Will Catalunya vote for independence?

In my opinion the two most significant political events this year are going to be the independence referendums in southern Kurdistan on 25 September and Catalunya on 1 October.

The reason I'm writing this post is because what I've seen in the media about Catalunya is only a part of the truth, and isn't providing us with an accurate picture of the way Catalans are likely to vote. For example, this is what was reported by the BBC yesterday:

How strong is the appetite for secession?

"It is hard to say," says the political scientist [Professor Arias-Maldonado] from Málaga University. "According to polls, secessionists are now around 41% of Catalans - numbers have been going down for some time. Around 49% are against it.

"These data come from the Catalan public polling body. How will the terrorist attack affect this situation? Who knows? But my bet is - not very much and if it does, it will reinforce the unionist side."

BBC - 20 August 2017

To get a more accurate picture, I think it's worth showing a few graphics from the CEO poll itself, which can be downloaded from here:


The first graphic does indeed show 41.1% support independence and 49.4% don't. The graph at the bottom shows that the figures were just about equal a year ago, but in fact the figures haven't really changed all that much in the last three years, fluctuating in the range between 40 and 50%.


The second graphic shows the breakdown by party. Junts pel Sí and CUP supporters almost entirely in favour; the PSC, PP and Cs supporters almost entirely against; and CSQP supporters split (their preferred option is a federal Spain, but that isn't on offer, which explains why they are more evenly divided).


But, as we should all know, even though people will express an opinion in an opinion poll, this doesn't always translate into actual votes at the ballot box. Even with the unusually high turnout in the Scottish Independence referendum of 2014, only 84.6% of the electorate actually voted. So it would have been possible to win that vote if only 42.4% of the total electorate had voted Yes.

According to the CEO poll, the turnout in the Catalan referendum would be a still-respectable 67.5% (roughly equivalent to recent Westminster general elections: 65.1% in 2010, 66.4% in 2015 and 68.8% in 2017) and that Yes would win a very substantial victory by a margin of nearly 25%, as shown in the graphic below:


I should perhaps explain that a "blank vote" is a valid vote for "none of the above" (a number of countries have long had this option, and in my opinion it should be available here too) and a "null vote" is the equivalent of a spoilt ballot paper here.

The reason for the apparent discrepancy is that those who want independence are very much more determined to get out and vote for it that those who don't. Of the supporters of the two parties in favour of independence, only a tiny percentage will abstain (3.1% and 1.9%) but the abstention figures for supporters of the three unionist parties are 39%, 28.8% and 28.3%, as shown in this final graphic:


Now of course, opinions could change between now and the referendum. But it's now less than six weeks away, and I don't think things will change much.

This means that after the referendum things will get very messy, with different people claiming different things. We can be sure that the leaders of the Catalan Government will point to the substantial margin of victory as justification for declaring independence.

We can be equally sure that the Spanish Government will say (as they have said all along) that the vote is illegal and unconstitutional. But I have no doubt that they will also claim that the reason Yes won was because those who abstained did so not because of the level of political apathy common to almost all western democracies, but because of a principled refusal to take part in an illegal and unconstitutional vote. That will be true, but only to a small extent, and we need to understand why that argument is a fallacy.

True, the unionist parties may not recognize the legitimacy of the vote; but the insurmountable problem they face is that a majority of their supporters clearly do recognize the legitimacy of the vote. As things stand at present, they are caught between a rock and a hard place. If a majority of their supporters were to boycott the election, they might well be able to claim that Yes only won because of the low turnout. But if 64.2% of PSC supporters, 55.9% of Cuitadens supporters and 66.1% of PP supporters defy their party leaders and go out to vote in the referendum (even if most of them vote No) then Mariano Rajoy—Prime minister of Spain and leader of the PP—is going to look pretty silly if he claims that the result is not democratic ... two-thirds of his own party supporters in Catalunya will not have agreed with him.

I've no doubt that he'll try it anyway, because the official referendum results won't be broken down by party allegiance. So he'll think that he'll be able to get away with it because people won't understand the underlying maths, and he'll hope that influential world leaders will commit themselves to rejecting the result on the basis of a specious argument before their research assistants get round to doing that maths.


Forewarned is forearmed. Those of us who want to see Catalunya become independent need to make the case that democracy, the will of the people expressed at the ballot box, is more important than the niceties of the Spanish constitution or any political inconvenience that a vote for independence might cause for other countries.

Bookmark and Share


Anonymous said...

The Leave vote in the Brexit referendum was 17,410,742 out of a total electorate of 46,500,001 = 37.4%

BoiCymraeg said...

Don't the poll figures take into account the likelihood a given person is to vote? This is certainly how all respectable UK political pollsters operate, that is, by weighing individuals' responses by their likelihood to vote so as to compensate for the fact that supporters for one side may be more dedicated than another. I don't know whether your 41%/49% figures are weighed in this respect or not, but I would expect them to be.

Michael Haggett said...

A good illustration 22:4. If you want an even lower figure, take our referendum of 1997. Yes won with 559,419 votes out of a total electorate of 2,222,533, which is 25.2%.


I'm not entirely familiar with the way opinion polls operate over there, BoiCymraeg. What I've done is highlight the answers to questions 31 and 81. It looks like the pattern over there is to ask lots and lots of questions, and try and glean an accurate overall picture from comparing all the answers; whereas over here the pattern is to ask far fewer questions, but try and compensate for that by adding various weightings to get accurate answers.

So the 41/49 figures in question 31 don't seem to be weighted for likelihood to vote, but likelihood to vote is specifically asked about to get the 62/38 figures in question 81.

All polls are just snapshots of opinion at the time the questions are asked, rather than predictions of the actual vote on the day of the referendum. But the Yes margin is so big that it's very hard to imagine opinion changing enough between now and then to overturn a Yes majority.

In fact the best hope of overturning it would be for the unionist parties (the PP, PSC and Cs) to urge their supporters to go out and vote. But in doing that, they would acknowledge the democratic legitimacy of the vote and, by implication, have to accept the result. That's why I said they were caught between a rock and a hard place.

Post a Comment