The Luck of Politics



Politics is shaped by luck to a far greater extent than has previously been realised. The luck of a ballot draw, the chance of the weather on polling day, or impact of unexpected events overseas can change political outcomes. Luck shapes preselection, individual candidates’ races, and the outcome of leadership ballots. In 2009, Tony Abbott won the leadership of his party by one vote. Had one of Malcolm Turnbull’s supporters not been told to stay home due to a health scare, the result might have been different.

The presence of chance in politics is more than a mere curiosity. Understanding luck should make us less inclined to put the successful on pedestals and kick the unsuccessful into the gutter. Hard work and intellect make a difference, but political life is also shaped by the winds of chance.

In particular, luck should make us even more concerned about extreme partisanship, since the political environment can act to magnify small gaps into large differences.

In 1971, researchers at Stanford University carried out an experiment in which 24 men were randomly assigned to play the roles of guards or prisoners. Within a day, those randomly appointed guards were abusing and tormenting those who had been randomly designated prisoners.

Similarly, the erosion of a political middle ground can mean that small chance events end up having large consequences. Imagine a landscape covered with balls, each of which represents a person’s ideology. If the ground is flat, then it doesn’t matter much if a ball is placed a little to the left or right. But if the ground is hilly, then small perturbations end up making a big difference. If a ball starts on the top of a steep hill, a small change in its placement will greatly affect where it ends up.

Moving from moderation to hyperpartisanship is like going from a flat political landscape to a hilly one. Not only is it harder to see eye to eye, but the luck of a person’s starting point exacerbates trivial differences into mammoth gaps. A recent analysis of media releases put out by members of the United States Congress found that one-quarter used exaggerated language to deride the opposition or their ideas. Attempts to undermine the legitimacy of individuals has manifested in the claim by ‘birthers’ on the extreme right that President Barack Obama was not born in the United States, and by birthers on the extreme left that Prime Minister Tony Abbott did not properly renounce his claim to British citizenship.

The partisan gap in the United States has been systematically measured by political scientists Keith Poole and Howard Rosenthal, who use congressional votes to place each legislator on the political spectrum. They show that during the interwar period, the gap between the average Democrat and the average Republican narrowed considerably. In the 1940s, half of all legislators were what they classify as ‘moderates’.

Since then, the gap between the major political parties has grown significantly. Today, partisanship in the US House of Representatives is higher than at any time since the late 1800s. At the same time as the gap has grown, the share of moderates has fallen – from above half to just one in ten. Across US counties, the same pattern can be seen. From 1976 to 2008, the share of Americans living in a ‘landslide county’ – where the incumbent had more than a 20% margin – rose from 27% to 48%.

In Australia, strict party discipline prevents a similar analysis of our partisan gap. However, it is possible to look at the gap in another way. Since 1996, researchers at the Australian National University have surveyed voters and candidates, asking them to place themselves on a left–right scale running from 0 to 10. Over the past two decades, the ideological gap between Labor and Coalition voters has grown from 1.5 to 2.1. For candidates, the partisan gap has increased from 2.4 to 3.3.

Another measure of partisanship is the share of ‘crossover’ candidates – either Labor candidates who are to the right of the average Coalition candidate, or Coalition candidates who are to the left of the average Labor candidate. In 1996, one in ten candidates fitted this description. By 2010, it was down to almost one in twenty.

From the late 1960s to the mid-1980s, Liberal MP Bert Kelly wrote a ‘Modest Member’ column for the Australian Financial Review. In 2012, the newspaper revived the ‘Modest Member’ column, this time written in turn by a group of Liberal Party parliamentarians. Analysing the columns from both eras, I estimate that 36% of the original ‘Modest Member’ columns criticised the Labor Party, while 29% criticised the Coalition. By contrast, the new ‘Modest Member’ columns criticise the Labor Party 70% of the time, but criticise the Coalition less than 2% of the time. Today’s ‘Modest Members’ are twice as likely to censure their political opponents, but fourteen times less likely to find fault with those on their own side.

The same pattern can be seen in an increasingly polarised media. From 1985 to 2013, the share of Americans who say that the press tends to favour one side rose from 53% to 76%. The rise of Fox News on the right and MSNBC on the left reflects a pattern in which media outlets are more likely to be critics or cheerleaders than straight reporters.

In the new world of hilly hyperpartisanship, understanding luck matters more than ever.


This is an edited extract from Andrew Leigh’s new book, The Luck of Politics, published by Black Inc. The Luck of Politics is being launched by the Melbourne School of Government on Tuesday September 1.

The Hon. Andrew Leigh

Shadow Assistant Treasurer and Member for Fraser

Published Date
August 31, 2015