This originally appeared as a guest post on James O’Malley’s substack. The reception to it was much better than I expected, and was shared by Ann Black, Michael Crick, and the Politico London Playbook amongst others.
Labour has won a remarkable victory. If, on the day that Keir Starmer became leader, you had offered me a majority of one at this general election, I would have taken it. I still have to pinch myself to believe that we have a majority of one hundred and seventy-four1.
Very many thanks must go to Keir Starmer, David Evans, Pat McFadden, and Morgan McSweeney for making this happen. Nothing justifies the election strategy quite as much as its resounding success. However, this strategy was not perfectly executed and has had consequences that will have to be addressed.
The strategy was, simply put, to win as many seats as possible. It shows from how far we have had to come that this even needs saying. More specifically, the plan was for our vote to be as efficiently distributed as possible by winning lots of seats by narrow majorities, even if meant that majorities went down in safe seats2.
And crucially it also meant not just abandoning, but actively suppressing and aggressively de-prioritising campaigning in non-target seats.
To achieve this, the campaign managers took three steps.
First, they continued the traditional practice of ‘twinning’ non-battleground seats with target seats, so that activists from one area would focus their time and energy where there was greater need.
Second, campaigning in non-target seats was strongly discouraged by party officials, and was made harder by the late selection of candidates.
Then thirdly, campaigning was made nigh impossible by cutting off access in non-target seats to Contact Creator, the voter database used by the party. Not having access to it is a bit like arriving at the crease to discover that the club secretary has tied your shoelaces together.
So in a sense, this election victory was partially thanks to incredibly effective targeting of campaigning resources. And though our targeting was clearly very effective, there is evidence that it wasn’t perfect, and in the future we should allow for near misses.
For example, North West Cambridgeshire was not a target seat, but returned Sam Carling as a Labour MP3.
And similarly, Peterborough, was a target, and was thought safe enough to redirect campaigners to Kettering with two weeks to go. And though Kettering returned Rosie Wrighting by a relatively comfortable 3,900 votes, Andrew Pakes only won Peterborough by 118 votes.
In total, there are twenty non-target seats where Labour lost by less than 2,000 votes. If we had allowed those seats to have selected their candidate earlier, it could have been enough to win them. Equally, there are target seats that had resources poured in where we did not win.
Campaigners are not fungible
I’m not suggesting that priority shouldn’t have been given to target seats – it absolutely should have been. I’m not even suggesting resources should have been allocated to non-target seats. However, our targeting is not so good that the national party should have hampered campaigning in non-target seats.
And even if our targeting is entirely accurate, not allowing campaigning in non-target seats can reduce our ability to campaign in target seats.
This is because campaigners are not a fungible resource4. The party cannot and should not assume that it can instruct volunteers where to go. Lots of campaigners – I emphasise here that ‘campaigner’ and ‘party member’5 are not synonyms – are only able to go out in their constituency, or perhaps even on their own street.
Others are willing to fly the flag further afield, but need encouragement and experience. At every election, new people will come out to get Labour and Labour candidates elected. You can usually convince them to travel to target seats, but not always straightaway.
In other words, by making it harder to campaign in non-target seats, the party made less campaigners available for target seats, particularly when more active members were alienated by the party machinery.
One of the things elections are good at is enthusing new and old campaigners to be active. But because the targeting strategy went so far, we have missed that opportunity. When coupled with late selections and no selections, a lot of Constituency Labour Parties (CLPs) have been left feeling demotivated and unloved.
And worse, they have been prevented from doing useful work for forthcoming local elections and metro mayor elections. While these are not as important as elections to the House of Commons, non-parliamentary elections, particularly for metro mayors, matter. They make a difference to people’s lives. And besides, if we must be mercenary, they also help build campaigning strength for general elections.
Missed opportunity
As I hope I have made clear, I am not arguing against targeting, but that it needs tweaking. Regardless of that, there are costs to targeting so ruthlessly. These costs should be addressed anyway, but particularly because of the volatile political times in which we live.
