Last year, to fill the Bermuda Triangle between Christmas and New Year’s Eve, I revived my tradition of making predictions for the coming year on Twitter, but with a twist: I asked followers to send me in questions about the year to come and offered predictions in response. This was a lot of fun, and I’ll be doing it again this year, and there’s a serious purpose too - predictions provide a great way to test the limits of our knowledge and identify our biases.
Wrong predictions are even better than right ones in this regard, provided we can figure out why we went wrong, and learn from it. Stephen Bush at the Financial Times is an established master of this “self-audit” format, and his regular reviews of what he got right and wrong in the past year (for example 2022 pt 1, 2022 pt 2, 2021, 2020) are both entertaining reads and great examples of how to reflect on your own biases and learn from mistakes.1 So here’s a look back at the predictions I made in the final days of 2022, with some thinking on what I got right, what I got wrong, and why.
THE HORSE RACE: “Will the polls begin to narrow in 2023?” “Will the Tories be better or worse placed to fight a general election in 2024 a year from now?” “What will the poll averages for December 2023 be?” “How many weeks will the Conservatives be less than 10% behind in the polls?” “Will Conservatives consistently go below 20% in the polls?”
My answers were “Yes, the polls will narrow”, because they usually do at this point in a cycle and “Yes, the Tories will be better placed to fight an election”, because they could hardly be worse placed than they were in December 2022, after losing two PMs in a year. I predicted that the polling would narrow from Lab 48 Con 26 on the New Statesman poll aggregator on 28th December 2022 to Lab 43 Con 28 on the same day in 2023. I said the Tories might manage two weeks in the year where they trailed by less than 10 points in the polling average, just through the combination of general averaging and outliers/volatility. I saw little chance of the Conservatives dipping below 20% , reasoning they were likely already down to their most committed core post-Truss.
How did I do? Well, there’s still a bit of time on the clock, and the New Statesman doesn’t seem to be updating its poll averages very often. But if we swap in the more up to date figures from Wikipedia (see below), there is evidence of some narrowing in the polls this year. This is more due to a Labour slide that a Tory rebound. Labour were averaging around 47-48 in December 2022, whereas so far in December 2023 they’re average around 42-43, a drop of about 5 points - in line with my prediction. The Conservatives, however, are averaging 25-26 so far - if anything slightly lower than a year ago. However, the Tories never came close to a substantial run under 20 per cent, though a couple of outlier polls did put them on 19 per cent.
I got three things wrong: the polls narrowed less than predicted, there was less volatility than I had expected, and (while this is more subjective) the Tories don’t look much better placed to fight an election now than a year ago. The first two errors stem from the same source - I expected recent polling volatility to continue, but it didn’t. As it turned out, the 2023 polling was remarkably stable - in this respect, at least, the expectation that a Sunak ministry would bring calmer and more predictable politics has proved accurate.
This stability does not, however, seem to be due to a more effective or organised Conservative party. Instead the mix of inconsistent messaging and strategy, basic incompetence and constant infighting that has characterised government in this country since the partygate scandal broke has continued, albeit with fewer really whopping scandals or blunders under Sunak. I think the general tendency for governments to recover in the polls as elections approach depends at least to some extent on past governments’ ability to deliver the basics: devising and agreeing a clear strategy, and uniting behind a leader with internal authority to deliver that strategy. Those basics have been absent in 2023, as they were in 2022. The recent total mess over Rwanda doesn’t suggest Sunak will be able to deliver these basics any time soon either, which doesn’t auger well for recovery in 2024.
Poll averages this Parliament so far
Source: By Ralbegen on Wikipedia Sources for individual data points can be found here: CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=90612156
THE UNDERCARD: “What will the polling for Lib Dems/Greens/ReformUK look like?”
One reason I overstated the expected shift in the big parties’ polling was failing to think through the logical consequences of my minor parties predictions. I predicted both ReformUK and the Greens would gain ground, and that the Lib Dems would be basically flat. But such votes had to come from somewhere. I should have seen that if multiple small parties go up, the big parties will go down.
In fact, the smaller parties advanced more than I expected. The Lib Dems are currently polling around 11 per cent, up two points on the same stage last year (though still below their 2019 result). The Greens are on around 6 per cent, up a point on last year (and double their 2019 showing). But the biggest gainers have been ReformUK, who are currently polling around 10 per cent, nearly double their score a year ago.
