The survival and demise of the State

What could cause a breakup of the State and what could preserve the union? When will the competing groups resort to violence? In the BSE Working Paper (No. 1028), “The Survival and Demise of the State: A Dynamic Theory of Secession,” Joan-Maria Esteban, Sabine Flamand, Massimo Morelli and Dominic Rohner build a theoretical model to shed light on these questions, particularly, at the dynamics and incentives at play among different disputing groups in a State.


Historically, most of the secessions from empires or states, or collapses, have been due to war or rebellion while some secession took place in peace. For instance, the Roman Empire voluntarily and peacefully split into two similarly large and similarly rich halves marked by some salient differences in social and religious norms. On the other hand, some breakups have been less harmonious such as in the case of Yugoslavia.

The academic literature on secessions is sizeable but mostly based on static models where trade-off between the preference heterogeneity of citizens and increasing returns to country size dominates the analysis. However, the authors note that the survival or demise of a state is an inherently dynamic phenomenon; the relative size of the groups, inequalities in affluence, and heterogeneity in preferences and social norms affect the inter-temporal trade-offs of the various groups which are involved in a strategic interaction. To address this gap in the literature, the authors propose a full-blown dynamic model of stability and break-down of states.


The authors build a theoretical model to offer insights on the incentives at play when many competing groups are involved in a conflict. Consider a country with two groups that can differ in size, economic productivity and preferences in the type of public good to be supplied. The types of public goods are culture, language, legislation, and other identity related collective decisions. Setting up or maintaining a State carries a cost. As long as a non-homogeneous State remains united, the group in power selects the public goods and determines how the surplus is shared. The rejection of the proposal by the opposition opens a period of costly conflict with a probabilistic outcome. If the incumbent group wins they retain power. But if the opposition wins they have the choice between seceding or becoming the group in power within the union and hence impose the public good and the allocation of the surplus. If secession takes place there is no further interaction with the other group. Otherwise, the two groups continue to form a union with possibly a new group in power.

Now vs. Future

Using a game theoretic argument, Esteban et al. are able to show that when no importance is ascribed to the future keeping the union is the best option for almost all pairs of population and surplus share. Importantly, as the value attached to the future becomes greater, keeping the union or a peaceful secession becomes less and less tenable and in the limit when people care about present and future at the same rate, the union is only possible when the opposition’s population share is sufficiently small. Consequently, when groups care a lot about future, conflict followed by secession becomes the most likely outcome.

The paper tries to convey some key messages such as remaining united implies that the public decisions need to be negotiated every period by groups with different preferences and priorities, while secession entails a cost today but no need to bargain with the other group ever again. In fact, through the lens of this model, they are able to shed light on many major conflicts and secessions that have happened in the past in many countries around the world.


In essence, Esteban et al. bring together two aspects of secession that have rarely been treated together and that provide novel predictions. In particular, in order to change the status quo, it is indispensable to challenge it. This has a cost and makes the outcome probabilistic. A second difference is that they model the problem as a dynamic problem so the increase in the cost of conflict also means that the other side knows that challenging the status quo in the future becomes less likely. However, authors note that finding empirical counterparts of the modeled aspects is tricky and difficult