European Union Referendum 2016

Transcription

1 BRIEFING PAPER Number CBP 7639, 29 June 2016 European Union Referendum 2016 By Elise Uberoi Contents: 1. The referendum on the UK s membership of the EU 2. Results 3. Characteristics associated with votes for Leave and Remain 4. Turnout 5. Opinion polls 6. Timeline of key events 7. Previous referendums 8. Data tables intranet.parliament.uk/commons-library

2 2 European Union Referendum 2016 Contents Summary 3 1. The referendum on the UK s membership of the EU 4 2. Results Analysis of results by counting region (including Northern Ireland) Analysis of results by counting area 10 In the East 11 In the East Midlands 12 In London 12 In the North East 13 In the North West 13 In Scotland 14 In the South East 14 In the South West and Gibraltar 15 In Wales 15 In the West Midlands 16 In Yorkshire and the Humber 16 In constituencies in Northern Ireland Spoilt ballot papers Characteristics associated with votes for Leave and Remain Voting behaviour Highest qualification and age Socio-economic indicators and immigration Rural and urban Turnout Relationships between turnout and age, and Leave votes? The weather Opinion polls Timeline of key events Previous referendums Data tables England Wales Scotland Northern Ireland 39 Contributing authors Paul Bolton (the weather) Matt Keep (maps) Richard Keen and Feargal McGuiness (data) Cover page image copyright: Verification underway, by Epping Forest District Council. Licensed by CC 2.0 / image cropped.

3 3 Commons Library Briefing, 29 June 2016 Summary In a UK wide referendum on 23 June 2016, voters were asked if they wanted the UK to remain a member of the European Union (EU) or leave the European Union. There were 17.4 million votes for Leave (51.9%). There were 16.1 million votes for Remain (48.1%). Votes for Leave and Remain were not evenly distributed across the UK. Leave won the highest share of the vote in England and Wales, while Remain won the highest share in Scotland and Northern Ireland. Out of the 12 counting regions (including Northern Ireland), 1 nine recorded more than 50% of the vote for Leave, and three recorded more than 50% of the vote for Remain. The highest vote share for Leave (76%) was recorded in Boston (Lincolnshire). The highest vote share for Remain (96%) was recorded in Gibraltar. The proportion of people in a local authority who voted Leave was strongly related to the proportion that voted for UKIP in the 2014 European Parliament election, and to the proportion of nongraduates. Relationships between Leave votes and other selected socio-economic indicators were weaker at local authority level. In England, urban areas were more likely to record higher vote shares for Remain than rural areas. High vote shares for Leave, however, were found in both urban and rural areas. Turnout was 72.2% across the UK. This was higher than at the 2015 General Election (66.2%) but lower than turnout at the 2014 Scottish independence referendum (84.6%). The highest turnout was recorded in Gibraltar (83.5%) and the lowest in Lagan Valley (Northern Ireland) (48.9%). The highest turnout (83.5%) in the UK was recorded in Chiltern (Buckinghamshire). 1 Northern Ireland was a counting area rather than a counting region but due to its size, and the fact that data has been released by constituencies within Northern Ireland, it is included in the discussion of regions.

4 4 European Union Referendum The referendum on the UK s membership of the EU On Thursday 23 June 2016, a referendum was held across the UK on the question: Should the United Kingdom remain a member of the European Union or leave the European Union? 51.9% of votes were to leave the EU. The referendum was first announced by Prime Minister David Cameron on 23 January He said that if a Conservative Government was elected to power following the General Election of 2015, it would hold a national referendum on European Union (EU) membership, framed on an in/out question, during the next Parliament. Shortly after the election, the new Conservative Government introduced the European Union Referendum Bill , which provided for the referendum to be held before the end of The date for the referendum was set in regulations introduced in February The franchise for the referendum was defined by the European Union Referendum Act It was based on the franchise used for UK Parliament elections, but additionally entitled Commonwealth citizens resident in Gibraltar and peers to vote. Votes were counted in counting areas, which in most cases were local authorities with electoral registration responsibilities (single tier authorities and lower tier authorities in two tier areas). There were 382 counting areas: 380 in Great Britain, one in Northern Ireland and one in Gibraltar. Votes were counted in eight locations in Northern Ireland but collated and declared as one result. Local results were collated for 11 counting regions, which were the same regions used for the European Parliament elections. The Chief Counting Officer, Jenny Watson, who is the Chair of the Electoral Commission, declared the national result from the national referendum count event held in Manchester on Friday 24 May at 07:51. The Parliamentary franchise includes British citizens, Irish citizens and qualifying Commonwealth citizens. British citizens living abroad can vote for up to 15 years after they were last registered in the UK. Box 1: Results data The results data used in this paper are as published by the Electoral Commission. Results by counting area and region are available to download from the Electoral Commission s website, Other data used for analysis in this Briefing Paper are published (online) in the accompanying dataset.

5 5 Commons Library Briefing, 29 June Results A total of 33,577,342 people voted in the referendum, a turnout of 72.2%. The Leave vote won by a 3.8 percentage point margin. National result % of valid Votes cast votes Leave 17,410, % Remain 16,141, % Winning margin for 'Leave' 1,269, % Total valid votes 33,551,983 Rejected papers 25,359 Total votes cast 33,577,342 The figures below show the total number of votes cast, and the national share of the vote, for Leave and Remain. Votes for Leave and Remain were not evenly distributed across the UK. Leave won the highest share of the vote in England and Wales, while Remain won the highest share of the vote in Scotland and Northern Ireland. Within England, London was the only region where Remain won a majority. Proportion of the vote across the UK Leave (%) Remain (%) England 53.4% 46.6% Wales 52.5% 47.5% Scotland 38.0% 62.0% Northern Ireland 44.2% 55.8% The highest vote share for Leave (76%) was recorded in Boston (Lincolnshire). The highest vote share for Remain (96%) was recorded in Gibraltar.

