The idea of a list order effect has been floating around for a number of years. For instance, in the 2017 local government elections, 81 per cent of the multimember wards—284 wards—had more than one candidate from the same party standing. We looked at the statistics and found that, in 73 per cent of those instances, the candidate for that party whose name was higher up the alphabet received the first preference. If Sue Bruce and I were standing for whatever party, I would get the second preference, even though I might be the better candidate, because she is “Bruce” and I am “O’Neill”.
That is the perception. Actually, we still do not know whether that is true—it is almost impossible to prove—but if you look at the preferential orders, which you can do because they are electronically counted, you can see that, in 82 per cent of Labour votes, 78 per cent of Scottish National Party votes, 68 per cent of Conservative votes and 64 per cent of Liberal Democrat votes, the candidate whose name was higher up the alphabet received the first preference.
You cannot say for certain that the reason why people would vote for Sue Bruce first and me second is that her name is higher up the alphabet than mine—there are millions of other reasons why people vote for particular people—but there is certainly a perception that the view that that is the reason probably has some validity. People worry about it and have thought about alternatives to alphabetic discrimination, as it is sometimes called.
The Scottish Government has indicated that it wants a review. In our report on the 2017 council elections, we said, “Whatever you do, you should test it to ensure that there are no unintended consequences,” and we volunteered to do the testing. In April this year, the Scottish Government asked us to assess the impact of alphabetic listing on voters, to look at the status quo, to consider the issue of drawing the order by lot—that is, using balls or whatever to order the list, which might result in a list that is all over the place—and the issue of adopting an alternating A to Z and Z to A ballot, where the order would be alphabetic on one paper and the reverse of that on another.
We employed Ipsos Mori to do some research with the public across Scotland to ensure that we got the views of voters. We talked with returning officers and deputy returning officers about the administrative impact of the proposals. We asked the political parties whether there was anything to do with campaigning that we needed to be aware of. We also talked to groups representing people with disabilities. In September this year, we published a report, which is in the Scottish Parliament information centre—no doubt many of you will have read it. What we found from the research—which was qualitative not quantitative—was that the ordering of candidates did not have an impact on the voters’ ability to vote for their preferred candidate.
One of the interesting things that we found from in-depth interviews was that many of the interviewees were not aware that the ballot paper was alphabetically ordered before we told them that it was. We talked to representatives of disability organisations, who were concerned that the changes might mean that disabled people would not be able to familiarise themselves with the ballot paper before they came to the polling station. We were told that disabled people quite often learn the order so they know how they are going to vote before they turn up, so adopting an alternating A to Z and Z to A ballot would mean that they would not know which list they would be facing and would have to learn both.
Administrators were confident that they could administer the process, but they wanted prescribed rules, because there are issues of transparency, particularly if we went down the road of using lots. That would probably mean that you would have to extend the election timetable. We got down to the level of thinking about how long it would take to do the pulling of lots and administrators thought that if you had, say, 20 wards with 15 lots per ward, it would take a long time—do the maths.
There was no consensus across the parties. SNP party officials were very keen on the A to Z, Z to A approach, and Liberal Democrats also supported that to a degree. The Conservative Party was for the status quo. The Labour Party and Scottish Greens did not have a view—to be fair, list order does not really have an impact on the Scottish Greens, because they tend not to stand more than one candidate in most places.