I was a bit bearish on Reform’s prospects a year ago, reflecting an overweighting of the problems posed by the loss of their previous biggest assets - Farage, Brexit and the Brexit Party brand - and an underestimation of the opportunities presented by a Conservative government eager to talk up the problems of immigration while having no coherent strategy to resolve them. While ReformUK continue to underperform their polling in local and by-elections - raising doubts about both polling methods and the party’s organisation - they have bright prospects in the year ahead if, as seems likely, the Conservatives continue to put the spotlight on their failings on immigration. Prospects will be even brighter if, as also seems plausible, Nigel Farage returns.
I was bullish on Green prospects a year ago as I felt that Starmer’s aggressive tacking to the centre-ground and eagerness to neutralise any attack line would leave a gap in the market for a party pitching to disgruntled progressives and social liberals. I think that gap is still there but, so far, the Greens haven’t really capitalised. But they have had a record year in local elections. I suspect their happiest days lie a little further ahead, as right now a strong desire to evict the Conservatives is probably keeping many of their natural voters in the Labour camp. They may get a big boost if and when they can pitch to progressives disappointed with a Labour majority government.
ELECTIONS: “GE this year?” “GE in 2024?” “Will COVID enquiry hasten early election?”
My answers were “No”, “Yes” and “No” - three correct but not difficult predictions. The fact that multiple people put these questions points to pervasive short-termism in our political discussion - a lot of people imagined that the disruptions of autumn 2022 would hasten an early election whereas my view was always that turkeys would not vote for Christmas, so a party which had collapsed in the polls or was being put through the wringer in a COVID enquiry wouldn’t be in a hurry to go to the electorate. That remains my view now, as once again chronic short-termism is leading to regular speculation about a May election - or even rather fantastical stories of a winter “stop the boats” election. Stranger things have happened. But not often.
BORIS “Will Boris make a comeback as PM/make a leadership challenge?” “Will Boris face a severe sanction from standards committee, triggering a recall election?
Myopia was also evident in politicos’ enduring Boris obsession. As it still is. No, Boris is not the Messiah. He’s a very naughty boy. Voters didn’t like him, voters still don’t like him, and by the end of 2022 sufficient Tory MPs also didn’t like him to make any comeback more or less impossible outside of Nadine Dorries fantasy fiction (sorry, “explosive behind the scenes account”).
Johnson himself tried his best to make this obvious by resigning from the House of Commons rather than face a recall petition (something I did not predict!). Yet here we are, a year on, still reading stories about how Boris will be parachuted into a safe seat (with Nigel at his side!) to save his party. He isn’t. He won’t. He’s gone. Get over it.
Further commentary with respects the prospects of a Boris Johnson ‘comeback’
“E's not pinin'! 'E's passed on! This parrot is no more! He has ceased to be! 'E's expired and gone to meet 'is maker! 'E's a stiff! Bereft of life, 'e rests in peace! If you hadn't nailed 'im to the perch 'e'd be pushing up the daisies! 'Is metabolic processes are now 'istory! 'E's off the twig! 'E's kicked the bucket, 'e's shuffled off 'is mortal coil, run down the curtain and joined the bleedin' choir invisible!! THIS IS AN EX-PARROT!!
SCANDAL & RECALL: “How many by-elections will there be in Conservative seats due to recalls or scandal?”
I said the following “So far this Parl there have been 9 by-elex triggered by resignations (driven either by scandal or to take up another job) but no recall petitions. With govt unpopular & divided, departure will be attractive for Con MPs, so I'll predict another 4 happening.” I got pretty close. In the end, we got five (listed below). The Cameron recall petition reform has proved to be one of the more consequential legacies of the Coalition, generating a now well understood and regularly exercised mechanism for removing scandal tainted MPs. Before recall petitions arrived, the frequency of by-elections had been in long term decline - parties sought to avoid them, and rising life expectancies meant less contests triggered by the grim reaper. Now they are on the rise again. I expect we may get as many if not more in 2024, as Conservative MPs seeing defeat approaching head for the exits. There are already likely to be two more in the first months of the year.