6 6 European Union Referendum 2016

7 7 Commons Library Briefing, 29 June 2016

8 8 European Union Referendum 2016

9 9 Commons Library Briefing, 29 June Analysis of results by counting region (including Northern Ireland) Leave won more than 50% of the vote in all counting regions except for three: London, Scotland and Northern Ireland. The West Midlands saw the highest share of the vote for Leave (59.3%), and Scotland the lowest (38.0%). Turnout was highest in the South East, and the South West and Gibraltar (76.7%), and lowest in Northern Ireland (62.7%). Results by region, ranked by highest vote share for Leave Counting region Remain Leave Rejected votes Total votes Remain (% cast (incl. of valid rejected) votes) Leave (% of valid votes) Turnout (valid votes as % of electorate) Turnup (total votes cast as % of electorate) Electorate West Midlands 1,207,175 1,755,687 2,507 2,965, % 59.3% 72.0% 72.0% 4,116,572 East Midlands 1,033,036 1,475,479 1,981 2,510, % 58.8% 74.1% 74.2% 3,384,299 North East 562, , ,341, % 58.0% 69.3% 69.3% 1,934,341 Yorkshire and the Humber 1,158,298 1,580,937 1,937 2,741, % 57.7% 70.6% 70.7% 3,877,780 East of England 1,448,616 1,880,367 2,329 3,331, % 56.5% 75.7% 75.7% 4,398,796 North West 1,699,020 1,966,925 2,682 3,668, % 53.7% 69.9% 70.0% 5,241,568 South West and Gibraltar 1,503,019 1,669,711 2,179 3,174, % 52.6% 76.7% 76.7% 4,138,134 Wales 772, ,572 1,135 1,628, % 52.5% 71.7% 71.7% 2,270,272 South East 2,391,718 2,567,965 3,427 4,963, % 51.8% 76.7% 76.8% 6,465,404 Northern Ireland 440, , , % 44.2% 62.7% 62.7% 1,260,955 London 2,263,519 1,513,232 4,453 3,781, % 40.1% 69.6% 69.7% 5,424,768 Scotland 1,661,191 1,018,322 1,666 2,681, % 38.0% 67.2% 67.2% 3,987,112 United Kingdom 16,141,241 17,410,742 25,359 33,577, % 51.9% 72.2% 72.2% 46,500,001 The figure below shows the net share of the vote for Leave in each counting region. The net share of the vote for Leave was highest in the West Midlands and lowest in Scotland. Net Leave vote by region (%) (Leave % - Remain %) West Midlands UK average 3.78% East Midlands North East Yorkshire and the Humber East of England North West South West and Gibraltar Wales South East Northern Ireland London Scotland -25% -20% -15% -10% -5% 0% 5% 10% 15% 20%

10 10 European Union Referendum 2016 As the number of voters varies by region, not all regions contributed equally to the overall national result: regions where more votes were counted had a greater impact on the final outcome. The figure below shows the weight of the contribution each region made to the final result. The South East counted the highest number of votes overall, and also the highest number of both Leave and Remain votes. Regional results as share of the national vote South East North West East of England West Midlands South West and Gibraltar Yorkshire and the Humber London East Midlands Scotland Wales North East Leave Remain Northern Ireland 0% 3% 6% 9% 12% 15% Note: in this chart, 100% equals the sum of Leave plus Remain votes across all regions 2.2 Analysis of results by counting area Out of the 382 voting areas, 258 recorded a vote share for Leave that was higher than 50%. Leave won 60% or more of the vote in 105 of these counting areas, and 70% or more in 14 of them. Remain won more than 50% of the vote in 115 counting areas. 46 of these recorded a Remain vote share of 60% or more, and 16 counting areas (10 of which in London) recorded a Remain vote share of 70% or more. The table below shows the highest level of support for Leave and Remain across all counting areas (including constituencies in Northern Ireland). Highest share of the vote for Leave and Remain Top 10 counting areas voting Leave (%) Top 10 counting areas voting Remain (%) 1 Boston 75.6% 1 Gibraltar 95.9% 2 South Holland 73.6% 2 Lambeth 78.6% 3 Castle Point 72.7% 3 Hackney 78.5% 4 Thurrock 72.3% 4 Foyle 78.0% 5 Great Yarmouth 71.5% 5 Haringey 75.6% 6 Fenland 71.4% 6 City of London 75.3% 7 Mansfield 70.9% 7 Islington 75.2% 8 Bolsover 70.8% 8 Wandsworth 75.0% 9 East Lindsey 70.7% 9 Camden 74.9% 10 NE Lincolnshire 69.9% 10 Edinburgh 74.4%

11 11 Commons Library Briefing, 29 June 2016 The vote was close (with either Leave or Remain winning by a margin of between 0 and 5 percentage points) in 75 counting areas. The most frequent margin was a 5-10 percentage point lead for Leave, which was recorded in 48 counting areas. Net Leave vote (% points), number of counting areas 50 to 55% 45 to 50% 40 to 45% 35 to 40% 30 to 35% 25 to 30% 20 to 25% 15 to 20% 10 to 15% 5 to 10% 0 to 5% 0% -5 to 0% -10 to -5% -15 to -10% -20 to -15% -25 to -20% -30 to -25% -35 to -30% -40 to -35% -45 to -40% -50 to -45% -55 to -50% Number of counting areas In the East In the East, Castle Point recorded the highest vote share for Leave, and Cambridge for Remain. Out of the 46 counting areas in the East, 5 recorded a vote share higher than 50% for Remain. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Castle Point 72.7% 1 Cambridge 73.8% 2 Thurrock 72.3% 2 St Albans 62.7% 3 Great Yarmouth 71.5% 3 South Cambridgeshire 60.2% 4 Fenland 71.4% 4 Norwich 56.2% 5 Tendring 69.5% 5 North Hertfordshire 54.4% 6 Basildon 68.6% 6 Watford 49.7% 7 Harlow 68.1% 7 East Hertfordshire 49.6% 8 Rochford 66.6% 8 Dacorum 49.3% 9 King's Lynn and West Norfolk 66.4% 9 Uttlesford 49.3% 10 Broxbourne 66.3% 10 Hertsmere 49.2% Turnout was highest in St. Albans (82%), and lowest in Luton (66%).

