Saturday, July 9, 2016

Immigration sentiment

Above, a lovely graph from The Conversation. A common story says that opposition to immigration comes from people in high-immigrant communities, who suffer externalities from the presence of many immigrants. It is not true.

24 comments:

  1. That is correct... but The Economist highlighted an even more striking fact, that places that endured high levels of growth of immigration recently did, in fact, vote most for leave.

    http://www.economist.com/news/britain/21701950-areas-lots-migrants-voted-mainly-remain-or-did-they-britains-immigration-paradox

    I'd venture a guess that, perhaps, this suggests that places unused to immigration experience a backlash against immigration at first, as the culture shock hits, but after a while, both immigrants start to assimilate and neighbours begin to realize that, as is always the case, immigrants are just human like them and begin to accept...

    ReplyDelete
  2. Doesn't actually disprove this. There are many stories of people moving to suburbs and other areas as they get older /because/ of migration and other factors pushing them outwards. This could reflect that. Additionally, since immigrants were allowed to vote, it is also helped by immigrants obviously voting remain.

    ReplyDelete
    Replies
    1. The only immigrants allowed to vote were those from Ireland and Commowealth nations:
      https://www.gov.uk/government/topical-events/eu-referendum/about
      So immigrants in general were not allowed to vote and in particular no EU immigrant (except for the Irish) was allowed to.

      The general anti-immigration sentiment caused the conservative government to vote laws against non-EU migrants such as the following one, presented by Theresa May, expected to become the new prime minister:
      http://www.bbc.co.uk/news/uk-politics-17204297
      There are anecdotes (unfortunately I don't know if there is any data about it) of non-EU migrants being against EU migrants as the latter can move to UK and get jobs easily without a visa and other restrictions. The silliest one comes from the Leave campaigner and UK Minister for Employment, Ms Priti Patel (of Indian origins):
      http://www.bbc.co.uk/news/uk-politics-eu-referendum-36330032

      Delete
    2. Wrong. All citizens were allowed to vote, whether they were immigrants or not.

      Delete
  3. It seems to me it's possible to interpret this a number of ways. One way is "communities with few immigrants prefer to keep it that way". Another is "immigrants are statistically more likely to vote for policies that favour immigrants", which is hardly a surprise.

    It's good to see a statistical approach, at least - there's too much black-and-white thinking going on in the media. Having concerns about the scale and scope of immigration does not mean you hate immigrants or want immigration to stop altogether.

    ReplyDelete
  4. All commonwealth citizens could vote in BREXIT. This means that Nigerians, Pakistanis, and Indians could vote. I expect they live in the areas with high concentrations of immigrants, and I expect that they voted Remain. It seems that the votes of these immigrants were not accounted for in the graph. Furthermore, is it possible to account for white flight? Lastly, the issue of rural voters in certain areas voting Leave in reaction to seasonal European migrant workers was well-covered by the media, and was reflected in the data.

    ReplyDelete
  5. Refutation of this: it doesn't look only at Eastern European immigrants, which is what the Brexiters were complaining about. Look at Boston, Lincolnshire, the place in Britain that most heavily voted Leave:

    "According to the 2011 Census, Boston is now home to a higher proportion of eastern European immigrants than anywhere else in England and Wales: 10.6 per cent of the town’s population of 65,000 comes from one of the “new” EU countries such as Poland, Lithuania, Latvia or Romania.

    More concerning, it has now been named as the least integrated place in England and Wales by a Policy Exchange report."

    http://www.independent.co.uk/news/uk/home-news/boston-how-a-lincolnshire-town-became-the-most-divided-place-in-england-a6838041.html

    I'm tired of people drawing up graphs without looking at a single datapoint.

    ReplyDelete
  6. Is this really that surprising? In areas with large immigrant populations, those immigrants themselves and/or their children are also voters, right? It would be interesting to see how the typical UK-born resident with UK-born parents etc voted in those areas.

    ReplyDelete
  7. Interesting graph but couldn't convince myself to keep reading the linked article after the authors confused P(A|B) with P(B|A) in the first paragraph. (The relevant poll http://bit.ly/28VSG6E does not show that 80% of leave voters think "multiculturalism is a force for ill." It's the other way around: 80% of the people who think this voted leave.)

    ReplyDelete
  8. "who **suffer** externalities from the presence of many immigrants"(?). Maybe it is about **enjoying** externalities.

    ReplyDelete
  9. Why should anti-immigration sentiment be stronger in areas with a higher percentage of immigrants? It is not true in New York state (New York city is less anti-immigrant than upstate), so why should it be true in the UK? Could it be that people who dislike immigrants self-select to live in places where there are fewer foreign born residents, or that it is easier to dislike immigrants when you don't know that many?

