The external military threat helps to create a strong European identity and

External military threats help build a stronger European identity and cooperation

Editors’ note: This column is part of the Vox debate on the economic consequences of war.

The ongoing war in Ukraine has challenged European citizens and decision-makers. Will the EU, together with its allies, be able to provide a strong and uniform response? Or will it be divided into different groups who cannot cooperate?

The EU is a particularly interesting case for a political union because of its historical significance, its enormous ambitions and the level of challenge for successful cooperation. Considering the vastly different member states in terms of language, culture and history, what has been achieved in the last few decades is remarkable.

However, many struggles remain, especially in times of crisis. A key issue for further – and successful – cooperation is the question of a common European identity. Currently most of the politics in the European Union is still governed by national considerations (Gehring and Schneider 2018). The European debt crisis, for example, has challenged the establishment of insurance processes and redistribution. A strong collective identity helps build trust and empathy within a group – a key condition for successful collaboration – and a willingness to share risks and support each other.

Identity determinants have recently become a popular topic in economics research. In the EU context, Dehdari and Gehring (2022) show that negative historical experiences, including inter-state wars and tensions with the Central States, are the main reason for influencing the power of regional identity. Gehring (2021) shows that in both support, surveys and actual voting for the EU, this negative experience can also be explained by the EU’s role in easing tensions between the region and the central states. For Eastern EU members, their membership and support are importantly linked to their historical experience with the Soviet Union.1

Threats, identity, and collaboration

Russia’s full-scale aggression in Ukraine has also sharply increased the perceived threat posed by possible Russian aggression to EU member states. Whether such external threats lead to a stronger common identity and further cooperation within a group has long been an important question.

Legend has it that the foundations of many nations were nurtured by an external threat. Think of the foundations of a united Germany after the American War of Independence against the British Empire or the war against France. The EU itself and its predecessors developed, at least in part, in response to the military threat from the Soviet Union and are thought to have had a unified effect of the Cold War (Bordalo et al. 2021).

However, no causal evidence has yet been found to support this claim outside of the laboratory setting. There were two major challenges in answering this question using real world examples. The first concerns the ability to distinguish the impact of an increased threat from other shocks. For example, increased threats are often accompanied by direct conflict, destruction, or actual cooperation, which makes it difficult to know whether there is an impact due to the conflict or threat (Todo and Kashiwagi 2021).

Second, relying entirely on comparisons before and after threats runs the risk of detecting a false relationship, but not necessarily a causal relationship. So, despite the popularity of threat-identity-conjecture, it is a fact and not just a pre-historic narrative.

Semi-experimental evidence from the 2014 Russian invasion of Ukraine

In a recent study (Gehring 2022), I used the 2014 Russian invasion of Crimea and parts of the Donbas region as a natural test to provide such evidence. Three features allow it. First, when the aggression was in Ukraine, it clearly affected the threat posed by Russia to the member states of the European Union. Second, the attack itself, and especially its precise timing, was unexpected at the time and thus could be considered an external shock (Gorodnichenko and Roland 2014, Gylfason and Wijkman 2014, Gylfason et al. 2014).

Third, there was a clear difference in the intensity of the threat among EU member states, creating cross-sectional differences in the intensity of that push. I would argue that the impact was greatest for Estonia and Latvia, as these two states have direct land borders with Russia and a large Russian minority population (used to justify Russia’s invasion).

Figure 1 Semi-experimental design

Comments: Figure 1 (a) reports a timeline for our analysis. Figure 1 (b) shows the treatment and control groups. Map based on Eurostat (2016). In Figure 1 (c) the minority shares are identified based on language. Figure 1 (d) shows a general average difference between our principal variables. The statistics show a gap of 95% confidence on average.

Using both qualitative and quantitative evidence based on newspaper articles and text analysis of Internet searches, I validate these assumptions. Citizens throughout the European Union feel more threatened, but the severity changes in line with this expectation.

With the data from the bi-annual Eurobarometer survey, I then experimentally test using a differentiated framework to see if the increased threat has a causal effect on EU identity, trust and willingness to cooperate within the group. The results are based on a representative sample of EU citizens for each member state.

The results clearly indicate a significant increase in the quality and quantity of a common EU identity. To put matters in perspective, the increase due to the increased Russian threat is of the same size as the initial difference between Poland (formerly strong EU identity) and Hungary (weak EU identity). This is more than double the initial difference between Germany (strong primary identity) and France (weak primary identity).

