Some psephologifying and random assorted thoughts on what future elections might have in store.
House of Commons, Scottish Parliament, Scottish Independence.
Make your pick and enjoy. Comment is free.
So this was the day. Mayxit Day. Theresa May, 13 July 2016 - 24 July 2019, all 1,106 days of it. She took her last PMQs this afternoon and by all accounts it was far short of a flamboyant exit. An outgoing PM suggesting that the Leader of the Opposition should step down too has to be the weirdest PMQ moment ever. I also think that Andrew Bowie's puerile attempt at fixing Theresa's Last Stand certainly did not help and only woodchipped any hopes Wee Andy ever had of becoming Scottish Secretary. The biggest irony of course is the job went to a nobody who had not publicly endorsed any candidate in the leadership race. Guess the Scottish Borista Quartet are fuming, especially Fanboy Ross Thomson who is delusional enough to have thought he had a shot at getting the job. Bully for you, Boaby Snatcher.
Theresa May's legacy will surely be closer to what Nicola Sturgeon described than to what Theresa May herself would want it to be. Especially as she totally failed to see the irony in having her last major speech in Scotland on American Independence Day and totally failed to make her point about the Preshoos Yoonion. It might get even worse though, with Welsh Independence and Irish Reunification also becoming serious future options. Englanders too should be happy to see her go as what she leaves behind is the biggest clusterfucked mess in living memory and then some generations, after three years spent setting new world records in dishonesty and incompetence. Though I have a hunch her successor has all it takes to break these records in nothing flat. Right now Treeza has already started retconning her Premiership but I doubt this will help. She will forever be remembered as the dismalest failure of a Prime Minister in recorded history. Quite an achievement when you consider she beats even Chamberlain and Cameron here.
General Election polling remains quite chaotic and contradictory but there are now some clues that it is slowly heading in a different direction, like a small realignment within the realignment. Ten polls have been fielded in July by five pollsters, with five predicting the Conservatives leading and five predicting Labour. None predicted the Brexit Party in the lead, a symbolic but significant change frome the June polls. This does not mean that we have already seen 'Brexit Peak' or that Tories and Labour are bound to regain all the lost ground from the last three months. Polls routinely show about 10% of respondents would definitely not vote and another 20% strongly lean towards abstention. Add 20% of undecideds to the mix and it is quite obvious that we are still in a period of uncertainty. Anyway the recent trends definitely show both Tories and Labour slightly up while LibDems and Brexit are both slightly down.
Only a small number of potential Brexit and LibDem voters have switched back to the two 'main' parties so far and it is too early to tell whether it's just a random variation or the start of a stronger trend. You might remember that I said recently that Labour could win the snap GE serendipitously, and not just because I like the sound of the word. Coincidentally Ipsos-MORI's Ben Page wrote an article where he concludes that a Labour victory would be nothing short of miraculous, which is basically saying the same thing with a different wording. I agree with Ben's demonstration here as I made basically the same points in previous posts, including the lack of credibility of Labour's Born Again Remainism. Tory voting intentions also don't show much of a Johnson Effect so far but this is probably bound to change in the light of events over the next few days, such as who is part of Johnson's Cabinet and who is not, and how the various Tory factions react to his first hours as Prime Minister. Next week's polls should enlighten us on all this.
Now is also as good a time as any to take a closer look at what psephologists and prognosticators call the pollsters' 'house effect'. This is the result of some sort of statistical voodoo and depends highly on how pollsters weigh their raw data before producing their headline results. Pollsters don't publish their detailed tweaking algorithms but only the basic principles of their weighting. Usual main ingredients are demographics, which can be off as pollsters don't use the exact same categories, recalled votes at past elections, which pollsters themselves admit can be wrong, likelihood of voting at the next election, which is volatile and possibly misleading as respondents might lie about it. So the mix of secret recipes and contradictory results naturally produces a healthy distrust in polls, especially when they get several elections wrong in quick succession. Or suspicion of bias, the most frequent case being YouGov, the most prominent and prolific pollster, being accused of systemic pro-Tory bias. Which is not necessarily supported by actual data.
Above are the weighted voting intentions from the ten polls fielded in July, sorted by pollster and with the numbers in brackets indicating the number of polls fielded by each. All four of YouGov's July polls have Conservatives in the lead, which might be read as pro-Tory bias as the other pollsters have Labour in the lead. But the Conservative voting intentions are on average the same in YouGov's recent polls as in other pollsters' so I venture the perceived pro-Tory bias is in fact the result of how their tweaks underestimate the Labour vote. I won't try to explain how and why YouGov find Labour 8% below the others' average while LibDems are 4% above, Greens 3% above and the Brexit Party 1% above. It obviously has to do with different weightings and tweaks on the raw data, but I am not privy to the recipes. So all we can do is taking the published headline results as they are, while fully aware that numbers don't always add up and might even not represent the exact state of public opinion. So be it.
My current Poll'O'Polls includes the six most recent ones for which full data and crosstabs have been made available online, fielded between 9 and 24 July, so you can't be more up-to-date than this. Super-sample size is 9,976 with a theoretical 0.95% margin of error, which is actually not really relevant when polls remain contradictory. The current weighted average of voting intentions should give nobody any reason to celebrate. The winning party, whoever they are, would represent less than 25% of voters, which would probably translate into just about 20% of the whole electorate. An all time low and another sign, if we ever needed one, of a deep crisis. The Westminster System is indeed broken and don't expect any of the English parties to offer a solution as they are all part of the problem.
