False choices about Labour’s recovery

by Sunder Katwala

If there has been one thing that has been symptomatic of Labour’s struggle to find a viable future strategy for electoral success, it is the penchant of too many in the party for daft debates about which voters the party does not want.

New Labour began by building the biggest tent British politics had ever seen, and ended by worrying endlessly about whether appealing too strongly to traditional Labour voters or Guardian readers would kill the project off. Meanwhile, the party’s left flank fretted about whether the support of marginal swing voters stopped Labour being Labour in government. If those were the problems, 29% of the vote would have been a solution. Neither Labour’s traditional base nor New Labour switchers saw the point of having Labour in government at all.

Well, here we go again.

The latest daft question: if Ed Miliband and the Labour party want to win the next election, should they seek to win votes from the Liberal Democrats, or from the Conservatives?

Uncut’s own Dan Hodges set up this choice at the New Statesman.

On the one hand, there is the compass analysis. The compass crystal ball has not proved infallible in the aftermath of the last election, but it now seems to mean pessimistically admitting that Labour will probably never ever win again under first past the post, so must negotiate a way to power with the Liberal Democrats.

On the other, we keep the New Labour flag flying by treating the collapsing Liberal Democrat vote as a distraction to be ignored entirely, because the only votes that count are those won from the Conservatives.
Another senior Labour insider put it this way:

“Ed has a clear choice. He can chase after a non-existent progressive majority, or he can try to bring middle and working class Tory voters home to Labour. Or, to put it another way, he can try to win on his own, or lose with Chris Huhne.”

It would be difficult to imagine a sillier debate about “electoral strategy”.

Perhaps the one thing that everybody serious about finding Labour’s path back to power could do is to refuse this framing, and to laugh at anybody who tries to start this debate. Neither Neal Lawson nor Dan Hodges are right about Labour’s route back to power.
Why? Well, let’s take it as given that Labour wants to win the election – by seeking to win as many seats as possible, and a majority of the House of Commons.

That is going to require close to 40% of the vote. Labour starts on 29% in May 2010.

Labour had 44% in the heady landslide of 1997, and held on to most of it in the 2001 replay.

Where did those lost votes go?

The Conservatives increased their share of the vote by five points from 1997 to 2010 – from 31% to 36%.

The Liberal Democrats increased their share of the vote by five points – from 18% to 23%.

I am not sure one needs a GCSE in maths to see a small problem with this “which votes don’t we want” debate.

We could win every Lib Dem lost voter (which Labour won’t) and it would not be enough to secure a Labour majority. You could also take every single voter who voted New Labour before going Tory in 2010 (which Labour won’t) and still be floundering around in the low-30s, unless Labour got the often affluent middle-class voters who went LibDem as well.

When New Labour put on 9% of the vote between 1992 and 1997, they were targeting a Tory vote which began on 42%.  It had 11 points to fall before reaching that English Tory bedrock which voted for Major in 1997, Hague in 2001 and would have voted for IDS in 2005.

What is striking about the Tory decline and recovery is how strongly the tribal Tory vote held up at 31-33% across their three election defeats, and how little progress Cameronism made, in hitting a ceiling of 36%, so far short of John Major’s winning share.

In 2010, Labour did lose votes more to the Conservatives than the Lib Dems, as Cleggmania failed to materialise, though it was not the AB swing voters who left in large numbers, but skilled and unskilled workers who were more likely to swing to the right.

In 2005, Labour lost 6% of the electorate, losing four times as much support to Charles Kennedy than to Michael Howard.

It is true that Labour has made most progress on the easier front. It has gained about two million votes from the Liberal Democrats over the last year.

One in five May 2010 LibDem voters are probably pretty firmly in the Labour camp, because of the formation of the coalition and the policy choices it has made which do not appeal to the most left-liberal segment of Lib Dem voters. The starting line for the next general election is probably something like Conservatives 36%, Labour 33% and LibDems 19%. That would put all of the parties into hung Parliament territory – before the battle begins. David Cameron needs to ensure he takes a swathe of more pro-coalition Lib Dem votes too, while there is another group of now undecided, perhaps more centrist, Lib Dems who all three major parties will hope to capture.

There is certainly going to be a Labour-Conservative front at the next general election, and Labour will need to make inroads into it. If the next election is closely contested, that may well ultimately prove the decisive battle. However, it will remain true that it has never before been a narrower front with voters fewer up for grabs (in either direction) that at any recent British general election. If there is not a complete economic meltdown, the Tories will surely poll 33-34% and very likely more. Ed Miliband’s task is to keep them below 35-36% while David Cameron needs to get up to 40% to govern alone, which is probably his only option next time.

