Does AV mean higher or lower effective number of parties?

There may be a conventional wisdom among people who study comparative electoral systems that the Alternative Vote (also known as Instant Runoff or Majority Preferential) tends to suppress the effective number of parties, compared to plurality (First Past the Post, or FPTP). Or maybe it is just me, but I will admit to having such a notion. After all, Australia is a pretty strict two-party system, isn’t it?

The correct way to approach the question of whether AV means a higher or lower effective number of parties (N) than FPTP is to ask: What we should expect N to be, given the country’s seat product?

As explained by Taagepera (2007) and further elaborated and tested by Li and Shugart (2016), the seat product is a country’s mean district magnitude (M), times its assembly size (S). The Seat Product Model says that the effective number of seat-winning parties (Ns) tends to be the sixth root of this product: Ns=(MS)1/6.

The model is logical, not a mere product of empirical regression work, although regression tests confirm it almost precisely (Li and Shugart, 2016).

When all districts elect just one member, thus M=1, the Seat Product is just the assembly size, S. Hence we take the sixth root of S to get an expectation for Ns. What if we do this for Australia’s House of Representatives? We get an expectation of 2.31.

The actual Ns for Australia’s elections since 1984, the year S was increased from 125 to 148 (subsequently it has increased to 150, a minor change) is… 2.53. However, I believe that figure (I am using Gallagher’s Election Indices) treats the Coalition parties as one in elections before 2010.

In the two most recent elections, Ns has been 2.92 and 3.23. The notes to Gallagher’s Election Indices indicate that for these elections the Liberal Party, the Nationals, and the Liberal National Party of Queensland are treated as separate parties. In my opinion they should be so treated, although I suppose one could have a debate about that.

The actual mean is thus above the expectation for a hypothetical FPTP of the same size assembly. If we use the figure of 2.53, it is obviously not much higher than 2.31 (the ratio is 1.10). However, if we consider the value, at least in recent elections, to be around 3.0, it is about 1.30 times the expectation value.

Contrast this with the UK, where elections of the same period (1987-2010) have a mean Ns=2.30. This is just what we expect for FPTP, right? Not much over 2.0. Not so fast! The UK has a huge assembly, and with S=650 (aprpox., as it varies over the period), we should expect Ns=2.94. The UK actually has one of the more under-fragmented assemblies, according to the Seat Product Model, with this recent-period average being only 78% of expectation.

So how about Canada, where AV is one of the potential reforms being considered? Over a similar period (1984-2011) we get Ns=2.63. With S around 300 during this time, we should get Ns=2.59. So Canada pretty much nails the expectation of the model.

So, should we expect Ns to go down if Canada were to adopt AV, as (what I characterized as) the conventional wisdom would have it? Or should we expect it to go up?

I would not be inclined to say ‘down’. I will just leave it at that for now.

New posts on Australia (and environs)

Four new posts are up today, each of them by either Alan or Henry. They are part of a series that begins now on the upcoming election in Australia (plus an update on Bougainville).

So scroll down, and enjoy. Thank you, Alan and Henry, for being F&V’s correspondents Down Under!

Has South Australia’s two-party system been Nicked?

As most of my readers will know, Australia’s federal election campaign has begun, with polling day scheduled for July 2. The campaign has been rather uneventful so far, but there is one interesting event that could take place in South Australia.

Nick Xenophon is a Senator from South Australia. His political career started when he was elected to the South Australian Legislative Council (which is elected eleven members every three years, using the single transferable vote) as a candidate opposed to ‘pokies’ (slot machines). He received 2.86% of the vote, which represented only about a third of a quota; nonetheless, he received group ticket preferences from seven other parties, and was elected.

During his six-year term, he developed substantial popularity in the state, and when his seat came up for election in 2006, his ‘Independent Nick Xenophon’ ticket received 20.6% of the vote for the lower house, only six points behind the Liberal Party. This allowed him to be elected, as well as his running mate Ann Bressington.