The weighted average fits with the overall trends with both the Brexit Party and the LibDems now back to below 20%. It shows that there was indeed a ripple effect from the European election in earlier polls and that it is now starting to wear out. Which does not mean that LibDems and Brexit have lost all influence on the result of the forthcoming snap GE. If, as seems likely, both finally reach a plateau on about 12%, the LibDems could still gain a dozenish seats and the Brexit Party could still make the Conservatives lose some and the election. Part of the reasons behind the recent shifts in voting intentions are shown by the comparison between the current voting intentions and the 2017 votes.
The Conservatives are in somewhat better shape than previously as they would lose 'only' 30% of their 2017 voters to the Brexit Party instead of the 50% they lost at the European election and in earlier polls. Labour apparently fare better but still lose 25-30% of their voters to the more reliably Europhile parties, and the probability of gaining them all back is slim indeed even after a policy U-turn. Again the SNP have the most faithful electorate. And these numbers tell only part of the story as the Labour and LibDems stats come from GB-wide polls and don't properly show the amount of their 2017 Scottish voters who have since switched to the SNP. Scotland-only polls say it might be as much as 10-15% of LibDem voters and 15-20% of Labour voters. The next round of Scottish polling, probably in the next few days, will tell us more about this. GB-wide we have another vision of the shifts in votes when looking at the voting intentions by region.
The Brexit Party have lost first party status everywhere and are now close to their national average in all English regions and Wales. It is not a good sign for them as an evenly distributed vote is a recipe for success only when you're the dominant party and with a strong lead, think SNP here, not when you're the challenger. The LibDems learnt that already the hard way at previous elections. The regional distribution also shows that the old Tory and Labour strongholds have not yet been restored. Only caveat here is of course the way FPTP works. With many seats coming out as three-way or even four-way marginals, a seat can technically be held or gained on 25-30% of the popular vote. Which of course could provide a neat restarting point for yet another debate on electoral reform. But that's another story. For now.
Election Night 2019 will indeed be quite suspenseful if the actual vote reflects current voting intentions. A hung Parliament again with Labour as first party but already seeking reliable coalition partners as they would be 63 seats shy of a majority. Or 60 shy with the almost certain addition of SDLP to the early mix. Then finding the way to a strong and stable full coalition, be it Rainbow Min or Rainbow Max, might prove to be quite a task. Basic math says the obvious first choice here is Lab-SDLP-Lib as it would bag 331 seats, an 18-seat majority. But in this ever changing political landscape, of course there is again more here than meets the eye at first glance.
So now we have Jo Swinson ruling out any deal with a Corbyn-led Labour. How unexpected…. naw, just kidding, she had already said it. This is darkly reminiscent of Nick Clegg in 2010 stating with a straight face he would consider a Rainbow Coalition if Gordon Brown stood down, when he had in fact decided from the start to go into coalition with the Conservatives no matter what. Is being a Clegg-clone the best Swinson can do? Then she might have to face a reality check some time soon, as this very issue might cause a rift between Orange Bookers and Beveridgers about the longer-term implications of not supporting Labour after voters handed them first-party status. Then Labour members themselves may have offered Corbyn a way out of the dilemma, or at least some food for thought if the last YouGov Topical poll is to be believed. It has a wide range of questions but let's just see what Labour members have to say on the juiciest three: coalition with the LibDems, coalition with the SNP, Scottish Independence. And how the views of Scottish Labour members and Momentum members, the Vanguard of Corbynism, differ from the average.
The most stunning result here is obviously that Scottish Independence is no longer the absolute taboo it used to be, with even a third of Scottish Labbers and a majority of Momentum members supporting is when you count undecideds out. A sign that Jeremy Corbyn should think it through and listen more to his closest supporters than to Richard Leonard. The overwhelming majority for an SNP alliance also denotes that Labour members are fully aware they have more in common with the SNP than with the Libdems. Even a majority of Scottish Labbers support it though they are the exception in preferring a LibDem alliance, probably based on fond memories of just such an alliance in Holyrood. What I see here is that Labour members would rather see a Corbyn government firmly anchored left of center by the SNP than being dragged into some incarnation of English Macronism by the LibDems. It is also significant that a majority of Momentum members oppose the LibDem alliance and that they prefer the SNP alliance by a much wider margin than average. Time will tell which way the party leadership and especially the PLP choose to go, at a time when election predictions show them to be much more resilient than the Conservatives to the LibDem and Brexit surges.
The breakdown of projected seats by meta-region outside Scotland is quite comforting for Labour as the would gain more seats from the Conservatives than they lose to the Brexit Party everywhere, except in London where they would lose more seats to the LibDems than the Conservatives and even lose a few to the Conservatives. Labour here also benefit from an unusual oddity of FPTP. The usual pattern is that a tie in the popular vote, as we have now, favours the Conservatives. But right now it works the other way round as the Brexit Party snatches far fewer Labour voters than Conservatives. So the split right-wing vote, also including Tory Remainers switching to the LibDems, delivers a massive bonus to Labour even in Southern Little England. What remains to be seen is if the Johnson Effect in the next batch of polls will change that, and my hunch is that it will, at least temporarily until a larger number of potential LibDem voters switch back to Labour to defeat Johnson. Again time will tell.
On current polling 174 seats would change hands, close to what we had in 1997. Of course the summary and cartography of gains and losses illustrate how different the 2019 realignment would be from the 1997 realignment, if 2019 can even be called a realignment at all. As Johnson was quick to appoint his Cabinet, we already have ten fatalities out of thirty-two Cabinet members officially announced tonight: Alister Jack to the SNP; Dominic Raab and Geoffrey Cox to the LibDems; Brandon Lewis, Mark Spencer and Kwasi Kwarteng to the Brexit Party; Amber Rudd, Nicky Morgan, Alok Sharma and Robert Buckland to Labour. And the body count will grow higher when all 120ish Government Payroll positions are filled. Despite doing better than ten days ago, Labour would still lose six members of the Shadow Cabinet, including the two from Scotland.
