Preference flows in the NSW Legislative Council

In the last few days, there has been much media interest in the decision of the Greens NSW to not direct preferences to the Labor Party in the Legislative Council on their how-to-vote cards.

This provoked a fierce response from Labor party figures such as former Premier Bob Carr and Labor MP Luke Foley, who have argued that the decision may result in right-wing minor parties such as the Shooters and Fishers and the Christian Democratic Party gaining the balance of power in the Legislative Council.

The Greens have responded by saying that it was the Labor government’s track record and it’s high levels of unpopularity that would increase the chance of a right-wing majority in the upper house, and not the Greens decision not to preference Labor. Retiring Greens MP Ian Cohen, however, is worried about the potential for right-wing gains in the Legislative Council, and publicly disagreed with his party’s decision.

So what are the facts? Would the Greens preferencing Labor improve the chances of the Greens maintaining the balance of power, rather than the position going to their conservative opponents?

First of all, current polling suggests that the Labor and Greens are running far short of winning the number of seats needed to obtain balance of power for the Greens in the upper house. There were nine Labor members and two Greens members elected in 2007. In order to hold half of the seats, the two parties need to collectively win ten more seats. The quota required to win these seats is approximately 45.5%. It may be possible to win ten seats on slightly less than that.

Even still, most polls have shown the two parties polling well below this level, largely due to a massive drop in the ALP vote. The Greens vote is at record levels, but this doesn’t entirely compensate for the fall in the Labor vote. The recent Galaxy and Nielsen polls each had the two parties collectively polling 35%, while the Newspoll had Labor and the Greens together polling 40%. While Newspoll is not too far off the range where it becomes winnable, it is the highest figure we have seen in a while. It also has the Greens polling 17%, which is the highest we have ever seen the Greens in NSW. It’s certainly a stretch to see the combined vote pushing up to 43% or 44%.

Even if it did reach the range where the balance of power became achievable, the question of preferences having an impact then comes into play.

Here I will quote Antony Green:

Overall at the 2003 election, 78.6% of voters used a single ’1′ only vote, 1.8% voted below the line and 19.6% used the new above the line preferences option. Of those, 66.0% gave a second preferences, 11% a third, 4.0% a fourth with 8.9% of people numbering all 15 above the line boxes. The usage varied by party, 9.7% for the Liberal Party, 25.1% for Labor, 26.3% the Greens.

Green suggests that a similar result occurred in 2007, although he doesn’t mention the specific data in his blog post.

At the 2003 and 2007 elections, preferences had absolutely no impact on the final result. In 2003, the first 17 seats were filled with full quotas based on each party’s primary vote. The final four seats went to the four candidates leading on primary votes after the first 17 had been filled. A similar result took place in 2007.

Antony Green has used this evidence to justify designing his Legislative Council calculator to not use any preferences when determining results. Yet when this evidence has been shown to those arguing for the Greens to preference Labor, they have held out the possibility that the system could spit out a different outcome in 2011.

Could this possibly happen? In an extreme scenario it could happen, but this assumes that there would be no political consequences of a decision to support a party that is currently one of the most unpopular governments in modern Australian history.

In 2003, only 26.3% of Greens voters decided to direct preferences above the line. This is despite the Greens issuing preferences above the line that year. The Greens advised voters to vote ’2′ for Save our Suburbs, then Socialist Alliance, No Privatisation People’s Party, Reform the Legal System, and then the Democrats. Interestingly, the Greens have not preferenced the ALP above the line in either 2003 or 2007, with the 2007 preferences going to the Democrats and a series of other minor parties.

In both those years, Greens preference arrangements were with small parties that shared many policies with the Greens. Even still, only a quarter of Greens even gave a number 2, and Antony Green’s statistics suggest far less would have followed the preference advice any further than that.

In contrast, the current Labor government has reached staggering levels of unpopularity in recent times. The three polls conducted this year have had Labor polling 20%, 22% and 23%, all far below their 39% vote at the 2007 election.

If the Greens were to issue a how-to-vote advising a number ’2′ vote for the ALP, the proportion of the vote that would actually flow to the ALP would likely be substantially lower. Not only would the proportion that would direct a preference be lower, it would be likely that many would rather preference the Coalition than the ALP in the current environment, further nullifying the impact of preferences to the ALP. I would wager that it would be hard for the ALP to gain preferences from 10% of Greens voters, and some of that would be cancelled out by Greens voters preferencing elsewhere. I would be surprised if the net effect was any more than 5% of Greens voters favouring Labor.

Even then, this vote would only have an effect in the following scenario:

  • The Greens vote produces a surplus that is too small to win a seat.
  • The ALP surplus is larger than the Greens, but is only slightly behind a right-wing candidate (Family First, CDP, Shooters, Coalition) in the race for the final seat.

Here’s one possible scenario:

  • The Greens poll 15%
  • The ALP polls 25%

On primary votes the ALP wins five seats, with 2.25% surplus. The Greens win three seats, with 1.35% surplus. If you assume that the ALP would get a 5% bonus from Greens preferences, you end up with 0.07% increase in the ALP vote. The candidate they are trying to beat would need to poll (or have a surplus of) less than 2.32% to be defeated. It is an extremely slim margin where preferences could make such a difference.

You should also note that in this scenario the Greens would still not get the balance of power, as the combined Labor/Greens numbers would only add up to 20 if the ALP squeezed out a win. To win an extra seat one or the other of the parties would need to be polling substantially more.

You also have to look at the possible impacts on the Greens vote of a decision to support the Labor Party in the upper house. The ALP is currently extremely unpopular in New South Wales, including amongst Greens voters. So far I have only found one seat where the Greens are reported to be preferencing Labor, while a large number have been announced as exhausting. The Greens have mounted a fierce attack on the policies and performance of the Labor government, and you would have to think that a decision to preference them would have an impact on the Greens vote, with plenty of voters switching to the Coalition. It is impossible to see the Greens holding the balance of power if this were to occur.

The primary reason why we are facing the possibility of a conservative upper house is thanks to the ALP’s collapsing primary vote. The Greens didn’t preference Labor in the upper house at either of the previous elections under the current system, and it had no effect, because the ALP was popular and won a large vote. If the right wing gains control of the upper house it is because of the massive unpopularity of the current government causing a massive landslide to the right.

Greens preferencing decisions have no impact on that outcome. I’m sure that the Labor types coming out and crying foul know that. They know that their policies have driven away a large proportion of their voter base, and are facing a massive disaster. The latest hysteria about Greens preferences seems to be nothing more than an attempt to blame the Greens for their impending defeat.

Elsewhere: After I drafted this post but before I posted it, Antony Green also posted an analysis of the impact of Greens preferences, interestingly using the same scenario of a 25% Labor vote and a 15% Greens vote.