Why are views about immigration different in Scotland?

Public opinion isn't formed in a vacuum

Two recent reports on immigration attitudes have highlighted an interesting divergence - while the most recent IPSOS-MORI tracker of overall British attitudes has shown evidence that voter concerns about immigration are beginning to rise again after a long post-Brexit decline, a new report from Migration Policy Scotland shows that Scottish attitudes about immigration are much more positive. While around half of British voters overall say they would like immigration levels to fall, the figure in Scotland is just 28%. A quarter of English voters now name immigration as one of the most important problems facing the country, nearly twice the 13% figure in Scotland. What might account for Scotland’s relatively relaxed and positive view about immigration?

Chart 1: Views on immigration levels in Britain and Scotland

Source: IPSOS-MORI/British Future immigration tracker August 2023; Migration Policy Scotland survey January 2023

The greater opennes of Scots to immigration is not new. Using data from the British Election Study internet panel (BESIP) we can track views about migration over the period from 2016 to 2022 in both England and Scotland. The share of voters favouring reductions in migration has declined quite substantially in both countries in recent years, part of a sustained positive shift in immigration attitudes which I have discussed on this substack and elsewhere. But the general decline has left national differences intact: at every time point the share of Scots wanting immigration reduced is about ten percentage points lower than the share of English respondents. The same patterns recur if we look at other BESIP questions on the economic and cultural impact of immigration. All the data point to the same conclusion: Scots are more positive about immigration, across the board, than the English. Why?

Chart 2: Share of English and Scottish voters who want lower immigration¹ 2016-2022

Source: British Election Study internet panel waves 7, 13, 14, 17, 20, 21,23

One possibility is that residents in the two countries are reacting to different migration experiences. The Scots may be more positive about migration because migrant flows to Scotland are smaller, and the need for migrants in Scotland, with an ageing population and a net outflow of internal migrants to the rest of the UK, may be more obvious.If this is the case, then the positive attitudes in Scotland should be matched by similarly positive attitudes in regions of England with similar features. But they are not, as chart 3 below illustrates. Scots are more favourable to immigration than English regions with similar experiences with low immigration. And English regions with low immigration have the same attitudes as English regions with high immigration, with one exception: London. London, like Scotland, tend to be much more pro-immigration. Yet London also has far higher migration inflows and migrant populations than any other part of the UK. There isn’t any clear linkage between regional levels of migration and views of migration.

Chart 3: Share wanting immigration reduced in Scotland, low migration English regions, high migration English regions and London, 2016 and 2022

Source: British Election Study internet panel wave 7 and wave 23

London’s unusual attitudes point to another possible explanation: demographics. London’s has a young, ethnically diverse population, with many graduates, and many residents who are migrants themselves or the children of migrants. These are all groups researchers have found to be more pro-migration. Perhaps Scotland’s distinctive pro-migration attitudes are also the product of distinctive demographics? Perhaps not. Pro-migration demographic groups aren’t over-represented in the Scottish population - quite the opposite. Scotland has relatively more old residents, more residents with few or no educational qualifications and relatively low migrant and ethnic minority populations. This would all suggest that, all else being equal, Scotland’s demographics should make it more migration sceptical than England, not less.

Could the geography of recent events explain the difference? The migration debate has in recent times been dominated by the arrival of irregular migrants crossing the English Channel on small boats. Scotland is a long way from the landing points, which are mainly in the South-Eastern most corner of England. Perhaps, therefore, Scots are less exercised by the “small boats” crisis as it is far away. Yet this is hard to square with the evidence too. For one thing, as we have seen, more favourable attitudes in Scotland long predate the small boats crisis. For another, attitudes to immigration do not seem to be more negative in the English regions nearer to the “small boats” landing areas. And for a third thing, both asylum seekers and resettled refugees do not remain in the areas of first arrival, but are spread around the country. Scotland hosts more asylum seekers and resettled refugees than London and most of South England, so it is just as exposed to the broader pressures from the asylum migration associted with “small boats” as much of England (though less so than Northern English regions).

Neither general migration experiences, nor demography, nor specific exposure to the “small boats” crisis seems an adequate explanation for the large and persistent divergence in Scottish and English attitudes to immigration. If the answer does not lie in actual outcomes and experiences, perhaps it lies instead in the media and political framing of the issue? Research by Geoffrey Evans and Jonathan Mellon² found that tabloid newspapers acted as a “transmission belt of concern” about immigration in the years prior to the EU referendum, with very strong correlations between levels of immigration and coverage of immigration in the tabloid newspapers, and then an even stronger linkage between coverage of immigration in tabloid papers and public expressions of concern about immigration levels. Their analysis “suggests that the vast bulk of concern over immigration is mediated by media coverage and is not the result of direct observation by voters”.³

If Evans and Mellon are right, and newspapers play a large role in shaping public concern over immigration, then three things should be true. Firstly, concern about immigration should be lower in regions where anti-immigration papers are less widely read. Secondly, the link between migration levels and concern about migration should be weaker in periods when there is less negative coverage of migration. Thirdly, the link between migration levels and concern about migration should weaken if readership of these newspapers declines.