All hope is not lost though for the predicted losers as 139 seats qualify as marginals, including half of those predicted to change hands. The updated cartography of marginals shows that far fewer now involve the Brexit Party, so for the alternate scenarios I have now returned to the more traditional Lab Max vs Con Max rather than the Brexit Min/Max in my recent projections.
Whichever way the marginals go, the Conservatives still lose the election. Only the sort of coalition Labour would have to build changes. Much would then depend on the LibDems' final decision. Is their stated ambition to oust Johnson stronger than their dislike for Corbyn? Now the funniest outcome would be the Lab Max scenario where they could tell the LibDems to get fucked and choose the SNP alliance that 83% of their members support. Guess that Corbyn would then find it a good idea to send Keir Starmer to Bute House to negotiate the terms, rather than going himself.
And now the die is cast. Boris Johnson is the new and probably last Prime Minister of the United Kingdom and some time soon the first First Minister of England. I would like to congratulate Jeremy Hunt on his outstanding success in bagging more than the 20% of the vote some polls predicted, and a majority among those who actually voted Conservative at the last European election. Now the UK has a new and major problem: the former PM was a walking dead and the new one is a running joke. Johnson's Premiership did not start in the best way as being accused of talking rubbish by the EU even before his appointment was official can sort of ruin your day. Then I guess Johnson does not really care as he has been accused of far worse all along his career, and each en every time it was true. But the Great Culling of the Huntistas certainly made his evening fucking enjoyable. Only the Scottish Tories now have a thousand reasons and some to be anxious, as sacking David Mundell was a deliberate humiliation aimed at Ruth Davidson, and what lies ahead for them Beyond Brexit is certain electoral annihilation. Bully for them.
People obviously have low expectations for a Johnson Cabinet and it might have a low life expectancy too if rebel-wannabes Tory MPs prove themselves to be as bitey as they are barky and make good on their threats to do whatever it takes to prevent a no-deal Brexit. Labour are already working on it with Keir Starmer initiating contacts with high-ranking Tories. Interestingly we don't have Tom Watson or even John McDonnell tasked with this, another sign in my opinion that Starmer is destined to play a bigger role in the near future than he is usually credited with. Don't rule out Deputy PM Starmer just yet, or even PM Starmer, as in one of my pet scenarios. The probability of a snap GE in the short term has also dramatically increased as many have already accepted the idea of the Johnson Government going down before the Conference Recess. Mark my words: the snap GE will be held on Brexit Day, 31 October, and it will be both tricky and a treat. Tricky as finding a new viable majority will be quite a task, and a treat at the sight of both Johnson going down and Corbyn struggling to find out how to approach the SNP without actually approaching them.
The next few weeks promise to be exciting so look out for all new upsets and gaffes, and stay tuned for further broadcasts.
The white bird of hope can be as elusive as the butterfly's ghost (Voyage to the bottom of the sea, episode And five of us are left, 1965)
Recent General Election polling has continued to follow an erratic and disconcerting pattern that makes an actual prediction quite challenging, even if I always try hard to give you a plausible one. One of the most intriguing recent polls is Comres' last, fielded between 5 and 7 July. Like YouGov who pioneered the genre (more on them later), Comres indulged in 'alternate reality' polling. This time with the classic baseline voting intentions, compared with what would happen under PM Johnson or PM Hunt. And this is where it beggars belief because, believe it or not, both contenders would deliver a better Conservative result than the baseline. Which is quite counter-intuitive unless you assume respondents subconsciously took the baseline question as 'how would you vote if Theresa May remained PM?', as that easily explains a weak Tory showing. All kidding aside, here are the voting intentions Comres found on the baseline question and the two alternates:
Now on to the seat projections under these voting intentions. Comres' Scottish subsample is distinctly less favourable to the SNP than other pollsters' or the current trend of Scotland-only polls which put the SNP on 50+ seats. But this is in fact irrelevant as the purpose of the experiment is to determine how Johnson and Hunt would perform relative to each other, not how well either would do against the SNP. Now the obvious lesson here is that both would lose the election, even with Johnson bringing a significant number of prospective Brexit Party voters back home. But the oppositions would also find themselves in awkward situations. Defeating PM Johnson would force Labour into a Big Tent coalition including both the LibDems and the SNP and even that would deliver only a slim six-seat majority. Defeating PM Hunt would put Labour in a slightly more comfortable situation as they could probably gamble a Small Tent minority coalition with the LibDems, SDLP, the Greens and Plaid Cymru. This would bag 320 seats overall, three shy of a majority, but might be worth trying as the SNP would be more likely to abstain or offer case-by-case support rather than defeating the Government. But of course this is just one set of possible alternate futures, and other polls predict strikingly different outcomes.
For the record, on the same vote shares Electoral Calculus predicts 345 Tory seats (obviously the source for the Daily Telegraph headlining 'Johnson would deliver 40-seat majority and wipe out the Brexit Party') and Election Polling predicts 316. The mysterious quirks of prediction models at work again. But I stick to my results as I don't see how Tories could bag more seats or even just as many as in 2017 while being 10% down on the popular vote and faced with both a strongish Brexit Party and LibDems doubling their vote share and targeting key marginals. I also think we should not underestimate the potential fallout of the UK crashing out of the EU only to become the de facto 51st State of the USA. With PM Johnson doing nothing to actually 'take back control' but acting as the subservient lapdog of the Trump administration who are hellbent on allowing their wealthy donors to make huge financial gains from any future USA-UK trade deal, including selling out the NHS to American Big Pharma and HMOs.