For example, our vote distribution was efficient6. That also means our coalition is fragile. We have seen collapsing and rising red and blue walls. We cannot assume that the winning coalition in four or five years will be the same as at this election, and that does mean we need to at least give ourselves the option of building new coalitions.
And this means that the CLPs that have been written off need to be given at least some support between now and the election. There are seats that we won this time round that we will struggle to hold at the next election and seats we didn’t win that we will have a real chance to win.
This brings me to another reason for wanting to make sure we are doing at least some campaigning across the country too: Reform.
The rise of Reform is multi-causal but part of it is a feeling of disengagement from politics coupled with a belief that the mainstream parties do not care about ‘people like me’.
This feeling of being taken for granted opened up a space that the BNP exploited in Barking and Dagenham in the first decade of this century. Part of the problem was Labour not doing enough work outside of elections7. And a way of bringing voters back from Reform to Labour, and from abstention to Labour, is literally just to have Labour people talk to them and take them seriously. But to do this, we need to allow our campaigners to campaign more broadly.
So, finally what to do now? I think there are several things that Labour high command can do to make nice with its activists.
The first step is just to be nicer. People who stood for selection should not have found out they were unsuccessful by social media, but by a phone call or email from the party. Regional officers could have said thank-you to candidates in non-target seats after the election. The pressures put on candidates in non-target seats and the manner in which those pressures were communicated were unacceptable and need to be checked ahead of the next general election.
And structurally, twinning needs to go both ways. CLPs8 who were the beneficiaries of twinning should be encouraged to pay back some of that support. Having a few experienced members come to a less well developed CLP for a day so that newer campaigners can learn the canvassing ropes would be a small but welcome start.
The party is going to need to do some work – not a huge amount, but some – to reassure some members. Agreeing a process for selecting candidates well in advance of the next general election that allows member involvement and then sticking to that plan9 would be a start. Promising not to take down Contact Creator two weeks before election day would be another.
On both fronts, making it clear that local election candidates will not be imposed by regional offices without references to members would be welcome, as well as either allowing or making sure that candidates are selected in good time for their campaigns too.
I don’t think we can go for a UK version of Howard Dean’s ‘50 state strategy’, not least because political parties here don’t have the resources. We need to keep targeting. However, we can’t afford to completely write off any constituency – if we are a national party, we have to act like it, but we also have to deal with the malaise that hangs around our democracy at the moment. A good way of doing that is to, well, do democracy or, at least, give ourselves the chance to do as well as we can across the country.
I am hopeful that the party is recognising some of the issues with its hyper-targeting campaign. But it should remember: Volunteers need to be encouraged, not instructed, and remember that targeting – which is the right thing to do – does cause problems and those problems should be fixed between general elections.
1 At the time of writing.
2 Jeremy Corbyn may have won the argument, but Keir Starmer won the election.
3 And the first MP to be born this century.
4 A pound coin is fungible in that any pound coin is very much like any other. A work of art is not fungible because the Mona Lisa is not the same as the Fighting Temeraire. Non-fungible tokens are a scam. This has nothing to do with fungi, although party members do sometimes feel like mushrooms – kept in the dark and occasionally having manure thrown over them.
5 There are lots of party members who pay their subs and don’t do much else. That’s absolutely fine, but you do get non-members enthused by a candidate who want to campaign in their local area that just aren’t going to go to another constituency for the good of the cause.
6 We won lots of seats by small majorities, instead of piling up votes in safe seats. In the UK, winning elections means winning the most seats, not the most votes. This may or may not be a good thing, but it is where we are.
7 Which is hard, unglamorous, unthanked, but essential work
8 Although some places have moved to all-member meetings, the traditional form of organisation is Branch Labour Parties covering a single ward or group of wards that send delegates to the Constituency Labour Party’s General Committee, which in turn elects the Executive Committee. There is then a regional party, a National Executive Committee, a National Constitutional Committee, and so on. Any similarity to the nine circles of hell is purely coincidental.
9 From a purely administrative point of view, CLPs can handle almost any system, but having to book a big hall and organise voting for several hundred people at short notice can be challenging for volunteers who actually run local parties.