Would being more anti-immigration win votes?

One thing we've seen over the last few months is pressure groups commissioning helpful polls showing that, lo and behold, adopting the policy that they are campaigning for would be really fantastic for the electoral prospects of whatever party they are trying to twist to their end. So, the Electoral Reform Society has commissioned polling that showed how backing a referendum on PR would be really good for Labour's electoral hopes. The League Against Cruel Sports has commissioned polling that shows that fox hunting is a really important issue that will affect loads of votes, and over the weekend Migration Watch published polling that shows that loads of votes are there for the Conservatives to take if they are more anti-immigration.

The finding that immigration was a major issue to voters was perfectly legitimate (and done using an unprompted question like MORI do - by far the best way of doing it). However, I'd recommend far more caution about the question showing 44% of people saying they are more likely to vote Tory if they adopted a particular immigration policy, or the general conclusion that it would be a politically advantageous thing for the Tories to do. I've been dismissive of questions asking "would you be more or less likely to vote x if they endorsed policy z" before, and I'll do it again here. People tend to use the question to express support or opposition to the policy in question, regardless of whether they would actually switch their vote. More or less likely is a very low bar. Focusing on a single issue gives it artificial prominence, and often people are already voting for the party they say they would be more likely to vote for, or would never vote for the party anyway. All in all, I wouldn't give it much credence.

The limitations of this particular question aside though, would going hard on immigration actually help the Conservatives? In many ways it is the Conservative equivalent to Labour's quandary over whether to soak the rich. In both cases polls show the policies themselves to be very popular, the question is whether it would have a negative effect upon a party's image.

In evaluating a policy there are actually at least three questions. The first is the straightforward one we are all used to seeing in polls - do people agree with it or not. In the case of immigration polls are consistent in showing that the majority of people want harsher restrictions upon immigration.

The second question is the salience of the issue, there are things that everyone agrees with, but that no one thinks is of much importance. For example, I linked to some polling for the League Against Cruel Sports earlier - a substantial majority of people think fox hunting is cruel, but aside from small numbers of very committed people at either extreme, for most people it is not a major issue at elections when compared to health, education or the economy: it is a low salience issue. This is sometimes slightly harder to measure - questions asking if people are worried about an issue give it false prominence. The correct method is the one used in this poll - an unprompted question on what issues are important, and it shows immigration is indeed a very salient issue, in this particular poll second only to the economy.

The third issue is the most nebulous and hard to measure. How do the issues and policies a party puts forward influence their broader party image? Does talking about a particular issue, or putting forward a particular policy make a party seem forward-looking, or caring, or negative, or bigoted or so on. This was the Conservative strategy for much of David Cameron's early leadership. His early emphasis on the environment was not something that would directly win votes, climate change ranks a long way behind more concrete things like crime, education and taxes when people come to vote, and those who do prioritise it are probably the sort of people least likely to vote Tory. However, the Conservatives no doubt believed that by championing it Cameron made the Conservative party look more moderate, caring, modern and so on.

We don't really have much direct polling evidence to judge this by, but the potential risk can easily be seen by looking back at 2005 and the exhaustive polling Michael Ashcroft privately commissioned in the run up to the election. Polling then was just as positive about a harsh immigration message as it is now. Questions showed anti-immigration policies met with overwhelming support and it was consistently rated as a very important issue. Presumably this influenced the Conservatives when choosing their 2005 strategy.

Every day during the campaign Populus asked 250 people what they recalled the Conservative party saying recently. Over 30% of people recalled the Conservative message on immigration after Michael Howard announcement at the end of January that the Conservatives would impose an annual limit on immigration. Apart from 3 days after the council tax announcement, it remained the most recalled message when people were asked about the Conservative party for the rest of January and February. In March it remained amongst the most recalled issues, but was topped for a while by opposing anti-terrorism legislation, sacking Howard Flight and cracking down on travellers. Once the election was actually called, on every single day throughout the whole of the campaign the most recalled Conservative message was anti-immigration. Immigration is indeed a very salient issue, and it completely swamped Conservative messages on health, taxes, policing and so on. At the end of the campaign Populus asked people to characterise this Conservative campaign which people had recalled as being almost wholly about immigration. The most popular options were negative and aggressive.

The polling now on immigration is almost identical to the immigration polling in 2004 and 2005. Back then one could be forgiven for coming to the conclusion that it made sense for the Conservatives to emphasise it. If nothing else, the 2005 campaign tested that hypothesis to its limit, and found it wanting. Put it this way. In 2004 a YouGov poll for the Economist found that 44% of respondents said they would be more likely to vote Conservative if they had "a harder policy on immigration", compared to only 6% who said they would be less likely - almost exactly the same figures as the YouGov polling for Migrationwatch today. Subsequently the Conservatives did indeed go into the 2005 election with a campaign that, in the eyes of the public at least, was utterly dominated by the message of a harsher policy on immigration. This produced no obvious gain in the Conservative vote at all - they ended up with almost exactly the same level of support as they had when YouGov asked the question back in September 2004.