12 12 European Union Referendum 2016 In the East Midlands In the East Midlands, Boston recorded the highest vote share for Leave, and Rushcliffe for Remain. Leave won more than 50% of the vote in 38 out of the 40 counting areas in the East Midlands. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Boston 75.6% 1 Rushcliffe 57.6% 2 South Holland 73.6% 2 Leicester 51.1% 3 Mansfield 70.9% 3 High Peak 49.5% 4 Bolsover 70.8% 4 Rutland 49.4% 5 East Lindsey 70.7% 5 Harborough 49.3% 6 Ashfield 69.8% 6 Nottingham 49.2% 7 Bassetlaw 67.8% 7 Derbyshire Dales 48.4% 8 Corby 64.2% 8 Charnwood 46.2% 9 North East Derbyshire 62.8% 9 South Northamptonshire 45.7% 10 Wellingborough 62.4% 10 Oadby and Wigston 45.4% Turnout was highest in Derbyshire Dales (82%), and lowest in Nottingham (62%). In London In London, Havering recorded the highest vote share for Leave, and Lambeth for Remain. Leave won more than 50% of the vote in 5 out of the 33 counting areas in London. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Havering 69.7% 1 Lambeth 78.6% 2 Bexley 63.0% 2 Hackney 78.5% 3 Barking and Dagenham 62.4% 3 Haringey 75.6% 4 Hillingdon 56.4% 4 City of London 75.3% 5 Sutton 53.7% 5 Islington 75.2% 6 Bromley 49.4% 6 Wandsworth 75.0% 7 Hounslow 48.9% 7 Camden 74.9% 8 Newham 47.2% 8 Southwark 72.8% 9 Redbridge 46.0% 9 Hammersmith and Fulham 70.0% 10 Croydon 45.7% 10 Lewisham 69.9% Turnout was highest in Richmond upon Thames (82%), and lowest in Newham (59%).

13 13 Commons Library Briefing, 29 June 2016 In the North East In the North East, Hartlepool recorded the highest vote share for Leave, and Newcastle upon Tyne for Remain. Remain won more than 50% of the vote in 1 out of the 12 counting areas in the North East. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Hartlepool 69.6% 1 Newcastle upon Tyne 50.7% 2 Redcar and Cleveland 66.2% 2 North Tyneside 46.6% 3 Middlesbrough 65.5% 3 Northumberland 45.9% 4 South Tyneside 62.0% 4 Darlington 43.8% 5 Stockton-on-Tees 61.7% 5 Gateshead 43.2% 6 Sunderland 61.3% 6 Durham 42.5% 7 Durham 57.5% 7 Sunderland 38.7% 8 Gateshead 56.8% 8 Stockton-on-Tees 38.3% 9 Darlington 56.2% 9 South Tyneside 38.0% 10 Northumberland 54.1% 10 Middlesbrough 34.5% Turnout was highest in Northumberland (74%), and lowest in Sunderland (65%). In the North West In the North West, Blackpool recorded the highest vote share fore Leave, and Manchester for Remain. Remain won a vote share over 50% in 7 out of the 39 counting areas in the North West. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Blackpool 67.5% 1 Manchester 60.4% 2 Burnley 66.6% 2 Liverpool 58.2% 3 Hyndburn 66.2% 3 Trafford 57.7% 4 Wigan 63.9% 4 South Lakeland 52.9% 5 Wyre 63.8% 5 Stockport 52.3% 6 Pendle 63.2% 6 Sefton 51.9% 7 Copeland 62.0% 7 Wirral 51.7% 8 Tameside 61.1% 8 Cheshire West and Chester 49.3% 9 Oldham 60.9% 9 Lancaster 48.9% 10 Rossendale 60.7% 10 Cheshire East 48.8% Turnout was highest in South Lakeland (80%), and lowest in Manchester (60%).

14 14 European Union Referendum 2016 In Scotland In Scotland, Moray recorded the highest vote share for Leave, and Edinburgh for Remain. Remain won the most votes in all counting areas in Scotland. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Moray 49.9% 1 Edinburgh 74.4% 2 Dumfries and Galloway 46.9% 2 East Renfrewshire 74.3% 3 Aberdeenshire 45.0% 3 East Dunbartonshire 71.4% 4 Eilean Siar 44.8% 4 Stirling 67.7% 5 Angus 44.7% 5 Glasgow City 66.6% 6 Highland 44.0% 6 Renfrewshire 64.8% 7 Shetland Islands 43.5% 7 East Lothian 64.6% 8 Falkirk 43.2% 8 Inverclyde 63.8% 9 North Ayrshire 43.1% 9 Orkney Islands 63.2% 10 Clackmannanshire 42.2% 10 South Lanarkshire 63.1% Turnout was highest in East Renfrewshire (76%), and lowest in Glasgow City (56%). In the South East In the South East, Gravesham recorded the highest vote share for Leave, and Oxford for Remain. Remain won more than 50% of the vote in 24 of the 68 counting areas in the South East. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Gravesham 65.4% 1 Oxford 70.3% 2 Dartford 64.2% 2 Brighton and Hove 68.6% 3 Medway 64.1% 3 Elmbridge 59.5% 4 Gosport 63.9% 4 Winchester 58.9% 5 Thanet 63.8% 5 Waverley 58.4% 6 Arun 62.5% 6 Reading 58.0% 7 Swale 62.5% 7 Vale of White Horse 56.7% 8 Havant 62.4% 8 Wokingham 56.7% 9 Shepway 62.2% 9 Guildford 56.2% 10 Dover 62.2% 10 Woking 56.2% Turnout was highest in Chiltern (83%), and lowest in Slough (62%).

