We are now less than six weeks away from finding out whether the UK will choose to remain part of the EU or decide to chart its own course outside of the bloc. Polls suggest the race is nearly impossible to call: the active vote is split down the middle and there are enough undecideds to swamp any small preference either way. Interestingly, however, the bookies (who gauge form and sentiment for a living) are betting heavily on the side of the campaign to Remain.
Who are we to believe? What does this disparity mean? And, most importantly for NZ plc., which way is the UK likely to jump on June 23?
There are many factors at play, but it’s worth highlighting a few potential influences:
Firstly, the betting odds referenced above do not include a call on the winning margin. Bookies only need a Remain majority to be correct. The bookies aren’t bothered with whether its close or not, they are simply confident Remain will get at least one more vote than Leave. Interestingly, looking at the poll data in the different way, while Remain and Leave have been neck and neck for months, Remain has beaten Leave about 70% of the time (60% in online polls).
Secondly, most Brexit polls (93 of 115 since September 2015, in fact) are conducted online, and online polls often give respondents three active options (Remain, Leave, Undecided). Bookies, on the other hand, choose between the two real world possibilities (Remain or Leave). This approach effectively assumes ‘Undecideds’ make a decision and ‘Would not Vote’s’ don’t vote, as would be the case at the ballot box. Bookies believe more ‘undecideds’ will go the way of Remain, perhaps in part because the status quo is usually viewed as the safer option.
This belief has some support in the poll data. Similar to the bookies, phone polls often force In/Out answers. They are also more likely to sample voters who are not regular internet users (who on average may be older and more conservative). In any case, on average since late 2015 phone polls have returned a 9% higher vote for Remain than Internet polls, while Internet polls have averaged a 6% higher vote for Leave. In one specific case (dated 17 April) pollsters ICM conducted both online and phone polls. The former showed Leave ahead by 1%, the latter gave the lead to Remain by 7%.
The implication here is that while both online and phone polls have their issues, phone polls may provide a more accurate read on the Brexit outcome and the results of these are closer to those of the bookies than the headline polls.
Thirdly, not all voters are easy to get hold of, and analysis by the BES (the British Election Study) suggests that those that are harder to poll are more likely to vote to Remain. In fact, when they did their sampling an initial lead for Remain after counting those polled on the first attempt progressively increased by 7% after 8 attempts.
Of course, the more attempts made to contact a voter, the more expensive the whole process becomes. For most pollsters it’s a balancing act. However, most pollsters (whether online or over the phone) are unlikely to try contacting someone anywhere near 8 times to get their view.
Again, this would suggest the headline Brexit poll data underplays support for Remain.
So while on the surface the pollsters and the bookies appear to be telling us very different stories, looking into the detail suggests a conclusion that isn’t so inconsistent. Things are close, but at this stage the Remain campaign looks to be in the strongest position. That said, six weeks is a very long time in politics. There is still ample time for both sides to put their case across and swing the result.
This was the final entry of our three part blog on Brexit. Part I is here and part II is here.
David Rigby
Senior Analyst
Disclaimer: This is intended to provide general information only. It does not take into account your investment needs or personal circumstances and so is not intended to be viewed as investment or financial advice. Should you require financial advice you should always speak to an Authorised Financial Adviser.