Berwickshire, Roxburgh & Selkirk

2015 Result:
Conservative: 19817 (36%)
Labour: 2700 (4.9%)
Lib Dem: 10294 (18.7%)
SNP: 20145 (36.6%)
Green: 631 (1.1%)
UKIP: 1316 (2.4%)
Independent: 135 (0.2%)
MAJORITY: 328 (0.6%)

Category: Ultra-marginal SNP seat


Main population centres:



Current MP
CALUM KERR (SNP) Born Gala. Educated at Peebles High School. Former consultant. First elected as MP for Berwickshire, Roxburgh & Selkirk in 2015.
Past Results
Con: 16555 (34%)
Lab: 5003 (10%)
LDem: 22230 (45%)
SNP: 4497 (9%)
Oth: 729 (1%)
MAJ: 5675 (12%)
Con: 13092 (29%)
Lab: 7206 (16%)
LDem: 18993 (42%)
SNP: 3885 (9%)
Oth: 2212 (5%)
MAJ: 5901 (13%)
Con: 6533 (23%)
Lab: 4498 (16%)
LDem: 14044 (49%)
SNP: 2806 (10%)
Oth: 916 (3%)
MAJ: 7511 (26%)
Con: 8337 (24%)
Lab: 5226 (15%)
LDem: 16243 (47%)
SNP: 3959 (11%)
Oth: 1166 (3%)
MAJ: 7906 (23%)

2015 Candidates
JOHN LAMONT (Conservative) Born 1976. Educated at Kilwinning Academy and Glasgow University. Solicitor. Contested Berwickshire, Roxburgh and Selkirk 2005, 2010. Member of the Scottish Parliament for Ettrick, Roxburgh and Berwickshire.
KENRYCK LLOYD JONES (Labour) Educated at Huddersfield University. Public affairs and policy manager.
MICHAEL MOORE (Liberal Democrat) Born 1965, Dundonald. Educated at Jedburgh Grammar School and Edinburgh University. Chartered accountant. MP for Berwickshire, Roxburgh and Selkirk 1997 to 2015. Secretary of State for Scotland 2010-2013.
PETER NEILSON (UKIP) Contested Berwickshire, Roxburgh & Selkirk 2005, Scottish European election 2009.
PAULINE STEWART (Green) Medical herbalist.
CALUM KERR (SNP) Born Gala. Educated at Peebles High School. Consultant.
JESSE RAE (Independent)
Comments - 562 Responses on “Berwickshire, Roxburgh & Selkirk”
  1. It appears that the review might not be so damaging for the Conservatives as it appeared on face-value.

    In fact an Ayr South, Carrick & Galloway seat would actually benefit the Conservatives over the existing Dumfries & Galloway seat. It seems pretty likely that we’ll have a notional Conservative seat in the Borders which excludes Galashiels and a notional marginal constituency based in Dumfries by 2020.

  2. @NTYUK My comment was more about the map that Partick posted, which seems to suggest that this seat would remain, with no additions, and the removal of the Galashiels area. That isn’t a legal seat under the current rules, as I understand them.

    On your proposal, I think the problem is likely to be why you would exclude Gala from the seat. There’s not really a rational reason to do so that I can see, except to create a Tory-friendly seat. From my knowledge of the area, Galashiels should, if possible, be in the same seat as Hawick, Selkirk and Melrose. Probably Jedburgh too. They have stronger links with each other than with the Berwickshire parts of the seat.

    I’d guess that the most likely change is just to add more or all of the Tweeddale East ward to the seat, which won’t impact much on the political balance. Electoral Calculus suggests adding the whole ward would increase the SNP lead by about 500, but it would still remain hyper-marginal. Then again, even your proposal would only have a Tory majority of around 1000, so whatever they come up with will be very marginal. The only way that doesn’t happen is if they go back to something like the historical Berwickshire and East Lothian seat, which would be reasonably safe for the SNP due to split opposition. I doubt they’ll do that, because East Lothian can continue on current boundaries, and makes an eminently sensible seat.

  3. Max, the problem for the Tories in Perthshire is much more that they don’t get enough votes than that the boundaries are awful. Going on Electoral Calculus, they were only ahead in one ward in either of the two Perthshire seats in 2015. The only way the boundaries could make much of a difference is by splitting the town of Perth between two seats, but there’s no reason to do that, and plenty of legitimate reasons why it should all be in the one seat.

