Morning After Poll

Five snap thoughts on a snap election

There is nothing so wonderful as Wikipedia, whose unknown soldiers have already created a map of the General Election results from last night. The overall theme of the results was unsurprising - a huge Labour landslide, the collapse of the Conservative benches, large votes for Reform and the Greens that didn’t really translate into seats. But there were of course surprises. Labour’s vote share underwhelmed, even as their seat share fit firmly within the error bands of most predictions. A large number of independents rode to Parliament on a wave of discontent about Gaza. The SNP just completely imploded, potentially along with Scottish independence for the next decade. Oh and I got a lot of things right and some things wrong in my prediction - but we’ll get to that. Here are five conclusions.

1. We have a Two-Party Electoral System but Multi-Party Politics

I recognise I am a broken record about the demerits of First Past the Post but it has really excelled itself this time. We are in a situation where Labour, with just over a third of the vote have just under two thirds of the seats in Parliament. Reform, with fourteen percent of the vote have four seats. The Greens with seven percent also have four. Somehow the SNP have gone from being massively over-represented by FPTP to getting half their proportional share.

Because the election gods have a dark sense of humour, where FPTP produced its most proportionate outcome it was the Liberal Democrats receiving twelve percent of seats and of votes. This was due to the Lib Dems finally deciding to join FPTP rather than beat it and combined ruthless geographic targeting with a whole bunch of luck. Last time around the Lib Dems got almost the same vote and single digit seat numbers. Ironically, it was those newfound fans of PR, the Tories, who had the most PR-like outcome, in the sense that they have around 19% of seats on a vote share in the low twenties, spread quite thinly across the country.

In all, this is the most disproportionate election in modern British political history. Heinz Brandenburg, the University of Strathclyde political scientist has done the calculations - on the Gallagher index, the standard measure we academics use to describe disproportionality, this election scores 24, beating the previous record of 21 in 1931, an election in which Labour were reduced to a rump 55 seats by the National Coalition. By my reckoning this means the new Parliament will be the fifth most disproportional in the world, after St Lucia, Barbados, the Bahamas, and Bhutan - classic combo.

Is this really tenable when, as Heinz, also notes, the ‘effective number of parties’ in British elections is also at a peak of almost five? Well, electoral systems are stable as long as the people who benefit from them enter power and then have the right to change - or not change - them. Now cries for some kind of PR won’t come only from the progressive left but from the right, particularly Reform. That the Conservatives did better than anticipated in the polling suggests they might choose to stick, though, rather than twist.

2. Voting is Strategic not Sincere - You Go with the System you Have

One of the key innovations of my voting app (see I lasted five minutes without mentioning it) is allowing tactical voting. When you have an incumbent party that people are desperate to get rid of, then tactical voting makes a good deal of sense - you are more concerned with the bad party you hate then how good the winning party will be. There was an enormous anti-Conservative vote across the UK and if anything is the takeaway from this election, it’s that.

This in turn means that we should stop obsessing about the raw vote totals (except in terms of turnout, which was very disappointing). Yes Jeremy Corbyn did really well in 2017 - I too was conscious and alive then - but he didn’t win a majority. Keir Starmer may end up with fewer votes than 2017 but will have a huge majority. But that’s because unlike in 2017, far more people voted for third parties, in part deliberately to get Conservatives out. I’d warrant that most of the Liberal Democrat voters in the new single-party Lib Dem dictatorship of rural Oxfordshire are totally fine with Keir Starmer as Prime Minister. And many would have voted Labour if Labour had been situated in second place versus the Conservatives. But they weren’t and so they voted Lib Dem. This will also be true for many Green voters. And if the election had been tighter, I suspect some of the voting for left-leaning independents would have shifted back to Labour.

What matters politically in the United Kingdom is your control of Parliament. That. Is. It. You can’t convert your votes into any other power than that. You can moan, you can gnash your teeth but you won’t magically get extra seats from it. Everything about political strategy is determined by the idiosyncrasies of our (mad) electoral system. Keir Starmer’s team figured out how to use it to their best advantage (as did Ed Davey’s). Jeremy Corbyn’s team didn’t. Thems the breaks.

