Will Scotland Set a Turnout Record? Your Definitive Guide

An exceptionally high turnout is predicted in Scotland’s independence referendum next week. Voter registration is at record levels and the Chief Counting Officer has been urging people to avoid the busy times at the polling stations to beat the queues. But how high does turnout need to be to be exceptional? What are the records that could be broken? Alan Renwick here provides the definitive guide.

Will Scotland set any turnout records when voters go to the polls next Thursday? In order to find out, we need to investigate some points of comparison. Let’s start with UK general elections. Figure 1 shows turnout across the UK at all general elections since the arrival of universal male suffrage and something approaching universal female suffrage in 1918. The high-point came in 1950, when 84.0 per cent of eligible voters – the highest proportion in any poll in UK democratic history – made their mark. The low-point – aside from the unusual post-war election of December 1918 – was in 2001, when only 59.1 per cent of registered voters turned up.

To beat the UK record, therefore, more than 84 per cent of Scotland’s 4,285,323 registered voters need to cast a ballot. If our comparison is with more recent general elections, then 70 per cent is enough to look good.


Figure 1. Turnout in general elections – UK-wide

UoR blog - Scottish referendum - 2014-09-12 (Figure 1)


But this referendum is happening in Scotland, not the rest of the UK. So the Scottish figures might be of greater interest. It won’t be difficult to surpass turnout at Scottish Parliament elections: these have failed to capture the imagination of the Scottish electorate. The high-point came in 1999 – the first Scottish Parliament election – when 59.0 per cent of those who could vote did so. Since then, the numbers have been 49.7 per cent in 2003, 53.9 per cent in 2007, and 50.4 per cent in 2011.

Scottish voting figures for UK general elections are healthier. In fact, as Figure 2 shows, they are very similar to those for the UK as a whole. Note that the data that I have here go back only as far as 1945. The highest turnout comes, as in the UK as a whole, in 1950, when 80.9 per cent of Scotland’s registered voters cast a ballot. A record Scottish poll is thus easier to achieve than a record UK poll. The low-point was in 2001 – again the same as for the UK as a whole – at 58.2 per cent.


Figure 2. Turnout in UK general elections – Scotland only

UoR blog - Scottish referendum - 2014-09-12 (Figure 2


Still, however, this might not be the comparison we are most interested in: this is a referendum, after all, not an election. So what are the figures for past turnout in referendums? Figure 3 shows turnout for major UK referendums. The Scottish figures are highlighted in yellow.


Figure 3. Turnout in UK referendums

UoR blog - Scottish referendum - 2014-09-12 (Figure 3)


As is apparent, turnout is generally lower at referendums than at general elections. The highest proportion of Scottish voters ever to have cast a ballot at a referendum is 63.6 per cent, in the 1979 devolution referendum. In 1997, by contrast, only 60.2 per cent of those who could vote did so. There ought to be no trouble in beating these records. The only exception to the rule of low referendum turnout is the vote on the Good Friday Agreement in Northern Ireland in 1998, when 81.1 per cent of registered voters turned out. This was, like the current referendum, a vote on an issue of great import that mobilized the overwhelming bulk of the electorate.

So far this analysis has been a bit parochial. As passions rise and the rhetoric becomes ever more florid, Scots might aspire to something higher than beating mere UK records. What of rewriting the record books on the world stage?

Brits (or whatever we like to call ourselves) are, in fact, relative slouches in the turnout stakes when compared with many other nations. The Voter Turnout database of the International Institute for Democracy and Electoral Assistance usefully provides turnout data for just about any election you could want since 1945. There are, of course, some oddities: the Bolivian parliamentary elections of 1978 and those in the Bahamas in 1972 share the distinction of achieving turnout greater than 100 per cent of registered voters. The People’s Democratic Republic of Laos reported a figure of 99.9 per cent in 2002, as did Somalia in 1984. What happened in the Bahamas I don’t know, but the others were all elections in non-democracies.

Even among democracies, however, there are some truly impressive figures. In some cases – such as Australia, where turnout has been between 93 and 96 per cent at every post-war election – that is achieved by making voting compulsory. Among democratic countries without compulsory voting, the prize for the highest turnout (calculated as a proportion of registered voters) in any parliamentary election since the Second World War goes to New Zealand in 1946, at 97.6 per cent. Looking at elections over the last few decades, the highest turnout is in Malta: 97.2 per cent of registered voters cast a ballot in 1996. In twenty-first-century elections, tiny Nauru heads the pack, with turnout of 96.9 per cent in 2013. Scots will have to do something that, for Scotland, would be off the historical scale in order to challenge these records.

But comparisons with such small countries as Malta and Nauru is perhaps unfair. Figure 4 shows turnout at the most recent general election in every country in the European Union.


Figure 4. Turnout in the latest parliamentary election, EU countries

UoR blog - Scottish referendum - 2014-09-12 (Figure 4)


The UK is bang in the middle of the chart, and Scotland on its own would occupy exactly the same position. Aside from tiny Malta and Luxembourg, and also Belgium, where voting is, at least in theory, compulsory, the highest turnout – surprise surprise – is in famously happy Denmark, where 87.7 per cent of registered voters cast a ballot in 2011.

