Local Government and Communities Committee, 05 Dec 2007

Meeting date: Wednesday, December 5, 2007

Official Report 388KB pdf

---

“Scottish elections 2007”

We move to item 4 on the agenda. The committee will take evidence from Sir Neil McIntosh CBE, the electoral commissioner; Peter Wardle, chief executive of the Electoral Commission; and Andy O'Neill, the commission's head of office for Scotland. Gentlemen, you are warmly welcome.

Previously the committee took evidence from Mr Ron Gould, the author of the independent report on the Scottish Parliament and local government elections. The Electoral Commission has now responded to that report. If you wish to make an opening statement before we move to questions, please proceed.

I will keep it brief. Thank you for giving me the opportunity to appear before the committee. On my left is Peter Wardle and on my right is Andy O'Neill. We are here to discuss the report by Ron Gould, which was commissioned as an independent report in the aftermath of the elections. We are pleased that the report has generally been seen as having identified a number of the key issues that need to be addressed. We have made written submissions, so I will not take up any more of the committee's time, if members wish to move to questions.

I appreciate that. Do you agree that the main problems that were experienced in May 2007 related to the Scottish Parliament election and that, to some extent, the local government election was caught in the slipstream?

We must see the problems as combined, because the elections were combined. The problems that arose can be placed in three categories: combination, legislation and fragmentation. The problems applied across the piece and, inevitably, they affected both elections.

Would decoupling the elections, which you support, solve many of the problems that arose in May 2007?

The commission's position is that it has become apparent that it is not possible to administer the Scottish Parliament and local government elections together with total effectiveness and that there is a strong argument for decombination. As you know, convener, that argument was advanced first by the commission on local government and the Scottish Parliament, secondly by the Arbuthnott commission on boundary differences and voting systems, and now the Electoral Commission has recognised that the issues still need to be addressed and that separating the elections will be the most effective approach.

As you say, decoupling will address some of the issues that emerged in the election. However, Mr Gould's recommendations go considerably beyond decoupling. Does electoral legislation in Scotland need to be dramatically overhauled and, if so, in what terms?

It needs to be overhauled not only in Scotland, but in the United Kingdom. After all, the Electoral Commission has an interest in what happens on the Scottish scene, but its role is much broader than that. I invite Peter Wardle to comment on consolidating legislation.

For a number of years now, the commission has been saying that electoral legislation across the UK is a mess. Indeed, we saw evidence of that in Scotland in the 2007 elections. I know, both from his report and from talking to him, that Ron Gould found it very difficult to get his head around the provisions in Scotland for running the two elections that were the subject of his inquiry. We have raised the same point not only in Scotland but in other parts of the UK.

There are many very good and well understood reasons for the current situation, particularly the length of time democracy has been running in the UK. However, Ron Gould's comments have reinforced the feeling of the commission and others that the legislation is getting too complicated and that it is time for a real overhaul. By that, I mean that the legislation should certainly be consolidated, though I would go further than that and suggest that we need to consider how it might be simplified. In fact, in his report, Ron Gould commented that the legislation for e-counting was made simply by taking existing legislation for paper-based counts and replacing the person counting pieces of paper in a sports centre with a computer. However, it does not work that way. We need a much more radical examination of these issues if we are to get e-counting right in future. Consideration of how the legislation might be consolidated and simplified and how it interacts with different forms of elections is certainly important and must be given much greater priority.

The situation in Scotland is complicated further by the fact that legislation on the electoral process and the administration of local and national elections comes from different sources. Does the Electoral Commission take a view on that issue?

The commission does not have a position on legislation itself, which should be discussed by the constituent Parliaments. However, administration is a different matter. As you know, Ron Gould has suggested that the Scottish Parliament and local government elections be handled by one administrative process with one chief returning officer.

Having answered the general question, I think that I will stop there rather than take up more of the committee's time. However, I am happy to develop the point.

Surely the commission's response and advice to other bodies with regard to decoupling and the handling of elections applies right across the board to European elections, UK elections, Scottish Parliament elections, local government elections and referendums.