15 15 Commons Library Briefing, 29 June 2016 In the South West and Gibraltar In the South West and Gibraltar, Torbay recorded the highest vote share for Leave, and Gibraltar for Remain, followed by Bristol. Remain won more than 50% of the vote in 10 out of the 38 counting areas in the South West and Gibraltar. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Torbay 63.2% 1 Gibraltar 95.9% 2 Sedgemoor 61.2% 2 Bristol 61.7% 3 Weymouth and Portland 61.0% 3 Bath and North East Somerset 57.9% 4 Torridge 60.8% 4 Isles of Scilly 56.4% 5 West Somerset 60.6% 5 Cheltenham 56.2% 6 Plymouth 59.9% 6 Exeter 55.3% 7 Purbeck 59.1% 7 Stroud 54.6% 8 Christchurch 58.8% 8 South Hams 52.9% 9 Forest of Dean 58.6% 9 Cotswold 51.1% 10 Gloucester 58.5% 10 Mendip 51.1% Turnout was highest in Gibraltar (84%), and lowest in Bournemouth (69%). In Wales In Wales, Blaenau Gwent recorded the highest vote share for Leave, and Cardiff recorded the highest vote share for Remain. Leave won a vote share of over 50% in 17 out of the 22 counting areas in Wales. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Blaenau Gwent 62.0% 1 Cardiff 60.0% 2 Torfaen 59.8% 2 Gwynedd 58.1% 3 Wrexham 59.0% 3 Ceredigion 54.6% 4 Caerphilly 57.6% 4 Vale of Glamorgan 50.7% 5 Pembrokeshire 57.1% 5 Monmouthshire 50.4% 6 Neath Port Talbot 56.8% 6 Isle of Anglesey 49.1% 7 Merthyr Tydfil 56.4% 7 Swansea 48.5% 8 Flintshire 56.4% 8 Rhondda Cynon Taf 46.3% 9 Newport 56.0% 9 Powys 46.3% 10 Bridgend 54.6% 10 Carmarthenshire 46.3% Turnout was highest in Monmouthshire (78%), and lowest in Merthyr Tydfil (67%).

16 16 European Union Referendum 2016 In the West Midlands In the West Midlands, Stoke-on-Trent recorded the highest vote share for Leave, and Warwick for Remain. Leave won more than 50% of the vote in 29 out of the 30 counting areas in the West Midlands. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 Stoke-on-Trent 69.4% 1 Warwick 58.8% 2 Cannock Chase 68.9% 2 Birmingham 49.6% 3 Walsall 67.9% 3 Stratford-on-Avon 48.4% 4 Dudley 67.6% 4 Malvern Hills 47.8% 5 Tamworth 67.5% 5 Worcester 46.3% 6 North Warwickshire 66.9% 6 Bromsgrove 44.6% 7 Sandwell 66.7% 7 Coventry 44.4% 8 Nuneaton and Bedworth 66.0% 8 Stafford 44.0% 9 South Staffordshire 64.8% 9 Solihull 43.8% 10 Staffordshire Moorlands 64.7% 10 Rugby 43.3% Turnout was highest in Wychavon (81%), and lowest in Birmingham (64%). In Yorkshire and the Humber In Yorkshire and the Humber, North East Lincolnshire recorded the highest vote share for Leave, and York for Remain. Leave won more than 50% of the vote in 18 out of the 21 counting areas in Yorkshire and the Humber. Highest vote shares for Remain and Leave (counting areas) Leave Remain 1 North East Lincolnshire 69.9% 1 York 58.0% 2 Doncaster 69.0% 2 Harrogate 51.0% 3 Barnsley 68.3% 3 Leeds 50.3% 4 Rotherham 67.9% 4 Sheffield 49.0% 5 Kingston upon Hull 67.6% 5 Craven 47.2% 6 Wakefield 66.4% 6 Hambleton 46.3% 7 North Lincolnshire 66.3% 7 Bradford 45.8% 8 Scarborough 62.0% 8 Kirklees 45.3% 9 East Riding of Yorkshire 60.4% 9 Ryedale 44.7% 10 Selby 59.2% 10 Calderdale 44.3% Turnout was highest in Craven (81%), and lowest in Kingston upon Hull (63%).

17 17 Commons Library Briefing, 29 June 2016 In constituencies in Northern Ireland Northern Ireland was one counting area in the referendum, but results have been published at constituency level by the Electoral Office for Northern Ireland. North Antrim recorded the highest vote share for Leave, and Foyle for Remain. Leave won more than 50% of the vote in 7 out of the 18 constituency areas. Highest vote shares for Remain and Leave (constituencies) Leave Remain 1 North Antrim 62.2% 1 Foyle 78.3% 2 Strangford 55.5% 2 Belfast West 74.1% 3 East Antrim 55.2% 3 Belfast South 69.5% 4 Lagan Valley 53.1% 4 South Down 67.2% 5 Upper Bann 52.6% 5 West Tyrone 66.8% 6 Belfast East 51.4% 6 Newry & Armagh 63.1% 7 South Antrim 50.6% 7 Mid Ulster 60.4% 8 Belfast North 49.6% 8 Fermanagh & South Tyrone 58.6% 9 East Londonderry 48.0% 9 North Down 52.4% 10 North Down 47.6% 10 East Londonderry 52.0% Turnout was highest in South Antrim (68%) and lowest in Lagan Valley (49%). 2.3 Spoilt ballot papers The proportion of spoilt ballot papers varied from 0.00% in the Isles of Scilly (where there were no spoilt ballot papers) to 0.24% in Leicester. 7 out of the 10 counting areas with the highest number of spoilt ballot papers were in London. Top 10 counting areas, spoilt ballots 1 Leicester 0.24% 2 Brent 0.22% 3 Haringey 0.18% 4 Orkney Islands 0.18% 5 Lambeth 0.17% 6 Tower Hamlets 0.17% 7 Luton 0.16% 8 Camden 0.16% 9 Newham 0.16% 10 Westminster 0.16%

