Chapter 8 Formality issues
8.1
At the 2007 election more than 510,000 electors cast votes in the
House of Representatives elections were deemed to be informal by electoral
officials, thereby excluding them from the count. While this represents a
decline of around 95,000 informal votes compared to the 2004 election, the
disenfranchisement of hundreds of thousands of electors, effectively disqualifying
them from deciding who should represent them and form government is of
continuing concern.
8.2
This chapter examines the range of factors that contributed to
informality at the 2007 election for the House of Representatives and the
Senate and possible ways to reduce informality. Issues arising out of the Court
of Disputed Returns’ decision on the McEwen petition and the Australian
Electoral Commission’s (AEC) response to the decision along with the AEC’s subsequent
review of formality guidelines are also examined.
Background
8.3
As noted in chapter 2, the rate of informal voting in the House of
Representatives and Senate elections declined across all jurisdictions at the
2007 election compared to the 2004 election. This was the first decline in the
national rate of informal voting since the 1993 election.[1]
While the decline is significant, it is of note that there were still over
510,000 informal votes cast for the House of Representatives.[2]
8.4
The AEC’s analysis of informality following the 2007 election found that
there appear to be four significant influences that correlate strongly with
higher than average informality rates:
n divisions with high
candidate counts;
n differences in state
and federal electoral systems;
n proximity to other
election events; and
n a high proportion of
citizens from non-English speaking backgrounds.[3]
8.5
These four areas of influence on informality have been consistently
identified as strong predictors of informality by the AEC.[4]
8.6
A ballot paper may be ruled informal for a number of reasons including:
n the ballot paper is
not marked at all;
n the ballot paper does
not have the official mark and has not been initialled by the polling official
and the ballot paper is not authentic in the opinion of the Divisional
Returning Officer;
n the ballot paper has
writing on it which identifies the voter;
n in the case of an
absent, postal or provisional vote the ballot paper is not contained in the
declaration envelope; and
n the voter has not
marked a vote correctly for it to be considered acceptable in accordance with section
268 of the Commonwealth Electoral Act 1918.[5]
8.7
The AEC notes that Australia traditionally has one of the highest rates of
spoiled or informal ballots among established democracies, given the compulsory
enrolment and voting nature of this country’s electoral system.[6]
House of Representatives
8.8
At a national level, informal voting in House of Representatives
elections declined for the first time since the 1993 election. Changes in
informality at a national level from election to election have generally been
consistent across the states and territories, although some jurisdictions, most
notably, NSW and South Australia, have recorded higher informality for a number
of elections (table 8.1).
Table 8.1 Informal voting, House of Representatives, 1983
to 2007 elections (per cent)
State/ territory
|
1983
|
1984
|
1987
|
1990
|
1993
|
1996
|
1998
|
2001
|
2004
|
2007
|
NSW
|
2.2
|
5.7
|
4.6
|
3.1
|
3.1
|
3.6
|
4.0
|
5.4
|
6.1
|
5.0
|
VIC
|
2.2
|
7.5
|
5.3
|
3.5
|
2.8
|
2.9
|
3.5
|
4.0
|
4.1
|
3.3
|
QLD
|
1.3
|
4.5
|
3.4
|
2.2
|
2.6
|
2.6
|
3.3
|
4.8
|
5.2
|
3.6
|
WA
|
2.0
|
7.1
|
6.6
|
3.7
|
2.5
|
3.2
|
4.2
|
4.9
|
5.3
|
3.9
|
SA
|
2.0
|
7.1
|
6.6
|
3.7
|
4.1
|
4.1
|
4.5
|
5.5
|
5.6
|
3.8
|
TAS
|
2.3
|
5.9
|
5.0
|
3.3
|
2.7
|
2.4
|
3.1
|
3.4
|
3.6
|
2.9
|
ACT
|
2.2
|
4.7
|
3.5
|
3.0
|
3.4
|
2.8
|
2.9
|
3.5
|
3.4
|
2.3
|
NT
|
4.4
|
4.6
|
5.8
|
3.4
|
3.1
|
3.4
|
4.2
|
4.6
|
4.4
|
3.9
|
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, p 7.
8.9
There are a number of factors to consider when looking at the longer
term trend in informality at House of Representatives elections:
n the spike in
informality at the 1984 House of Representatives election is generally
attributed to the introduction of above-the-line voting in Senate elections;[7]
and
n between the 1984 and
1996 elections, ballot papers that were assumed to have been accidentally
marked non-consecutively for the House of Representatives (1,2,3,3,…) were
counted as formal votes. The ballot was accepted as formal and preferences
distributed up to the point where the mistake of numbering began. These ballot
papers then became ‘exhausted’.
8.10
Following each election, the AEC undertakes a survey of informal ballot
papers to identify the possible causes that influence informal voting at
federal elections. The largest proportion of informal votes at the 2007
election were those with a ‘1 only’ (30 per cent), followed by blank ballot
papers (20 per cent), non-sequential numbering (18 per cent), and
marks and scribbles (15 per cent) (table 8.2).
Table 8.2 Informality, by category and jurisdiction, 2007
election (% of total informal vote)
|
ACT
|
NSW
|
NT
|
Qld
|
SA
|
Tas
|
Vic
|
WA
|
Blank
|
25.8
|
18.2
|
15.0
|
15.4
|
26.9
|
29.3
|
22.3
|
23.5
|
1 only
|
25.9
|
36.2
|
24.7
|
36.4
|
24.3
|
17.3
|
21.6
|
18.0
|
Incomplete numbering
|
3.1
|
5.3
|
3.6
|
5.3
|
3.3
|
4.5
|
2.9
|
4.6
|
Tick or cross
|
10.2
|
11.0
|
15.2
|
9.4
|
12.8
|
7.2
|
8.1
|
8.3
|
Non sequential
|
9.9
|
15.8
|
24.4
|
15.2
|
15.9
|
15.0
|
21.7
|
26.3
|
Marks and scribbles
|
22.4
|
11.5
|
13.3
|
16.2
|
14.9
|
25.1
|
20.3
|
16.9
|
Voter identified
|
0.1
|
0.1
|
0.2
|
0.0
|
0.0
|
0.0
|
0.1
|
0.1
|
Other
|
2.6
|
1.9
|
3.6
|
2.2
|
1.9
|
1.6
|
2.9
|
2.3
|
Total
|
100.0
|
100.0
|
100.0
|
100.0
|
100.0
|
100.0
|
100.0
|
100.0
|
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, p 11.
8.11
The overall number of informal House of Representatives ballot papers
declined by over 95,000 at the 2007 election when compared to the 2004
election. The categories of informal voting that made the largest contribution
to the decline were the number of ‘1 only’ ballot papers (down by over 45,000),
the number of blank ballot papers (down by over 26,000) and the number of
ballot papers with marks and scribbles (down by almost 14,600).[8]
8.12
The 10 divisions with the highest overall informality in 2007 were all
in western Sydney (figure 8.1), with five of these divisions also having the
highest ‘poor English’ ranking, indicating that they have the lowest levels of
English proficiency out of all of the national divisions.[9]
The AEC noted that:
In fact the five divisions with the highest levels of
informality in 2007 also featured in the top ten for the 2004 and 2001
elections, indicating that any voter confusion over different systems is likely
to be extenuated by other factors. The persistence of this high informality in
certain divisions suggests other factors are also important, given state
elections in NSW do not all necessarily fall in close proximity to the federal
event.[10]
Figure 8.1 Divisions with the highest informality rates at
the 2007 election
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, p 13.
8.13
AEC research on informality at the 2007 House of Representatives election
confirmed previous results that attributed informality to the four factors of (1)
divisions with high candidate counts, (2) differences in state and federal
electoral systems, (3) the proximity to other election events and (4) the
proportion of citizens from non-English speaking backgrounds.[11]
The AEC noted that:
n divisions which have
a higher than average number of candidates are generally more likely to also
have higher than average rates of informality. Number of candidates was found
to have a strong correlation with informality, explaining approximately
one-quarter of change within informality rates. The decrease in informal voting
across the past two federal elections coincided with a decrease of the average
number of candidates per division (7.27 in 2004 to 6.66 in 2007);
n differences between
state and federal electoral systems has a significant impact on unintentional
informal voting. The evidence suggests that electors who can cast a valid
ballot with a ‘1 only’ preference in state elections (NSW and QLD) may be
conditioned to do the same for the federal House of Representatives ballot, not
knowing this is actually informal in the federal system;
n proximity to another
electoral event may increase the level of informality in a federal election.
There may be confusion for those voters coming from a state electoral system,
such as NSW, that has different criteria for casting a formal ballot. The
confusion surrounding how to cast a formal vote appears to contribute to
unintentional informality in some states; and
n it is possible that
linguistic and cultural barriers experienced by some NESB electors may amplify
problems associated with high candidate numbers and state/federal electoral
differences. This issue appears particularly relevant in NSW, which has a
comparatively high proportion of NESB electors. In 2007, five of the top six
divisions with the highest rates of informality, were also the top five
electorates with the highest proportion of NESB. The 14 divisions with the
highest rates of informality in 2007, all border each other in a concentration
based around western Sydney.[12]
8.14
Optional preferential voting is used for House of Assembly elections in
NSW and Queensland; with electors only required to indicate a single preference
(‘1’) on the ballot paper for a vote to be counted as formal. Lower house
elections in all other states and territories generally use full preferential
voting systems (box 8.1).
