All this has happened before. That doesn’t mean it’ll happen again (or: Why we need to be thinking about a wipeout)
All this, as Admiral Adama intones in Battlestar Galactica (the greatest modern sci-fi series), has happened before. And all this will happen again.
At times it seems that’s the default status of our political commentary. To look back at scarce data points and declare that because this pattern took place in, say, 1997 it must be followed in 2023/4. The only way for Labour to win is to ape Blair. 97 is the worst it can get for the Tories.
Thus, the insane debate a few months back on whether we’re in the 1992/1997 timeline; can the Tories close the gap like Major? Or is it a Blair style ride to victory for Labour? No other avenue is contemplated. I took on that particular debate in my blog back in Feb, one whose basic conclusions I think have held up well.
At that time it was argued that Rishi could close the gap and that the polls would close. That we were in 1992, not 1997. 9 months on and… the Tories are arguably in a worse electoral spot. In February they were 21% behind Labour in Britain Predicts poll of polls. They’re now… 20% behind and the gap has widened since the summer.
More than this the one positive for the Tories in Feburary, Rishi’s ratings, has collapsed. In Feb he was on -29, not good but not catastrophic by modern political standards. There was the prospect he could drag the Tories' ratings up to meet him. This hasn’t happened; his approval in October hit -40, it's at -34 this month, and pretty much every one of his polling rating has collapsed. By comparison Starmer is on -14%. Good luck with a presidential contest.
To underline the point today's Yougov has Labour with a 24% lead, the Tories hitting 23%, one of a number of polls showing them on below 25%. Put that YouGov poll into electoral calculus and, even without tactical voting, you get this:
Stick in tactical voting at a moderate level and the Tories are reduced to 35 seats.
Obviously this is an extreme example, but equally it's not a unique one. Put in any present poll and you’ll get a similar result; Electoral Calculus’ own poll of polls has the Tories on 133 seats. The Tories haven’t hit over 30% in any poll since June. Labour’s polling has remained broadly stable with a average 15%+ lead since Trussageddon over a year ago.
But, all this has happened before, right? In 1997 Labour also chalked up huge leads but were pegged back by the time of the election. In IPSOS/MORI at the same point in the cycle, for example, Labour notched up a 27% lead. Other polls around the same period had their lead between 16% — 34%. A year later Labour won by 12.5%. And if all this has happened before, it must happen again.
Well… maybe. When we look at 1997 it’s worth considering two key factors; the Tories in 1997 presided over a robust economy, which provided a reason for polling to close. 1997 also saw the polls overestimating Labour’s lead, making those large leads in 1996 look a little less substantial in retrospect.
The economics seem particularly key. By the time 1997 rolled around the UK had seen three years of strong growth, with falling unemployment, low inflation/interest rates and a smaller deficit. This gave Clarke room for a relatively aggressive budget in 1996, cutting tax and giving the Tories at the very least an electoral case. In addition to this Major, while unpopular, was at the very least a proven politician who had won an election and equally Blair in terms of trustworthiness — the electorate's verdict was, to quote Kellner, ‘nice man, shame about the party’. In short: There were decent reasons to vote Tory. A argument.
Today none of those factors apply. The economy is in a poor state, Sunak’s ratings are underwater, as are those of his government and party. The party is deeply split with cabinet ministers freelancing with no attempt to reign them in. And the polling appears far more accurate than in the pre-1997 period, bourne out by repeated by-elections.
So, while there are some similarities between 1997 and today, there are as many or more differances. All this hasn’t happened before, and there is no reason for the same thing to happen again.
Our obsession with the past means we view 1997 through the frame that it was a disaster for the Tories, the worst possible result. But what if it wasn’t? What if it was the best-case scenario? After all, it was the largest polling closure in recent history, driven by very strong economic growth driving it. What happens when you remove that?
All of which is to say this: People are massively underestimating the chance of a wipeout at the next general election. Not just a major loss like 1997 but a Canada 93 scenario where the Tories are reduced to <100 seats.