18 18 European Union Referendum Characteristics associated with votes for Leave and Remain The referendum result suggests the UK is split almost down the middle on the issue of Europe. Polls before and after the referendum have attempted to analyse this split; who are the people who voted Leave and Remain? The analysis below shows there are clear relationships between the proportion of people in a local authority who voted Leave and the proportion voting UKIP in the 2014 European Parliament election, and between the proportion of Leave votes and the proportion of non-graduates. Weaker relationships exist between Leave votes and social grade, and immigration (where higher levels of immigration are associated with lower levels of Leave votes). The Telegraph summarised pre-referendum polling findings that suggested associations between intentions to vote Leave and age (60+), political party support (UKIP), social grade (C2DE) and level of education (non-graduate). Conversely, polls also found associations between intentions to vote Remain and age (18-29), political party support (Green), social grade (ABC1) and level of education (graduate). 2 Lord Ashcroft carried out a post-referendum poll that supports these suggestions. 3 This poll also found that the main reason people gave for voting Leave was the principle that decisions about the UK should be taken in the UK. The main reason people gave for voting Remain was the risks of voting to leave the EU looked too great when it came to things like the economy, jobs and prices. People who voted Leave and Remain were also found to have opposing views on the desirability of developments like multiculturalism, social liberalism, feminism, and the internet. Ballots are secret, but some of these polling findings are confirmed by the following analysis of the relationship between selected social characteristics and voting patterns, looking at their prevalence at local authority level. Referendum results are not available at local authority level for Northern Ireland, and Census data is not available for Gibraltar; the analysis below focuses on Great Britain. Deprivation is analysed only in England: data on deprivation in Scotland, Wales and Northern Ireland are not suitable for comparison. Social grade is a widely used measure to classify the population on the basis of occupation. Grades ABC1 refer to people in professional and managerial roles; grades C2DE refer to people in skilled, semi- or non-skilled manual jobs, the unemployed, casual workers and pensioners. 2 EU referendum: which type of person wants to leave, and who will be voting to remain?, The Telegraph, 22 June 2016, 3 How the United Kingdom voted on Thursday and why, Lord Ashcroft Polls, 24 June 2016,

19 19 Commons Library Briefing, 29 June 2016 Box 2: Data The analysis uses Census data on age, highest qualification obtained and social grade. The Census is a ten-yearly population count that covers all people who are usually resident in the UK. It was last held in Data from Library Briefing Paper 14/32 European Parliament election 2014 is used to analyse voting behaviour. The Department for Communities and Local Government ranks local authorities in England in terms of how they perform on indices of deprivation, where 1 is the most deprived and 326 is the least deprived. This data is used to analyse deprivation. HMRC estimates of child poverty (August 2013) and households claiming tax credits (April 2016) proportion calculated using the Annual Population Survey January-December 2014 were used to analyse socio-economic background. The ONS Annual Survey of Hours and Earnings provides information on median earnings by local authority, and the ONS Nomis claimant count dataset provides an estimate of the proportion of Job Seeker s Allowance claimants by local authority (May 2016). ONS data is used for the proportion of population by country of birth by local authority (August 2015). 3.1 Voting behaviour The charts below show a strong relationship between the proportion of people who voted Leave and the proportion who voted for UKIP in the 2014 European Parliament election. 4 A weaker relationship exists between the proportion of people in a local authority who voted for the Green Party in 2014 and the proportion voting Remain. Note that the proportion of people voting Green in any local authority did not exceed 24.5%. No significant relationships exist between the proportion of people who voted for Labour or the Conservative Party and the proportion of people voting Leave. This suggests a split in the support base of these parties on the issue of the EU (and related topics). 4 The characteristics of people who vote UKIP are discussed in more detail in Library Briefing Paper 5125 Membership of UK political parties and in Steven Ayres, Common Threads Among UKIP Voters, 2014,

20 20 European Union Referendum 2016 Voting behaviour: voting patterns at the 2014 European Parliament elections and vote shares for Leave (red) and Remain (blue) 3.2 Highest qualification and age The charts below show a strong correlation between the proportion of non-graduates in a local authority and the proportion voting Leave. The relationship between age and voting Leave or Remain was weak. Note that the proportion of people aged over 60 did not exceed 38% in any local authority. Likewise, the proportion of people aged did not exceed 32% in any local authority.

21 21 Commons Library Briefing, 29 June 2016 Highest qualification and age and vote shares for Leave (red) and Remain (blue) 3.3 Socio-economic indicators and immigration The charts below show a relationship between the proportion of people in social grade C2DE (manual and casual workers, the unemployed and pensioners) in a local authority and the proportion of people voting Leave. A weaker (negative) relationship exists between the median weekly earnings in local authorities and the proportion of people voting Leave. There is no obvious relationship between the rank of a local authority in terms of deprivation and vote shares for Leave. There is a weak relationship between the proportion of households claiming tax credits in a local authority, and the proportion of people voting Leave. A (negative) relationship exists between the proportion of the population that is foreign-born, an indication of immigration, and the vote share for Leave. 5 5 Data on immigration is not available for Ryedale, Derbyshire Dales, Hinckley and Bosworth, North Warwickshire, City of London, Isles of Scilly, West Somerset and Rossendale. These local authorities are excluded from the analysis.

22 22 European Union Referendum 2016 Socio-economic indicators: social grade, deprivation, earnings, tax credit claimants and immigration, and vote shares for Leave (red) and Remain (blue) There is no obvious relationship between the proportion of people claiming Jobseeker s Allowance, and the proportion of people voting Leave in a local authority. Likewise, there is no obvious relationship between the proportion of children living in low income households in a local authority, and the proportion of people voting Leave. This suggests that votes for Leave and Remain are not strongly related to this set of socio-economic indicators individually. 3.4 Rural and urban In England, urban areas recorded slightly higher vote shares for Remain than rural areas. High vote shares for Leave, however, were found in both rural and urban areas.