It will prove more difficult for David Cameron to take votes from those who stuck with Gordon Brown, and probably to appeal to first-time voters too. So, essential to a David Cameron majority is that he needs to damage the Lib Dems in the south as much as Labour does in the north. (Cameron will also benefit and take seats on the basis of a Labour recovery in Lib Dem-Tory marginals, as some keep the Tories out tactical voting unwinds; and his greatest triumph is to make a trade with the Lib Dems where they voted for new electoral boundaries and a smaller Commons, getting in return the slap in the face from the Tory-led campaign in the AV referendum).

It will matter where these voters are (and we now have the electoral system for the next election, but not yet the constituencies). Tim Horton’s snap post-election report on the electoral impact if a Tory-Lib Dem coalition were formed demonstrated in detail of why the Lib Dems should have been able to anticipate paying a heavy electoral price. What has been overlooked is that this affects the Labour-Tory battleground too, not just the university seats gained by the LibDems.

As Horton set out, there are 25 Tory marginals where Labour would start with a lead in the two-party battle on the basis of one in five Lib Dems changing sides. These include Hendon, Thurrock, Broxtowe, Bedford, and Hove in the south and midlands, as well as the Tory gain of Stockton South in the north – before the Tory-Labour battle for votes begin.

One paradox for those in the Labour tribe who want Ed Miliband to “go it alone” and avoid seeking to negotiate a route to power after another hung Parliament have an even stronger reason to appeal to those voters who deserted Labour for the LibDems over the last two or three elections, as well as to win voters from the Conservatives. The only way to avoid having to talk to Lib Dem politicians after the next election would be to have forged a Labour case that appealed much more strongly to a large section of their voters than Labour did in either 2005 or 2010.

The electoral argument between the two Milibands in the leadership contest was important, but the differences were perhaps overstated and misunderstood.

Both were pretty clear that Labour’s electoral ambition needed to extend beyond trying to win the 2005 general election again. (There has been little or no advance in Tory detoxification since May 2010 – perhaps rather more the opposite – but Labour’s electoral hopes can not be pinned on the return of Michael Howard).

That was why Ed Miliband was against the “one more heave” politics of returning to some old New Labour winning formula a generation later. It is why he stressed the case for the party to change more rather than less, shifting a party culture and organisation which was built in response to the communications environment of a generation ago. It is odd that his critics seem to believe that he offered to craft a winning electoral coalition for the Labour party within a year of a collapse to 29% of the vote. That might have been possible if not being the Tories, and not being the coalition, was enough. The whole point is that it wasn’t.

To rebuild a winning electoral coalition, Labour will need to win voters from Lib Dems and Tories, and to regain support for the general election from the SNP too. It will need working-class votes, concerned about economic insecurity, and those concerned about crime, immigration and cultural anxieties. It will need affluent middle-class voters, whether they are concerned most about the economy and taxes, spending on the NHS and childcare or civil liberties.

Sometimes, there are important trade-offs between the views of voters. Often, their interests are less zero-sum than so much pro and anti-New Labour electoral analysis consistently implies. As John Denham has argued most consistently, the point of New Labour’s pre-1997 appeal was that it sought to make a broad one nation argument, yet seemed to forget that too soon after entering office.

Labour needs a compelling political narrative and project again. It won’t get back to power by the alphabet soup of electoral segmentation.  Finding the argument and strategy that can build a broad enough electoral coalition to provide a coherent governing project is not going to be easy.

Still, that is the price of political power.

The one thing that is a recipe for permanent opposition is yet another debate about which voters Labour would be better off without.

Sunder Katwala is general secretary of the fabian society. He blogs his personal views – read more at www.nextleft.org.

Tags: , , , ,

11 Responses to “False choices about Labour’s recovery”

  1. oldpolitics says:

    Good post. I think the timing issue is absolutely correct – we’re doing better against the Lib Dems because their voters from 2010 haven’t got what they wanted. Tory voters, by and large, have got what they expected – cuts. At the moment, a lot of talk about cuts and feeling good about reducing the deficit, but very few actual cuts.

    As the real cuts get made over the coming year, we’ll find out how many Tory voters are as keen on cuts that affect them in practice as they were on cuts in theory (which always only affect other people). Cuts at the sharp end will make great stories, even for state-hating quasi-tabloids like the Mail and Express. At that point, the split between those who decide the cuts aren’t a price worth paying, and come to Labour, those who think they are, and stay Tory, and those who believe they could be avoided if we stopped giving so much money to foreigners (UKIP) will be fascinating to watch.

    We also need to keep an eye on the ‘blame game’, not just those who blame Labour for the cuts, but those who blame Councils for Council cuts, despite the fact they are simply passporting central cuts. As Labour takes over more Councils, we need to win that battle. In the meantime, however, we’re doing far better than it was reasonable to expect at this point, and we shouldn’t be ashamed to say so.

  2. CS Clark says:

    Excellent stuff. If I have one criticism, it’s that I can imagine at least one sillier debate about “electoral strategy”. It’s one where the correctness of a policy isn’t even balanced against how popular it might be.

    If you pushed me I might be able to do another five before breakfast.