Precisely so. As Scottish Parliament and local government elections are held and managed in a Scottish context, some might well argue that they might be better handled with proper co-ordination. However, the commission needs to consider how such moves would fit into the wider scene and structure of other elections and to think about what lessons can be learned about the approach to elections, not only by Scotland but by other constituent nations in the UK.

Do you agree that the furore over the number of spoiled ballot papers in the Scottish Parliament elections masked problems that arose in the local government elections? I have serious concerns that many members of the public did not realise that they could vote 1, 2, 3, 4, 5 so they simply put one X in one box.

Another concern is the alphabetical ordering on the ballot paper. It is clear that, the higher up the alphabet their name, the more likely candidates are to get elected. The Conservatives carried out research that concluded that 92 per cent of its successful candidates were high up the alphabet where more than one Conservative candidate was elected. I think that a similar pattern could be seen—across the party divide—in every North Ayrshire ward. Indeed, even if we lump together candidates from the same political party, the share of the votes among them also went 1, 2, 3. For example, of the three Labour candidates in one ward, the highest on the list received significantly more votes than the next highest up the list, who in turn received significantly more votes than the third. Of course that could be for other reasons, but that seems a clear flaw in the single transferable vote system.

Let me deal with those two points separately.

On the generality of the STV system, I would not call that a flaw in the system. We need to analyse exactly what happened and what the background reasons were. For instance, the rate of rejected ballot papers in the local government elections was obviously less than in the Scottish Parliament elections. At 1.89 per cent—let us call it 2 per cent—the rate of rejected ballot papers in the local government elections equated broadly to a rate that was seen in Northern Ireland when there were combined elections. One issue is whether the combining of elections in which electors are required to write an X as well as 1, 2, 3, 4 creates an added level of complexity. That said, you are quite right that Xs were used, including multiple Xs in some cases. We need to address the issues of voter information and of process. However, if you are asking me to say that STV has a fundamental flaw compared with other systems, I could not say that.

Surely there is a flaw. For the Charter 88 and Unlock Democracy report, 26 per cent of the volunteers who were recruited to monitor the election did not understand how STV worked. There was a furore of party-political broadcasts and leaflets coming through doors at election time. Much of the information that was issued to educate voters must surely have been lost in that mêlée. We did not have a significant enough education programme to let people know about the system.

How will we even know the level of public concern about the system if we take only a two-dimensional view by considering only the spoiled ballot papers? I believe that many people did not know that they could vote 1, 2, 3. Obviously, to a large extent, the onus is on voters to know about the system before they enter the polling booth. However, ballot papers on which an X was written against three candidates—whether or not those candidates were independents or from different political parties or a mixture—were automatically discounted because the votes were not prioritised 1, 2, 3. Is there any way in which such votes could be treated as having an equal priority, perhaps by giving them a reduced weighting, rather than discounting those ballot papers altogether?

There are some difficulties with that suggestion, but you are absolutely right that we should consider those issues. To look ahead, we need to learn lessons from the past. Although separate elections would allow public information to concentrate on the voting system for local government, such issues still need to be addressed. We need to explore the sort of thinking that you have set out. That is perfectly reasonable and proper.

Should we consider, for example, randomising the order of names on the ballot paper?

Let me turn to that, which was the second point that you made.

The commission has established quite clearly that the first name on a list benefits from that position, so we will study the grouping of candidates in order to find out how a better and fairer expression might be produced. The list could be varied simply by lot. Rotation is probably the fairest way, although it starts to become very complicated in that setting. As you will appreciate, the issue is about not just the use of the alphabetical approach to candidates' names, but to parties' names. Peter Wardle might want to say something about how the commission is approaching that general point.

Mr Gibson made the point that messages to voters can be lost. One of the lessons that I take from Ron Gould's report is that most voters do not actually think about what they have to do with the ballot paper until they turn up at the polling station—that is fairly well agreed. What worked best at the Scottish elections was having good advice in the polling station so that voters could get good guidance from the staff and had good information officers there to help them when they thought, "What do I do now?" That was when they needed the information. The commission, Ron Gould and others will all have seen patchy performance on that and we want to take that lesson away, so that as much effort as possible is put in to give the voter the information when they need it, which is when they come to vote.