  4. @ Simon – I have already provided my reasoning, but I’ll do it again.

    Firstly electoral calculus is heavily inaccurate in Scotland through it’s failure to account for the referendum: it is effectively universal swing applied to local council elections for each respective constituency, which isn’t so relevant when on a national scale Labour’s best constituency had the largest swing in Scotland, and their most marginal turned out to be their only constituency in Scotland.

    I have a notional Conservative lead of 2% in the proposed Bordees constituency.

    The reasons in favour of grouping Midlothian with Galashiels (as I have already mentioned) and joining Berwickshire, Roxburghshire and Selkirkshire with Tweeddale are as follows:
    1) In the proposed review area of Midlothian and Scottish Borders a more proportionate electoral disparity between the two seats allocated can be drawn by joining Galashiels with Midlothian (giving the two seats in the area similar electorates).
    2) Galashiels has closer ties with Midlothian in comparison to Tweeddale, having formed part of the county of Midlothian from 1890-1975.
    3) The electorate in Berwickshire, Roxburgh and Selkirk is too small to form its own constituency without drawing upon a significant number of electors from Tweeddale West (I will double check the figures but I believe it would need to take in the whole ward). Dividing Tweeddale between two constituencies is less desirable than joining Galashiels to Midlothian as it would see the town of Peebles split into two separate constituencies.

  5. The SNP probably took all wards in Perth & Kinross in 2015, it’s unwinnable in the current political context in Scotland.

    As I have already mentioned the third ‘best’ potential consistency for the Conservatives would be a seat covering Ayr West, Ayr East, Maybole, North Carrick & Coylton, Girvan & South Carrick, Stranraer & Rhins, Wigtown West, Dee, Mid Galloway and Castle Douglas & Glenkins. They would be behind by 10% of the vote, with room to make significant gains in the South Ayrshire section of the constituency and moderate gains in the Dumfries and Galloway side of the seat.

    It’s still unwinnable with the referendum results in mind, but with a moderate Labour recovery it is by far the third best potential constituency for them in Scotland in the 2018 Review.

    You’ve really flattened out the Conservative vote in the Scottish Borders there, with a huge spike in Galashiels… I am very satisfied with my notionals there, which have the Conservatives well ahead in Kelso & District (20% lead), Mid Berwickshire (16% lead), Jedburgh & District (12% lead) and East Berwickshire (12% lead). Moderately ahead in Hawick & Hermitage (3%) and moderately behind in Hawick & Denholm (3%), Selkirkshire (3%) and Tweeddale West (4%). Behind in Tweeddale East (10%), Leaderdale & Melrose (25%) and Galashiels & District (37%).

  6. I will do, unless the Commission manage to come up with it by themselves 🙂

    In a sense I’m rather disappointed in the fact that Ayr will have to be split into two constituencies yet again, although the ‘Ayr’ seat could be created (joining Ayr, Prestwick and Troon with the adjacent rural area) by joining Galloway with Carrick and Cumnock this would contradict the Commission’s guidelines for minimalising the number of council areas which a seat can cover (2). Then again, the unionist in me awaits ‘Ayr South, Carrick and Galloway’.

  7. Galashiels clearly has much closer links to the other nearby “rugby towns” than to Midlothian. Peebles is more linked to Edinburgh and the surrounding area than anywhere else in the Borders, because it’s more of a commuter town these days. I suppose it depends how much the Commission weights things like this against historical boundaries and how undesirable they find splitting towns.

    The margins of error on any guesstimates by ward must be pretty huge, given that there was no breakdown of referendum results in this part of the world, and local elections have a number of independents and the Borders Party. For example, in the two Hawick wards, independents took half the vote in 2012. You’d need very good local knowledge to unpick that with any accuracy.

    The end result probably depends where the Boundary Commission start. They could start in the Southwest and create a seat that starts in Stranraer and goes as far through D&G as is necessary. That could leave the remains of D&G with most of the Borders, except Berwickshire, and then they’d have to go with Berwickshire going in with a chunk of East Lothian. If they start in the Southeast, and work west, then they get something like we’ve been discussing, and they probably have to then take the Galloway seat into Ayrshire (or the DCT-style seat further into South Lanarkshire).

