Winner of the New Statesman SPERI Prize in Political Economy 2016

Tuesday 23 March 2021

As things stand, the chances of defeating Johnson at the next election are minuscule


I don’t think this is understood by the leadership of the main opposition parties, along with many others. The Conservatives are way ahead in the polls at the moment, because the government had the right policy on vaccine procurement and delivery. That will fade, but it could quickly be replaced by an economic recovery bounce. The economy is likely to grow rapidly both this year and next, so the recovery bounce could last until 2023. As usual, much will be made of ‘record growth rates’ and rather less to where we are in comparison to pre-pandemic. For this reason, I suspect Johnson will call an election well before he needs to (having repealed the FTPA). Their erroneous reputation for managing the economy will appear enhanced. The Conservative election campaign will be slick, with vast amounts of money spent on social media, and the electorate will be rigged by insisting on ID. Johnson's victory will be assured.

Of course, as a former Conservative PM is said to have said, “events, dear boy, events”. For example Johnson could rush on with ending his third lockdown even though R becomes greater than one, and combined with not enough people being vaccinated this leads to another wave of COVID cases, possibly involving virus variants that make some vaccines less effective. But even during the second wave, the severity of which was entirely down to government ineptitude, Labour never took a consistent lead in the polls. [1] For reasons including those outlined in my last post, everything has been stacked against the opposition for some time.

Yet it is imperative that this government is defeated as soon as possible. It is an authoritarian government with immense power because of its solid majority, and the longer it stays in power the more difficult it will make the life of any opposition. So the only relevant question is what can the opposition to this government do to maximise its chances of winning before it becomes even harder to win.

The most obvious and familiar example of how it can be done is the United States, and the first lesson has to be for all of the opposition parties to cooperate in fighting seats in the next election. What that has to mean is that just one opposition party fields candidates in key marginals, as the LibDems and Greens did in 2019. Biden would not have won if there had been another significant liberal party contesting virtually every state.

Another way to see why a socially liberal (aka progressive) alliance is essential is to divide voters using two dimensions, the familiar left right and social liberalism or conservatism (sometimes called authoritarian). This breakdown is essential to understand politics today, and for those unfamiliar with it I outline the reasons why it is essential in an appendix. Without doubt nearly all socially conservative right wing voters will vote Tory, and socially liberal left wing voters will not vote Tory, and the two groups are similar in size. The battle in any election is to win over those in the other two quadrants: left wing social conservatives and right wing social liberals.

Once you see this, the folly of dividing the English socially liberal vote among three parties becomes obvious. Cooperation of the kind mentioned above will greatly increase the chances that the Conservatives will not win. Given the strong possibility of an election at the end of 2022 or early 2023, the sooner discussions start the better. An alliance is a clear win for the two main UK wide opposition parties (by seats). Labour will play a leading role in the next government rather than staying in opposition. The Liberal Democrats will win more of their target seats where there are currently many wasted Labour votes.

But such an alliance alone is not enough, as the appendix explains. Each party needs to play to its strength. The LibDems target seats are winnable in part because right wing social liberals could be persuaded to vote with their liberalism rather than their pocket. But that in turn means Labour have to focus on winning socially conservative left wing voters. The Tories know that is their route to power, which is the reason for all this nonsense about wokeness, being nasty to asylum seekers, ending the right to demonstrate and so on.

I should say at the start what this post is not about. It is not an argument for Blue Labour. Labour’s base today is socially liberal, and if Labour were to become a socially conservative party its core vote would go elsewhere.

Nor is this post about the left vs centre-left within Labour. What I have to say is relevant to Labour led by the Left or Centre-Left. After my last post, many on the left pointed to 2017 as proof that Labour could do relatively well without compromising its liberalism. But in 2017 the issue that divided social liberals and conservatives in a way no issue has for decades was Brexit (see appendix). Because Labour in 2017 supported Brexit, then this together with a strong left wing economic package helped win many social conservative left wing voters.

