Support for an AV referendum high... but vulnerable

Questions asking about support for referendums are tricky. Almost whatever you ask, people support referendums. I suppose it's not really surprising - asking if people support a referendum on X is pretty much the equivalent of asking "Should you get a say on X, or should politicians make the decisions for you?" As far as I recall, I have never seen any poll showing people opposed a referendum on anything. To some extent this is a reflection of reality - people like the idea of more democracy, but I suspect it is wrong to take from this that referendums are always a political plus that can't become a vulnerability for a government.

Back in March 2009 YouGov did a poll on the Scottish referendum that tried to open out attitudes to a referendum a bit, seperating the principle of the referendum from the practicalities of it. At the time, YouGov found people supported a referendum on Scottish independence, but didn't think it was appropriate that year.

YouGov carried out a similar set of questions for tomorrow's Sun - on the principle of a referendum on electoral reform, there remains overwhelming support - 69% support a referendum being held, only 12% oppose it.

The support is soft though, asked if they thought it was appropriate given it would likely cost £80 million pounds at a time when the government was cutting expenditure (the sort of argument that Conservative party and media critics of the referendum are making) only 35% thought it was, 46% thought it wasn't. In other words, the government are vulnerable to criticism along these lines (assuming they get traction of course, given that all three party leaderships support the referendum). Support for the referendum is very high, but is not necessarily very robust.

YouGov also have the voting intention from the daily polling this week up on the site, the most recent figures from Thursday are CON 42%, LAB 36%, LDEM 15% - they have been very similar all week, though the continuing downwards drift of the Lib Dems is worthy of note.

UPDATE: Fun point from James Graham - the £80 million cost originally came, I think, from Prof Robert Hazell and was quoted a lot in the last Parliament when the Labour government's bill went through Parliament, the current government have not, as far as I'm aware, commented upon the likely cost so it is unclear how cost relates to whether it is held in conjunction with other elections or not, but it's logical that it will be more to do it standalone, less to combine it. With that in mind, the other controversial point about the referendum plan is whether it is right to do it on the same day as the elections next year - on that point, these findings are an argument in favour of the government's position, since they are saving money by combining polls!