QUOTA pr logo NOTES

 

Newsletter of the Proportional Representation Society of Australia

 

 

   QN2013C                     September 2013           www.prsa.org.au

 


2013 federal poll results show reform is needed

 

Australia changed its government after 22 seats changed hands following a national swing above 3.6% to the Coalition on 7 September 2013, taking it to a two-party-preferred majority of votes in all states and territories except the Australian Capital Territory.

 

HOUSE OF REPRESENTATIVES: Nominations were up 40% to 1,188, there being on average 7.9 candidates per seat with a median of 8. Nearly three-quarters were male, roughly the proportion of elected candidates that are male (74%). Victoria, Tasmania and the territories led the major influx of additional candidates, followed by Queensland, and only South Australia had fewer than in 2010, by two.

 

Victoria had 14 of the 26 seats with 10 or more candidates and an average of 9.3 with median 9, matched most closely by Western Australia with the same median, but a mean of 8.5. Both the mean and median were lowest at six in South Australia. Voters’ choice in provincial (mean of 8.6 candidates) and rural (8.1) divisions was greatest, and about equal in inner (7.6) and outer (7.7) metropolitan divisions.

 

With voters required to mark preferences for all candidates except one allowable blank square, all eight seats with informality levels above 10% were in NSW, which had 7.6% of ballots informal, compared with the national average of 5.9%. Use of optional preferential voting at state level, and the high percentage of voters whose first language was not English, were again major factors in this substantial discrepancy. The Northern Territory was next worse with 6.3% of ballots informal. The other mainland states had around 5% rejected, and Tasmania and the ACT about 4%. As in other recent federal elections, Labor’s ordinary votes tended to be 3 or 4 percentage points higher than its pre-poll or postal support levels.

 

Of the four seats with margins below the benefit likely from the donkey vote (0.25%), three (Barton, Indi, and perhaps Fairfax) were won by the candidate that drew the higher position on the ballot paper. Nationally Labor suffered about 5% swings in its marginal electorates, compared with 3% in fairly safe seats and around 1.5% in safe ones. It went backwards by nearly 3% in marginal Liberal or National electorates, and around 6% in others the former opposition held.


Overall the greatest swings against Labor were in inner metropolitan (over 5%) and provincial (over 4%) areas. Just over 40% of electorates had margins of 6% or less, with the rest usually classified as fairly safe or safe. Of the close outcomes, 12 including just three ALP-held ones would change with a swing of 1% or less, 9 (five ALP) with swings from 1 to 2%, 10 (five ALP) in the range 2-3%, 13 (eight ALP) with movements of 3-4%, 10 (four ALP) with 4-5% and 7 (four ALP) with 5-6%. In eleven seats, the last two candidates were not just from the ALP and Coalition, while 53 seats were decided on first preferences alone.

 

Ten seats changed hands in NSW on a swing of around 3% to the Coalition (the independents Tony Windsor and Rob Oakeshott did not recontest), four in Victoria following a swing above 5% (including Sophie Mirabella’s defeat by independent Cathy McGowan in Indi), three each in Tasmania (on an 11.3% swing) and Queensland (after a swing of around 1%, perhaps including capture of vacated Fairfax by Clive Palmer heading his new party), and one each in South Australia (which had a 5.5% swing) and Western Australia (a swing above 1% saw just O’Connor narrowly reverting to the Liberals after the National incumbent, Tony Crook, did not recontest).

 

As shown below, the percentage of first preferences or two-party-preferred support bore little relationship to seats won, which were generally distorted well beyond cube rule expectations, as so many seats were safe. Rural areas and metropolitan pockets had even greater local concentrations, given the winner-take-all system.

 

House of Representatives First Preference Votes and Seats

 



Coalition

Labor

Greens

Others

NSW

votes

47.3

34.5

8.0

10.2


seats

62.5

37.5

0

0

Vic

votes

42.7

34.8

10.8

11.7


seats

43.2

51.4

2.7

2.7

Qld

votes

45.7

29.8

6.2

18.3


seats

73.3

20.0

0

6.7

WA

votes

47.3

28.8

9.7

14.2


seats

80.0

20.0

0

0

SA

votes

44.5

35.7

8.3

11.5


seats

54.5

45.5

0

0

Tas

votes

40.3

34.8

8.3

16.6


seats

60.0

20.0

0

20.0

Aus

votes

45.1

33.4

8.7

12.8


seats

60.0

36.7

0.7

2.7


SENATE:
An increase in Senate candidates of 50% nationally meant the unjustifiable requirements for casting a formal ballot (at least 90% of the individual squares to be marked with no more than three departures from sequential numbering) were particularly onerous, demanding at least 22 and 25 numbers in the territories and ranging between 49 in Tasmania, and a new unwanted record in Australian public elections of 99 in NSW. There were increases of more than half since 2010 in Victoria, SA, Tasmania and the territories, while WA alone had a modest rise of 13%. Of the candidates, 28% were female while 35% of those elected were female, including three each in WA and Tasmania.