I don't know how recent events would affect the results of another such multiverse poll fielded now, and we will probably never know as nobody will find it worthwhile to have one. The ITV Clown Show only showed that both contenders were just eager to prove the other one is unfit so serve as PM, and sadly demonstrated that both are right. Then Darrochgate shed a new disturbing light on what to expect from a Johnson Premiership: a whole country driven into self-destruct by a bombastic egomaniac bully taking orders from the psychopathic warmongering leader of another country. And I don't mean North Korea here. Then it's too late as the ballots for the Conservative leadership vote had already been sent out before all this, and probably already sent back with the Johnson box ticked, and anyway Tory Little Englanders want just what Johnson has to offer: crashing out of the EU into a WTO-only world no matter what, with Mauritania the only trading partner left, while entertaining delusions of grandeur and past glory. Good luck with that when they find out Nigel Farage is the next Ambassador to the USA. Oh wait….. they would actually love it, they voted for him at the European election, didn't they?
YouGov's newly found appetency for exploring alternate timelines has led them to indulge in some on their own initiative (more on this later). But some clients also thought it was good value for good money to poll some other multiverses on top of that. So last week Britain Elects commissioned YouGov for one such endeavour that was unfortunately not much publicized. Unlike other recent multiverse polling there is no calibration baseline question in this one as YouGov took respondents directly into a near future where Boris Johnson is Tory Leader and still Prime Minister, Jeremy Corbyn is Labour leader, Jo Swinson LibDem leader and Nigel Farage still Brexit Party leader and not yet Ambassador to the Mothership in Washington. So far so good as this is most likely what we will have two weeks from now. Then there are two options. First how would you vote if the election happened before Brexit is delivered? Which would mean an early October snap GE, something I think we can already rule out. Second how would you vote if the election happened after Brexit had been delivered? Pointing to a possible early November snap GE, which I find much more likely. And here are the voting intentions after undecideds and non-voters removed:
The most salient feature of this poll is that it is totally at odds with what regular polls over the same timeframe predict. Even YouGov's own polls find Tories no higher than 2% ahead of Labour and they notoriously have a Tory-leaning house effect. So Tories possibly leading by 10% is definitely an outlier. I also seriously doubt that Jo Swinson's name recognition would propel the LibDems above 20%. If anything ,other recent polls show that she has little if any name recognition outside East Dunbartonshire. And what little she has is more likely to have a negative effect than a positive one. This kind of multiverse polling brings to mind the obvious question: why should explicitely naming the party leaders change the voting intentions when all of them are already widely known? Admittedly this makes sense when a leadership contest is happening, as in pitting Johnson and Hunt against each other, but then pollsters should also test the LibDem vote with Ed Davey as leader. When no leadership contest is in sight it makes less sense unless the pollster also links a hypothetical change of leader with a just as hypothetical change in party policy (see one such below). Then I guess the appetite for alternative reality polling will fade away quite soon when the Conservative and LibDem leadership contests come to an end, and possible changes in the parties' manifestos become known. Now what sort of seat distribution would the 'alternate Boris' options possibly deliver?
With all the usual caveats and more, this projection nevertheless show the limits of the Johnson Effect. In both options offered here Tories would lose the election anyway, which is obviously what Britain Elects wanted to demonstrate. Then it is quite surprising that Brexit Unbound alone would be enough to switch 100+ seats back to the Tories in elections supposedly held only weeks apart. There is also some ambiguity about what 'delivering Brexit' actually means in the respondents' mind. Putting pressure on Johnson with a Brexit Party vote when he has repeatedly pledged to risk even a no-deal Brexit makes little sense. In fact it would be more sensible for hardline Leavers to support Johnson to make sure he has the votes to make good on his promises. The projection of the post-Brexit votes is just as flabbergasting. If the intention is actually to reward Johnson for delivering, then it fails utterly with Tories some 30 seats away from a majority, and finding an alternative government coalition quite a difficult endeavour. But enough now with the alternative realities and let's go back to what regular polls have to say, which is in fact just as puzzling.
Recent GE polls are just as inconsistent and disconcerting as they have ever been since the European election. Fun part though, after a recent controversy among Wikipedia editors, is that Survation no longer offers Change UK and UKIP as prompts, but bundles them with 'Others' who bag a massive 1.2% in this week's poll. YouGov still includes them in their prompts and both bag exactly 0% of voting intentions in this week's poll. Quite the irony here (read why in James Kelly's article linked above). Polls over the last month are still all over the place and the most recent batch has done nothing to make the situation less confusing. The spread of voting intentions for all parties is way beyond the margin of error so basically everybody can find at least one poll that makes them happy and one that makes them decidedly unhappy.
As usual the best we can do is rely on the weighted average of the most recent polls. Only caveat is than an average of such contradictory results is likely to not define an actual snapshot of current public opinion, but just to reflect regression to the mean. In other words vastly different results will just cancel each other out and will not help determining any real trend. Anyway that's all we have and my current Poll'O'Polls includes the six most recent ones for which full data tables and crosstabs are available, fielded between 2 and 11 July. Super-sample size is 9,832 with a theoretical 0.96% margin of error, and here's what it says:
Tories and Labour tied, while the Brexit Party and LibDems still do rather well, opens the door to a lot of possibilities that will depend mostly on how a small number of votes will determine the results in marginal seats. Major lesson is that the common wisdom that the Tory vote would be boosted by PM Johnson just does not show in polls right now. Potential LibDem voters might be willing to switch to Labour to keep any Tory PM-wannabe out of Number Ten. Tactical voting from potential Brexit Party voters is also a strong possibility to make sure the next Tory PM will feel the necessary support or pressure to get Brexit done. One of the key factors will be the way the Brexit Party voting intentions evolve in comparison to their European election result, and how the evolution of UKIP voting intentions in a similar context five years ago can be taken as a prediction of future Brexit Party votes, or not.
