Does foreign aid undermine political institutions?

Many scholars, including the recent Nobel Laureate Angus Deaton, would tend to agree. Others? Not so much.

November 12, 2015

Sam Jones, Finn Tarp

Does foreign aid undermine political institutions?

Many scholars, including the recent Nobel Laureate Angus Deaton, would tend to agree. While Deaton qualifies this view as being applicable to countries that receive very large inflows of foreign aid relative to their government budgets, the basic argument goes something like this: When a country receives foreign aid, the government becomes less dependent on domestic taxes to meet its financial needs. This makes the state less inclined to abide by its social contract of providing basic public goods and services, in turn weakening the imperative to build well-functioning institutions.

Not surprisingly, researchers are divided on whether this story is borne out in reality. However, using new and more extensive data on aid flows in 104 countries over the period 1983-2010, our recent paper published in the Journal of Development Economics takes a fresh look at the issues. The key to making progress is to recognize the diversity of aid types, each of which has its own objectives and predictability. Overall, we found no evidence for the claim that aid has a negative effect on political institutions. Instead, stable flows of aid that are explicitly targeted towards governance have a moderate positive effect.

Revisiting an Old Question with New Data

We used AidData’s most recent research release, which has richly-coded data on each aid project’s stated objectives (i.e., the data is “purpose-coded” in a consistent format). The data allowed us to disaggregate aid into various types, and study the effects of each type on institutional outcomes. Specifically, we disaggregated aid to a given recipient into three main types: (1) governance aid, defined as aid for the strengthening of government policies and plans, public sector and civil society institutional development, as well as human rights and conflict prevention activities; (2) economic aid, which primarily encompasses support to production and trade sectors; and (3) other aid such as social sector funding, humanitarian relief and debt-related actions.

We first looked at the cross-section relationship between the amount of different types of aid received and changes in political institutions over the full period. In addition, we looked at the dynamic relationship (within-countries over time). Both methods gave similar results. They suggest a small positive overall effect of aid on political institutions, but this is driven primarily by flows of governance aid. The impact of both economic and other aid is much less clear.

We also employed dynamic estimates, which allow us to distinguish between more stable and temporary flows of each type of aid. Here an even clearer picture emerges. The positive relationship between governance aid and changes in political institutions is itself driven by the stable inflows of governance aid. In other words, positive political developments are much more strongly associated with credible and consistent flows of governance aid. In turn, since aid provided explicitly for governance purposes makes up a relatively small share of total aid, this explains why we find a small but positive effect of aggregate aid.

Two other results are also worth mentioning: aid has a stronger effect on political institutions in countries that have better institutions initially, and the effects are stronger for the post-Cold War period between 1995 and 2010.

The Bottom Line

The idea that aid weakens political institutions is based on the argument that increases in aid may substitute for domestic revenue, making governments less responsive to the needs of its people. Our findings challenge this simplistic story. The institutional effects of aid depend on the type of aid and the frequency or credibility with which it is delivered. Since governance aid flows are expected to be most closely associated with institutional performance requirements, it is encouraging to find that this type of aid, especially when less volatile, has a robust positive effect on institutional outcomes.