Roger Awan-Scully


The first Welsh Political Barometer poll of election year has arrived. This latest measure of public attitudes delivers the very latest evidence on the political state of play, just a few months before the scheduled Senedd election in May. And it shows that contest shaping up to be an increasingly close, three-way contest between Labour, the Conservatives and Plaid Cymru.


With fewer than four months to go until the scheduled date for the Senedd election, our new poll once again asked about voting intentions at both the devolved level and Westminster. Polling for the Senedd once again included 16- and 17-year-olds, to take account of the newly expanded franchise; respondents in this age group were, however, excluded from the sampling of voting intentions for a UK general election.


Given the looming devolved election it is natural to look first at Senedd voting intentions. Here are the results of our latest poll for the constituency ballot (with changes in support levels for each party since our last poll, published in early November, indicated in brackets):


Labour: 34% (-4)

Conservatives: 26% (-1)

Plaid Cymru: 22% (+2)

Greens: 6% (+3)

Reform UK: 5% (no change on Brexit Party last time)

Liberal Democrats: 4% (+1)

Others: 4% (no change)


These numbers appear to show a clear weakening of Labour support since before Christmas. Before we get carried away with interpretations of such an apparent change, however (for instance, attributing it to the Welsh Labour Government’s handling of Covid-19 in recent weeks), we should note that this change simply reverses a four-point Labour rise in our last poll. That November poll may have thus been a slight outlier, with matters reverting to a more normal reading in our latest measure of public attitudes. However, with Labour’s rating apparently down, and Plaid Cymru’s support level apparently edging up slightly (although again by an amount well within any sampling error), the picture painted by our poll is much more of a serious thee-way contest for the Senedd than November’s poll suggested. Then, Labour’s reported support was almost twice that of Plaid; now there are only just over ten percentage points separating the top three parties.


None of the other parties appear to be pushing themselves into the constituency contest for the Senedd as serious players; while the Greens will be pleased at a relatively good showing, any chances they have of winning a Senedd seat surely come in the list vote. Probably the most important feature of the constituency vote findings for all these smaller parties is the continued lack of significant improvement for the Liberal Democrats. With their national poll ratings being so dire, and with Kirsty Williams standing down in their one remaining seat, there must currently be a very good chance of the party finally losing any Senedd representation in May (or whenever we end up voting).


A uniform swing projection of the changes in party support since May 2016 indicated by this poll suggests that three constituency seats in the Senedd would change hands, with the Conservatives very narrowly gaining the Vale of Glamorgan and Vale of Clwyd from Labour, and Labour also losing Llanelli to Plaid Cymru.


And what about the regional list vote? Here our new Barometer poll produced the following results (with changes since the previous poll again indicated in brackets):


Labour: 30% (-3)

Conservatives: 25% (+1)

Plaid Cymru: 22% (+2)

Abolish the Assembly: 7% (no change)

Greens: 5% (+1)

Reform UK: 4% (-1)

Liberal Democrats: 4% (no change)

Others: 1% (-2)


Just as for the constituency vote, we see for this ballot a notable decline in Labour’s support. With the Conservatives and Plaid Cymru’s rating edging up marginally on this ballot, it again reinforces the suggestion of a close, three-way contest for the Senedd – only eight percentage point separate the three leading parties on the regional vote. It is also interesting that the rise in support for the anti-devolution Abolish the Assembly party seen in the last Barometer poll has been sustained into 2021.


Allowing for the constituency results already projected, and as per usual 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 Conservative, 1 Abolish the Assembly

Mid and West Wales: 2 Labour, 1 Conservative, 1 Abolish the Assembly

South Wales West: 2 Conservative, 2 Plaid

South Wales Central: 2 Conservative, 2 Plaid

South Wales East: 2 Conservative, 2 Plaid


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


Labour: 26 seats (24 constituency, 2 regional)

Conservatives: 16 seats (8 constituency, 8 regional)

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

Abolish the Assembly: 2 seats (2 regional)

Liberal Democrats: 1 seat (1 constituency)


Labour must still be long odds-on to emerge as the largest party in the Senedd after the election. But our new poll does suggest that they may be pressed harder by their Conservative and Plaid opponents than had looked likely in late 2020.


What about the picture for Westminster? Our final poll of 2020 had suggested that Labour were possibly extending their lead in Wales. Our latest set of general election figures are shown below (with shifts since the previous Barometer poll in brackets):


Labour: 36% (-7)

Conservatives: 33% (+1)

Plaid Cymru: 17% (+4)

Reform UK: 5% (no change)

Greens: 4% (+1)

Liberal Democrats: 3% (no change)

Others: 2% (no change)


These figures show a striking decline in Labour support in Wales – one that is notably larger than for the devolved level. Again, this may be something of an outlier – at the very least it will be something to keep an eye on across future polls. The main apparent beneficiaries of Labour’s slump in Westminster support are Plaid Cymru. Their 17 percent support in our new poll is the highest they have ever recorded for Westminster in any Welsh poll this century of which I am aware (narrowly edging out their 16 percent in a July 2016 YouGov poll). Other than this, all apparent changes in party support levels since we last reported in November are small, and well within any ‘margin of error’.


What might these numbers from the new Barometer poll mean in terms of parliamentary seats? Using the standard method of projecting swings since the last general election uniformly across Wales generates the following projected outcome in terms of seats (with changes from the December 2019 election result indicated in brackets):


Labour: 21 (-1)

Conservatives: 14 (no change)

Plaid Cymru: 5 (+1)


Only two seats are projected by this poll to change hands: the Conservatives are projected to narrowly gain Alyn and Deeside (having narrowly lost it in the general election), but the Tories are also projected to narrowly lose Ynys Mon to Plaid Cymru. Of course, these projections are based on the current constituency boundaries, with forty Welsh seats. That number will very likely be reduced to 32 by the time of the next general election, with substantially revised boundaries.


Overall, our new poll indicates a tightening of the race in Wales as the next devolved election approaches. Quite when we will be voting, and what sort of campaign the election will witness, remains uncertain. The latest evidence from the Welsh Political Barometer suggests that the outcome of the Senedd election is also rather less certain than some might have thought.


Author biography

Professor Roger Awan-Scully is the Chair of the Political Studies Association and previously the Head of Politics and International Relations and Professor of Political Science at Cardiff University. This article was first published on Cardiff University's Election in Wales blog and has been reposted with the permission of the author.Image credit: UK Parliament/Flickr.

The Welsh Political Barometer poll, for ITV-Cymru Wales and Cardiff University, had a sample of 1,018 Welsh adults aged 16+ and was carried out online by YouGov from 11-14 January 2021. Figures for Westminster voting intention include only those respondents aged 18 and over.