Following an initial announcement trailed in March at the conclusion of Part One of the Home Office’s PCC Review,[1] the government confirmed that it would be scrapping the Supplementary Vote, which is used for all mayoral elections in England and Police and Crime Commissioner elections in England and Wales, and extending First Past the Post to these contests.[2] This change was included as a late amendment to the Elections Bill,[3] after the legislation had already received its second reading and was being scrutinised in committee stage.
This briefing describes how the Supplementary Vote works and shows how it increases voter choice and allows candidates to be elected with a stronger democratic mandate than under FPTP.
What is the Supplementary Vote?
The Supplementary Vote is the voting system currently used for mayoral elections in England (in unitary authorities, combined authorities, and London) and for Police and Crime Commissioner (PCC) elections in England and Wales.[4]
Under the Supplementary Vote, voters can express a first and second preference for their two preferred candidates, though they do not have to mark a second preference if they do not have one.
If a candidate obtains more than 50 percent of the vote, they are elected on the basis of first preferences alone. If no candidate obtains more than 50 percent of the vote, the two candidates with the most votes in the first round proceed to a second, ‘run-off’ round, while all other candidates are eliminated. In the run-off, the second preferences of voters whose first-choice candidate was eliminated, are reallocated. The candidate with the most votes at the end of the second round is declared the winner.
The Supplementary Vote means politicians need a wider base of support than under First Past the Post, with the winning candidate having 50 percent or more of votes in the second round. By being able to express a second preference, voters have a greater say over the final outcome of the election and fewer votes are wasted. Voters whose first-choice candidate was eliminated, but who expressed a second preference for one of the two remaining candidates in the run-off, help decide who the eventual winner will be.
Voters are used to the Supplementary Vote
In the press release announcing the decision to scrap the Supplementary Vote (SV), the government claimed that First Past the Post (FPTP) would ‘further strengthen the accountability of elected mayors and PCCs to their electorate, making it easier for voters to express a clear choice’, seemingly implying that voters find SV to be particularly complicated or confusing. The evidence, however, does not seem to suggest this is the case.
The Supplementary Vote has been in use for over 20 years for directly-elected mayors and was chosen by the Conservative-led coalition government for elections to the newly created role of Police and Crime Commissioners (PCCs).[5] Since 2000, when the first mayoral election was held in London, 212 elections have been held using the SV across England and Wales (the latter for PCC elections only).[6]
The Supplementary Vote means fewer votes are ‘wasted’
Unlike FPTP, where only votes cast for the winning candidate matter, fewer votes do not count at all towards an election under the Supplementary Vote. First preference votes cast for the eventual winner and runner-up determine who goes to the run-off, and second preference votes for the two remaining candidates affect the final outcome of the election. Further, second preference votes cast for the eventual winner count towards electing that candidate, providing them with a broader democratic mandate than FPTP.
Table 1: First and second preferences by type of election
| Type of election |
Total first preferences |
First preferences for winner under SV |
First and second preferences for winner under SV |
Votes for winner under FPTP |
Second preferences for remaining two candidates under SV |
| London Mayoral |
13,330,599 |
5,531,554 |
6,344,543 |
5,531,554 |
1,444,302 |
| Combined Authorities Mayoral |
3,159,128 |
1,292,115 |
1,532,416 |
1,309,951 |
425,272 |
| Unitary Authorities Mayoral |
2,296,757 |
898,889 |
1,077,287 |
904,977 |
320,886 |
| Abolished Unitary Authorities Mayoral |
328,493 |
93,067 |
119,983 |
93,463 |
50,707 |
| PCCs 2012 |
4,296,878 |
1,562,533 |
1,983,579 |
1,577,243 |
721,774 |
| PCCs 2016 |
7,893,625 |
3,034,092 |
3,727,662 |
3,034,092 |
1,245,555 |
| PCCs 2021 |
8,091,116 |
3,545,396 |
4,104,024 |
3,545,396 |
1,204,449 |
ERS analysis. Data taken from House of Commons Library (commonslibrary.parliament.uk) and London Elects (www.londonelects.org.uk).