23 23 Commons Library Briefing, 29 June 2016 The Office for National Statistics has classified English local authorities into six groups according to their rurality: mainly rural (>80% rural), largely rural (>50% rural), urban with significant rural, urban with city and town, urban with minor conurbation, and urban with major conurbation. Among the 144 local authorities in rural categories, 25 (17%) recorded a Remain vote share greater than 50%. 44 (31%) recorded a Leave vote of 60% or more. Out of the 181 local authorities in the three urban classifications, 54 (30%) recorded a Remain vote share greater than 50%. 59 (33%) recorded a Leave vote share of 60% or more. In the most urban category, Remain took over 50% of the vote in a higher number of local authorities (38 out of 75) than Leave (34 out of 75). As the table below shows, high vote shares for Leave were found across local authority types. Top ten vote shares for Leave, by local authority type Leave Local authority Classification votes (%) Boston Urban with Significant Rural 75.6% South Holland Largely Rural 73.6% Castle Point Urban with City and Town 72.7% Thurrock Urban with Major Conurbation 72.3% Great Yarmouth Urban with Significant Rural 71.5% Fenland Largely Rural 71.4% Mansfield Urban with City and Town 70.9% Bolsover Urban with Significant Rural 70.8% East Lindsey Mainly Rural 70.7% NE Lincolnshire Urban with City and Town 69.9%

24 24 European Union Referendum Turnout Turnout across the UK was 72.2%. This was higher than turnout at the 2015 General Election (66.2%) and the most recent national referendum, on the introduction of the Alternative Vote electoral system, held in 2011 (42.0%). It was also higher than turnout at the previous referendum on the EU (then the EEC) in 1975 (64.0%). However, it was not as high as turnout at the Scottish independence referendum held in 2014 (84.6%). Turnout at general elections was higher than the 2016 referendum turnout between 1950 and 1992, with the exception of 1970 (72.0%). Across the UK, turnout in the referendum was 73.0% in England, 71.7% in Wales, 67.2% in Scotland, and 62.7% in Northern Ireland. Among the counting regions, turnout was highest in the South East and the South West and Gibraltar (76.7%), and lowest in the North East (69.3%). Turnout by region, ranked highest to lowest Turnout Remain (% Leave (% (valid votes Counting region of valid votes) of valid votes) as % of electorate) Electorate South East 48.2% 51.8% 76.7% 6,465,404 South West and Gibraltar 47.4% 52.6% 76.7% 4,138,134 East of England 43.5% 56.5% 75.7% 4,398,796 East Midlands 41.2% 58.8% 74.1% 3,384,299 West Midlands 40.7% 59.3% 72.0% 4,116,572 Wales 47.5% 52.5% 71.7% 2,270,272 Yorkshire and the Humber 42.3% 57.7% 70.6% 3,877,780 North West 46.3% 53.7% 69.9% 5,241,568 London 59.9% 40.1% 69.6% 5,424,768 North East 42.0% 58.0% 69.3% 1,934,341 Scotland 62.0% 38.0% 67.2% 3,987,112 Northern Ireland 55.8% 44.2% 62.7% 1,260,955 United Kingdom 48.1% 51.9% 72.2% 46,500,001 The table below shows turnout varied across local authorities in Great Britain and constituencies in Northern Ireland. Six out of the ten areas with the lowest turnout were in Northern Ireland. Highest and lowest turnout, by counting area Top 10 highest turnout Top 10 lowest turnout 1 Gibraltar 83.5% 1 Lagan Valley 48.9% 2 Chiltern 83.5% 2 Glasgow City 56.2% 3 Hart 82.6% 3 Belfast North 57.2% 4 St Albans 82.4% 4 Strangford 57.4% 5 Waverley 82.3% 5 Newham 59.2% 6 Mole Valley 82.1% 6 Manchester 59.7% 7 Richmond upon Thames 82.0% 7 Belfast East 59.7% 8 Derbyshire Dales 81.9% 8 North Lanarkshire 60.9% 9 Horsham 81.6% 9 North Down 61.6% 10 East Hampshire 81.6% 10 Foyle 61.7%

25 25 Commons Library Briefing, 29 June 2016

26 26 European Union Referendum Relationships between turnout and age, and Leave votes? The charts below show that there was a weak relationship between the proportion of people aged over 60 in a local authority, and the proportion of people turning out to vote. There was no significant relationship between higher turnout levels and higher levels of support for Leave. 4.2 The weather There is no evidence to suggest weather impacts on turnout, but the link has long been the subject of speculation. On 23 June 2016, there were thunderstorms overnight and into the morning, particularly in the South East of the country. These were accompanied by spells of particularly heavy rain which caused flash flooding in London and South East England. Polling stations in Kingston and Barking had to close due to flooding before new locations were found. The wettest place in the country was South Farnborough where 55 mm or two inches of rain fell in the 24 hours to 2100 on 23 June. 6 This was more than the average rainfall for the whole month in this region. 7 Manston in Kent was the warmest at 25 o C. Katesbridge in Northern Ireland the coldest with a minimum of 3 o C. 5 June 1975, the date of the EC Membership referendum, was drier than 23 June 2016 and around 3 o C cooler on average across the country. 8 6 Met Office UK weather last 24 hours (24 June 2016) 7 Met Office Hadley Centre Observation Data (UK precipitation series) 8 Based on daily England and Wales precipitation series and daily mean Central England Temperature series;