The commission is examining ballot paper design, and party names and descriptions. In our document about moving on from the Gould report, we said that we would take on the role of developing UK-wide standards for ballot paper design. That would pick up a number of the issues that Ron Gould and others identified in relation to the May elections in Scotland.

We will also look at party descriptions. We had already been doing that following the elections in Wales in May this year, in which there were issues about the use and the numbers of party descriptions. We plan to complete the first stage of our consultation on that by the end of January 2008. We will be looking at all the issues that Ron Gould mentioned to do with party descriptions, alphabetisation, randomisation and so on, and we will try to come forward with some options.

It is not necessarily the case that every one of Ron Gould's options has to be taken up in order to solve the problem. I am aware of some concern from the voters' perspective. For example, voters in Scotland are used to folding their ballot papers and in May we saw a lot of people who got confused when they were asked to do it differently. We also need to bear it in mind that a lot of people in Scotland are used to seeing an alphabetical list, and so any departure from that will need to be considered; it might solve one problem but create another.

I am sure that you will do a lot of work on that. I will pursue two points that were well rehearsed with Ron Gould. First, given the Electoral Commission's role in informing people, I am interested that providing information might be pointless because no one reads it or pays attention to it until they arrive at the polling station. We might come to that later. Secondly, there was some polling work that showed that people had a problem with lack of information. How does the Electoral Commission respond to that and will you try to reduce that problem?

Also, ballot papers were made available to Ron Gould and his team. Was the Electoral Commission able to study any of those ballot papers to support Kenny Gibson's contention that the failure rate was more than 2 per cent? We are focusing on decoupling, but there was only a 2 per cent failure rate in the local government elections, which is within the acceptable threshold. We all know from anecdotal evidence, however, that a paper on which someone put a single X was not counted as a failed paper. Do we know how many ballots were passed, although not completed correctly?

I will respond to one of your points, but Peter Wardle will come in first.

On your first point, I am sorry if I did not make myself clear. I was not saying that information campaigns are not worth while. My point is that the lesson from Ron Gould was that none of us did well enough in providing consistent information to voters at the point of voting on polling day.

We have heard politicians apologise and say that they did not do well enough, but what did the Electoral Commission not do well enough?

The Electoral Commission has made it clear that we did not do two things well, both of which relate to the pre-election day campaign. The first thing concerns the research that we conducted at the request of the Scotland Office in relation to the parliamentary ballot paper. We have accepted Ron Gould's criticism that we should have used a different methodology. We used the best methodology that we were advised was available at the time but, clearly, it was not good enough and we will not use it again.

The second thing that we did not do well was to do with the evaluation of our campaign, which is a rather detailed point that is covered in full in one of the annexes to Ron Gould's report. We did not conduct the evaluation sufficiently close to when the campaign had taken place to get good enough statistics. We will take on board both those points.

As a more general point, we have taken on board the general criticism directed at all those who were involved in the election, which is that we failed to place voters' interests at the heart of the planning, the legislation, the preparations and so on. We share that responsibility and we have certainly committed to renewing our efforts to put the voter first—as have, I am glad to say, a number of other people.

With regard to the day of the election, we gave guidance on how the information officers should do their job and the polling clerks should advise electors about, for example, how many votes they had on the STV paper. That guidance was perfectly adequate—it was good and right—but it was not always followed consistently and thoroughly and we will want to do more to ensure that it is in future.

You asked about the access to the ballot papers that was given to Ron Gould and his team. The Electoral Commission did not, separately, use that power to do any of its own analysis. As Ron Gould makes clear in his report, he was concerned that he should do nothing that could be construed as questioning the validity of the results of the election or interfering with the secrecy of the ballot. For that reason, he looked only at images of the face of the ballot papers. He did not look at the actual ballot papers, so that there could be no suggestion that he had turned them over and seen identifying marks or whatever else. The analysis of the ballot papers that was carried out is the analysis that is in Ron Gould's report. We have not carried out any further analysis.