 

There were reductions in support for Labor (around 5%), the Greens (over 4%) and the Coalition (about 1%), with the primary beneficiaries being the Liberal Democrats (3.9%, more than double their previous support level) and the new Palmer United Party, which secured places in Tasmania, Queensland and perhaps Western Australia on a 4.9 % nationwide vote.

 

The combined Labor and Coalition Senate vote has declined from around four-fifths in both 2004 and 2007 to under three-quarters in 2010 and just over two-thirds in 2013. With the quota at just over one-seventh, it is not surprising that other candidates won two places in two states, and won a record three in South Australia where the combined Labor and Liberal first preference vote was only 50.11%, the lowest since proportional representation was introduced in 1948. Independent Senator Nick Xenophon took nearly 25% of the vote, just under the Liberal vote and ahead of Labor. As few preferences flowed to his column and therefore his running mate, the Greens and Family First secured election after starting with respectively 7.1% and 3.8% of first preferences. The Proportional Representation Society of Australia is delighted that Senator Xenophon has foreshadowed legislation to abolish party boxes and to institute a form of optional preferential voting instead.

 

The table below illustrates how nationally there was quite close correspondence between first preferences and senatorial places won by the Coalition, Labor and the Greens and that, with the exception of the Palmer United Party, other parties or candidates obtained less representation than their level of support indicated.

 


Coalition

Labor

Greens

Palmer

Others

Votes

37.2

30.1

8.6

4.9

19.1

Seats

42.5

32.5

7.5

7.5

10.0

 

Much attention was focused on ‘deals’ made or agreements not adhered to when Group Voting Tickets were not lodged for four related parties in Victoria. In any ‘deal’ between two parties, each hopes to benefit from a transfer, usually at the exclusion of the other. There is normally no way to know in advance which will be excluded first, as micro-parties are increasingly holding their votes within a tight grouping not necessarily consistent with policy directions, in the hope that at least one will claw a way to a quota.

 

Vocal complaints were made - some only following disclosures after the election - about persons or small groups forming multiple parties, agreed financial and other support arrangements, failure to lodge Group Voting Tickets as promised, misleading advertising about GVTs and their possible or likely impact, and pressure to participate in tight networks of preference ordering, orchestrated particularly by Glenn Druery.

 

Examination of the ABC calculator treating all votes as above-the-line alerted the nation early to the possibility of senators being elected even though their group started with much fewer than 1% of first preferences. The full compilation of WA votes and the extent of Tasmanian below-the-line marking illustrated the broad indicative nature of such modelling approaches, as a narrow exclusion at a critical juncture halted the projected progress of some candidates that did extremely well from a low base. For instance, there was a reversal of about 1,000 votes in the relative positions of the successful Palmer United Party candidate and the Australian Sex Party candidate that the calculator showed for Tasmania’s last place.

 

In NSW, it immediately became evident that the leftmost column position of the Liberal Democratic Party saw it gain many votes of its 9.5% intended for the Liberal National grouping near the middle of 44 group columns. The party also benefited greatly from a more favourable position on the ballot paper than the Coalition’s in Tasmania, South Australia and Western Australia, obtaining between 2.3% and 3.5% of first preferences, whereas it achieved just 0.7% in

Queensland where the Liberal National Party column came much earlier. In Victoria, it failed to register a Group Voting Ticket, so its position second from the left was of no practical benefit.

 

Patterns of voting above- and below-the-line suggest that the Labor and Coalition vote was slightly inflated by unsuccessful attempts by those supporting other parties to vote formally below-the-line, as the attrition rate has been found to be very high in studies after previous elections. In mainland states, 97-99% of those voters voted above-the-line, but for Greens supporters that range was 82-92%, and for smaller parties or groupings it was 92-97%, though there were several parties consistently below even 80%.

 

The table below sets out some of these major differences in each state and territory for parties or groups where Group Voting Tickets were lodged. In Victoria, South Australia and Tasmania, non-lodgement of Group Voting Tickets meant that some supporters could only use the below-the-line option.

 

Use of above-the-line party boxes, by State and Territory (%)

 

 

Total

Coalition

Labor

Green

Others

Lowest

NSW

97.9

99.3

98.7

91.9

96.9

77.0

Vic

97.3

99.3

98.8

90.8

94.0

75.1

Qld

97.0

98.8

98.0

88.4

94.9

70.4

WA

96.2

98.8

97.7

88.5

93.6

67.4

SA

93.5

97.5

95.9

81.7

91.6

56.8

Tas

89.7

93.4

91.9

73.0

88.4

57.7

ACT

80.1

90.9

84.3

62.3

68.3

44.5

NT

91.9

95.2

93.8

78.5

87.1

78.5

 

Four vacancies were filled at the start of the scrutiny in four states and three in each of SA and WA.