Figure 2 The main result

Comments: The figure shows the difference-in-difference coefficient by measuring the effect of the increased Russian threat with a 90% and 95% confidence interval (95% in light gray). All results are standard. Controls all regression for individual characteristics, including gender, age, level of education, labor market status, urban vs. rural areas, marital status and child attendance, specific effects of time and specific effects of member states. Standard errors are clustered at the regional level The number of pre-clinical measurements varies from two to five, and the number of post-treatment observations varies from one to three, depending on the availability of variables. The number of observations for EU identity is 24,885. For other results, it ranges from 25,569 to 68,408.

Further highlights the assumption that the effect lasts over time. It is even stronger for those who had personal experience with the Soviet Union and those who had personal or indirect experience with state repression in the Soviet era.

Predicted by socio-psychological theories, the increased identity translates into higher confidence in EU institutions as well as higher support for cooperation at the EU level. This desire for cooperation is not limited to defense policy, but extends to areas such as general foreign policy, taxes and regulations.

Assessing the impact of the Russian invasion of Ukraine in 2022

How can we assess the ongoing full-scale attack by Russia? In terms of the scope of the operation, the threat should be at least large and trigger a significant response. Eurobarometer survey results are not yet available for a quantitative assessment. However, an initial (‘flash’) Eurobarometer indicates a response consistent with the results of my paper.

Politically, Eastern Europeans are much more united than ever. Even a strong former member state with Russia like Bulgaria has taken a clear position. Western states, such as Italy and Germany, were initially hesitant, but in support of the referendum, both governments eventually lined up with other members for a joint response.

Of course, there may be motivations against common actions that cannot be overcome by a strong collective identity. Hungary provides a sad example. The state-controlled media has the power to restrain the response even while providing a biased view of actual events. However, unlike the previous period as part of the Visegrad group, Hungary is now at least explicitly excluded from the member states of the Eastern European Union.

Thus the unified effect of facing external threats seems to be clearly visible without additional quantitative evidence. Finland and Sweden are forced to abandon their neutrality and join NATO. Denmark is likely to opt out of EU defense policy in an upcoming referendum. The future will show whether a strong European identity will also help increase cooperation in areas beyond defense and foreign policy.

There can be hope. Many studies suggest that a common identity is a prerequisite for overcoming collective action problems and, for example, a common social security system (Bagues and Roth 2021). ). But in the end, it is up to the EU commissions and member state governments to turn this support into effective institutions and policies that justify the trust of its citizens.


Bagues, M and C Roth (2021), “International Communication and National Identity”,, 3 January.

Bordalo, P, M Tabellini and D Yang (2021), “Issue Salience and Political Stereotypes”,, 20 January.

De Grauwe, P (2012), “Confidence among eurozone leaders can produce self-fulfilling prophecies”,, 13 July.

Dehdari, S and K Gehring (2022), “The Origin of Common Identity: Evidence of Alsace-Loren”, American Economic Journal: Applied Economy 14 (1): 261–292.

Fouka, V and HJ Voth (2013), “Remembering Revenge: Car Selling During the German-Greek Conflict and the Euro Crisis”, CEPR Discussion Paper No. 9704.

Gehring, K. and S. Snyder (2018), “Towards a Greater Better? Nationality of EU Commissioners and Allocation of Budget to the European Union “, American Economic Journal: Economic Policy 10 (1): 214-239.

Gehring, K. (2021), “Overcoming History through Exit or Reunification – A Deeper Source of Support for the European Union”, American Political Science Review 115 (1): 199-217.

Gehring, K (2022), “Can External Threats Promote European Union Identity? Evidence of Russia’s Invasion of Ukraine”, The Economic Journal 132 (644): 1489-1516.

Gorodnichenko, Y and G Roland (2014), “What are the risks in Crimea?”,, 10 March.

Gylfason, T and P Wijkman (2014), “Russia’s Challenges to the EU’s Eastern Partnership”,, 25 January.

Gylfason, T, I Martínez-Zarzoso and P Wijkman (2014), “A Way Out of the Ukrainian Quagmire”,, 14 June.

Ochsner, C and F Rösel (2017), “Active History – The Case of the Turkish Siege of Vienna”, CESifo Working Paper Md. 6586.

Todo, Y and Y Kashiwagi (2021), “The Impact of Natural Disasters on the Perception of Others”,, 21 December.


1 Current events may enable or strengthen such dependencies, as shown by Ochsner and Roesel (2017).

Leave a Reply

Your email address will not be published.