Visualizing Values Mismatch in the European Union



Join 36.9K other subscribers

In my July 1 post, Brexit as Destructive Creation, I argued that one significant cause for the European dysfunction was the choice made by the European elites to expand the union too fast too far. Why do I think this was a mistake?

As I have said on numerous occasions (in this blog and in my other writings), it is hard to get people to cooperate, especially in large social groups. Successful cooperation requires that people share values and institutions. Values tell us why we want to cooperate: what is the public good that we collectively want to produce? Norms and institutions tell us how we are going to organize cooperation. Mismatched values and institutions may doom a cooperative effort even before it has a chance to get going.

In my opinion, the expansion from the original six nations (Benelux, France, Germany, and Italy – I will refer to them as the “core” EU nations) to the current 28 was a big, big mistake. We can use the data collected by the World Values Survey (WVS) to visualize just how bad this mistake was.

WVS has been collecting data on people’s beliefs in many countries since 1981. One interesting result of analyzing these data was that much of variation between populations of different countries can be mapped to just two dimensions: (1) Traditional values versus Secular-rational values and (2) Survival values versus Self-expression values. When average values for each country in the sample are plotted in a two-dimensional space defined by these two axes, we have what is known as the Inglehart–Welzel Cultural Map. Here it is for the latest (sixth) survey:



Inglehart-Wetzel group culturally similar countries into “Catholic Europe”, “Protestant Europe”, “English Speaking”, and so on.

But I am interested to look at this mapping from a different point of view. Accordingly, I color-coded all countries into the following categories:

  • Core (red): the original six countries that formed the European Economic Community
  • EU (brown): the other 22 members of the European Union
  • Europe (green): two Western European countries that are not in the EU
  • Candidate (yellow): current candidates for the EU
  • World (grey): the rest of the world (I omitted country names that would clutter the infographic too much).

Note: the reason Italy* has an asterix is because, for some reason, it was not included in the sixth wave, so I used its values from the fifth wave data.

And here’s what it looks like:


The pattern is so striking it almost doesn’t require commentary, but let’s spell it out anyway. The original six (“Core Europe”) group together very closely. There are only two other countries that are part of the same cluster, Austria and Switzerland. Remarkably, the modern territories of both of these countries were encompassed by the boundaries of the Carolingian empire (see Is this the Beginning of the End for the European Union?). It looks like the “ghost” of the Charlemagne’s empire has more influence on today’s cultural values than such later distinctions as Catholicism versus Protestantism.

The current 28 members of the European Union, on the other hand, don’t cluster at all. On the contrary, they span three-quarters of world variation in values. Only African-Islamic countries and central America end up outside the ellipse that encompasses all 28 EU members.

Given such normative mismatch, is it so surprising that the European Union in its current composition is a dysfunctional organization?



Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
al loomis

perhaps it works as well as can be expected, as a ‘fourth reich?’

Jan Mikkelsen

Very interesting. Iceland should probably be green like Norway and Switzerland rather than brown.

Luigi Forlai

It’s true there are shared values in the European Core but I do not think there is a common mythology
(foundation stories) that allow an easy interaction among the populations within the core. For ex. he foundation stories in Germany, France and Italy differ deeply and there is a major mistrust that is not easily overcome.

Federico D'Onofrio

You don’t offer any comparative measure: how close would chinese provinces, US states or russian subjects group? Neither you address the issue of trends. Are european countries converging or diverging? (Not to mention the methodology behind the data)


It would be very informative to break this down by regions.
I would expect the difference between Lombardy and the Mezzogiorno to be as big as the difference between Germany and Poland (Italy on average is between the 2).

Same with Catalonia and Andalusia. Likely as different as Austria and Greece (Spain on average is between the 2 right now and on the edge of the oval with Italy).

Ross Hartshorn

I am curious, how much of the traditional-vs-secular is predictable from something like attendance at religious services, which might be more widely available (e.g. at the state or province level)? And how much of survival-vs-self-expression is predictable from median income or life expectancy (which also might be available at state or province level)? Perhaps a proxy measurement of values could be made for within-nation differences.

Paolo Ghirri

early railways network can be a factor on the Survival vs Self-Expression?