  8. I assume the population was too small to meet the quota.

  9. @ Simon – The Commission do not operate in that way. They begin their review by chopping Scotland into separate ‘review areas’ following council areas. It’s most likely that they’ll group Dumfries & Galloway with Ayrshire and Midlothian with Scottish Borders. Creating a review area across South Ayrshire, East Ayrshire, Dumfries & Galloway, Scottish Borders, Midlothian, East Lothian and Edinburgh is illogical when better alternatives exist. A similar structure of Review Areas submerged as part of the Zombie Review.

    I strongly disagree with your idea that rugby leagues represent communal ties on any level??? By this logic Berwick-upon-Tweed should be part of the Scottish Borders. Peebles is part of the Scottish Borders League anyway? I really don’t understand your logic here, however town splitting (and ward splitting) is not a desirable characteristic of parliamentary boundaries, which the Commission tend to avoid at all costs. Especially when a simple, logical alternative exists.

    Again, clear, descernable ties exist between Midlothian and Galashiels which just isn’t the case for Midlothian and Tweeddale. Typically the Commission do not even consider said factors, operating exclusively within the confines of local area council/council boundaries in Scotland (examples: forming Ayr, Carrick & Cumnock rather than seperate Ayr and Carrick, Cumnock & Loudoun constituencies in 2005 despite clearly established communal differences/Splitting Hamilton between Rutherglen and Lanark despite the fact that the boundaries could be re-arranged (giving Bothwell to Rutherglen) to keep the town in a single constituency).

    Although there are many exceptions to the rule that towns should not be split between constituencies (primarily a consequence of the Commission’s own incompetitance) it is generally frowned upon – ward splitting even more so.

  10. I think the only tie that you’ve suggested between Gala and Midlothian is that they were once in the same county a long time ago (although I thought Galashiels was actually in Selkirkshire). In my view that’s not much of a link at all. I think that the people of Gala would see themselves as having much more in common with the people of Hawick, Selkirk or Melrose than they would with the people of Dalkeith, for example. My assessment, from actually knowing most of these places is that if you are adding one ward to Midlothian, Tweeddale West is the best candidate by some margin.

  11. @ Simon – well we can agree to disagree, although I can say with quite some confidence that the people of the town of Peebles have more alike than Galashiels does with Selkirkshire, Roxburghshire and Berwickshire.

  12. * And by that I mean Peebles should remain in one constituency to ensure that the communal ties of the town are not unrepresented by the constituency.

  13. Paddy Power predicts that the SNP will gain every constituency except Ettric (although odds for Orkney and Shetland).

    Labour in third place in Glasgow Kelvin, Eastwood and Dumfriesshire.

    Paddy Power predicts larger SNP leads over second placed Tories in Eastwood and Ayr but a close result in Galloway & West Dumfries.

    SNP 1/7, Con 10/3, Lab 80/1 and LD 200/1

    SNP 1/6, Lab 3/1, Con 100/1 and LD 200/1

    SNP 1/10, Lab 5/1, Con 66/1 and LD 175/1

    SNP 4/5, Con 9/4, Lab 5/2 and LD 100/1

    SNP 1/14, Lab 11/2, Con 50/1 and LD 100/1

    SNP 5/6, Con 15/8, Lab 10/3 and LD 100/1

    Con 1/4, SNP 5/2, LD 100/1 and Lab 100/1

    SNP 5/6, Con 7/4, Lab 40/1 and SNP 200/1

    Glasgow Kelvin
    SNP 1/25, Green 10/1, Lab 14/1, Con 80/1 and 150/1

    Glasgow Maryhill
    SNP 1/20, Lab 7/1, Con 100/1 and LD 200/1

    Glasgow Pollok
    SNP 1/20, Lab 8/1, Con 100/1 and LD 200/1

    Glasgow Provan
    SNP 2/5, Lab 7/4, Con 50/1 and LD 100/1

  14. That’s some terrible odds for the Tories in Ayr, Galloway and Eastwood.

  15. (And equally dreadful in Dumfriesshire)

  16. The Conservatives need to win this seat, or whatever successor seat includes most of its area, if they are to make any sort of comeback in Scotland. What are the current betting odds in this seat?

