When Boris Johnson was re-elected as London mayor in 2012 he came top of the pile, by taking 44 per cent of “first-choice” votes: a lead of 3.7 points over Ken Livingstone. If it had been a first-past-the-post election, like that we use to elect MPs, he’d have been home and dry with that.
But since London’s mayoralty was created in 2000, the job has been filled using the “supplementary vote,” where electors also cast a “second choice” which comes into play when no candidate gets more than half of the first preferences. Specifically, second preferences cast for the top two candidates from voters whose first-choice pick comes in third or worse get added to their tally—so in 2012, what mattered were votes for either Boris or Ken from voters whose first preference was for neither Boris nor Ken. Ken won 102,355 of those to Boris’s 82,880. Boris still won overall, but—on a like-for-like comparison with those first-preference votes—now with a reduced lead of 2.8 percentage points.
After the mayoral elections this May, it was reported that the government would press ahead with plans to switch all elections for combined authority mayors (metro mayors), the mayor of London, and police and crime commissioners (PCCs) from the supplementary vote towards the old first-past-the-post system—reverse electoral reform.
Such a move will affect a lot of voters. Some 41 per cent of people in England now live in London or an area with a metro mayor. Everyone in England and Wales is represented by a PCC, or a mayor with the same powers (as in London, Manchester and West Yorkshire). People may not know it, but PCCs have the power to appoint and dismiss chief constables and set police budgets and priorities. They are increasingly voted for on partisan lines, and so the partisan effects of electoral system change matter.
Johnson’s 2012 experience with the supplementary vote system was not unique. He also won fewer of the transferred second-preference votes than Livingstone when he was first elected in 2008. Indeed, Conservative candidates for London mayor have always gained fewer votes on the second round than their opponent. In 2016, Sadiq Khan won nearly twice as many of the transferred second-preference votes as Zac Goldsmith did.
Nor is this just a London thing. In this month’s local elections, most of the supplementary vote elections involved a transfer round because no candidate won more than half the first-preference votes: such a transfer was triggered in seven out of 12 metro mayor contests, and 27 of the 39 PPC elections.
Whenever they were involved in a transfer round—and because they are normally placed either first or second, they usually were—Conservative candidates got fewer transfer votes than their rival, with just two tightly fought exceptions. (In the Hampshire PCC election the Conservative won 53 per cent of the transferred votes, while in the Surrey PCC election the Conservative candidate won 42,856 to 42,803 for the Lib Dems—a virtual transfer tie.)
Elsewhere, Conservatives were typically trounced on transfers. On average across all the PCC elections, when a Conservative was involved in the running for second preferences, they took just 41 per cent of transfers, compared with 58 per cent for Labour candidates and 64 per cent for the Lib Dems who made the “second round.” This failure in picking up transfer votes was even more stark in the five metro mayor contests that went to an instant run-off featuring the Conservatives. There, on average, the Tories won little more than a third of transfers.
This is hardly surprising, given that Labour, Liberal Democrat, Green and Plaid Cymru supporters tend to prefer each other’s parties to the Conservatives. The problem of division of among the socially liberal left is getting worse, as my analysis of this month’s local election results for Prospect demonstrated.
The supplementary vote alleviates that problem somewhat, as supporters of parties on the liberal left can put each other down as their second preference. But that only goes so far. Although most of the metro mayor and PCC elections this month did go to a second round, the winner of the first-preference votes normally has a big enough lead to see them elected even if they win fewer transfers than the runner-up. Just one mayoralty (Cambridgeshire and Peterborough) and one PCC election (North Wales) flipped on the transfers. So it is relatively rare for it to make the difference, but in both these cases the upshot was that a Conservative win on first preferences was overturned after transfers.
In North Wales, the Conservative won the first-preference vote with just 32 per cent. Labour was not far behind, with 29 per cent for first-preference votes. Labour also won some 66 per cent of the transferred votes, giving it 52 per cent of the total votes after transfers. If first-past-the-post had been used with the same first-preference votes, then the Conservative candidate would have been elected on less than a third of the vote.
Jacob Rees-Mogg justified a change from the supplementary vote because “first-past-the-post is better for democracy because the most popular candidate wins.” But a candidate with 32 per cent of the vote and the votes of just 14 per cent of the electorate is not obviously the most popular.
What is clear, however, is that unless a lot changes in the structure of party preferences, a switch from the supplementary vote to first-past-the-post would benefit Conservative candidates in England and Wales.
Not in Scotland though, where the additional member system used at Holyrood is even more different from the old Westminster system. But don’t expect the Conservatives to agitate for first-past-the-post Scottish Parliament elections any time soon. Reverse electoral reform north of the border would have produced an overwhelming SNP majority this month, and would likely reduce all other parties to a rump for as far as the eye can see.