May-Day – and May’s Week?

There is no post-war precedent for having a general election campaign already underway whilst the local elections take place. Living in a competitive ward myself, in a normally marginal seat, I’m feeling the seasonal love that an election brings, and being bombarded with the literature coming through the door. This Thursday will see local elections in some councils in England, throughout Scotland, and across all twenty-two local authorities in Wales. But which party will be flying the flag of victory come May 5th, and will this be a taste of things to come on June 8th?

Of course we can expect variations in the council elections because of particular local politics and the strengths or failings of different local parties. But there is no doubt that the national mood does and will influence the results we get in the local elections. The most striking recent example of this is the fate of many Liberal Democrat councillors between 2011-15; the national unpopularity of the party after it entered the coalition government translated into many hard-working local Liberal Democrats losing their council seats.

The national mood at present is defined by Conservative dominance. Though many still view them as “the nasty party” (a term coined fifteen years ago by the current Prime Minister), the reality is that the Tories currently face no serious or credible opposition outside of Scotland and Nicola Sturgeon’s SNP. Jeremy Corbyn has not persuaded most people that he offers effective leadership, while the Liberal Democrats are only at the start of a long road back after their crushing defeat in the 2015 general election.

The strong position of the Conservatives, and the weakness of the other parties, was reflected in the recent Welsh Political Barometer poll. This pointed to significant Tory gains, and substantial Labour losses, in the general election. Something similar is also likely this Thursday. Colin Rallings and Michael Thrasher from the University of Plymouth have estimated council seat gains and losses based on the Barometer poll’s question on local election voting intentions. This, they suggest, points to the following overall changes:

Labour: -130 seats
Conservatives: +90 seats
Plaid Cymru: +20 seats
Liberal Democrats: No change
Independents/Others: +20 seats

After last week’s dismal polling results, suffering three-figure losses of councillors would be another body blow for Welsh Labour. And making substantial gains would be another huge boost to the Welsh Conservatives. In practice, I would be surprised if the Tories picked up quite this many seats – their performance in local elections in Wales in recent years has been distinctly patchy. I would also be surprised if the Liberal Democrats make no net gains at all.

However, even before the local election takes place, I can tell you who some of the winners will be. This isn’t because I am psychic, but is because of the distressingly common phenomenon of uncontested seats.

Roughly eight percent of all the seats up for election in Wales have only one candidate. This is very similar to the proportion that were uncontested in 2012 – something which stood in stark contrast to Scotland, where there were no uncontested seats. The full list of uncontested seats that have thus already been won by each party is as follows:

Plaid Cymru: 28 (+6 on 2012)
Independents: 26 (-7)
Labour: 18 (+1)
Conservatives: 5 (+3)
Liberal Democrats: 1 (-1)

The pattern of who wins these uncontested seats is explained by where they are: heavily concentrated in rural mid and west Wales, with particularly large numbers in Gwynedd, Pembrokeshire and Powys. We see few uncontested seats in the more heavily populated parts of the country, by contrast: none at all in Cardiff, Newport, Swansea or Wrexham.
Why are there uncontested seats? The parties often have difficulty finding people who want to be candidates – a problem likely to be greater in more sparsely-populated parts of the country. To many people being a councillor looks like a thankless task, with little reward, that they are not willing to commit to.

More obvious is that we just have too many elected councillors in Wales. Wales has a population of a little more than three million compared to Scotland’s of just over five and a quarter million. Yet Wales has more elected councillors: 1,271 here to Scotland’s 1,219. That gives us roughly one elected council representative for every 2,360 of us, compared to around one for every 4,343 people in Scotland.

Scotland also has a strong advantage over Wales in that it uses the proportional STV system for its local elections. That means that every (multi-member) ward is competitive: everyone has something to fight for. In Wales, under a non-proportional system, we have lots of safe wards where parties are reluctant to fight if they are certain they will lose. There is a strong, and in my view compelling, argument for moving to STV for local elections in Wales.

Whoever ends up ‘winning’ an uncontested council seat, the losers are the public and democracy.


NOTE: The April Welsh Political Barometer poll was conducted by YouGov for ITV-Cymru Wales and Cardiff University’s Wales Governance Centre. It had a sample of 1029 Welsh adults, with interviews being conducted via the internet between 19-21 April 2017.


  • J.Jones

    When are YouGov going to release the complete data sets?

  • Geraint

    You raise the point of numbers of uncontested seats in the west. Powys one of the counties you named has 73 wards. To stand a candidate you need to find someone who lives in or very close to the ward. Another 10 individuals who live in the ward are required to sign the nomination papers. To put one name on the ballot paper requires 803 electors across the county. To have a contested election requires at least 1,606 electors provided they live in the right places. Considering only 73,831 individuals voted at the last UK general election and 53,967 voted at the last assembly election in the county a bar of at least 1,606 just to get a contested election seems to me to be very high.

    In urban multimember wards to have a contested election you need far fewer electors. A four member seat can have an election if there are only five candidates. So if there were four candidtes from one party standing and one from another, three would be effectively returned unopposed. I do not known if the same nominators can nominate different people on the same party list in a multimember ward but if this is the case this could be a contributing factor to this urban rural difference.

