Skip to main content
Coronavirus (COVID-19): Latest updates.


The June Welsh Political Barometer Poll

4 June 2020

If the last few years in politics have taught us nothing else, they should surely have made us aware that political fortunes can change very rapidly. In a context where a far lower proportion of voters are life-long supporters of one party than was once the case, what goes up can also go down – and sometimes very rapidly and to a very substantial extent. And if anyone had not yet learned this, our new Welsh Political Barometer poll should drive the lesson home rather effectively.

Our most recent Barometer poll, in April, was quite literally the best opinion poll ever for the Welsh Conservatives. For both Westminster and the National Assembly, the Tories were at their highest ever rating, and in clear leads: twelve percentage points ahead of Labour for a general election; and six points ahead for the Senedd on the constituency vote and nine points ahead for the regional list vote. This exceptionally strong Welsh Conservative position was in line with the trends observed in Britain-wide polls, where a post general election boost in the Conservative position had been further enhanced by a ‘rally to the flag’ effect in the first weeks of the Covid-19 crisis.

Things have changed. Recent weeks have seen increasing public scepticism in Britain-wide polls about the Johnson government’s handling of Covid-19. We have also seen steadily improving ratings for the new UK Labour leader, Sir Keir Starmer. More recently, the Dominic Cummings saga appears to have had substantial public cut-through, and caused significant damage to public evaluations of the Johnson government and the Conservative party.

All of these factors are reflected in the new Barometer poll. As I will discuss elsewhere, the poll shows a substantial decline in public assessments of the UK government’s handling of Covid-19, as well as a large setback in the Prime Minister’s own ratings. But there have also been significant changes in the standing of the parties. As always, our poll examined voter preferences for both a general and a devolved election. First, Westminster: here are the voting intention figures for a general election (with changes from the April Barometer poll in brackets):

Labour: 39% (+5)

Conservatives: 35% (-11)

Plaid Cymru: 15% (+4)

Liberal Democrats: 5% (+1)

Greens: 3% (+1)

Brexit Party: 2% (-1)

Others: 1% (+1)

These figures demonstrate a substantial turnaround in party fortunes since April. Labour are back in the lead. But while they have improved their position in the last two months, the change is primarily about Conservative decline from the extraordinarily strong position they had last time. While the Welsh Tories’ position in Wales remains strong by all historic standards (and is only just below their vote-share in last December’s general election), to fall eleven points between two polls is highly unusual. Of the other parties, Plaid Cymru will be most cheered, with the Conservative decline benefitting them almost to the same extent as it does Labour.

What might this mean in terms of parliamentary seats? Using the standard method, of projecting uniformly across Wales swings since the last general election, we generate the following projected outcome in terms of seats (with changes from the December 2019 election result in brackets):

Labour: 21 (-1)

Conservatives: 15 (+1)

Plaid Cymru: 4 (no change)

Our previous poll had seen the Conservatives projected to gain eleven parliamentary seats, on top of the fourteen they won in the December 2019 general election. Those eleven projected gains have now been cut to one: the Alyn and Deeside seat that the Tories very nearly won in December is now projected to be very narrowly gained by them.

The changed political landscape can be seen every bit as well by looking at devolved voting intentions. The April Barometer poll saw the Conservatives in unprecedented leads on both votes, and projected to be the largest party in the Senedd. None of that is any longer the case. Here are the figures for the constituency ballot (with changes in support since April once again indicated in brackets):

Labour: 34% (+2)

Conservatives: 31% (-7)

Plaid Cymru: 22% (+3)

Liberal Democrats: 5% (+1)

Brexit Party: 3% (-1)

Greens: 3% (+1)

Others: 3% (+2)

Our previous Welsh poll showed that the surge in Conservative fortunes in Wales was not limited to Westminster. But the strong influence of British-wide political tides on the electoral fortunes of parties in Wales – an influence that is these days very much weaker in Scotland – can hurt as well as help a party. Our new poll demonstrates that the setbacks the Tories have experienced in recent times are impacting them at the devolved level as well. And as at Westminster, this appears to have worked to the benefit of both Labour and Plaid Cymru.

