In today’s post I’m following up on a pre-election blog post where I looked at what preference flow data tells us about the underlying political trends in the crossbench seats, now with 2023 data.
I’m going to start with a chart looking at how primary votes for the twelve crossbenchers flowed on a two-party-preferred basis.
The preference flows from the three Greens MPs were very strong to Labor, and more so than they had been in 2019.
Preference flows from Alex Greenwich voters favour Labor, moreso than they did in 2019. Greg Piper's preference flows to Labor imrpoved, but his voters still overwhelmingly exhaust and only weakly favour Labor.
New independents Michael Regan and Judy Hannan have surprisingly strong preference flows to Labor, and lower exhaustion rates.
The regional independents had very low rates of preferences to Labor, but the Coalition preference flows were also weak, with very high rates of exhaustion.
Helen Dalton had the lowest rate of preference flows to Labor, even less than in 2019.
That chart then flows through to this next chart, which shows the overall 2PP - including the votes of these elected crossbenchers as well as the other candidates' votes.
Labor managed a truly astounding 2PP in Kiama of almost 70%, but when you factor in exhausted votes, it looks far more modest. Labor managed less than 47% of the vote after preferences, but with so much of the right-wing vote exhausting before it reached the Liberal, that translated to a wide Labor 2PP majority. Still, less than a third of all votes exhausted on the 2PP in Kiama, while it exceeded 40% in Orange, Lake Macquarie, Barwon, Wagga Wagga and exceeded 50% in Murray.
Exhaustion rates generally were higher in the Coalition-leaning and rural seats. Only 12-16% of voters exhausted in the Greens seats, and 22-26% in Sydney, Wollondilly and Wakehurst.
Labor "won" the 2PP in six out of these twelve seats, although there were only two out of the twelve where the winner managed more than half of the total formal vote - Balmain and Newtown.
These 2PP figures contribute to the calculation of the statewide 2PP, and they also could in theory be used to calculate the swing required if a seat were to return to a classic contest in 2027, as is quite likely in Kiama. But the huge exhaustion rates show that to be a bit silly, and should make people cautious in drawing conclusions. I have no doubt that most Ward voters who exhausted would happily vote Liberal 1 if Ward was not standing, and thus that Labor 2PP majority would quickly vanish.
When so many voters exhaust their votes, it's hard to use that information to assess the relative political standing of those seats, so instead I'm going to look at how these seats voted for the upper house.
Balmain and Newtown have very large progressive votes, combining Labor, Greens, Animal Justice and Legalise Cannabis. Sydney, Ballina, Lake Macquarie and Kiama all lean slightly to the left.
Wakehurst came up as surprisingly progressive - of the identified parliamentary parties in the above chart, neither the left nor the right won a majority there.
Wollondilly, Wagga Wagga and Barwon all had a right-wing majority, with Orange and Murray having a bigger right-wing majority.
Finally, I've made a chart comparing the 2PP and the alternative 2CP in a series of interesting races where there was a strong minor party challenger. This includes many of the crossbench seats, but also a few others, including races like Lismore and South Coast that ended up being classic races.
It is interesting to see which seats had very different exhaustion rates - specifically Murray, Kiama, Orange, Wagga Wagga. Lake Macquarie and Barwon. In these seats, the 2PP is quite distorted because so much of the electorate has opted out.
It's interesting to compare how Labor performed relative to the teal in some of the weaker teal seats. The independent only slightly outperformed Labor in North Shore and actually underperformed Labor in Lane Cove.
It's also interesting to compare Labor and the Greens in the races where one of them was up against the Coalition. The Greens do better than Labor in Ballina, but the gap between Labor's 2PP and the Greens 2CP in Lismore is now about 15%. The Greens actually won the 2CP narrowly, but Labor won the seat quite safely. I wrote about this phenomenon prior to the election.
Part of the story here is that candidates with a primary vote usually do better, but in seats like Lismore it looks like Saffin has attracted a larger group of voters who don't have interest in preferencing the Greens. The Greens-Nationals 2CP in Lismore has a much higher exhaust rate.
That's it for today, but I'll be back with a few more posts over the coming days.
Working hard over the past few days Ben. Another interesting analysis.
The only thing I would question you about (remembering I am no Psephologist) is that you said that Kiama would return to the Liberal fold, but that does seem to be born out in your upper house analysis where Labor and Greens seem to have a healthy share that would (IMO) make then hard to run down.
Ben, Neil: it aint necessarily so? In Kiama there was basically no Liberal HTV distributed at all on the day or by mail. As I remember, the Libs were along the right of the ballot paper so if you wanted to vote Lib you had to work hard, have reasonable eyesight (a problem for the Lib demographic in Kiama) & pass by all the sexy alternatives.
I didn’t say that Kiama would return to the Liberal fold. I said the Labor majority (the 69% one) would vanish. They could win, but it won’t be that sort of huge margin.
There’s strong preference for Labor as the ‘preferred major party’ in the Green seats and to a lesser extent, Sydney. Kiama leaned Labor, when excluding Gareth Ward. Sydney is interesting and would be tricky to estimate the real 3CP without an independent. The Greens/Labor vote has been subdued under Alex Greenwich and Clover Moore and their presence has encouraged tactical voting and sucked up the progressive vote. Methinks this would be a Labor seat as this area is more Liberal and isn’t as Green as Newtown and Balmain.
Thanks Ben. That looks like a diverse bunch of electorates.
Could the high exhaustion in the alternative 2PP in some seats be simply accounted for by the benefits of incumbency? After all, it would be reasonable for a voter to assume that they could vote 1 for their crossbench MP and their vote would be very unlikely to exhaust. The exception to that is Sydney, which is perhaps more of a multi-party contest than the regional seats.
