Tactical voting and the Brecon by-election

How did we win the Brecon and Radnorshire by-election? To state the obvious, people changed their vote between 2017 and 2019. Contrary to most of the comment in the press from both Tories and Liberal Democrats, the main reason we won was not because of a remain alliance. The numbers are very clear. We gained 14.3% of the vote between 2017 and 2019.

Where did those votes come from? They certainly didn’t come from the Conservatives. The vote for the ‘pro-Brexit’ parties (CON, BREX and UKIP) stayed remarkably stable: they took exactly 50% of the vote in 2017 and 50.3% in 2019. Some of the votes came from Plaid but only 3.1%. So, apart from Plaid, we gained 11.3% from elsewhere. And there is only one place these came from – Labour. The Labour vote completely collapsed from 17.7% to 5.3%, a drop of 12.4%. Almost all of this seems to have moved to the Liberal Democrats. There is a discrepancy of 1% which equates to the Monster Raving Loony Party vote – where that came from, heaven only knows. Perhaps some ex-Labour voters decided Screaming Lord Sutch was a better leader than Jeremy Corbyn.

So, it is clear, the reason we won was because of Labour defections. If we had only the Plaid vote, the Conservatives would still hold the seat. The scale of the Labour collapse is unprecedented – historically, from the end of the war until the arrival of Thatcher, this was a safe Labour seat.

So why did the Labour vote collapse and why were the Liberal Democrats the recipients? It’s tempting to say that Liberal Democrat policies were the reason, particularly Brexit. The anecdotal evicence suggests otherwise. Many Labour voters on Twitter were encouraging their colleagues to vote Liberal Democrat but the reason they gave, more than anything else, was a desire to get rid of the Boris Johnson government.

The phenomenon of Liberals and Labour voting tactically against the Tories is hardly new. In a First Past The Post system it is often the only way to defeat a Conservative MP. I would suggest that it is going to be more commonplace in the future because of the extreme stance of this government. The Boris Johnson government must be defeated, not only because of Brexit but because it will be the most illiberal in recent history. Once the current phase of election bribes is over, it will be cutting taxes for the rich, closing childcare, starving the NHS of funds and negotiating disastrous treaties overseas.

The evidence from Brecon is that we will not gain a substantial number of seats without tactical voting from Labour in appropriate seats (remember 90% of our target seats are held by Conservatives). How can we encourage such tactical voting? There is no scope for national agreements with Corbyn, but might there be room for some co-operation at local level? What we must do is ensure that we don’t frighten off potential Labour defectors. Aiming to stress the points of agreement rather than points of difference would help. We can’t stop criticizing Corbyn’s party but we can focus criticism at the specific policies of the leadership rather than at the principles of social democracy that underpin both of our parties.

* Richard Taylor is retired, having previously working in IT. He is a Lib Dem member since 2010 who lost enthusiasm during coalition but didn't resign. He believes we have to regain our position on the left of politics.

Read more by or more about or .
This entry was posted in Op-eds.


  • Chris Bertram 5th Aug '19 - 11:18am

    “Perhaps some ex-Labour voters decided Screaming Lord Sutch was a better leader than Jeremy Corbyn.” Er, Screaming Lord Sutch died in 1999. Though when you think about it, he might still be a better option than the current “leader” of the opposition.

  • Nonconformistradical 5th Aug '19 - 11:22am

    “Screaming Lord Sutch died in 1999. Though when you think about it, he might still be a better option than the current “leader” of the opposition.”

    Or maybe even a better option as prime minister than the current incumbent…?

  • I think we did gain some votes from the Tories, with some Labour switching to the Brexit party.

    If we want more Labour supporters to lend us their votes in future elections we must be careful not to use that against the general party. Blame their current position and hold Corbyn to account, but predictions that it means the party is doomed will result in cold feet.

  • adarynefoedd 5th Aug '19 - 11:31am

    Good observations. As a former member, life long supporter of the Labour Party and B and R resident, I found it very hard to vote tactically and some of the leaflets were off-putting eg Labour cannot win here (Labour actually held the seat from 1945 – 1979 and only lost through boundary changes). What helped was working together for Remain and then Wales in Europe, better working together at Council level and our Lib Dem AM joining the Welsh Cabinet. What is now putting me off again are statements like ‘I will not work with Jeremy Corbyn’. I think if the Lib Dems are joining the main battle, they need to work with everybody who is pro-Remain.

  • Andrew McCaig 5th Aug '19 - 11:49am

    Agreed Fiona,
    The BXP candidate said their canvas returns indicated significant votes from Labour, and the polling data shows that as well.
    In more Remainy seats we will get a lot more votes from Tory Remainers especially if there is a no deal Brexit, and if trying to defeat a Tory, every one of those is worth two votes off the majority.

