Ethnic Federalism in a Dominant Party State: The Ethiopian Experience 1991-2000 Lovise Aalen r 2002: 2

What determines the functioning of federal systems?
An understanding of the operation of federal systems requires an analysis of
more than formal governmental structures. As Watts (1994:13) has pointed
out, it requires an analysis of the interaction of societies, structures and
processes. The functioning of the formal and institutional framework and the
character of intergovernmental relations is dependent on the underlying
political, social and economic structures and the roles played by inter alia
interest groups, individual leaders, informal elites and political parties.
Livingston (1956) claims that federalism is a function not of constitutions
but of societies. He judges a state as federal not by its federal government and
its legal structures, but by the way social, political and economic interests were
organised. His socio-cultural theory of federalism implies that a society is
federal in the sense that its major linguistic, cultural, ethnic, economic or other
interests are territorially grouped, but the diversities are not so great that there
is no room for political integration. A federal government is then the device by
which federal qualities of a society are articulated and protected. King (1982:
126-127) criticises Livingston by arguing that his socio-cultural theory does not
make us more able to distinguish federal states from non-federal states. A
society with a “federal character” does not necessarily choose a federal
governmental system, but could choose other forms of proportional
representation. Nevertheless, Livingston’s theory adds more reality to the study
of federalism by paying attention to the underlying social realities, and
modifies the heavy emphasis being given to constitutional change (Tarlton
1965: 866).
Symmetry and asymmetry
Tarlton (1965) bases his study on the same assumptions as Livingston, that
federations are products of society. But he looks more into how social,
economic and political conditions are affecting the relationship between the
federal government and its constituent units. He points to a fact that had been
largely ignored by earlier studies, that federal systems may be more or less
federated throughout its parts, and that the “federalness” and commitment to
federalism is disparate among the various regions in a federal system (Mullins
and Saunders 1994: 42-43). In an ideal symmetrical federal model, the
constituent units have equal size of territory and population and the same
economic features, cultures, social groupings and political institutions. They

---

C M I
19
share interests and concerns with the centre and with the system as a whole,
and thus, all the regional states have a strong commitment to federal unity and
have all the same relationship to the central authority (Tarton 1964: 868). In
an asymmetrical federal model, the federation is composed of regional units
which correspond to the differences in interest, character and make-up that
exist within the whole society. The diversity in the larger society finds political
expression through the regional governments, possessing varying degrees of
autonomy and power. In an asymmetrical system, it would therefore be
difficult to discern interests that could be clearly considered to be mutual or
national in scope, and the various regions all have different types of
relationships and commitments to the central authority and the system as a
whole (Tarlton 1964: 869).
All federal systems have elements of both symmetry and asymmetry. Tarlton
suggests that the higher degree of symmetry a federal system has, the more
likely it is that the federation will be viable and suitable. The more a system is
asymmetrical, the more unlikely is it that the federation will develop
harmoniously. When diversity dominates, the potential for secession is higher,
and this necessitates increased central control and authority to make the system
work. In current studies of federations, the question of constitutional
asymmetry is frequently discussed as a way of ensuring unity in multi-national
states (Saunders and Mullins 1994). Constitutional asymmetry means that the
constitution has defined that regional units have different relationships to the
federal level in the state and different degrees of autonomy from the central
authority. If we follow Tarlton’s argument, constitutional asymmetry will
enhance the degree of diversity in the federation and thereby decrease the
chances of a harmonious relationship between regions and federal government.
This contradicts the assumptions made of those who claim that constitutional
asymmetry will enhance stability and decrease conflict, because minorities are
given protection against domination. According to Tarlton’s theory, federalism
in general, and constitutionally asymmetric federalism in particular, might not
be as good an instrument for accommodating differences as has been claimed
by others. For federalism to be viable, the common interests must outweigh the
potentially divisive regional interests.
Fiscal issues
Tarlton’s asymmetry theory focuses on the fact that there is always a tension
between the formal network and the experience and behaviour of the players
within the network (Saunders and Mullins 1994: 44). In federal systems, this
tension is present between the legal and institutional framework and the actual
relationship between regional units, the central authority and the system as a
whole, which is determined by underlying social, economic and political
variations. The regional states are able and willing to implement the
constitutional provisions of regional autonomy and representation on federal
level according to their political, economic and social situation. A key in
determining the actual implementation of constitutional rights is the fiscal
relationship between the regions and the centre and among the regions. Fiscal
issues are essential parts of intergovernmental relations, and reflect the

---

C M I
20
interplay between structures (the provisions in the federal constitution),
processes (the actual fiscal policies that are implemented), and society (the
economic realities of the federation).
According to Watts (1994:129) the allocations of fiscal resources to each
level of government within the federation is important for two reasons. Firstly,
these resources enable or constrain the governments in exercising their
constitutionally assigned legal and executive responsibilities. Secondly, taxing
power and expenditures are in themselves important instruments for regulating
the economy. In most federations, the constitution defines the expenditure
responsibilities and revenue sources of each level of government, but it has
become apparent that it is difficult to create a federal constitution where the
allocation of autonomous revenue resources match precisely the expenditure
responsibilities. Imbalances are therefore created, and there is a need to rely on
financial transfers to correct these. Some imbalances are vertical, between the
higher and lower levels of government, while others are horizontal between the
various constituent units. In most federations, vertical imbalances are corrected
by sharing of specific tax earnings and conditional or unconditional grants to
improve the capacity of the lower units, while the horizontal imbalances are
corrected by general equalising transfers and grants in aid to even out the
regional disparities. When taxes are shared between the central and the
regional levels of government, some proportion of the amount collected in the
jurisdiction of the regional government is returned to that regional government.
This amount could be used directly in the region, and the fiscal planning of the
regional government is improved with certainty. This finally enhances the
constituent units’ economic independence and autonomy from the centre (Bahl
and Linn1994: 8). The disadvantage of this kind of policy is that poorer
regions with a meagre local tax base will loose and the horizontal imbalances
will increase. This again will lead to what Tarlton warns of, increased
asymmetry in the federal system. The policy of minimising horizontal
imbalances through equalising grants from the centre, however, might
undermine the incentives for regional revenue mobilisation and thereby
increase the regional governments’ reliance on central transfers. This might, in
the end, undermine the constituent units’ autonomy. But if we follow Tarlton’s
theory, correction of horizontal imbalances through equalisation will lead to a
more sustainable federation in the long run, because it will lead to more
symmetry among the units. Although the correction of vertical imbalances will
increase the regional governments’ autonomy in the short run, the following
asymmetry between the constituent units will in the long run lead to more
control and coercion from the centre because the disparities between the units
make unity more fragile and hard to maintain.
Political parties
The structure of party systems and the role played by political parties are other
important determinants of the operation of federal systems. The way political