Conflict and Coffee in Burundi 145The effects of this on the economy of Burundi were adverse and have been
observed to have been partly responsible for the disturbances during the sec-
ond half of that decade.136
Failure to renew the agreement in 1989 is partly attributable to the waning
and eventual collapse of Cold War politics. The United States' interest in the
agreement was part of its Cold War zero-sum competition for producer coun-
try allies with the defunct Soviet Union. It hoped that by stabilising coffee-
producing countries' receipts from the crop export by playing a role in shield-
ing them from world market price volatility, they would be less vulnerable to
'poaching' by the Soviet Union. With the diminishing Cold War discourse in
the late 1980s, the ICA's strategic relevance turned into an unjustifiable loss
to the powerful coffee-roasting lobby on Capitol Hill. Additionally, discontent
was growing among some major producers that the main winners in the
scheme were Brazil and Colombia, that some smaller producers had taken
advantage of the scheme to surpass their allotted quotas and had in fact been
exporting to East European countries. The ICA collapsed in 1989, the victim
of a changed and changing world.
Between 1986 and 1992, the price of coffee dropped by 75% and Burundi's
average growth in export revenues correspondingly dropped from 3.3% for
the period 1965-1980, to -1.9% for the period 1980-1990. This was thus a
period of heightening tension between the increasingly impoverished and neg-
lected rural population and the privileged urban elite. Nkurunziza and
Ngaruko's "predation, repression and rebellion" model helps in explaining the
outbreak of the 1988 Ntega-Marangara violence in relation to the shrinking of
the government's clientalist reservoir. The model amply illustrates the posi-
tive correlation between state predation and declining government capacity
owing to a shrinking prey. During periods of declining GDP per capita, preda-
tion rises. This in turn provokes rebellion by the peasants. The government
reacts with predictability, heavy-handedly putting down the rebellion."'
The relevance and place of international financial aid is pertinent in this
discussion. The export revenue deriving from coffee in Burundi provides less
than half of government development and recurrent expenditure estimates.
Multilateral and bilateral donors and development partners have filled this
gap. In the period between 1989 and 1991 more than half the finances spent
by the Burundi government came from these sources and international non-
governmental organ is at ion^.^^^ This would explain the relative closeness
between Bumndi peasants and these organisations as contended by Laely.'j9
External development financing mainly put up the infrastructure projects
built during the Bagaza period.
The twln phenomena of the Burundi government's disproportionate
reliance on external financial receipts and the country's rural population's
near indifference to central authority is made more compelling when one
looks at the post-1993 events and the 'creeping coup' that came to pass in