Transborder Ethnic Ties and Repression of Ethnic Minorities

Abstract

Transborder ethnic ties constitute an alignment between an ethnic minority group and its external ethnic kin in another country. Similar to interstate alliances, in which a powerful state extends its deterrent capability to protect its protégé state against another state, a powerful external kin can extend its deterrent capability to protect a minority group against the target government. I employ deterrence theory and develop a game-theoretical model to analyze the triadic strategic interaction between a government, a minority group, and this minority group’s external kin. I argue that a minority group can only be effectively protected by its external kin when the external kin possesses a credible and capable threat of using force. I test the predictions of my model with a newly developed dataset covering 20 minority groups and 5 powerful ethnic kin. The results show that a highly credible and capable threat by an external kin significantly decreases the likelihood of government repression in the form of active and targeted political discrimination. Specifically, previous economic support to the minority group by their militarily powerful external kin constitutes a strong signal of a credible military threat, contributing to general deterrence and diminishing the probability of repression. This research contributes to the literature by moving beyond the mostly invariant factors, such as geographical proximity or the nature of ties between the groups, to understand the effect of an external ethnic kin on a minority’s security.

RESUMEN

Los lazos étnicos transfronterizos constituyen un alineamiento entre un grupo étnico minoritario y sus parientes étnicos externos residentes en otro país. De manera similar a lo que ocurre con las alianzas interestatales, en las que un Estado poderoso puede extender su capacidad de disuasión para proteger a su Estado protegido contra otro Estado, estos parientes étnicos externos poderosos pueden ampliar su capacidad de disuasión con el fin de proteger a un grupo minoritario contra el Gobierno objetivo. Empleamos la teoría de la disuasión y desarrollamos un modelo teórico de juegos con el fin de poder analizar la interacción estratégica triádica existente entre un Gobierno, un grupo minoritario y los parientes étnicos externos de este grupo minoritario. Argumentamos que un grupo minoritario sólo puede ser protegido de manera eficaz por sus parientes étnicos externos cuando éstos suponen una amenaza creíble y tienen capacidades para usar la fuerza. Ponemos a prueba las predicciones de nuestro modelo mediante un conjunto de datos recientemente desarrollado, que incluye 20 grupos minoritarios y 5 parientes étnicos poderosos. Los resultados demuestran que el hecho de que los parientes étnicos externos supongan una amenaza altamente creíble y capaz disminuye de manera significativa la probabilidad de represión gubernamental en forma de discriminación política activa y dirigida. En concreto, el apoyo económico que los parientes étnicos externos militarmente poderosos han concedido, anteriormente, al grupo minoritario constituye una fuerte señal de amenaza militar creíble, la cual contribuye a la disuasión general y disminuye la probabilidad de represión. Esta investigación contribuye a la literatura ya que llega más allá de los factores mayoritariamente invariables, como la proximidad geográfica o la naturaleza de los vínculos entre los grupos, para poder comprender el efecto que ejerce un parentesco étnico externo sobre la seguridad de una minoría.

RÉSUMÉ

Les liens ethniques transfrontaliers constituent un alignement entre un groupe d’une minorité ethnique et ses parents ethniques étrangers d’un autre pays. À l’instar des alliances interétatiques, dans lesquelles un État puissant fait profiter un État qu’il protège contre un autre État de sa capacité de dissuasion, les parents étrangers puissants peuvent faire profiter un groupe minoritaire de leur capacité de dissuasion à l’égard d’un gouvernement cible. J’emploie la théorie de la dissuasion et crée un modèle de théorie des jeux pour analyser l’interaction stratégique triadique entre un gouvernement, un groupe minoritaire et les parents étrangers de ce groupe minoritaire. J’affirme qu’un groupe minoritaire ne peut être protégé efficacement par un parent étranger que lorsque celui-ci est capable de menacer de façon crédible un recours à la force. J’évalue les prédictions de mon modèle à l’aide d’un nouvel ensemble de données couvrant 20 groupes minoritaires et 5 parents étrangers puissants. D’après les résultats, une menace concrète et extrêmement crédible d’un parent étranger diminue de façon importante la probabilité de répression gouvernementale sous la forme d’une discrimination politique active et ciblée. Plus précisément, un soutien économique antérieur apporté à un groupe minoritaire par son parent étranger puissant sur le plan militaire constitue un puissant signal de crédibilité d’une menace militaire. L’effet dissuasif est fort et la probabilité de répression chute. Ce travail de recherche contribue à la littérature en dépassant les facteurs majoritairement invariants, comme la proximité géographique ou la nature des liens entre les groupes, afin de comprendre les effets d’un parent ethnique étranger sur la sécurité d’un groupe minoritaire.

