The Prime Minister’s proposals for electoral reform are too limited and too late. But despite that I support them.
Not only because I believe the introduction of Alternative Vote is a key step on the way towards the introduction of a genuinely proportion (and more psephologically interesting) electoral system but also because the rabid response of the right has convinced me that Brown must be on to something. This has ranged from Cameron’s none-too-subtle barbs about rigging the electoral system at PMQs, through to the ill-advised playing of the Mugabe card by Reading East Tory Rob Wilson MP*.
As I like to give all politicians the benefit of the doubt (stop sniggering at the back, there) I’m prepared to be convinced that if we weren’t 100 days from a general election then the response would be a bit more considered. Because if this is an attempt to rig the electoral system, it would be an astoundingly cack-handed way of doing it. Alternative Vote makes no significant amendment to the UK’s constitutional settlement, it is highly unlikely to break the dominance of the two major parties and will leave the vast majority of seats in the same hands as currently, albeit with a little more legitimacy for the sitting MP.
Whether the AV transition is likely to happen this time round or not is a moot point. But I remain convinced in some degree of historical inevitability of electoral reform in the UK, and AV seems a very likely first step whenever it comes around. So what would it mean? I don’t have the time or inclination to go through each of the UK’s 650 constituencies, but I thought I’d have a run through the London region: not only because it’s my home, but also because we have some experience of this sort of system. The Supplementary Vote system used for electing the Mayor is a hybridised form of AV, in which the voter is limited to expressing two preferences, rather than being able to number all the way down the ballot paper. So there is a bit of evidence, albeit somewhat unwieldy, as to how voters might react to a preferential system.
At the next general election, London will elect 72 MPs, on new constituency boundaries. If the 2005 voting patterns were exactly replicated (using the notional results supplied by Anthony Wells), 27 of these – almost 38% – would produce results where the winning candidate took over 50% of first preference votes and would therefore be elected on the first round under AV (18 Labour, 7 Conservative, 2 Liberal Democrat). In one case (Orpington), this would be with a margin over 50% of just 22 votes, whilst in another case (Camberwell & Peckham) the winning candidate would be more than 12% clear of the threshold. The map below shows these seats, and also the 19 seats where the winning party in 2005 took between 45% and 50% of the vote, with a clear margin (10%+) over their nearest competitors (12 Labour, 6 Conservative, 1 Liberal Democrat). Were such results to be replicated, it would be almost impossible for the leading party not to pick up enough transfers in the first few rounds of redistributions to not be guaranteed of victory.
None of these figures are a guarantor of safety, it should be pointed out, particularly not in the bottom category. I consider at least three of the Labour seats in this category (Dagenham & Rainham, Ealing North, Feltham & Heston) to be looking decidedly shaky at the coming election. Similarly, the precursors of two of the Tory seats in this category were won by Labour in the 1997 landslide. However, the key point is that the sort of swing needed to dislodge the occupants of these seats under plurality rule** would equally dislodge them under AV. I am contending, therefore that in almost 64% of London’s seats, the introduction of AV would never, or almost never, have any effect on the result.
What, then, of the remaining 26 seats – those left white on the map above? 14 of these are Labour held (or notionally held), 7 Conservative, 4 Lib Dem and 1 Respect. Of these, I am immediately throwing one Conservative seat (Uxbridge & South Ruislip) into the “electoral system won’t make a difference category”, as to overcome the 13.5% Tory lead, Labour would be having to take an unfeasably large proportion of Liberal Democrat second preferences in an area where Lib Dem voters don’t tend to be of the left leaning variety.
Let’s take a skip through the other 25. One of the tools we can use, particularly where the Liberal Democrat vote could be key, is the break down of second preferences in the 2008 Mayoral elections. The London Liberal Democrat voter base is remarkably varied – running from the social liberals of Hampstead, who might well place Labour as their second choice (perhaps after the Greens); the caring capitalists of Kingston, where a Conservative second preference seems more likely, through to Simon Hughes’ working class base in Bermondsey, whose second preferences could just as easily split to the BNP as to Labour. It is impossible, therefore, to make a city-wide generalisation as to where the bulk of Lib Dem second preferences would go. The overall breakdown of Paddick’s second preferences tells us nothing much – they split 31% to Livingstone, 30% to Johnson, 39% to other candidates or left blank. However, I believe that a very rough idea of how Lib Dem voters in a particular constituency are likely to split can be obtained by looking at the split of second preferences obtained by Livingstone and Johnson, given that a high proportion of these will be from Liberal Democrat voters. Similarly, and importantly in a number of seats, the Green vote is hard to pin down. My working assumption is that a bulk of the second preferences will move towards the Lib Dems in the first instance, or potentially directly to Labour, depending in part on individual candidates. It seems likely, though, that in Labour-Conservative fights, a sizeable portion of the Green vote would eventually end up in the Labour column (Sian Berry’s second preferences split 47% Livingstone, 14% Johnson, 39% other/blank) – or perhaps I’m just a bit too hyped up about the value of ‘progressive coalitions’ after last Saturday’s conference.
