Brian Rudman 's Opinion

Brian Rudman is a NZ Herald feature writer and columnist.

Brian Rudman: 'Donkey votes' can make an ass of council election

55 comments
Mayor Len Brown and the nine of 20 councillors whose names start in the top third of the alphabet, should declare a vested interest in the outcome and not vote. Photo / Sarah Ivey
Mayor Len Brown and the nine of 20 councillors whose names start in the top third of the alphabet, should declare a vested interest in the outcome and not vote. Photo / Sarah Ivey

Tomorrow, Auckland councillors decide whether candidates' names in this year's ballot papers should be listed in alphabetical order.

Mayor Len Brown and the nine of 20 councillors whose names start in the top third of the alphabet, should declare a vested interest in the outcome and not vote.

For that matter, perhaps the four who are used to being listed last on any ballot paper - the Walkers, Sir John and Wayne, Penny Webster and George Wood - should do likewise. For research suggests that candidates listed at the top or the bottom of a ballot paper have an unfair advantage over those stuck in the middle.

The Australians, bless 'em, even have a name for the effect: they call it the "donkey" or the "moron" vote.

Across the Tasman, a compulsory voting law forces the ignorant, the apathetic and the protesters to vote.

As a result, some tick a few names at the top and the bottom of the voting paper, regardless of party, to avoid a fine.

But plenty of research around the world suggests those at the top of the list have an advantage even without the compulsion.

Researchers from City University in London examined the relationship between vote ranking and the position on the 2010 local government election ballot paper of 5000 candidates in the Greater London area.

This was a first past the post election, won by the top three polling candidates in a ward. Most candidates were on party tickets and names were in alphabetical order.

The outcome was that "on average, a candidate listed first in their party was 6.3 times more likely to get the most votes in their party than a candidate listed third."

The researchers concluded that "ballot position did indeed strongly influence the number of votes received by candidates ... and that some of those who are currently representing London may have benefited from this effect, just as those who are not, suffered from it".

The authors said there was "some evidence that the strength of this effect is sufficient to overcome voter preference for party, most likely in marginal seats ..."

The Local Government Commission, in a July 2008 review, acknowledged a similar effect in the 2007 New Zealand local elections. Candidates whose names appeared early in the alphabetically listed voting papers and candidate profile booklets "were up to four per cent more likely to be elected than those whose names were later in the alphabet".

It also found "a significant bias in favour of candidates in the left column of voting documents".

Its call for further research has not been followed through.

The report to tomorrow's meeting by new Electoral Officer Bruce Thomas offers councillors three options to chose from, but gives none of the above background.

It warns that selecting the fairest option, which is to print the ballot papers in a truly random order, would cost an extra $100,000. But what price democracy?

The other option is to print the names in "pseudo-random" order, which involves drawing the names out of the proverbial hat and printing all ballot papers in that "random" order. However, this just transfers the unfair advantage from the alphabetically-selected top of the pile to another group of "winners".

The fairest option is to produce a truly randomly ordered set of ballot papers. This would involve programming a printer to randomly change the order of candidates as it prints each ballot paper. This wouldn't stop a proportion of voters still favouring those at the top of the poll, but it would ensure every candidate had an equal chance to benefit, by being at the top of an equal number of ballot papers.

By selecting the random system tomorrow, Auckland councillors would only be playing catch-up with the country's other big cities.

In 2010 Wellington, Christchurch, Nelson, Porirua and Tauranga used it. So did seven district health boards and five district councils. Last week, Hamilton councillors decided that for democracy's sake, the extra $10,000 on election costs of around $300,000 was worth it in this year's election.

The extra cost for Auckland is $100,000. The report does not include the estimated total cost of running the election.

But based on 2010 figures, my guess is there'll be little change out of $5 million, which makes the $100,000 rather small, especially when shared among a million or so eligible voters.

- NZ Herald

Brian Rudman

Brian Rudman is a NZ Herald feature writer and columnist.

Brian Rudman's first news story was for Auckland University student paper Outspoke, exposing an SIS spy on campus during the heady days of the Vietnam War. It resulted in a Commission of Inquiry and an award for student journalist of the year. A stint editing the Labour Party's start-up Auckland newspaper NZ Statesman followed. Rudman decided journalism was the career for him, but the NZ Herald and Auckland Star thought otherwise when he came job-hunting. After a year on the "hippy trail" overland to London, he spent four years on Fleet St with various British provincial papers. He then joined the Auckland Star, winning the Dulux Journalist of the Year award for coverage of the 1976 Dawn Raids against Polynesian overstayers. He has also worked on the NZ Listener, Auckland Sun, and since 1996, for the NZ Herald as feature writer and columnist. He has a BA in History and Politics.

Read more by Brian Rudman

Have your say

We aim to have healthy debate. But we won't publish comments that abuse others. View commenting guidelines.

1200 characters left

Sort by
  • Oldest

© Copyright 2014, APN New Zealand Limited

Assembled by: (static) on red akl_a3 at 02 Oct 2014 18:01:31 Processing Time: 851ms