Average Swings at Federal By-elections 1983-2013 antony green

« 2013 Federal Election Offcuts | Main

January 23, 2014

Average Swings at Federal By-elections 1983-2013

Since 1983, the average swing against the government of the day in by-elections that finished as two-party contests has been 5.1 percentage points. The average swing against first term governments was a smaller 1.7 percentage points. The swing has been against the government at six of the eight first term two-party by-elections held since 1983.

Altogether there were 36 by-elections held between 1983 and 2013. Of these, 18 finished as two-party contests between Labor and the Coalition, two saw one of the major party candidates excluded, and 16 by-elections were not contested by the government of the day.

It has become more common in recent years for governments to not contest by-elections in safe Opposition seats. Labor under Hawke and Keating contest 16 of the 22 by-elections between 1983 and 1996, the Coalition under John Howard contested three of nine by-elections, and Labor under Kevin Rudd contested one of five by-elections.

Looking at the 18 two-party by-elections between 1983 and 2013, the average swings were as follows:

  • The average swing against the government after preferences for all 18 by-elections was 5.1 percentage points. The average change in first preference votes was 6.2 percentage points against the Government, 4.2 percentage points towards the Opposition, and 2.0 percentage points to Others. Governments tended to get their best by-election results in their first term.
  • The sample includes eight by-election in the first term of a government. The average swing against the government after preferences in the first term was 1.7 percentage points. The change in first preference vote share was Government -1.9, Opposition +3.3, Others -1.4.
  • The sample includes three by-elections held in the second term of a government. The average swing against the government after preferences in the second term was 5.9 percentage points. The change in first preference vote share was Government -6.7, Opposition +3.7, Others +3.0.
  • The sample includes three by-elections held in the third term of a government. The average swing against the government after preferences in the third term was 10.4 percentage points. The change in first preference vote share was Government -12.0, Opposition +8.0, Others +3.9.
  • The sample included no by-election in a fourth term but four by-elections held in the fifth term of a government. The average swing against the government after preferences in the fifth term was 7.3 percentage points. The change in first preference vote share was Government -10.2, Opposition +3.8, Others +6.4.
  • There were only three by-elections that recorded a swing to the government. These were a 0.5% swing to Labor at the 1984 Richmond by-election, retained by the National Party. There was a 1% swing to Labor at the 1994 Fremantle by-election when popular former WA Premier Carmen Lawrence was elected to Federal Parliament. There was a 5.0% to the Liberal Party at the 1996 Lindsay by-election when Jackie Kelly was re-elected after her general election victory had been voided by the Court of Disputed Returns.
  • Despite massive leads in opinion polls, the Hawke government sufferred swings against it at five of the six by-elections held in its first term 1983-84.
  • Despite a massive lead in opinion polls, the Rudd government suffered an 8.1% swing against it at the Gippsland by-election in 2008.

While it can be argued there is no such thing as an ‘average’ by-election, the above figures provide some historical context for the upcoming Griffith by-election.

For background on the Griffith by-election and profiles of the candidates, check out my Griffith by-election page at http://www.abc.net.au/elections/federal/2014/griffith/. The site also includes more detail on past elections, including the source data from which the above by-election analysis has been compiled.

Posted by on January 23, 2014 at 10:48 PM in By-election,

Leave a Reply