  • It is not necessarily just tactical voting by Labour voters. Polls show that the Labour share of the vote has dropped 12-15% since the 2017 election while we have increased by about 11%. Many of those former Labour voters were going to vote for Jane Dodds before any tactical voting came into play. Stopping a hard Brexit is the reason as much as dislike of a Johnson-led government. The object now is to keep those voters.

  • Chris Leeds 5th Aug '19 - 12:12pm

    As Fiona says, the Tory vote standstill is probably more complex than is immediately apparent.

    That said, I agree with the general tenor of your analysis Richard. We have to entice Labour voters. By all means attack Corbyn and Labour policies, but don’t attack the concept of their Party. Many of them hold their Party dear, even as they despair of its current state.

    There’s no prospect of an electoral pact, even restricted to individual seats, with Labour. They’re too tribal for that, both at national and local level. Therefore, while we may have formal pacts with other parties, with Labour it will be a matter of them standing in every seat, but we and others peeling away their vote.

  • Paul Barker 5th Aug '19 - 12:12pm

    An Electoral Alliance with The Greens (outwith Scotland) & Others would make it easier for “Left” Labour Voters to Vote for us. Lots of Lefty/Liberal types have a soft spot for The Greens, allowing them leeway that they won’t give us.

  • Liberal Neil 5th Aug '19 - 12:22pm

    It is a lot more complicated than voters switching because EITHER they dislike Johnson OR because the support a Remain alliance.

    These issues are intertwined, along with many others, not least the impact of running a proper campaign on the ground.

  • @Martin nails it here “When canvassing I understood that how people voted in the referendum was the strongest indicator of whether they might vote for us. In constituencies across the country we have to be able to attract voters form both Labour and Conservatives to win seats. No, despite agreements, we cannot be assured that all Green votes would transfer to us, though a local Green endorsement can be a great help. In Layla’s constituency, local political cooperation helped, but not all Green voters will have transferred.”

    We have the usual suspects calling for a leftwards turn to attract left-wing voters. Complete waste of time – left wing voters have many options to vote for and won’t vote for us*.

    In order to deny a Conservative majority, we need attract votes from across the centre ground of British politics, from the centre left to the centre right. Stripping the Conservatives of their voters reduces the number of seats they will win. Now Johnson is attracting back leavers from BXP it’s more important than ever we remove soft Cons, otherwise they will win a majority on c33% of the vote.

    *-as the usual suspects never tire of reminding us, they think we’re yellow Tories.

  • John Marriott 5th Aug '19 - 12:48pm

    When Mr Taylor talks about ‘tactical voting’, I am sure he is referring to general elections. But what about ‘tactical voting’ in Parliament? Unless it is recalled, we can write off the rest of this month, whilst Johnson and Cummings (more likely Cummings) will be putting together the blueprint for Vote Leave 2.0. It’s apparently now too late to squeeze in a General Election before 31 October so, rather than using any possible Motion of No Confidence when Parliament returns to trigger one, I hope that wiser heads on both sides of the house are considering trying to form a government of National Unity, with the sole aim initially of avoiding a No Deal Brexit. I know that Ed Davey for one wouldn’t rule this out, nor would a few enlightened members on the Labour and Tory back benches.

    You can call me naive if you like and the maths is distinctly dodgy; but what else is on offer to try to stop us crashing out of the EU on Halloween? So, how might it work? The big problem will, of course, be Labour. In all my thirty years of dealing with Labour councillors in local government, they always proved the most inscrutable and difficult to budge. Theirs is the true faith. Everything else is surrender.

    Of the other opposition parties in Parliament, the SNP, whether we like them or not, have claimed to be prepared to seek compromise. Of course their idea of compromise might be a pledge from a National Unity Government to agree to hold another Scottish Independence Referendum. I say we do what Cameron did over AV, namely allow a vote and then campaign like hell to get it rejected.

    So what if, within the statutory fourteen days, a coalition of Labour, Lib Dem, SNP, Change UK and Clarke/Grieve Tories were to emerge? It could start by revoking Article 50 on the basis that it could be invoked again if another referendum supported it. In the meantime, this new ‘government’ could approach the EU to reopen negotiations on a new ‘Deal’. If they knew we meant business, they also, given that all is not well in their organisation, might be prepared to compromise. If a new deal could be negotiated and agreed by Parliament then a preferential referendum could be held, with the choice between Deal, No Deal and Remain. Whichever choice eventually got over 50% would decide the direction of travel of the U.K. Pie in the sky? Probably yes!