  • Geraint

    The University of Plymouth projections based on current opinion polling show Labour losses, Liberals standing still and gains for Tories, Plaid & Independents/ Others. Accordng to the BBC, the last elections in 2012 (Ynys Mon did not vote so it is not clear if the islands results were part of the 2008 baseline showing gains or not) showed Labour gained 231 seats, Plaid lost 41 seats, Conseratives lost 61 seats and Liberals lost 66 seats. So if the opinion poll results actually predict the real results correctly the Labour party will keep 101 of the seats gained, the Tories will gain 29 seats, Plaid will be down about 20 seats and the Liberals will have made no progress on their 2012 melt down.
    Considering the tone of news reports and the dire reporting around the Labour leadership surely this would be a reasonable result for Labour, show modest gains for the Tories, disappointing for Plaid and be an on going disaster for the Liberals. Labour would remain the dominant party on about 450 seats. The Tories and Plaid (results from Ynys Mon will probably give Plaid a slight lift) would be fighting it out for second place and the Liberals would show no movement from 2012.

  • J.Jones

    Here in Anglesey the expectation is that Plaid will be the biggest party. They have pretty well flooded all wards with candidates, three out of 5 where I live…I only cast two of my allotted 3 votes.

  • Eric Willis

    I don’t think people appreciate how disastrous the council elcetions were for Labour in Wales. They are down to 7 councils and outside South Wales they are virtually non existent. The Valleys “donkey”vote is starting to collapse under any sort of organised opposition, and Cardiff and Newport was only won as a result of the Muslim block vote (as in many English cities.) In the medium term this dependence on the Muslim block vote is not going to make Labour popular with the general elctorate. Of course the Lib Dem collapse helped in all 3 cities. Plaid did well outside South Wales. They effectively run 4 councils and have a record of good governance which will sustain them in the long term in these areas. In South Wales they rely on their success on a huge amount of local effort to combat the lack of Brirish media exposure. This works in the Rhonnda and Fairwater. However Plaid’s brand of “socialism with a Welsh accent” is not attracting the new generation of successful aspirational and committed Welsh people that Plaid needs to move forward. Plaid’s core workers in RCT (outside the Rhondda), and Caerphilly are now getting on a bit, and a very large number of Plaid activists in the 1970’s 1980’s and 1990’s have drifted away, and have not been replaced by their children’s generation.The anti Brexit stance of Plaid, did them no favours and lost them votes in areas like Creigiau Pentyrch and Dinas Powis. Plaid (the “Party of Wales”) really should listen to the views of the Welsh people, and concentrate on making Wales Brexit ready, and a dynamic place to do business. Policies like opposing the new stretch is the M4 makes them a laughing stock.
    Generally Plaid needs to reinvent itself as the party of good governance and the party of the aspirational, successful and committed Welsh people. “Labour with a Welsh accent” is a failed product. Without change Plaid will tread water, which is a shame.

  • J.Jones

    Interesting to look at the prediction against the actual; Labour losses overestimated by 23 seats and Plaid success underestimated by 13. Conservatives success overestimated by 10 and independents under by 7. The LibDem losses (-11) not predicted.

    I would suggest that country wide polling isn’t a great predictor at local government level.

    What will be interesting is whether Wales follows a national trend for young people to register to vote. If it does then we may see the Tory surge nullified to a degree…particularly in University towns.

  • Tomos

    One of the most interesting elements of the poll was the 8% for ukip predicted in the local elections (not that far behind the independents on 12%). Yet the independents won over 300 seats and ukip didn’t get close anywhere.

    I think there’s a major concern that the new weighting by brexit vote is oversampling ukip and conservative voters. The kind of former labour voters who voted brexit on the panel are probably quite different to the population they are being used to represent. By sampling this group heavily there is a serious risk that UKIP and the Tories seem to be doing much better.

    Will the next barometer be weighted by local election vote? I think this would be a lot more reliable than the method used for the local elections poll.

    Since these are the only Wales polls they risk setting the agenda. After the brexit referendum which the barometer got very wrong and now the local elections which overestimated UKIP (and I suspect the conservatives but haven’t seen national projections yet) surely it’s time to try some sensitivity analyses of your weightings?

  • Chris

    Eric Willis; do you have any idea of how big the Muslim community in Newport and Cardiff is? It’s about 4% which is nowhere near big enough to sustain Labour’s dominance of either council even if it might be telling in a handful of wards

  • J.Jones

    Adult Muslim population Cardiff 2011 5.5%. Newport 3.9%. If they all vote it still wouldn’t be a game changer in an election unless you can be sure that they all vote for one party. It seems to me that all main parties have at least some Muslim’s support.

  • Christian Schmidt

    So let’s compare the forecast to the results.

    Forecast was:
    Labour -130 seats
    Tories +90
    Plaid +20
    LDs +/-0
    Ind/Others +20

    Labour -107
    Tories +80
    Plaid +33
    LDs -11
    Ind/Others +5

    The swing between Labour and Tories nearly 20 seats less, Plaid better, LD and Ind worse.

    How does this fit with previous differences between polls and results, and what if anything does that tell us about the general election (other than local elections results are always more patchy…)?

  • Chris


    the figures I am hoping to see are the comparison between the Local Election opinion polls and the actual votes cast as that is much more “like for like”; there are too many local vagaries when it comes to actual seats; don’t know if Roger has done this?

  • S.

    When you have 5 candidates trying for 3 seats and 3 are from the same party its obvious they have worked to use a block voting system. in my eyes an individual has not much of a chance to get a seat this is where one vote shoud stand and the top 3 have their seats

Leave a Reply

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