This report examines connections between the Indian Ocean tsunami of December 26, 2004 and
Sri Lanka‘s ongoing civil war between the Sinhalese majority-dominated government and the
Liberation Tigers of Tamil Eelam (LTTE), better known as the ?Tamil Tigers?. An initial
question of interest was whether the tsunami and ensuing relief efforts precipitated a renewal of
fighting and the eventual end, almost three years to the day after the tsunami, of a ceasefire
agreement signed in 2002. Recent Government of Sri Lanka (GoSL) victories on the battlefield
raise a more provocative question: did the tsunami and relief efforts tilt the balance of power in
favor of GoSL? While many studies have looked at the role of corrupt agents and insurgents in
co-opting relief funds, few have looked at the possibility that disasters and disaster relief could
create incentives for a government to return to war.
While this report comes to no definitive conclusions, it garners considerable evidence supporting
an affirmative answer to this question. The author has linked district-level data on financial
assistance flows with more detailed community-level data on population distribution, tsunami
damage, and recovery. Government data sources were complemented by a survey of 141
communities in five heavily affected districts in Eastern and Southern Sri Lanka.
With respect to tsunami damage, the report highlights the following key facts
? As is now well known, over two-thirds of tsunami damage was concentrated in Northern
and Eastern Provinces, home to much of the Tamil and Muslim minorities.
? Less well understood is the extent of Tamil suffering in what some have perceived to
have been a Muslim tragedy. Estimation of the ethnic distribution of damage using
community-level data suggest that 48% of destroyed homes belonged to Tamils, 23% to
Muslims, and 29% to Sinhalese. A similar proportion of deaths were to Tamils.
? The complex political ecology and history of population transfer in Eastern Province
placed large, vulnerable Tamil communities in close proximity to the coast. In majorityMuslim
Ampara, the most heavily affected district, only 30% of the population was
Tamil yet 52% of destroyed homes belonged to Tamils. Eastern Province Tamils once
constituted the principal base of financial, logistical, and military support for the LTTE.
? The comprehensive population transfer policies of the LTTE had forced out the Muslims
of Northern Province long before the tsunami, rendering a substantial Tamil impact.
? The LTTE sustained an uncertain amount of damage to military, logistical, and
administrative capacity during the conflict, though probably minimal.
With respect to the post-tsunami relief effort, the report reinforces media reports of regional bias
in the aid delivery and housing reconstruction
? Expenditures in the all important housing sector were considerably higher, on a perdestroyed-home
basis, in Sinhala-majority Southern and Western Province than in
Northern and Eastern Province. Expenditures were particularly low in the most heavily
affected areas of Eastern Province, Ampara and Batticaloa District.
? Housing expenditures extraordinarily high in Sinhalese-majority areas with high levels of
support and identification with the ruling Sri Lankan Freedom Party (SLFP), President
Mahinda Rajapakse, elected in late-2005 on an anti-ceasefire platform, and SLFP‘s
virulently nationalist coalition partners the Janatha Vimukhti Peramuna (JVP).
Rajapakse‘s home district of Hambantota received four times the average expenditure per
home and six times that received by the most heavily affected eastern district.
? Some of this bias reflected the failure to convert funding allocations into expenditures,
possibly due to the conflict, but it primarily reflected bias in initial donor allocations.
? The bias towards the south and Hambantota in particular were consistent across bilateral,
multilateral, and non-governmental donors. Bilateral donors tended to exert a strong bias
against areas of LTTE control while multilateral donors balanced these tendencies.
? Funding biases resulted in divergent rates of housing reconstruction. By the end of 2006,
only 20% of homes had been rebuilt in Ampara, 30% in Batticaloa. In Southern Province,
Hambantota had received more than three new houses for every one destroyed (4,065
built versus 1,290 destroyed) and Matara had almost reached the break-even point, while
the opposition stronghold of Galle had seen only 50% of homes replaced.
Taken together, the tsunami and the relief effort constituted at minimum a net $150 million
dollar transfer to the southwest of the country relative to the northeast before taking into account
the cost in loss of life. The unfavorable distribution of within-district impacts for the Tamil
population could only have exacerbated LTTE challenges in resource extraction and manpower
recruitment. While these imbalances could not have been decisive in a renewed conflict, they
could certainly have altered the GoSL calculus for assessing the potential costs and benefits to
pursuing a military solution. The two years immediately following the tsunami saw annual
increases in defense spending amounting to 40% per year. While inflation rose precipitously,
some of the burdens of new expenditures, particularly on the SLFP support base, were mitigated
by access to tsunami relief. GoSL has promised, credibly, that by the fourth anniversary of the
tsunami they will have regained administrative and military control of the entire country.
The report ends with some modest recommendations and difficult questions for the way forward.
To avoid misallocations of humanitarian resources in complex political ecologies, it is
recommended that social vulnerability surveillance mechanisms be established, with possibilities
for rapid linkage to disaster assessment mechanisms. Challenging questions persist for the future
of aid delivery. It is easy to say that the renewal of conflict, even for those who would be happy
to see the end of the LTTE, represent a suboptimal outcome from a humanitarian relief
standpoint due to the damage, the instability, the further deterioration of conflict resolution
mechanisms, and the high likelihood that the GoSL victory merely represents a shift to a
guerrilla war, not an end to war. Yet it is difficult to identify approaches to the delivery of
humanitarian assistance in such a contested context that would not create the incentives for
belligerence.