Comparative Political Studies
44(5) 546 –571
© The Author(s) 2011
Reprints and permission: http://www.
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0010414010364350
http://cps.sagepub.com
Understanding the
Multinational Game:
Toward a Theory
of Asymmetrical
Federalism
Christina Isabel Zuber
1
Abstract
This article presents a baseline theory of asymmetrical federalism in multinational
states. Two arguments building on a game-theoretic foundation link central
and regional elites’ strategic choices to questions of federal stability. The first
argument concerns the creation of asymmetrical institutions. In a confrontation
game between the center and national minorities credibly threatening to exit
the framework, the center’s decision to grant asymmetrical autonomy ensures
mutual cooperation. Yet by extending the level of autonomy for minority
regions, federal asymmetry creates a third player, the nonadvantaged regions.
Consequently, the second argument models asymmetrical federalism as a “nested
game” where events in the ethnonational arena determine the payoffs in the
federal arena. Asymmetrical federal rules turn out to be inherently unstable
from a perspective that takes all actors in all arenas into account. A narrative of
the development of Russian federalism exemplifies the theory.
Keywords
asymmetrical federalism, federal stability, secession, multinational states,
Russian federalism, nested games
1
University of Cologne, Germany
Corresponding Author:
Christina Isabel Zuber, University of Cologne, Seminar für Politische Wissenschaft,
Gottfried-Keller-Strasse 1, D-50 931 Cologne, Germany
Email: zuberc@uni-koeln.de
Konstanzer Online-Publikations-System (KOPS)
URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-0-369490
Erschienen in: Comparative Political Studies ; 44 (2011), 5. - S. 546-571
https://dx.doi.org/10.1177/0010414010364350
Zuber 547
Multinational states result from the incorporation of different nations with
traditional homelands into a single state (Kymlicka, 1998, pp. 113-119). As
a consequence, they are confronted with the challenge of having to accom-
modate national minorities, defined as cultural groups voicing claims to
self-determination on a territorial basis (Keating, 2001, pp. 43, 61-62; Stepan,
2001, p. 327).
1
Multinational federal systems address these claims through
the institution of asymmetrical autonomy (Stepan, 2001, pp. 327-328). A
national minority is given special (i.e., asymmetrical) autonomy to realize on
a substate level what the dominant group enjoys on a state level—
self-determination as a ruling majority (Kymlicka, 1998, p. 133).
Some degree of asymmetry is a feature of all, not just multinational fed-
eral, political systems (Watts, 1999c, p. 29).
2
Substate units vary with regard
to size, demographic strength, or climate conditions. These given geographic,
cultural, social, and economic factors are commonly termed de facto or polit-
ical asymmetries. They have to be differentiated from de jure or constitutional
asymmetries where constitutional law treats constituent state units differ-
ently, granting some more self-governing powers than others (Burgess & Gress,
1999, p. 53; Kymlicka, 1998, p. 130). This can be observed in full-fledged
federations such as Canada, Belgium, Spain, India, and Russia but also in
constitutionally decentralized unions such as South Africa and the United
Kingdom (for a comprehensive overview of de jure asymmetrical autonomy
arrangements, see McGarry, 2005, p. 304).
Although asymmetrical solutions are of high practical relevance because
they try to answer to the specific challenges that arise from the nature of political
cleavages in multinational settings threatened by secessionist conflict, aca-
demic understanding of the dynamics they introduce and the mechanisms
through which they function is still wholly insufficient (Congleton, Kyriacou, &
Bacaria, 2003, p. 170; Keating, 1999, p. 86). Mostly on normative grounds,
scholars dealing with multinational states agree on the need for establishing
asymmetry to accommodate politically relevant claims to self-determination
on a territorial basis (Agranoff, 1999b, p. 97; Burgess & Gress, 1999, p. 53;
Watts, 1999a, p. 134). The potential of asymmetrical systems to remain
stable, however, is disputed, leaving open whether these solutions are desir-
able or detrimental (Watts, 1999c, p. 28). The original account of federal
asymmetry provided by Tarlton (1965) was followed by a long period of
scholarly silence. A seminal volume edited by Agranoff (1999a) rediscovered
asymmetrical federalism in 1999. Since then, some scholars have contrib-
uted to the subject, however, mostly at the level of typology or with analyses
of asymmetry in specific individual contexts (e.g., Burgess, 2008; Palermo,
Hrbek, Zwilling, & Alber, 2007; Rao & Singh, 2004; Swenden, 2002;
548 Comparative Political Studies 44(5)
von Beyme, 2005). As a consequence, we still lack an analytical understand-
ing
of asymmetrical federalism that could yield an answer to the question,
can federal asymmetry stabilize multinational states?
3
Asymmetry can be defined as a feature of multinational federal systems
through which nationality-based units (NBUs; home to national minorities)
de jure enjoy more rights than regional-based units (RBUs; home to the
national majority population) and maintain a differentiated relationship to the
center (Kymlicka, 1998). This suggests that the stability of asymmetrical fed-
eral political systems depends on the interaction of a complex set of actors
defined by both the majority–minority and the center–regional divide. This
article accounts for these patterns of interaction and presents a baseline
theory of asymmetrical federalism in multinational states that draws on a
game theoretical foundation.
Section 1 presents the first argument, which concerns the creation of
asymmetrical institutions. Given a situation where a national minority con-
fronts the center and credibly threatens secession, the center’s readiness to
grant asymmetrical autonomy to the minority region solves the prisoner’s
dilemma and ensures mutual cooperation. Yet by extending the level of
autonomy asymmetrically for minority regions (i.e., NBUs), federal asym-
metry creates a third player, the nonadvantaged regions (i.e., the RBUs).
