Exposing a Rubbish Statistic – there were more Labor than Liberal voters at the 2019 Federal election

The Morrison government was re-elected in 2019 with a three seat majority. The government won 77 seats in the 151 seat House of Representatives against 68 Labor and six crossbench members. The Coalition recorded 51.53% of the national two-party preferred vote and gained a 1.17% swing in its favour. The Coalition also improved its position in the Senate, gaining several seats that had been won by small right-of-centre parties at the 2016 double dissolution election.

Labor recorded its lowest first preference vote in more than 80 years, a lower vote than on the defeat of the Rudd government in 2013. The Labor Party accepted that it lost the election, the party’s postmortem largely blaming its own policy and campaign failures for the defeat. One external factor it did highlight was how much Clive Palmer was allowed to spend on advertising.

Yet some in the twittersphere have not accepted the legitimacy of the Morrison government’s victory. They put forward various arguments about why the government didn’t or shouldn’t have won.

It cheated with “sports-rots” and mis-leading signs in Chinese. Clive Palmer’s spending was illegal (it wasn’t) and his nomination of ineligible candidates should have invalidated some contests (it didn’t). Somehow the Australian Electoral Commission (AEC) should have stopped these things happening, despite the AEC’s lack of legal authority to do so. And why was the result so different from the pre-election opinion polls?

Labor to its credit doesn’t back the above claims. The Labor Party does want a proper investigation into the administration of the various sports grants programs before the election, but no one seriously suggests these grants won the election for the government.

Amongst the arguments thrown around on twitter to de-legitimise the Morrison government is this one – that Labor won more votes than the Liberal Party.

If you look at the House of Representatives party vote totals on the AEC’s website, the Labor Party did poll more votes, 4,752,160 votes (33.3%) to 3,989,404 (28.0%) for the Liberal Party.

But to argue these two numbers mean the Morrison government did not legitimately win the 2019 election is to ignore history, to ignore the existence of the Coalition, to ignore that the LNP is a branch of the Liberal Party, and to ignore that while Labor contests all seats in Australia, each of the four Coalition parties contest only part of the country.

For decades it has almost always been the case that Labor has outpolled the Liberal Party. What four decades ago might have been a whinge against preferential voting, these days is almost totally irrelevant. Labor’s vote has been sliding election after election since the emergence of the Greens. To continue to point to Labor having more first preference votes than the Liberal Party becomes a less and less relevant argument as Labor’s vote slips and the party relies more and more on preferences.

Spouting that Labor polled more votes than the Liberal Party is an example of a rubbish statistic, something that is true but irrelevant and meaningless.

Let me explain why.

Australian Elections are Labor versus Coalition contests, not Labor versus Liberal

There have been only three Federal elections fought out solely between Labor and Liberal Parties. These were the 1910, 1913 and 1914 elections. After the 1916 Labor split, the 1917 election was a Labor-Nationalist contest, 1919 and 1922 saw the arrival of the Country Party, after which the Nationalists and Country Party formed the first Coalition government.

Every election since has pitted Labor against a two-party conservative coalition, with only the party names changing. The Nationalists became the United Australia Party in 1931 and the Liberal Party in 1944. The Country Party became the National Party in the 1970s. But apart from two brief periods in government in the 1930s, and several occasions in opposition, the two parties have always fought elections as a Coalition.

The creation of merged state branches in Queensland (Liberal National Party or LNP) and the Northern Territory (Country Liberals or CLP) means there are technically four parties in the Coalition. The Liberal Party does not nominate candidates under its own name in either Queensland or the Northern Territory.

The arrival of what became the Country Party led to the introduction of preferential voting ahead of the 1919 Federal election. Since then, comparing first preference votes by party has become a meaningless statistic. The best measure is always the count after preferences, what we know as the two-party preferred vote. This measures voter’s preferred parties, not an artificial adding together of party first preference votes.

So in 2019, the four Coalition parties polled 41.44% to 33.34% for Labor. If you add Labor and the Greens together, even though they are not related parties, they add to 43.74%, more than the combined Coalition. But as I said, first preference comparisons don’t matter under preferential voting. After the exclusion of all other candidates in 2019, the Coalition won 51.53% to Labor’s 48.47%. And if you use the two-party result in the six crossbench seats, there were 81 seats with a Coalition majority in 2019 compared to 60 with a Labor majority.

Preferential voting has allowed the Coalition to exist as a long-term electoral entity, something that wouldn’t have happened under first past the post voting. Preferential voting allows the Coalition parties to govern together and campaign with a common platform, while being able to compete against each other in individual contests.

It is an arrangement that would fall apart under first past the post voting, where one party would have to stand aside to avoid splitting the vote and delivering seats to Labor.  Even optional preferential voting creates difficulties for the Coalition, as seen by the disappearance of three cornered contests at NSW and Queensland elections.