  • Michael Sammon 5th Aug '19 - 1:16pm

    Good way to make all of those new members feel unappreciated who joined us previously from voting Conservative over the last three years. We do have different principles to Labour as well. We share a lot of common principles between liberals, social democrats and moderates on all sides.

  • I think the voting patterns are complex. The Tories lost votes to BXP and us, Lab lost votes to BXP and us, we gained votes from everyone. It is noteworthy too that our majority was larger than the number of votes polled by Plaid Cymru at the last election.
    Nationally we need to hit Labour areas hard, their vote is in danger of collapse.

  • We have an interesting by election in Worcester this week. Lib Dem about 200 behind Cons but here is a smallish Labour and Green vote to squeeze. Judging from trends over the past week I would expect the Cons vote to hold. Let us see what happens. 3 weeks ago this seat may have been a doddle for us, but this week……..

  • @Mike Sammon – “Good way to make all of those new members feel unappreciated who joined us previously from voting Conservative.”

    Well, quite. We need to be reaching out to everyone who broadly shares our Liberal values regardless of where they might have placed themselves previously. If we are serious about achieving power we have to assemble a coalition of support that straddles the centre.

    However, take heart from the fact that 2/3 of our membership have joined in the last four years and are young, enthusiastic and Liberal. Those who style themselves “retired … who lost enthusiasm during [the] coalition … [and] believe we have to regain our position on the left of politics”, whilst constituting a vocal portion of the commentators here, are unrepresentative of the party as a whole.

  • Richard Underhill 5th Aug '19 - 2:19pm

    Screaming Lord Sutch is dead. He used to present electoral results by adding the number of abstainers to the Looney vote. His biography shows a picture of his Rolls Royce. He explained the party policy debate (split?) on what would happen if they won an election, immediately resign? or what?
    The leader of the Monster Raving Looney Party is Howling Lord Hope who was present at the count, and sometimes complains about lack of funding.

  • Tony Greaves 5th Aug '19 - 2:34pm

    It is clear from here and from many Council elections that apart from a few diehards (most of whom would jump in the canal rather than vote LD, and haven’t changed that view throughout their lives) the wall of resistance to tactical voting for us by Labour voters which built during the Coalition has largely disappeared.

  • @Richard Taylor

    It’s almost impossible to model voter switching by comparing votes for respective parties between 2 elections. Fact that Party A’s vote went up 10% and Party B’s vote went down 10% does not at all mean that Party A is the direct benefactor of voters from Party B. There are Partys C, D and E to think about, as well as people who didn’t vote last time but did this time and vice versa.

    Short of opinion polling asking how voted in 2019 compared to 2017, you get a much better idea by undertaking a switch analysis of canvass returns (internal party data, it’s never going to be made public).

    I suspect a large number of Labour votes went to Brexit Party (and some to Lib Dems and Conservative too of course). And a fair number of Lib Dem votes to Conservative. But a larger number of votes from Conservative to Lib Dem. Plus I suspect better GOTV in 2019 compared to 2017 meaning more Lib Dems turned out to vote. And I suspect a fair number of Conservatives didn’t vote this time

  • Richard Taylor 5th Aug '19 - 3:40pm

    Thankyou for all the comment. I would accept that the vote movement I have suggested is an oversimplification. Having said that I would still maintain that it was Labour tactical voting that was the largest factor. I would also suggest that in LibDem Tory marginals, squeezing the Labour vote is going to be exremely important and we don’t want to scare off potential support.

  • David Evans 5th Aug '19 - 5:23pm

    Richard, I’m afraid your view that Labour tactical voting that was the largest factor, while persuasive, is not borne out by the facts. Labour’s vote fell by 5,600. Our vote rose by only 1,800. Even if all of these were Labour tactical voters, 3,800 Labour votes went elsewhere or didn’t turn out.

    Even bigger though was the fall in the Tory vote of 7,400. Whether they didn’t vote, voted Brexit or whatever, this was the key factor. What they and people like them do in a general election will be crucial.

    Of course all this assumes that our voters from 2017 stayed very largely solid. But of all the hypotheses about 2017 voters, I think that is the most certain of all.

    As an aside, one very concerning factor is that with politics in a state of turmoil and the biggest decision our country has had to make in over 40 years imminent, the turnout was under 60%. In the 1985 by election it was nearly 80%. That should be a massive concern for all parties.