The April Barometer poll had projected (under uniform national swings) the Conservatives to capture nine constituency seats from Labour. Now those gains have been cut to five seats: in current order of marginality, those are Vale of Glamorgan, Vale of Clwyd, Gower, Wrexham, and Cardiff North. Meanwhile, Plaid are projected to gain Llanelli.

For the regional list vote, the new Barometer poll generated the following results (with changes since the April Barometer poll once again in brackets):

Labour: 32% (+3)

Conservatives: 28% (-9)

Plaid Cymru: 24% (+6)

Liberal Democrats: 5% (+1)

Abolish the Assembly: 4% (no change)

Brexit Party: 3% (-1)

Greens: 3% (no change)

Others: 1% (-1)

These results reinforce the picture of Conservative decline putting Labour once more in the lead, but also benefitting Plaid Cymru to a significant extent.

Allowing for the constituency results already projected, and once more assuming uniform national swings since 2016, our new poll projects the following overall results for the Senedd’s regional list seats:

North Wales: 2 Plaid, 1 Labour, 1 Conservative

Mid and West Wales: 3 Labour, 1 Conservative

South Wales West: 2 Conservative, 2 Plaid

South Wales Central: 2 Conservative, 2 Plaid

South Wales East: 2 Conservative, 2 Plaid

From all these figures we can thereby generate the following overall projected result for the Senedd:

Labour: 25 seats (21 constituency, 4 regional)

Conservatives: 19 seats (11 constituency, 8 regional)

Plaid Cymru: 15 seats (7 constituency, 8 regional)

Liberal Democrats: 1 seat (1 constituency)

The last few weeks have seen the greatest public health crisis for many years. In that context, with suffering and deaths having been widespread, the polling fortunes of our political parties may seem of little concern. But what people think of those who lead our governments, and our political parties, does matter for many reasons. In Wales we are less than a year away from the next Senedd election, and the political fallout of the current crisis is likely to play a big part in shaping the fortunes of all the parties in the run up to that campaign.

Our previous Welsh Political Barometer poll produced historically good results for the Conservatives. But I said at the time that the electoral landscape could alter very rapidly – and so it has.

The poll, for ITV-Cymru Wales and Cardiff University, had a sample of 1,021 Welsh adults aged 18+ and was carried out online by YouGov from 29 May to 1 June 2020.


  1. Paul Culloty

    Certainly, the Tory collapse (comparative to April at least) is largely attributable to their handling of the Covid crisis, and the perceived dichotomy in approaches between London and Cardiff, with Drakeford gaining the “rallying vote” that has buoyed various European incumbent parties recently. Perhaps little surprise either that Plaid has also gained, with Yes Cymru trumpeting a membership surge on their social media platforms – but is this a false dawn for the nationalists, or can they use this as a foundation for more lasting progression?

  2. Jac Pro-Biotic

    These numbers are interesting, yes, but the exceptional volatility shown since the last poll demonstrates how (and apologies for the bluntness) meaningless they are.

    The only numbers that matters these days are the death toll from the virus. And apologies again for the miserableness, but to understand these figures you have to do the comparisons. Compare the death toll in the ‘United’ Kingdom from the virus with that of other European states, and you will see that more people are dying daily in the UK than in the EU27 put together. That doesn’t look good for the powers-that-be in Westminster. Compare the death rate per capita in Wales with that in England and it looks even worse for the Conservatives, and shows that at least Welsh Labour is doing some things right. (Then look at how there is a concerted campaign by a lot of Tory MPs and ASs to demand that we follow England’s lead … and wonder how these people can sleep at night).

    The death rate is, of course, explicitly NOT about numbers but about flesh-and-bone – fathers, mothers, uncles, aunts, grandparents, siblings, cousins, friends … Voters will remember this the next time they are invited to place their X in a box.

    You reap what you sow. Look out Welsh Conservatives – your numbers are going down.

  3. Richard Harris

    Could those figures be why RT Davies finally lurched into spluttering life this morning and offered some VERY belated criticism of Dominic “Total Fantasia” Cummings? Surely not…

  4. Recce

    In response to Jac Pro-Biotic, trying to compare counties is pointless. Spain’s numbers have plummeted because they have changed how they record the deaths. Hence their drop from over 250 / day to an average of 1 / day.