  3. MattNW5 says:


    Of course that thesis relies on people being stupid enough to believe that Labour cuts wouldn’t be anything like as significant as the coalition ones. Coming up with a coherent strategy is surely far more important just than electoral maths

  4. Dan Hodges says:

    I see Sunder has joined the ranks of those reassuringly telling Ed he doesn’t have to make a decision.

  5. Edward Carlsson Browne says:

    Thank you. A thoughtful corrective to some ideas that, as you say, do not fit with the electoral mathematics.

  6. Sunder Katwala says:

    Dan’s response borders on the silly, as it relies on ignoring what the piece says in the last 4 paras. As I have argued before, I think Ed Miliband, and others around him, need to do quite a bit more this summer and Autumn to articulate more of his strategy and direction ahead of the party conferences.

    What I am questioning, in broader debatss rather than just Dan’s piece, is zero-sum thinking about the electorate, such as thinking an Observer report in which the Labour leader calls for LibDems to support Labour is a barrier to winning votes from the Conservatives. Perhaps somebody could identify where the content did that.

  7. Guido Fawkes says:

    There is of course a scenario where the LibDems or a faction of the LibDems stick with the Tories. And why not if the economy has bounced?

  8. Alan L says:

    Some interesting points, but your election number crunching does not take into account the significant variations in turnout in the differing elections.

    If you factor this in, you see that Labour + LibDem votes cast is pretty constant over 1992 – 2010, suggesting that most traffic has gone between these two in the last 20 years. Similarly Tory + Did not vote is also pretty constant, suggesting that disgruntled ex-tory voters do not vote, rather than switching to Labour.

  9. Dan Hodges says:


    To be mature then, if you want to appeal to Lib Dems, you can’t simply say, “ditch your sandals and join us brothers”.

    You have to articulate a policy agenda on immigration, crime, civil liberties, trade unions, welfare reform and electoral reform, to name a few, that are diametrically opposed to those you would offer Tory/Lab switchers.

    It’s comforting for some in the movement to think that choice doesn’t have to be made, but it does.

  10. Sunder Katwala says:


    Thanks for a reply.

    Yes, politics involves choices, for sure.

    But the specific claim is very zero sum, and I think too much so, because “diametrically opposed” just overstates the level of tensions on the issues you list, though these may be significant on some specific challenging issues for Labour – such as crime, immigration and Europe. But there are other issues (NHS most clearly) where the interests align.

    I agree that it may well be too little to say ‘you were betrayed, so you have to vote for us’ if that shows a sense of entitlement … But your NS piece seemed to argue that was enough to ignore LDs. and/or that the LD voters don’t matter.

    Civil liberties …. Labour should be clear that it is going to place more weight on civil liberties, and find some significant symbols of that shift. I suspect that would be useful with some possible Tory switchers as well as more important to LibDems

    …. For example, I don’t believe that ID cards were popular and salient with Tory/Lab swing voters. They may have polled well as a general idea, but that is very different from charging people for them after massively overspending on systems, etc. But they were very salient with a liberal niche. We should have ditched them publicly. (On the other hand, I think CCTV does involve a choice between different views. I would keep it, and defend it, where people want it, and argue that liberals who find that difficult need to win the argument on the ground).

    Crime might be a good case for your argument, as we saw when last in opposition, except it maybe isn’t in a case where the Coalition sticks with a Ken Clarke agenda.

    Constitutional reform: I doubt electoral reform is coming back very soon, but I would be very surprised if our AV referendum commitment was an electoral liability with anybody. I suspect we lost votes to the Tories on other issues. I am not aware of evidence that our extensive 1997-2001 constitutional reform agenda offered any major barrier to Tory switchers, and it was certainly attractive to LibDems, as well as being important to Labour in Scotland post-1992. (Europe may be an exception here, where Labour is electorally vulnerable if it went super-integrationist or was perceived to have done so). Who is going to not vote Labour because we propose a fully elected Lords, say, or votes at 16?

    Trade unions: I don’t think there is a clash here between LibDems and Tory switchers, unless LibDems have somehow become massively in favour of unions, which I haven’t noticed … there may be a clash between Tory switchers and other Labour voters. Where there are “pro-union” policies that have very broad public support (as with minimum wage … living wage campaigns … ) then

    Immigration: yes, there may be a clash here between different groups of voters: Labour ended up seen as very lax on immigration and very harsh on asylum, and it seemed with no support from anybody for its managed migration approach.

    So, yes, there would be some tensions in a winning electoral coalition, which must surely be true of every winning electoral coalition since 1867. But I still think it makes considerably less sense in 2011 to say ‘its only really Tory voters we need’ than it did in 1993.

  11. john p Ried says:

    Sunder- I don’t believe that ID cards were popular and salient with Tory/Lab swing voters.
    More due to the money they’d cost rather than the liberty issue, Plus, I really don’t think it was the reason that anyone went form Laobur to tory, either

Leave a Reply