Parliamentary Voting System and Constituencies Bill

Lord Tomlinson Excerpts

Monday 7th February 2011 Lords Chamber

[Lord Campbell-Savours]

I had hoped to speak at some length on this amendment on Report but understand that agreements have been made on Part 1 so I do not intend to delay the debate. Perhaps I can start by explaining why I have been pushing SV during the course of this legislation.

The supplementary vote is a variant on the alternative vote. It is one of the three systems which we have discussed at length in Committee. We have the Australian system, the Queensland system, and the supplementary vote system. It is not my favourite system for electoral reform—my favourite is a PR list system or an AMS system—but is a compromise. If you have two options on the agenda—alternative vote, Queensland, or alternative vote, supplementary vote—then I will always pick the supplementary vote. The reason I want to present the preamble to my case today on that basis is that I intend to criticise some aspects of SV along with AV. I am criticising a family of systems which generally come under the alternative vote.

To get the preliminaries out of the way, the supplementary vote is already used in 13 cities in the United Kingdom. It is used in the mayoral elections, and it was used in the London election to elect Boris Johnson. Many people think when they walk into the polling booth in London and vote for a mayor for London they are voting under an AV system. They are not. They are voting under a particular system within the family of AV systems, the supplementary vote, which is not what is on offer in this legislation.

The key question we have to ask about those 13 mayoral elections is whether the supplementary vote changes election results as against a first past the post system. It has done on four occasions, where the second-placed candidate on the second count has won the seat and where the first-placed candidate on the first count has, therefore, lost. In that sense, therefore, it can influence election results.

Furthermore, the supplementary vote was the recommendation of the Plant commission, which was established by the Labour Party in 1990 to evaluate different electoral systems. In Committee I read on to the record a part of the Plant commission’s report and its recommendations.

The next debate that took place on the supplementary vote took place in 1998, when London mayoral elections were established in the system in legislation. Nick Raynsford, who was then the Minister, in conjunction with many outside bodies which lobbied him on behalf of the various systems, decided that the supplementary vote was the appropriate system. It is a used and tried system within the United Kingdom.

I now want to move the debate from pushing my system within the family of AV to another argument. Within the family of AV systems there is a problem which has never been debated in Parliament. To know what the problem is you have to look at a paper produced immediately after the landslide victory for the Labour Party in the 1997 election. It was called Remodelling the 1997 General Election: How Britain Would Have Voted Under Alternative Electoral Systems by Professor Patrick Dunleavy, Helen Margetts, Brendan O’Duffy and Stuart Weir. This is the only piece of good, clear evidence of what happens when you introduce alternative vote systems within the United Kingdom. Again, however, it is an extrapolation.

I could spend an hour quoting from the paper but I have taken out the salient paragraphs which should influence opinion. The writers simulated what would happen under AV under the landslide victory for Labour in 1997. They said:

“Our simulation approach developed over the two 1990s elections seeks to get as close as possible to how a new system might work via several innovations … asking survey respondents to complete alternative ballots for the rival systems, immediately after they have voted in a general election”.

In other words, after they voted in a general election they then asked them questions. The paper continued:

“In 1997 ICM Research interviewed a sample of nearly 8,447 people across 18 regions of Britain for the project, achieving a response rate of 82 per cent”.

That is a very substantial sample, asking questions about how people would have voted under AV in 1997. They combined,

“regional responses for each type of voter and information from the general election on first preferences to extrapolate how second and subsequent preferences would be structured under the alternative voting systems at the level of local constituencies”.

The authors then took Queensland AV and SV and found:

“To simulate an SV outcome” —

Remember we are talking about a sample of 8,500—

“we looked at all 301 constituencies where the winning MP in 1997 had only plurality support, identifying the top two candidates who would go to the second stage of the count, and also those candidates who would be eliminated … The outcomes were dramatic”.

Tory MPs should read this stuff because it then says:

“Across the country as a whole the Conservatives would have lost”—

a further—

“55 seats, cutting their representation in Parliament to just 110 MPs”.

[Lord Tomlinson]

What’s wrong?

[Lord Campbell-Savours]

My noble friend might ask, “What’s wrong?”, and there may well be people in the Labour Party all over the country repeating, “Yes, what’s wrong?”. I will tell you what is wrong: we know it is wrong. We know if we were being reasonable, we could never have cut back the Conservative Party to 110 seats in 1997. It would have been a ludicrous result, producing, as the paper states,

“less than 19 per cent of seats in Britain compared with their vote share of 31.4 per cent”.

So here we have it. This system, we are told, is about fair votes; it is about somehow matching the number of seats with the votes cast in a general election, turning out in 1997, in the Labour landslide victory, 19 per cent of seats in Britain compared with their vote share of 31.4 per cent. That is a huge difference, and it is wrong that we should be introducing a system that potentially can lead to results on that scale.

“Such an outcome would be the most severe under-representation of the Tories in British history. The biggest reduction in Conservative seats would occur in the south west”.

It might well be there would be those who would argue, “Well, they are only Conservative seats that are being lost”, but it works both ways because AV exaggerates results and swings. You can get huge swings against a party which could just as well be the Labour Party and we, too, could be reduced to a rump. The Conservatives have simply failed to understand the dangers inherent in the system they want to introduce.

The paper goes on to point out:

“Under SV the Liberal Democrats would have won another 38 seats on top of their existing 46”.

We now know why they want to introduce the system. It clearly distorts. Then what does it say?

“Under SV Labour would also have gained 17 more seats, buoyed up by extra transfers from supports of eliminated Liberal Democrats, further boosting their already disproportionate majority, giving them over 68 per cent of British seats in Parliament on the basis of 44 per cent of the vote”.

This is this super system that we are introducing. This is the system we are told is fair votes. On the basis of the 1997 general election, the landslide victory for Labour, we would have won over 440 seats. What a ludicrous system. What a ludicrous proposition has been put before Parliament.

I go back to the amendment and the question posed by the noble Lord, Lord Newton of Braintree. People have not done their homework, and something needs to be done about that. We need a referendum question that invites people to say no to first past the post. Then let us get the inquiries established because the homework has not been done.

On the classic Queensland AV, the authors go on to say,

“We assessed AV’s impacts by examining whether the tiny differences in second preferences from the SV ballot would have changed any of the SV simulation outcomes in any constituency but we could not identify any such cases”.

The proposition before the House is that we do something we should not be doing. The Tory Government should stop this, and stop it now. We are on Report. They should go away and come back at Third Reading having fully considered the implications of the Dunleavy work from 1997. I know that the Minister will get up and say, “It doesn’t matter. It’s all gone through. It was approved by the House of Commons”, but they did not know what they were doing. They did not understand the implications of this system. We are dealing now with a major change in the constitutional arrangements of the United Kingdom. If we produce exaggerated results that would have given Labour 444 in 1997 and a massive majority much larger than we actually had, we are making a major error, and I appeal to the Government to think again before it is too late.