Populations Choose Populism: Part Two: Factors of Populism and Nationalism

Populations Choose Populism: Part Two: Factors of Populism and Nationalism

In this article, we are going to take a variety of factors and analyse them using linear regression to see the relationship between differing factors and the rise of populist and nationalist parties in the European Election 2019, including both native and foreign unemployment, real GDP growth, asylum applications, and existing Islamic presence.

I am going to split this article into two parts. Part One shall be a simple summary of the results, and what we can derive from them. Part Two shall be the methodology and results in detail; feel free to skip this if you have no interest.

We have already covered the broad strokes of the European Election 2019 in Part One, and of the United Kingdom in particular here. I’d recommend giving them a read before reading this paper.

Part One: The Article

In my previous article, I made this picture below to show the spread of populism and nationalism across Europe. 

Simply noting the rise of these parties does nothing; we must ask ourselves why?

The Factors:

We shall measure whether the rise of right wing, left wing, or populist/nationalist parties have a correlation with the following factors. I shall then explain the results given from the linear regression step-by-step. Please note that this paper does not provide morality; I do not say if anything is good or bad, only that there is a relationship supported by evidence. I leave morality to politicians and Twitter.

Disclaimer; this is only a cross-sectional data involving 22 countries. The data itself has been gathered by governments and involve over 300 million people across Europe, but this is not a hard and fast experiment. It only shows the general trends of Europe, and provides some soft evidence on those trends.

The variables we shall look at are:

  1. Election Turnout: Often we hear that elections have unfavourable results because ‘not enough people voted’. If that is true, we should expect to see some correlation between people turning up to vote, and certain parties being elected.

  2. Native Unemployment Rate: People say that populists and nationalists are voted in because the people are disenfranchised and have no work.

  3. Foreign Unemployment Rate: People say that the foreigners are ‘taking our jobs’, or that ‘they are soaking up welfare’. If this have a relationship, either negatively or positively, that would support this.

  4. The Difference Between Foreign and Native Unemployment Rates: If the foreigners are taking the jobs, then they would have less unemployment than the natives. If they are indeed taking welfare, they’d have greater unemployment than the locals. If either of these is the cause of certain parties becoming more popular, we’d expect a relationship.

  5. Real GDP Growth: Some people (including myself here) say that populism or nationalism comes from a weakening economy. We would expect that lower rates of GDP growth would have a positive relationship with populist or nationalist parties election results.

  6. Tertiary Education: The common knowledge is that the smarter a population, the less it would vote populist or nationalist. The mirror to that is only the uneducated vote for these parties. If so, we would expect a negative relationship between having a higher tertiary education against the popularity of these parties.

  7. Total Population of a Country: The idea is that the smaller a country, the more it would notice a foreign population and react against it.

  8. Asylum Applications in 2018 and 2017: There is an idea that people are voting for these parties due to the massive European Migrant Crisis; therefore, we’d expect countries who have the most asylum claimants to their country would have the largest reaction.

  9. The Percentage of Asylum Seekers Compared to Total Population: 2,000 coming into a place of 2 million people is negligible. 200,000 people coming into the same place is 10%. We shall see if this has a relation.

  10. Percentage of Muslim Population: The final argument often used is that this is a reaction to the rise of Islam in Europe. If so, we could expect either a simple relationship (more Muslims, more nationalism) or maybe a non-linear relation (when there are enough Muslims, they will reduce the anti-Islamic reaction, or people may feel more comfortable with them). We shall see.

We shall pick the final results on which results best fit the data, and interpret from those results.

The Experiment:

Left-Wing Parties

Full results are in Part Two, but the fact is, there were only two significant relationships. Even then, the results do not show any great amount of correlation, and are not strongly supported (adjusted R-square 0.10%~; meaning only 10%~ of the data is supported by this model. Not great).

1) A country who votes left-wing is more likely to have asylum applications in 2018 (confidence level of >95%). I suspect that countries who vote left wing would be more open to immigrants, and countries with left-wing policies (such as greater welfare support) would be more attractive to immigrants.

2) Real GDP Growth has a negative relationship with voting left-wing (confidence level of >90%). Left-wing policies tend to stunt economic growth; equity over efficiency and all that.

Right-Wing Parties

I had to greatly alter the right-wing linear regression to find a model that had any kind of reliability. All variables were removed except for Muslim Population Percentage; the Percentage of Asylum Populations Against Population Size, and the Difference Between Native and Foreign Employment.

Simply put, all factors about immigration affected the rates of voting for right-wing parties, not economic nor educational factors, nor native employment. Once done so, I found very confident links between these factors and these parties. I go further into the effects of immigration on the working class here.

1) Muslim Population has a negative relation to voting right-wing; I assume people who live with larger populations of Muslims are less likely to vote right wing, but may be more likely to vote nationalist or populist. As such, more Muslims lead to lower right-wing voting, but more far-right voting (>99% confidence level).