    If the UK had followed Spain’s recording methodology, the UK would have recorded 20 rather than the 359 it did report. Of course if the UK changed it’s counting in such a way there would be cries of cover-up.

    Have a look at this thread from the FT’s data analysist John Burn-Murdoch

  5. Derek Baskerville

    Would have liked to have seen figures for the current estimates of seats to be won by each party compared with the actual number of seats won in the 2016 Senedd elections.

  6. Christian Schmidt

    There are now clear differences in Covid19 measure between England and Wales, and these are covered in the UK news – in fact I wonder if Wales has ever been mentioned more often UK national telly?

    But while the big mistakes by the Tories are obviously covered, as well as differences between England and Wales, Welsh Government mistakes are not given much prominence. So I think Wales appears better. And while Plaid is not in government, I wonder whether a process like ‘the news show Wales as different + don’t show the Welsh Government as incompetent as the UK government = devolution is good, therefore I support the party most associated with Welsh being different’ may be partly behind the good Plaid numbers?

    (Just a thought, but it would be interesting to see whether there is any correlation between Plaid’s number s and how often Wales [other than in a rugby context] is in the UK news…)

  7. dave taylor


    As we all know causes of death rate are very difficult to attribute, did you know INFECTION rate per 100k in Wales is 390 in England is 240 so we are 50% more like to catch it , however our DEATH rate per100k Wales 44.7 England 48.9 not a huge difference, but summary is we more like to catch it in Wales but less likely to die of it.

    This had me wondering why until I found an article giving data/papers from doctors and pathologist etc across the world suspecting this is a blood disease causing clotting etc not a respiratory one, it uses the lungs to access the blood cells. One of the results they came up with was that people on statins and blood thinners (smokers) or diabetes on control drugs are much less likely to catch it as their drugs use the ACE2 cell to access blood stream which is the same one that Covid wants to use and finds them blocked, however don’t start smoking because if you are in one of those groups and do catch it you are far more likely to die from it.

    Now how does that affect our figures in Wales, I hypothesise that we have less smokers and diabetics per 100k than england so our infection rate can be higher but our death rate will be lower

  8. Ddirpytnop

    Dave Taylor – The official death rate per 100k in England is around 63 – much higher than in Wales. At face value, the official figures suggest you are much more likely to contract Covid-19 if you live in Wales but much less likely to die of it. Suprisingly, I have seen no discussion of this in the media.

    • dave

      population Eng 56m = 560 blocks of 100k, Deaths Ebg 29673, deaths per 100k is 29673/560 = 53

  9. Owain Dafydd

    With the changes to the voting age in Senedd to allow votes at 16 will there be a change to the methodology of the next Barometer poll? And what, if any, impact might it show?

    • dave

      Yep it was passed by 41 to 19 it had to have 2/3rd (40) to pass pity the same rule did not apply to the original referendum, but foreign nationals who have zero allegiance to UK or Wales are they mad or desperate.

      Still thank heavens it won’t apply in UK parliamentary elections

  10. Ddirpytnop

    Dave – PHE’s official death toll in England is 35k not 29k. Not sure where you got the latter figure from . Look at the UK Government’s website for the most up to date figures. Alternatively look at ONS’ higher (and more accurate) figures.

    • dave

      The ONS figs include all death certs that mention covid whether they were tested or not the source I used cleared stated deaths were certified as tested, hence the difference. apols . However I have checked on and clearly the site I used is lagging, but there is a very interesting piece of data on gov site which does confirm my assertion you are far more likely to catch it in Wales than England

      Nation Total cases Rate
      England 153,807 274.8
      Northern Ireland 4,773 253.7
      Scotland 15,553 286.0
      Wales 14,238 453.6

  11. Petroc ap Seisyllt

    Jac probiotic appears to think the Covid deathntoll is relevant, actually total UK death rates are below the may norm. (And they no longer vote) The incompétent management is the issue. School are closed because elderly teachers are at risk not the pupils.

Leave a reply

Your email address will not be published. Required fields are marked *