By-elections in Conservative held seats 2023
Uxbridge (Boris Johnson, scandal, resigned rather than face a recall petition)
Selby and Ainsty (Nigel Adams, he resigned due to being omitted from Boris Johnson’s New Year’s Honours list - whether such a fit of pique counts as a scandal I leave to the reader)
Somerton and Frome (David Warburton, resigned following a lengthy but unfinished investigation into a range of behaviour sufficiently dubious that I think we can classify this as “scandal” )
Mid-Bedfordshire (Nadine Dorries, resignation in a huff over Johnson honours list and to make time for researching very serious and well grounded insider accounts of politics)
Tamworth (Chris Pincher, who eventually resigned rather than face a recall petition, due to a scandal whose botched handling helped bring down Boris Johnson)
From this list we can safely conclude that the recall petition has become one of the more influential legacies of the Coalition government. At least three seats changed hands in 2023 due to this mechanism - Margaret Ferrier lost her seat after a recall petition triggered by her breaking of COVID regulations, while both Boris Johnson and Chris Pincher resigned rather than go through the process. Tory politics would be even more volatile right now if Johnson was still in the Commons, and without his predecessor’s recall procedure, he might well be. And the recall petition is certain to continue bearing fruit in 2024 - with one by-election already certain after a petition to recall Peter Bone in Wellingborough succeeded this week, and other quite likely in Blackpool South, after Scott Benton was handed a sanction severe enough to trigger a petition. I doubt these will be the last - and if Conservative polling remains dire some MPs may head for the exits early even if scandal and sanction doesn’t force their hands.
PERSONNEL: “How many Chancellors? How many Home Secretaries? Which Cabinet minister will lose job first? How many Housing Ministers? Lee Anderson in a major role?”
I said Hunt would stay, because Sunak had strong incentives to keep him (steady hand at the tiller) and Hunt had little incentive to quit the most powerful job he’s ever going to get in politics. As indeed it has proved.
On Home Secretaries I said “I’m leaning towards “one”…but narrowly...Sunak will want to maintain stability at the top and Braverman would be a magnet for revolt on the backbenches. But she looks awfully error prone and HO is a tough brief.”
This came out pretty accurate - Sunak tried to keep Braverman in the tent but ultimately the chaos she kept generating proved too much. She has already demonstrated her capacity to cause headaches from the backbenches.
I went for Barclay (Health) or Harper (Transport) as first cabinet minister to lose their jobs, anticipating that the public sector strikes and transport chaos would claim one or both of them. Barclay did eventually lose the health brief to a more emollient figure, but he wasn’t first to go (that was Dominic Raab, who resigned following allegations of bullying). I confidently, and wrongly, predicted Lee Anderson would not be given a major job. I am not sure elevating him to Deputy Chair has done much for Sunak’s prospects.
I predicted two housing ministers in 2023 - instead we got three: Lucy Frazer lasted less than four months, Rachel Maclean managed just over nine months, and Lee Rowley is now one month into his tenure (though he did get a six week stint last autumn too - a kind of ministerial work experience placement). This brings us up to 16 housing ministers in 13 years of Conservative led government. Good thing there isn’t a housing and planning crisis in need of sustained political attention.
POLICY: “Will the NI protocol and Britain’s trading relationship with Europe be resolved? Will Labour announce a significant pro-EU policy shift? Will housing become more salient in polls? How disruptive will voter ID laws be?
On the NI protocol, I correctly predicted that Sunak’s strong desire to put the issue to bed, and pressure from the US for a deal, would result in a breakthrough. So it proved. This, of course, did not “resolve” Britain’s trading relationship with Europe as that relationship is a process not an outcome. So, like it not, we will be talking about it for many years to come. Which brings us to Labour…I predicted Labour would not announce any major pro-EU policy shift, as they really had no strong political or electoral incentive to do so, and would not want to jeopardise their rapidly improving standing with Leave voters. So it has proved, but the year has also seen a larger shift in opinion about Brexit than I would have expected, in particular among the growing group of Leave voters who now see Brexit as an economic failure. A growing majority for Rejoin will put pressure on a Labour party whose own support is very strongly pro-EU. I suspect they will continue to resist such pressure unless evidence emerges that doing so is harming them in the polls.
I predicted that housing would rise in salience, and so it has, hitting its highest score in over four years on the IPSOS-MORI issues index last month. With the planning system borked, house building levels still dismal, rents rising fast, and millions coming off fixed rate mortgages onto much higher interest rates, I suspect housing will continue to rise up the agenda in 2024. Keir Starmer’s conversion to the YIMBY cause may come to look prescient.
I predicted voter ID laws would have little major impact on local elections, as these attract high engagement voters who are more likely to be aware of the new rules. The effect does seem to have been pretty limited, but a general election is a very different affair, with more potential for problems. I think there is a good chance of some chaotic scenes at polling stations in areas with lots of less informed or engaged voters - or indeed with large numbers of poorer elderly voters, as the elderly are among the groups less likely to have the most commonly accepted forms of ID (such as passports).