But this should also remind the Conservatives not to settle back into their heavily-used copium couches. Already, Tim Shipman reports that there is a manic excitement among the Conservatives that the lowish vote share for Labour means there is a way back for them in 2029. Obviously that is entirely possible. But the Conservatives need to remember two things - first, that their own vote share was the worst they have ever had. And second, that tactical voting will play out differently if people think the Conservatives might get back in.

3. Don’t Miss the Lede

Speaking of the Conservative Party. There is one absolutely historic aspect to yesterday’s vote. The Conservative Party had their worst election ever, at least adjusting for the size of the Commons. Even without doing so, they are competing in raw numbers with the Tory party of the Whig dominated era of the mid-18th century. Maybe they need to find William Pitt’s great-great-great…. grandchild. Inasmuch as we know much about vote numbers of the past, it also appears to be the lowest share of votes they have ever received in a General Election.

The fact that the pre-election polls had suggested what always struck me as a very unlikely Canada style outcome has led to a weird wave of relief among Conservatives and a strangely snarky tone about Labour on the election-night coverage. Partly this was because the exit poll massively overstated how many seats Reform would win, thereby allowing every broadcaster to engage in their absolutely favourite topic of Nigel Farage and the wonders of the populist right. I hope in the clear light of day, people realise that was probably a mistake - Reform just beat the UKIP total of 2015.

But I think the more likely reason the election coverage has offered a somewhat disappointed tone about Labour is that people likely to have contrarian takes. That Labour would win a big majority and the Tories collapse was known ahead of time. Boring… move on.

But no, don’t move on. This is the big story. Let’s not be sophisticates, rambling on about how ‘actually…’ if you look at something else the real story is more complicated… No. No it isn’t. There has been a seismic shift between the two big parties of British politics. That’s what matters in the election. And perhaps more importantly in the policies that will get made over the next five years. Of course Keir Starmer and Rachel Reeves will have to make sound decisions to keep Labour popular. But I am sure they would much rather than over 400 seats and a slightly disappointing vote total than more votes and no majority.

And the Conservatives? They love power. They live for it. Now they don’t have and won’t for five and quite probably more years. They are in disrepair and lost in the wilds. Yes the press leans conservative and will cover them more than any other party with a similar number of seats. But this was very very bad for them. Don’t get gaslighted.

4. There is No Progressive Coalition nor a Reunited Right

If the story of this election (other than the massive shift in you know, who runs the country), is about a fragmented electorate, it might be tempting to wish away the fragmentation. Lots of people have already been adding the Conservative vote share to the Reform vote share and noting it edges ahead of Labour’s. Sure. Now add the Lib Dem and Green vote shares to Labour. Oh…

But actually, don’t do any of that. Because although there was a great deal of tactical voting, especially between Labour and the Lib Dems, the election shows that there was a great deal of disaffection about our two main parties and a cry for new options. Our electoral system makes that hard for third parties that don’t have a clear geographical basis (unlike the SNP and the new lords of Wessex, the Lib Dems). Reform and the Greens did well IMO to win four seats each under our system. Even more surprising were the handful of Labour seats won by independents upset with Labour’s position on Gaza. Indeed they took down major Labour figures including Johnathan Ashworth and Khalid Mahmoud, and and almost Wes Streeting and Jess Phillips.

Clearly not all the public can be herded back into the two-party system by telling them to just vote tactically. Indeed, it is quite rational for people with ‘populist’ right or ‘decolonialist’ left views to try and derail the Conservatives or Labour in order to push the big parties to take them more seriously. After all, this worked wonders for UKIP. But that doesn’t mean they will come back into line quietly after this election. They may develop the taste for niche parties. And those niches, as with Reform, can make lots of headway into the mainstream.