The activist, marginal voter divide

I recently wrote a post trying to justify Starmer’s strategy of avoiding being seen to champion social liberal issues. I think it is fair to say that many people hated that post. But to try and convince people that I was right in principle, I want to reference a fascinating discussion between Noah Smith and political data scientist David Shor. Although that concerns US politics, the parallels with UK politics are very close. What Shor is saying is very similar to what I was trying to say in my earlier post.

One of the first points Shor makes in this discussion is that party activists are the worst judge of what works at winning elections. He gives the example of the mirror ad used by Clinton in 2016, which used Trump’s derogatory comments about women. The Clinton people thought it was a great ad and put a lot of money behind it. What Shor found was that among the marginal voters that any Democrat needs to win over, the white socially conservative working class (and to some extent the non-white socially conservative), it actually lost votes.

In my view a great deal of comment on twitter is based on this fallacy. Activists typically want their politicians to talk about the issues they care about, and get upset when they don’t. But if a party leader just talks about what activists want them to talk about, they will probably do badly in an election. Specifically for Labour, a leader who wants to win elections needs to appeal to socially conservative left wing voters. Labour almost certainly, and a progressive alliance probably, cannot win without these voters. No Labour leader in opposition is going to change these voters socially conservative beliefs (for what does, see later). That means a Labour leader has to focus on economic issues from a left wing perspective, and not champion social liberal issues and concerns.

A Labour leader who understands the argument above has to walk a tightrope. If they go so far as to advocate for socially conservative issues, they risk losing their socially liberal base to one of the other social liberal parties (or to not voting). The clearest case of that was Labour during the course of 2019. Labour leaders do best when they don’t adopt socially conservative issues, but don’t push socially liberal issues either. (Issues where many social conservatives agree with liberals are fine.) A progressive alliance avoids that tightrope to a large degree, because social liberals fed up with Labour’s silence on these issues don’t get the chance to vote Green or LibDem in key marginals. But it does not eliminate it entirely because voters may not vote, but more importantly it is crucial to keep your base onside [2].

Because it is a tightrope, it is very easy for Labour leaders to get this wrong. In particular a leader, or the team around them, may become so focused on their target voter that they go too far in denying their social liberalism. [3] The recent bill allowing the police to prevent noisy protests or protests that cause annoyance is a good example of where Starmer made a mistake and corrected it. The best thing to do is turn an issue around so that people think about it differently. Whatever you think about Blair, his ‘being tough on crime and the causes of crime’ did exactly that.

What you do, and what you talk about

Many of the reactions to my earlier post failed to make the distinction between what an opposition talks about to win power, and what it then does in government. Not talking about certain issues in an election campaign is not the same as being indifferent to them in government. All I am talking about is what a Labour opposition uses its scarce airtime to talk about.

This point is so obvious to Conservatives that it is second nature. A Conservative party will not campaign on privatising the NHS but that does not stop them doing it when in office. Equally the Labour party under Blair did not campaign for an independent Bank of England, but it still enacted one.

What many people remember is a Labour government, both with rhetoric and actions, trying to appease the anti-immigration mood created in large part by the right wing press. That was worse than pointless, because it failed to put the case for immigration and validated the idea that immigration was a problem. It failed to use the government’s ability to put a case (which a Labour opposition hardly ever has), and alienated many of its own voters. Once again, that is not what I’m talking about here.

Election campaigns, which for oppositions last five years, involve promoting your most popular policies. For successful Labour oppositions that is going to involve left wing economics policies but not socially liberal policies. Governments have much more latitude in how they conduct election campaigns. Because they get much more exposure in the media, they can focus on the popular things they have done rather than the more unpopular. To take an example, Shor argues for the US that on gay marriage making those issues partisan is dangerous.

Another reaction I got to the earlier post was that someone needed to champion social liberalism when the Conservatives are in power. But Shor argues that what changes public attitudes in a liberal direction is the broadcast media. The broadcast media, either through documentary or fiction, can normalise minorities among those who were previously antagonistic to them. The reason the broadcast media promotes social liberalism (and has done so during the steady advance of social liberalism over the last 50 years) is that programmes are typically made by younger people with university degrees, who tend to be socially liberal.