Giorgios Papadopoulos

The culture of the Greeks orthodoxs is more closely to the culture of catholics Italians than to it of countries like Russia, Ukraine or Bulgaria. There is the same religion in the surface but under of it all is different

Federico D'Onofrio

Without data at subnational level and about historical trends, this graph doesn’t mean much. If variation within countries that have succesfully existed for centuries turned out to be more significant than variations between countries, or if the trend showed converging dynamics then you could draw exactly the opposite conclusions. There are similar problems with Emmanuel Todd’s criticism of the Eu


Does the entirely of variations between countries actually matter, or does the character of the ruling elites and in modern states the middle class matter most of all?

The higher within the state you go the consistent, and usually more historical, it gets.

bob sykes

You need to explain the axes. As it is, they are largely unintelligible: “”survivial vs. self-expression,” what do the numbers mean? Is the far right highest survival, lowest self-expression? That would be the standard English interprestion: the ratio has survival in the numerator. I had to go back to the original paper to figure out what each axis actually represents.

Loren Petrich

The World Values Survey researchers define what they mean by those values in — here are their summaries of those values:

Traditional values emphasize the importance of religion, parent-child ties, deference to authority and traditional family values. People who embrace these values also reject divorce, abortion, euthanasia and suicide. These societies have high levels of national pride and a nationalistic outlook.

Secular-rational values have the opposite preferences to the traditional values. These societies place less emphasis on religion, traditional family values and authority. Divorce, abortion, euthanasia and suicide are seen as relatively acceptable. (Suicide is not necessarily more common.)

Survival values place emphasis on economic and physical security. It is linked with a relatively ethnocentric outlook and low levels of trust and tolerance.

Self-expression values give high priority to environmental protection, growing tolerance of foreigners, gays and lesbians and gender equality, and rising demands for participation in decision-making in economic and political life.


Hi Peter, to pull this together with your other work on the centrality of Steppe peoples as political and social catalysts, what do you make of Robin Hanson’s observation that the placement of a cultural group towards the Survivalist end of the Survivalist-Self-Expression value axis is substantially correlated with proximity to steppes or semi-deserts from which pastoralists have historically emerged?

It seems that populations in places with a ‘Steppe-Eurasian’ dynamic to their political history, all cluster towards the Survivalist end, especially if this history is recent.


Here it is:
His idea is that the US, more representative of an island/Oceanic outlook, have values that emphasize the functioning of individuals and families, while the historical primacy of outside invasion in populations with a ‘Eurasian’ outlook gives the functioning of the group/community as a whole more salience.

Personally I think this is quite confused; since there are many factors that influence family and community size, and there are many societies where extended families or tribes are the basic political unit, not the nuclear family. I suspect what he means by ‘family’ is excessively biased towards the WEIRD nuclear family, and the orientation of individuals in such societies towards gaining a mate in a field with freer mate choice and less coercive marriage practices.

But it seems there is a powerful germ of an idea here, since family vs community naturally raises interesting connections to multilevel selection, especially because the distribution of family types and family size/cohesion across Eurasia seems to show some links to the influence of the Steppe as well.


Excellent analysis!
(In retrospect it’s surprising no one has applied WVS to the EU before.)

But of course shared values may have nothing to do with it: Junker himself has clearly indicated that what will keep the rump EU stable is threats, punishment, humiliation. (Not my own, Josef Stiglitz.)

Moritz Reichelt

Where can one find the original data? It would be interesting to know, which questions were asked leading to these parameters displayed in the two axes.

Shilla Nassi, MD, PhD

i was curious as to whether you’ve come across the parasite-stress theory.

i note that, until recently, malaria was endemic in southern europe, reaching even the north

here’s a critique of the theory:

Clinton Reilly

There is a similar situation (probably a much more advanced state of breakup) in the Middle East. Although some think ISIS can be defeated and that this will bring a ‘new dawn’ for the Middle East, it is also likely that another organisation will take ISIS place and the turmoil will continue. Do you think it is possible to apply a quantitative history approach to predicting what is a likely workable political future for the Middle East?


One of the things that I find most fascinating in maps like this are the large geographic correlations. With the exception of Northern Australia, all of the tropical countries are in the lower left quadrant. This aligns with the parasite stress theory. I think it has more to do with the selectional force of cold winters. Weak, loosely structured social organizations simply will not survive freezing conditions. There are a lot of other factors looslely correlated with latitude, such as jungles vs plains. Why did the great tropical civilizations of Central America, Southeast Asia, and even Africa all die out? Perhaps the annual climate change of seasons produces greater resliency to exceptional stresses such as long-lasting droughts.

stefan szymczyk

very interesting grafs, need too commit more time 2 analize it better

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. Visualizing Values Mismatch in...

© Peter Turchin 2023 All rights reserved

Privacy Policy