援助削减是否助长非洲暴力冲突？如何促进和平

The last 18 months have seen a historic decline in development aid budgets from various donor countries, in a period where many of them are earmarking more funds for rearmament. The biggest waves have been made by the abrupt and massive reduction in American aid.

Less than a week after President Donald Trump’s inauguration in January 2025, his administration ordered the immediate suspension of all programmes run by the United States Agency for International Development (USAID) – historically the world’s largest national humanitarian donor. Officers had hours to vacate their posts, local contracts were terminated, and medical and food supply chains ground to a halt.

Researchers have already issued warnings of the severe health toll. A study published in Lancet Global Health estimates that USAID helped prevent nearly 92 million deaths between 2001 and 2021. It also projects more than 14 million additional deaths by 2030 if the cuts continue.

The rapid and massive decline in foreign aid experienced by various recipient countries can have very substantial consequences, ranging from health outcomes to armed conflict. Recent evidence suggests what types of international support could allow countries to build back better and foster peace and prosperity for coming generations.

Development aid also matters for various socio-economic outcomes. Hence one would expect aid cuts to have wide-ranging societal consequences as well. Drawing on our past work on political economic topics and development economics, we have decided to investigate this question.

Our new study finds that the sudden cuts to USAID programmes are associated with surging armed conflict in various regions across Africa.

We cross-referenced geolocated data on historical USAID disbursements with violent incidents recorded by the Armed Conflict Location and Event Data (ACLED) project across 870 African regions and over nearly two years. The project monitors conflicts.

The data paint a clear picture: in areas that had received the most American aid, the probability of conflict increased by 3.1 percentage points after January 2025 relative to the control group of places not benefiting from USAID. To account for the fact that places with US aid and others may be different from each other, we filter out the time-invariant conflict risk of particular places and focus solely on changes over time.

This amounts to a relative rise in conflict risk of 6.5%. Battles increased by nearly 7%, protests and riots by more than 5%, and conflict-related deaths by roughly 9%. These effects appear within the first few weeks and intensify over time.

International aid has always divided opinion. Some economists see it as a lever for stability; others, as fertile ground for corruption and conflict, by creating resources worth fighting over. Academic work has indeed found evidence pointing in both peace-promoting and conflict-fuelling directions when it comes to gradual aid flows. But a massive, sudden withdrawal follows a different logic – and our results confirm this.

When aid vanishes abruptly, economic opportunities contract very quickly. The cost of rebellion mechanically decreases, as participants have less to lose, yet many of the underlying reasons for conflict – economic rents, territorial disputes, ethnic tensions, political grievances – remain intact.

This mechanism may explain why violence flares up precisely where aid had been most present. Our data also show that institutions play a buffering role: where governance is stronger, the destabilising effects are markedly weaker.

We find no evidence, however, that the presence of Chinese aid projects softens the impact of USAID’s withdrawal.

What to do next?

Sub-Saharan Africa, where USAID primarily funded health, food security and basic services, is also the continent where state fragility is most widespread. Our estimates likely represent a lower bound: other European donors (including France) have begun reducing their own contributions. If these cuts accumulate, the effects could exceed what we are measuring so far.

Since today’s conflicts are the best predictor of tomorrow’s, a spike in violence can quickly become a trap that is very hard to escape.

Hence, the policy implications of these results are manifold. As far as other major donors are concerned, one interpretation of our findings is that at present they may want to act slowly and cautiously. After having witnessed a very significant drop in global aid and a surge in armed fighting, a rapid and massive disengagement of other major donor countries may very well accentuate the serious effects that we have documented.

Rethinking development aid

In the wake of USAID’s dismantlement there is an urgent need to rethink development aid. It is the right moment to ask difficult questions and reassess the type of international support that is best suited to promote peace and prosperity. Many questions arise: how can aid be made more resilient, less dependent on a single donor, more firmly rooted in local institutions?

A growing academic literature on optimal policies for peace can serve as a useful guide. A synthesis of recent research highlights the role of sound institutions, security guarantees and productivity-promoting policies. While institution building is often home-grown, international cooperation is key for the other two aspects. In particular, a series of recent studies documents the paramount importance of UN blue helmets for guaranteeing security. The presence of United Nations troops prevents the worst atrocities, suggesting that what the international community should do is to increase the blue helmets budget rather than cutting it.

Relatedly, financial aid can play a key role when it comes to investments into the economy, with wide-ranging consequences. When people have opportunities and perspective in life, they are much less likely to engage in armed conflict as leaving legal employment would entail much larger opportunity costs. Hence, it is not surprising that policies such as school construction, better health treatments and labour-market access have been found to promote peace.

Sound public policies and safety nets are crucial. As shown in a recent article, public employment programmes can provide insurance in the face of adverse shocks, leading to lower conflict levels.

Build back better

Now that the previous foreign aid system has been largely broken, the international community should not aim to restore much of the same. Rather we should aim to “build back better”. This starts with favouring types of investments that cannot be easily appropriated – physical capital can be stolen while human capital cannot. We need to make sure that money gets invested in ways that boost productivity and economic perspectives for everybody.

Together with domestic pushes for the strengthening of institutions and of inclusive governance, such international financial aid can bear fruits for generations to come.

Dominic Rohner receives funding from CEPR through the research initiative Reducing Conflict and Improving Performance in the Economy (ReCIPE) funded by FCDO. He is affiliated with CEPR/ReCIPE.

Austin L. Wright receives funding from CEPR through the research initiative Reducing Conflict and Improving Performance in the Economy (ReCIPE) funded by FCDO. He is affiliated with CEPR/ReCIPE.

Oliver Vanden Eynde receives funding from CEPR through the research initiative Reducing Conflict and Improving Performance in the Economy (ReCIPE) funded by FCDO. He is affiliated with CEPR/ReCIPE.

Uwe Sunde receives funding from CEPR through the research initiative Reducing Conflict and Improving Performance in the Economy (ReCIPE) funded by FCDO. He is affiliated with CEPR/ReCIPE. He is affiliated with CEPR/ReCIPE.

Jing-Rong Zeng does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.