Battersea Currently Labour held, majority 0.8% over Con This is an ultra-marginal, which could well become more favourable for Labour under AV. Labour was 9.4% short of the threshold, with 18.8% worth of Lib Dem and Green votes to come into play and every indication that these would be weighted towards Labour in the redistributions: there was a 56.3:43.7 Livingstone-Johnson split on second preferences here.
Bethnal Green & Bow Currently Respect held, majority 3.4% over Lab This seat would be fascinating to watch under AV (as it is already!). Both of the main contenders are a long way short of the threshold and it’s hard to judge how the preferences of Lib Dem, Conservative and Green voters would break down. Mayoral second preferences show strong support for Livingstone, but there was no Respect candidate for Mayor, so we can’t know if Lib Dem or Green voters would number Labour above or below Respect. The identity of the Respect candidate and whether Conservative voters choose to number Labour above Respect in a perceived anti-extremist gesture could decide this.
Brentford & Isleworth Currently Labour held, majority 8.0% over Con An intriguing seat which could almost be considered a three-way marginal (38.8/30.8/23.3). Even if they can’t quite reach contention, the size of the Lib Dem share is easily large enough to take the Tories to victory if their second preferences lent right. However, Livingstone had an 8% lead over Johnson on second preferences here, suggesting that the opposite might be the case. A curve-ball here could be a Green boost as a result of Heathrow Third Runway unpopularity, possibly translating into a boost for the Lib Dems on second preferences – could it in fact create a Labour/Lib Dem run-off after all?
Carshalton & Wallington Currently Liberal Democrat held, majority 2.5% over Con Along with a number of other marginal south west London constituencies, AV would seem likely to make Carshalton & Wallington a much safer prospect for the Liberal Democrats. There is a 17.2% Labour share to milk lower preferences from, the vast bulk of which would be very likely to end up in the Lib Dem column.
Croydon Central Notionally Labour held, majority 1.2% over Con Boundary changes bring this seat marginally back into the Labour column, but even without a swing against Labour, AV would stack the odds against it holding on – firstly there is a 2.2% UKIP vote share which could be expected to swing heavily towards the Tories, thus overturning Labour’s lead. Add to this the fact that on second preferences, Johnson had a 2% lead over Livingstone which might suggest that the 13% Lib Dem block would preference the Conservatives over Labour.
Ealing Central & Acton Notionally Conservative held, majority 0.2% over Lab A fascinating 3-way marginal has been created here, with the largest parties within 3% of each other. Despite being only a handful of votes behind the Conservatives, Labour would be fearful that the large 4.9% Green share would push the third-placed Lib Dems ahead of them on redistributions, thus leading to a Con-Lib Dem run off. In such a scenario and with a huge Labour block to harvest from, the Lib Dems would seem assured of a win. If Labour can avoid this scenario and ensure a run-off with the Tories, the odds are in their favour: Livingstone led Johnson by 10% on second preferences in this constituency.
Eltham Currently Labour held, majority 8.1% over Con Despite Labour’s reasonably healthy lead in this constituency, the weighting of the smaller parties towards the right must be of concern, with a combined BNP+UKIP vote share of 5.2% – if transferred overwhelmingly to the Tories, as seems likely, this would immediately half Labour’s lead and leave them needing to pick up a high proportion of Lib Dem second preferences. Johnson’s 7% lead on second preferences here suggests that might be tricky.
Enfield North Notionally Conservative held, majority 2.3% over Lab A 4% vote share for UKIP+BNP looks likely to significantly strengthen the Conservative position here. The Lib Dems are weak here (11.5% vote share) and Livingstone’s second preference advantage relatively small (2.3%), making it hard to see enough lower preferences coming Labour’s way to overturn the Tory advantage.
Enfield Southgate Currently Conservative held, majority 2.8% over Lab The major party results for the two Enfield constituencies are disturbingly similar, but what happens beneath them is almost a mirror image. The right wing fringe parties are much weaker here (1.2%), whilst there is a small but significant Green share (2.6%). This might suggest a more hopeful outlook for Labour than in North, until you note that in this seat Johnson had a 3.8% second preference advantage over Livingstone, which doesn’t bode well for the crucial Liberal Democrat split.