27 27 Commons Library Briefing, 29 June Opinion polls Opinion polls tracking voting intentions for the EU referendum were regularly published from 4 September 2015 up until polling day. 9 The chart below shows polling trends over this period (including all polls where fieldwork was carried out before the polling day). The trend lines show the average of the last ten polls (each individual set of poll findings is shown as dots). 60 EU referendum polling data: % leave / remain/ don't know Sep 15 Oct 15 Nov 15 Dec 15 Jan 16 Feb 16 Mar 16 Apr 16 May 16 Jun 16 Source: BBC News EU referendum poll tracker, What UK Thinks Most polls published close to polling day suggested a Remain win, and YouGov s polling day poll, based on a sample of people who had already voted, also predicted a 52% win for Remain. Polling methodology had received media attention in the run up to the referendum, after opinion polls had suggested an outcome that differed from the actual result in the 2015 General Election. Before the referendum, commentators noted the difference between polling findings in online and telephone polls. 10 The chart below shows how online and telephone polls suggested different levels of support for Leave between September 2015 and June Monthly polling averages for Leave online phone 50% 40% 30% 20% Polling companies aim for samples that are representative of the population, and they weight their data to achieve more representative results. These methodological decisions impact on the polling information they publish. 10% 0% September November January March May Source: BBC News EU referendum poll tracker, What UK Thinks It is not clear whether online of phone polls provided a more reliable indication of the referendum result. However, it is clear that most polls got it wrong. 9 See What UK Thinks EU for details of all published polls. The Telegraph and BBC News have published regularly updated poll-trackers. 10 See Why the difference between phone and online in EU referendum polling, YouGov, 23 February 2016

28 28 European Union Referendum Timeline of key events 23 January 2013: Prime Minister David Cameron promised a referendum if the Conservative Party were to win the next general election; During the 2013/14 and 2014/15 Sessions, Conservative MPs introduced Private Members Bills to introduce a referendum on EU membership based on a draft European Union (Referendum) Bill published by the Conservative Party on 14 May 2013; 27 May 2015: the European Union Referendum Bill was announced in the Queen s Speech, and introduced the day after. It had its Second Reading on 9 June 2015 and received the Royal Assent on 17 December 2015; 7 September 2015: the referendum question was changed to Should the UK remain a member of the European Union or leave the European Union, in line with the Electoral Commission s recommendations; 19 February 2016: Prime Minister David Cameron completed his negotiations with the other EU countries to reform the UK s relationship with the EU; 22 February 2016: regulations setting the date of the referendum (and other details) were laid before Parliament, and made on 3 March 2016; 13 April 2016: the Electoral Commission designated Vote Leave and The In Campaign as lead campaign organisations ; 15 April 2016: the referendum period, during which controls applied to the campaign, started. The purdah period, during which public bodies were restricted from publishing information related to the referendum, started on 27 May; 7 June 2016: the deadline to register to vote in the referendum. Following technical issues, it was extended to 9 June; 23 June 2016: polling day. Lead campaign organisations are entitled to benefits including higher spending limits, public funding, and referendum broadcasts

29 29 Commons Library Briefing, 29 June Previous referendums The table below summarises referendums that have taken place in the UK since 1975 at a national or regional level. The table does not include referendums held at local authority level. Summary of referendums since 1975 Subject of referendum Relevant area Date Outcome Continued membership of the European Community UK June 1975 Devolution Scotland March 1979 Devolution Wales March 1979 Devolution Scotland September 1997 Tax-varying powers for Scottish Parliament Scotland September 1997 Devolution Wales September 1997 Establishment of Greater London Authority London May 1998 Northern Ireland May 1998 Winning vote share Turnout UK remained in European Community 67.2% 64.0% Devolution did not proceed (threshold requirement not met) 51.6% 63.6% Devolution did not proceed 79.7% 58.8% Scottish Parliament established 74.3% 60.2% Scottish Parliament given taxraising powers 63.5% 60.2% National Assembly for Wales established 50.3% 50.1% Greater London Authority established 72.0% 34.0% Support given for provisions Good Friday Agreement of Agreement 71.1% 81.0% Elected North Regional Establishment of elected East Assembly not Regional Assembly England November 2004 established 77.9% 47.1% National Assembly for Wales given Expansion of Welsh greater law Assembly law making making powers Wales March 2011 powers 63.5% 35.2% Alternative Vote system Voting system UK May 2011 rejected 67.9% 42.0% Independence Scottish independence Scotland September 2014 rejected 55.3% 84.5% Source: Commons Library Briefing Papers 14/50 Scottish independence referendum 2014, 11/44 Alternative Vote referendum 2011

30 30 European Union Referendum Data tables 8.1 England Results by local authority, England (including Gibraltar) Counting area Leave Leave (%) Remain Remain (%) Spoilt Spoilt (%) Electorate Turnout (%) East Babergh 29, % 25, % % 70, % Basildon 67, % 30, % % 132, % Bedford 44, % 41, % % 119, % Braintree 52, % 33, % % 112, % Breckland 47, % 26, % % 98, % Brentwood 27, % 19, % % 58, % Broadland 42, % 35, % % 99, % Broxbourne 33, % 17, % % 68, % Cambridge 15, % 42, % % 80, % Castle Point 37, % 14, % % 68, % Central Bedfordshire 89, % 69, % % 204, % Chelmsford 53, % 47, % % 129, % Colchester 51, % 44, % % 127, % Dacorum 43, % 42, % % 108, % East Cambridgeshire 24, % 23, % % 62, % East Hertfordshire 42, % 42, % % 106, % Epping Forest 48, % 28, % % 100, % Fenland 37, % 15, % % 71, % Forest Heath 18, % 9, % % 38, % Great Yarmouth 35, % 14, % % 72, % Harlow 29, % 13, % % 59, % Hertsmere 28, % 27, % % 73, % Huntingdonshire 54, % 45, % % 128, % Ipswich 38, % 27, % % 91, % King's Lynn and West Norfolk 56, % 28, % % 113, % Luton 47, % 36, % % 127, % Maldon 24, % 14, % % 49, % Mid Suffolk 33, % 27, % % 78, % North Hertfordshire 35, % 42, % % 99, % North Norfolk 37, % 26, % % 83, % Norwich 29, % 37, % % 96, % Peterborough 53, % 34, % % 120, % Rochford 34, % 17, % % 66, % South Cambridgeshire 37, % 56, % % 114, % South Norfolk 41, % 38, % % 102, % Southend-on-Sea 54, % 39, % % 128, % St Albans 32, % 54, % % 104, % St Edmundsbury 35, % 26, % % 81, % Stevenage 27, % 18, % % 62, % Suffolk Coastal 41, % 37, % % 98, % Tendring 57, % 25, % % 111, % Three Rivers 27, % 25, % % 67, % Thurrock 57, % 22, % % 109, % Uttlesford 26, % 25, % % 64, % Watford 23, % 23, % % 65, % Waveney 41, % 24, % % 90, % Welwyn Hatfield 31, % 27, % % 78, %

