A guest blog by Dylan Difford. Dylan is a researcher specialising in elections and voting systems, as well as a regular contributor to the Electoral Reform Society blog.
Disclaimer: Make Votes Matter does not advocate for any form of Proportional Representation. With large enough districts, the Single Transferable Vote system does meet the standards of the Good Systems Agreement. But any voting system should be decided by the people, in an evidence-based, deliberative process.
At the start of May, voters across the UK went to the polls in local and devolved elections. But there was a big divide in terms of how they voted and how fairly their votes were translated into seats. Voters in Scotland and Northern Ireland were able to fully express their voting preferences and were rewarded with councils and a legislative assembly that accurately reflected their views thanks to the Single Transferable Vote (STV). Voters in England and Wales weren’t so lucky – First Past the Post (FPTP) has yet again delivered heavily skewed results and denied many voters a representative of their choice. Excitingly, Welsh councils can now choose to switch to STV, with this year’s results providing a convincing case for doing so.
But what could the results of these elections have looked like if everybody had been given the opportunities afforded to Scottish and Northern Irish voters? To get a taster, let’s project the results of the elections in the 32 London boroughs onto STV, utilising high-quality data on voter preferences from the British Election Studies. It might not be the whole country, but London offers us a rare complete set of whole council elections and many areas that are microcosms of wider types of council found elsewhere in England.
What is STV?
The Single Transferable Vote (STV) is a form of proportional representation that uses preferential voting – with voters ranking individual candidates according to their preferences. Candidates are elected if they exceed a quota based on the number of votes cast, this quota is normally "total votes / (total seats + 1)". For example, in a district electing four MPs, the quota would be roughly 20% or one in five votes.
Any surplus votes above the quota are transferred to each voter’s next preference. If no candidate beats the quota in a particular round of counting, the candidate with the fewest votes is eliminated and their votes are transferred to their next preference until all seats have been filled.
STV elections typically take place in small constituencies of between three and six members. It is used for most elections in Ireland and Malta, elections to the Australian Senate and Northern Ireland Assembly, as well as local elections in Scotland and Northern Ireland.
The Citywide Picture
The main headline of the local elections in London was Labour solidifying their dominance in the capital – they took just shy of two-thirds of all councillors and hold a majority on 21 of the 32 councils. While part of this dominance is genuine – the party has a 16-point lead over the Conservatives in votes – a significant chunk of it has been manufactured by the voting system. The number of seats they won is over 150% of what they would be entitled to under a fully proportional system. Most of this surplus came at the expense of the Greens, who returned just 18 councillors on 12% of the vote.
Had the election been conducted under STV, the number of councillors won by each party would have been far more reflective of their share of the vote. Labour would retain a degree of overrepresentation, though this would be significantly smaller than that seen in the actual results and would be expected for the largest party under a proportional system with fairly small constituencies. The other three main parties would all have been more fairly represented under STV – with the improvement in the Green’s fortunes being particularly striking. Overall, the average vote-seat deviation of the STV projections is 8%, a clear reduction on the average 30% score for the FPTP elections.
STV wouldn’t just mean that councils were more representative of voters in aggregate, but also at a more individual level. At present, only around 56% of London voters have a councillor that they chose. There is also huge variation between parties – four in five Labour voters have at least one Labour councillor, but just half of Conservative voters do and only one in twelve Green voters do. Had the elections been conducted under STV, however, nine in ten voters would have at least one councillor from their first-choice party – including 99% of Labour voters, 92% of Conservative voters and roughly 80% of Lib Dem and Green supporters.
STV would also mean a fairer deal for the parties in terms of which councils they are represented on. At present, no party has councillors in every borough – with no Labour representation in Kingston or Richmond upon Thames and no Conservative seats on seven councils. These projections show a Conservative and Labour delegation on every council, with Lib Dem groupings on 30 (up from 13) and a Green faction on 26 (up from 8).
But it’s important to remember these are local elections and everything isn’t about the London-wide picture. While there isn’t the time to get into the deep and dirty of every council’s results, we can still take a closer look at some of the biggest wrongs perpetuated by FPTP and how PR could have righted them.
Lewisham
Lewisham is one of Britain’s several urban councils that it would be an understatement to say that Labour were the dominant party in. Along with Barking and Dagenham, it is also one of two London councils where Labour faces no opposition at all. But these 100% of seats were won on just 52% of votes – the half of Lewisham electors who didn’t vote Labour have no representation at all, again.
Rather than 54 Labour councillors, Lewisham would have a more balanced picture, with 31 Labour, 14 Green, 6 Liberal Democrat, and 3 Conservative councillors.
Under STV, Labour would retain a majority on the council, but would instead face three healthily sized opposition parties who would be able to hold them to account.
Croydon
Croydon was one of the six London councils to change hands in this set of elections – with Labour losing it to no overall control. But, although Labour remain the largest party in seat terms, the Conservatives actually won more votes – with FPTP producing a wrong-winner result (an outcome that previously befell Wandsworth in 2018).
Instead of its wrong-winner election, under STV Croydon would have had a more proportional result, with 27 Conservative, 25 Labour, 11 Green, 6 Liberal Democrat, and 1 Independent councillor.
With a proportional voting system, like STV, the risk of a wrong-winner result is significantly lower. Had it been in place for this election, the Conservatives would rightfully be the largest party on Croydon council – though their 37% of the vote wouldn’t have been enough to get them a majority.
Kingston upon Thames
The Liberal Democrats have long been one of FPTP’s biggest victims – being serially and significantly underrepresented at Westminster. But, in the suburban south west London boroughs, their old foe has helped them to some rather unfair results. Kingston upon Thames, where the Lib Dems took all but four seats, produced the second least proportional result in London this time (after Lewisham).
Rather than its 44 Liberal Democrat councillors, STV would see Kingston upon Thames with 23 Liberal Democrats, 11 Conservatives, 7 Labour and 5 Green, and 2 Residents’ Association councillors.
STV would have reduced the vote-seat deviation in the borough from 46% to just under 5%, as well as helping the Conservatives, Labour, Greens and a local independent residents’ group to fairer representation on the council.
Kensington and Chelsea
While the Conservatives are increasingly underrepresented in London local government, there are still a few pockets where they hold on to power such as Kensington and Chelsea, which, including predecessors, has had a Conservative majority since 1900. But the borough is no longer as true blue as it once was and the Conservative majority there is now a manufactured one, being won on just 44% of the vote.
Instead of its Conservative majority, the three-way marginal council of Kensington and Chelsea would be represented as such by STV, with 22 Conservative, 18 Labour, and 10 Liberal Democrat councillors.
While it is sometimes possible for a party in the mid-40s to win a majority under STV, that isn’t the case here with the Conservatives only winning a plurality.
First Past the Post: Bad for Westminster, Bad for Local Government
These projections confirm what we’ve learnt from Scotland since 2007 – replacing the outdated plurality voting systems currently in use with a proportional voting system like STV offers a significant opportunity to improve representation in local government. It makes councils and parties more representative of their voters and gives more of those voters a councillor that they actually voted for and support. It brings an end to unrepresentative one-party fiefdoms and guarantees opposition parties fair representation – giving greater ability to hold council leaders to account.
Though these simulations are limited to London, there is no reason whatsoever to believe that these same benefits would not apply to the rest of England and Wales as well.