LOCAL ELECTIONS: “How bad will locals be for Cons? Will Sunak face leadership challenge if they’re bad?”
I chickened out of making a specific prediction on seat losses. While I did predict there would be a lot of anti-Conservative tactical voting, I’m fairly certain any number I gave would have been too low. Which points I think to another theme of the year for me - I like many others have been reluctant to take the growing body of evidence that things are truly dire for the Conservatives at face value. Two of the more reliable heuristics in British politics are that the Conservatives always find a way back, and Labour always find a way to disappoint. Therefore month after month of massive Labour poll leads - a situation which would have looked wildly implausible as recently as summer 2021 - still has an air of unreality about it. Surely the Tories will bounce back as polling day approaches. Look at the don’t knows. Look at Starmer’s mediocre approvals. Look at the improving economic figures. Well, maybe.
But, maybe not. I recently went back to the Nuffield election book on the 1997 contest to see how the mountain of evidence of a coming Labour landslide was treated in the dog days of the major government. It was the same story as we see today. Plenty of people simply refused to believe the evidence before them. Conservatives dismissed it out of partisan hope - “the voters will see sense, we are the greatest vote winning machine in political history, the economy is coming good, Blair’s a phony”. Labour dismissed it out of partisan fear - “the press will kill us in the campaign, the Tories will dredge up ancient fears of tax hikes and inflation, the economy is coming good, Blair’s a phony”. Neither side really confronted the obvious truth - that a once in a generation political shift was imminent - though the evidence was there for all to see. Perhaps we are at a similar moment now. I would certainly rate the chances of an historic Tory collapse and a massive Labour victory much higher now than I did a year ago.
THE SCOTTISH PLAY: Support for the SNP? Will Sturgeon still lead SNP?
While Britain wide polling was stable this year, there was one major sea change in opinion, which I completely missed. I very confidently predicted that Sturgeon would continue her dominant leadership of the SNP and depart on her own terms, probably after the general election. I was also sceptical that Scottish Labour would make any further inroads into the SNP’s support, reasoning that much of the unionist low hanging fruit had already been plucked. I was completely wrong on both counts.
Polling in Scotland during this Parliament
Via Wikipedia By Ralbegen - CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=111137453
Sturgeon’s departure, and the subsequent slump in SNP support, provide an important lesson. Events can upend any prediction. The unusual stability in British polling over the past year in part reflects the absence of any massive external or internal shocks to the system - there has been nothing of the magnitude of the COVID pandemic or the Ukraine invasion this year, nor has there been political drama on the scale of Johnson’s collapse or the kamikaze Truss premiership. Not so in Scotland, where a caravan drove a coach and horses through Scottish politics, and helped end the career of the SNP’s most successful and popular politician.
There is no guarantee that the relative calm of 2023 will continue - indeed, given the feverish state of Tory internal politics, there are good reasons to expect a stormier year ahead. But predicting when, where or in what form shocks will arrive is beyond the capabilities of any electoral Nostradamus. Expect the unexpected - it will happen soon enough.
That seems a good note on which to end this prediction review. The big lessons I’m taking away from this year’s exercise are to resist wherever possible short term thinking, focus on the bigger picture, but also remember that past patterns will only repeat if the underlying drivers of those patterns also repeat. Governments which unify around a clear message and deliver positive outcomes will tend to recover ground as an election approaches. This government hasn’t delivered these basics so far. There is still time, but not much.
What will 2024 bring? Send me your questions
I’ll be putting up posts on Twitter (@robfordmancs) and on bluesky (@robfordmancs.bsky.social) shortly asking for your questions about the coming year in politics. You can also ask me questions here and I’ll add them to the thread. I’ll do a post in the new year gathering them together and maybe adding a few further predictions of my own.
Now its time to log off - a pierogi laden Polish festive season awaits. Merry Christmas to all Swingometer readers! I’ll leave you with the only acceptable pie chart, at this time of year or any other:
Sam and Lawrence Freedman of Comment is Freed are also great self-auditors. Sam’s audit of 2023 just came out and is well worth a read (as is Comment is Freed in general):
Excellent read as ever. Happy Christmas Rob
Very interesting. Unforeseen events dear boy. Same then. Same now. Tories are done. Lab are bad but no other choice. Yet. SnP are in the toilet. Good. Lets tax the income of consultant surgeons even more, btw we are about 600 down on consultants, but thats fine. Now Dublin seems to want to go to diplomatic war, little Leo trying to play hard ball to the Shinners. Crack on. Wales falls further into self induced chaos. PC have the answer. Ho hum.