Published on EUI Times on 2, October 2014

After one of the longest and most intense campaigns in British political history, the referendum for the independence of Scotland is now over. The final result confirms what almost every opinion poll had predicted over the last year, a clear majority for the pro-union side. Yet, in the very last weeks of the campaign everything seemed to change and for the first time the ‘impossible’ outcome of a ‘Yes’ victory became conceivable.

The only point which everyone agreed upon on the eve of the vote was that the result would be too tight to call. Probably not even the strongest supporter of the union would have predicted a final difference of around 10 percentage points; 400,000 votes in favour of the pro-union camp. How can this somewhat unexpected result be explained?

First, it could be that opinion polls simply got it wrong. An analysis of 16 recent constitutional referendums shows that in 12 cases “the average vote for ‘Yes’ (which in each case was against the status quo) in the final polls was higher than was found in the ballot boxes”. It seems that polls tend to overestimate the popularity of change  – especially when turnout is expected to be high – due to different motivations underlying the choices. While ‘Yes’ voters tend to be more visible and more keen to express their opinions – as everyone could notice walking through the streets of Edinburgh in the weeks before the referendum – ‘No’ voters, on the other hand, might have been more reluctant to express their opinions because of fear of being seen as unpatriotic.

Secondly, opinion polls themselves could have contributed to mobilising the pro-union side, especially (and paradoxically) when for the first time, only ten days before the vote, they attributed a tiny majority to the independence side. A long-standing debate in political science identifies two opposite and co-existing reactions to opinion polls: a band-wagon effect and an underdog effect. The ‘pro-independence’ polls may have triggered an unexpected underdog effect, meaning that, in addition to boosting confidence among the ‘Yes’ supporters, they probably convinced reluctant pro-unionists to vote against the risks that are supposedly related to leaving the UK.

In this sense, the spectacularly high level of turnout could be a signal of this hidden mobilisation of a share of the population that traditionally does not go to the polls. This part of the electorate includes voters who are traditionally against big changes (such as the prospect of becoming a new state) for various reasons, ranging from personal interests to political apathy. Thus, in the very last few days of the campaign they were probably motivated to break the habit of a lifetime, leave their houses and go to the polls to secure the status quo.

Finally, opinion polls are just one of element of the many that help to explain the final outcome. The ‘emergency’ strategy carried out by the Better Together campaign in the last two weeks – with an increasing number of promises of further devolution –certainly contributed to the pro-union victory. Not only did all the main Westminster political leaders travel north to intensively campaign for the union in the final two weeks, but the vast majority of mainstream media also rushed to openly support the ‘No’ side, as two editorials in the Herald and the Scotsman made clear in the very last few days before the vote.

This silent mobilisation of pro-union supporters seems to partially explain the unexpectedly clear victory of the ‘No’ side. Convinced either by the need to secure personal interests or by politicians and the media, pro-unionists had the final word on Scottish independence – at least that is, until the next referendum.