On the indications from some of the pre-election research of some of the problems that might be experienced by voters, I have commented on that already in relation to the parliamentary ballot papers. The research on the STV ballot paper was carried out by the Scottish Executive, not the Electoral Commission. I cannot say much about that, except that we clearly looked at some of the issues that emerged from that public testing of the ballot papers and tried to ensure that our guidance to returning officers about the information that voters would require addressed those problems. As I said before, we need to ensure that that guidance is followed more consistently in future.

I am conscious that you might want to move on to other points, convener, so I will not speak at this point.

Others have mentioned the issue of information officers. Mr Gould said that they were

"a last-minute desperate attempt to try to resolve the problem of lack of voter information".—[Official Report, Local Government and Communities Committee, 21 November 2007; c 280.]

You said that you felt that the provision of information officers was patchy, as was the information that they had. I am sure that members of the committee have anecdotes to that effect. I remember one voter telling me that they had asked the information officer in the polling place, "Should I make just one mark on the parliamentary paper?" and that the information officer, who, in retrospect, obviously thought that they were asking whether they should make only one mark in one column, said yes. That voter was, therefore, disfranchised on the second half of the parliamentary paper. Do you see information officers having a role in future? If so, what should that role be, and by whom should it be supervised?

The information officers were not an afterthought. In his comments, Ron Gould was referring to the late stage at which funding was approved to permit them to be used. In some authorities, they were used in 2003. In 2005, Andy O'Neill came back from the Northern Ireland elections and proposed to the steering group that the use of information officers was important, so it has been an underlying theme.

Information officers are important because, even with all the advertising in the world, individuals entering a polling station may not have been reached. In that situation, a personal contact can be more than helpful. Where information officers operated effectively, that was appreciated. However, provision was not consistent across the piece and could have been better. Returning officers, who were responsible for recruiting, appointing and training information officers, with support from the commission, would accept that. There is a role for information officers at polling stations in the future, but it must be developed, sustained by effective training and much more focused.

I return briefly to the issue of ballot paper design in the Scottish Parliament election. The research that the commission commissioned from Cragg Ross Dawson involved only 100 participants but, significantly, it indicated a 4 per cent rejection rate for ballot papers. As Mr Gould records in his report, that was very close to what happened on 3 May. You have indicated that you do not believe that the best methodology was adopted for that research and that it could be improved on. Was the problem that the sample was so small that the warning sign of a 4 per cent rejection rate was not given the significance that it deserved?

I invite Andy O'Neill to talk you through the detail of the issue.

As Peter Wardle said, we accept that the methodology for the research was not right. However, what the Scotland Office asked us to do in the summer of 2006 was to take soundings on the aspects of the ballot paper that people liked and did not like. We did that. In August 2006, we passed the report to the Scotland Office, highlighting a number of areas in which we felt further work should be done. At that point, our involvement in the design of the ballot paper ended. As Mr Gould points out in his report, it was for the Scotland Office and DRS, the e-counting supplier, to finalise the ballot paper. The fundamental point is Ron Gould's recommendation that a full, in-depth research programme on the ballot paper, leading to what the voter in the polling booth would see on 3 May, should have been undertaken. We accept that it was not and will seek to ensure that that happens in future.

Andy O'Neill has indicated when our involvement ended, but the commission accepts that it should not have ended at that point. Although the matter was not our responsibility, we should have recognised that there was an issue and should have pressed for action to be taken.

So you think that, with the benefit of hindsight—which is a wonderful thing—you should have pursued the matter or, having passed the report to Government, should have monitored Government's response or maintained a dialogue on the issue.

Precisely. Everyone who saw the message should have said that more needed to be done and that we needed to press forward on the issue. With hindsight, we see that clearly.

One of the problems that Mr Gould identifies is the lateness of legislation. Because we were trying to do a lot of other things, we failed to carry out the task that we should have performed of ensuring that the Scotland Office took up the matter.

