As of yesterday, the results of the 2022 federal election have been finalised and move to the results archive. One of the last pieces of data to be compiled is a breakdown of two-party-preferred preference flows for each party.
The two-party-preferred vote, by definition, splits up every vote between Coalition and Labor. Since voters are required to number every box, all votes end up in one pile or the other.
Labor ended up winning 52.13% of the two-party-preferred vote. But for today’s blog post I’m most interested in the flows of preferences from every other party. You can see the raw numbers and percentages (calculated as a share of the minor party’s total vote) on this page. The AEC has also published the same numbers by state, but we don’t have breakdowns at the seat level in the non-classic seats where the 2CP and 2PP are not the same.
Over 85% of Greens voters preferenced Labor, which is the highest of any party on the list. Kevin Bonham has conveniently compared the preference flow rates for the largest minor parties in 2019 and 2022:
Key preference flows (2019 in brackets)
Green 85.66 to ALP (record high) (82.21)
UAP 61.86 to L-NP (65.14)
PHON 64.30 to L-NP (65.22)
IND 63.77 to ALP (59.40)— Kevin Bonham (@kevinbonham) July 21, 2022
Labor improved its flow of preferences from the Greens, United Australia, One Nation and independents.
There’s a lively debate about the value of 2PP stats in our multi-party system.
On the one hand, major parties will often claim mandates based on winning the two-party-preferred vote, even though their primary vote didn’t come close to a majority. Polling can be overly focused on this single number, boiling politics down to a simple Labor-Coalition contest which minimises the role of other political players.
But I don’t think the 2PP lacks value entirely. While there are more seats than ever that are non-classic (where the top two candidates are not Labor and Coalition), about five sixths of all seats remain 2PP contests, and calculations of uniform swing continue to do a reasonably good job of predicting how many seats will change hands between the major parties, even if they are completely blind to contests involving independents or minor parties.
In particular, I think the 2PP does a good job of identifying which major party has more of a claim to leading the government, even if it doesn’t really provide a valid basis for a “mandate” for a monopoly over government power.
We don’t directly elect a prime minister in this country, but the 2PP can be seen as an indication of whether voters prefer a Labor-led government with a Labor prime minister, or a Coalition-led government with a Coalition prime minister. Even if hung parliaments are more likely than ever, we are a very long way away from any other political forces taking a leading role in government.
At this point, with the major party vote dropping below 70%, there’s no denying that we no longer have a simple two-party system, but we have more of a multi-party system. It’s common in multi-party politics that parties group themselves into “camps”. Some political systems which use proportional representation have seen pretty much every political party sort themselves into one of two camps, so that you know that, for example, a vote for a Green candidate will contribute to a Labour-led government. In other countries like France, which uses a system of single-member electorates elected over two rounds of voting, the parties tend to cluster into blocs which compete with each other but ultimately cooperate against parties outside their bloc.
We don’t have those same kinds of alliance building in Australia (the role of preference deals in the House of Representatives is massively exaggerated, and the major parties generally form government on their own), but in a sense the voters do it themselves. Labor has constructed a two-party-preferred majority which includes their own primary vote but includes about one fifth of all voters who preferenced Labor after voting for someone else.
If we were in the US, you would refer to the bloc of voters who deliver victory to a party as their “coalition”, but that word has a different context in Australia. But the point of this blog post is to analyse the Coalition of voters who delivered Labor a two-party-preferred majority.
Labor didn’t win the election just off the back of the 32.58% of voters who gave them a first preference. They won due to a combination of those voters, and a further 19.55% of voters who preferenced Labor over the Coalition.
This next chart shows primary votes for the bigger parties split between those that preferenced the Coalition higher, or Labor.
About 74.6% of the Coalition's two-party-preferred vote came from primary votes for the Coalition, but only 62.5% of Labor's base came from Labor primary votes.
This is interesting in the context of the current debate around climate targets, although you could also apply this argument to any of the other issues coming up. Labor claims a mandate for their policies, while the Greens insist on further negotiation.
10.5% of all votes were Greens primary votes which eventually flowed to Labor.
Unfortunately I can't break down the independent vote between those who can be classified as "teal", but I previously estimated that about 70% of all independent votes were cast for teal candidates (including votes for Rebekha Sharkie, the sole Centre Alliance candidate who clearly fits in the teal category). About 3.5% of all votes were independent primary votes that flowed to Labor, which translates to about 2.5% as teal votes that flowed to Labor.
If you combine that 10.5% Greens voters, and 2.5% teal voters, that's at least 13% of the vote that flowed to Labor from candidates with more ambitious climate policies. That's almost a quarter of Labor's two-party-preferred vote.
So Labor's primary vote makes up about five eighths of the two-party-preferred vote, with the Greens and teals contributing another two eighths. That's a crucial part of the puzzle, but doesn't bring Labor to a majority. So what is in the remaining one eighth of Labor's two-party-preferred coalition? About 6.5% of the total national vote is made up of other votes that flowed to Labor.
Just over half of those remaining votes came from the larger minor parties of the right. One Nation, United Australia and the Liberal Democrats contributed 6.98% to the Coalition, but also 3.83% to Labor.
There are also a bunch of left-wing minor parties who contributed to Labor's total. Animal Justice contributed 0.38%, and there was a further 0.42% contributed from the Victorian Socialists, Fusion, Socialist Alliance, Indigenous-Aboriginal Party, The New Liberals, Legalise Cannabis, the Local Party, Sustainable Australia, Democratic Alliance, Reason and the Progressives.
About 0.03% of the total national formal vote were primary votes for the lower-polling Liberal or Nationals candidate where the parties ran against each other and flowed to Labor.
So overall, the coalition that delivered Labor its majority has a great deal of diversity (even before considering the diversity of thought within the 32.6% who voted 1 Labor). About a quarter of the vote came from the Greens and the teals, but outside of those groups Labor doesn't gain many preferences from the left. A small part of Labor's preferences come from the minor right-wing parties, but that is enough to make the difference between a two-party-preferred minority and majority.