Finchley & Golders Green Currently Labour held, majority <0.1% over Con With a Labour majority of just 31, this is one of the most marginal seats in the country. There’s little question that this will go Tory in 2010, but with a 16.7% Lib Dem share to be redistributed, hope might not be fully extinguished for Labour under AV. There are no real clues from the Mayoral election here, where Livingstone practically replicated the parliamentary situation, leading Johnson by just 0.2% on second preferences.
Hammersmith Currently Labour held, majority 13.5% over Con Hammersmith only just failed to get into my “electoral system wouldn’t make a difference” category – although it is by no means a shoe-in for Labour as demographic change works against them. On the current figures, however, there is little hope that the Tories could take anything like a large enough proportion of the 19.2% Lib Dem share (+3.9% for the Greens). With a whopping Livingstone lead of 15.5% on second preferences that seems a very tough call.
Hampstead & Kilburn Currently Labour held, majority 2.7% over Lib Dem With a strong Lib Dem challenge, Labour’s best hope of hanging on here under the current system is that a Tory surge splitting the opposition vote lets Glenda Jackson ride through the middle. This route is pretty much shut off under AV, with the large Conservative 24% share likely to favour Liberal Democrats on lower preferences, at least under current circumstances. It seems clear, however, that AV would end forever any hopes the Tories might have in this seat, placing it in a pretty much guaranteed Labour/Liberal Democrat alternation.
Harrow East Currently Labour held, majority 6.2% over Con The effect of AV on this seat would be heavily dependent on how the Lib Dem’s 14% vote share splits on redistribution, with a small margin in the Conservatives’ favour being enough to overhaul Labour, especially with a UKIP share of around 2% already swinging in their favour. However, the mixed demographics and electoral history of this area suggest that preferences might be pretty evenly split: Livingstone had just a 0.4% lead on second preferences here.
Hendon Currently Labour held, majority 7.6% over Con Labour’s situation here is quite similar to that in Harrow East, with a tight majority and a 14% Lib Dem share to be redistributed. However, more strongly than in Harrow East, there would seem to be good reasons to think that the Liberal Democrat second preferences would break rightward, not least the small but significant 2% bias towards Johnson in second preferences.
Holborn & St Pancras Currently Labour held, majority 16% over Lib Dem In percentage terms, this is the safest seat to be examined here. However, it is one where AV could have a devastating effect for Labour if it was unable to convince progressive voters to provide it with valuable second preferences. Labour is 6% short of the threshold, and there is a very strong 8% Green vote. If Labour can get even half of Green voters to second preference it, its position would be unassailable. The danger is if the seconds go overwhelmingly to the Liberal Democrats, boosting their vote share into the mid-30%s before the 20% Conservative share is redistributed. This could propel the Liberal Democrats to a narrow victory on the back of a weird coalition of progressive and ‘anyone but Labour’ right-wing votes. My opinion is that Labour would need to hold onto Frank Dobson, or someone of his ilk, as the candidate here for as long as possible.
Hornsey & Wood Green Currently Liberal Democrat held, majority 4% over Labour This end of Haringey appears to be solidifying behind the Liberal Democrats and it is hard to see how AV would do anything other than assist that at a Parliamentary level. Whilst the Conservative vote is very small (12.7%), it is still likely to transfer overwhelmingly into the Lib Dem column. Barring a major shift in voting behaviour, Labour’s only realistic chance here under AV would be to convince most of the Green’s 5% share to second preference them and hope that most of the Conservative vote doesn’t preference as far down as the Liberal Democrats.
Ilford North Currently Conservative held, majority 4% over Labour It seems unlikely that AV would help Labour take this seat back: the Conservative margin would be boosted by transfers from UKIP’s 2% share, and the Liberal Democrat split, whilst of less than a 14% share, is likely to be unpromising in an area such as Ilford. Johnson had a margin of 4.4% over Livingstone on second preferences here.
Islington South & Finsbury Currently Labour held, majority 1.6% over Lib Dems Holding this seat would be a very tough challenge for Labour under AV, starting as it does with less than 40% of the vote share and the Liberal Democrats effectively neck-and-neck. The two big blocks for redistribution are Greens (4.8%) and Conservatives (14.8%), both of which are likely to favour the Liberal Democrats. Labour’s best hope would be if the Liberal Democrat’s local government travails in Islington (or perhaps the incumbent’s record on environmental issues) were to convince enough Green or Conservative voters to preference Labour above them. I fear this is a forlorn hope.