Voting intentions for the Brexit Party have indeed gone down recently, though polls contradict each other here. Main point is that they go down at a slower pace than UKIP's voting intentions five years ago. Over the same time period in 2014 UKIP were down 15% on their European election vote and already credited with the 12% they would get at the 2015 GE. Brexit Party now are down only 10% on this year's European election. On about 20% of current voting intentions, they will not gain as many seats as earlier polling predicted but will still make their presence felt, for the most part by denying Conservatives a number of key marginal seats. The irony here being of course that the Brexit Party will switch a number of these marginals to Labour and the LibDems and weaken the Conservatives, no matter who the PM will be next month and how he will deal with Brexit.
Current voting intentions would again deliver a hung Parliament, one that would elicit more sighs of relief than smiles of joy from Labour ranks as they would come out as the first party only serendipitously again, and with a rather unsatisfactory result on a handful fewer seats than in 2017. Of course it wouldn't match the Conservatives' utter humiliation with their worst result since the days of Michael Howard. Again cases of champagne would be delivered only to Great George Street, SW1 and Gordon Lamb House, EH8. That is before phones start ringing at both with urgent calls from Victoria Street, SW1E as the real fun would start only at dawn after a long suspenseful Election Night. And then Corbyn's path to Number Ten would indeed be a long and winding road 🔊. Oops…. sorry…. this one's McCartney's not Lennon's…. won't happen again…. Scout's honour.
Here a Lab-SDLP-Lib coalition would bag 322 seats, a perfect tie with Sinn Féin still not taking their seats and technically one seat short of a majority. Rainbow Min including the Greens and Plaid Cymru too would bag 329 seats or a narrow 14-seat majority. It would surely be viable without SNP support, but some sort of deal with a Section 30 Order squarely on the table would be safer. That this is the Golden Road To Number Ten has apparently just dawned on Kezia Dugdale though that option has been in the cards for quite a time now. Then you just have to wonder what sort of influence Kezia or her think-tank have on Jeremy Corbyn, and here 'sweet fuck all' springs to mind. Then we may also have to consider the possibility of LibDems propping up Tories again, as Jo Swinson was deliberately ambiguous on the matter, and there was nothing constructive about it. Of course I don't see even Swinson helping Brexit-crazy Tories to stay in power, but strangest things have happened. Then the extra layer of irony in LibDem ambiguity would be that it would literally force Labour to strike a deal with the SNP and make a Second Scottish Independence Referendum inevitable, no matter how loud LibDems would oppose it.
The breakdown of projected seats in the meta-regions outside Scotland again tells a lot about the why and how of Labour's victory. Labour proves surprisingly resilient in the North and Midlands and would even make gains in the South. Then perhaps the Corbynistas were right and party loyalty is stronger than any lingering doubts about the exact party line on Brexit, and Leave voters once tempted by the Brexit Party would come home as the prospect of defeating Boris Johnson becomes more credible. The only black spot would be Corbyn's own backyard in London where a sizeable part of Remain-leaning ABC1s would stay unconvinced and switch to the LibDems. A similar pattern would work against the Conservatives in the South where right-wing Europhiles would make the LibDems the strongest threat to Tory dominance there, and no longer the Brexit Party as we had in earlier projections closer to the European election.
A technical aside seems appropriate too as you might wonder why my model often delivers different projections from my competitors, most noticeably Electoral Calculus who seems to be the one most commonly used by media and prognosticators alike. The basic reason is how projection models handle Uniform National Swing (UNS) and Proportional National Swing (PNS), the two concurrent methods used for election projections. My own model is tuned on 70% UNS + 30% UNS and here is a comparison between what it predicts on current polling and what we would have if tuned on pure PNS and pure UNS, but otherwise using the exact same projection algorithms.
UNS is the method commonly used by British prognosticators, sometimes with a modicum of sophisticated tweaks as those used by Marin Baxter in the Electoral Calculus engine. Interestingly Martin freely admits the tweaks result in a 15-seat 'bonus' for the Conservatives compared with pure UNS. But even pure UNS massively dampens the effects of the LibDems and Brexit surges and boosts the Conservatives' seat projection more than Labour's, and this how some scenarios predict a Tory majority even on a lower share of the popular vote when processed on UNS. It is worth remembering though that all UNS-based models failed to predict a hung Parliament in 2017 and sometimes by a stunningly wide margin. Conversely PNS was a better predictor of the Canadian Federal Election in 2015 which saw a massive swing from both right and left to the Liberal Party. And in due course the Snap GE of 2019 will tell us who was right or almost and who was wrong but then not as much as you'd thought.
On current polling 187 seats would change hands. Far from the all-time highs we had on other recent projections but on a par with 1997, the last major realignment election, when 185 seats changed hands. The breakdown of gains and losses shows a widely different pattern though with seats switching in all directions instead of an en masse migration from Conservatives to Labour. Labour returning the same share of seats as in 2017 would be good news for Jeremy Corbyn as this would remove the threat of a leadership contest in the near future, especially with the prospect of him becoming the next PM, even if he had to go through complicated negotiations with other parties from the former oppositions before achieving this.