Box 8.1 Voting systems
by jurisdiction
New South Wales |
|
Legislative Assembly |
Optional allocation of preferences |
Legislative Council |
Optional allocation of preferences above the line and partial
preferential if voting below the line |
|
|
Victoria |
|
Legislative Assembly |
Full preferential voting |
Legislative Council |
Single preference when voting above the line (plus group
ticket voting) and full preferential when voting below the line |
|
|
Queensland |
|
Legislative Assembly |
Optional allocation of preferences |
|
|
Western Australia |
|
Legislative Assembly |
Partial preferential voting |
Legislative Council |
Single preference when voting to the left of the line (plus group ticket voting) and full preferential when voting to the
right of the line |
|
|
South Australia |
|
House of Assembly |
Full preferential voting |
Legislative Council |
Single preference when voting above the line (plus group ticket voting) and full preferential when voting below the line |
|
|
Tasmania |
|
House of Assembly |
Partial preferential voting |
Legislative Council |
Partial preferential voting |
|
|
ACT |
|
Legislative Assembly |
Partial preferential voting |
|
|
Northern Territory |
|
Legislative Assembly |
Full preferential voting |
Source Electoral Council of Australia,
‘Electoral Systems’, viewed on 18 May 2009 at http://www.eca.gov.au/systems/index.htm;
Bennett S and Lundie R, ‘Australian Electoral Systems’, Research paper no 5
2007-08 (2007), p 31.
8.15
The AEC noted that informality at Legislative Assembly elections in NSW and
for the Legislative Assembly in Queensland, were lower than for House of
Representatives elections in these states since 2003 (table 8.3).
Table 8.3 Comparison
of state and federal elections – overall informality rate, New South Wales and
Queensland, 2003 to 2009 (per cent)
Election
|
2003
|
2004
|
2006
|
2007
|
2008
|
2009
|
NSW Legislative Assembly election
|
2.62%
|
|
|
2.69%
|
|
|
NSW – House of Representatives federal election
|
|
6.1%
|
|
4.95%
|
|
|
QLD Legislative Assembly election
|
|
1.99%
|
2.08%
|
|
|
1.94%
|
QLD - House of Representatives federal election
|
|
5.2%
|
|
3.56%
|
|
|
Source Australian
Electoral Commission, submission 169.17, p 8.
8.16
While the average number of candidates per division declined in the 2007
House of Representatives elections, there were 10 divisions where
informality increased when compared to the 2004 election. In seven of these
divisions, the number of candidates increased from the previous election (table
8.4).
Table 8.4 Informality
rate and number of candidates in divisions where informality increased at the
2007 election
Division
|
Number of candidates
2007 election
|
Number of candidates
2004 election
|
Difference between 2007 and 2004 elections
|
Change in informality 2007 election compared to
2004 election (percentage points)
|
Murray
|
9
|
7
|
+2
|
+1.06
|
Richmond
|
7
|
8
|
-1
|
+0.67
|
Bendigo
|
9
|
5
|
+4
|
+0.67
|
Grayndler
|
7
|
5
|
+2
|
+0.56
|
Bennelong
|
13
|
7
|
+6
|
+0.38
|
Parkes
|
7
|
5
|
+2
|
+0.25
|
New England
|
0
|
0
|
0
|
+0.11
|
Mallee
|
6
|
6
|
0
|
+0.10
|
Page
|
10
|
8
|
+2
|
+0.07
|
Riverina
|
5
|
4
|
+1
|
+0.05
|
Source Australian
Electoral Commission, ‘Virtual Tally Room, House of Representatives, House of
Representatives Results, Candidates’, viewed on 26 May 2009 at http://results.aec.gov.au/13745/Website/HouseCandidatesBySurname-13745-A.htm;
‘Virtual Tally Room, House of Representatives, Candidates’, viewed on
26 May 2009 at http://results.aec.gov.au/12246/results/HouseCandidatesBySurname-12246-A.htm.
8.17
The number of divisions with more than nine candidates at the 2007
election (8), was lower than that for the 2004 election (13), with the division
of Bennelong having the highest number of candidates at the 2007 election (13),
one less than the 14 candidates who stood for the division of Greenway at
the 2004 election.[13] The strong relationship
between the number of candidates and informality gives rise to risks that
‘ballot flooding’ (running large numbers of candidates to make the ballot paper
bigger and voting more difficult) can impact on election results by leading to
a significant increase in informality.
Senate
8.18
All states and territories recorded a decline in informality at the 2007
Senate elections (table 8.5). Informality in Senate elections at a national
level has been lower than House of Representative elections since the 1998 election.[14]
Table 8.5 Informal voting, Senate, by jurisdiction, 1996
to 2007 elections (per cent)
State/territory
|
1996
|
1998
|
2001
|
2004
|
2007
|
NSW
|
3.8
|
3.3
|
3.5
|
3.5
|
2.2
|
VIC
|
3.6
|
3.8
|
5.6
|
5.1
|
3.2
|
QLD
|
3.3
|
3.0
|
3.0
|
2.8
|
2.3
|
WA
|
3.5
|
2.7
|
3.6
|
3.5
|
2.4
|
SA
|
3.3
|
2.8
|
3.1
|
3.5
|
2.4
|
TAS
|
3.2
|
3.1
|
3.3
|
3.4
|
2.6
|
ACT
|
2.5
|
2.0
|
2.3
|
2.5
|
1.7
|
NT
|
2.8
|
2.0
|
2.8
|
3.1
|
1.9
|
National average
|
3.5
|
3.2
|
3.9
|
3.8
|
2.5
|
Source Australian
Electoral Commission, Electoral pocketbook: election 2007 (2009), p 56.
8.19
The rate of informality in Senate elections is generally below that for
House of Representatives elections. However, the AEC noted that at least eight
divisions (Gorton, Maribyrnong, Bruce, Hotham, Batman, Scullin, Melbourne and
Calwell) reported a higher informality rate for the Senate than the House of
Representatives at the 2007 election.[15]
8.20
When ruling a Senate ballot paper informal, the categorisation of
whether the informality related to an attempt to vote above the line or below
the line is problematic.[16] The AEC noted that:
Each Senate ballot paper essentially contains two different
voting systems on the one ballot paper, which creates difficulties in clearly
separating informal above-the-line (ATL) votes and informal below-the-line (BTL)
votes. Some types of informal votes, such as blank ballot papers or a ballot
paper incorrectly marked both above and below the line cannot be inferred to be
informal ATL or informal BTL. Inferring intention in other types of informally
marked ballot papers is also problematic – is a ballot paper marked with only a
‘1’ next to the first candidate of a below the line grouping an informal ATL
vote or an informal BTL vote?[17]
8.21
The AEC did not conduct a survey of informality for Senate ballot papers
following the 2007 election. In 2001, a national informal ballot paper survey
was conducted, which found that over one-half of the informal votes cast in
South Australia’s 2001 half Senate election resist categorisation as either
informal above-the-line or below-the-line. (That is, the categories of ‘blank
ballot paper’, ‘writing, slogan, poetry’ and ‘other’) and another quarter of
informal votes could be interpreted as being either informal above the line or
informal below the line (ballot papers marked with a first preference only
below the line) (table 8.6).[18]
Table 8.6 Types of informal voting, 2001 South Australian
Senate election
Category
|
Number
|
Proportion
|
Blank Ballot Paper
|
10,375
|
34.0%
|
1st Pref Only marked below
the line
|
7,742
|
25.3%
|
Writing, slogans, poetry
|
2,854
|
9.3%
|
Less than 90 per cent of
boxes numbered below the line
|
2,690
|
8.8%
|
Other
|
2,623
|
8.6%
|
More than one number 1 above
the line
|
1,346
|
4.4%
|
More than one number 1 below
the line
|
845
|
2.8%
|
Large number of repeating
numbers or missing numbers below the line
|
830
|
2.7%
|
Ticks and crosses below the
line
|
617
|
2.0%
|
Combination of ticks and crosses
above the line
|
413
|
1.4%
|
Use of letters above the
line
|
154
|
0.5%
|
No 1st preference below the
line
|
67
|
0.2%
|
Total
|
30,556
|
100.0%
|
Source Australian
Electoral Commission, submission 169.6, p 2.
Efforts to reduce informality at the 2007 election
8.22
The AEC reported to the committee that strategies implemented in 2007 to
reduce informality included:
n the AEC analysed ABS
data at the polling place level to identify polling places with both high
informality at the 2004 election and high populations of ‘culturally and
linguistically diverse’ groups. On this basis, the AEC expanded its recruitment
drive to employ staff for selected polling places who could speak the targeted
language(s) for those communities. These polling officials were provided with a
badge indicating the language that they spoke (in 21 languages).
n some polling places
played a DVD of translated formality television advertisements, either in a
loop of all languages or in selected languages according to their elector
profile.
n the three questions
issuing officers are required to ask of electors were translated into 21
languages and made available for divisions to download and have available at
polling places as required.
n how to vote guides
(instructions on how to vote in English and translated into 21 languages in a
flipchart format) were supplied to each polling place, mobile polling team and
early voting centre.
n Polling staff in divisions
with the highest 2004 informality levels were provided with extra training.[19]
8.23
In addition to this specific work, the AEC noted that a range of
publications included information about how to vote. The AEC’s communication
strategy also included elements aimed at reducing informality.[20]
8.24
The committee noted that in relation to NSW, the AEC intended to continue
to research and analyse the informal voting figures from the 2007 election to
understand which mix of strategies may have had the greatest impact in working
to reduce informality levels at the 2007 election.[21]
Comments on informality by inquiry participants
8.25
Inquiry participants’ comments on informality at the 2007 election
related mainly to the need to continue with strategies to reduce informality
and examine the harmonisation of federal and state/territory voting
arrangements.