Poplar & Limehouse Currently Labour held, majority 11.5% over Con Despite this looking pretty safe for Labour on paper, demographic change in the old East End and the intervention of George Galloway are pushing it towards three-way marginal status. With the Liberal Democrats clearly in fourth place, it is the division of their not-insignificant 14.8% share which would probably decide this seat under AV, given it could make all the difference as to which of the three remaining parties is knocked out. On the face of it, Livingstone’s stomping 16% lead on second preferences here suggests Labour might be in line for a significant benefit, but it has to be remembered that there was no Respect Mayoral candidate. Would a sizeable portion of Lib Dem second preferences end up with Respect? My gut feeling is that AV will help Labour here, but it does need to be wary of the danger of not picking up enough transfers in the penultimate redistribution and being knocked out at this stage. A Labour/Conservative run-off would, under AV, almost certainly lead to a Labour victory.
Putney Currently Conservative held, majority 4.9% over Lab Providing a major surge doesn’t take the Conservatives too much closer to the 50% threshold here (on 42.4% currently), AV could help Labour’s chances here. There is a not insignificant Liberal Democrat vote to be redistributed (16.3%) – and my feeling is that the Conservative already took most of the right-leaning Lib Dem vote in order to win the seat back in 2005. Despite Wandsworth’s right-wing heritage in local government terms, Livingstone racked up healthy second preference leads across the borough, including 9% here in Putney.
Richmond Park Currently Lib Dem held, majority 7.2% over Conservatives I would suspect that Zac Goldsmith, for all his claimed reforming credentials, would not be a strong advocate of AV. It certainly wouldn’t help his chance of de-seating Susan Kramer in this seat: as with much of the rest of South West London, Lib Dem hegemony would become much more entrenched with preferential voting. Labour’s vote share here is risible (9.3%), but it is enough – especially combined with the Greens’ 2.7% – to catapult Kramer over the threshold (she was only 3% short in 2005), even if the bias in preferences was only narrowly in her favour.
Sutton & Cheam Currently Lib Dem held, majority 6.7% over Conservatives I could simply say ‘See Richmond Park’, given that the 2005 vote shares are almost identical. Whilst there is no Green presence here, the Labour share is slightly higher at 11.9%, more than enough to keep the Lib Dems safely ensconced.
Tooting Currently Labour held, majority 12.2% over Conservatives All three Wandsworth seats become better prospects for Labour under AV. There is a big Lib Dem pool (19.4%) for Labour to fish in for second preferences, plus they could hope for some transfers from the Greens’ healthy 4.1%. The 2008 elections suggest that there would be no difficulty for Labour in achieving a significant bias in their favour from these transfers – Livingstone was 18.6% ahead on second preferences here.
Westminster North Currently Labour held, majority 9.1% over Conservatives This is another demographically mixed seat where the final outcome under AV would be very much dependent on how the Liberal Democrat transfers split. The Lib Dem share is strong (18.7%), but Labour would need to take over half of the second preferences to be assured of victory. Given that Livingstone was 11.7% ahead of Johnson on second preferences here, that’s not too great an ask.
Wimbledon Currently Conservative held, majority 5.5% over Labour I don’t think the wind is likely to be blowing in Labour’s direction in this seat any time soon, but there is a sizeable ‘progressive’ pool for it to fish from in happier times under AV - 18.2% for the Liberal Democrats, 3.2% for the Greens. I was surprised to find that Livingstone took a 5% lead over Johnson on second preferences here, so maybe an AV-assisted win for Labour might not be so far off here after all.
Without trying to make actual predictions (we don’t, after all, know what the political climate would be when the first AV elections were fought), this run through certainly doesn’t seem to be a very effective vote-rig in Labour’s favour, at least not in London. If anyone wins, its the Lib Dems, with a number of seats becoming much safer for them and a handful of gains from Labour looking clearly cemented in. None of this, of course, considers whether people’s voting behaviour might change under a reformed electoral system (there’s little evidence of it doing so under the current variety of electoral systems in the UK). The key fact is that the introduction of AV would force every party to think differently about how it fights marginal seats and which voters it targets. This could be a pretty dramatic change.
A final point – I haven’t mentioned the cause celebre of Barking here. Not wishing to be complacent, but I do not consider it to be a marginal seat (the real danger from the BNP is over control of Barking & Dagenham Council). Nevertheless, I would strongly recommend Next Left’s post on the issue, which demonstrates very clearly the value of preferential voting in creating a pro-democracy bloc against an extremist threat.
*As an aside on Mr Wilson, I discovered today that he has a Chief of Staff. A Chief of Staff. The President of the USA, the Prime Minister and the Mayor of London have Chiefs of Staff. Opposition Departmental Whips don’t. Or apparently, they do.
** Still can’t bring myself to call it First Past the Post after being threatened with lost marks if we did so by a university lecturer ten years ago. As he put it, “the key feature of so-called First Past the Post, is that there is no post”.