NSW councils – measuring the magnitude


One of the main things that determines the shape of an electoral system is the number of elected members elected for each constituency. It’s a key feature that sets the direction of how elections work in each NSW local council. In this post I’m going to explore how this varies across New South Wales.

NSW legislation requires a consistent magnitude for all wards in a council. This is in contrast to Victoria, which often draws ward boundaries with different numbers of councillors elected for each ward. I mapped out the magnitude of Victorian councils in this blog post before the 2020 council elections.

There are no wards in New South Wales which only elect a single councillor. The last single-member wards died with the old Botany Bay council when it was amalgamated into Bayside council in 2016.

Up until the 2012 election, a different voting system was used for two-member wards. This system was similar to the old preferential block vote used for the Senate until 1946. Voters marked preferences. The first seat was elected using a single-member election and then all the votes were returned to the original pile, the first elected candidate excluded and their votes passed on at full value, then another election was conducted. This voting system was extremely majoritarian and usually produced a clean sweep in each ward for a ticket and often led to massive majorities. When I was first getting involved in politics around 2004 it was still being used for Ku-ring-gai, Botany Bay, Cabonne, Wollongong and Shellharbour, amongst others, but it no longer is an option.

The dynamics of an election play out very differently even with the same voter preferences when a council is divided up until small wards compared to a council with a higher magnitude.

All but one council with a magnitude of six or above has only a single ward, while magnitudes of 2 to 5 involve more than one ward in a council. The only exception to this rule is Fairfield, which is switching in 2021 from three four-member wards to two six-member wards.

Councils with 6 or more tend to have reasonably proportional results, with a wide array of groups running. On the other hand, 2-member wards tend to result in fewer options, both because the options available are split between the wards, but also because less popular options simply don’t bother. I briefly touched on how the 2-member system helps reinforce deadlocks on Ku-ring-gai council in a recent post.

The perfect recent example is Shellharbour, which switched from a single 7-member ward in 2017 to four 2-member wards (plus a directly elected mayor) in 2021.

Ten groups ran for Shellharbour in 2017. Labor topped the poll with just 32.7% and won three seats, with four different groups winning the remaining four seats.

In 2021, there is only one ward with more than two options. One ward has three groups running, while two others have two groups. In the fourth ward, it appears that the second Labor candidate withdrew to allow for an uncontested election. There will be a mayoral ballot, but with only two candidates. Labor will almost certainly win one seat in each ward, plus a mayoralty giving them a majority. I highly doubt they would have polled a majority of the vote with a more open voting system.

This chart shows what share of councils in NSW use each magnitude.

The most common option is nine councillors, with the next most common is three. To keep it simple, three is the default option for urban councils, while nine is the default for rural councils. Those councils with 3-member wards are much more populous than most of the 9-member unwarded councils, as you can see in the next chart, which shows the number of electors enrolled in each type of council.

Over 46% of voters in councils up for election in 2021 will vote in wards electing three members. When you add in those councils using 2-member wards, more than half of electors will cast votes in wards with 3 councillors or less.

There are different dynamics that play out in 2-, 3-, 4- and 5-member wards. Once you get past that point the differences are more subtle, although an election in 15-member Campbelltown can look quite different to a 6-member unwarded council.

3-member wards, at least where partisan politics exists, tends to result in one safe seat for each of the two main “sides”, with the third seat in play. With a quota of 25%, you need about 50% to win two seats, which is uncommon but does happen (see Blacktown with its solid Labor majority). While most magnitude-3 councils in NSW don’t have a one-party majority, it’s easier to achieve than in a place like Campbelltown with 15 members and no wards. I touched on the impact of 3-member wards on local party systems in this blog post.

4-member wards have their own peculiarities. You only need 40% to win two out of four seats, which makes it a bit easier to win half the seats than it is with 3-member wards. Under a system of three 4-member wards, a party can win half the seats with a uniform 40% vote. Under a system of four 3-member wards, a party needs to achieve 50% in two of the four wards to achieve the same outcome.

It’s reasonably common for councils with 4-member wards to also have a directly elected mayor to produce an uneven number of councillors and effectively break ties – see Wollongong, Lake Macquarie, Wollondilly and Shoalhaven, and Ryde is considering such a switch in 2021. Labor won half the seats on Wollongong council in 2017 with just over 40% of the primary vote, and were only deprived of a majority by losing the mayoralty.

I have also mapped these different magnitudes across the state.

There is a serious urban-rural divide. Urban councils tend to have wards (most commonly 3) while rural councils tend to be unwarded. In the Greater Sydney region there are only two large councils without wards: the City of Sydney and the City of Campbelltown. There are also a handful of small councils that escaped amalgamation in 2016: Burwood, Strathfield, Canada Bay and Mosman.

In rural areas, 9-member wards are the most common, as explained above. But there are rare 12-member councils and quite a few 11-member councils. It’s quite common on the mid-north coast of NSW to have 8 councillors elected alongside a directly-elected mayor.

There are a surprising number of magnitude-2 councils in rural NSW, including one created in 2016. Dubbo, Tenterfield, Walcha, Dungog and Lachlan all follow this model. These councils tend to have one dominant town, and 2-member wards allow for the creation of a “rest of the LGA” ward (never mind that this minority would have more options with an at-large election).

The magnitude can often be influenced by limitations imposed by legislations. Councils must have between six and fifteen councillors. The most populous councils (including most of the newly-amalgamated councils) have fifteen members. Fifteen fits nicely with three-member wards. But the mayor is counted in that 15-councillor limit, so if you have a directly elected mayor you can only elect up to 14 other councillors, and 14 is an awkward number. It can only be divided into two wards of 7 or seven wards of 2. So councils with mayors tend to have a maximum of 13 councillors.

When councils change their mayoral election method they often don’t want to change the number of councillors at the same time, so it requires a change to the ward magnitude. North Sydney has switched from three 3-member wards to two 5-member wards with the end of their directly-elected mayoralty, while Wollondilly switched from three 3-member wards to two 4-member wards with the creation of their new directly-elected mayoralty.

Finally, councillor numbers (which dictate options for ward magnitude) seem to require a constitutional referendum to change, so they can be locked in. Switching between a ward system and an at-large election requires a referendum, but changes to the ward magnitude in itself does not. There are three referendums on this topic in 2021 – two unwarded councils are proposing a reduction in the council size, while a third council is proposing an abolition of their wards. But I will return to that topic in another post.

Liked it? Take a second to support the Tally Room on Patreon!


  1. The multi-member ward acts against independents. When standing as an individual, you end up in the unwinnable “ungrouped” list on the right. As a consequence, independents need to find someone to stand with them. Being the second or third candidate in a group is the equivalent of being listed 5 or 6 in a group standing for the senate. Obviously, parties have any number of candidates who are willing to run in an unwinnable position, willing to do so either for the party good or to get in line for next time. An independent needs to find a person willing to do all the work for no reward.

    If councils wish to produce an odd number of councillors, there needs to be both an odd number of wards and an odd number of candidates eg 3 x 3 = 9.

    If the Mayor is elected directly by the people in a presidential style election, the winning candidates are predominately a “celebrity” known across the LGA rather than someone selected from the councillors who know the best administrator.

    Perhaps the solution is to allow individuals to be placed on the ballot paper in the same way as a group is.

  2. I am not a fan of the requirement that groups contain a minimum number of candidates, but I still think higher district magnitude makes things easier for independents, not harder. If you can’t find say 2 candidates to run with you then you’re not going to do well.

Comments are closed.