The Conservative frontbench would again suffer badly despite getting better results than in previous projections. Penny Mordaunt and Matt Hancock would lose their seats to the Brexit Party. Solicitor General Lucy Frazer and Attorney General Geoffrey Cox would both be unseated by the LibDems. Eleven Ministers of State, six junior Ministers and four Parliamentary Private Secretaries would also lose their hard-earned jobs. Not as much of an outright massacre as in previous projections but still a heavy toll. Labour would fare better, losing only six members of the Shadow Cabinet, with our BFFs Sweeney and Laird as usual among them.
The most rejoicing Portillo Moment would obviously be Amber Rudd losing Hastings and Rye to Labour. All this energy spent on reneging on her strongest beliefs and sucking up to Boris Johnson for a Cabinet post would have been in vain. Another upset would be John Bercow losing Buckingham to the Brexit Party. Downside here would be that his successor would probably be Nigel Farage who already tried his luck here once before. Just wonder who Labour would then choose as Speaker as it is now their turn after ten years of Bercow in the Chair. Tom Watson anyone? Just to avoid having him in Cabinet embarrassing the PM.
This time 149 seats would qualify as marginals, again more than twice as many as in 2017. 109 would involve the Brexit Party either as the projected winner or the runner-up. Again, and possibly for the last time if the anti-Brexit trend in polls continues, what could happen in these seats would be the major factor in trying to determine alternate outcomes to the election.
The possible alternative scenarios would be better than before for Labour. The Brexit Party underperforming would bring only 15 more seats to Labour but that would be enough to take the Lab-Lib coalition to 334 seats and the Rainbow Min coalition to 341, thus avoiding the awkward necessity of negotiations on confidence and supply with the SNP.
At the other end of the spectrum, the Brexit Party could gain another 36 seats from Labour. They would still come out as the first party but this would jeopardize all chances of a Lab-Lib coalition or even Rainbow Min to pass a Queen's Speech. Once again the SNP would find themselves the kingmakers as their votes would take the Rainbow Max alliance to a comfortable 56-seat majority on 350 seats. With the added bonus of being in a position to extract almost any concession from Labour. Devolution of everything save foreign policy and defence would be a good start. Or wouldn't it?
YouGov, once known only as the UK's most prolific pollster who would poll anything from the proper pronunciation of 'lieutenant' to the appropriateness of wearing kimonos (real examples) if you paid the fee, have started a new career as political fiction testers. Their latest self-inflicted foray into this field deals with a near future where a snap GE takes place with Boris Johnson as PM and Jo Swinson as leader of the Liberal Democrats, with Nigel Farage still leading the Brexit Party. But the Labour leader would possibly be Jeremy Corbyn if the ongoing investigation into the Civil Service proves he is well enough to stay as leader, or might be Keir Starmer. A scenario quite to my liking as you might remember I have mentioned Starmer as a possible PM several times already over the past months, so I guess the screenwriters at YouGov read my blog. Then the heart of the matter is what Brexit policy each of the two Labour leaders would pursue, however out-of-character these may look. Here are the four options YouGov laid out on the table before flabbergasted respondents, in their own wording:
YouGov sold this one as a separate endeavour on their main topics page, but it is in fact a set of variants of the regular generic poll they fielded 24 and 25 June, which conveniently provides us with a baseline with which to compare the alternate futures. You will notice the alternate futures are not quite consistent with the baseline, which is probably the price to pay when you ask oiks to think twice or thrice about a once-in-seven-generations issue that promises to deliver the most momentous turning point in the country's history since the French Invasion, just seconds before you ask them if they have ever met Erskine May and how they pronounce 'shkedule' (real ones too). Actually the whole line of inquiry is probably irrelevant now, after Labour's convoluted reboot of their Brexit policy and Corbyn's apparent conversion to Remainism, but it's still worth having a peek at what public opinion thought. So let's start with the voting intentions in the baseline calibration poll and the four alternative futures:
First obvious result is that 'Corbyn on Alternative Brexit' and 'Starmer on Alternative Brexit' deliver the same voting intentions. You might infer from this that what matters is not how sexy the Labour leader is, as it doesn't seem to make a difference, but how much the core policy sucks, which in this case in obviously 'a fucking lot'. We have a significantly different picture in the two 'New Referendum' options, with Labour doing better whoever the Suryong is, but Starmer delivering the best better of the two. So you can argue that, with public opinion considering a widely more popular policy choice, then the leader's personality makes a difference. You might also conclude it is not so much a personality issue per se but rather a credibility issue, as people would have a hard time trusting Corbyn actually believing in a People's Vote U-turn, while Starmer has been definitely more consistent in his opposition to Brexit and his search for feasible alternatives. Now we know voting intentions tell only a part of the story as FPTP can play strange tricks, so here are my seat projections on the baseline and the four options of the multiverse:
First obvious point here is that Tories would lose the election whatever the scenario, which does not mean Labour would actually win it. In fact all results point to an inextricable mess, with the actual level of inextricability depending on the LibDems' shares of votes and seats. With the LibDems now also entertaining some sort of constructive ambiguity about what they would do in a hung Parliament, we have a new level of uncertainty. In my opinion, LibDems would have no incentive to support a Labour minority if Labour stuck to the unicornish 'alternative Brexit' policy. But they certainly would never prop up a hard-Brexiteer minority Tory government. Only outcome would then be for the LibDems to switch from kingmakers to king-unmakers and force a repeat snap GE, which would then most likely be fought with Starmer as Labour leader on a New Referendum manifesto. In that case Labour would get their best results and LibDems their worst of all multiverses. Then PM Starmer could consider a coalition with Labour as the strong senior partner, that would bag 340 seats with SDLP in too, a 36-seat majority, and avoid any embarrassing need for a deal with the SNP.