There was a 4 per cent failure rate for ballots for the Scottish Parliament election and failure rate of nearly 2 per cent for ballots for the local government election. What do you regard as an acceptable failure rate or as the norm? There will always be spoiled ballot papers, but what is the target for properly completed ballot papers for which we should aim?

I will not try to dodge your question, but you will appreciate that the obvious answer to it is zero. One would like all ballot papers to be completed correctly. Of course, we should not forget that 3 per cent of the 4 per cent of papers that you mentioned were blank. That could mean that people were confused, but we know that some voters demonstrate their opinions by not completing their ballot papers. Therefore, an element of voter choice comes in, but that would not have created the skew that there was. We must consider international comparisons to find out what is happening. With respect to STV, at the very least, we should try to achieve a pattern of no more than 1 per cent of ballot papers being spoiled.

It is important to the commission that what actually happened in the elections is considered and people drill down to what caused papers to be rejected. We must ask what was in voters' minds and what was missing. Some 1 per cent of papers could be spoiled in an election, but a serious issue might still need to be addressed. I am loth to say that there is an acceptable percentage of rejected ballot papers because, by definition, there is not. No voter should lose their vote for any reason other than because they have chosen not to exercise their choice.

I want to return to decoupling. You chaired a committee that considered local government and the timing of elections, and you obviously have a lot of professional experience. I think that you mentioned the Arbuthnott committee. You may take off your Electoral Commission hat when you answer my questions if you need to. What is your view on the timing relationship between Scottish Parliament and local government elections? It is obvious that there will have to be a transitional arrangement if we want to decouple them. How should that transition be facilitated?

I will not take off my Electoral Commission hat, convener.

Indeed. You are here to represent the Electoral Commission.

Yes. However, it may be helpful to say, bearing it in mind that my term of office ends in the coming January—

You could come back in February.

—that an option that was raised with the commission was that local government could have a five-year term. Its elections would therefore be moved a year behind parliamentary elections. It could then have a second five-year term, which would take local government elections to the mid-point between Scottish Parliament elections. Terms would then run from there. That would be consistent with the views of the commission on local government and the Scottish Parliament and Arbuthnott. We would probably like elections to be clearly separated. One would not want one set of elections to be held one or two months after the other set.

I want to ask about one of the confusions that the elections caused. There were 32 returning officers—one for each local authority—but much of the responsibility for elections in Scotland is held in London. Mr Gould thinks that there should be a returning officer for the whole of Scotland. What is the commission's view on that? How can it be progressed? As it is possible that we could have overall control of Scottish elections in Scotland, should we consider the benefits of a single voting system, whether an STV system or another system? Would there be less voter confusion?

I should clarify that the 32 returning officers are not controlled from London. Each is independent in his or her own right. The fact that there are 32 of them inevitably creates problems of consistency, which is why Ron Gould suggested a co-ordinating role in Scotland for Scottish elections.

Broadly, the commission agrees that more co-ordination is needed. There was fragmentation, which creates its own problems, but the delivery of more co-ordination is inevitably more complicated, and things must be seen in a United Kingdom context. There is no point in addressing fragmentation in elections in Scotland and ending up with fragmentation across the UK. It is important that principles and best practice are established. However, in broad terms, having a Scottish co-ordinating returning officer role would provide one means of addressing the issue. That covers that point.

It is inevitable that voting systems will be determined by Westminster or the Scottish Parliament—that depends on the situation. The commission does not get into the political—with a small "p"—consideration of that.

Mr Gould suggests that each local authority should have a full-time officer who is responsible for dealing with electoral administration. Should that be a new post or could electoral registration officers be given that responsibility in each local authority area?

I do not necessarily think that electoral registration officers could take that responsibility; they have a clear role. Any returning officer set-up includes the nominal returning officer, who tends to be the chief executive, as all members know, and an electoral administrator, so authorities have such professionals. Ron Gould has made an interesting point about carrying that professionalisation through to a higher degree, which raises issues about responsibility.