8.26
The Liberal Party of Australia welcomed the fall in informality at the
2007 election and commended the AEC on the consultative approach it took after
the 2004 election to ensure its advertising and information campaigns were
effective.[22] The Liberal Party of
Australia noted that:
We believe it is important the AEC continues to produce
advertising and information campaigns that target informality. In doing so we
encourage the AEC to again work in consultation with the parties to ensure the
most simple and effective campaigns are developed.[23]
8.27
Mr Eric Jones considered that greater resources should be directed to
reduce informality, including utilising part-time ‘political and youth
political officers’ who speak the major Non-English language in high informal
rate areas to conduct voting and general democratic educational programmes for
the year prior to an election.[24] Mr Jones supported such
an arrangement being delivered by the AEC or through funding provided to the
political parties:
Why should the political parties do this type of work? To be
seen to be putting something back into the democratic process. There is a lot
of cynicism about politicians and politics out in the broader community and
perhaps this might help in building up their position in the eyes of the
community.
However, if the above is not acceptable, at least try it on a
trial basis by funding and having the AEC employ such people for the year
before an election.[25]
8.28
University of Melbourne academic Dr Sally Young considered that various
strategies should be continued to reduce informality including:
Civics education, the AEC conducting education programs and
running advertisements on voting (including during non-election periods as a
way of better reaching voters before the sound and fury of an election campaign
begins and when many other political messages are then competing for their
attention), as well as writing to voters from non-English speaking backgrounds
in their own languages to advise them on voting procedures.[26]
8.29
Participants advocating that voting systems be harmonised across the
federal and state/territory elections considered that such an approach would
contribute to reducing informality, with some suggesting a move to optional
preferential voting in the House of Representatives.[27]
Mr William Bowe referred to research that indicated confusion between state and
territory systems contributed one percentage point to informality in those
jurisdictions where optional preferential voting was used at a state level.[28]
8.30
The Nationals supported harmonising voting systems as a means of
addressing informality, noting that:
The Nationals recommend the Committee work with the State and
Territory jurisdictions to get a common voting system nationwide. Compulsory
preferential voting has been the traditional voting system in Australia since
federation and is the system used in the majority of State jurisdictions. The
Nationals recommend it should be implemented in every State and Territory to reduce
voter confusion and informal voting.[29]
8.31
The New South Wales Government expressed its concern about the
continuing high proportion of informal votes in the state, noting that the 15
electorates with the highest rates of informal voting are all located in the
state.[30] The New South Wales
Government supported the adoption of optional preferential voting at a federal
level, noting that:
A significant factor is the difference in voting arrangements
between New South Wales, where preferential voting is optional, and the
Commonwealth, where preferential voting is compulsory. This leads to voter
confusion, resulting in a higher number of otherwise valid votes being
classified as informal.
I encourage the Committee to consider the benefits of
optional preferential voting in the interests of removing inconsistency between
Stale and Federal voting arrangements. State and Federal education programs
designed to improve voter understanding of the electoral process would no doubt
be more effective if consistent voting arrangements applied across all
jurisdictions.[31]
8.32
The Democratic Audit of Australia considered that the adoption of
optional preferential voting, which requires voters to only number ‘1’ for a
vote to be counted, should be considered for House of Representatives elections
and for voting below the line for Senate elections.[32]
8.33
As an alternative to optional preferential voting, the Democratic Audit
of Australia proposed the re-introduction of the ‘savings’ provision that was part
of voting arrangements between 1983 and 1998.[33]
Committee conclusion
8.34
The committee welcomes the reduction in informality recorded at the 2007
election compared to the 2004 elections. While the decline in the overall
informality rate is a positive outcome, the committee remains concerned about
the persistently high levels of informality recorded in some divisions,
particularly in south western Sydney.
8.35
Although harmonisation of voting systems appears to provide some
opportunity to reduce informality, the committee does not consider that it is
necessary to harmonise this aspect of electoral arrangements — decisions about
what voting system is appropriate for each jurisdiction should be left to each
respective parliament to determine.
8.36
The committee does not support a change to adopting optional preferential
voting for House of Representatives elections. However, the committee has
examined a range of savings provisions that could be adopted. These are
discussed later in this chapter.
8.37
With the drivers of higher informality generally well understood, it is
important that the AEC continue its efforts to address informality,
particularly in areas that consistently record relatively high levels of
informality.
Recommendation 34 |
8.38
|
The committee recommends that the Australian Electoral
Commission increase efforts to improve electors’ understanding of the federal
voting systems and take appropriate measures to reduce the rate of informal
voting, especially in electorates with a high percentage of electors from
non-English speaking backgrounds.
|
‘Saving’ informal votes
8.39
As pointed out in chapter 2, a set of protections or savings provisions
apply to ballot papers and operate to ensure that ballot papers are included in
the count as long as they express valid preferences. These protections are
known as formality provisions.
8.40
For Senate elections, an elector may vote by placing the single figure ‘1’
in one and only one of the squares above the line, or by placing consecutive
numbers beginning at 1 until all squares below the line are numbered.
8.41
A number of savings provisions ensure that where mistakes are made in
expressing preferences on Senate ballot papers the ballot papers may still be
regarded as formal. These include provisions that allow for:
n a first preference
mark to be indicated by the presence of the number ‘1’, or a tick, or a cross
in a square above the line;[34]
n less than complete
numbering below the line;[35] and
n a ballot paper which
is informal below the line to be counted if it is formal above the line.[36]
8.42
Similarly, a number of savings provisions ensure that where mistakes are
made in expressing preferences on House of Representatives ballot papers, those
ballot papers may still be regarded as formal. These include a provision that
allows for ballot papers with a first preference for one candidate and an order
of preference for all the remaining candidates except one to be regarded as
formal, with the blank square deemed to be the voter’s last preference, and
where there are only two candidates in a House of Representatives election and
the ballot paper contains a first preference for one candidate and the other
square is blank or contains a number other than two it may be deemed formal.[37]
8.43
Prior to the 1998 federal election, a further provision existed in
the Commonwealth Electoral Act that ‘saved’ votes where electors had made
numbering errors in marking their House of Representatives ballot paper,
keeping such votes in the count up to the point where the error had been made.
8.44
A different savings provision still exists in elections for the House of
Assembly in South Australia, serving to keep votes in the count where voters
mark only the single preference ‘1’ despite a full preferential voting system
being in place there.
Savings provision for the South Australian House of Assembly elections
8.45
Elections for the South Australian House of Assembly use compulsory
preferential voting, with a savings provision in place to include as many votes
as possible with incomplete or out of sequence preferences. Elections for the
South Australian Legislative Council are similar to the Senate voting
requirements, with voters able to make a single preference above the line or
complete all preferences below the line.
8.46
Under the voting system for the House of Assembly, all candidates in an
electorate have the right to register one or two ticket votes. On election day,
copies of each of the tickets registered in an electorate are displayed on the
voting screens of every polling place in that electorate. These provide a guide
to voters on how to fill in the sequence of further preferences for their first
choice candidate. A candidate’s how to vote material is required to match the
registered ticket vote and parties are banned from advocating a vote that does
not include preferences.[38]
8.47
The approach adopted for South Australian House of Assembly elections ensures
that votes marked with a single preference, which can include a single number
‘1’ as well as a tick or a cross, are included in the count, with preferences
beyond those preference expressed directed according to the registered tickets.[39]
8.48
The impact of this savings provision on the rate of informality for
South Australian House of Assembly elections is to markedly reduce the
informality rate, which would, at a minimum, be twice as high without the
savings provision (table 8.7).
Table 8.7 South
Australian House of Assembly informal and ticket voting, 1985 to 2006 state elections
(per cent)
|
1985
|
1989
|
1993
|
1997
|
2002
|
2006
|
Total informal votes
|
3.5
|
2.8
|
3.1
|
4.0
|
3.1
|
3.6
|
Accepted ticket votes
|
4.1
|
6.0
|
5.9
|
4.9
|
4.0
|
4.6
|
Source Green
A, submission 62.1, p 14.
8.49
The effect of the savings provision at the 2006 South Australian state
election was noted by Mr Antony Green, who told the committee that:
At the 2006 South Australian election, a total of 35,029
informal votes were recorded, a rate of 3.6 per cent, compared to 5.2 per
cent in the Legislative Council. Compared to Victoria, Western Australia, and
Commonwealth elections in every state, South Australia is the only state using
compulsory preferential voting where lower house informal voting is less than
upper house informal voting.