Following this result, he decided to run for the Australian Senate in 2007, as an independent candidate. He won a seat with 14.8% of the primary vote: just above a quota.

The results of that election saw the Labor Party win a majority in the House of Representatives, and form government. However, the party was relatively weak in the Senate; they had only 32 seats, to 37 for the Liberal/National coalition. With 39 votes needed to pass legislation, this meant that Xenophon shared the balance of power with five Green Senators and one member of the religious conservative Family First Party.

This meant that Senator Xenophon was in a position to heavily influence government legislation. And, as a Senator without a particularly strong ideological affiliation, he used this power to direct Federal funding to South Australian projects; for example, when the government needed to pass an economic stimulus bill, Senator Xenophon blocked it at first, in order to increase funding for conservation of the Murray-Darling River.

Following the 2010 election, the Labor Party lost its majority in the House of Representatives, but was able to come to agreements with a number of independent members of the House to hold onto a majority. However, in the Senate, Labor and the Greens together had a majority, allowing the government to avoid negotiation with the other Senators.

Despite this reduction in his influence, Xenophon was easily re-elected in 2013, securing 24.9% of the vote, narrowly beating the Labor Party. Such a vote put him well on the way to a second seat; however, most parties (including Labor and the Greens) submitted group tickets against his running mate, allowing Family First’s Bob Day to be elected off 3.76% of the primary vote.

Much has changed in the Senate since then. The 2013 election brought the Liberal/National coalition to power, but the government only won 33 seats in the Senate, meaning that they needed six other Senators to pass a bill. Assuming Labor and Green opposition to a bill, this meant that Xenophon and two other Senators could block a bill.

During this parliamentary term, Xenophon announced he would form a national political party. While he had registered a party before his 2013 election called the ‘Nick Xenophon Group’ (in order to have his name appear above the line on the ballot paper) it only contested the Senate in South Australia. In 2014, the party was turned into the Nick Xenophon Group, which would contest elections outside South Australia and without Xenophon as a candidate.

The government also made substantial changes to the electoral law towards the end of this parliamentary term. The electoral system for the Senate was changed from the single transferable vote with compulsory preferences, where voters could either number every single candidate, or vote for a party’s ticket of preferences. Large numbers of candidates mean that most voters voted for a party’s ticket of preferences. This system meant that small parties were able to swap preferences to other small parties with very strong preference flows, allowing candidates to win with very small shares of the vote. Dr Kevin Bonham sums up the issues with the system very well here.

The system was replaced early this year, following an agreement between the Liberal/National government, the Greens and Xenophon. It was replaced by an optional preferential system, where voters can either vote preferentially above the line (with a preference for a party representing a vote for the candidates of that party in order) or for individual candidates below the line.

Shortly following this change, the government announced that they would recall Parliament in order to put several industrial relations bills to a vote. This was the second time these bills had been put to the Senate (they had been rejected the first time), and if they were rejected again, the government would have a trigger for a double dissolution (an election for all members of both houses of Parliament). As it turned out, the bill was blocked, and a double dissolution was called.

The Xenophon Team has been announcing candidates for selected House of Representatives seats, and will run a Senate ticket in every state. The fact that this election is a double dissolution will have a number of advantages for this party.

First of all, the threshold for election in the rest of the country will be halved, as twelve seats will be up for election. This will make it easier for minor parties of all sorts to be elected, and Xenophon’s party is minor outside of South Australia.

Second of all, and more importantly, it will allow Xenophon himself to be attached to the party’s Senate ticket in South Australia for all twelve Senators, rather than just six. Why is this important?

Well, when Xenophon was re-elected to the South Australian Legislative Council in 2006, his ticket won two seats. While his original running mate turned out to be a disaster, the replacement for his seat (John Darley) ran for re-election in the 2014 election as the “Independent Nick Xenophon Team”. He received only 13% of the primary vote, but held his seat.