    Thre is a substantial LibDem vote which the Conservatives might hope to dig into now that the LibDems have gone from first to third. Much may depend on whether the LibDems can hold onto their support, and if not where it goes to.

  17. @ Frederic – The Scottish Parliament follows different boundaries than the UK Parliament. The equivalent seat in the Scottish Parliament is the Ettrick, Roxburgh and Berwickshire constituency, probably the only constituency which could be seen ‘safe’ for the Conservatives. It excludes areas where the SNP did better in such as Galashiels and Melrose, and is a pretty strong notional Conservative seat.

  18. Yes….see the Conservative Ettrick odds above…the Holyrood version of this constituency…Con 1/4…

  19. I have been looking at the difference between the “Berwickshire, Roxburgh & Selkirk” Westminster constituency and the “Ettrick, Lauderdale & Berwickshire” Holyrood constituency.

    While they look very similar, the Westminster constituency also includes “Galashiels & District Ward” which is currently estimated to be SNP 4200, Con 2000, Lib Dem 1300 and Lab 600.

    This would give the Conservatives a lead of just under 2000 in “Ettrick, Lauderdale & Berwickshire” last May.

    “Ettrick, Lauderdale & Berwickshire” is clearly the hardest SNP Gain but not impossible.

    If the SNP poll well over 50% across Scotland, its possible that “Ettrick, Lauderdale & Berwickshire” could also fall, particularly if the Labour vote collapses, regardless of Lamont’s personal vote.

    Just look what happened to David McLetchie in 2011.

  20. My notional for Ettrick, Roxburgh & Berwickshire is:

    Conservative – 42%
    Scottish National – 33%
    Liberal Democrat – 18%
    Labour – 4%
    Others – 3%

    The Berwickshire, Roxburgh & Selkirk seat covers Galashiels and Lauderdale, the two worst wards for the Conservatives in the Scottish Borders.

  21. *Those wards are not covered by the Ettrick constituency. A 2,000 Tory majority here is far too slender.

  22. No.

  23. 20% Conservative? There’s no precedent in the polls.

  24. That probably wouldn’t give the Tories over 50% of the vote here, 16-17% seems more realistic.

  25. I don’t think you ever learn 😛 The Tories polling average from 1 April until just now is 17%: usually the polls tend to be more accurate for the Conservatives over most other parties in Scotland, 18% is towards the upper end of the polls, 16% the lower end.

    Last year the upper figure for the Tories was around 17%. Look how that turned out.

    I believe they could take around 16 to 17% of the vote on the constituency ballot.

  26. A silver-lining for Labour is that they will “defy the odds” by beating the Tories into second place in Scotland.

  27. Today’s polls make even that quite questionable; Labour have a slender lead in the constituency poll, the Tories a one point lead in the regional list polls (all c. 18-20%). On those sorts of numbers one would expect them to finish with the same number of MSPs.

  28. Depends, as ever, on the accuracy of the polls. From memory, Holyrood polling has historically overstated the Tories, and they have always struggled to garner much support beyond their core vote of 17% or so.

    Of course, things may be different this time. Then again, they may not.

  29. The best result for the Conservatives was in 2003 where they took 16.6% of the constituency vote and 15.5% of the list vote.

  30. Indeed, which suggests that the size of the Tory core vote is even smaller than I’ve stated. There is a long-standing trend in Scottish politics, which is that Tory voters are moving away from the party and towards the crematorium. There are no real indications that anything has happened to reverse that trend.

  31. I suspect they have picked up some votes, more due to the various other parties’ situation rather than anything they’ve done themselves. If you’re strongly unionist with broadly centrist views and don’t live in a seat where the Lib Dems are relevant, then Ruth Davidson is probably more appealing that Jeremy Corbyn or Nicola Sturgeon. There aren’t really a whole lot of other options.

  32. The Scottish Tories’ problem has long been more tribal than ideological or socio-economic, in that they do significantly less well than you’d predict from issues polling or the proportion of ABC1 voters in comparison to England.

    They haven’t done much to solve their tribal problems, but I do think that the referendum and Ruth Davidson have done some things to change that.

