The 2017 Local Elections in Wales

Amidst everything else that has been going on, I have not had time to devote sufficient attention on this blog to the results of the recent local elections. However, with the very considerable help of my excellent PhD student Jac Larner (@Jaclarner on Twitter), I have been able to assemble some relevant information and a few thoughts.

Here, first of all, is a table with the total number of councillors won, the number of councils under majority control, and the net gain/loss in council seats, for each party in Wales:


Party Total Councillors Councils Net Seat Gain/Loss
Labour 472 7 (-3) -107
Plaid Cymru 202 1 (no change) +33
Conservative 184 1 (+1) +80
Lib-Dems 62 0 (no change) -11
UKIP 0 0 (no change) -2
Green 1 0 (no change) +1
Independents 322 3 (+1) +13
Others 6 0 (no change) -7


And here is the total numbers of votes cast, and the percentage of the total for each party:


Party Total Votes %
Labour 560,860 35.05
Plaid Cymru 240,061 15.00
Conservative 330,475 20.65
Lib-Dems 107,560 6.72
UKIP 13,247 0.84
Green 11,835 0.74
Ind/Others 335,909 20.99


At least a couple of issues arise from these figures. The first is the significant differences between the percentage vote totals in the table immediately above and those in the Welsh Political Barometer poll in late April. That poll gave the following figures on voting intention for the local elections:


Labour: 28%

Conservatives: 26%

Plaid Cymru: 19%

Liberal Democrats: 7%

UKIP: 8%

Others/Independents: 12%


Why should the poll have been so far out from the actual voting figures? There are several potential reasons. First, the poll was conducted around two weeks before the election, and many voters may have changed their minds in the intervening period. Polling companies normally like to conduct their ‘final call poll’, which is how their accuracy compared to the results is usually judged, as close as possible to an election – and certainly not two weeks beforehand. Second, local elections are particularly hard to poll in many places, including much of Wales, because of the strong tradition of local independent candidates and groups. Third, and perhaps most importantly, in many seats there were not candidates from all of the parties. Labour stood in the clear majority of Welsh seats, but the Conservatives only in about half, Plaid Cymru in somewhat fewer than half, the Lib-Dems in below one-quarter of all contests, and UKIP and the Greens in a much lower proportion even that that. Many voters may have intended to vote for a party in the local elections but been unable to do so for want of a local candidate.

Nonetheless, while there are legitimate reasons to excuse at least some of the inaccuracy in our poll, we should also be open to the possibility that it reflected some genuine error in measuring party support. If so, those errors were in the direction of under-stating Labour support, and over-stating support for the Conservatives, Plaid Cymru and UKIP. This should be born in mind over the next few weeks. Overall, our poll for the local elections certainly gave some indication of the direction in which the political wind was blowing, but not too accurate a reading of the strength of that wind.


Second – how should we interpret the results for each party? Here is my brief assessment:

Labour: I think Welsh Labour should be fairly pleased at these results. In England and Scotland the party had some pretty dreadful results. That was not the case in Wales. Yes, they did lose more than 100 seats, and control of three councils. But overall Labour losses were towards the lower end of expectations – my own were in the 100-200 seat range. Moreover, Labour held up particularly well in Cardiff, Swansea and Newport – something that is encouraging for the general election where all three cities have potentially losable seats that Labour is defending. It was not a great set of local elections for Labour, but the results were much better than some in the party will have feared. As with last year’s Assembly election, Welsh Labour showed considerable resilience in difficult circumstances. But will this transfer over to a general election where the focus will be on the UK-wide parties and leadership?

Plaid Cymru: Despite the more substantial gains made by the Welsh Tories, Plaid continue to be the second party of local government. They did make some gains, including winning seats in several areas of historic weakness. But I do think that some in Plaid, including Leanne Wood, rather over-spun the degree of Plaid success. I incurred the online wrath of some Plaid figures and supporters for suggesting this. But if finishing with less than 43 percent of the number of councillors that Labour has in Wales is really a great result for Plaid, it speaks volumes about the level of ambition in the party. These local elections were a step forwards for Plaid, certainly – but quite a small step.

Conservative: This was certainly a good set of local elections for the Welsh Tories: they made a net gain of eighty council seats, and regained control of Monmouthshire council. But they didn’t regain the Vale of Glamorgan while, for all their gains, the Conservatives still remain the third party of local government in Wales. There were some encouraging signs for potential general election gains – such as the advances the party made in Bridgend. But these were hardly gains on the scale seen in much of England, nor even in Scotland.

Liberal Democrats: A truly dreadful set of results. Beforehand I had scarcely thought it possible that the Lib-Dems could fail to make some net gains against the baseline of their appalling local election performance in 2012. But that is exactly what they managed to do. They did make modest advances in a few places. But they lost further ground in places like Swansea and Cardiff, where until a few years ago they were involved with running the council. These local elections have left the future looking bleak for the party.

UKIP were wholly irrelevant to the Welsh local elections in 2012, and they were so again. With their poll ratings falling all the time in Wales, UKIP seem to be sinking back into minor party status before our eyes.


  • Harry Hayfield

    Running the actual vote shares through my UK-Elect file for the locals, the numbers come up as very similar. Lab 484, Plaid 182, Con 173, Lib Dem 60, Green 0, Ind and Others 322

  • Anne Greagsby

    But why the rise in support for labour? Was this due to the general election being announced? The Tory wrap a round adverts in local newspapers causing anti tory panic?

  • Harry Hayfield

    The changes on 2012 in terms of vote share was: Lab -5.24%, Con +7.27%, Plaid +0.05%, Lib Dem -1.75%, UKIP +0.36%, Green -0.27%, Ind and Others -0.42% (suggesting a national swing from Lab to Con of 6.26% in line with national polls)

  • Owen

    How much would unopposed returns skew the national vote share? Voters in around 100 wards couldn’t cast a vote, with the large majority of these being in weak areas for Labour.

  • Dewi Harries

    The biggest distortion surely is the number of uncostested seats. Plaid got about 30 of these…a reasonable assumption would be that Plaid would have polled better than Wales average in these seats if contested. Thus actual figures underestimate Plaid support?

  • Tomos

    Did the latest barometer ask the question about how people voted in the local elections? Since that was after the election it shouldn’t have the same problems about parties not standing (as people would hopefully remember who they voted for).

    It would be really interesting to weight the poll by how people actually voted in the local election. Unless we see that, I remain very dubious of the large tory leads in Wales which are getting so much publicity and are driving political discussion in Wales.

    I really think there’s a responsibility to make that data available if it exists.

    If it wasn’t asked it might be worth asking in future as an alternative option for weighting.

Leave a Reply

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