Brexit: Nationalism won the battle, will they win the war? Four facts that say no.

I am probably not the only one surprised by the results of the Brexit referendum. Even prediction markets got it wrong. On June 22, one day before the election, Predict It, one of the world’s most popular prediction markets was giving the “out” option about a 30% chance of succeeding.

But we all know what happened. “Out” won. Now the U.K. is on its way out of the European Union, and Scotland, who voted in, may be thinking of running a new referendum to leave the U.K.

But the bigger question here, at least to me, is whether this referendum signifies a global shift away from globalization and towards nationalism, or if it represents nationalism’s last stand. Of course, I understand that the referendum is not about nationalism, but about leaving a formal institution. Nevertheless, since much of the rhetoric from the campaign to leave contained anti-immigrant propaganda, it is reasonable to interpret this elections in the context of preferences for nationalism or globalization.

But to evaluate whether the Brexit signifies a global turn towards nationalism, or nationalisms last stand, we need to look beyond the aggregate results of this election and look at who voted for the “in” and “out” options. By looking at who voted for what, instead of the aggregates, we may get a better view of potential long term trends.

In the next paragraphs I will argue that this election is more likely to represent one of the last stands of nationalism even though we may see more results like this in the next two, three, or even four decades to come. This election, however, showed four clear signs that imply a shift away from nationalism over the long run. These four signs are: (i) the age difference among the voters that preferred the “in” and “out” option, (ii) the urban rural divide, (iii) the positive correlation between people voting for the “in” option and the number of immigrants in their area, and (iv) the correlation between the level of education of voters and their preference for the “in” option. Also, I should mention that these four facts probably are collinear (since urban centers tend to gather more educated people), but since I did not have access to the data, only the charts, I could not run a multivariate analysis, so I am presenting each fact separately.

Now, before I present the facts I should disclose that, while I am a scholar, I am person with personal views on this issue. I am personally an immigrant born in a family that has a recent history of immigration, so if there are even genes that make a person likely to switch countries, I probably have them. Also, I am a strong supporter from global labor mobility, to the point that I’ve advocated for labor mobility as a human rights policy in panels at the OECD. So I do have a personal view on this topic. Nevertheless, this personal view should not change the facts that I am about to present, so if you wish to attack this post, focus on the facts and avoid the ad-hominem logical fallacy of attacking the argument by attacking the person behind it.

Now that the disclaimers are out, let’s look at the four trend suggesting facts:

1 Younger voters clearly preferred the in option.

This fact is well known and its implications are quite obvious. If young voters do not change their preferences substantially as they age, and the even younger voters of the future have a preference for globalization as high as them, or ever higher, then a shift away from nationalism is inevitable, and mostly, a matter of time. Of course, there are a few caveats to consider here, like the fact that people do become more conservative as they age. But the gap appears to be substantial enough to believe that when Millenials will be as old as Boomers nationalism will be more of a minority opinion.

2. Urban Rural Divide

The second clear pattern of the Brexit referendum was the clear rural urban divide. When you consider England and Wales, the two countries that on average supported the “out” option, you still see that most mayor cities (London, Manchester, Liverpool, Cardiff, York, and Bristol), voted to stay in the EU. In Northern Ireland and Scotland, who on average voted “in,” the largest fraction of “in” votes also came from their urban centers (Glasgow, Edinburgh, and Belfast).

Now you may ask, why would the vote of cities matter more for the future? The idea here is that there is a global trend for people to live increasingly in dense cities, and if that trend continues, most of the votes will come from urban centers. So cities — which tend to be more liberal and accepting of foreigners— should become increasingly dominant in future elections.

3. Correlation between fraction of immigrants and “in” votes

The fact that places with more foreign born residents supported the in option more strongly is the most interesting of the three facts to me. I found this fact in a chart shared on Facebook by Ricardo Baeza-Yates, another scholar. Of course, this fact is probably collinear with the first two, since these places may be urban centers, and also, places populated by younger individuals. Still it suggests that people living in places where they find foreigners more regularly are more likely to accept them. So immigration is seen as positive in places that have more immigrants. And that’s good news for people like me, who believe in the possibility of global integration. Immigrants tend to reinforce a positive view of immigration. Those against immigrants, are more likely, those who do not have them.

4. Correlation between people voting to remain in the EU and their level of education

Finally, as a bonus, I present another chart showing the correlation between education levels and fraction of people voting to remain in the EU. Nevertheless, the long term trend here is also in favor of more acceptance of foreigners since there is an steady increase in the levels of education of people across the world.

Together, these four facts suggest that, while preferences for nationalism won the battle, they may fail to win the war. The Brexit election was one marked by clear differences in the demographics of those who voted for the “in” and “out” options, and these differences suggest that the same election would not have had the same result in the future.