The proposal merits consideration. We accept the thrust of the argument that elections are no longer run only once every four or five years but regularly and that there is a strong public interest in running them effectively and consistently and in their being seen to be effective. That needs a new approach to uprate the management of elections—and perhaps even its resourcing—in Scotland, quite apart from the rest of the United Kingdom.

You have said, in response to the Gould report, that you generally accept the recommendation to have a chief returning officer for Scotland. The debate is about what that officer's role should be. Why would that role not include clear responsibility for providing information, education and training, and support for information officers? I felt that the Electoral Commission's response to that argument was a bit defensive. The commission certainly has reservations—it devotes four or five paragraphs to the division of responsibilities.

We have made it clear in our response that a chief returning officer could easily work in Scotland for the elections that Ron Gould has examined. As you say, we have set out several points that would need to be considered and worked through if that solution were adopted. Separately, we have said that the idea relates to a wider debate that was brewing in other parts of the UK and which Ron Gould's recommendation has strengthened. In the UK, a chief returning officer has been proposed in Scotland, and Northern Ireland already has a chief electoral officer. If Scotland and Northern Ireland were to have those posts, why not Wales? What would we do about the very fragmented and diverse group of returning officers in England? As part of its UK-wide role, the commission wants to consider that as a separate issue.

Ron Gould's finding is that, subject to his reservations, which I do not criticise and which the commission accepts, the commission did a good job on public information. In several reviews, including one that the National Audit Office conducted for the Speaker's Committee of the House of Commons only in the past two years, the commission has been supported strongly for the work that it has done and the effectiveness of that work on public information.

The point in our response on following up the Gould report is that we would like the cost effectiveness of hiving off a nationally administered role to national or regional levels throughout the UK to be addressed. The issue is probably easier in Scotland, because of how the media work here, but they do not work in the same way in other parts of the UK. What you detect from our full response rather than our Scotland-specific summary is that the proposal gives rise to questions in our mind when we consider whether the solution that might work in Scotland would work in the rest of the UK.

Will you clarify whether it is your view that not all Ron Gould's recommendations need to be implemented for us to address the problems that were highlighted in the elections? You have already expressed reservations about some of them, so it would be reasonable to say that you do not necessarily agree that they should all be implemented.

I mentioned a particular point on possible solutions to the way parties and party descriptions appear on the ballot paper. Ron Gould said that party descriptions and the order in which party and candidate names appear could be addressed and that alphabetisation, random draws or rotations could be options. I am not convinced that we need to implement all those to deal with the problems that we experienced.

The Gould report identifies some key issues to do with legislation, timing, planning, co-ordination, accountability and supplier management, to name the top few. Those need to be addressed. Ron Gould develops a range of options on them in the body of his report and concludes with a particular set of recommendations, which is one way of solving the problems. It is not necessarily the only way and there are other options in the report. The important thing is that whatever is done is tested against the problems that he identified and that we decide how well the solutions will work.

Would it be reasonable to say that, from your perspective, the issue was not where the legislation emanated from but that you required administrative coherence in its implementation? That is to say that the argument that we need a chief—one responsible person in Scotland—is entirely different from determining who legislates on how elections are run.

At the beginning of the meeting, we distinguished between the management of elections and legislation on elections, which is not within the commission's scope.

I will ask about fairness in the ballot. Do you agree that there is a distinction between what political parties or individual politicians regard as fair and what the voters experience as unfair? I will give you an example that Kenny Gibson used. If a supporter of a party that put three candidates on the ballot paper made it clear that they wanted to vote for their party alone by indicating their support for all three candidates, their ballot paper would be discounted, but the vote of a supporter of a party that put only one candidate on the paper, who therefore put only one X on the paper, would be counted. Is that right?

Yes.

Would it be reasonable to say that, as a consequence, what was perceived as fair—putting candidates on the ballot paper in alphabetical order—disadvantaged strong supporters of particular parties that had more than one candidate?

You are touching on the sensitivity of the ballot and why it is important that it is considered against the backdrop of voter interest.

