Western Australia’s first step on local council voting reform

The Western Australian Labor government has been recently going through a process of changing the structures of local government elections in the state. The process involves the replacement of first-past-the-post with preferential voting (including with the single transferable vote where more than one councillor is elected in a ward), but it has also involved some efforts to standardise other features of council elections: the use of directly-elected mayors, the size of the council and whether the council uses wards have all become more aligned with the population size of the council area.

In this post I try and explore what they have done and assess how proportional WA elections have really become.

To understand WA’s local council elections, there are three facts you need to know:

• Council elections are held every two years, with roughly half of the council elected at each election for overlapping four year terms. In most cases wards elect the same number of councillors at each election, but not always. A ward could elect three at one election and two at the other, or some small wards could only elect a councillor once every four years. There are plenty of odd-numbered councils without wards where, say, the council elects four councillors at one election and three at the next. Directly-elected mayors serve four year terms, with these elections split between the two election cycles.
• Elections, up until 2021, were conducted using first-past-the-post. It appears to be the case that in multi-member wards, a voter would get to vote for as many candidates as there are vacancies, a system known as plurality block voting. It’s as far as you can get from proportional representation, even moreso than single-member plurality systems.
• Western Australian councils are very small. The state has 137 local government areas, including 29 in Perth. Compare this to 129 in New South Wales, 79 in Victoria and 77 in Queensland, all of which have at least twice the population. There are quite a few inner-Perth councils with a tiny population.

The recent electoral changes have firstly done away with first-past-the-post and replaced it with preferential voting. For mayoral elections and for single-member ward elections, it is a single-member preferential system. For multi-member ward elections or undivided council elections, the old system has been replaced by single transferable vote.

Regular readers will be familiar with the role of district magnitude in proportionality, and the half-term elections have a big impact on the maximum proportionality in WA councils, even after theoretically introducing a proportional electoral system. There are quite a lot of wards represented by two councillors which could at least be slightly proportional, but instead function as separate single-member contests. There are a small number of councils which elect six councillors at a time, but that’s quite rare.

For this analysis, I have tried to collate a dataset of the council structure in each council, both before and after the reforms. Unfortunately the WAEC website is lacking council election results for some very small councils, and it’s made more complicated due to the half-council elections. Overall I was able to make a valid comparison for how each council’s elections will be run between 2019 and 2027 in 88 out of 137 councils, so for some of the analysis further down I’ll just be looking at how things are changing in those 88.

On top of the change in system, there have been three different ways in which council electoral structures have been changed:

• The total number of councillors on each council have been adjusted in line with the total population of that council.
• Small councils have had their wards abolished, while larger councils have been shifting towards having more wards, in most cases electing just one councillor per ward at each election.
• All larger councils have been required to switch to a directly-elected mayoralty.

This page lists every council as part of one of four groups of councils. These groups seem to mostly be based on population size, but not entirely so. Class 1 have the biggest councils, and Class 4 the smallest.

Class 1 and 2 councils have all switched to a directly elected mayor. Of those 88 councils where I have data, just 26 had directly-elected mayors in 2019 or 2021. All of those 26 are maintaining that model, and a further 21 are switching over in 2023 or 2025. Most of those switches were in councils in the first two classes, although four councils in Class 3 also switched over.

All but two councils in the Perth metropolitan area will have a directly-elected mayor in 2023 or 2025. The only exceptions are the tiny Peppermint Grove and slightly larger Bassendean.

The reforms have also imposed a range of council sizes depending on the range of population in the council which has had a downward effect on the average council size. In the 88 councils I could compare, the total number of councillors was 815 after the 2021 elections. This will drop to 770 in 2023, and 753 in 2025. 50 of these 88 councils will be the same size in 2025. Seven councils (including 5 in Perth) will increase by one seat. Another 31 councils will shrink in size. Five councils will lose four seats each, which is quite a lot. The most common option is to lose two seats – 15 councils are in that position.

While there may be some logic in the relative council size, I think there’s case for offsetting any reduced council size in small areas by growing the size of councils which cover larger populations, and that hasn’t happened. The trend is towards smaller councils in less populous areas and councils of a consistent size in more populous areas.

Finally there has been a shift in the type of ward structure being used. Councils in classes 3 and 4 have been obliged to abolish their wards and elect the whole council at large. There hasn’t been such a blanket rule in the opposite direction, but it appears that there has been a shift amongst the larger councils away from having multi-member wards towards having a clear system of single-member wards (with each ward being represented by two members elected at different elections).

Amongst the 88 I have been able to analyse, the number of councils with some multi-member wards has reduced from 26 to 9. The number of councils with single-member wards has increased from 12 to 16, while the number of unwarded councils has increased from 50 to 63.

Almost half of the councils in Perth are now using single-member wards, while in regional WA most councils use no wards.

This means that the same voting system plays out very differently between the two, with most regional elections involving a district magnitude of at least 3, while much of Perth is stuck using single-member wards.

This map shows the change in ward structure for Perth councils, and you can click on each council to see some other stats.

In this sense this resembles the change we’ve been seeing in Victoria, which has led to more proportional elections in regional areas but the end of PR in Melbourne, although the smaller Perth councils will have more proportional elections.

The amount of proportionality is pretty low – at its most proportional it is about as proportional as the least proportional NSW elections, and doesn’t compare to Tasmania. It would be similar to the scale of proportionality in South Australia, which isn’t much.

I would argue that the next goal of reformers in Western Australia should be the end of the double-election overlap. This used to be a common practice around Australia but WA is now the only state that doesn’t elect whole councils at the same time. This would instantly make each ward twice as proportional. Once that is done, there will be more flexibility to draw low-magnitude proportional wards. For example, some of those new two-member wards can be restructured into three-member wards.

Western Australia probably had the worst local government electoral system in Australia prior to this change. Now it has a system that is still probably in the bottom half, but doesn’t stand out as so terrible. There is a lot more to do, but it’s an encouraging first step.

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