2) The more people seek asylum in your country, the more likely you are to vote right-wing (>99% confidence level). I assume to protect your borders, or to vote to keep Christian and other historical institutions. This has also an effect over 4000% times that unemployment differences; this is the most significant reason for right-wing voting.

3) The more Muslims are unemployed than the native population by percentage, the more people will vote right-wing (>95% confidence level). The assumption is, I suppose, a perception of free-loading off of the state, as well as not integrating into society.

Nationalist and Populist Parties

Firstly, election turnout has no relationship with nationalists or populists getting elected; they don’t get elected just because ‘people didn’t turn up’.

This is the most reliable data in this experiment; all of the variables have over 90% confidence, and 4/6 variables have over 99% confidence level. The formula explains 62% of the results.

The Real GDP growth has a strong relationship with voting Populist/Nationalist (>95% confidence level). My assumption is that countries that are doing well economically have attracted greater numbers of Muslims, which in turn leads to a rise of this type of support. This relationship is large; 8%-18% of real GDP growth can provide an additional 1% of support for nationalist and populists.

The more people have tertiary (university level) or higher education, it leads to lower support for populist/nationalist parties (>90% confidence). We have seen the average green or liberal voter in this article. It does seem that the idea that only the undereducated vote nationalist or populist; this could represent that the working class affected by illegal labour are voting more (see more here), or perhaps a greater share of the elderly voting who are more likely to have not gone to university. However, the negative link between education and voting for these parties is easily proved.

The strongest relationship in this entire article is that of the size of your country; the larger your country’s population, the more likely you are to vote nationalist or populist (>99.99% confidence level). This could be due to several reasons. Perhaps a “We’re full” mentality? I suspect it may be due to the fact that a large country would have a history of being successful in terms of economics and war, and as such, would be more likely to drive you to pride (see it in India here, who has a nice large population). Every 600,000-760,000 of population has an additional 1% of support for nationalists and populists.

There is a negative relationship between asylum applications in 2018 and the support for nationalists and populists (>99% confidence level). My suspicion is that countries known for voting for nationalists and populists are less likely to attract asylum applications in 2018.

There is a positive relationship between asylum applications in 2017 and the support for nationalists and populists (>99% confidence level). This is the year after the Syrian migration. In 2018, the countries pro- and against asylum have become none; at this point, many countries were accepting asylum more evenly regardless of political thoughts.

Finally, the second largest correlation is that between existing Muslim populations and voting for nationalists and populists (>95% confidence level). For every 3.2%-8.4% of Muslim population will create 1% of support for these parties, meaning it is a very important reason for this phenomenon. 

The Final Conclusion

The above is worrying for another reason; long-term, Islamic population in Europe will increase drastically in Europe, with many countries becoming majority Islamic in this century, and will therefore trigger massive nationalist and populist uprisings.

This paper shall several conclusions.

Left-wing parties have negative relationships with Real GDP growth (probably due to the massive deficit spending of socialist policies) and asylum in 2018.

Right-wing parties are losing votes to the nationalists and populists from the rise of the Islamic population. It seems that support for right-wing parties are strongly linked purely as a reaction to Islamic migration, even as they lose support to the nationalist and populists.

The nationalist and populists are, as many suspect, heavily supported on issues of Islam and migration. However, they also strongly correlate to the size of a country, and how well the economy is growing, and as many say, are more likely to be uneducated.

So is all of this a result of the rise of Islam as some say (such as Hillary Clinton). Not entirely. There are strong economic factors that are also causing this, as well as demographic issues. We can also see, that as the right wing shrinks, the populists grows. This bore out in the European Elections 2019.

If the issue of Islam and asylum in Europe is not solved; if the economics of the common man are not solved; we can see further and further growth of people who promise to solve the problem. The data is there; let’s see what they do.

Part Two: Linear Regression

I have taken a cross-sectional snapshot of Europe in 2019 with the most official and recent data I can find. 6 countries have been removed from this regression: Bulgaria, Croatia, Cyprus, Lithuania, Malta, and Romania. This is due to the fact that these countries are missing large amounts of data from the OECD, and I want to keep the methods for calculating data as similar as possible. 

Anyone asking for a literature review of these variables is welcome to read Part One and this article.

The dependent test variables are below:

 The independent variables chosen are as such:

Another issue came with the percentage of Muslims within a population; multiple countries such as Germany, France, and Belgium do not report these numbers anymore due to fears of stoking Islamophobic hate. As such, we have taken data from 2016, the last year these countries reported this data.

The full table of data with legible tables shall be put into the appendix.

We shall be running the linear regression with using the Ordinary Least Square model, robust standard errors, using GRETL.

