War and Relatedness

I am told that the economics commentator Noah Smith linked to a 2016 paper of mine on Twitter today. I am not on Twitter and likely will never be on Twitter, so I can't link to this thread. Yet the paper is quite relevant for the current situation between Russia and Ukraine, so I can't resist commenting on it here.

In "War and Relatedness", Enrico Spolaore and I argued that societies that are more closely related in terms of culture,  religion, ancestry, and language, are more likely to go to war with each other. This contrasts with a view that interstate wars feature a "clash of civilizations", a hypothesis articulated by political scientist Samuel Huntington in a 1996 book. This view implies that societies that differ in terms of religion, language, values, and common ancestry are more likely to fight. It is starkly at odds with the empirical evidence, which strongly supports the opposite view. Using a dataset of interstate conflict between 1816 and 2001, Enrico and I showed that a simple measure of relatedness between countries strongly predicts the likelihood of interstate conflict. This is a very robust relationship, which holds across the world, within Europe, and across the world after excluding Europe. It holds in different periods, and after controlling for an exhaustive list of geographic proximity measures. This is not about geography, but about human relatedness.

Why might there be a greater propensity to fight with culturally closer societies? Enrico and I articulated two mechanisms. First, closely related societies are likely to care about more similar resources. Many wars are over the control of good land or of natural resources such as oil. More similar societies are more likely to have preferences for the same rival goods, and to develop conflicts over goods or issues they both care about. Second, interstate wars are often about territorial disputes. When a country wishes to conquer territory from another, it cares about its ability to manage the new territory in the event of victory. It is easier to administer and integrate a more similar population after conquest, than a very different one.

The example of Russia and Ukraine (very closely related societies in terms of ancestry and culture) provides a very nice example of our theory. Moreover, Russia seems mostly interested in regions of Ukraine (Crimea, Donbas) where a strong presence of ethnic Russians will make post-conquest administration easier. In fact, in our 2016 paper, Enrico and I explicitly mentioned the case of Crimea:

"For instance, the recent conflict between Russia and Ukraine over Crimea and the Eastern region of Ukraine (not included in our sample, which stops in 2008) is consistent with our prediction that states with closely related populations are more likely to engage in conflict over control of rival goods and populations." (p. 937)

The thesis of this paper has held up quite well in light of recent events - if I may say so myself!

Comments

[Lisa Trueblood]

That is a very interesting connection and something I hadn't though of before!

I found another interesting article that argues that countries that are democratic, trade heavily and belong to lots of international bodies fight each other less often than authoritarian, isolationist states do.

The article also looked at the correlation between democracy and peace and determined that the countries most prone to wars appear to be neither autocracies nor full democracies, but rather countries in between. A similar finding applies to prosperity. Middle-income countries are more warlike than very poor or rich ones.

I would argue that Russia falls into all of these categories. The article is linked here in case you're interested in reading more: https://www.economist.com/graphic-detail/2018/11/08/which-countries-are-most-likely-to-fight-wars

[Romain Wacziarg]

Lisa - thanks. Yes, the idea that countries that are democracies and that trade a lot with each other tend not to go to war is known as "liberal peace theory" or "democratic peace theory". You can read about it here:
https://journals.sagepub.com/doi/10.1177/0022343396033001002
and here:
https://en.wikipedia.org/wiki/Democratic_peace_theory
In my view these are among the most robust correlations in the field of international relations.
In my work with Spolaore we controlled for both democracy and trade and the effect of relatedness did not change - greater cultural distance was still predicting lower conflict. These findings did not make it in the published version of the paper but we still have them somewhere.

[Mareena Haseeb]

I also had never considered the argument presented in your paper, Professor. The argument that societies that are more closely related in terms of culture, religion, ancestry, etc. are more likely to go to war is fascinating to me as I always assumed at a high level that wars were a result of a “clash of civilizations.” However, this got me thinking about India and Pakistan. While the fighting is rooted in religious differences, both countries share pieces of culture, religion, and language. Not to mention that those who migrated to Pakistan when it was founded (like my grandparents) were born in India, so there is definitely shared ancestry. Despite all of the similarities, they have gone to war over their differences. If we were to apply Huntington’s view here, we would assume that India and Pakistan would not fight as the societies do not differ (at a high level) in terms of religion, language, values, and common ancestry. As much as I would like for them not to fight, what we see is the opposite, the view shared in Professor’s paper, and what is also happening with Russia and Ukraine, that closely related societies are more likely to go to war with each other. In both cases, countries have developed conflicts over goods or issues they both care about (for example, we could parallel Russia and Ukraine over Crimea to India and Pakistan over Kashmir).

With respect to Lisa’s article, I don’t find it that surprising that countries that are democratic, trade heavily, and belong to lots of international bodies fight each other less often than authoritarian and isolationist states. I would assume that they want to maintain peace to ensure they don’t break trade ties which would ultimately impact their economy. It seems that my assumption is correct based on Professor’s sharing of the literature on the liberal/democratic peace theory!