If it was the latter then I can understand the anger of people like Neal Lawson of Compass who wanted the referendum to coincide with the next General Election. However, the problem with that strategy is that the issue of PR could become subordinate to the other main issues at the election. Then there is the other problem of double jeopardy. Either scenario will mean that two separate electoral hurdles must be overcome before electoral reform can become a reality. With four possible permutations and only one that will deliver electoral reform the odds are not good.

As Ben Raue points out, the proportionality of AV+ depends on the number of top-up seats. However, there are three other important factors that most people overlook when considering the merits of any type of electoral reform.

Firstly, you cannot judge the merits of a proposed electoral change for the House of Commons without considering how such a change interacts with the electoral process for the upper chamber (assuming we ever get one). The two bodies should be designed to work in unison with each complementing the other and compensating for the deficiencies of the other, not merely reproducing the political balance or composition of each other. A bicameral system only serves any real purpose if it introduces a separation of power or powers, and creates constitutional checks and balances.

Secondly, people wrongly assume that if electoral reform is implemented everything else will stay the same, that the constituency boundaries will be the same and the amount of tactical voting will be unchanged. Neither of these are true. If the voting system changes then the Boundary Commission will inevitably change the size and shape of constituencies in order to make the outcome of any election fairer, and people’s voting strategies will then be influenced by both of these changes.

Finally there is the elephant in the room that no politician will talk about. That elephant is the Parliament Act of 1911 and its 1949 amendments that give the House of Commons ultimate supremacy over the upper chamber. It’s justification is the greater democratic legitimacy of the elected House of Commons (HoC) over the unelected House of Lords (HoL). But if the HoL is reformed into a democratic chamber elected by PR using the list system, and the HoC remains with FPTP or merely changes to AV or AV+, how can the Parliament Act continue to be justified? The reformed HoL will have at the very least equal democratic legitimacy, and so should have equal power and status. Unfortunately, many MPs will never countenance such a transfer of power, and presumably neither will many of them countenance any real electoral reform either. In that respect these MPs demonstrate their true anti-democratic credentials.

But perhaps the most worrying aspect of Brown’s speech was the one reform that was glaring by its omission. What is the Government’s policy on House of Lords reform? Until we know that everything else will be put on hold.

It’s also worth noting that AV+ as proposed by Jenkins is not particularly proportional. Jenkins proposed electing 80-85% of MPs through constituencies. This in itself is problematic, as it is quite possible and indeed likely that parties could win a majority just in the constituency seats without getting anywhere close to a majority.

In 2005 Labour won 413 seats off 40.7% of the vote. If you assume that AV wouldn’t particularly hurt Labour at the very least, and assume the bare minimum of constituency MPs, then Labour would stil win about 330 seats. They wouldn’t need a single list seat to form a majority government on less than 41%.

Even worse than this, Jenkins proposed dividing list seats amongst 80 regions across the UK. Even Northern Ireland was given two regions.

If you look at this map: http://www.archive.official-documents.co.uk/document/cm40/4090/annex-c.htm, you see that the vast majority of top-up areas would only get a single top-up MP, producing very severe lack of proportionality.

Many, many top-up areas would elect constituency MPs solely from one party (in the last few elections, Labour, but there are other areas where this applies for the Tories). In those areas, the opposition party would get the one seat, but this wouldn’t come close to proportionality. It would be unlikely the Lib Dems would win many in any place where they came third unless the constituency seats were divided very evenly. I think the chances of the Greens or UKIP winning any seats would be virtually zero. If they were to win any, it would have a lot more to do with them standing in a region with an even split of major party constituency seats rather than actually polling particularly highly.

To see this problem of disproportionality, it’s worth looking at the 2007 Welsh Assembly election. In four of the five regions, the proportional result was not achieved due to a party getting an ‘overhang’ in the constituency seats. Unlike in Germany and New Zealand, where an overhang is solved by adding extra seats, in Wales (and Scotland too I think) this is solved by simply taking away a seat from the party that wins the last proportional seat. In Wales this cost the Conservatives three seats. Labour won more seats than its proportional share in three regions. This would have made a significant difference in the Assembly. Remember Wales elects four top-up AMs in every region, which is much more proportional than AV+ would be for the House of Commons.

If you wanted to make AV+ work you would need to increase the proportion of list seats to 30-40%. Germany has 50%, Wales has 33%, New Zealand has 40% or so. In addition you would need to increase the size of top-up regions. Germany has 16 top-up regions to elect 299 top-up seats. I think the correct magnitude would be to elect say 200 list MPs to represent the nine English regions and the three other countries (the Euro constituencies) or maybe break Scotland and some of the English regions in two.

As for an assessment of what would have happened in 2005, I did see something the other day that suggested Labour would have got an increased majority (something like 82, up from 66) but I cannot for the life of my find where I saw it. If anyone else spots it please let us know and we can hopefully do another post based around it.

Excellent article – question I have is whether anyone has done any serious analysis of how AV might have impacted on the 2005 election?

I started doing one myself but found it methodologically too complex to calculate.

Other question would be whether GB’s reference to AV allows an option to choose between AV and AV+?

As one can hardly now conceive of a majority Labour govt resulting from the next election, the only scenario I can imagine a referendum happening at all is if a minority Labour government is forced to seek Lib-Dem support – n which case they would certainly demand at least an AV+ option.

What gets tricky is turning it into a genuine multiple choice – one that allowed a choice between no change, AV, AV+ and STV would almost certainly be indecisive in that no clear majority would vote for any one of these options.

I’d also suggest that if we do by some largely undeserved miracle get a hung parliament rather than a Tory landslide then having a referendum on any form of PR under a weak minority government is hardly conducive to a strong vote a system that will be presented by the media as delivering more weak minority govts.

Lets face it this could have all been settled in 1998 but Tony Blair blew it – now we’ve no hope of seeing any change for at least another decade.

However, AV+ also gives those who vote for minor parties a stake, as the greens and even (perhaps) the SDP would get the odd seat in the top up section. The job of those minor parties elected in the top up would be to represent the interests of their voters nationally on whatever their core issues are. While being representative in this respect, and increasing the scrutiny of the House of Commons over legislation, it retains the link between the constituecy and its MP.

Surely this is a panacea?