31 31 Commons Library Briefing, 29 June 2016 Results by local authority, England (including Gibraltar) Counting area Leave Leave (%) Remain Remain (%) Spoilt Spoilt (%) Electorate Turnout (%) East Midlands Amber Valley 44, % 29, % % 96, % Ashfield 46, % 20, % % 91, % Bassetlaw 43, % 20, % % 85, % Blaby 33, % 22, % % 73, % Bolsover 29, % 12, % % 58, % Boston 22, % 7, % % 39, % Broxtowe 35, % 29, % % 83, % Charnwood 50, % 43, % % 133, % Chesterfield 34, % 22, % % 79, % Corby 20, % 11, % % 43, % Daventry 28, % 20, % % 61, % Derby 69, % 51, % % 171, % Derbyshire Dales 24, % 22, % % 57, % East Lindsey 56, % 23, % % 107, % East Northamptonshire 30, % 21, % % 68, % Erewash 40, % 25, % % 87, % Gedling 37, % 30, % % 88, % Harborough 27, % 27, % % 67, % High Peak 27, % 27, % % 72, % Hinckley and Bosworth 39, % 25, % % 85, % Kettering 32, % 21, % % 70, % Leicester 67, % 70, % % 213, % Lincoln 24, % 18, % % 63, % Mansfield 39, % 16, % % 77, % Melton 17, % 12, % % 37, % Newark and Sherwood 40, % 26, % % 87, % North East Derbyshire 37, % 22, % % 78, % North Kesteven 42, % 25, % % 86, % North West Leicestershire 34, % 22, % % 73, % Northampton 61, % 43, % % 144, % Nottingham 61, % 59, % % 195, % Oadby and Wigston 17, % 14, % % 42, % Rushcliffe 29, % 40, % % 86, % Rutland 11, % 11, % % 29, % South Derbyshire 34, % 22, % % 73, % South Holland 36, % 13, % % 65, % South Kesteven 49, % 33, % % 105, % South Northamptonshire 30, % 25, % % 71, % Wellingborough 25, % 15, % % 54, % West Lindsey 33, % 20, % % 73, % London Barking and Dagenham 46, % 27, % % 115, % Barnet 60, % 100, % % 223, % Bexley 80, % 47, % % 170, % Brent 48, % 72, % % 186, % Bromley 90, % 92, % % 231, % Camden 23, % 71, % % 145, % City of London 1, % 3, % % 5, % Croydon 78, % 92, % % 245, % Ealing 59, % 90, % % 212, % Enfield 60, % 76, % % 198, % Greenwich 52, % 65, % % 168, %

32 32 European Union Referendum 2016 Results by local authority, England (including Gibraltar) Counting area Leave Leave (%) Remain Remain (%) Spoilt Spoilt (%) Electorate Turnout (%) London (cont) Hackney 22, % 83, % % 163, % Hammersmith and Fulham 24, % 56, % % 114, % Haringey 25, % 79, % % 150, % Harrow 53, % 64, % % 162, % Havering 96, % 42, % % 183, % Hillingdon 74, % 58, % % 193, % Hounslow 56, % 58, % % 165, % Islington 25, % 76, % % 144, % Kensington and Chelsea 17, % 37, % % 83, % Kingston upon Thames 32, % 52, % % 108, % Lambeth 30, % 111, % % 210, % Lewisham 37, % 86, % % 197, % Merton 37, % 63, % % 136, % Newham 49, % 55, % % 176, % Redbridge 59, % 69, % % 189, % Richmond upon Thames 33, % 75, % % 132, % Southwark 35, % 94, % % 195, % Sutton 57, % 49, % % 140, % Tower Hamlets 35, % 73, % % 167, % Waltham Forest 44, % 64, % % 162, % Wandsworth 39, % 118, % % 219, % Westminster 24, % 53, % % 120, % North East County Durham 153, % 113, % % 389, % Darlington 30, % 24, % % 77, % Gateshead 58, % 44, % % 145, % Hartlepool 32, % 14, % % 70, % Middlesbrough 40, % 21, % % 94, % Newcastle upon Tyne 63, % 65, % % 190, % North Tyneside 60, % 52, % % 156, % Northumberland 96, % 82, % % 240, % Redcar and Cleveland 48, % 24, % % 103, % South Tyneside 49, % 30, % % 115, % Stockton-on-Tees 61, % 38, % % 141, % Sunderland 82, % 51, % % 207, % North West Allerdale 31, % 22, % % 74, % Barrow-in-Furness 21, % 14, % % 53, % Blackburn with Darwen 36, % 28, % % 100, % Blackpool 45, % 21, % % 102, % Bolton 80, % 57, % % 197, % Burnley 28, % 14, % % 64, % Bury 54, % 46, % % 141, % Carlisle 35, % 23, % % 80, % Cheshire East 113, % 107, % % 285, % Cheshire West and Chester 98, % 95, % % 259, % Chorley 36, % 27, % % 84, % Copeland 23, % 14, % % 54, %