Between mid-April 2014 and the end of June, odds offered for bets on the independence referendum suggest bookmakers estimate the probability of a ‘No’ vote increased from 70 per cent to 79 per cent.
That’s the conclusion of the University of Stirling’s Professor David Bell, who has been monitoring bookmakers’ odds as a means of forecasting the outcome of the referendum. Bookmakers’ odds have a strong record in forecasting political events.
The economist, who is funded by the Economic and Social Research Council (ESRC), compared opinion poll data with figures drawn from 23 bookmakers’ published odds over the same period.
He observed considerable variability in the opinion poll data, but no change in momentum for one outcome or the other since the beginning of 2014.
Similarly, he observed no significant movement in the overall trend of the betting data – either positive or negative – with the implied probability of a ‘No’ vote remaining between 70 and 80 per cent throughout the period from January to June 2014.
However the betting data, which varies less from week to week than do the opinion polls, was much better at picking up short-term trends - and Professor Bell observed a shift in sentiment towards a ‘Yes’ vote from mid-March to mid-April and then a subsequent return to pre-March levels of support for ‘No’.
Professor Bell said: “Increased support for the ‘No’ outcome has offset the improved prospects for a ‘Yes’ outcome which was evident between mid-March and mid-April. This increase in support for ‘Yes’ was perhaps a reaction to Mr Osborne’s announcement that an independent Scotland would not be allowed to share the pound.
“Some of the change in odds will also be a response to punters who are willing to bet heavily on a ‘No’ outcome, including the individual reported widely in the media recently for placing a bet of £400,000 on Scotland staying part of the UK.”
Those betting on the referendum outcome are assessing not just the current state of the opinion polls, which reflect current voting intentions, but also the likelihood of a shift or change in these intentions over the rest of the campaign.
Professor Bell said: “Whether punters win their bet or not depends only on the outcome of the referendum, so they have to assess the chance of any major swing in voting intentions among the Scottish public.
“There may be a quite small margin between the ‘Yes’ and ‘No’ campaigns as the referendum approaches. But so long as punters do not think there is much chance of a sudden swing in voting intentions, their money will go increasingly towards the leading campaign.
“Bookmakers, who have to make a profit to survive, will respond by shortening the odds. And by doing so, they are implying that the chance of the other campaign winning is becoming very small.”
Professor Bell’s study is for the “Future of the UK and Scotland project” – an ESRC-funded programme which brings together a range of academics looking at Scotland’s constitutional future.
See accompanying image - figure 1: Opinion Polls and Implied Market Probability of a No Vote. The upper panel (in red) shows the proportion of committed voters (don’t knows are excluded) that would answer ‘No’ to the question ‘Should Scotland be an independent country?’ between March 2013 and June 2014. Data is drawn from the ‘What Scotland Thinks’ website run by John Curtice. The lower panel (in blue) shows Professor Bell’s estimate of the market expectation of a ‘No’ outcome drawn from published bookmaker’s odds over the same period. Opinion polls show much more variability than market expectations. John Curtice explains the reasons for this variability here. These include the method and size of sampling etc.
Professor David Bell is a Professor of Economics at the University of Stirling Management School. Since 2011, he has been providing policy advice to the Scottish, Westminster and Irish Governments. He also advises the Scottish Parliament’s Finance Committee, Scotland Bill Committee, Health and Sport Committee and Education and Culture Committee (in relation to economic and social conditions, Scotland’s funding position and potential changes in the financing regime). Professor Bell’s earlier paper ‘The Independence Referendum: Predicting the Outcome’ explains how the market expectation probability is calculated.
The Future of the UK and Scotland programme aims to both inform the debate in the run-up to the referendum and assist in planning across a wide range of areas which will be affected by the outcome of the vote - whether for Scottish independence or the Union. The programme – funded by Economic and Social Research Council - has its own website which presents research findings and information to support the debate - helping voters make an informed judgement about their decision in the referendum for Scottish independence. Visit http://www.futureukandscotland.ac.uk/ You can also follow the debate on Twitter: @UKScotland.