Consequently, Section 2 models asymmetrical federalism as a “nested game”
(Tsebelis, 1990) where events in the ethnonational arena (structured by the
majority–minority divide) determine the payoffs in the federal arena (struc-
tured by the center–regional divide). Asymmetrical rules are shown to
sim ultaneously facilitate outcomes aimed at abolishing and at maintaining
the asymmetries and therefore turn out to be inherently unstable from a per-
spective that takes all actors in all arenas into account. Section 3 exemplifies
theoretical insights through a narrative of Russian federal institution building
in the 1990s, during the course of which the center and secessionist-minded
regional minority elites negotiated what Watts (1999b) described as “the
most complex current example of constitutional asymmetry” (p. 67). Section
4 concludes.
1. Why Multinational States Follow
Asymmetrical Rules
Within game theoretic approaches, “a self-enforcing institution is one in which
each player’s behaviour is a best response” and “no one has an incentive to
deviate from the behaviour associated with the institution” (Greif & Laitin,
2004, p. 633). Some of the more recent advances in federal theory have applied
Zuber 549
this concept of self-enforcing institutions to questions of federal stability, emp-
hasizing that federal rules have to be self-enforcing for officials at all levels of
government (De Figueiredo, McFaul, & Weingast, 2007; De Figueiredo &
Weingast, 2005; Filippov, Ordeshook, & Shvetsova, 2004). The theory of asym-
metrical federalism presented here takes up these insights and links asymmetry
and stability at the level of central and regional actors’ choices to yield explicit
arguments about whether asymmetrical arrangements restrain rule-deviant
and motivate rule-abiding behavior or whether the opposite holds.
The most imminent threat to stability in any federal system is secession
(Filippov et al., 2004, p. 13). A group of actors not only deviates from some
but also aims at abandoning the whole given set of rules. In multinational
states where national minorities aspire to self-determination, this problem is
aggravated because the most obvious strategy to realize this goal is by creat-
ing another sovereign nation-state (Hechter, 2000, p. 116). The stability of
multinational federal systems is therefore systematically challenged by their
very nature as multination states where the center has to confront sovereignty
claims of peripheral nations within the state’s borders. The first argument
about the origin of asymmetrical federalism departs from this point where a
national minority challenges the existing multinational framework and aims
at secession. At the same time, this starting condition determines the scope
of the theory. It does not hold universally across multinational systems but
depends on exogenous conditions that mobilize national minority secession-
ism. Possibly the most significant of such conditions would be a transition to
democracy (Hechter, 2000, p. 93). But also established democracies such as
Spain or Canada encounter secessionist challenges.
Young (1994) summarized the following particular characteristics of
secession games as opposed to games describing the formation of polities:
In instances of potential secession, however, the starting point is the
status quo. The definition of the community is contested. And rational
individuals may be concerned more with their potential losses through
secession than with their potential gains through the formation of new
communities. Finally, real existing governments are prepared to deploy
policies and power in order to influence those calculations of future
costs, and to alter them in reality. (p. 222)
In line with these assumptions, consider the following scenario of a contested
status quo in a multinational state:
4
Territory T is the traditional homeland of
a national minority. Both the national majority and the minority want to
maximize power over T: the majority because it is part of the overall state
550 Comparative Political Studies 44(5)
territory in which it constitutes the majority (Chapman & Roeder, 2007, p.
679),
the minority because it believes to have “some prima facie historical
claim” (Kymlicka, 1998, p. 140) over T. Both minority and majority are
represented by their respective political elite, hereafter termed Mi and Ma.
At the heart of T’s significance lie natural resources, for example, an oil
field. Revenue (R) from extracting the oil field constitutes the payoff to be
gained from controlling T.
5
Now consider the impact of institutions. In the most extreme case, a uni-
tary framework without power-sharing institutions, Ma controls key state
institutions and the state territory including T in virtue of demographic domi-
nance. Mi is denied participation in decision making at the central level and
special regional authority over T. Given oppression, any outcome is prefer-
able to the present one for Mi, hence no rational choice between strategies is
available (Young, 1994, p. 228). In a federal power-sharing system, how-
ever, the situation is more complex. Ma still controls key state institutions in
virtue of demographic dominance, yet the federal units enjoy a certain degree
of subnational authority and keep some amount of the revenue from regional
resources. The existing order thus allows Mi a rational choice over whether
to stick to the rules associated with the status quo.
In a symmetrical federal system, Mi is likely to disagree with the institu-
tional status quo because of a perceived inequality: Ma dominates not only
central state institutions that receive the federal share of all regional revenue
but also all RBUs keeping their respective share. Mi on the other hand con-
trols only its NBU and keep only the regional share of R. Mi thus considers
itself to be a relative loser within the symmetrical framework (McGarry,
2002, p. 434). Meanwhile, Mi could gain exclusive power over T through
secession, shifting R fully to Mi, a strong incentive to abandon the status quo.
Conversely, incentives for Ma at both the regional and the central level of
government to maintain the status quo are high.
At one point in time, Mi’s discontent leads to credible threats to secession.
Ma can insist on maintaining the contested status quo (defect) with the risk
of having to fight a secessionist war in due time if Mi realizes its threat or
agrees on a change of the status quo toward a new institutional framework
that is more attractive for Mi (cooperate). Mi on the other hand has to choose
whether to really push through with secession at any cost (defect) or agree to
remain within the multinational state (cooperate), a choice that will depend
on weighing the benefits of a new framework against the costs of secession,
potentially involving a military confrontation with Ma.
In a conventional symmetrical military confrontation, the costs of con-
flict for the central state would be comparatively low, whereas the
