There’s been a lot of chatter online in the last few days challenging calculations about the low primary vote for the newly-elected Labor government, not by disputing that figure, but by arguing that it is higher than the vote for the Liberal Party. It’s been all over Twitter from some of the more ferocious ALP supporters. This has been a faulty logic, because it doesn’t include the vote for the other Coalition parties.
After trying to explain why this is the correct way to calculate these totals, I thought I would put my thoughts down here. When does it make sense to think of the Coalition as a single major party, as two parties, or even as four parties?
The Coalition was formed by the Nationalist and Country parties in the 1920s, so it’s been around for about a century, but it’s changed over time – in the early days, it was common that the two parties would run candidates for the same seat, but it’s now very rare.
The relationship between the parties varies from state to state. There is a sizeable Nationals party in both New South Wales and Victoria. In Queensland and the Northern Territory the parties have actually merged (as the Liberal National Party and the Country Liberal Party respectively). The is no Nationals party in Tasmania or the ACT, and they are very small in South Australia and have no MPs. The WA Nationals are actually the larger party in the state parliament but are a minor factor in federal politics, only running in one seat in 2022.
So there are actually four federally registered parties who are part of the Coalition: Liberal, Nationals, Liberal National Party and Country Liberal Party. LNP and CLP members elected to the federal parliament are designated to sit with one of the two coalition partners, which is how it’s possible that the leaders of both federal parties are members of the Queensland LNP.
There are a few different ways you can analyse the Coalition partners.
If you are comparing results to Labor, you need to look at the whole package. The first reason for this is that no single Coalition party runs in every seat, whereas Labor runs everywhere. If you just look at each individual party’s share of the national vote, you are looking at a vote diluted by many seats where that party received zero votes.
The second factor is that the Coalition parties don’t really run against each other these days. Out of 151 electorates, there was only contests in four seats. In two of these (Barker in SA and Durack in WA) they were contested by local Nationals who aren’t really proper parts of the Coalition, but there was also contests in the Victorian seats of Indi and Nicholls where there was no incumbent Coalition MP. In the other 147 seats, a Coalition voter only gets one option. They might prefer Liberal, but if the Nationals are running in that seat, that’s all the choice they get.
So when it comes to analysing vote patterns, it’s silly to try and separate out the Liberals and Nationals. Most of the time, there’s just one option.
This table gives a sense of how much of the country each party contests, but also how the Labor vs Coalition primary vote varies between the different zones. While the Liberal Party contests about two thirds of all seats, that includes a lot of Labor seats. The LNP is stronger in Queensland, and the Nationals significantly outpoll Labor in their seats. Indeed the Nationals (excluding the LNP) hold most of the seats they contest. They just aren’t a presence in too many marginal seats, which means they become a more significant part of the Coalition in opposition but don’t have much prospect to win back seats.
ALP prim %
L/N prim %
% of votes
# of seats
Lib vs Nat
Of course, that’s not to say the differences within the Coalition don’t matter. They clearly operate in different ways within federal parliament, and use different branding that could make a difference to voters. But when you are aggregating votes to the national level and analysing broader trends, separating out the parties is a distraction.
So there are two purposes I can see for analysing the Coalition’s national primary vote which require it to be treated as a single unit.
Firstly, looking at the national level of support. One Nation, Labor, United Australia and the Greens all ran in most if not all electorates. Comparing them to separate parts of the Coalition is misleading since those Coalition parties are not national in the same way.
Secondly, when analysing the level of support for a government. When I analysed how the declining major party vote meant that the primary vote for each majority government has been shrinking, I needed to look at the whole Coalition. The Liberal Party doesn’t win government on its own, so you need to include all of the votes for all of those parties.
I have been critical of how our electoral system has translated a 33% primary vote into a majority of seats and thus a single party majority government. I had the same criticism for recent Coalition governments too, and I think the solution is proportional representation. But there has been some more opportunistic and hypocritical chatter from the right trying to argue that it’s somehow unfair for Labor to win an election while the Coalition has polled about 3% more on the primary vote, and this has sometimes tied into an argument that First Past the Post would be a fairer system. This argument is nonsense: neither party should be winning sole power on the votes they’re getting now, it’s not just a criticism of Labor.
Some have tried to suggest that the Coalition should fairly be compared to Labor and the Greens, but I don’t think that makes sense. Labor and the Greens run against each other across the country, and have completely separate identities. They also don’t have any kind of coalition partnership, let alone the kind of permanent alliance that exists on the right wing.
And Labor is winning government in its own right, without Greens seats, which is not how things work for the Liberal Party, whether you mean just the party that exists distinct from the Nationals in five states and a territory, or if you also include the Liberals in Queensland. It’s not Labor and the Greens forming government, it’s just Labor on their own.
If people want to propose a different method of analysis, they need to justify why it makes sense. Wanting to make Labor’s electoral statistics look better is not a good reason.