  • Yes I don’t disagree (and I don’t think anyone here would either) that squeezing the Labour vote in Conservative-facing target constituencies is crucial. It’s been recognised for a long time that squeezing 3rd, 4th, 5th etc place parties is essential to win. That 3rd party may be Labour, Conservative, Green, even sometimes UKIP or a regional nationalist party (some UKIP voters are just really “anti-establishment” protest voters who intensely hate Conservative and Labour). Appealing to those voters to be squeezed doesn’t involve aping their party of usual support’s policies, but merely presenting the situation as a pragmatic decision

    I think since 10,000 people didn’t vote in 2019 compared to 2017 (circa 17%), that voter to non-voter was the biggest movement of votes, with the Conservatives (and probably Labour too) the biggest losers. Differential turnout and getting the Lib Dem vote out was certainly a bigger factor than squeezed Labour votes

  • Was turnout affected by the election being in August? Many teachers away on holiday etc.

  • Richard Taylor,

    historically, from the end of the war until the arrival of Thatcher, this was a safe Labour seat.

    I don’t consider a majority of 4.43% in 1951 as a safe seat and the majority of 5.2% in October 1974 wasn’t large enough for this to be a safe seat or the Conservatives wouldn’t have gained it in 1979.

    I agree we need to stress our policies which would appeal to Labour voters, however we also need not to scare people into voting Conservative by saying the Labour Party is extreme. I consider the 1983 Labour manifesto extreme and the 2017 Labour manifesto was not as extreme as that.

    John Marriott,

    A “National Unity Government” would need 320 MPs to support it. I think we can assume that at least 150 Conservative MPs and 150 Labour MPs would not support it (on the basis that about this number are “ministers” of one sort or another). The 10 DUP MPs will not support it. That leaves 329 possible supporters and if 10 of them voted against a “National Unity Government” being formed it would not be formed.

    It is therefore unrealistic to believe a non-Corbyn led coalition government could be formed. It has to be Corbyn led and maybe have Philip Hammond as Chancellor of the Exchequer with a commitment to reverse austerity and replace the existing budget rules with the one suggested by Neil O’Brien in the think tank Onward report ’Firing on all Cylinders’ – to just keep the debt to GDP ratio falling gently each year when not in a recession.

    Perhaps there are about 60 or 70 Conservative MPs who would vote against a Johnson government taking us out of the EU with no deal, but there are at least 26 Labour MPs who don’t support another referendum or a Remain position. If 60 Conservative MPs did support a Labour led coalition to have another referendum and a hold a general election by October 2020 it is likely with our support and the support of others this government could have a majority of over 30.

  • I suppose you could have a minority national unity government if the group forming it made up the largest whip grouping in the Commons.

  • @David, before you can realistically talk about votes gained or lost since the last election, you have to adjust for the lower turn-out. This may not be even across the board, but if turn-out for all parties is similarly impacted by the by-election downturn, then it would mean reducing each figure by approximately 12% so the 2017 results at 2019 turnout would be:

    Con: 20 081 => 15 663
    Lib: 12 043 => 9394
    Labour: 7 335 => 5721
    Plaid: 1 299 => 1013
    UKIP: 576 => 449

  • @Fiona I think it’s almost certain that some parties will experience a higher depression of their vote than others.

  • @Fiona

    But turnout % between elections can’t be assumed to apply evenly between parties. I suspect that most of those that voted Lib Dem in 2017 turned out to vote again in 2019 (not all necessarily for the Lib Dems, but the system of Lib Dem voter data collation and campaigning will heavily get those that are previously known to vote Lib Dem to vote). By far the biggest drop in turnout (by numbers) is almost certainly those that previously voted Conservative, though I suspect Labour were heavily affected by reduced turnout.

    By-election campaigning methods aim to ensure your support turns out to vote, that you switch as many of your opponents supporters to vote for you or to waste it on a 3rd, 4th or 5th placed party, that you maximise squeeze of 3rd, 4th, nth placed parties to vote for you, and that your opponents’ support is suppressed. Whether by design of the Lib Dem campaign, or by the Conservatives’ reduced appeal, the Conservative vote was undoubtedly heavily suppressed with enough staying at home to let in a Lib Dem win.

  • Agreed James. I said as much in my post. We can’t assume that reduction in turn-out is even, but we very definitely can’t assume that every Labour voter that turned out in 2017 turned out last week.

  • David Evans 5th Aug '19 - 11:05pm

    Fiona, I think you have missed the entire point of my post. You can’t simply adjust all parties by the same percentage. Indeed it is my point that it is a total misunderstanding of how by-elections need to be analysed to do so. To use percentages is easy, straightforward and totally misguided simply because it immediately removes the most important factor – differential turnout.