In total, 43,553 votes were admitted to the count after being
‘saved’ by the use of registered ticket votes. All of these votes would have
been informal under the Commonwealth Electoral Act. Under Commonwealth
formality rules, the South Australian lower house informal vote would have been
8.1 per cent, not 3.6 per cent.[40]
Savings provisions for non-sequential numbering in the Commonwealth
Electoral Act
8.50
As previously noted, between the 1984 and 1996 federal elections a
savings provision existed in the Commonwealth Electoral Act that meant that a House
of Representatives ballot paper would be deemed formal provided that it bore a
unique first preference, and numbers - any numbers, in all of the remaining
squares, or in all but one (with that last square left blank).[41]
The provision was part of section 270 of the Commonwealth Electoral Act
and applied only to House of Representatives ballot papers and read:
‘(2) Where a ballot-paper in a House of Representatives
election in which there are 3 or more candidates-
(a) has the number 1 in
the square opposite to the name of a candidate;
(b) has other numbers in
all the other squares opposite to the names of candidates or in all those other
squares except one square that is left blank; and
(c) but for this
subsection, would be informal by virtue of paragraph 133 (1) (c), then-
(d) the ballot-paper shall
not be informal by virtue of that paragraph;
(e) the number 1
shall be taken to express the voter's first preference;
(f) where
numbers in squares opposite to the names of candidates are in a sequence of
consecutive numbers commencing with the number 1- the voter shall be taken to
have expressed a preference by the other number, or to have expressed
preferences by the other numbers, in that sequence; and
(g) the voter shall not be
taken to have expressed any other preference.
(3) In considering, for the purposes of subsection (1) or
(2), whether numbers are in a sequence of consecutive numbers, any number that
is repeated shall be disregarded.’[42]
8.51
As noted by the AEC, the parliament recognised the possibility that
section 270(2) might appear to offer optional preferential voting for the House
of Representatives in contradiction to the requirement for full preferential
voting in section 240, and accordingly enacted section 329(3) to make it an
offence to distribute how-to-vote cards that might induce electors to vote
otherwise than in accordance with the instructions on the ballot paper.[43]
8.52
Commencing at the 1987 election, a series of campaigns were conducted by
several individuals advocating that electors not vote at all, cast an optional
preferential vote under then section 270(2), or that they vote informal. The
AEC noted that:
At the 1987 election, a campaign was run in Victoria advising
electors not to vote at all, or to cast an optional preferential vote under
then section 270(2), or to vote informal. The AEC sought injunctions against
the campaigners (Mr van Moorst and Mr Langer) on the basis of these three
campaign objectives. The Court awarded injunctions to prevent the campaigners
from advocating not voting at all, and to prevent them from inducing electors
to vote otherwise than in accordance with the instructions on the ballot paper.
However, the Court decided that as it was not unlawful to vote informal, it could
not be illegal to advocate informal voting.
At the 1990 election, the campaigners did not proceed with
their planned advocacy, so there was no need for the AEC to initiate court
proceedings. However, as a consequence of an indication that such campaigns may
be run in the future, the 1990 JSCEM recommended to Parliament a further
tightening of the penalties to protect the full preferential voting system.
Section 329A was enacted in 1992:
n ‘(1) A person must
not, during the relevant period in relation to a House of Representatives
election under this Act, print, publish or distribute, or cause, permit or
authorise to be printed, published or distributed, any matter or thing with the
intention of encouraging persons voting at the election to fill in a ballot
paper otherwise than in accordance with section 240.
n In this section:
'publish' includes publish by radio or television.’
At the 1993 election, Mr Langer indicated that he was
intending to run a campaign advocating informal voting and optional preferential
voting. After receiving warnings from the AEC, on 5 March 1993 he applied to
the High Court for an injunction to prevent the AEC from intimidating him, and
a declaration that section 329A was unconstitutional. The High Court dismissed
his injunction application, but referred the constitutionality of section 329A
to the Full Bench.
On 7 February 1996 (8 days after the issue of the writs for
the 1996 election) the High Court decided3 that section 329A was a valid
enactment of Parliament.
At the 1996 election, Mr Langer again indicated that he was
intending to run a campaign advocating informal voting and optional
preferential voting and after he published an advertisement which was clearly
in breach of section 329A, the AEC obtained an injunction against him from the
Victorian Supreme Court. Mr Langer immediately defied that injunction, and was
sent to jail for contempt of court. Mr Langer then appealed the injunction
to the Federal Court and lost. He then appealed the contempt order and was
given early release from jail.[44]
8.53
Following the 1996 election, the then Joint Standing Committee on
Electoral Matters examined the operation of the savings clause and the impact
of sections 329(3) 329A on the electoral process. The committee found that in
its view, ‘the Langer affair has clearly shown that section 329A is an
ineffective and heavy-handed provision’ and recommended that section 329A and
related provisions should be repealed, while the wording of section 240 should
be clarified.[45]
8.54
In 1998, the former sections 270(2), 329(3) and 329A of the Commonwealth
Electoral Act were repealed by the Electoral and Referendum Amendment Act
1998, and section 240(2) was introduced, so that section 240 now reads:
240 Marking of votes in House of Representatives election
(1) In a House of Representatives election a person shall
mark his or her vote on the ballot-paper by:
(a) writing the number 1
in the square opposite the name of the candidate for whom the person votes as
his or her first preference; and
(b) writing the numbers 2,
3, 4 (and so on, as the case requires) in the squares opposite the names of all
the remaining candidates so as to indicate the order of the person’s preference
for them.
(2) The numbers referred to in paragraph (1)(b) are to be
consecutive numbers, without the repetition of any number.[46]
8.55
The AEC noted that the current wording ‘clearly prescribes full
preferential voting, and both the savings provision and sanctions for
advocating other than full preferential voting have been removed’.[47]
8.56
In 1996, when non-sequentially numbered ballot papers were last admitted
as formal and were classified by the AEC as ‘exhausted’, a total of 48,979 such
votes were cast out of 10,883,852 formal votes (0.45 per cent). The exhausted
votes were of a similar magnitude for each of the states and territories, NSW
(0.46 per cent), Vic (0.49 per cent), QLD (0.26 per cent), WA (0.62 per cent),
SA (0.51 per cent), TAS (0.21 per cent), ACT (0.49 per cent), NT (0.48 per
cent).[48]
8.57
At the 1998 election, the first following the abolition of the savings
provision, the level of informality for the House of Representatives election
rose from 3.4 per cent to 4.2 per cent.[49]
It is not known how much of the increase was due to the removal of the savings
provision. Three years later at the 2001 election, around
15,000 Langer-style votes were identified by the AEC as part of its
informal vote survey for the election. The proportion of ‘Langer style’
informal votes as a proportion of total informal votes varied significantly
across jurisdictions (table 8.8).
Table 8.8 ‘Langer style’ informal votes, by jurisdiction,
2001 election (per cent)
State/territory
|
Langer
style
|
Non
sequential
|
Total
informal votes
|
New South Wales
|
2.37%
|
22.52%
|
5.42%
|
Victoria
|
3.22%
|
10.49%
|
3.98%
|
Queensland
|
2.00%
|
14.15%
|
4.83%
|
Western Australia
|
4.18%
|
21.75%
|
4.92%
|
South Australia
|
1.05%
|
13.40%
|
5.54%
|
Tasmania
|
6.88%
|
13.17%
|
3.40%
|
Australian Capital
Territory
|
0.83%
|
7.66%
|
3.52%
|
Northern Territory
|
14.56%
|
15.06%
|
4.64%
|
National
|
2.68%
(15,564)
|
17.18%
(99,946)
|
4.82%
(580,590)
|
Source Australian
Electoral Commission, Research report number 1: Informal vote survey, House of
Representatives, 2001 election (2002), p 8.
8.58
The AEC estimates that if such ballot papers had been able to be counted
at the 2007 election, up to 90,149 additional ballot papers would have been
admitted across Australia, representing 17.79 per cent of total informal
votes at the 2007 election.[50]
8.59
The AEC notes that those ballot papers that were saved by the operation
of section 270(2) are likely to fall into three categories – those so marked
deliberately to take advantage of the savings clause; those so marked
accidentally in the belief that preferences were optional and those so marked
accidentally. The AEC considered that it is not possible to identify the
motivation behind ballot paper markings, although previous AEC research reports
have examined the possible environmental determinants of ballot paper
informality.[51]
8.60
It is of note that for the 10 divisions in NSW recording the highest
rates of informality at the 2007 election, votes classified as ‘non sequential’
are more likely to feature as a reason why votes were ruled informal compared
to the NSW average (table 8.9).
Table 8.9 Highest informality divisions, by informality
type, 2007 election (per cent)
Division
|
Totally blank
|
‘1' only
|
Incomplete numbering
|
Ticks and crosses
|
Non sequential
|
Marks and scribbles
|
Other (symbols, illegible, other)
|
|
Blaxland
|
19.04
|
34.54
|
7.61
|
10.85
|
15.03
|
9.98
|
2.95
|
Watson
|
16.57
|
46.15
|
5.30
|
10.10
|
7.82
|
10.72
|
3.34
|
Chifley
|
18.51
|
28.56
|
8.53
|
11.57
|
25.73
|
6.09
|
1.01
|
Prospect
|
18.44
|
34.76
|
3.26
|
19.12
|
11.16
|
8.87
|
4.39
|
Fowler
|
14.61
|
42.62
|
2.05
|
21.53
|
6.53
|
10.99
|
1.67
|
Reid
|
19.53
|
32.81
|
6.67
|
10.3
|
14.34
|
14.52
|
1.83
|
Parramatta
|
18.81
|
30.63
|
11.22
|
7.64
|
20.7
|
9.43
|
1.57
|
Werriwa
|
19.12
|
41.8
|
4.24
|
13.31
|
10.34
|
10.36
|
0.83
|
Banks
|
18.87
|
40.23
|
5.28
|
10.38
|
11.54
|
10.15
|
3.55
|
Bennelong
|
11.31
|
22.15
|
11.99
|
5.31
|
32.79
|
9.77
|
6.68
|
NSW average
|
18.21
|
36.23
|
5.29
|
11
|
15.78
|
11.51
|
1.98
|
|
|
|
|
|
|
|
|
|
|
|
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, pp 11 and 15.