That does not mean activists should just sit back and wait for this progress to happen. There is a key role for non-partisan campaigns to put issues on the agenda, and to expose government hypocrisy on issues like climate change. But a Labour leader fighting an election campaigning on social issues that divide social conservatives and liberals is unlikely to persuade any marginal voters and will just lose Labour votes.

[1] The broadcast media is critical here. The Conservative politician that persuaded Johnson not to follow the scientific advice in September was the Chancellor Sunak, yet Sunak remains the most popular politician. The reason is that the broadcast media never mention this key point, so it remains largely unknown to most voters.

{2] This is one reason why Starmer attacking Corbyn is a bad idea. Not only does it not play well with many activists, it also reminds voters of party divisions.

[3] Basing policy too closely on red wall focus groups can be a big mistake in many ways. To give one recent example, I agree with Michael Walker here that Starmer failed to present an alternative way of thinking about the pandemic. He needed something positive rather than the negative of Johnson acting too late. In particular he needed to bust the excuse the government uses constantly of having to trade off lives with the economy, and he needed to quote numbers for the deaths the government caused. What might appear negatively as an aggressive and unsupportive stance among focus groups will reap dividends in the period which matters, which is post pandemic.

Appendix: Two dimensional mapping of voters

Rather than just categorise voters as somewhere on the left-right continuum, more recent voter analysis has looked at two dimensions, where the second axis is social liberal to social conservative. To assess where voters are on this axis, a standard method is to look at answers to questions of national surveys. You can then see how those individuals voted on key issues. Here is an example from the Financial Times at the end of 2017.

The labels given to the extremes of the ‘cultural’ axis vary a lot. I prefer social liberal to libertarian for obvious reasons. Here is an example of the kind of questions used to plot where voters are, taken from a presentation by Paula Surridge.

The FT diagram is typical: most Labour voters are somewhere in the social liberal/left quadrant, and most Tory voters are in the opposite quadrant. It is the two other quadrants that are more contestable by the two main political parties. I’m going to follow an excellent
paper by Noam Gidron and label those voters who are left wing but socially conservative 'welfare chauvinist', and those that are right wing but liberal 'market cosmopolitan'. The trick for parties of the left or right is to get voters in these two groups to vote for them. Gidron argues that the decline in left wing parties in Europe is due to both groups tending to vote for right wing parties rather than left wing parties.

The policy in the UK that did this most successfully for the Conservatives was Brexit, as the diagram shows. It is clear that Brexit appealed almost exclusively to social conservatives, capturing the welfare chauvinist group. It follows that this was why Corbyn’s 2017 campaign was the most successful for Labour since Blair: Labour appeased social conservatives (by accepting Brexit, and also talking about police funding) and focused on left wing economic policies, thus capturing many welfare chauvinists.

In both the UK and US, geography together with unusual voting systems mean that the marginal voter is left wing but socially conservative (welfare chauvinists). Social liberals are concentrated in large cities or university towns, and are normally in the minority elsewhere. That means that in a battle between social conservatives and liberals (like Brexit) voters may be evenly balanced in numbers but social conservatives will win far more seats. The 52% to 48% Brexit referendum, if fought over seats, is estimated to have produced 406 Leave seats and just 242 Remain.seats.


  1. I agree on the strategy for Labour less of the 'virtue' issues and more of the stuff that affects 90+% of the electorate. Sadly the socially liberal issues are the only thing that unites Labour's 2 wings?

    For Labour Liberals and Greens to unite is a very broad church without any historical factors and without the symbiosis that binds the two sides of Labour unhappily together.

    They need to play on their field. The Dems being outsmarted by Bannon for instance in USA. He made a quip that 'they' would be better for black people and the Dems disappeared down a rabbit hole of their love of black people* for 2 weeks and revived the Trump campaign by playing on their opponents chosen field.

    I asked a friend who watched Labour debates did they discuss trans rights and he was yes at length. Next what was their stated economic philosophy? Never mentioned.

