Post WWII elections by winning two party preferred vote

Here are the federal elections from 1949 onwards in order of two party preferred vote of the winning party*.

All columns self explanatory.                                                           Learn about preferential voting.

Harold Holt's Vietnam War poll tops the list, while Malcolm Fraser's 1975 landslide comes second in votes but received the biggest seat majority. (See second last column). The big wins are all Coalition. Notice that most elections in Australia are reasonably close - closer than, say, US presidential ones, eg Ronald Reagan getting 59 percent of what we would call the primary vote in 1984.

                                                   Continue reading underneath table ...

Election results by winning two party preferred vote (1949 - 2001)

Go to results my winning margin

Notice ...

• The bottom five won government with less than half the vote. Depending on who you ask, this is due to luck or  marginal seat strategical genius, but all were incumbents going for re-election. 

Notice also how Hawke's 50.8 percent of votes in 1987 [rank 15] got him 58.1% of seats, while Beazley's 51.1% in 1998 [rank 22] translated into a meagre 45.3 percent of seats.

Governments are better equipped to pork barrel the marginals, and spend up on elections in general. That's one reason they tend to get more seats for the vote buck. Probably more importantly, voters in the typical marginal seat - regional or outer suburban - tend to be conservative and hard to shift. Howard's success in 1998 was undoubtedly largely due to key voters feeling that tossing a government out after one term wasn't quite cricket.

Another factor is that Labor habitually wastes votes in safe areas, which is why the current hand wringing about soft support in western Sydney - Labor's biggest vote gobbler - is difficult to understand in anything other than sentimental terms. .

• The Howard electoral record

After the 2001 election, Dennis Shanahan wrote in The Australian that "the Liberal team [Howard] has led in three election victories has won some of the biggest votes and parliamentary majorities since Federation."

Let's test that. Howard's initial 1996 win was the sixth biggest since 1949, so we can give that to Dennis - just. But his two re-elections have been deeply ordinary, and one of them - in 1998 - was with the smallest (successful) vote on record. 2001 was also well in the bottom half [rank 13]. 

And in seat terms, 1996 saw a big majority, but the other two were historically modest. Note the government's percentage of seats actually went backwards by a fraction of a percent in 2001 compared with 1998.

Howard's electoral record is, like Shanahan's analysis, unimpressive. He's been re-elected twice, but that's a record no non-Labor government for ninety years has failed to match.

• Primary votes

We hear much about Labor's dire primary vote at the last poll - at 37.8%, its worst since the Great Depression. But for context, have a look at the 1954 election result [no. 21], at which Labor's primary vote was 50 percent - its highest since the introduction of preferential voting in 1919. Labor lost in 1954. See also the 1990 poll [rank 18] that Labor won with primary support of 39.4%.

Two party preferred support is what matters.

• Grumpy old Alan Ramsey thinks primary votes are more important, which might explain his claim several months ago that Howard gave Labor "a hiding" in 2001. In fact the Coalition's proportion of seats went down slightly from 1998 (Labor's dropped by a larger amount). In seat terms, 1998 and 2001 were similar. They were comfy wins but competitive contests.

• Ranks 13 to 22, the bottom 10, are where both sides got between 51 and 49 two party preferred. Exactly half went to the party with most votes. And eight of those ten were Coalition wins. 

So historically, when the vote reasonably close, our electoral system has got it right about half the time - if you reckon the winner of the two party preferred vote should get government. 

And if the vote is close, the Coalition usually wins.

• Further observations as they develop. See also votes and seats graph

* All numbers from the AEC. Note: before 1983 not all preferences were distributed. Two party preferred votes from 1949 to 1980 are therefore the Commission's estimates.  

All look reasonable except 1949, where 51 to 49 two party preferred seems too small a lead from 50.3 to 46.0 in primary votes and a 27 seat majority in a 121 size house. But in absence of better estimate, am using that. For results since 1901 (sans two party preferred) visit the AEC.