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

Statistics and public reports
These sources are basically used as background information, providing
numbers and figures about federal-regional relations and regional affairs. The
statistics include for example figures on selected socio-economic variables in the
various regions, regional and federal budget allocations, financial utilisation,
federal-regional transfers, the manpower situation and the ethnic composition
of the regions. Statistics are often seen as reliable and unbiased sources of
information, particularly in a Western context, because they are meant to
records of relatively neutral phenomena and are based on allegedly objective
sources and measurements. But statistics are also social products, often made to
serve certain purposes. Most of the statistics collected for my thesis are
government produced, and might have been made deliberately to strengthen the
government’s image. Even statistics such as the popular census, which is

---

C M I
35
considered to build on objective numbers, should be critically examined. In
Ethiopia, the previous regimes tended to manipulate the population figures of
ethnic groups and the ethnic composition of regions to serve political interests.
This might also be the case during the current regime, particularly with the
federalisation along ethnic boundaries in mind. I observed several times that
informants, particularly those from minority ethnic groups, contested and
raised doubts about the accuracy of the census. Although I cannot verify the
validity of their points, they indicate that the popular census does not have as
much credibility as it should.
The public reports include plans, evaluations and assessment studies
compiled by governmental bodies and the World Bank, concerning the
federalisation process. Government-produced reports are obviously political
products and will be analysed and interpreted with that in mind. The reports
made by the World Bank are probably to a lesser extent results of political
considerations than the government produced ones. Still, the World Bank has
co-operated with governmental bodies in the collection of data, and their
objectivity should therefore not be taken for granted. Some of the reports are
published only in Amharic, but were translated into English by my interpreters.
During the analysis of the statistics and reports, I did not experience that the
findings directly contradicted the findings from the other kinds of sources. But I
experienced that the documents did not include many of the essential aspects
that came through in the interviews and the literature. An important reason for
this is probably that these aspects are primarily found outside the formal
sphere and are therefore not perceived as matters relevant for public statistics
and reports, or are of a politically sensitive nature.