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

party advisors. Tigray, on the opposite side, has enjoyed relative independence
from the central party, mainly due to the fact that the regional party of Tigray
and the central party leaders have shared interests.

---

C M I
100
Chapter 7: Conclusion
This thesis has aimed at analysing both the legal and the practical aspects of
the federalisation process in Ethiopia. The first objective was to discuss whether
the Ethiopian federal model de jure corresponds or contradicts with the central
principles of federalism and federal practices. The second objective was to
explore the de facto relations between the federal and regional levels of
government and uncover which factors determine the nature of these relations.
The overall conclusion of the analysis is that although the Ethiopian de jure
model meets the requirements of a federal system, the de facto relations
between the central government and the federal units are so centralised that the
federal division of power is severely undermined. A major reason behind this is
the centralised structure of the party system. The asymmetry of the federal units
and their lack of an independent financial base do also contribute to a
weakening of the federal division of power.
A summary of the findings
Various aspects of theories on federalism and federations provided a useful
framework for the analysis of the Ethiopian federal system. Studies of
federalism, the normative basis for federations, underlined the importance of
looking beyond ideological rhetoric in the justification of federations, and
instead explore the political use of federalism. Who benefits from a
reconstruction of the state along federal lines and under which circumstances is
the federal model introduced? The central argument in chapter 4 is that the
federalisation of Ethiopia was introduced in a way that made it lose legitimacy
even before it was implemented. The architect of the ethnic federal model, the
TPLF, had a commitment to the principle of national self-determination during
the struggle against the previous regime, and a desire to reconstruct the
Ethiopian state in order to neutralise the old repressive elites. But the concern
for its own survival as the ruling power led it to impose the federal system from
above, practically excluding the other political forces in the country. This
essentially undemocratic process made the political opposition view the new
federal model as a purely TPLF-dominated project and not as an undertaking
favouring the whole Ethiopian population.
King’s (1982) minimalist definition of federations and other studies of the
institutional features of federations was a starting point for the analysis of de
legal framework for the Ethiopian federation. In King’s definition, the
Ethiopian system is qualified as federal. The Ethiopian constitution guarantees
the constituent units representation at the federal level, and this representation
cannot be changed without the consent of the majority of the constituent units.
But there are some peculiarities in the Ethiopian model, which make it quite
dissimilar to other federal systems of the world. In the constitution, the
sovereignty of the state is not given to the federal units as in other federations,
but to “the nations, nationalities and peoples” of Ethiopia. The right to
national self-determination includes the right to secede. But this extreme right

---

C M I
101
does not fit in with the powers of the federal units, which are relatively meagre,
and gives the central government the upper hand. The odd combination of the
right to secession and the meagre regional powers distinguishes the Ethiopian
federal model from other federal systems that are relevant to compare with.
The Russian, Canadian and Spanish federations do not offer the federal units a
right to secession, but have given them other extensive remedies for self-
determination. Other peculiarities in the Ethiopian legal framework are the
absence of an independent constitutional court and the lack of checks on the
head of the central executive, the Prime Minister. These aspects give the party
in power at the centre a large space of action.
Tarlton’s (1965) and Riker’s (1964) theories underline that the functioning
of federal systems cannot be predetermined or understood by only looking at
the ideological justifications or the constitutional framework for federations.
This argument was the point of departure for the analysis of the de facto
federal system of Ethiopia in chapter 6. Tarlton’s theory on symmetry and
asymmetry assumes that the more dissimilar the constituent units of a
federation are, the more would the central government intervene in regional
affairs to keep the federation together. The exploration of the Ethiopian federal
units revealed that the federation is highly asymmetrical when it comes to
ethnic composition, interethnic relations and socio-economic development level.
The study of fiscal relations between the regional states and the central
government uncovered that the central government has been granted the most
lucrative revenue sources in the country. Consequently, the regional states’
financial bases are so weak that they remain dependent on federal grants and
subsidies to carry out their constitutionally defined responsibilities. Especially
the lowland regions, which have the lowest level of socio-economic
development, are totally dependent on federal transfers, while the four highland
regions at least manage to fund about a fourth of their own expenditures.
Riker’s studies of federations underline the importance of exploring the
political context in which a federal system is implemented. He claims that the
single most important factor in determining the character of a federal system is
the operation of the party system. The analysis of the party system in Ethiopia
showed that the coalition in power controls all the regional states in the
federation either through its member parties or through affiliate parties. The
EPRDF’s centralised organisational structure makes the regional governments
in the various units accountable to the party hierarchy above instead of to the
regional institutions and the electorate. The four lowland regions, which are
run by affiliates of the EPRDF, face the most serious interference from the
central government. Advisors from the Regional Affairs Department at the
Prime Minister’s Office together with informal advisors from the central party
virtually control the regional governments there. The four highland regions,
which are run by EPRDF member parties, do not experience the same form of
central interference. Nevertheless, the informal and unistitutionalised
intervention from the central party through TPLF cadres undermines their
chance to act independently from the federal level. The studies from SNNPRS
and Tigray show that the status of the regional EPRDF parties and their
relationship to the central leadership are decisive in determining the degree of
central party interference. In SNNPRS, where the regional party has a weak

---

C M I
102
position, TPLF/EPRDF cadres interfere directly in regional affairs and are
active even at local level to ensure the interests of the EPRDF. In Tigray, the
position of the regional party is strong and the regional party leaders share
interest with the central party leaders. The regional government is therefore
able to act relatively independent from the central government.