    About 9,500 fewer voters voted in the by-election than in 2017. The whole aim of our party in a winnable by-election is to get out more of our vote than the other parties. This we did. This is how we win even when times are tough, and starting in a by-election 8,000 votes behind is really tough, but it is what we have to do to win. Jane, Kirstie, the B&R Lib Dems and all the Lib Dems who helped did a great job. But if we are to learn from it, and so win more by-elections, we can’t simply wash away the biggest factor at the very start.

  • David, you are literally telling me I’ve missed something that I explicitly said in the first post you ‘corrected’, and you are doing it again after I politely pointed that out to you.

    Or course you can’t assume that all would be adjusted equally, which is why I pre-empted my illustrative revised numbers with that particular caveat in case it wasn’t obvious to everyone. This was only intended to be a starting point, and I was giving people credit for realising that those were not absolute figures and that they’d need further adjustment. The point being that the 2017 figures are not an appropriate starting point either. Adjusting the 2017 ones to 2019 turnout is a better, if not perfect, starting point for a lot of the electoral arithmetic. But I concede, yet again, that it’s only illustrative.

    I absolutely get that it’s about trying to keep our votes up, and hopefully we did that better than the other parties, but that doesn’t mean that we don’t need to factor in the reduced turnout, which is normal for by-elections and will be relevant to all parties to a greater or lesser extent, into the analysis.

    My post doesn’t contradict your point about how we worked to keep our vote out better than others, although I do disagree that we all need to be worried about the general reduced turnout. It’s higher than most constituencies and held up well compared to most byelections.

  • Teresa Wilson 6th Aug '19 - 9:00am

    John Marriott

    Cameron’s tactic of allowing a vote then campaigning like hell against it didn’t always end well.

  • Sue Sutherland 6th Aug '19 - 12:05pm

    It’s important to consider the tactics of voting because for a good few years the general public seemed mostly to understand what they needed to do to vote out a Government or LA ruling group they had grown to dislike. For those years we mostly targeted the Labour vote for squeezing and then suddenly we changed tack and went into Coalition with the Tories. Was it any wonder that our support shattered? In Labour facing seats it was a direct result and in Tory facing seats we lost the Labour squeeze votes. Relying on squeezing the vote rather than converting it was always a risky strategy.
    I think the difference now is that a lot of people actually want to vote for us rather than against another party. This is evinced by members of the Labour Party resigning their membership and joining us and some Tories doing the same. Obviously our Remain stance is an important part of this, but it’s also that the Left/Right ideologies are dominating both Labour and the Tories and we are not bound by those ideologies.
    What we are bound by is the Preamble to our Constitution. If we create policies in accordance with that statement of our beliefs we can appeal to many voters and remain true to our values.

  • Paul Barker 6th Aug '19 - 12:29pm

    With No-Deal Brexit 86 Days away & Parliament mostly unwilling to stop it, I am not sure that I get the point of this thread.

  • Richard Underhill 10th Aug '19 - 5:42pm

    Hazel Jones, a retired teacher, aged 71 is pictured on page 11 of The Times of 10/8/19.
    She is smartly dressed and cares for the future of her four grandchildren.
    She is quoted as saying ” We all have to do our bit and I think it’s very important that people are made aware of the imminent catastrophe that we will be faced with if Brexit goes through. My generation has fouled up the prospects of younger people. . .”
    She was a lifelong Labour supporter, but now votes Liberal Democrat because of the party’s commitment to a second referendum. She has determination and has received praise on social media.
    Perhaps she could be considered for a role as a borough candidate in a Labour held ward in Wakefield?

  • Richard Underhill 10th Aug '19 - 6:19pm

    Des Wilson was quite open about tactical voting in the 1992 general election.
    “We are in favour, providing it is understood that the tactic is to vote for us.”
    The late Enid Lakeman constantly explained that allowing people to express their preferences through the Single Transferable Vote would end the need for tactical voting.

  • I agree with Richard’s comments. The votes gained in order to win the seat of Brecon and Radnor were from previous Labour supporters NOT from Plaid or Greens. Those who supported the Plaid or Greens might well have voted for someone else rather than for us. We shouldn’t burn our bridges with Labour either because they have a much larger vote and many are prepared to vote Lib Dem tactically.

Post a Comment

Lib Dem Voice welcomes comments from everyone but we ask you to be polite, to be on topic and to be who you say you are. You can read our comments policy in full here. Please respect it and all readers of the site.

If you are a member of the party, you can have the Lib Dem Logo appear next to your comments to show this. You must be registered for our forum and can then login on this public site with the same username and password.

To have your photo next to your comment please signup your email address with Gravatar.

Your email is never published. Required fields are marked *

Please complete the name of this site, Liberal Democrat ...?


Recent Comments

    No recent comment found.