A guest article by Rod Crosby

AV – Labour’s Lock on Power?

Since the election of David Cameron as Conservative leader the polls have consistently shown a Hung Parliament as the most likely result of the next Election. Apart from the obvious practical difficulties that would follow, such an outcome would arise amid novel and explosive circumstances.

• An unelected (and unselected) Prime Minister has lost his inherited parliamentary majority at his first electoral test.

• The Conservatives are ahead of Labour by as many as 1.5 million votes UK-wide, and have a clear plurality – or even a majority – of English seats, and are ahead in English votes by a country mile.

• A unelected Scottish PM, after losing control of a devolved Scotland, is now defeated in the UK, yet attempts to remain PM on the back of Scottish Labour MPs – who already have little say in the running of Scotland, and ought not to have any say in the running of England!

Any of these outcomes alone would be unprecedented in the period of mass suffrage, but taken together would throw the UK into a full-blown constitutional crisis.

Last year, I did an analysis of the prospects of a Hung Parliament under FPTP and found it to be highly likely – and that’s still the case. However, a few weeks ago I cautioned those who might take a punt on a HP to consider what would happen if Gordon Brown introduced the Alternative Vote.

AV, sometimes called Instant Runoff Voting, is a system whereby voters rank candidates in existing constituencies, and if no candidate gets 50%, the lowest candidates are successively eliminated, and second and further voting preferences are redistributed. It is not a system of proportional representation, and while it cures some of the anomalies of First Past The Post, it adds a few new ones of its own.

A Hung Parliament is quite likely under AV, but it all depends what we mean by “likely.” If we model all plausible shares of the vote for all parties, the “area” comprising hung parliaments is about the same (perhaps even slightly larger) under AV as under FPTP.

However, if we use as our starting point the last election, or current opinion polls, then Labour and LDs would benefit significantly at the expense of the Tories, reducing dramatically the prospect of a HP.

Firstly, the introduction of AV would push the danger zone of a Hung Parliament much further away from Labour than it is now, saving about 15-20 Labour seats, due to third-placed LibDem voters breaking for Labour. Under some circumstances Labour could actually make a few net gains from the Tories, even if there was a swing against them.

Secondly, for plausible shares of the vote, the LibDems would benefit by some 15 seats, drawn mostly from the Tories. Seats like Eastleigh, Cheltenham and Romsey, which might otherwise be lost would be easily held, while the likes of Bournemouth West, Broadland, Chelmsford, Harborough, Totnes and Wells could move into the LibDem column.

Overall, calculations show that AV could seriously handicap the Tories, depending on how the critical LibDem second preferences split. The “handicap” is the increase in their 1st Preference lead the Tories would require to achieve the same result under AV as they would under FPTP.

Lab/Con lead in
LibDem 2nd Preferences (%) Con Lead Handicap
(%)
0 0.1
10 1.5
20 3
30 4
40 5.5

These figures are based on 100% transfer of 2nd preferences, and Conservative and Labour at level-pegging in seats. Specific Conservative seat totals could require higher leads. If only half of voters actually used their 2nd preferences, the Tory handicap figures would (roughly) halve also, and so on.

These handicaps may at first glance not seem large. However they must be placed against the backdrop of the existing HUGE electoral bias against the Tories. A 4% handicap, for example, would mean for the Tories to become the largest party their required lead increases from an already massive 6% to 10% – just to be 1 seat ahead of Labour! A Tory majority could require as much as a 15% lead over Labour.

If Gordon Brown has a long-range plan for a “progressive alliance”, AV would deliver a 30-60 seat bonus for Labour and LDs combined over the Tories.

AV could effectively lock the Tories out of power for the foreseeable future.

Interestingly, academic indices of “proportionality” show that AV is slightly more proportional than FPTP if the Tories have a modest lead in votes, although they don’t take account of a perverse rank ordering of seats and votes.

Could the apparent huge unfairness of AV be justified in any way as a reform? This is a very complex question. All that can be said is that a Tory lead in 1st preference votes could be translated into a Labour overall vote lead, after second preferences were distributed – provided the Tory lead was not too large, and the LibDem 2nd preferences continue to break for Labour. More immediately, AV could neutralize the ticking time-bomb of a FPTP hung parliament – or at least muddy the waters significantly. How the situation would be perceived would also depend on whether Labour and the LibDems were in some form of pact.

The AV-Swingometer shows likely outcomes under both FPTP and AV, and also includes a probabilistic method similar to the one which produced the famous ‘66’ majority forecast in 2005. www.titanictown.plus.com/AVswingometer.rar
I am indebted to Anthony Wells for kindly providing his notional figures upon which this spreadsheet is based.

Rod Crosby