    Labour needs to right itself first or this coalition would merely presage chaos given it would need the neo liberal nationalist SNP probably.

    * Irony Obama Admin': foreclosed on poor people 5.5 Mn evictions disproportionately black: Obama personally told poor black people to drink the poison water in Flint: allowed Big Ag to openly discriminate against small Black farmers etc now this was not racism but class hatred as their failure to tackle the opioid epidemic showed. Plus when these murders by the cops occur who is the Mayor and whose Govt's passed crime bills and welfare 'reform' and armed the police? Who backed mass incarceration in the 90s etc etc

  2. “The Conservatives are way ahead in the polls at the moment, because the government had the right policy on vaccine procurement and delivery. That will fade, but it could quickly be replaced by an economic recovery bounce.”

    I think this is too simplistic. It is not just that the Government has done well on vaccination. It is that its performance is in contrast to that of other Governments (notably within the EU) that have done badly. So too will it be on economic recovery.

    Potentially, it looks as if the Eurozone is lined up for Euro-crisis mark II. Greek debt to GDP is now more than 200%, for Italy it is 160%. Spanish unemployment was 14% even before the pandemic. Maybe, the EU can avoid a crisis if it is agile, but its recent performance on the vaccine issue doesn’t inspire confidence. It is quite conceivable that the later the Government goes to the electorate the bigger will be the disparity between the EU and UK recovery.

    The EU’s problems largely arise from the North-South disparity and it is the latter which has suffered most economically from the curtailment of tourism. No doubt UK ministers are aware of this, so expect overseas holidays to be restricted this summer. All for health reasons of course.

  3. Excellent article. Thank you. This has helped "crystallise" my understanding of this question.

  4. Minuscule! I considered being picky about that, but I am happy that nature has taken its course without me.

  5. It was clear to many (though to not enough at the time) that the 2019 general election was the most serious in our recent history. An authoritarian, Bannon inspired, govt was going to ride a wave to a majority; with a liberal coalition the only thing that could stop it. I have much time for both Lib Dems and Green parties, but they should be chastising themselves for treating 2019 GE as just 'another election', as that approach led us to what we have now. Jo Swinson was reckless. Corbyn for all his faults realised Brexit couldn't be put back in the bottle, but needed to be owned by a responsible govt. We should have had a left coalition govt headed by Corbyn temporarily until the dust settled.

  6. The lib-right quadrant looks rather empty except around the edges. I suspect though that of the four quadrants, lib-right ideology has the most elite support (including among MPs) relative to general population support, for the same reason that a member of the elite is far more likely than the average voter to be a university-educated landlord. Elite disapproval of your political programme creates a very difficult environment in Parliament, the media, funding etc, even if your policies individually have very large majorities of voter support (as is true of a number of "left-wing" economic policies).

    It may be the Lib Dems would be the natural home of this quadrant in other circumstances, but it's not clear how that gets the Tories out of power: at the voter level, there are not enough lib-right voters in marginal seats to make a difference, and at the elite level, there would be an awkward clash of messaging between them and Labour, given that Labour doesn't want to speak up on social liberalism but is overtly opposed to the lib-rights on economic policy. Perhaps the best we can hope for from lib-right influencers is a kind of British Lincoln Project, i.e. trying to create an alternative "liberal Conservative" narrative that critiques the corruption, incompetence and authoritarianism of the present government, without saying much about Labour directly.

  7. I am sorry to be coming late to this post. An interesting and helpful post on which I will reflect. However you make no reference to Scotland. It remains to be seen what is the outcome of Holyrood elections on 6th May. In the 2010 General Election, the Labour Party won 41 of the 59 Scottish Westminster seats; in the 2019 election it won just one. Surely the Labour Party needs to win back some of these and even then it will certainly need the support of SNP MPs. However, unless it recognises the increasing demand for independence/"devo max", it hasn't a hope of regaining much support in Scotland. It needs to make concrete proposals not vague promises of a Constitutional Commission. I am a former Labour voter but to me voting in Scotland, the Labour Party is now simply irrelevant .


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.