NB: These figures relate solely to elections held under SV, thus excluding de facto FPTP contests and the 52 elections which were won on the basis of first preferences alone. They also exclude elections which were declared void and subsequently re-run (i.e. 2014 Tower Hamlets mayoral election and 2021 Wiltshire and Swindon PCC election).
Table 1 displays ERS’ analysis of first and second preference votes across the 160 elections held under SV which were not won in the first round,[7] highlighting which votes count towards the election and how – by allowing voters to express a second preference – the eventual winner under SV has the support of a larger swathe of the electorate than would be the case with only first preference votes. By contrast, the only votes that would have mattered in the 160 contests had these been held under FPTP would have been the first preference votes for the eventual winner (second column from the right).[8] The final column shows the total second preference votes cast for the remaining two candidates in the run-off across all elections, which affect the second round of the election and allow more voters’ preferences to have an impact on the eventual election outcome.
The Supplementary Vote means winners need greater support
In addition to allowing voters to express two preferences, if they wish to, and to ensuring the winning candidate has broader support than under FPTP, the Supplementary Vote guarantees that the winning candidate receives a clear majority (at least 50 percent of the vote) in the second round. Unlike other transferable and preferential voting systems, this is far from perfect, as the winning candidate’s vote share at the end of the second round does not necessarily represent a majority of the overall vote share.[9] But it is still an improvement on FPTP, where the winner only needs to receive one vote more than the runner-up and can win, as seen in FPTP elections to the Commons, with as little as 24.5 percent of the vote.[10]
Table 2: Comparison of vote share under FPTP and SV by type of election
| Type of election |
Average vote share of winner under FPTP (first preferences) |
Lowest/highest vote share of winner under FPTP (first preferences) |
Average vote share of winner under SV (second round) |
Lowest/highest vote share of winner under SV (second round) |
| London Mayoral |
41.2% |
36.8% / 44.2% |
55% |
51.5% / 57.9% |
| Combined Authorities Mayoral |
39.4% |
27.3% / 48.7% |
56.5% |
50.4% / 74% |
| Unitary Authorities Mayoral |
40.1% |
22.8% / 49.6% |
59.4% |
50% / 74.3% |
| Abolished Unitary Authorities Mayoral |
27.9% |
21.9% / 42.1% |
56.8% |
50.6% / 71.6% |
| PCCs 2012 |
36.4% |
24.8% / 48.4% |
57.2% |
51.3% / 67.8% |
| PCCs 2016 |
38% |
24.4% / 49.9% |
58.1% |
50.5% / 68.1% |
| PCCs 2021 |
43.3% |
31.6% / 49.97% |
57.2% |
51.0% / 68.2% |
ERS analysis. Data taken from House of Commons Library (commonslibrary.parliament.uk) and London Elects (www.londonelects.org.uk).
NB: As in the previous table, figures relate solely to elections held under SV (not de facto FPTP contests) and exclude elections won on the basis of first preferences alone. For the 17 elections which would have had a different outcome under FPTP, the vote share of the FPTP (not SV) winner is used for the FPTP results.
If we look at the average share of the vote the winner received under SV and compare it with what the winning candidate would have received had these elections been held under FPTP (table 2), we can see that, under SV, the winning candidate can at least claim to have received a majority of first and second preference votes cast for the two remaining candidates in the run-off. Under FPTP, by contrast, candidates would have won with as little as 21.9 percent of the vote (the 2005 election in the now abolished unitary authority mayoralty of Torbay).
While these figures are not directly comparable – for FPTP, they refer to the winner’s share of the vote as a proportion of first preferences alone, whereas the SV figures relate to the winner’s share of the vote in the second round of the election only – they discredit the argument that ‘losers’ can win under SV and have less accountability to the people who elect them.[11]
Rejected and spoilt ballots
One of the reasons underlying the government’s proposed changes is that the Supplementary Vote allegedly leads ‘to hundreds of thousands of votes being wasted.’[12]
There was a high number of spoiled ballots at the 2021 London mayoral contest – but this has been widely attributed to the ballot design,[13] rather than considered to be a direct result of the electoral system used. In May 2021, a record number of candidates standing to be mayor of London meant that a new ballot paper design had to be used to allow for the inclusion of all 20 candidates. Rather than presenting candidates as one continuous list, as had been the case in previous London mayoral and other SV elections, the ballot presented candidates across two columns.