8.61
The AEC considered that any reintroduction of the savings provision,
while appearing to be relatively simple, would instead ‘reinstate policy
conflicts in the Commonwealth Electoral Act remedied by the 1998 amendments’.[52]
8.62
Some of the basic policy conflicts identified by the AEC in 1996 were
that with the savings provision in place, the Commonwealth Electoral Act:
n required full
preferential voting;
n prohibited inducing
optional preferential voting;
n prohibited the
advocacy of optional preferential voting;
but
n allowed certain
optional preferential votes as formal.[53]
8.63
The AEC considered that these contradictions would occur again if a
savings provision were introduced. The AEC noted that:
While various courts had consistently upheld the Parliament’s
intentions in enacting the above provisions of the Commonwealth Electoral Act,
there was a perception that there was a ‘loophole’ in the Commonwealth
Electoral Act that allowed for the avoidance of the requirement for full
preferential voting. This perception may have been exacerbated by the increased
availability of optional preferential voting at state elections; it remains
possible that some electors are confused by the different ballot paper marking
requirements across Commonwealth and state elections.
It is also clear that the reintroduction of a savings
provision alone would see an increase in the number of House of Representatives
ballot papers that were not fully preferenced. Not only could non-aligned
campaigns re-emerge to advocate less than full preferential voting, but
experience in those states and territories with optional preferential voting
show that political campaigns themselves move towards advocating exhausting a
ballot paper to limit preference flows to other groups and parties.[54]
8.64
The AEC considered that without a reintroduction of the sanctions that
prevented advocating other to vote in a way other than in accordance with full
preferential voting ‘the effect would be to allow for open and possibly
widespread advocacy of optional preferential voting at federal elections’.[55]
The AEC noted that:
This would send a clear signal that Parliament was accepting
in principle that optional preferential voting should exist as an alternative
to full preferential voting for federal elections, although the Commonwealth
Electoral Act may not clearly state as much. The question would then arise as
to why Parliament does not expressly provide for optional preferential voting
in the Commonwealth Electoral Act, rather than allowing it to exist only as a
‘loophole’.
Of more concern is the possibility that if Parliament were to
introduce savings provisions without prohibiting advocacy in relation to those
provisions, public confusion about the real intentions of the legislators on
the method of voting required under the Commonwealth Electoral Act can be
expected to increase under the pressure of well-organised public campaigns in
support of optional preferential voting. The AEC does not believe that this
potential confusion can be properly and appropriately addressed by AEC
education campaigns alone.[56]
Committee conclusion
8.65
The committee supports the retention of full preferential voting for
House of Representatives elections. That said, it is important that where an
elector expresses a clear preference but makes a mistake when completing the
ballot paper, that the vote should be included in the count up to the point
where the mistake is made.
8.66
The savings provision used for South Australian House of Assembly
elections significantly reduces informality and would have potentially ‘saved’
almost 154,000 votes at the 2007 federal election, had such a provision
been included in the Commonwealth Electoral Act. However, the committee
considers that the South Australian model which also saves votes where only a
single preference is expressed (including a ‘1’, a tick or a cross), is a step
too far, in that it may actively encourage optional preferential voting rather
than operating as a genuine savings provision.
8.67
The AEC has provided evidence that up to 90,000 votes may have been
cast in 2007 where the ballot paper included non-sequential numbering.[57]
The number of votes in this category has remained at this level for the past
three elections. One-third of these informal votes are cast in NSW, with
significant numbers also recorded in Victoria, Queensland and Western Australia
(table 8.10).
Table 8.10 Number
of informal votes attributed to non-sequential numbering, 2001 to 2007
elections
|
2001
|
2004
|
2007
|
New South Wales
|
40,040
|
33,914
|
33,375
|
Victoria
|
14,683
|
25,122
|
23,136
|
Queensland
|
9,750
|
11,729
|
13,290
|
Western Australia
|
10,695
|
11,143
|
12,416
|
South Australia
|
6,972
|
7,829
|
6,180
|
Tasmania
|
1,303
|
964
|
1,473
|
Australian Capital
Territory
|
492
|
363
|
524
|
Northern Territory
|
615
|
840
|
960
|
Total
|
84,550
|
91,904
|
91,354
|
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, p 10; Australian
Electoral Commission, Informal Vote Survey House of Representatives 2001
Election (2003), Research Report number 1; Analysis of Informality during House
of Representatives 2004 Election (2005), Research report number 7.
8.68
Further, there are clear risks associated with ballot flooding given the
strong relationship between higher number of candidates and higher informality
rates. Of the 10 divisions with the highest informality rates in NSW at
the 2007 election, those with the highest numbers of candidates (Blaxland,
Chifley, Reid and Parramatta) were associated with a higher than average
proportion of ballot papers ruled informal because of non-sequential numbering
errors (table 8.11). In these 10 divisions alone, 10,091 ballot
papers were ruled informal due to non-sequential numbering.
Table 8.11 Informality
in the 10 NSW divisions with the highest informality rates, 2007 election
Division
|
Informality rate (%)
|
Change in informality
compared to 2004 election
(% points)
|
Number of candidates 2007
election
|
Change in number of
candidates compared to 2004 election
|
Proportion of informal votes
due to non-sequential numbering (%)
|
Number of votes ruled
informal due to non-sequential numbering
|
Blaxland
|
9.49%
|
-0.78
|
8
|
0
|
15.03%
|
1,221
|
Watson
|
9.05%
|
0.05
|
6
|
-1
|
7.82%
|
625
|
Chifley
|
7.99%
|
-2.11
|
9
|
+1
|
25.73%
|
1,795
|
Prospect
|
7.73%
|
-1.68
|
5
|
-1
|
11.16%
|
742
|
Fowler
|
7.67%
|
-1.44
|
4
|
-1
|
6.53%
|
424
|
Reid
|
7.57%
|
-3.77
|
7
|
-1
|
14.34%
|
933
|
Parramatta
|
6.56%
|
-1.97
|
10
|
-1
|
20.70%
|
1,238
|
Banks
|
6.36%
|
-1.57
|
6
|
-1
|
11.54%
|
647
|
Werriwa
|
6.53%
|
-1.45
|
6
|
-1
|
10.34%
|
576
|
Bennelong
|
6.22%
|
-0.24
|
13
|
+6
|
32.79%
|
1,890
|
NSW average
|
4.90%
|
-1.20
|
7
|
-0.6
|
15.78%
|
681
|
Source Australian
Electoral Commission, Analysis of informal voting: House of Representatives
2007 election (2009), Research report number 11, pp 13, 38–86; ‘Virtual
Tally Room, House of Representatives, House of Representatives Results,
Candidates’, viewed on 26 May 2009 at http://results.aec.gov.au/13745/Website/HouseCandidatesBySurname-13745-A.htm;
‘Virtual Tally Room, House of Representatives, Candidates’, viewed on
26 May 2009 at http://results.aec.gov.au/12246/results/HouseCandidatesBySurname-12246-A.htm.
8.69
The committee’s preference therefore is to reinstate the savings provisions
that existed in the Commonwealth Electoral Act between the 1984 and 1996
elections to include those ballot papers where there are non consecutive
numbering errors in the count up to the point at which the numbering errors
began.
8.70
While the committee acknowledges the AEC’s concerns in relation to the
potential re-emergence of campaigns advocating for optional preferential
voting, the committee considers that these concerns do not justify the
exclusion of up to 90,000 votes where electors have expressed clear
preferences for a number of candidates but may have made mistakes in numbering
their ballot paper. Such a savings provision will also provide some insurance
that election results are not affected by deliberate attempts to increase the
number of candidates (and thereby leading to a rise in the informality rate) to
influence the outcomes of an election.
8.71
Under the committee’s proposal, votes marked with a single preference
(either a ‘1’, a tick, or a cross), will not be saved. The committee therefore
does not consider that the reinstatement of the savings provision should be
seen as accommodating optional preferential voting.
8.72
The committee recognises that the reinstatement of such a provision
would need to be accompanied by an appropriate penalty provision to deter the
advocacy of a vote other than in accordance with full preferential voting.
