Last week’s dramatic by-elections provide evidence of a secret weapon that could boost the Conservatives at the next general election. Deployed effectively it could win them the seats they need to deny Labour victory. But if it blows up in their face, they are doomed to lose heavily.
The evidence comes not from those who voted last week, but from those who didn’t. With turnout below 50 per cent in all three seats, more electors stayed at home than cast a vote. And most of those who voted at the last election but not last week were former Conservatives.
To some extent, it was ever thus. Turnouts are almost always lower at by-elections than at general elections – but not to the extent that is now commonplace. The last time when the Conservatives suffered at the hands of a resurgent opposition was in the 1990s. They lost all eight of the seats they defended in the 1992-97 Parliament. But they did so on average turnouts of 64 per cent. Voters did not just desert the Conservatives, they switched in large numbers to other parties.
That has now changed. Far fewer people vote in by-elections these days. In the eight seats the Tories have defended in the current Parliament (lost six, held two), the average turnout has been just 45 per cent. Such a big drop has multiple causes. it cannot be explained by Conservative abstainers alone. However, what is striking is that the numbers of people voting Conservative have NOT shrunk in line with turnout. In fact the loyalty rates have been much the same: 41 per cent of general election voters stayed with the Tories in by-elections in the 1992-97 Parliament, 39 per cent since 2019. Both figures are terrible, but they are not getting significantly worse.
The difference is what has happened to the non-Tory vote. It rose by-electionsin the Nineties by 5,500 on overage. In the present Parliament the average is for non-Tory votes not to rise at all, but fall by 1,000.
In short, the story is of similar numbers of Conservatives deserting the party in both eras, but with more of them now staying at home than switching to other parties. Rishi Sunak’s problem is still serious, but it is not quite as awful as the bleak prospect that faced John Major. In his era, the evidence from by-elections, confirmed by Tony Blair’s landslide in 1997, was that voters had made up their minds; they deserted the Conservatives in droves for other parties. That has not happened nearly as much this time. Rishi Sunak has the opportunity to woo lost Tory voters back before they cross a political Rubicon and back a rival party.
This is confirmed by recent polls. YouGov has a large database of people who said at the time how they voted in the 2019 election. This enables them to compare what respondents say would do today with how they voted then, according to what they said at the time (rather than rely on the memory of what they did four years ago). YouGov’s latest poll finds that 32 per cent of people who voted for Boris Johnson’s party either don’t know who they would back today (24 per cent) or say they would not vote (8 per cent). That’s the equivalent of 4½ million people who are excluded from YouGov’s voting intention figures.
Decades of experience tell us that people who vote in one election are very likely to vote in the next, and that Conservative supporters are especially determined to use their vote (not least because so many of them are elderly, who consistently turn out more than younger voters).
How many of these lost Tories could Sunak reasonably expect woo back? Given troubles of recent times – from Partygate to high inflation, rising mortgage rates to stalled living standards – if these lost Tories have not yet moved across to other parties, then Sunak must feel he has a chance to win many of them back. A stretched but plausible target would be 2.5m, with the rest distributed in smaller groups – say half a million Labour, half a million other parties and one million sitting out the next election.
That would reduce Labour’s lead over the Conservatives by two million, or six percentage points. That’s 3,000 voters per constituency – enough to save up to 40 seats the Tories would otherwise lose, and change the complexion of the next Parliament.
Of course, Sunak might fail in this endeavour. Perhaps a majority of these lost Tories will prove immune to his blandishments. Either way, the future decisions of the people who stayed at home last week, and the millions like them around the country, will be crucial when we are all asked to choose our next government.
This analysis was first published by the Sunday Times
Peter Kellner. The Politics Counter 