted by Covid.
Such a declaration has been made for these four by-elections. The regulations put in place will see all voters sent ballot papers in a postal vote pack. The period for return of postal votes after polling day has also been extended from 3 days to 10 days. The draw for ballot position took place on Friday 28 January and postal vote backs are currently being dispatched.
Pre-poll and polling day voting is still available, but all voters will be sent a postal vote as well. The change this make to counting will be explained below.
Under normal rules, only voters on the permanent postal vote register would automatically be sent postal votes. All other voters would first have to apply for a postal vote.
NSW elections are conducted with only two weeks between close of nominations and polling day, the shortest period in the country. Postal votes normally have to be returned by the Wednesday after polling day. Under these rules, remote voters struggle to have their postal vote returned in time. Postal voting from overseas is a waste of time as almost no votes can be returned in time.
This is why the NSW Electoral Commission has developed iVote, as a replacement for postal voting.
iVote has been used at all state elections and by-elections since 2011. Last December it was used at the Local Government elections for the first time. This caused a big increase in usage, the increase magnified when the government agreed to allow on the day Absent votes to be cast using iVote. There is normally no absent voting at local government elections.
Allowing Absent voters to use iVote caused a huge surge in the iVote registration, verification and voting on election day, a huge load compared to the load planned in iVote's design. iVote was meant as a replacement for postal voting where this load was spread over the campaign, not all on polling day.
iVote ran into serious problems as a result and the Electoral Commissioner decided that it could not be re-configured to be used for the by-elections.
With the limited postal voting period, and without iVote, it has been decided to switch to the full postal mail-out option while still allowing in-person voting.
Changes to the Count
Conducting elections where every voter is sent a postal votes creates problems with the usual rules for scrutinising returned postal votes. Under normal election rules, postal vote applications are checked against the roll on receipt before ballot papers are dispatched. When the postal vote envelopes are returned, the details on outside are checked against the original application. Related rules are applied for postal votes returned from those on the permanent postal roll.
With no application to check against, different rules are being applied for the count at these by-elections.
The biggest concern is stolen postal ballot papers and multiple voting. To get around this, the usual rules on pre-processing postal votes on return have been suspended. Instead, postal vote processing will not begin until all the hard copy electoral rolls from pre-poll and polling day have been scanned.
No postal vote will be admitted to the count if the returned postal envelope is from a person marked as having voted in person. An in person vote voids any postal vote cast in that person's name.
To allow the roll scanning, postal vote scrutiny will not begin until the Wednesday after polling day, Wednesday 16 February. All postal envelopes will be checked against the roll mark-off data, and if the person hasn't voted, the vote will be admitted to the count.
Initial postal vote scrutiny will begin on Monday 14 February, limited initially to obvious errors with the envelopes that will prevent them being admitted to the count. Once the in-person voting election rolls are scanned, scrutiny checking postal votes against in-person turnout will begin. Counting of the ballot papers from the postal votes will not begin until Saturday 19 February, the week after the election.
Depending on how many votes are cast in person, it means we may not know the result on election night 12 February.
After voting pre-poll the other day, I was discussing the likely increase in double voting at this election. We were informed by the election officer that we would receive one, and that we should destroy it, so this should help. But we haven’t received ours yet, so there could be quite a gap between advice and arrival, so increased opportunity for errors.
COMMENT: As I explain in the post, counting of postals won’t start until the Wednesday after polling day to allow the rolls to be scanned. Any postal vote for someone marked as having voted will be rejected.
Off topic, but given the leadership speculation within the federal Liberal Party at the moment, could Scott Morrison call an early election to head off a leadership challenge? One of my friends who is an avid watcher of all things political claims that the Governor General would refuse to allow the writs of the election under this scenario. Is that correct?
COMMENT: This was suggested as a course of action by Malcolm Turnbull in 2018 and that didn’t eventuate. The request would be for a dissolution and an election. There may be questions on whether supply is necessary which is relevant at the moment. Having to arrange Senate writs by the states would delay the dissolution. There are enough ways to delay the process to give politics the chance to resolve the problem.
Boring – and not even informative!
I found the presentation was a bit dry (similar articles relating to pre poll vote rates have been shown as graphs rather than tables). However, it is quite informative as it shows how many people use pre poll or postal compared to the previous general election.
COMMENT: I’ve now added graphs. It was a bit difficult to show a graph until I had a few days of data. The point of the post was also to explain the change in counting method which hasn’t been explained anywhere else.
Great précis, Antony. Thanks, Alan
I always vote postal. If someone goes to a voting centre on election day and votes under my name, will I be notified that a duplicate vote has been recorded, or do they just discard the postal and not notify me.
COMMENT: It depends on the relevant electoral act. Some jurisdictions reject the vote, others don’t. The rules for these by-elections are different because the opportunity to steal postal votes is so high. Normally the details are checked against the postal application as a check of validity. Pre-poll votes are always counted where there are several reasons why Postal votes won’t make it into the count.