While the proportion of rejected ballots in London was an unusually high 4.3 percent, in all SV contests held in May 2021, the rate was 2.7 percent overall. Rather than being the rule, the high number of spoilt ballots in the London mayoral election this year was uncharacteristically high – for comparison, the rate of rejected ballots was 1.9 percent in the 2016 London mayoral race.
Voters spoil ballots for a number of both intentional and unintentional reasons, and one cannot ascribe this solely to the electoral system. The Electoral Commission’s post-poll research on all local elections in England in 2021 (held under both SV and FPTP) found that around one in 10 found it difficult to fill in their ballot paper, the reasons varying among unclear instructions, the ballot paper being complex/confusing, confusion caused by voting in multiple elections, and there being too many candidates.[14]
A worse deal for voters
Scrapping the Supplementary Vote in favour of First Past the Post would be a step back for voters. The Supplementary Vote has been used for over 20 years and, while far from perfect, is an improvement on First Past the Post, which forces voters to vote tactically and prevents them from expressing a second preference.
Rather than making elections more accessible and enhancing participation, this latest addition to the Elections Bill would lead to voters having less of a say in our democratic processes.
[1] Home Office (2021). Home Secretary to strengthen Police and Crime Commissioner role. 16 March. https://www.gov.uk/government/news/home-secretary-to-strengthen-police-and-crime-commissioner-role
[2] https://www.gov.uk/government/news/first-past-the-post-to-be-introduced-for-all-local-mayoral-and-police-and-crime-commissioner-elections
[3] https://bills.parliament.uk/bills/3020
[4] https://www.electoral-reform.org.uk/voting-systems/types-of-voting-system/supplementary-vote/
[5] The Police and Crime Commissioner Elections Order 2012. https://www.legislation.gov.uk/uksi/2012/1917/contents/2014-07-24
[6] This figure excludes the 2014 Tower Hamlets mayoral contest, which was later declared void, the May 2021 Wiltshire PCC election (which was re-run in August 2021 due to the original winning candidate being disqualified), and four elections which were de facto FPTP contests due to only two candidates standing.
[7] Fifty-two elections were won in the first round, as the winning candidate obtained over 50 percent of the vote, and are therefore excluded from this analysis.
[8] Seventeen contests would have resulted in a different winner if they had been held under FPTP. The FPTP totals in this table, therefore, reflect the first preferences which would have gone to the different winner had these contests used FPTP.
[9] Given the lack of available data on second preference votes for all candidates across all elections, the winner’s share of the overall vote cannot be determined. Thus, the focus is on the winner’s share of the vote at the end of the second round only.
[10] McBride, S. (2017). Record-breaking MP who won in extraordinary circumstances ‘quietly confident’. iNews, 6 June. https://inews.co.uk/news/politics/record-breaking-mp-won-extraordinary-circumstances-quietly-confident-70234
[11] Hope, C. (2021). Exclusive: Losers can no longer win as ‘first past the post’ reintroduced for all elections. The Telegraph, 14 September. https://www.telegraph.co.uk/politics/2021/09/14/exclusive-losers-can-no-longer-win-first-past-post-plans-elections/
[12] Ibid.
[13] Baston, L. (2021). Too many Londoners were confused by the ballot paper in 2021, but preferential voting should remain. OnLondon, 12 May. https://www.onlondon.co.uk/lewis-baston-too-many-londoners-were-confused-by-ballot-paper-in-2021-but-preferential-voting-should-remain/
[14] Electoral Commission (2021). Report on the May 2021 elections in England. https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums/past-elections-and-referendums/police-and-crime-commissioner-elections/report-may-2021-elections-england