From that result, I don’t think it’s unreasonable to say that there is a dropoff in support for Xenophon-affiliated candidates when Xenophon himself is not on the ballot, and thus him being on the ballot for all Senators could boost the party’s representation somewhat. While this may be extrapolating too much, I also think that this dropoff makes it unlikely that the Xenophon Team, despite optimistic polls (I suspect voters may be confused between the Senate and the House, but to be honest this is mostly speculation).

The other interesting thing about the Xenophon Team is that it has the possibility to lead to some very interesting results. While his House candidates are likely to be irrelevant outside of South Australia, his Senate support in the state, as well as strong polling, suggests that his party could be competitive in House seats.

The most positive individual seat polling for the Xenophon Team took place before the Liberal Party replaced Tony Abbott as leader with Malcolm Turnbull. Seat polling commissioned by trade unions a year ago suggested that a Xenophon Team candidate could win Education Minister Christopher Pyne’s seat. However, more recent polling has been somewhat less optimistic.

What will be interesting is that we will get a better picture of how preferential voting works in a multi-party environment. The importance of finishing order will be substantial, and there could be opportunities for tactical voting.

In most of the seats the Xenophon Team is targeting, the Liberal Party will almost certainly finish first in the primary vote. It is who finishes second that will be important. If, after all the other candidates are excluded, Labor is second, then the Liberals will have a good chance of victory. While the Xenophon Team is still undecided about how it will distribute their preferences, one key candidate has said that they will not file a specific how-to-vote card, and it seems likely that voters for the Xenophon Team will be more centrist.

If the Xenophon Team is second, however, then all bets are off. Labor has not yet announced its preference decisions in South Australia; however, I think it likely that they would not be too unhappy to see the defeat of a prominent Liberal MP.

Tactically speaking, it would be sensible for Labor to run a minor campaign in those seats, to ensure that the Xenophon Team finishes second on primary votes. On the other hand, the Liberals might want to surreptitiously encourage people to vote Labor (although such a campaign has not, to the best of my knowledge, been tried in Australia before).

How preferential voting works in a three-way race will become more relevant if Canada adopts it, as part of the Liberal electoral reform process. The results in South Australia, and the campaigns in the districts, might give a clue as to whether it works effectively in a multi-party environment.

the Emerald City and the sticks

Sydney has been known as the Emerald City since David Williamson’s play was first performed in 1987. Politics in Sydney are quite different from what happens ‘out in the sticks’ in the rest of the state.

New South Wales is the largest state by population and by GDP, the second by per capita income, and the third largest by area. (Both territories have higher per capita incomes) The only sister state relationship in the US is with California. Americans could think of the place as New Yorkifornia with Sydney playing many of the roles that New York, Los Angeles and San Francisco play in the US.

Apart from 1961, no-one has ever won a federal election without winning a majority of seats in NSW. The state elects 47 of the 150 MHRs and 12 of the 76 senators. Population shifts over the last few decades have tended to transfer 1 seat to either Queensland or Western Australia at each re-apportionment. The state has provided 6 of the last 10 prime ministers. ABC Vote Compass has thoughtfully posted a list of the most left-leaning and right-wing electorates in the country. NSW has 3 of the most left-leaning and none of the most right-leaning seats.

In the 1920s NSW briefly adopted STVPR. This led to a hung parliament where the Industrial Socialist Labor Party (an organisation close to the International Workers of the World) elected a single MLA who held the balance of power. STVPR was abolished in 1926, in reaction to the horror of a Wobbly balance of power.

Australia 2016

Australia will elect all members of the house of representatives and the senate on 2 July. This is an extended campaign by Australian standards. The timing was dictated by a series of interlocking provisions in the constitution relating to double dissolutions.

Australia has a general election on 2 July for all seats in the house of representatives and the senate. The constitutional reasons for this double dissolution are canvassed in the advice tendered to and accepted by the governor-general which has been published. Sidebar: you often read British and Canadian authors who argue that conversations and advice between head of government and head of state must be private but Australian governors have been publishing ministerial advice and their response on major issues since 1975. I am not usually an enthusiast for Sir John Kerr, but in this at least he established a good precedent.