This makes me suspicious of claims that the Conservatives simply need to ‘reunite the right’. Reform voters voted against you because they hate you guys. You already tried throwing them lots of bones with Rwanda etc and it didn’t work. They hate the system and you were the system for the past fourteen years. In time it might be possible but they are not prodigal children who will come back to dine with you because you kindly ask them. Similarly, Labour have a liability on the left and now that people have the taste for voting Green or for pro-Gaza independents, they may not come back to the fold quickly. One potentially good portent for Labour though is that they managed to turf out George Galloway in Rochdale, not least because they had a strong candidate in Paul Waugh.

5. Predicting is Hard

Finally, the book-consumption moment. How did my prediction go? Much like the immortal line from Chernobyl “not great, not terrible”. I did an updated prediction on election day morning, adjusting for the fact that Labour’s polling had already slipped below 40 percent. So let’s use that. Recall I entered the polling averages into my app, used proportional swing (which Dylan Difford still thinks was the right way to go), and had 30% tactical voting and 20% Reform voters coming back to Conservatives. I’m happy with these assumptions, not least because about 20% of the final Reform vote share did drop off and looks to have made up the polling gap for the Conservatives. Labour’s vote share was substantially lower than the last polling average but what can you do? And yes, it was more a proportional swing election than UNS I think.

Here were my final scores on the doors:

Conservatives: 91

Labour: 434

Lib Dems: 69

SNP: 28

Reform: 6

Plaid: 2

Greens: 1

And what did we end up with (with just 2 seats to declare - Inverness, Skye and West Ross-shire (prob SNP) and South Basildon and East Thurrock (who knows? Reform?, Conservatives?, Labour?))). I’ll put the difference between actual results and mine in parentheses.

Conservatives: 121 (+30)

Labour: 412 (-22)

Lib Dems: 71 (+2)

SNP: 9 (-19)

Reform 4 (-2)

Plaid 4 (+2)

Greens 4 (+3)

Independents 5 (+5) I think

So where did I go wrong. Firstly, the Conservative vote share was more like 24% not 20.4%. That clearly saved a lot of seats for them that might otherwise have gone Labour or Lib Dem (see the narrow wins by for example Mel Stride and Jeremy Hunt). Second Labour’s vote share was substantially lower than 39% in the end. So I think the big picture stuff between the two main parties basically comes from polling error (not my fault guv). On Reform, my bank of the envelope guess was pretty good and better than the exit poll (ya boo sucks to them). But I didn’t see the Greens or Plaid doing as well as they did. And most of all, my guesstimate that I should deduct 10 seats from the SNP seats that my app produced turned out to be not ambitious enough. They have been totally ravaged by this election.

Where did I go right? Well it turns out that my tactical voting part worked really well for the Lib Dems. Mark Pack, send your cheque in the mail. I’m frustrated it didn’t pick up the Greens too but I suspect it will do better the larger the starting shares are for each party - hence why it worked well with the LDs. I also think my overall guess was not too shabby. I never had the Tories on sink-circling levels like the infamous Survation MRP. But it’s also true I didn't have them in triple digits. My Uniform National Swing model was much more generous to the Conservatives, giving them 147 to Labour’s 388. But that turned out to be wrong in the opposite direction! So in reality the swing was probably somewhere between a UNS and proportional model and well, that’s what error bands are for.

But the big question. Will I eat a book. A nice chunky copy of Why Politics Fails, available at all good bookstores and delicatessens. Well, here’s what I said to Iain Mansfield when he asked where my book-eating would begin: “Book eating begins at over 150 Conservative seats since that’s when UNS is wrong too!”

Phew…

---

It’s been a wild few weeks on Political Calculus what with all the predicting and what not. Should you wish to hear a more reasoned and insightful chat about democracy, and in particular the role of the free press, do listen to this week’s What’s Wrong with Democracy where I achieve a life ambition and interview Eliot Higgins from Bellingcat. Go on, it’s really fascinating stuff.