There is some indicative evidence to support all of these propositions. Firstly, the readership of the main migration sceptical newspapers (The Daily Mail, The Daily Telegraph, The Daily Express and The Sun) is lower in both Scotland and London than in the other regions of England. And it is in Scotland and London that concern about immigration is lower. Scotland also has a distinctive media market with both Scotland only newspapers and separate Scottish editions. If Scotland specific papers, and Scottish editions of British papers, are less critical of migration this could further increase the difference in how the issue is portrayed and understood in Scotland compared with England. I don’t have any data on this, but it seems plausible given the progressive stances of some Scotland only newspapers and would be an interesting thing to look into further.

Secondly, the marked positive shift in migration attitudes between 2016 and 2021 has indeed coincided with a period when migration remained very high, but the attention of migration sceptical newspapers was largely focussed elsewhere. While I have not been able to find systematic analysis of media coverage, all of the main migration sceptical newspapers were also Leave aligned and running intensive Brexit campaigns in the years from 2016 to 2020. This left less time and space for immigration coverage. The COVID pandemic also dominated media coverage for a long period, displacing all other issues. There was therefore unusually low coverage of migration in all of these papers for a sustained period, even as migration levels remained very high by historical standards.

Chart 4: Share of BES respondents reading migration sceptical newspapers by region, 2016 and 2022

Source: BES Internet Panel waves 7 and 23. Migration sceptical papers defined as the Daily Express, the Daily Mail, the Sun and the Daily Telegraph

Thirdly, the reach of migration sceptical newspapers has declined in every part of the country during the past few years of improving migration attitudes, as chart 4 above illustrates. Around a third of voters in England outside London reported regularly reading one of these papers in 2016; by 2022 the figure had fallen to around a fifth, and even lower in London and Scotland. The shrinking readership for migration sceptical newspapers may help explain both why attitudes have become more positive in recent years, and why the recent, intensive campaigns against irregular migrants on small boats have produced a more muted public response than earlier periods of critical coverage.

The ‘small boats’ issue also underlines another important point: media are not the only actor generating information and arguments for voters though. Political parties, particularly those in government, also have the power to set the agenda, and to direct media and voter attention towards particular issues. Here, too, we see a potentially distinctive Scottish story. The Westminster government is headed by a Conservative Prime Minister who has made “Stop the Boats” one of his central political pledges, resulting in a great deal of negative political messaging and media coverage of immigration (or at least one aspect of it) in the past year. This has coincided with a sharp rise in the share of voters naming immigration as a pressing issue - a rise which has thusfar been almost exclusively found among existing supporters of the governing party running the “stop the boats” campaign. It seems a Conservative government, with assistance from Conservative allied papers, can encourage Conservative voters to worry about immigration. So far there is little evidence they can get anyone else to prioritise the issue.

Scotland is once again different. The current SNP First Minister Humza Yousaf has made clear his strong support for migrant and refugee rights and his opposition to current Westminster policies in this area, and has promised a more open citizenship model if Scotland becomes independent. This continues the approach taken by his predecessor Nicola Sturgeon, who called in 2020 for Scotland to be granted powers to set its own, more liberal, immigration policy. The two political systems are sending very different cues to their voters, and the divergence in attitudes is consistent with voters responding accordingly to those cues. Conservative voters and English voters have become more concerned about immigration as a problem. Labour voters and Scottish voters remain relatively sanguine about the issue.

Chart 4: Share naming immigration as one of the most important problems facing the country, August 2022 and August 2023

Source: IPSOS MORI “Issues Index”, August 2022 and August 2023

Politicians and poll watchers sometimes talk of public opinion in terms of pressures and tides, language which points to an underlying model of the public as having fixed views and demands, to which the political class must respond. But public opinion is not fixed or simple, and it isn’t formed in a vacuum. Instead, opinion on any issue is jointly authored by citizens trying to understand it and respond, media and other information sources seeking to package and explain it, and politicians seeking to frame arguments and dividing lines.

The persistent differences between English and Scottish opinion on immigration are an example of this. Immigration attitudes are not free-floating - they have clear roots in voters’ demographic characteristics, social identities and political values. But similar voters in the two countries nonetheless view the issue differently. These differences are, at least partly, the product of different media and political ecosystems which shape how the issue is framed and understood. No issue is simple or open to only one interpretation. Change the frame of reference used to explain an issue, or the political conversation in which it is embedded, and chances are you can, for some voters, some of the time, also change how voters understand an issue and what action they want to see.

1. The precise question asked on the BESIP is a bit different to those asked in the other surveys. Respondents are asked whether they think more or fewer immigrants should be allowed. They place themselves on a 0-10 scale with 0 marked “many fewer” and 10 marked “many more”. The charts show the share giving answers of 4 or less on the scale.

2. The paper is open access, so available for anyone to read. It is highly recommended for anyone interested in the politics of immigration.

3. Evans and Mellon found a 0.97 correlation between coverage of immigration in the Daily Mail and the salience of immigration in “most important problem” questions. This is about as strong a correlation as we ever see in social science. However, the correlation between immigration levels and Daily Mail coverage was also, at 0.89, very high indeed.