Recommendation 35 |
8.73
|
The committee recommends that:
n Section
240 (2) of the Commonwealth Electoral Act 1918, which provides that
the numbers on House of Representatives elections ballot papers are to be
consecutive numbers, without the repetition of any number, be repealed, and
n the
savings provision contained in paragraph 270 (2), repealed in 1998, which
provided that in a House of Representatives election in which there were more
than three candidates, and where a full set of preferences was expressed on
the ballot paper, but there were non-consecutive numbering errors, the
preferences would be counted up to the point at which the numbering errors
began, at which point the preferences were taken to have ‘exhausted’, be
reinstated to the Commonwealth Electoral Act 1918, and
n the
Government amend the Commonwealth Electoral Act 1918 to provide a
penalty provision sufficient to deter the advocacy of ‘Langer style voting’.
|
Implications arising from the McEwen petition
8.74
The Court of Disputed Returns’ (CDR) decision on the McEwen petition and
the subsequent review of the AEC’s ballot paper guidelines and recount policy
have significant implications for the administration of elections by the AEC.
8.75
The committee has examined the decision and throughout the Inquiry has
canvassed various options designed to mitigate against similar experiences
occurring in future elections.
Court of Disputed Returns process and decision
8.76
As noted in chapter 2, a petition was filed with the Court of Disputed
Returns on 25 January 2008, relating to the conduct of the recount in the
division of McEwen.
8.77
A summary of the events commencing with the count of ballot papers in
the division of McEwen through to the court’s decision is outlined in table 8.12.
8.78
The final decision by the court was made on 2 July 2008, with the court
ruling that the final margin in favour of Ms Fran Bailey was 27 votes.[58]
8.79
In coming to this view, the court conducted a review of
643 ‘reserved’ ballot papers that had been set aside during the recount
when scrutineers challenged the decisions of the Divisional Returning Officer.
As a result of the court’s review of these ballot papers, the Court reversed
154 of the decisions made by the Australian Electoral Officer during the
recount in respect of the 643 ballot papers on which it ruled.[59]
8.80
The court also made a number of important observations in respect of
issues associated with ruling on the formality of ballot papers and developed a
set of ‘principles’ (the first two ‘cardinal’ principles and the second three
‘subordinate’ principles) that reflected past practice in ruling on formality
including elements of various judgements by courts on these matters. The
principles developed were:
n That the ballot,
being a means of protecting the franchise, should not be made an instrument to
defeat it;
n Doubtful questions of
form should be resolved in favour of the franchise where there is no doubt as
to the real intention of the voter;
n When seeking to
determine the voter’s intention resort must be had, exclusively, to what the
voter has written on the ballot paper;
n The ballot paper
should be read and construed as a whole; and
n A voter’s intention
will not be expressed with the necessary clarity unless the intention is
unmistakeable and can be ascertained with certainty.[60]
8.81
Three ballot papers, which did not bear the initials of the presiding
officer, were ruled to be informal under s 268(1)(a) of the Commonwealth Electoral
Act.[61] The three ballot papers
were of different appearance and of these, two appeared to be initialled on the
back by the DRO as being admitted to the recount accompanied by the text ‘DRO
convinced the ballot paper came from a legitimate pre-poll envelope through the
dec exchange’.[62]
8.82
While paragraph 268(1)(d) of the Commonwealth Electoral Act provides
that a ballot-paper will be informal if ‘it has upon it any mark or writing…by
which, in the opinion of the Divisional Returning Officer, the voter can be
identified…’, the court ruled that three ballot papers, which had on them what
appeared to be initials, to be formal votes. [63] This was consistent
with, and confirmed established AEC guidance on formality, which provides that
a person’s initials annotated on a ballot-paper will not usually identify a
voter, noting that on a divisional roll with around 100,000 voters there will
frequently be several, if not numerous people with the same two initials.[64]
Table 8.12 Timeline of events, division of McEwen, 2007
election
Date
|
Event
|
2007
|
|
Saturday
24 November
|
Polling
day: ordinary, provisional and absent votes were cast by electors at polling
places between 8 am and 6 pm. Ordinary votes were counted at the polling
places in McEwen, for the purposes of providing an election night tally,
which at the conclusion of counting showed Mr Mitchell leading by 558 votes.
|
Monday
26 November
|
A fresh
scrutiny commenced of ordinary ballot-papers received at the Divisional
Counting Centre from polling places across McEwen. The fresh count included a
check of ballot-papers for formality, first preference totals and tallies of
the preferred vote for Mr Mitchell and Ms Bailey. The counting continued
until 10 December and, in addition to ordinary votes, included postal votes,
which may be accepted up to 13 days after polling day, namely 7 December
2007.
|
Monday
10 December
|
The DRO
for McEwen (the DRO) concluded the distribution of preferences, resulting in
a majority for Mr Mitchell of 6 votes as follows:
Mr Mitchell 48,416
Ms Bailey 48,410
Informal 3,823
(3.8%)
Total 100,649
Ms
Bailey wrote to the DRO setting out a number of reasons supporting a request
for a recount. Independently of the specific issues raised in Ms Bailey’s
letter, the Australian Electoral Officer (AEO), in close consultation with
the Electoral Commissioner and in accordance with section 279 of the
Commonwealth Electoral Act, directed the DRO to conduct a recount of all
ballot-papers.
|
Wednesday 12 December
|
The DRO commenced the recount
at the Divisional Counting Centre. In total, 4,116 ballot-papers were
declared to be informal in the recount. The DRO estimates between 1200 and
2000 ballot-papers were referred for his personal decision on formality.
Scrutineers disagreed with the decision of the DRO in respect of 643
ballot-papers and these were reserved for decision of the AEO.
|
Thursday 13 December
|
The AEO advised candidates
that he would commence consideration of reserved ballot-papers the next day
at his office in Melbourne and that they were each entitled to appoint one
scrutineer to observe the process.
|
Friday 14 December and Monday
17 December
|
The AEO made decisions in
respect of 406 ballot-papers on 14 December and decisions on the
remaining 237 ballot-papers on 17 December.
|
Wednesday 19 December
|
The AEC announced that the
recount of all ballot-papers resulted in a majority for Ms Bailey of 12 votes
as follows:
Mr Mitchell 48,253
Ms Bailey 48,265
Informal 4,116 (4.1%)
Total 100,634
The recount identified a
number of errors that contributed to a net decrease of 15 ballot-papers from
100,649 to 100,634. As well, the recount took account of the AEO’s decisions
on the formality of reserved ballot-papers which contributed to the increase
in informal ballot-papers of 293 – from 3,823 to 4,116.
|
Thursday 20 December
|
The DRO declared Ms Bailey as
the elected candidate for McEwen.
|
Friday 21 December
|
The EC certified in writing
that Ms Bailey was the elected candidate, attached the certificate to the
writ for the general election relating to the members of the House of
Representatives to be elected from Victoria, and returned the writ to the
Governor-General.
|
2008
|
|
Friday 25 January
|
Mr Mitchell disputed the
outcome of the election by petition to the Court of Disputed Returns (CDR),
complaining that a significant number of the 643 reserved ballot-papers had
been wrongly rejected by the AEO.
|
Thursday 21 February
|
The petition was heard before
Justice Crennan of the High Court. The Court ordered that the matter be
remitted to the Federal Court of Australia (FCA) and set a timetable for
submissions to be filed by each party. The Court also ordered that the AEC
deliver to the Victorian Registry of the FCA the 643 ballot-papers reserved
for the consideration of the AEO.
|
Friday 28 March
|
The first directions hearing
before Justice Tracey of the FCA, sitting as the CDR, between Mr Mitchell
(Petitioner), and Ms Bailey (First Respondent) and the AEC (Second Respondent).
During the directions hearing submissions were made by Counsel on what, if
any, access would be provided to the reserved ballot-papers. Arising out of
the hearing the parties undertook to provide submissions to the Court on the
principles of formality and an agreed submission on the process to be adopted
for the hearings.
|
Tuesday 22 April
|
The Court handed down its
reasons for its decision that the Petitioner and the First Respondent could
not view copies of the 643 reserved ballot-papers (Mitchell v Bailey (No 1)
[2008] FCA 426).
|
Friday 2 May
|
In a second directions
hearing, Justice Tracey indicated that he was considering providing Counsel
for the Petitioner and Counsel for the First Respondent with access to the
reserved ballot-papers.
|
Wednesday 21 May
|
This was the first day of two
days of full hearing before Justice Tracey. The Court ordered that Counsel
for Mr Mitchell and for Ms Bailey were to have access to the reserved
ballot-papers under Court supervision so that they could identify the
ballot-papers where the decision of the AEO was disputed. Subsequently, in
their submissions to the Court, Counsel for both parties disputed only 285 of
the 643 ballot-papers.
|
Tuesday 17 June
|
On the second day of the
hearing, Counsel made submissions to the Court on ballot-paper formality.
|
Wednesday 2 July
|
The CDR handed down its
judgement that the decisions of the AEO in respect of 153 of the 643 reserved
ballot-papers should be changed, finding that 12 ballot-papers should have
been treated as informal rather than formal, and that 141 should have been
treated as formal rather than informal (Mitchell v Bailey (No2) [2008] FCA
692) (Mitchell v Bailey). The Court’s decisions resulted in an increased
majority for Ms Bailey of 31 votes as follows:
Ms Bailey 48,339
Mr Mitchell 48,308
Informal 3,987 (4.0%)
Total 100,634
|
Friday 11 July
|
The CDR handed down its
judgement that the Commonwealth should meet the legal costs of both Ms Bailey
and Mr Mitchell. The Commonwealth had not opposed the making of such orders
(Mitchell v Bailey (No3) [2008] FCA 1029).
|
Source Henderson
A, Review of ballot-paper formality guidelines and recount policy (2008), pp 2-5,
(exhibit 4).
