Thread by @glad_1723

Does aid increase conflict?

This week on #econfriday: @KaiGehring1, @LennartKap, & @mhl_wong investigate the effect of aid projects on conflict at the subnational level in Africa. Aid doesn't seem to be that bad at all!

Here's the WP: papers.ssrn.com/sol3/papers.cf…

Development aid is thought as one important way to help poor countries. However, prior research raises concern that aid can lead to
civil conflict (@DrNathanNunn & Qian 2014).

There is quite some divide between micro-level & macro-level studies on the relationship between aid & conflict. This paper closes a gap between micro-level country studies and macro-level cross-country studies.

With geo-referenced aid data (@AidData), the authors look at two specific donors: The World Bank and China.

While the WB prefers high conditionality (human rights and ecology) for allocating aid, China prefers to attach fewer strings and
focus on mutual economic benefits.

The FE & IV identification considers the cautionary note of @paulchristianWB & @cbb2cornell (2017) on interacted IVs with many robustness tests & correcting problematic trends.

Here is an example for the WB on problematic trends to their estimation and their actual trend used:

Now to the results. WB and Chinese aid reduce(!) rather than fuels conflict, on average. This is especially true for the transport and finance sector.
The OLS FE results turn out to be very similar to the IV findings.

A closer look reveals some interesting heterogeneity

- Aid leads to less lethal violence by governments against civilians:

Potential shaming by the international aid community imposes opportunity costs of losing aid.

- But: Chinese aid seems to make non-lethal government repression more likely in recipient areas: This corresponds to previous evidence (@roudabehkishi).

Whereas aid doesn't seem to fuel conflict, this paper definitely fuels interest in the topic.

We think: An important read for academics as well as policymakers to find one's position in aid questions.