Foreign Aid Is a Marketplace of Competing Motives—Separating Why We Give from How It Works

Foreign aid is often discussed as if it follows a single logic—either altruism or self-interest, either geopolitics or poverty reduction. But in practice, it looks more like a crowded marketplace of motives and methods. Actors who want very different things can end up endorsing the same policy tools, while people who share broad goals can disagree sharply on what to do. One way to cut through that confusion is to separate two questions that are frequently tangled together: why provide aid, and how to deliver it.

The “why” is fundamentally about priorities and beneficiaries. It asks which problems deserve attention and who is expected to gain. In some cases, the intended beneficiaries are primarily people in recipient countries—reducing poverty, improving health, or expanding access to education. In others, the donor country anticipates returns in the form of economic ties, political influence, or softer forms of power and prestige. A third rationale focuses on outcomes that serve everyone, such as global public goods: knowledge, disease control, or other benefits that spill across borders regardless of who funds them.

Seeing these motivations side by side helps clarify what tensions are real and what disagreements may be overstated. Two stakeholders might appear to be fighting about strategy, when they are actually pursuing different end goals. Conversely, apparent consensus on a program can mask very different expectations about who benefits and why.

Once the goal is chosen, the “how” becomes the central challenge: what approach can address the target problem most effectively, and at the lowest cost for a given level of impact? Here, evidence is not a luxury; it is the core input. The principle is straightforward: use rigorous evidence when it already exists, and create it when it does not. That does not only mean evaluating whether a program “works,” but also understanding for whom it works, under what conditions, and how it can be improved.

A particularly powerful set of aid strategies aims to strengthen governments’ ability to both use evidence and generate it. Investments in state capacity—data systems, evaluation skills, and decision processes that rely on results—can advance multiple objectives at once. They can help recipient countries deliver services more effectively and sustainably. They can also align with donor interests by making aid dollars more productive and improving governance relationships. And by producing credible findings that others can learn from, they contribute to a broader global public good: transferable knowledge about what works.

Crucially, the process implied by the “how” is not a one-time decision followed by implementation. It is iterative. Programs should be treated as opportunities to learn: test interventions, adopt what succeeds, adapt based on what the evidence shows, and repeat. Then share results so that other agencies and countries can build on them rather than starting from scratch.

These themes are the focus of an upcoming article in the Journal of Economic Perspectives by economist Dean Karlan. Karlan notes that the piece benefited from extensive feedback and research support, while emphasizing that any errors are his own. He also discloses that he served as Chief Economist at the U.S. Agency for International Development from November 2022 through February 2025 and that he founded Innovations for Poverty Action and sits on its board. He reports no financial conflicts. The work is associated with the National Bureau of Economic Research, which distributes research through several free publications and other public-facing formats.

Why it matters

How governments justify foreign aid shapes not only budgets, but also which communities and challenges receive attention. Separating goals from methods can reduce unproductive arguments and make trade-offs explicit—especially when donor interests and recipient needs only partially overlap. The next key test is whether aid systems invest more in evidence and in local capacity to generate it, allowing programs to improve over time rather than repeating familiar, unproven approaches.