So only comparing votes cast for the Labor and Liberal parties ignores two institutions of Australian politics that have been in place for a century – preferential voting and an anti-Labor Coalition.

The LNP is the Queensland Branch of the Liberal Party

If you go to the AEC’s register of parties, you will find that the Liberal National Party of Queensland is registered as a branch of the Liberal Party of Australia. However, the LNP chooses to nominate candidates under its state name rather than its national parent’s name. For that reason the AEC reports results for the LNP separate from the Liberal Party.

If you acknowledge this registration, and add the two parties together, then the Liberal Party polled 36.66% of the first preference vote, more than Labor’s 33.34%.

Anyone arguing that the Labor Party polled more votes than the Liberal Party is arguing that all Labor’s votes in Queensland should be included in its national totals, but the Liberal Party should be given zero votes in Queensland. That would be distorting the truth given Labor’s dreadful result in Queensland at the 2019 election.

The Pattern of Candidates

Labor nominated candidates for 151 seats at the 2019 election, the Liberal Party only 107. There were 96 seats where the Liberal candidate was the only Coalition representative, and another 11 three-cornered contests where there was also a National Party candidate. There were a further 12 seats with only a National Party candidate, 30 seats in Queensland with only LNP candidates, and two seats in the Northern Territory with only Country Liberal candidates.

When you break down the election result by contest type, you get the party percentages shown in the table below.

	% Party Vote by Labor v Coalition Contests
Labor versus (Contests)	LIB (96)	LNP (30)	LIB & NAT (11)	NAT (12)	CLP (2)
Labor	37.0	26.7	26.8	25.6	42.3
Coalition	39.6	43.7	47.1	45.5	37.5
Greens	11.1	10.3	7.5	7.7	10.2
One Nation	1.3	8.9	3.3	2.7	..
United Australia	3.2	3.5	3.7	5.1	2.8
Others	7.7	6.9	11.5	13.4	7.2
Two-Party Preferred
Labor	52.3	41.6	42.4	39.9	54.2
Coalition	47.7	58.4	57.6	60.1	45.8

• In the 96 Labor-Liberal contests, the Liberal Party outpolled Labor 39.6% to 37.0%. Labor did win a majority of the two-party preferred vote in these contested, 52.3% to 47.7%.
• In the 30 Queensland seats that were Labor-LNP contests, the LNP had an overwhelming majority 43.7% to 26.7%, also winning the two-party preferred vote 58.4% to 41.6%.
• In 12 seats where there was only a National candidate, the Coalition primary vote was 45.5% to 25.6% for Labor, also winning the two party preferred vote 60.1% to 39.9%.
• In the 11 three cornered contests, the Liberal Party polled 37.7% and the Nationals 9.4% against 26.8% for Labor, also winning the two-party preferred vote 57.6% to 42.4%.
• In the two Northern Territory seats, Labor led the Country Liberals 42.3% to 37.5% and 54.2% to 45.8% after preferences.

So in summary, the Liberal Party did record fewer votes than the Labor Party, but that is a wholly irrelevant statistic as it ignores non-Labor votes in Queensland, the Northern Territory and in most rural parts of the country.

So was it preferences that Lost the Election for Labor?

In short, no. One Nation preferences didn’t win the election for the Coalition. Clive Palmer’s preferences didn’t win the election for the Coalition, though his advertising spend might have helped in the Coalition’s campaign against Labor.

The preferences of both One Nation and the United Australia Party flowed about 65% to the Coalition and 35% to Labor, but that didn’t allow the Coalition to win any seats where they started the count trailling on first preferences. All that One Nation and United Australia Party preferences did was widen a lead the Coalition already held on first preferences.

While preferential voting may have been introduced in 1918 as an anti-Labor tactic, to make it easier for competing conservative candidates to contest seats without splitting the conservative vote, it has been three decades since preferential voting favoured the Coalition.

And so it was at the 2019 election. The Coalition won 77 seats, 31 on first preferences and 46 after leading on first preferences. Of Labor’s 68 seats, it won 14 on first preferences, 44 after leading on first preferences and another 10 wins from second place on preferences. Of the six crossbench members, Independent Andrew Wilkie won a majority of the first preference vote, Rebekha Sharkie in Mayo and Helen Haines in Indi won from second place after preferences, the three other crossbenchers being elected after leading on the first count. (You can read about how preferences flowed at the 2019 election, and how the impacted on the result, in this post.)

Whatever way you look at it, the argument that Labor won more votes than the Liberal Party mis-represents the nature of the 2019 election. Labor had a record low vote. Labor had a swing against it. Minor party preferences didn’t prevent Labor winning. And above all, it is the Coalition vote that should be compared with Labor’s, not just the Liberal Party’s vote.