I also recall your pre-election post regarding various scenarios for minority government in NSW. Of course the result that we have from this election is close scenario #7. It’ll be interesting to see how this minority situation plays out this term… But perhaps the split upper house takes on a more defining role?
And the IND (not Teal though) only beat Labor by around 3% in Willoughby (2% in North Shore).
When Lane Cove, North Shore and Willoughby are looked at collectively, more votes exhausted in the IND vs LIB contest than in the ALP vs LIB contest. Makes you think what’s going on. I am struggling to see other seats where the same occurred (seats with GRN rather than IND are equivalent as there is a different dynamic going on there).
I’ll take a punt on why I think those results are different, Insider. I think the lower north shore has both a larger “swinging” vote (mainly Chinese), and also a larger legacy public sector / working class vote (RNSH nurses in St Leonard’s flats), both in high density along the railway line and the highway, than either the northern beaches or eastern suburbs – and that much of that vote wants nothing to do with installing an independent MP. If they couldn’t have a Labor MP this time, then the attitude of some was “tough”. A bitter pill for IND supporters to swallow.
@Insider The reason why in Lane Cove and North Shore, the exhaustion rate for the LIB vs ALP contest was lower than the LIB vs IND contest was because the independent candidates’ “number every box” campaign in these seats were so powerful that there were a higher proportion of independent voters preferencing a major party than ALP voters preferencing LIB or IND. The effectiveness of “number every box” campaign run by independents was also the reason why in Willoughby and Wollondilly, the exhaustion rate for the LIB vs ALP contest was only marginally higher than the LIB vs IND contest. The “number every box” campaign ran by independents was targeted at ALP and minor party voters, but it was the independent voters themselves who were most likely to follow this advice. Just like at every NSW state election, the Coalition’s “Just vote 1” campaign is targeted at minor party and independent voters, but it is the Coalition voters who are more likely to just vote 1 than anyone else.
The rate of preference flows from minor party and independent voters in LIB vs ALP contests and from Labor and minor party voters in LIB vs IND contests are shown below for six Northern Sydney seats with strong independent challengers.
2023 Lane Cove LIB vs ALP preference flow: 2782 (17.25%) to LIB, 8578 (53.19%) to ALP, 4768 (29.56%) exhausted
2023 Lane Cove LIB vs IND preference flow: 2110 (11.73%) to LIB, 9305 (51.73%) to IND, 6574 (36.54%) exhausted
2023 Manly, LIB vs ALP preference flow: 3915 (17.58%) to LIB, 9798 (44.00%) to ALP, 8554 (38.42%) exhausted
2023 Manly, LIB vs IND preference flow: 1777 (12.09%) to LIB, 6662 (45.33%) to IND, 6257 (42.58%) exhausted
2023 North Shore LIB vs ALP preference flow: 2900 (15.57%) to LIB, 9132 (49.02%) to ALP, 6598 (35.42%) exhausted
2023 North Shore LIB vs IND preference flow: 1732 (10.60%) to LIB, 7802 (47.74%) to IND, 6808 (41.66%) exhausted
2023 Pittwater, LIB vs ALP preference flow: 4659 (20.86%) to LIB, 10528 (47.14%) to ALP, 7148 (32.00%) exhausted
2023 Pittwater, LIB vs IND preference flow: 1228 (12.77%) to LIB, 5005 (52.03%) to IND, 3387 (35.21%) exhausted
2023 Wakehurst, LIB vs ALP preference flow: 4745 (19.16%) to LIB, 8437 (34.05%) to ALP, 11595 (46.80%) exhausted
2023 Wakehurst, LIB vs IND preference flow: 1615 (11.56%) to LIB, 6159 (44.11%) to IND, 6190 (44.33%) exhausted
2023 Willoughby, LIB vs ALP preference flow: 3120 (16.23%) to LIB, 10088 (52.49%) to ALP, 6013 (31.28%) exhausted
2023 Willoughby, LIB vs IND preference flow: 1695 (10.77%) to LIB, 8213 (52.20%) to IND, 5826 (37.03%) exhausted
In fact, as the preference flow rates above shows, in Lane Cove, North Shore and Willoughby, ALP got stronger preference flows in the LIB vs ALP contest than INDs got in the LIB vs IND contest, due to the fact that independent voters in these seats tended to number more preferences on their ballot papers than Labor voters did (@benraue You can check the ballot completion rate data to verify this), despite IND preferences in these seats being less likely to be distributed than ALP preferences.
Seems surprising right? After all, in the 2022 federal election, among 14 L/NP vs IND seats, only 1 seat, Kooyong, had the ALP candidate received a higher rate of preference flow in the L/NP vs ALP contest than the IND candidate received in the L/NP vs IND contest. In fact, under CPV, assuming exhausted votes will split in the same way as non-exhausted votes, the IND candidates in Lane Cove, North Shore and Willoughby would still received a higher rate of preference flow in the LIB vs IND contest than the ALP candidate would have received in the LIB vs ALP contest. However, OPV distorted preference flows due to IND and ALP voters having different rates of ballot paper completion.
By the way, @benraue, if you check the 2PP preference flows from independent challengers in Lane Cove, North Shore, Willoughby, Manly and Pittwater, the results could well shock you. Their rate of preference flows to Labor could well exceed that of any elected independents including Alex Greenwich, with exhaustion rates approaching or even lower than some Greens MPs.
The whole discussion of assessing the effectiveness of “number every box” campaigns in Northern Sydney seats with strong independent challengers is well worth a whole new blog post in itself.
@benraue I think the 2PP preference flows from unsuccessful independent challengers in Northern Sydney seats, as well as the 2PP and upper house votes in these seats, should be added to this post, so that some more surprising findings can be revealed and debated.
Comments are closed.