αParty = ßElecTurn+ ßNatUnemployPERC + ßForUnemployPERC + ßDIFUnemployPERC+ ßRealGDPGrowth+ ßTerEduPERC+ ßTotPopMil+ ßAsyApp2018+ ßAsyApp2017+ ßPERCAsyAppPop+ßMusPERC+ ε

αParty represents each grouping of parties; Left-Wing, Right-Wing, and Populists/Nationalists. We assume that ε (the error rate) is 0. ß is each case is the correlation between the αParty and the independent variable.

The final models selected shall be the models that best fit the data; higher adjusted R-squares, lower criterions. The higher the confidence level, and the greater the correlation, the more important the relationship. Anything significant shall be highlighted in pink.

The Results:

Right Wing Parties

Originally, there were no relationships at all, and the adjusted r-squared value was negative. Only this version of the regression yielded results that are somewhat reliable (these results cover more than the left-wing results), but those results are very strong.

αParty = ßDIFUnemployPERC + ßPERCAsyAppPop+ ßMusPERC+ ε

Left Wing Parties

Populist and Nationalist Parties

Like the right-wing model, the original regression had no strong relationships. Having only Islam-migration factor had no relationship. Only economic factors had no relationship, nor did just employment factors. However, under this model:

αParty = ßRealGDPGrowth+ ßTerEduPERC+ ßTotPopMil+ßAsyApp2018+ ßAsyApp2017+ ßMusPERC+ ε

Using the above formula, adding any of the other variables to the above has no change to the above relationships, but do not yield any other information. As such, some economic factors, some demographic factors, but 50% migration factors finds significant relationships between all factors and the rise of nationalist and populist parties. We find an adjusted r-square value of 0.62, meaning that 62% of the data can be explained by this regression.

Screenshot 2019-06-20 at 17.06.32.png

Appendix: The Data

The table of data I’ve assembled can be seen here.

Cross-Sectional Data Of Factors Against Different Parties Election Results for the European Election 2019

Screenshot 2019-06-20 at 17.10.43.png
Screenshot 2019-06-20 at 14.07.37.png
Screenshot 2019-06-20 at 14.09.16.png

Source: (BBC, 2019a, 2019c; OECD, 2019a, 2019b, 2019c, 2019d; World Population Review, 2019; Jewish Virtual Library, 2019; Eurostat, 2019)

Countries in Red Are Missing Data and Excluded from Linear Regression


Some sources of the Bibliography are not for this article, but for Part One.

In all articles, I provide as much information for sources as possible, including links. I encourage everyone reading this article to read deeper, and make their own conclusions. For students, links are here so they can read the original source themselves.

AP (Associated Press), (2019), “The Latest: Salvini says populists will control 150 seats”, published by the Associated Press, retrieved from https://www.apnews.com/d09a7028c0f04c60bd604b4bd8056a0f on 16th June 2019.

AFP, (2019), “Inquiry launched after 73-year-old woman seriously hurt in Nice”, published by the Local, retrieved from https://www.thelocal.fr/20190325/inquiry-launched-after-73-year-old-woman-seriously-hurt-in-nice-yellow-vest-protest on 17th June 2019.

BBC, (2019a), “European elections 2019”, published by the BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/topics/c7zzdg3pmgpt/european-elections-2019 on 1st June 2019.

……, (2019b), “European elections 2019: Results in maps and charts”, published by BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/world-europe-48417191 on 1st June 2019.

……, (2019c), “European elections 2019: Country-by-country roundup”, published by BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/world-europe-48420697?intlink_from_url=https://www.bbc.com/news/topics/c7zzdg3pmgpt/european-elections-2019&link_location=live-reporting-story on 16th June 2019.

……, (2018a), “Solvenia, Slovakia, and the constant confusion between the two”, published by the BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/world-europe-43419675 on 16th June 2019.

……., (2018b), “Migration to Europe in charts”, published by the BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/world-europe-44660699 on 17th June 2019.

……, (2016), “Clinton: Half of Trump supporters ‘basket of deplorables’”, published by the BBC, London, United Kingdom, retrieved from https://www.bbc.com/news/av/election-us-2016-37329812/clinton-half-of-trump-supporters-basket-of-deplorables on 17th June 2019.

Boffey, D., (2019), “Belgium government loses majority over UN migration pact”, published by the Guardian, retrieved from https://www.theguardian.com/world/2018/dec/09/belgium-government-loses-majority-over-un-migration-pact on 16th June 2019.

Dorpe, S.V., Cerulus, D., and Cokelaere, H., (2019), “Far-right surge in triple election shocks Belgium”, published by Politico, retrieved from https://www.politico.eu/article/belgium-far-right-surge-elections-2019/ on 16th June 2019.

Economist, The, (2019), “Syriza, Greece’s ruling party, is heading for defeat”, published by the Economist, retrieved from https://www.economist.com/europe/2019/03/21/syriza-greeces-ruling-party-is-headed-for-electoral-defeat on 16th June 2019.

