When will the Tory rot stop?

So just how wrong have I been about the UK general election? My first prediction was that there would be a limit to the Conservative decline (at about 100 seats), and that there was no chance that the Liberal Democrats would overtake them. The Reform UK insurgents would be unable to replace the Tories. In my second post I suggested that the Conservatives would narrow the gap on Labour by squeezing Reform, limiting the overall size of their loss.

The second prediction is not faring well. I was wrong-footed by Reform’s leader, Nigel Farage, changing his mind about standing himself, and being the formal party leader. His initial declaration got Reform off to a bad start, and offered the Tories an opportunity. But the Tories made a weak start, and Mr Farage is back, attracting media like bees to a honeypot, including the supposedly establishment-biased BBC. The Conservative attempt to squeeze Reform supporters has stalled, at least at the national level, and even gone into reverse. More than one MRP poll now suggests that Reform might even pick up a handful of seats.

But I’m holding fast to my first prediction. There is quite a lot of breathless talk about the Tories being crushed, with well under 100 seats retained. One or two MRP polls suggest that they will indeed by eclipsed by the Lib Dems. Reform cheerleader Matt Goodwin, whose Substack is one of my main sources of information on the populist right, is talking of an inflection point, with Reform overtaking and crushing the Conservatives. Even the Financial Times‘s Stephen Bush finds himself dealing with the speculation that the Lib Dems might overtake the Conservatives in terms of seats won – though he still considers this unlikely.

My logic is rather different from that of other commentators, though. It is that the actual results in many constituencies are determined by the activity of local activists on the ground – canvassing, leaflet delivery, posters and such, often referred to as “the ground war”, as opposed to the “air war” of media. The ground war helps voters decide who is really in contention in their seat, and then to decide on the least-worst option. This is more likely to change minds than anything that happens in the air war, that attracts most of the attention. I am a case in point. I have voted for the Lib Dems or their predecessors in every general election from 1983 (my first vote, in 1979, was for the Conservatives). The only thing that could possibly persuade me otherwise this time is if I thought my local Green candidate was in serious contention in my constituency of East Grinstead and Uckfield – she is my local councillor, and the only one of the candidates I know – and whom I’m quite impressed with. She claims that this is the case based on local election results; but the Lib Dem candidate makes the same claim based on 2019 election results (complicated by the fact it is a new seat); meanwhile the tactical voting recommendation from Best for Britain is for Labour, based on an MRP poll. It’s as clear as mud, though I think the Labour claim to be ridiculous. The Tory candidate (Mims Davies, the MP for the now-defunct Mid-Sussex constituency) has not given up, and has been canvassing in our village, in the extreme south of the constituency. She will surely have quite an easy ride back into parliament, even if she fails to get 40% of the vote, which is likely.

The Conservatives lack opponents with a proper ground-war capability in so many of their well-off rural strongholds. Labour have never been strong in these seats; the Lib Dem infrastructure is only there for some seats (locally to me in Lewes and Tunbridge Wells, for example), and the Greens in even fewer (the non-rural Brighton Pavilion in my area). Reform is not constructed as a grassroots organisation, and will struggle without a charismatic candidate and an experienced local organiser – which applies in a maximum of three or four seats, I suspect. The MRP poll methodology only picks up the ground war factor indirectly and underplays it. This makes its constituency predictions unreliable in each individual case, though more plausible in aggregate, which is how their success is usually judged. My guess is that the unique conditions of this campaign mean that their aggregate conclusions are out too in most cases – and that the Conservatives will do better than forecast. Though to be fair, there is quite a bit of variation in the MRP seat projections. The ground war factor mean that it is unlikely for the Lib Dems to pass the 50 seat mark, or the Greens 3.

Furthermore, I think Mr Goodwin is running well ahead of the evidence in his claims about Reform. So far I think only one poll, by YouGov, has placed them ahead of the Tories, and that was a week ago. Aggregated poll trackers tend to put them several points behind. Mr Farage’s volte-face gave the party a boost but it doesn’t look like sustained momentum. They may pick up a lot votes in areas where the ground war is weak, such as in safe Labour seats, which might lead them to widespread second places, but little impact on the overall result (a bit like the SDP-Liberal Alliance in 1983 and 1987, and the Lib Dems in 1992 – though in their case it was mainly safe Conservative seats where they did well).

Having criticised recent MRP polls, though, the overall result predicted by Ipsos-Mori (Labour 453, Conservatives 115, Lib Dem 38) is not far from where I think things will end up – though the individual constituency projections are wayward – giving Labour far too big a share in Lib Dem target seats, for example). MRPs are a poor way to guide for tactical voting decisions. The Tories may do a bit better than this, and Labour worse. I hope the Lib Dems will do a bit better too, but then my hopes have run ahead of the actuality for every election since 1983, apart from 1997.

So, to answer the question posed by my blog title, I expect the Tory rot to stop soon, with their ground war strengths serving to limit the damage. What happens next is the big question – but let’s wait until after the election to consider that.

Postscript

Just after I posted this, three new MRPs came out – and Matt Goodwin breathlessly reported a new poll showing a spectacular surge in Reform support to 24%, with the Conservatives languishing on 15%. This poll was conducted by Mr Goodwin’s own organisation, People Polling, for GB News, another Reform promoter. This is curious. Mr Goodwin is an academic and I have no reason to doubt his professionalism in conducting polls. But it is an outlier. It is clearly becoming harder to conduct opinion polls, and a lot of the result depends on how adjustments to the raw results are made. There is always a lot of soul-searching after elections amongst pollsters, which then leaves the impression that they are refighting the last war rather trying the win the current one. Mr Goodwin says that the key change is that a lot of the previous “don’t knows” are making up their minds, and that Reform are doing very well here and the Tories very badly. There is other evidence for the second of these contentions (though that suggested Labour doing well) – a disaster for Rishi Sunak, whose whole campaign was based on the idea that he could rally “don’t knows” who had previously voted Tory.

I don’t think the new MRPs add anything to my analysis, though I haven’t looked at them closely. They show the Tories coming in at roughly 50, 100 and 150 seats. My opinion is that the higher end of the range is the most realistic forecast. I think generally that the MRPs are over-reporting Labour strength in seats that the party isn’t targeting – probably to the benefit of the Tories, but maybe the Lib Dems too.

20 June 2024