There was a political argument about that as the legislation came through the Parliament. The parliamentary committee took a decision that candidates ought to be presented by alphabetical order. In effect, that decision meant that less information was provided on the ballot paper for people who were interrogating it than if it had been grouped by party, for example. We test the system by how fair the outcome is in terms of party choice and judge an electoral system to be fair if the share of seats reflects the share of the vote, but the ballot paper was not grouped by party, which would have given voters the information that would have allowed them to express their political preferences.

That, too, emphasises the importance of thorough research on any ballot paper development or design that takes the voter interest and experience into account. It also emphasises the importance of an independent report going before the legislative body to ensure that the public interest is put before elected members when they take the decision.

But do you accept my premise that the decision by the parliamentary committee not to group the candidates under political parties disadvantaged people who were voting for a political party that had more than one candidate? Such parties must have had a disproportionate number of discounted votes.

That has to be set against all the factors and issues. I would not want to be locked into saying that one element was absolutely secure.

I want to pursue the point about alphabetical order. I agree with the reservations that you expressed. As someone who has worked with people with literacy difficulties, I know that the alphabet gives people a tool with which to interrogate a piece of paper. If we were going to randomise the ballot paper, it could be done by political party, but it would need to be done in alphabetical order. We would have to find some way to make it easier for people to access the ballot paper.

I want to pursue two points about the disproportionate number of people in disadvantaged communities who spoiled their ballot papers. The turnout was lower, and there is evidence that the level of spoiled ballot papers was higher in some of the poorest communities in Scotland. What do you think we should do to address that?

I will give you two examples. First, what do we do to the ballot paper to make it easy and accessible to people who face certain challenges? Secondly, what do we do with our information officers? There is evidence that some information officers gave wrong information or were entirely hands-off in case they were seen to be interfering. We presented people with a complicated ballot paper and we disfranchised them by not supporting them to work their way through it. Voters were told that even if they thought they had made a mistake, they could not do it again. All the information programmes in the world are a waste of money if people do not pay attention to them. Surely the investment should be in proactive officers inside the polling station giving people information.

That is a fair and valid point. The answer comes round to careful selection of information officers with the qualities that we are looking for, to intensive training to ensure that those qualities are carried through, and to monitoring to see how effective that training has been.

The other point was about the areas of social deprivation, which the University of Glasgow report picked out as an issue. That brings us back to the theme of Ron Gould's report—putting the voter first. We have to address the issues, not as a generality across all electors but on specific voter education and, ultimately, some of the underlying issues of social deprivation. It is not an issue that we can ignore.

In taking specific action, the starting point is to try not to present people with a ballot paper of the complexity of the one that was used in May 2007. As I mentioned before, we are going to come up with standards for the design of ballot papers. The Electoral Commission has a good track record on that since our establishment seven years ago. Those standards will certainly include some on issues related to literacy, disadvantaged communities and accessibility. I am sure that those issues will feature, which will be new for some people who have not taken them sufficiently into account in the past. That is the starting point.

We have already talked about the need to ensure that information officers are well briefed and that what they can and cannot do is clear. As Sir Neil McIntosh said, and as Ron Gould's report points out strongly, it is important that, whoever ultimately takes the decision about the design of ballot papers, the information on which the decision is based and the factors that are taken into account are clear to the electorate—accepting that there may have to be compromises and that there will be no perfect solution.

I should say that I am talking about additional work. There is no realistic suggestion that the information campaign carried out in the run-up to the election was a waste of money. There were high levels of awareness of the campaign, which is recorded in the appendices to the Gould report, and there was a significant increase among those surveyed in their understanding of the voting system and what they needed to do. We now have to address what needs to be done for the people whom the campaign did not reach. Inevitably, for various reasons, there will be a significant number of people in that category. We will have to consider what can be done at the beginning, to make the process as simple as possible, and at the end of the process, when people turn up to vote. We all need to do better to address those issues.

Yes. You accept, of course, the Gould report's criticism around information staff and the involvement of the Electoral Commission. We are all culpable, but there is a bit of reluctance to accept that fully.

There is no reluctance on the commission's part to say that we were part of that.

I just detected a bit of that in Mr Wardle's answer—just now and previously. We have all found it difficult to apologise to the voters who were let down. We were all culpable in bringing the situation about.