Henderson review
8.83
Following the decision of the Court of Disputed Returns (CDR) decision,
the AEC commissioned Mr Alan Henderson PSM to conduct a review to identify
action that should be taken by the AEC to ensure that processes and procedures
are in pace for future elections to address the matters identified in the
Court’s decision.[65] The review was to:
n consider the specific
ballot-papers and the Court’s decision in Mitchell and any implications in the
way in which electoral officials are supported by AEC policies, guidelines,
procedures, manuals, and training in making decisions about the formality of
ballot-papers;
n consult with key
stakeholders about the impact of the Court’s decision on the scrutiny process
for electoral events;
n identify measures to
improve the quality, consistency, transparency and accountability of
decision-making by electoral officials on the formality of ballot-papers; and
n identify any
necessary changes to the existing policies, guidelines, procedures, manuals and
training produced by the AEC on the formality of ballot-papers.[66]
8.84
Given the potential link between decisions on the formality of ballot-papers
and the case for undertaking recounts, the review was also required to consider
the AEC’s policy on recounts and identify possible criteria for accepting or
rejecting requests for a recount.[67]
8.85
Judicial redress based on the formality of ballot-papers in respect of
House of Representatives election outcomes has been exceedingly rare, with two
or three cases in the period since the passage of the original Commonwealth
Electoral Act 1902.[68] Mr Henderson considered
that two factors seem relevant in explaining the rarity of court challenges:
(a) the size of the winning margin; and (b) the effectiveness and transparency
of electoral administration. Mr Henderson noted that:
These factors are related. In situations where there is great
confidence in the effectiveness of administration it is less likely that
results will be subject to challenge. Transparency is important in building
confidence not least because administrative decisions and procedures that are
subject to scrutiny by stakeholders are likely to be more rigorous and
effective than those that are never subject to external scrutiny.[69]
8.86
Mr Henderson noted that close results are rare, with only five divisions
in the last five federal elections having a winning margin of less than 100
votes.[70] Court challenges to the
outcome of House of Representative divisional elections have also been rare –
with at most three cases identified since federation.[71]
Mr Henderson noted that:
The evidence from these cases shows that the impact of
disagreements on formality is muted because they do not consistently favour
particular candidates. While the Court reversed the decisions on the formality
of 153 of the 100,634 ballot-papers in McEwen, the total number of formal votes
for one candidate increased in net terms by 74 and the votes for the other
candidate increased by 55.[72]
Guidance on formality and Australian Electoral Commission manuals
8.87
The key guidance in the CDR decision in Mr Henderson’s view, emphasised
the requirement for ballot-papers to be read and construed as a whole, with one
or more poorly formed numbers to be deciphered in the context of a consecutive
series of numbers rather than as single number in isolation.[73]
Mr Henderson considered that it was the interpretation of this principle of
‘the ballot-paper as a whole’ that accounted for the great majority of the 153
ballot-papers where the CDR reversed the decision of the AEO.[74]
8.88
After reviewing the AEC’s guiding documents on formality, Mr Henderson
found that the main AEC documents, the Polling Place Procedures Manual,
Training of Operational Staff Manual and Scrutineer’s Handbook ‘do not directly
address the whole of ballot-paper principle’. Mr Henderson also considered that
the Formality Policy, which is prepared primarily for the information of
permanent AEC officials’ does address the whole of ballot-paper principle and
the interpretation of unclear markings or poorly formed numbers. In relation to
the Formality Policy, Mr Henderson considered that there was some divergence
between the Formality Policy and the CDR decision. Mr Henderson noted that:
The Formality Policy document refers to one unclear marking
in a consecutive series of numbers, whereas the CDR found in Mitchell v Bailey
that the ‘discernable sequence of numbers’ test may mean that a formal ballot-paper
could include more than one unclear marking. In sum, there was a gap in AEC
guidance for the consideration of the admission or rejection of ballot-papers.
This can be addressed by incorporating the ‘discernable sequence of numbers’
test supported by realistic illustrations of poorly formed numbers in AEC
manuals. The provision of publicly available comprehensive guidance for the
consideration of the admission or rejection of ballot-papers should provide the
basis for greater consistency and accuracy in decision-making on the formality
of ballot-papers.[75]
8.89
Mr Henderson made a number of recommendations to provide additional
guidance to AEC officials and scrutineers on ballot paper informality. Mr
Henderson recommended that:
n The following
guidance provided by the CDR in Mitchell v Bailey should be incorporated
in AEC manuals, handbooks and training:
§
Ballot-papers should be read and construed as a whole, with one
or more poorly formed numbers to be deciphered in the context of a consecutive
series of numbers rather than as single numbers in isolation;
§
Poorly formed numbers must bear a reasonable resemblance to
identifiable numbers;
§
Unconventional but recognisable numbers such as continental 1s
and 7s are acceptable;
§
Initials annotated on a ballot-paper will not usually identify a
voter and therefore does not provide a basis for rejecting a ballot-paper; and
§
If a ballot-paper lacking official markings is considered
authentic, then the annotation made by the Divisional Returning Officer (DRO)
under subsection 268(2) of the Commonwealth Electoral Act should specify that
the DRO is ‘satisfied that it is an authentic ballot paper’– Section 6(e).
n A single
comprehensive set of information on formality, including guidelines and
illustrative ballot-papers for decision-makers on the admission or rejection of
ballot-papers should be made available on the AEC website at www.aec.gov.au.
The illustrative ballot-papers should include actual examples of poorly formed
numbers on formal ballot-papers as well as extremely poorly formed numbers that
would render a ballot-paper informal – Sections 7(b) and 7(c).
n To assist
decision-making in the potentially tense environment of close election counts:
§
The briefs (manuals or handbooks) documenting guidelines on
formality for decision-makers made available to scrutineers should be as
comprehensive as, and identical in their relevant wording to, those available
to electoral officials;
§
Officials should brief scrutineers on the guidelines at the
commencement of counting processes;
§
Copies of the guidelines on formality should be readily available
in counting centres;
§
Officials should be prepared to fully explain their reasoning by
reference to the guidelines in relation to their decisions on specific
ballot-papers; and
§
At least in recounts, scrutineers should be prepared to explain
their reasoning for seeking the reserving of ballot-papers by reference to the
guidelines – Section 5.[76]
Recount policy
8.90
Mr Henderson noted that the current AEC recount policy provides that ‘there
is no minimum number under which a recount will occur’ because ‘given the
checks and balances’ in the scrutiny system significant sorting errors are
highly unlikely to go undetected.[77] Mr Henderson
considered that the evidence for the period since 1984 supports this judgement,
noting that:
The average difference is 22 votes for the 6 recounts where
the margins for both the initial count and the recount are available. The size
of the change between the initial count and the recount in McEwen, 18 votes,
was close to the average. However the direction of the change combined with the
narrow initial margin meant it was very significant because it changed the
result of the election.[78]
8.91
In considering whether there should be an ‘automatic’ threshold for
conducting a recount, Mr Henderson noted that the AEC generally undertakes
recounts when the margin is less than 100 votes. Mr Henderson considered that
there would be some merit in introducing 100 votes as an automatic threshold
for recounts, and for ‘State and National Office executives to monitor progress
in close seats, to ensure that additional support is readily available to
address the inevitable pressures that arise in close counts’.[79]
8.92
Mr Henderson made a number of recommendations in relation to the AEC’s
recount policy, recommending that:
n A threshold for an
‘automatic’ recount should be introduced with the key elements of recount
policy revised to read as follows: A request for a recount needs to
identify specific ballot-papers and associated significant counting process
errors or irregularities that could change the result of an election within a
division, unless the margin of votes on the initial count is less than 100, in
which case a recount will be undertaken as a matter of course – Section 8(c).
n The details of all future
recounts and requests for recounts should be systematically documented and
assessed by National Office – Section 8 (c).
n Consistent with the
practise of identifying close seats to be followed during counting, senior
State and National Office executives should monitor progress in those seats, to
ensure that additional experienced support and resources arereadily available
to address the inevitable and appropriate increase in expectations in regard to
transparency and attention to detail that arise in close counts – Section 8(d).
n Senior management
should emphasise the importance of the existing policy whereby DROs and AEOs
are expected to consult senior managers in the State and National Office
respectively, including informing the Deputy Electoral Commissioner, before
deciding whether to undertake a recount – Section 8(d).[80]
8.93
The committee notes that the AEC already commenced administrative action
to implement the recommendations from the Henderson Report relating to training
and information materials.[81] The AEC noted that:
It is anticipated that this action will assist in addressing
any reasonable concerns about the AEC’s handling of disputed ballot papers. The
CDR decision has provided the AEC with clear guidance on the application of the
formality rules in the Commonwealth Electoral Act. Additionally, the AEC now
has a wealth of examples and precedents of disputed ballot papers that have
been ruled upon by the CDR and which have been analysed and adapted to form the
basis of new manuals, handbooks and training.[82]
Process for handling disputed ballot papers
8.94
During the course of the inquiry the committee explored the option of
replacing the single decision-maker for dealing with reserved ballot papers in
a recount situation with a panel of three comprising the relevant AEO and two
other AEC officers at the Senior Executive Service level or equivalent.[83]
The committee also examined whether such a process should involve discussions
between the three members in the presence of (or absence of) scrutineers, and
whether it was necessary for the panel to come up with a written statement of
reasons as to the reasons behind the rulings on individual ballot papers.[84]
8.95
The rationale behind such a move would be to increase confidence in
rulings on disputed ballot papers thereby avoiding a lengthy CDR process and
the associated uncertainty of an election result. The replacement of a single
AEO ruling on disputed ballot papers would not change the existing process
whereby parties can petition the CDR disputing an election result.
