援助的主要目的在於受贈國的發展，而許多國家發生內戰歸咎於經濟情況不佳，那麼，援助是否可以降低內戰的風險？現今的文獻認為援助是具有戰略性的，也就是必須要考慮援助國以及受贈國的互動，因此援助可視為偏向政府的干預，本文認為這樣的干預會惡化政治不穩定國家中的承諾問題（commitment problems）、增加權力分享（power sharing）的成本，同時也降低戰爭成本（war costs），阻礙衝突團體間和平協議的達成而增加內戰爆發的風險。本文使 用縱橫資料（time-series and cross section data），以多層次「二元勝算對數模型」 （binary logit model）進行分析，結果發現當援助越多，內戰爆發的機率反而會上升。   

Civil war often occurs in poor countries that also attract foreign aid from developed countries, which raises the question of whether foreign aid is capable of reducing the likelihood of civil war. Despite such aid being known as charitybased foreign policy, previous studies have pointed out that aid is delivered strategically. Since aid targets the incumbent government, it could be seen as a pro-government-biased intervention. I argue that it is the biased interfering signal that causes commitment problems and changing costs of power-sharing and civil war between warring groups to discourage peace agreements. This paper analyzes time-series and cross-sectional data from 1947 to 2017 with multilevel binary logit models, and finds a positive relationship between foreign aid and the onset of civil war, which also confirms the theory.