 

In New South Wales, a Liberal Democrat, David Leyonhjelm, was the fifth elected, on Australian Sex Party preferences, while a Liberal, Arthur Sinodinos, was the last, picking up large transfers from the Palmer United Party and major components from the Australian Sporting Party, which increased its total from 55,000 to 285,000 (0.4 quotas) at exclusion.

 

In Victoria, timely boosts saw the progress total of Rick Muir of the Australian Motoring Enthusiast Party rise from 17,000 first preferences to above a quota (483,076), with moves of 160,000 from the Palmer United Party (on whose Group Voting Ticket he was a very early preference) and 140,000 upon the partial exclusion of the Australian Sex Party. Janet Rice, a Green, was fifth elected, on Labor preferences. Her surplus put the Sex Party just above Palmer United for the first time. After the initial surpluses were dealt with, the progress total of the third Coalition incumbent, Helen Kroger, rose by only 50,000.

 

In Queensland, the third LNP candidate started just 40,000 votes short of a quota (374,209), the Palmer United Party, headed by former rugby league star Glenn Lazarus, began with 285,000 votes and the Greens with 155,000. Mr Lazarus picked up over 35,000 votes when the Australian Fishing and Lifestyle Party was excluded and then over 100,000 votes from Katter’s Australian Party to be elected. The exclusion of the first Help End Marijuana Prohibition Party candidate saw one-third of his 180,000 votes place the Coalition well past its third quota.

 

Western Australia received much attention when the possibility arose of the Australian Sports Party climbing from 2,000 votes to achieve the fifth quota (187,349). Fourteen votes separated the Australian Christians and the Shooters and Fishers at a critical juncture in the scrutiny on which the last two places would turn. Palmer United’s Zhenya Wang was just 2,539 votes ahead of the last Liberal Democrat before picking up 55,000 votes from his exclusion and then 65,000 from the Shooters and Fishers to be elected. The distribution of his surplus under the flawed unweighted transfer procedure placed Labor’s Senator Louise Pratt 1,400 votes above a quota while Senator Scott Ludlam, of the Greens, was 2,000 below. Were the High Court to severely strike out provisions surrounding lodgement of multiple voting tickets, a different order of exclusion would arise. A recount of above-the-line and informal votes in Western Australia has been ordered.

 

In South Australia, the second Liberal, Senator Simon Birmingham, was just 12,000 short of a quota (148,348) when exclusions began. Family First’s Bob Day remained 4,000 ahead of the last Liberal Democrat when one of them had to be excluded, and almost doubled his progress total at that point. The exclusion of the Labor powerbroker, Senator Don Farrell, propelled a Green, Senator Sarah Hanson-Young, above a quota, after an earlier 26,000 boost from Palmer United. Her surplus went mainly to Bob Day and his largely went to elect the second Liberal.

 

In Tasmania (quota 48,137), Palmer United’s Jacqui Lambie started 7,000 in front of the third Liberal and 9,000 behind the first Green after four surpluses were dealt with. The Australian Sex Party was just 244 behind Labor’s Senator Lin Thorp when one had to be excluded. Green Senator Peter Whish-Wilson was re-elected on her exclusion and his surplus put Palmer United 1,276 ahead of the Liberal Democrat, whose progress total had nearly quadrupled to approach 30,000 votes, and able to secure election. The Liberals only advanced by 11,000 votes as exclusions occurred.

 

In the ACT, the ALP’s Senator Kate Lundy was 2,000 votes above a quota while the combined Liberal vote was 700 below. The high level of below-the-line votes saw Zed Seselja easily above a quota with the lead Bullet Train and Green candidates still unexcluded.

 

In the Northern Territory, the combined Labor vote was just 400 short of a quota (34,494) while the Country Liberals were well above.

 

Genuine reform will involve clearly identifying the problem and tackling it in the simplest manner possible. Most voters have no idea of the numbering in Group Voting Tickets that they feel dragooned into endorsing because of the imposition surrounding below-the-line formality. Party boxes would not have been introduced had Dr H V Evatt paid more attention to suggestions from Liberal MPs about formality requirements when proportional representation was being introduced for the Senate in 1948, instead of stubbornly persisting with full compulsory marking.

 

Letting a party or group register two or three Group Voting Tickets without giving voters an opportunity to indicate which they endorse is on sufficiently doubtful constitutional ground for Section 272(5) of the Commonwealth Electoral Act 1918 to have a savings provision trying to at least preserve the numbering common to all tickets in such circumstances.

 

There is simply no need for party boxes if voters can readily cast a formal ballot based on the way they assess the candidates and policies. The experience in Eire and Malta over many decades, and in the ACT since 1995, with a single first preference sufficing has been perfectly satisfactory, just as it has been when a number of preferences not exceeding the number of vacancies have been required in Tasmania, NSW and Victoria, and for various municipal polls.

