S3M-8127 Local Electoral Administration (Scotland) Bill

The Presiding Officer (Alex Fergusson): The next item of business is the stage 3 debate on the Local Electoral Administration (Scotland) Bill.

09:24
... ... ...
09:39

Stewart Stevenson (Banff and Buchan) (SNP): The debate is perhaps an opportunity to look at the changing nature of how we run elections. If we go back to the UK election that took place in 1832, which is the earliest one for which I have been able to find records, 658 members of the House of Commons were elected and 827,776 people cast votes, so the number of votes per member of Parliament was just a wee bit over 1,000. That was a very different environment from the one in which we live now. Indeed, fewer votes were cast for each MP than we would now expect to be cast for each member of a local authority.

If we fast-forward to the Westminster election of 1945, we had multimember seats and seats for which the alternative vote or the STV system was used. We are looking at changing the electoral system for Westminster elections, but the Conservatives, in particular, will not be in favour of the multimember first-past-the-post system that Brian Donohoe proposed yesterday in a House of Commons debate as a replacement for the list system for Scottish Parliament elections because, of course, in 1922, when Churchill stood for re-election in Dundee, he came third in a two-member seat. He was defeated by a Scottish prohibitionist, Edwin Scrymgeour and by the Labour candidate. The results are not always what we expect.

In 1945, when three members were elected to the Combined Scottish Universities seat by STV, a form of alternative voting, the third person who was elected on the second ballot obtained only 4.15 per cent of the first preference votes and was elected despite losing their deposit. Therefore, the systems that we have had over the years can lead to various differences.

Moving forward to the general election of October 1974, the turnout in Scotland was 74.81 per cent. That was a highly memorable election. After it, Westminster had more nationalist members than it had Liberal members.

Jamie Stone (Caithness, Sutherland and Easter Ross) (LD): Shame!

Stewart Stevenson: There were 13 Liberal members and 14 nationalist members, including three Plaid Cymru members and others. It is clear that, over the piece, there were changes in the way things were done. In 1945, it was a fortnight after the election before the results were known because, in days before the advent of the internet, the service vote took some logistical organisation.

I add to the commendation that there will be for Duncan McNeil’s contribution on the subject in October 2008, when he reported to Parliament on his committee’s deliberations. The committee’s work was vital in underpinning what we are discussing. Its report highlighted a general point that I and my colleagues and, I think, some others would make, which is that having different bodies and different parliamentarians responsible for the rules for different elections is a potential source of difficulty. It is certainly the case that in 2007 the Scotland Office did not cover itself in glory.

Even though a vote on the use of AV for Westminster elections is coming up shortly, it has not led to a single question from an elector to me so far. The SNP has just completed two days in Glasgow at our party conference. In my hearing, the subject never arose, although it may have arisen in other people’s hearing.

We have heard about some of the difficulties in 2007. It is certainly important that the Electoral Commission should report on how elections have gone. An illustration of when a report by the Electoral Commission might have been useful is the referendum that was held on 1 March 1979. I was at the count in Lothian. Members who are old enough to remember the campaign may remember that the “no” campaign bought lots of poster space. The posters had a picture of the ballot paper with the words “yes” and “no” on it. Opposite the word “no”, instead of an X, the word “no” was written. More than 2,000 electors in Lothian chose to write the word “no” opposite the “no” option.

We might think that that was fair enough. Most of us here might think that the electors’ intentions were relatively clear, and that is the normal test. However, on that occasion, the returning officer decided that, because the electors had written “no” opposite the word “no”, those votes should count as a “yes”. Being a campaigner for the “yes” campaign, I was not greatly upset by that decision, although I was astonished by it. On appeal, the returning officer of that count agreed that those votes would be counted as spoilt papers. That is an example to show that it was not just in 2007 that we have had difficulties; there have been previous occasions on which it would have been right and proper to examine what went on.

When we have complex elections, it is important that the electors know what is going on. One of the rules in the forthcoming election, as in all previous elections, prohibits election communications from referring to other elections, which might help people to understand the nature of other, simultaneous elections. That prohibition might be thought to be unhelpful and the Electoral Commission might have to look at that.

As someone who spent 30 years in computers, I will make a wee reference to the nature of some of the difficulties that might arise with computer systems. We computery people always used to apply a rule of thumb when we were given numbers relating to the throughput of a computer system. The rule of thumb was that marketing people always get estimates wrong by a factor of 10. It was the computer people’s job to work out whether to divide or multiply. In some ways, that is exactly what part of the problem was in 2007. We did not anticipate that more than 20 people would be standing on some of the lists, and there was a limitation in the software. In Lothian, the number standing on the list exceeded that limit so there was a last-minute ad hoc redesign of the ballot form that caused the computer systems great difficulties. I hope that the stress testing that will take place in the autumn will focus on some of the more unlikely boundary conditions that might occur, because that is where computer systems almost invariably fail.

I am pleased to see the legislation coming through Parliament. I sniff not a whiff of dissent and I hope that the motion will be carried unanimously at decision time.

09:47