Denmark happens to have a population very similar to Scotland’s – 5.6 million compared with 5.3 million. So if 87.7 per cent of Danes can turn out to vote in a routine election, perhaps we should not get too excited about the rejuvenation of Scotland’s democratic community unless at least a similar level is reached in this far-from-routine referendum.

But we need also to do some comparisons with referendums around the world. A Swiss institute runs a comprehensive archive of referendum results. Less helpfully, they have not set it up in a way that gives the user a handy spreadsheet, so I have just looked for national referendums since 1970 in countries that are now in the European Union and that were established democracies at the time of the referendum. In order to avoid skewing things too much towards the countries that frequently hold referendums, I have looked at no more than the most recent ten referendums in any country. Even with these restrictions, there are 102 referendums, so in Figure 5, which shows turnout from them all, it is not possible to label which is which.


Figure 5. Turnout in referendums in Europe since 1970

UoR blog - Scottish referendum - 2014-09-12 (Figure 5)


Despite the lack of labels, Figure 5 shows quite a lot. First, turnout in referendums varies enormously. The low is 9.5 per cent in a referendum in Slovakia in 1997, which the opposition parties successfully boycotted. The high is – again – in Malta, this time in 2003, when 90.9 per cent of registered voters turned out to give their view on whether their country should join the EU. Average turnout is low – just 49.9 per cent – and median turnout is a fraction lower still, at 49.4 per cent. In only eighteen of the 102 referendums has turnout exceeded 70 per cent. It therefore should not be too hard for Scotland to find a place at the referendum turnout top table.

But that, of course, is because most referendums are nothing like the current referendum in Scotland. Many of those with very low turnouts are citizen-initiated votes that fail to capture the imagination of the general public or that the political establishment successfully squashes through silence. Many referendums are on relatively technical issues that excite few passions. Figure 6 therefore narrows in on the referendums with turnout above 70 per cent to see what Scotland is really competing against.


Figure 6. Turnout in referendums in Europe since 1970: cases above 70 per cent only

UoR blog - Scottish referendum - 2014-09-12 (Figure 6)


Thirteen of these eighteen referendums were on EU matters: six on whether to join the EU itself, two on whether to join the euro, and others on a variety of EU treaties. Those happy Danes managed to secure 76.2 per cent turnout in 1998 on the Amsterdam Treaty – which most voters in the UK were not even aware of. The remaining topics were diverse – from the voting age in Denmark in 1971 to the creation of a united Republic of Cyprus (the figure here is only for the Greek part of Cyprus) in 2004.

Aside from the Cyprus vote, it would be difficult to argue that any of these referendums had implications as deep for the people as the country concerned as the current Scottish referendum. So, again, we shouldn’t get too carried away with claims about the exceptional depth of Scots’ engagement with the democratic process unless we see a turnout towards the top of this range.

Though some of these European referendums achieved very high turnout, none of them is the record-holder if we look across the world as a whole. The highest turnout I know of was in a referendum that has already been much discussed in Scotland in the last few weeks: the referendum on independence for Quebec in October 1995, when 93.5 per cent of registered voters cast a ballot. The Scots have quite a mountain to climb to beat that. (If anyone knows a case with higher turnout, please let me know!)

To sum up then, here are the various points of comparison to bear in mind when deciding just how high turnout has been as the results come in next Thursday night:


UoR blog - Scottish referendum - 2014-09-12 (Table, corrected)

Scottish voters seem likely to break some records next Thursday, but just how many remains to be seen. The more of these markers are passed, the more we will be able to say that a remarkable mobilization of the Scottish electorate has taken place.


Data sources:

  • UK-wide election turnouts: David Butler and Gareth Butler (2011). British Political Facts, 10th ed. Basingstoke: Palgrave Macmillan, pp. 265-71.
  • Turnouts in Scotland for UK general elections: David Butler and others, The British General Election of [year], 1945-2010, appendices in each edition.
  • Turnouts in Scottish Parliament elections: Paul Cairney and Neil McGarvey (2013). Scottish Politics, 2nd ed. Basingstoke: Palgrave Macmillan, p. 80.
  • Turnouts in UK referendums: David Butler and Gareth Butler (2011). British Political Facts, 10th ed. Basingstoke: Palgrave Macmillan, pp. 493 and 498; Paul Cairney and Neil McGarvey (2013). Scottish Politics, 2nd ed. Basingstoke: Palgrave Macmillan, pp. 26 and 30; David Butler and Uwe Kitzinger (1976). The 1975 Referendum. London: Macmillan, p. 266; BBC News: reports on the London referendum and North East referendum; Electoral Commission (for the 2011 AV and Welsh devolution referendums).
  • Election turnouts around the world: International Institute for Democracy and Electoral Assistance, Voter Turnout database, accessed 12 September 2014.
  • Referendum turnouts around the world: Datenbank und Suchmaschine für direkte Demokratie, accessed 12 September 2014.


Note: An error in the table was corrected on 16 September 2014.

This entry was posted in British Politics. Bookmark the permalink.

Leave a Reply

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