 

Proposals to impose an exclusionary threshold usually betray either misunderstanding of how the single transferable vote works, or seem aimed at propping up the failed system of party boxes, without recognizing the potential for unstable or distorted results to arise whenever arbitrary interventions are made.

 

The process of election and exclusion is soundly defined except for transfers of surpluses. If arbitrary thresholds apply, whether parties or candidates fall just short or a little above may hold the key to more than one place. Depending on the rules imposed, large numbers of votes could be wasted or directed to a candidate who would not otherwise have benefited.

 

In 2013, parties starting with percentages around 11.7, 10.8, 9.9, 9.5, 7.1, 6.6, 5.0, 3.8 and 0.5 all had senators elected because a candidate attracted sufficient further votes to exceed a quota. By what principle should any of them be ruled ineligible at the outset? Would a different threshold apply at a double dissolution poll?

 

Smaller parties and independents collectively obtain less representation than their aggregate support levels indicate. Were larger parties to endorse better candidates and articulate more attractive policies, their additional votes would translate into more senators.

 

In Tasmania and the Australian Capital Territory, where Robson Rotation within group columns applies, candidates without half a quota of strong support rarely succeed. In Eire and Malta, where casting a formal vote is straightforward and no attempt is made to get party supporters to follow a particular numbering, candidates with few first preferences get excluded early in the scrutiny. It is only where party boxes exist and draconian formality requirements make below-the-line voting almost an imposition that candidates with few first preferences can hope to cobble together a quota on above-the-line transfers about whose possibility very few voters are aware.


Nearly 16% of votes wasted in German election


 

While two parties obtained more than 5% of the party list vote in respectively six and seven of the sixteen states in the German elections held on 22 September 2013, they just failed to do so nationally, and therefore did not qualify for the apportionment of seats to the Bundestag. The bickering Free Democrats (4.8%), who were in coalition with Dr Angela Merkel’s Christian Democratic Union, but unable to deliver tax-cutting promises, lost two-thirds of their previous record support (14.6%), and their unbroken Bundestag presence since 1949. The new eurosceptic Alternative for Germany (4.7%) had its lowest support in some of the larger western states, and also just fell short of the arbitrary 5% national exclusionary threshold. That meant a much higher bonus for each of the four parties or groups with aggregate 84.3% support that succeeded.

 

With some 41.5% of the determining party list vote, the CDU (including its separate Bavarian CSU partner) obtained 311 of the 630 seats. The Social Democrats won 30.5% of the seats for their 25.7% support, while the Greens finished with 10% for 8.4% support and The Left, with support concentrated in eastern regions (where they vie for or achieve second place), gained 10.7% arising from 8.6% of the vote. Turnout was 71.5%. The Chancellor, Dr Merkel, immediately set about the negotiations that will be necessary to form a new coalition government.

 

Had the threshold been set instead at 4%, both the Free Democrats and Alternative for Germany would have emerged with more than thirty party list seats surrendered by those with an actual augmented incoming Bundestag presence, and wasted votes, beginning with the Pirate Party at 2.2%, would have been limited to 6.2%.

 

German voters have both a constituency and party list ballot of which respectively 1.6% and 1.3% were invalid. Parties winning more constituency seats than their statewide proportional entitlement using ordinary d’Hondt divisors keep the extra “overhang” seats. At the second attempt, following Federal Constitutional Court rulings that such overhang seats could not be nullified by the workings of the separate national process of party list allocation, acceptable legislation was passed to ensure that the benefit of statewide overhang seats always remained.

 

The latest outcome is a salutary reminder of the very unstable role that an arbitrary exclusionary threshold tends to play in any scheme of seat allocation.

 

Call for Nominations for Elections of the four PRSA Office-bearers for 2014-15

 
The Returning Officer is Mr Patrick Lesslie of the PRSA's NSW Branch. Under the PRSA Constitution, the Returning Officer rotates among the Branch Secretaries. The order, by precedent, is VIC-TAS, NSW, SA, WA, and the ACT.


Nominations, for President, Vice-President, Secretary and Treasurer, need be signed by the candidate only, as consent to nomination, and must be with Mr Lesslie, at 4/64 Grosvenor Crescent SUMMER HILL NSW 2130 or patrick@lesslie.com.au by Thursday, 14 November 2013.


 

© 2013 Proportional Representation Society of Australia

National President: Bogey Musidlak 14 Strzelecki Cr. NARRABUNDAH 2604

Editor, Quota Notes: Geoffrey Goode 18 Anita St. BEAUMARIS 3193

Tel: (02) 6295 8137, (03) 9589 1802 Mobile 04291 76725 quota@prsa.org.au