    It’s worth recalling that the Scottish Tory vote went up in 2015 at Westminster – their vote share fall was entirely the product of an increased turnout.

    Simon’s analysis seems plausible to me, but I would add that the Scottish Tories also struggle by association with the Westminster Tories. If you look at Gary, Tank Commander’s interviews on Youtube/the BBC with the various Scottish party leaders, Nicola Sturgeon and Ruth Davidson are the only ones who come across as effortlessly likeable and at ease.

  33. I think Willie Rennie pulled off some impressive feats during his interview Gary Tank Commander/leadership debates, although that could be down to the fact that he is a relatively unknown figure within Scotland’s mainstream political scene.

    I would be shocked if the Conservatives make any real advance beyond their high-water mark of 18 MSP’s.

  34. CON Hold (Ettrick, Roxburgh & Berwickshire).

  35. 15-20 is a more reasonable range over 20-24, which just seems to be sustained wishful thinking on your part for some reason.

  36. Max has never knowingly made a prediction that’s bad for the Tories.

  37. Simons definitely got a point here.

  38. If you applied that same logic onto Scotland last year the Tories would have took 18% of the vote: in the event they took 15%.

  39. The Conservatives seem to have taken more numerical votes in Berwickshire, Roxburgh and Selkirk over 2015: the Lib Dem vote collapsing comprehensively.

    Perhaps this recognition that this is a two-way fight between the Conservatives and the SNP.

  40. After this result I make the Tories favourites for the Westminster seat in 2020. It seems like the Lib Dem vote is being squeezed in their favour.

    That’s not considering boundary changes, though.

  41. I still doubt the Tories could win WA&K unless the landscape changes markedly again. Ruth Davidson is a far more popular (or less unpopular) leader in Scotland than Cameron is, and the same will be true whoever is PM going into the 2020 GE, I would have thought. There are plenty of Scots voted Tory in Holyrood elections because they are unionists, but would still not want a Conservative government at Westminster.

  42. Poll Troll- I share your scepticism to some extent. I certainly don’t see the Conservatives managing more than three Scottish seats in 2020. However, I’d now be fairly surprised if they won one or none.

  43. On current boundaries Berwickshire, Roxburgh and Selkirk would be a very likely to Tory gain in 2020… the ex-LD vote there has collapsed drastically, mainly to the Tories advantage, since 2010. You’ve got to think Dumfries and Galloway, and West Aberdeenshire and Kincardine would also be competitive between the SNP and the Tories. But there will be boundary changes and Polltroll is right that it is very unlikely the Tories will perform quite as strongly at a GE where the Scottish constitutional question isn’t likely to be as high on the agenda.

  44. More importantly the low turnout of last Thursday’s election pulls the Conservatives out of competition in Dumfries & Galloway, East Renfrewhire and West Aberdeenshire & Kincardine.

  45. In saying that we should have a lower turnout in 2020, which would benefit the Conservatives assuming that they can sustain some momentum.

  46. Last year I believe he predicted some 5 constituencies would go Conservative: equally miles off.

    The pollsters were well off this time!

  47. NTY UK,

    So a good record if you aggregate the two elections, but a bad one if you take them individually?

  48. I believe so 😛

  49. Other than NE Fife and Edinburgh West the Lib Dems collapsed on mainland Scotland.

    Their vote has emploded in the 2 seats in the Scottish Borders, collapsed in Aberdeenshire West, Dunfermline, Edinburgh Southern, Edinburgh Central and all three Scottish Highland seats.

    The only seat that the made any kind of recovery was Argyll & Bute.

  50. On the basis of Thursday’s results I’d watch out for an LD resurgence v Salmond in Gordon come the GE. The Tories surged in the Holyrood seat but if there is a chance for the LDs to take it many of their voters may go LD, as they seem to have in North East Fife and Edinburgh Western.

Leave a Reply

NB: Before commenting please make sure you are familiar with the Comments Policy. UKPollingReport is a site for non-partisan discussion of polls.

You are not currently logged into UKPollingReport. Registration is not compulsory, but is strongly encouraged. Either login here, or register here (commenters who have previously registered on the Constituency Guide section of the site *should* be able to use their existing login)