Intermediatedevelopment
Foreign Aid: Does It Work?
Overview
Rich countries give ~$200 billion yearly in official development assistance — for vaccines, roads, schools, and budget support. Does it work? The debate is fierce: Sachs says scaled-up aid could end poverty; Easterly says it feeds corruption; the evidence says both have a point — micro-interventions succeed while macro growth effects are elusive. This lesson covers aid's forms, the great debates, what rigorous evidence shows, and the future (cash benchmarking, China, private flows dwarfing aid).
Learning objectives
- Distinguish ODA, humanitarian, and private flows
- Summarize the Sachs-Easterly-Moyo debates
- Evaluate the macro evidence on aid and growth
- Identify micro-interventions with strong evidence
- Describe how aid is evolving: China, cash, private capital
What aid is — and isn't
Official Development Assistance (ODA): government grants and concessional loans for development, ~800B+ — dwarfing ODA). Crucially, FDI and commercial flows exceed aid several-fold for most recipients. Aid matters most where other flows don't reach: the poorest, most fragile states.
The UN's 0.7%-of-GNI aid target (1970) is met by only a handful of Nordics. The US gives most in absolute terms but ~0.2% of GNI — the generosity debate in one statistic.
The great debates
Sachs (The End of Poverty): poverty traps need a 'big push' of scaled-up aid — calculable, achievable, moral. Easterly (The White Man's Burden): top-down aid fails because planners lack local knowledge and feedback; markets and freedom, not planners, develop nations. Moyo (Dead Aid): aid props up corrupt governments, distorts accountability, and should be phased out for market finance. Each captures truth: big pushes worked (Marshall Plan, East Asian infrastructure), planner failures abound (white-elephant projects), and aid dependence does corrode governance. The synthesis emerging: aid works when it funds what works, with feedback and local ownership.
Mosley's puzzle: rigorous evaluations find many aid projects succeed, yet cross-country regressions find little growth effect. Resolution: successful micro doesn't aggregate when fungibility, Dutch disease, and governance effects operate at macro scale.
What the evidence shows
Macro: cross-country studies find small, fragile aid-growth effects — positive in good-policy environments (Burnside-Dollar, debated), negligible otherwise. Micro: strong successes — GAVI vaccines (over a billion immunized), PEPFAR (25 million on HIV treatment), deworming, bed nets. Budget support: mixed — it funds health and education spending but risks fungibility (aid frees government money for other uses). The most robust conclusion: aid's average effect is modest, its best uses are excellent, and its harms (Dutch disease via currency appreciation, accountability erosion) are real but manageable with design.
Aid for schools may just let governments shift their own school money to the military. Evaluating aid requires tracking total spending, not just the aid-funded project.
How aid is changing
Four shifts. Cash benchmarking: USAID now compares every program against just giving cash — a brutal, healthy test. China: the Belt and Road Initiative made China the largest bilateral infrastructure lender, offering fast, condition-free finance — and debt-trap controversies (Sri Lanka's Hambantota port). Private flows: philanthropy, impact investing, and remittances dwarf ODA; 'beyond aid' agendas focus on trade, tax, and climate finance. Localization: the push to fund local organizations directly (currently ~2% of aid goes local) rather than Western contractors — decolonizing development, slowly.
For most developing countries, trade access, tax cooperation, migration, and climate finance matter more than ODA volumes. The aid debate increasingly looks quaint next to these.
Making aid work better
The reform consensus: fund evidence-backed interventions (vaccines, cash, deworming); use country systems rather than parallel bureaucracies; make it predictable (multi-year commitments, not stop-start); coordinate to cut transaction costs (some countries face 1,000+ donor missions yearly); evaluate rigorously and publish failures; and shift power to recipients (cash transfers and direct local funding do this structurally). None of this ends the macro debate — but it describes aid that helps real people measurably, which is what the enterprise should be judged on.
Key points
- ODA is ~$200B/year, but remittances, FDI, and philanthropy are larger flows for most recipients.
- The Sachs-Easterly-Moyo debate: big push vs. planner ignorance vs. aid harms — each partly right.
- Macro growth effects are elusive; micro successes (vaccines, cash, bed nets) are strong.
- Fungibility and Dutch disease are the real macro risks, manageable by design.
- Aid is evolving: cash benchmarking, Chinese lending, localization, and 'beyond aid.'
Common misconceptions
✓ Truth: For most developing countries, trade, investment, and remittances dwarf aid. Aid is critical mainly for the poorest fragile states other flows skip.
✓ Truth: Corruption exists, but rigorously evaluated programs (vaccines, cash transfers, deworming) deliver extraordinary returns. The waste critique overgeneralizes from real but fixable failures.
✓ Truth: Absorptive capacity, governance, and what the aid funds matter more than volume. The Marshall Plan worked in industrialized Europe; the same dollars elsewhere bought less.
Practice questions
Q1. The 'micro-macro paradox' in aid evaluation refers to:
- Aid workers being short
- Many individual aid projects succeed in evaluations while cross-country studies find little aggregate growth effect
- Microfinance always failing
- Small countries getting more aid
Mosley's paradox: project success doesn't aggregate — fungibility, Dutch disease, and governance effects at the macro level dilute micro wins. Both findings can be true simultaneously.
Q2. Which flow is largest for most developing countries today?
- Official development assistance
- Remittances from migrants abroad
- Food aid
- Military aid
Remittances exceed $800B yearly — multiples of ODA — and go directly to households. For most recipients, migration policy matters more than aid policy.
Q3. The strongest evidence-based criticism of scaling up aid focuses on:
- Donor generosity
- Absorptive capacity, fungibility, Dutch disease, and governance effects — not just funding gaps
- Recipient laziness
- Exchange rates
Skeptics don't dispute that money can help; they argue more money hits institutional constraints — the binding limits are governance and capacity, not finance.
Click a question to reveal the answer and explanation.
Frequently asked questions
What was the Marshall Plan and is it relevant?
America's $13B (1948–52) rebuild of Western Europe — aid's legendary success. Its relevance is limited: Europe had skilled workers and institutions; the money unlocked existing capacity rather than building it from scratch.
What is 'tied aid'?
Aid requiring recipients to buy donor-country goods and services — effectively a subsidy to donor firms. Untying aid (now the norm among DAC donors) raises its value ~15–30%.
Does Chinese aid differ from Western aid?
China's flows are mostly commercial-rate infrastructure loans without governance conditions — faster and welcome by borrowers, but raising debt-sustainability and transparency concerns. It's development finance, not charity.
What is cash benchmarking?
USAID's practice of comparing every program's cost-effectiveness against just giving beneficiaries cash. Many programs fail the test — which is precisely the point: it disciplines the aid industry.