Thread by @KaiGehring1

Russian invasion in Crimea 🡺 Russian threat for Eastern EU states ⇑

DiD design comparing EU states with higher vs. lower risk

External threat ⇑ 🡺 stronger EU identity

Higher trust in EU institutions & higher willingness to cooperate at EU level

papers.ssrn.com/sol3/papers.cf…

Research question: Can external threats really create a joint identity and foster cooperation? This is a key question in social psychology, that is also relevant for economics and political science, as identity influences political preferences &decision-making in federal systems?

Sociologist Georg Simmel in his seminal work: heterogeneous groups ”will easily break apart unless a danger, shared by all, forces them together”

Montesquieu: “Republics require a constant fear of an external threat to hold them together”

Empirical challenge in prior papers:
1) Disentangle effect of threat from (a.) conflict, (b.) serving in an army, (c) occupation, (d) destruction
2) Find valid counterfactual, or else only before/after time series.
Mixed evidence for Cold War Threat on bipartisan coop in US.

My setting:
Struggle between EU & Russia over control in Eastern Europe
Former Soviet&Warsaw Pact states first become independent, in 2004 EU members

Ukraine last buffer state between blocs,association treaty negotiations create further tensions

Russian actions deemed unlikely

Russian invasion in Crimea as natural experiment, unexpected shock provides time-series variation

Compare high and low threat EU member states, provides cross-sectional variation

Implement 2x2 quasi-experimental design, Difference-in-Differences

High-threat def:
a.) Direct land border with Russia
b.) Significant Russian minority (played crucial role in Ukraine)

High-threat: Estonia and Latvia

Control: other eastern EU, also former Warsaw pact

Validity check:
Use google searches for Russian army threat, higher jump in high-treat states

Measurement of identity:

Questions from @EurobarometerEU (three for robustness)

(Bi-)yearly with 1000 obs per member state

Stated preferences, but shown to influence behavior in experiments and real-world (revealed pref.)

Why EU identity and not national?

Based on social psychology, the identities that are activated by threat

a.) are the ones that are most salient (conflict between EU and Russia) and
b.) that can supposedly help in a potential real conflict (small states incapable)

Econometrics checks (among others):

a.) No differential trends in composition in low and high threat states
b.) Rule out influence of other events in between survey measurement (e.g. drop observation overlapping with refugee crisis)

Results: Clear increase in joint European identity.

First causal evidence outside a laboratory experiment that external (military) threats can strengthen identity.

Effect of threat, while there is not (yet) an actual conflict

Size of effect of military threat: Very sizeable impact

- 2/3 of existing cross-country standard deviation in EU identity

- Similar to change in EU identity in Ireland after Brexit (large economic shock)

Other measures, channels and consequences:

Psychological change: Related to higher trust in EU institutions

Policy: Higher willingness to cooperate at EU level, both on policies that offer potential protection and on EU enlargement

More details in the paper on
-Theory
-Econometrics
-Many robustness tests

Beautiful graphs and coefplots thanks to @jan_gromadzki, who was a great help with this paper! I could not have done it without you.
Any feedback appreciated, dm or email to [e-mail]

Who might be interested? @austinlwright @austinmstrange @CatherineDVries @sarahobolt @mcopelov @mhallerberg @AndrewWattEU @fuchs_andreas @valentin_lang @ValentimVicente @FuestClemens @vz_eich @D_Langenmayr @glad_1723 @essobecker @stefwalter__ @garritzmannj @tabouchadi

Also thanks to any friends and colleagues with way more followers than me for sharing (not sure this touches directly upon your specific research, but help appreciated):
@schnellenbachj @BachmannRudi @DinaPomeranz @MaxCRoser @jsuedekum @SDullien
@econ_uzh @IPZ_ch @polito_uzh

Happy to provide some more background and details for journalists interested in European Union, EU integration, EU identity.
@schieritz @PhilipPlickert @derspiegel @NZZ @NZZ @welt @EURACTIV

@threadreaderapp unroll