The headline view is that Labor appears to be gaining, but it is by no means clear that they will gain enough seats for a majority government. It is equally unclear who would be best placed to form a minority government in the event of a deliberative, balanced or ‘hung’ parliament. The prime minister’s personal standings have been tanking for some time but that appears to have levelled off. The opposition leader’s standings have been rising significantly. Newspoll, widely regarded as definitive, has a snapshot of how they see the situation here. Historically pollsters have relied on second preferences from the previous election to estimate preference flows because that tends to give a more accurate result than asking poll respondents.

Here is the Gazetted enrolment (30 April 2016) published by the AEC.

NSW 107 323

VIC 105 929

QLD 101 302

WA 97 228

SA 106 398

TAS 74 009

ACT 139 025

NT 65 444

The dramatic gap between the ACT and NT is a rounding error, the NT is just over 1.5 of the uniform quota and the ACT is just under 2.5. The low figure for Tasmania is a consequence of their minimum representation of 5 in the house. In my view a variation from 139 025 to 65 444 is not really acceptable in the 21st century.

The election will probably be decided in New South Wales and Queensland. Labor did quite well in Victoria in 2013, against the national trend, and has probably peaked at 19 out of 37 seats. The nest result for Labor in the snaller states and the territories would be gains of 1 seat each in South Australia, Tasmania and perhaps the Northern Territory. Western Australia is showing a marked swing to Labor and as many as 4 seats could change hands, Labor would be looking at a net gain of 6 outside NSW and Queensland. The Newspoll page has a convenient calculator that lets you estimate seat gains for particular swings.

That takes us to the battleground states with 47 seats in New South Wales and 30 in Queensland.

The magic number to form a majority government is 76. From the Australian Financial Review:

If Labor wins the July 2 election it will become the first opposition in 85 years to regain government after just one term. To do so, it needs to win a net minimum of 19 seats.

The Coalition, to avoid becoming the first government to lose power after one term since the Scullin Labor government lost in 1931, needs to lose fewer than 15 seats.

The situation is complicated by minor party and independents who may (again) be called on to decide who forms the government. I plan to do separate posts for the situations in New South Wales, Queensland and the minor parties and independents. My guess is that there will be a minority government but it is too early to say who will form government.

Labor is making loud noises about never again making a confidence and supply agreement with the Greens. Both major parties seem to think the solution to a deliberative parliament is to call a second election.

They made similar noises in Tasmania in 2010 and were rebuffed by the governor. It is hard to imagine the governor-general granting a second election for the exact some reasons that the governor of Tasmania refused to grant one in 2010.

 

 

 

 

 

 

Australia’s impending double dissolution

The latest news from Australia:

Electoral reform abolishing Group Voting Tickets and establishing partially optional preferential voting ‘above’ and ‘below’ the line was passed in both houses last week, and barring an unlikely High Court decision to the contrary, it will go into effect at the next election. Now that the electoral system is no longer an obstacle (and perhaps due to the electoral reform, which aroused the ire of most Senate crossbenchers), Prime Minister Turnbull has all but called the expected double dissolution election, threatening to do so if controversial industrial relations bills do not pass the Senate at the next session, scheduled to start on April 19th.

A double dissolution is the deadlock-breaking mechanism provided by Australia’s Constitution. Though half the Senate is usually elected alongside elections to the House of Representatives, the Senate has a fixed six-year term. This can only be shortened by a double dissolution election where all seats of both houses are up for election. A double dissolution can only be brought about by the government when triggered by a disagreement between the houses, as spelled out by section 57 of the Constitution; if the disagreement with regards to a bill continues after the double dissolution election, those bills can be put to a joint sitting of the houses, where the government is likely to prevail due to the houses’ relative numbers.

Although section 57 gives the government the power to threaten to dissolve the Senate if it does not pass its legislation, there are various factors that complicate this procedure, making it rather cumbersome for the government. There is, of course, also the risk of losing the election. The procedure has only ever been used six times, and the last double dissolution election took place in 1987.