Our sense of political normality rebels against this. You look at the seats which would need to fall for this to happen and it seems unreal; my home seat of Skipton and Ripon, for example, would go red and the concept of a Labour MP for Kilnsey is just… weird. But at this point it’d be remiss t just dismiss the possibility just because it hasn’t happened before: A Tory MP for Bolsover would have seemed bizarre in 1997, when Labour had a 27k majority, but there we are. The incumbent side always won national referendums until they didn’t. Labour would always hold Scotland. Labour can’t win Tamworth or Mid-Bedfordshire.
Politics is broadly causal. There aren’t natural laws, but instead voters who respond to the situation they are in, the wider picture and what they see from politicians. So its always good to think about the why and, put simply; Why would the polls close from here?
The only real arguments I can see are: ‘Don’t know’ voters return home to the Tories as we get closer to the election, and ‘Reform votes’ are squeezed in an election. The former has some issues — Don’t know voters aren’t exclusively drawn from ex-Tories and more would consider Labour than the Tories — and it’s unclear if those Reform votes exist in real life to be squeezed (and any effect may well be cancelled out by Labour squeezing Green votes and tactical voting).
Aside from that it’s hard to see what drives poll closure over the next 11 months. The broader strategic picture looks very poor for the Conservatives. Consider:
- The economy is likely to remain flat-lining. There are strong indicators of a recession. Inflation and interest rates will remain high, directly attributed to the Tories; any rise in real wages will be minimal and there is precious little fiscal headroom for any pre-election sweeteners. Rishi’s pledge on inflation is likely to be missed and is misunderstood anyway
- The NHS winter crisis is likely to be more severe that last year, when ambulance services basically stopped, and there is no sign of strike action disappearing. The waiting lists will continue to rise with Rishi’s pledge on them looking like it’s been abandoned. Strike action will continue.
- Public services will continue to deteriorate. Our prisons are overflowing and the criminal justice system is collapsing. Police don’t attend crimes. Schools will continue to struggle with RAAC. The trains are a mess and will continue to get worse.
- Small boats will continue to arrive, though perhaps in smaller numbers. Conflict over the asylum, immigration and wider culture wars escalate.
- The housing crisis will continue to develop, with more mortgage holders, landlords and renters being hit by a mix of rising costs, interest rates and supply squeezes
These are just a few of the issues that the government will continue struggle with over the next few months — in broad order from most to least important. You could add another half dozen, from foreign policy to parliamentary challenges. For any government, this range of crises would be overwhelming and we’d expect it to impact their polling negatively. To expect the government to make gains in this environment seems… unlikely. And that is without considering any black swan events or crises that are likely to spring up.
It’s also worth considering the election itself and how it might play out. What exactly is the Conservative line going to be? Given the above, what is its argument for another term, beyond trying to undermine Starmer? They will be led by a deeply unpopular leader (who has another year of his ratings dissolving), face massive and seemingly efficient tactical voting and a well-funded, strategic Labour party with a leader who is tactically leaden-footed but strategically and politically strong.
None of the above screams poll closure to me. The reasons for the gap to hold or widen seem to outweigh those for a closure. The biggest reason for the polls closing seems to be ‘its what they do;’ it’s happened before, and will again. But as I argued in February this isn’t always the case — we’ve seen gaps go out in the past.
I’m not saying it’s certain that the Tories will be wiped out in 2024. My personal expectation right now is a 1997 style (or 2019 for Labour) defeat with the Tories reduced to 200 or so seats. The mountain for Labour to climb from 2019 is vast, the electoral geography hard and they won’t be able to spread their resources too thin. Some votes will return to the Tories and I don’t think tactical voting will be as efficient as we’ve seen in by-elections.
But what we should be doing is taking the prospect of a Tory wipeout seriously and seeing it as a very plausible outcome — far more so than a Tory majority or hung Parliament — rather than dismissing it as a outcome. This should inform our commentary and expectations. And with every day that the polls continue to hold it becomes more likely. Their predictive power increases as we get closer.
Our present discourse doesn’t reflect this reality at all, but it should. We need to be focusing on what the Tories can do to avoid wipeout, not to win another election. How would Starmer behave with such a huge majority? What would it enable Labour to do? Who would be elected? None of this is being covered or event discussed. We seem to be sleepwalking towards a major (and quite obvious) political change with very little awareness of whats happening.
But whatever happens it’ll be very different to what has come before. So say we all.