EP Spokesperson, (2019a), “How many MEPs?”, published by the European Parliaments Spokesperson’s Department, retrieved from http://www.europarl.europa.eu/news/en/faq/12/how-many-meps on 16th June 2019.

……, (2019b), “What are the European Parliament’s powers and legislative procedures?”, published by the European Parliaments Spokesperson’s Department, retrieved from http://www.europarl.europa.eu/news/en/faq/17/what-are-the-european-parliament-s-powers-and-legislative-procedures on 16th June 2019.

Eurostat, (2019), “Asylum and first time asylum applicants by citizenship, age, and sex Annual Aggregated data (rounded)”, published by Eurostat, Luxembourg, retrieved from https://ec.europa.eu/eurostat/web/asylum-and-managed-migration/data/database# on 17th June 2019.

Gutiérres, P., and Clarke, S., (2019), “Spanish General Election full results”, published by the Guardian, retrieved from https://www.theguardian.com/world/ng-interactive/2019/apr/28/spanish-general-election-2019-full-results on 16th June 2019.

Hawkins, R., (2012), “Biden tells African-American audience GOP ticket would put them “back in chains””, published by CBS News, retrieved from https://www.cbsnews.com/news/biden-tells-african-american-audience-gop-ticket-would-put-them-back-in-chains/ on 17th June 2019.

Jewish Telegraphic Agency, (2019), “Belgian far-right party triples share to be second largest in federal election”, published by the Jewish News, United Kingdom, retrieved from https://jewishnews.timesofisrael.com/belgian-far-right-party-triples-share-to-be-second-largest-in-federal-election/ on 1st June 2019.

Jewish Virtual Library, (2016), “Islam: European Muslim Population”, published by the Jewish Virtual Library, retrieved from https://www.jewishvirtuallibrary.org/european-muslim-population on 17th June 2019.

Meredith, S., (2018), “Europe’s migration crisis is ‘make-or-break’ for the EU, Germany’s Merkel says”, published by CNBC, retrieved from https://www.cnbc.com/2018/06/28/europes-migration-crisis-is-make-or-break-for-the-eugermanys-mer.html on 17th June 2019.

OECD, (2019a), “Foreign born unemployment (indicator)”, published by OECD Publishing, Paris, France, retrieved from https://data.oecd.org/migration/foreign-born-unemployment.htm on 17th June 2019.

……., (2019b), “Native born unemployment (indicator)”, published by OECD Publishing, Paris, France, retrieved from https://data.oecd.org/migration/native-born-unemployment.htm#indicator-chart on 17th June 2019.

……, (2019c), “Real GDP forecast (indicator)”, published by OECD Publishing, Paris, France, retrieved from https://data.oecd.org/gdp/real-gdp-forecast.htm#indicator-chart on 17th June 2019

……., (2019d), “Adult education level (indicator)”, published by OECD Publishing, Paris, France, retrieved from https://data.oecd.org/eduatt/adult-education-level.htm#indicator-chart on 17th June 2019.

Oltermann, P., (2019), “Austrian chancellor Sebastian Kurz is ousted in no-confidence vote”, published by the Guardian, retrieved from https://www.theguardian.com/world/2019/may/27/austrian-far-right-party-likely-to-back-motion-to-oust-chancellor-eu-election on 16th June 2019.

Rankin, J., (2019), “MEPs create biggest far-right group in European parliament”, published by the Guardian, retrieved from https://www.theguardian.com/world/2019/jun/13/meps-create-biggest-far-right-group-in-european-parliament on 16th June 2019.

Shuster, S., (2019), “Austria’s Young Chancellor Sebastian Kurz Is Bringing the Far-Right into the Mainstream”, published by Time Magazine, retrieved from https://time.com/5466497/sebastian-kurz/ on 17th June 2019.

Verhofstadt, G., (2019), “Guy Verhofstadt celebrates ‘new balance of power’ in EU Parliament”, published by Euronews (in English), retrieved from https://www.youtube.com/watch?v=_uaDOnUAaX0 on 26th May 2019.

Wintour, P., (2018), “Hillary Clinton: Europe must curb immigration to stop rightwing populists”, published by the Guardian, retrieved from https://www.theguardian.com/world/2018/nov/22/hillary-clinton-europe-must-curb-immigration-stop-populists-trump-brexit on 17th June 2019.

World Population Review, (2019), “Population of Countries in Europe 2019”, published by the World Population Review, retrieved from http://worldpopulationreview.com/countries/countries-in-europe/ on 17th June 2019.

More than Money: The Sino-U.S. Trade War

More than Money: The Sino-U.S. Trade War

Populations choose Populism: Part One: European Election 2019

Populations choose Populism: Part One: European Election 2019