Brian Galligan is Professor of Political Science at the University of Melbourne. He has published extensively on Australian politics, including more recently as joint author Australian Citizenship (2004) and Australians and Globalization (2001). He is joint editor with Winsome Roberts of The Oxford Companion to Australian Politics to be launched by the Chancellor Ian Renard on 15 November.

The most likely outcome of the 2007 election is a comfortable win for Labor with minor parties holding the balance of power in the Senate. If so this will be only the sixth change of party in national government since Labor under Curtin came to office in 1941 – Menzies in 1949, Whitlam in 1972, Fraser in 1975, Hawke in 1983, and Howard in 1996.

Incumbency is usually the best predictor of electoral outcomes. Governments lose elections because of weak leadership (McMahon in 1972), or economic mismanagement (Whitlam in 1975). A third factor is widespread belief that they have run their course and become out of touch with public sentiment (Fraser 1983, Keating 1996, and probably Howard in 2007).

Crucial for change is a credible alternative prime minister: Latham showed that negatively last election; Rudd positively this time.

Winning a majority of votes is not enough: Peacock did so in 1990 but lost by eight seats; Beazley with 51 per cent of the vote lost by 12 seats in 1998. Nor are swings uniform: a 4.9 per cent uniform swing would give Labor the 16 seats needed to win, but that would include Howard’s Bennelong seat, and two Western Australian seats where Coalition support is holding up. More likely is an uneven six per cent swing that delivers about 16 to 20 seats to Labor.

A bolder scenario is a Labor landslide of eight per cent swing or more and 30 plus seats that the opinion polls have consistently suggested. But as Hewson’s losing to Keating in 1993 showed, election campaigns can turn things around.

A counter scenario is Howard doggedly hanging on; perhaps governing with the support of maverick National-cum-independent Queenslander, Bob Katter.

Incumbency after a decade in office is a poor indicator of election outcomes (even Menzies nearly lost in 1961). Howard’s pragmatic opportunism has had its decade and squandered the recent good times. Change is likely and welcome.