8.96
The AEC’s final response to this issue took account of the findings of
the Henderson review and the CDR decision on the McEwen petition. The AEC noted
that the report by Mr Henderson sets out for the AEC the broad guidance given
by the CDR on formality and recommended that a single comprehensive set of
information on formality be developed.[85] The AEC was of the view
that the provision of such information was likely to increase transparency in
the process, noting that:
The development of such information, and the training of AEC
officers and the stakeholders on the formality rules flowing from the CDR
decision and the Henderson Report, will provide greater transparency in the
decision-making process by the AEO and will assist in both identifying those
ballot papers that are really in dispute and preventing unnecessary challenges.[86]
8.97
In the AEC’s view, changing the existing single decision-maker to
include some panel arrangements involved a number of risks that would not
necessarily reduce delays in the return of the writs.[87]
The AEC noted that:
To replace the AEO with a panel that includes the AEO and two
other senior AEC officers could give rise to concerns that the relatively
straightforward process that currently exists will be elevated to a court-like
setting with the panel needing to retire to make final decisions and then
returning before the scrutineers to notify them of the panel’s decision. An
alternative process would be to have the panel discuss the matters before the
scrutineers and reach a decision in their presence. This alternative process
then raises the issue about how any minority views are resolved in front of
scrutineers.
Inherent risks of increasing the lodging of petitions to the
CDR challenging the panel’s decisions in the absence of transparency appear to
exist if either of these two processes were to be adopted. The risks would
clearly increase where the panel decision was not unanimous and that lack of
unanimity occurred in the presence of scrutineers. This would also run the risk
described by Associate Professor Graeme Orr in Submission No. 187 that the
panel ‘would be sitting in judgment on the decisions of other electoral
officials in a court setting’.
The AEC is concerned that delays in the return of the writs
would occur if any new process involving a proposed decision-making panel
results in the need for a written statement of reasons to be prepared and
published. The imposition of a requirement to provide statements of reasons
would appear to be inevitable to preserve transparency in decision-making if
the panel was to retire from the presence of scrutineers to consider the
disputed ballot papers.[88]
8.98
In his review, Mr Henderson recommended that officials should be
prepared to ‘fully explain their reasoning by reference to the guidelines in
relation to their decisions on specific ballot papers’.[89]
The AEC noted that presently, this full explanation of the reasoning of the
decision-maker occurs when the AEO conducts the scrutiny of the reserved ballot
papers in the presence of the scrutineers. This is all done orally as only the
final decision itself (i.e. ‘admitted’ or ‘rejected’) is required to be
recorded in writing (see subsection 279B(7) of the Electoral Act).[90]
8.99
In the AEC’s view, this recommendation does not require the
decision-maker to provide a formal statement of reasons and that the adoption
of such an alternative process would create additional risks. The AEC noted
that:
This would clearly create risks that additional time may be
required to formally record the basis for decisions and to obtain legal advice
in the preparation of formal reasons for decision.[91]
8.100
The AEC therefore suggested that rather than adopt a three person panel
model, the effectiveness of the measures to be adopted to implement the
recommendations of the report from Mr Henderson should be reviewed after the
next federal election in which an AEO is required to review reserved ballot
papers under sections 279B, 280 and 281 of the Commonwealth Electoral Act.[92]
Committee conclusion
8.101
The committee recognises that the closeness of the result in the
division of McEwen and the resulting Court of Disputed Returns’ petition was a
relatively rare event in the context of federal elections.
8.102
That said, the reversal of almost one-quarter of the AEO’s decisions in
respect of the 643 reserved ballot papers is of concern and may be seen as
putting community confidence in election results at risk. There is also the possibility
of increased disputation, as candidates in tight election contests may be encouraged
to take their chances by having the results of elections reviewed by a different
decision maker.
8.103
The committee supports the review process adopted by the AEC following
the decision by the Court of Disputed Returns on the McEwen petition.
8.104
The committee endorses the AEC’s proposed response in implementing the
recommendations by Mr Henderson, which the committee agrees, should provide for
a greater understanding by electoral officials and scrutineers about rulings on
formality.
8.105
The additional transparency associated with the publication of
guidelines on formality, which incorporate illustrative ballot papers based on
the judgement by the Court of Disputed Returns in the McEwen petition, is also
welcomed.
8.106
The committee supports the AEC’s arguments in continuing with a single electoral
official (the AEO for the respective jurisdiction), to rule on formality in a
recount situation, given the AEC’s adoption of the recommendations in the
Henderson review.
8.107
The committee also supports the AEC proposal that the process be reviewed
after the next federal election in which an AEO is required to review reserved
ballot papers under sections 279B, 280 and 281 of the Commonwealth Electoral
Act.
8.108
It is of concern to the committee, however, that those ballot papers
which were considered formal by the DRO even though they did not contain the
initials of an issuing officer nor a watermark were not annotated by the DRO in
such a way as to reflect the requirements of s268(2) of the Commonwealth
Electoral Act.
8.109
The committee also agrees with the recommendation by Mr Henderson
that section 268(2) be amended to require a Divisional Returning Officer who
rules a ballot paper to be formal despite the ballot paper not containing
either the initials of a issuing officer or the official mark, to annotate the
ballot paper with the words ‘I am satisfied that this is an authentic ballot
paper’.
8.110
This move is complementary to the committee’s support of moves by the
AEC to amend the wording of s209A in order to allow for ballot papers to be
printed with a ‘feature approved by the Electoral Commission’. In combination,
these amendments will serve to eliminate confusion about ballot paper formality
(see recommendation 36 below).
8.111
Accordingly, the committee recommends that the AEC adopt all of the
recommendations contained in the report prepared by Mr Henderson, with the
exception of recommendation A(v) from that report which is the subject of
recommendation 37 of the committee.
Recommendation 36 |
8.112
|
The committee recommends that the Australian Electoral
Commission adopt all recommendations contained in the report entitled Review
of Ballot-Paper Formality Guidelines and Recount Policy prepared for the
Australian Electoral Commission by Mr Alan Henderson, except for
recommendation A(v) which is the subject of recommendation 37.
|
Recommendation 37 |
8.113
|
The committee recommends that section 268(2) of the Commonwealth
Electoral Act 1918 be amended to provide that in those cases where the
Divisional Returning Officer responsible for considering the question of the
formality of a ballot paper, is satisfied that the ballot paper is not
informal, because the Divisional Returning Officer is satisfied that it is an
authentic ballot paper on which a voter has marked a vote, the Divisional
Returning Officer be required to annotate the ballot paper with the words ‘I
am satisfied that this is an authentic ballot paper’.
|
Clarification of permitted official marks, and removals to ‘on-demand’ printing
of ballot papers
8.114
While ballot papers are typically printed on water marked paper, the
Commonwealth Electoral Act includes provisions that give the AEC some
flexibility to print ballot papers utilising regular printers. Such provisions
are usually used by the AEC in the early stages of postal voting and at
Australian embassies overseas.[93] Where ballot papers are
printed locally, the formality requirements of the Commonwealth Electoral Act
require that the ballot paper is authenticated by the initials of the presiding
officer.[94]
8.115
According to the AEC, it had received advice from the Australian
Government Solicitor following the CDR decision on the McEwen petition that put
a narrower construction on the meaning of paragraph 209A(b). The Australian
Electoral Commission noted that:
It is restrictive because of the legal interpretation of the
word ‘overprint’. What we have done, to some extent, may not have been strictly
legal in terms of the act but we need the flexibility. It is just the wording
in that section of the act. For example, we need the flexibility when we print
ballot papers from files sent overseas before they actually get ballot papers.[95]
8.116
In the AEC’s view, it was clearly not intended, at the time that
paragraph 209A(b) was inserted in the Act, that there would be a critical
distinction between ballot papers which had acquired their colour through
printing with a coloured wash, and ballot papers which had acquired their
colour through having been printed on dyed stock.[96]
Furthermore, the AEC noted that:
The advice has highlighted the need to have a provision
sufficiently flexible to enable the use of evolving technology for security
printing (including ‘on demand’ printing) – which these days can include
methods which do not fall within the definition of ‘overprinting’, such as the
use of stock with markings visible under ultraviolet light, or even the
inclusion of holograms.[97]
Committee conclusion
8.117
The committee notes the AEC’s legal advice and considers that there is
benefit in providing the AEC with additional flexibility in the printing of
ballot papers.
Recommendation 38 |
8.118
|
The committee recommends that paragraph 209A(b) of the Commonwealth
Electoral Act 1918 and paragraph 25A(b) of the Referendum (Machinery
Provisions) Act 1984 be repealed, and replaced with the words ‘a feature
approved by the Electoral Commission’.
|