    ReplyDelete
    Replies
    1. I would not expect a high percentage of immigrants to favor anti-immigrant policies. The people most likely to hold anti-immigrant sentiments are those most likely to have been hurt, i.e. people who have had their livelihood threatened by low-paid immigrants, esp. in the building trades.

      Delete
    2. Not necessarily. These are also people most likely to have worked with immigrants and have immigrant friends. In my experience, the people with the stronger anti-immigration attitudes are people who live in relative homogeneous but struggling communities and become convinced that immigrants are to blame, or who simply dislike them for other reasons (cultural, religious, fear of terrorism, etc.)

      Delete
  10. Is it possible that what's driving this negative trend line is simply that non-Uk born residents tend to be pro-immigration and vote 'stay'? Unless I'm missing something, it's not immediately clear from this chart whether natives are more or less likely to vote 'leave' when located in high immigration areas.

    ReplyDelete
    Replies
    1. In fact, if all immigrants were citizens and could vote and opposed Brexit, and voted at no higher a rate than the native citizens, then we'd expect a trend line with a slope of -1 in the graph. In fact, the slope is much less than -1. This suggests that the percentage of native citizens who voted for Brexit did rise with percentage of immigrants.

      Delete
  11. Or do they? http://www.economist.com/node/21701950/print

    ReplyDelete
  12. Fascinating. Still, so many possible confounders - income level (higher incomes are less impacted by immigration, much higher median income in , eg, London?), unemployment (higher in "leave" regions, esp. long term unemployment. ?). I suspect a much more "micro-demographic" analysis would be revealing.

    ReplyDelete
  13. Oh, and of course one should probably remove naturalized immigrants able to vote. Presumably there is a higher percentage of them in communities on the right of the chart? Maybe not enough to move the outcome significantly...

    ReplyDelete
  14. Another possible (although not entirely alternative) explanation is that those communities which are not yet melting pots look forward and vote against converging to the state of those communities which instead are.

    ReplyDelete
  15. Arguably, the people most affected by immigration in the UK are the Cockneys, the old London working class. They don't really exist any more. They and their children moved further east to Essex. And they voted Leave. The places they used to live, now disproportionately Bangladeshi, voted Remain. It would be interesting to know how other places adjacent to areas with high immigration voted, especially if these adjacent areas contain people who left the areas with high immigration (whether EU immigration or other).

    ReplyDelete
  16. Globalization's Impact
    Does an economic decision by Corporate America to open a processing plant in a small community affect the cultural and political makeup of that community -- with regard to economic opportunity, income, racial/ethnic/religious makeup, attitudes about immigration, and the social fabric with a way of life in rural society?

    Bureau of Labor Statistics: Occupational Employment Statistics;
    Slaughterers and Meat Packers

    http://www.bls.gov/oes/current/oes513023.htm

    A Location Quotient greater than 1.0 means that the particular job/industry has a major impact on the locality. The LA Metro Area and Fresno, CA employ about the same number of people -- Location Quotient for LA Area is 0.34; for Fresno, 6.42. What is the impact in the old Dust Bowl area of the U.S. -- CO, KS, OK, TX?


    In a nation governed by a democratic process, where people vote and participate in their local government, what happens in such a community?...especially with regard to the ability to change?

    ReplyDelete
  17. OF COURSE the data will slope that way, as non-UK born citizens are allowed to vote, too, and are included in the sample. More interesting would be a poll of just UK-born voters.

    ReplyDelete
  18. Now pretend your prior is such that you are skeptical of this claim -- what methodological problems do you see in such a simplistic analysis? Surely at least... a few.

    Perhaps the communities that were traditionally completely homogeneous also interact with a shared culture (omitted variable bias here), and are more sensitive BOTH culturally and on production factors to immigration shocks.

    Maybe not. But a single variable, a few line plots, and a story, is only sufficient evidence for someone already convinced.

    ReplyDelete
  19. The same observation can be made in Switzerland (Not sure it exists in English https://de.wikipedia.org/wiki/Eidgen%C3%B6ssische_Volksinitiative_%C2%ABGegen_Masseneinwanderung%C2%BB). Areas in Switzerland with the highest percentage immigration voted against limiting immigration, while areas that have almost no or a low percentage of immigrants voted for the "limiting immigration" law. The next country with the same "pardox" voting behavior might be Germany. Elections in Mecklenburg-Vorpommern showed a very strong rise in votes for the populist party AfD which sole objective is to criticize Merkel's immigration laws. Paradoxically Mecklenburg-Vorpommern has the least percentage immigrants in Germany and strong economic growth. The behavior does not seem to be UK specific.

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.