Keywords:

• Deterrence
• ethnicity
• minority
• repression

Acknowledgments

I am sincerely grateful to the anonymous reviewers and the editors for their invaluable and insightful feedback, which greatly improved the quality of this manuscript. Additionally, I extend my heartfelt appreciation to Stephen L. Quackenbush, Sheena Chestnut Greitens, Bryce W. Reeder, and Tola Olu Pearce for their thoughtful comments and suggestions on earlier drafts of this work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Named perfect deterrence theory because their conception of credibility aligns with the perfectness criterion of subgame perfect Nash equilibrium, as defined by Selten (1975).

2 The game-theoretical model and the original dataset partially draws upon the author’s unpublished dissertation, titled “Transborder Ethnic Alignments and the Interplay between Domestic and International Violence” (Yazici 2019).

3 Other forms of support by the External Kin, such as economic or diplomatic support, are considered credible signals of potential future military support or threat.

4 It should be noted that the Minority’s resistance to Repression always leads to Intrastate War (IW). This implies that when there is an MID, there might be an ongoing Intrastate War (IW) at the same time. They are considered separate outcomes because, in the presence of an MID between an External Kin and the Government, the conflict becomes internationalized, with the MID prevailing as the dominant conflict type.

5 See the Supplemental Material for the detailed solutions of the other equilibria.

6 See the Supplemental Material for the proof.

7 Ibid.

8 Ibid.

9 Ibid.

10 See the Supplemental Material for more information on the sample selection process.

11 Bilateral trade between the External Kin and Government is excluded from the empirical models due to high amount of missing data.

12 The composite Credibility variable is the sum/extraction of the z scores of these variables. See Supplemental Material for detailed operationalization of the variables.

13 The Composite Index of National Capability (CINC) score aggregates the six individual measured components of national material capabilities (total population, urban population, military personnel, military expenditures, primary energy consumption, and iron and steel production) into a single value per state-year.

14 I could alternatively use time dummies or splines as suggested by Beck, Katz, and Tucker (1998), however the cubic polynomial approximation outperforms the time dummies and splines. Also, cubic polynomial approximation is easier to implement and simpler to interpret (Carter and Signorino 2010).

15 See Carter and Signorino (2010, 282) for the original version of the equation.

16 Table 9 in the Supplemental Appendix shows that results of the random effects and fixed effects linear regression models. Even though OLS is not the ideal estimator for a binary dependent variable, the results show the directions of the effects are consistent with the logistic regression models.

17 Time-invariant variables are automatically excluded in the mixed effects model, which estimates the fixed effects over time within each dyad.

18 The results of robustness tests, as presented in the linear models in Table 7, align largely with the findings of the main models. Notably, diplomatic support from the External Kin and military alliance between the External Kin and Government show statistically more significant effects on the likelihood of repression in these linear models.

19 Results of the F-test: F = 10.285, p-value < 2.2e-16.

20 Estimation of these quantities of interest are based on the pooled logistic regression.

21 In the second scenario, either the maximum or minimum values of the credibility variables are used depending on the direction of their expected effects. The ratio of CINC scores is set equal to its third quartile. Control variables are held at their average values.