Precisely.

"It wisnae me."

Even round the table, people are saying that it was nothing to do with them.

Johann Lamont made an important point about literacy and voting intentions. There are some aspects of what she said that I disagree with, but she made an important point about party groupings on the ballot paper. Although I disagree with what she said about that, I am open minded on the idea of a lottery or randomising the order in some way. Johann Lamont and I both have a party-political perspective on that, as well as a fair and open-minded perspective, I hope. That is the nature of politics. There was a danger that people perceived that ministers were acting with party-political motivation. That is a different aspect of the issue altogether.

I am interested in some of the evidence that you have given the committee. Your report says:

"It is essential that there should be a formal consultation process"

for

"future ballot paper design".

It goes on to say:

"the Commission will publish a plan for the development of a set of UK-wide standards on the accessibility, design and usability of ballot papers and associated stationery by 30 September 2008."

If the research and the consultation are of a very high standard, are persuasive and do not just use four clusters of 25 ballot papers—as was the case for the previous elections—the conclusions could be so overwhelming that they supersede any party-political advantage or any question of ministers trying to exercise an advantage.

I want to home in on the date of 30 September 2008. Will you publish a plan and then consult after September 2008? Will consultation come before that? Your report mentions a UK-wide perspective and "fragmentation of the legislation". There is a feeling that if there are going to be local government elections in Scotland in 2012, which is a real possibility, Scotland has to act now to stop fragmentation in all aspects of the political process, not just for Scottish parliamentary elections but for local authority elections. The timescale is important, and I would like to know more about that. Is the rest of the UK geared up? Scotland cannot wait for the rest of the UK to fall in line; we have to find Scottish solutions for elections in Scotland.

The commission entirely agrees that we must not let other parts of the UK get in the way of implementing the important recommendations relating to Scotland. Our review was of the elections in Scotland; the recommendations and solutions need to be found in a Scottish context. The review has raised questions for the rest of the UK, but we are determined that the rest of the UK should not get in the way of the implementation of Scottish solutions.

We are conscious of the need to ensure that any recommendations that we make are implementable in time for the next likely set of local government and Scottish parliamentary elections. It might be helpful if the commission gives the committee a note on the details that we have about the timetable and the work plan for the standards for ballot paper design. However, I am sure that we want to consult all the stakeholders in Scotland—of which the committee is one.

And that will be a far more robust piece of research than one involving 100 sample ballot papers?

I am sure that it will involve more than 100 ballot papers.

I am sure we agree that time pressure should not effectively exclude voters and that any conclusions that are reached should not be taken by us politicians—the usual suspects. Do you agree—to go back to my original question—that if we decoupled the elections, we would resolve many of the pressures and problems with management, the problem of having two, different, ballot papers on the one day and so on?

Yes. At the root of this is the matter of combination, speaking candidly. If we remove that factor, we remove an area of potential confusion and added complexity, which has presented a difficulty. Nevertheless, even if you do that, it does not take away the need for effective co-ordination, management, ballot design and staff training. There are a range of issues of consistency there, which political parties have been concerned about over a number of years. All those issues remain.

I know you are not doing this, but it would not be in Scotland's interests to take the simple view that, if we decombine the elections, that will be it, and we can sit back and relax because there will not be an issue any more. Elections are part of our lives, and the various issues must be addressed.

It was interesting speaking with Mr Gould. We were speaking about elections becoming part of our lives, and we asked him about voter fatigue. His response about how that might be handled referred to compulsory voting. Have you any view on compulsory voting?

This is a fascinating point at which to draw to a close. Could I perhaps come back with an answer to that in February? Compulsory voting is not part of the culture of United Kingdom elections, and we would have to be very careful about building it into the equation. Nevertheless, as with everything else, it should not be forgotten about. However, I would not wish to go any further on that point now.

Some would say that STV is not part of the culture of British voting, but we have it.

Good thing too.

It is good for some.

Thank you very much for your attendance today. That was a good session.

Meeting suspended.

On resuming—