Of course chances of the Starmer-Referendum scenario ever happening are slimmish at best. And even if it did happen, Starmer would need time to restore some semblance of trust in Labour and assert his personal credibility as standard-bearer of a massive U-turn back to sanity. Some within Labour advocating support for another option that already failed thrice, what you might dub Brexit The Third Way, certainly wouldn't help. But I now guess that Corbyn's born-again Europhilia has made all alternative proposals irrelevant, and indeed this poll itself. Then on the kind of scenarios it explored, it it always interesting to see what would happen in London and Scotland, the two most Europhile areas in the UK. So London first as befits the Capital of the English Empire, voting intentions and seat projections:
Voting intentions again show how sticking to an alternative Brexit would have hurt Labour in London, massively switching centrist ABC1s to the LibDems, while taking a stand for a second EU referendum brings back some of them and limits the self-inflicted damage. Oddly a LibDem tsunami would switch 'only' a third of Labour's London seats, as many as their MPs there are as deeply entrenched as the ones in the Northern Powerhouse sinkholes. Yet as many as seventeen Labour MPs could go down in this scenario, including a couple of famous names (Jon Cruddas, Margaret Hodge, Teresa Pearce, Tulip Siddiq, Wes Streeting, Kate Hoey and even Emily Thornberry). The impact of YouGov's scenarios is less spectacular in Scotland, as the SNP's dominance has produced some kind of immunity to the somersaults of English politics. LibDems could hope to gain North East Fife from the SNP, which I have already said I find highly unlikely even if the math says so. But in the extreme scenario they would also gain Edinburgh South from Labour, which is not as odd as it may seem in an alternate world where LibDems would quadruple their Scottish vote and Labour lose three quarters of theirs. The irony here is that Labour's best result under PM Starmer, snatching 5% from LibDems and 5% from the SNP, would get them exactly zero extra seat but just switch one from the SNP to the Conservatives. Vote Red, Get Blue, as they say….
YouGov's intention here was openly to prove that Labour need both a change of leadership and a change of policy. The latter is obviously true and has already been delivered, even in a quite obscure way, but clearly the jury's still out on the former. And it's no coincidence that we now see the Guardian turning against Corbyn when they endorsed him just two years ago. So I guess we may soon see the rebirth of some media-inflated push for the emergence of some incarnation of English Macronism, which would have to be the bastard spawn of Orange Bookers and post-Blairites, and YouGov's findings could be twisted and spun to support that. But they would need to find the proper leader now that Chuka Umunna has wrecked all his chances at ever morphing into Chuka Emmanuel. And I doubt even the Guardian can fancy Jo Swinson filling the position, as the keyword of Macronism is 'neither left nor right' or even 'and left and right' some days, and Swinson is more like 'neither left nor left' which would quite defeat the very purpose of the whole stunt. So just wait and see who and what the Guardian will propel next. For the record I do think Corbyn is not the best man for the job right now and Starmer would probably be a wiser choice, but I would also hate to see Corbyn going down as the result of some media-engineered self-fulfilling prophecy.
You know Unionists have something big to hide when the Cabinet Office turns down a Freedom Of Information request issued by the SNP at the same time the Prime Minister commissions a review of devolution, managing to infuriate both the Yes camp and the two Tory PM-wannabes. Your guess is as good as mine about what Lidington's secret poll says, but it seems reasonable to think it shows even bigger support for Independence and opposition to Brexit than recently released polls did. We just don't know it yet but Lidington does and he does not want the truth out lest English Civilization As We Know It be fucked and doomed thrice over. On the other hand, May's review of devolution looks like a thinly veiled attempt to justify and amplify the power grab that has already been going on for two years or so. For example by allowing direct investment from the English Government into devolved domains, so they can brag 'Westminster better than Holyrood'. But the stunt will fool only those who want to be fooled or have a vested interest in fooling others. Then you know…. fool me once, fool me twice, more fool me…. But hopefully people will be able to see through the flimsy cover story once they get all the facts.
English Tories on the campaign trail also found it wise to share some thoughts about Scotland, and even do it from Scottish territory. Fun moment was when Jeremy Hunt thought it would be a good idea to tell Scots what he would 'allow' us to do if he became PM. I venture that Hunt is actually thick enough to not have realized his statement might have made some converts to an Unilateral Declaration of Independence, and at the very least switched some votes to the SNP and Yes, like in 'git tae fuck ye glaikit'. And the backlash came quickly, from Angus Robertson and Ian Blackford as you might expect, and also from the 13k AUOB marchers in Ayr who again demonstrated how strong support for Independence is now. And Unionists of all shades, not just Jeremy Hunt, should sit back and let the truth dawn on them: the days of The Cringe are over. Nobody any more thinks that Scotland is 'Too Daft, Too Wee, Too Poor' to become a successful independent country. Quite the opposite actually and nobody can tell Scotland what we can and can't do, no matter if you tell it from Pacific Quay or from Whitehall.
Boris Johnson's frontal attack on the Barnett Formula should be taken more seriously than Jeremy Hunt's ramblings, at least until BoJo U-turns on that too and claims to have been misunderstood. Of course describing Scots as 'subsidy junkies' always goes down well in Little England, but the truth is a wee smitch more complex and ambivalent than this. First the money 'sent back' to Scotland does not grow on Surrey Money Trees, it's the product of Scottish taxpayers' money being sent directly to the Treasury and then injected into the Barnett mechanism. So Scottish taxpayers' hard-earned money actually funds reserved domains such as an overbloated Foreign Service (over 14k personnel by their own official publication, compared to 13k for the USA including CIA operatives, 11k for Germany and 10k for France) that fails to provide support to the First Minister of Scotland on her official trips abroad, interests on a mammoth debt that Scotland bears no responsibility for, or nuclear submarines that Scotland would gladly send to anyplace in England. Too bad we can't ask for refunds.
Now if the English Government really thinks Barnett is flawed, and there are indeed many reasons to think it is, there is an easy and quick way out of this, short of Scottish Independence: Full Fiscal Autonomy. Then the flow of money is reversed and Scotland would indeed be visibly and unquestionably funding England's vanity projects. The Scottish Government would have every reason to go for selective funding: send back money to the English Treasury based on the actual added value for Scotland and not the infamous 8.3% population-share flat rate. Of course we could also charge a reasonable fee for collecting taxes on behalf of the English Treasury, or couldn't we? Then in our 'here and now' I reckon that James Kelly (naw, not that one) might well be right: Yes has already won, it's just that nobody got the memo. Now that Nigel Farage has come out in support of Scottish Independence, anything can happen.
Right now the future of the UK looks pretty much like 'Final Destination 6: No Deal', as it is clear that insisting on further negotiations with the EU would only take the next PM straight to the corner of No and Where, SW1. Quite obviously now the only way to avoid the Thelma and Louise scenario is to go straight to a no-confidence vote leading to a snap GE, though even this scenario is not without its own brand of quirks. But Tories have found a neatish way to get around that, for the immediate future at least. The results of the Gollums' Beauty Pageant are to be announced on the 23rd but Theresa May will take her last PMQs the next day and formally remain PM until around suppertime on the 24th. Then the next PM will be officially appointed only around lunchtime on the 25th which kills any prospect of an immediate no-confidence vote as MPs leave for their yearly Benidorm Summer Camp Recess precisely that day, like 4PMish to avoid traffic, unless of course John Bercow decides to make himself even more popular than he already is and postpones the Summer Break a week or two, but I wouldn't count on that. So here's the trick they found to delay any vote until after the Brecon and Radnorshire by-election but then it will most likely backfire as disgraced recalled former MP Christopher Davies is bound to lose heavily to the LibDems.
Commons will then reconvene on 3 September unless the next PM stages a constitutional coup by proroguing Parliament during the Benidorm Recess and until after Halloween. Of course the coup can happen only if the next PM is ready to drag Elizabeth Windsor into the mud of partisan politics, which the Civil Service and a majority of the Privy Council would strongly advise against, not to mention the tsunami of outrage this would spark all across the country. And of course the case against prorogation has been made stronger by Commons amending the Northern Ireland Bill in a way specifically designed to make it impossible, pending expert legal advice or a possible court challenge. So the likely outcome is that Commons will reconvene as planned and sit until 12 September, the likely starting date for this year's Conference Recess. And even if the coup succeeded, the legal case has already been made that Brexit in whatever incarnation can't be dealt with on Royal Prerogative, be it Henry VIII or Richard III or whatever, and that a Commons vote would be required anyway.
Failed PM-wannabe Sam Gyimah stated recently that some 30 Tory MPs would be prepared to vote against the government in order to prevent a no-deal Brexit. The first week of September might then be the perfect window of opportunity for the new PM to call their bluff, if it is indeed one. First remember that the current Session of Parliament, the one ending eleven days from now, started two years ago for two years, an exception to the convention that says Sessions last only one year. So if prorogation doesn't happen, as is the most likely scenario, Commons will reconvene on 3 September for the whole shebang: State Opening of Parliament and a Queen's Speech. From here it's basic math. On 635 valid votes on divisions (650 minus Sinn Féin, Speaker, Deputy Speakers and Tellers) the theoretical headcount is 321-314 as long as DUP and Sylvia Hermon prop up the Tories. What happens next depends on the Queen's Speech explicitely including No Deal as the preferred option, or not. And then all bets are off….
I venture that No Deal would see Hermon switching from Aye to No, making the headcount 320-315. Then the Queen's Speech could be voted down by only one vote à la Callaghan with only three Tory defectors switching to No. But you can also expect a number of luminaries like Kate Hoey, Frank Field, John Woodcock or Stephen Lloyd to abstain so some safety margin would be needed. Like 10-12 Tories switching to No or abstaining, which is not such a wild scenario if you factor in Philip Hammond making good on his veiled threat to take down any No Deal PM. And if all else fails we still have the RoryVerse Option of an Alternate Parliament sitting in open defiance of the PM. Which I don't see happening but would be the best popcorn moment ever. I venture the actual outcome would be the Queen's Speech voted down, which would not be a first but just a rare occurrence as it already happened thrice (1886, 1892 and 1923). Back then the opposition just took over as government but in today's context this would lead to a snap GE. There would be some legal hurdles to clear as the Commons' schedule would be conflicting with the provisions of the Fixed-Term Parliaments Act, with the Conference Recess starting soon, but surely nothing that can't be ironed out, and we would get the long-awaited snap GE possibly on 31 October, ironically right on Brexit Day and back to the beginning. How polls are inconsistent and all that….
More strange days ahead so stay tuned for further updates and keep the cooshite detector on
The opportunity to secure ourselves against defeat lies in our own hands But the opportunity of defeating the enemy is provided by the enemy himself
