The Aid Paradox: Helping or Hurting the Poor?

Foreign aid is often presented as a moral act in international politics. Wealthy nations seem to selflessly transfer resources to poorer ones, aiming to “defeat poverty, tackle instability and create prosperity in developing countries” (HM Treasury and Department for International Development, 2015). Judged by the aims alone, aid appears difficult to criticise.

Aid has undeniably contributed to remarkable humanitarian achievements: internationally funded vaccination campaigns helped to eradicate smallpox in 1980 (World Health Organization, n.d.), reduce polio prevalence by more than 99% since 1988 (World Health Organization, 2025), and dramatically reduce child mortality worldwide (World Health Organization, 2026). If foreign aid has saved millions of lives, surely the answer to the question posed is obvious.

However, the answer relies principally on the definition of “help”. Saving lives and making them liveable are not the same thing. Whilst the former requires identifiable interventions, the latter involves contested goals and is shaped by countless local decisions that outsiders cannot easily direct or even foresee.

This article argues that humanitarian relief and development differ fundamentally, since whilst relief requires resources, development requires knowledge. External actors cannot fully understand or replicate this local knowledge, meaning that the more control they exercise over how aid is used, the more their limited knowledge substitutes for the recipient’s own, thereby limiting aid’s effectiveness. Therefore, in general, the success of aid from a development perspective is inversely related to the amount of control exercised by the donor.

Firstly, the term “poor people” requires definition. Poverty exists in both absolute and relative forms, including within donor countries: almost 2.5 million Londoners live in poverty, accounting for housing costs (Trust for London, 2026), while Cornwall has historically qualified for EU Objective One grants (Hansard, 1998). Despite this, the UK spent £13,036 million on Official Development Assistance in 2025 (Foreign, Commonwealth & Development Office, 2026). Every pound spent on foreign aid cannot simultaneously be spent on housing, healthcare, education, or anti-poverty programmes at home. This introduces an unavoidable opportunity cost. Aiding those abroad could improve their welfare whilst undermining the welfare of home residents by reducing the resources available to others; hence, it is arguable that foreign aid hinders poor people in the home country, rather than helping them.

However, proponents of aid may argue that this comparison is misleading due to the varying nature of poverty. Since many aid recipients experience absolute poverty, whereas poverty in donor countries is often relative, a pound transferred from a wealthy country may generate far greater improvements in welfare abroad than at home.

Even if the overarching category of “poor people” is confined to those living within recipient countries, individuals are affected differently by aid interventions. The case of Haiti’s rice markets illustrates this. Cheap imported rice benefited poor urban consumers by reducing food prices but concurrently harmed poor Haitian farmers who could no longer compete with subsidised imports (Oxfam, 2008). This example exposes a flaw in the question itself. Donors of aid often assume that “the poor” constitute a single group with shared interests. In reality, aid frequently creates winners and losers among the poor. Consequently, asking whether aid helps or hurts poor people depends entirely upon which individuals are considered and which interests are prioritised. Aid, therefore, cannot be judged solely according to its aggregate effects on “the poor”, because no such unified constituency exists.

The relevant question is therefore not whether aid has an opportunity cost, but whether the benefits exceed the costs imposed on poor people, both within the donor country and among the recipients themselves, and whether donors are capable of directing those benefits effectively.

Sometimes aid achieves both immediate and lasting improvements; for example, studies of disaster risk reduction suggest that every $1 invested can generate approximately $15 in avoided future disaster recovery costs (United Nations Office for Disaster Risk Reduction, 2025). Yet such successes share an important characteristic: they involve relatively clear objectives and measurable outcomes. The challenge becomes far greater when aid attempts to engineer economic development itself.

Development differs from disaster prevention because it involves no single objective and no universally accepted solution. Even among development economists, there is disagreement regarding what development should achieve. Amartya Sen (1999) defines development as the expansion of human capabilities and freedoms, whereas Walt Rostow (1960) views it as economic modernisation and growth. This disagreement suggests that development is not merely a technical term but a normative one. Before aid agencies can act to promote development, they must first decide what development should mean. In doing so, they inevitably substitute their own conceptions of progress for those of the people they seek to help.

Even if a consensus on the definition of development could be reached, Friedrich Hayek’s theories (1945) suggest that achieving it may lie beyond the human capabilities of external planners. Hayek argued that the information necessary to organise an economy is dispersed among millions of individuals, and no planner can ever possess all the knowledge held by the people participating within it. Although Hayek developed this argument in response to central economic planning, it also applies to many forms of foreign aid. Whilst much knowledge is simply dispersed, some of the knowledge required for development cannot be known in advance, since it arises from the combination of countless individual decisions and thus can only emerge through natural experimentation. Consequently, development is not simply a problem that can be solved, meaning that prescriptive solutions will continue to struggle.

Despite this, the flaw remains widely unidentified because of the seemingly logical presumption that experts possess knowledge that poor people lack. A development economist has access to extensive statistical data, yet this remains an imperfect substitute for the personal knowledge, derived from lived experience, that shapes individual choices. Unlike vaccination campaigns, where the desired outcome is obvious, prosperity emerges from countless context-specific decisions that cannot be centralised. Consequently, aid providers who feel fully equipped by technical expertise conflate it with practical knowledge. The result is cookie-cutter programmes designed for a generic 'poor person' who does not exist.

The Haiti example demonstrates how interventions designed by external actors can generate unintended consequences by altering incentives and market behaviour in ways that policymakers fail to anticipate. The issue was not a lack of good intentions, but the impossibility of possessing the dispersed knowledge necessary to predict how aid would interact with a complex economic system.

In rebuttal to this argument, Esther Duflo (2017) contends that aid should be evaluated through evidence rather than ideology, with individual interventions being assessed according to their outcomes. Although this is an entirely valid criterion, many development programmes subjected to precisely this standard fall short, for example the TOMS shoe donation programme.

The TOMS shoe donation programme provides a useful illustration of the knowledge problem in practice. Through its “one-for-one” model, the company donated a pair of shoes to a child in a developing country for every pair purchased elsewhere. Yet a randomised evaluation in rural El Salvador found that its overall effects were largely negligible (Wydick et al., 2018). Most significantly, researchers discovered that the vast majority of children already owned shoes before the intervention took place.

This aid-style programme was flawed from the beginning. Donors had incorrectly identified what recipients needed, meaning that while it was smoothly implemented, it failed because the underlying assumption was wrong. TOMS’ subsequent move away from this fixed model (TOMS, n.d.) only reinforces this point: development interventions are weakest when they prescribe needs in advance.

Although perfect information is only possible in theory, partial knowledge may sometimes be available. However, aid is rarely purely humanitarian (Dreher, Lang and Reinsberg, 2024). Donor governments frequently use aid to advance strategic, economic or political objectives, including tied aid arrangements requiring recipients to purchase goods and services from donor countries (OECD, n.d.). If aid is designed partly to serve external interests, it becomes even less responsive to the local knowledge and priorities of those it is intended to help. The issue is therefore not merely that donors lack information, but that they often possess incentives that differ from those of the recipients.

Nonetheless, Jeffrey Sachs (2005) presents one of the major arguments for aid: local knowledge alone cannot generate development if individuals lack the infrastructure and institutions necessary to act upon it. Sachs uses the term “institutions” in the sense commonly found in growth economics, referring broadly to property rights and the rule of law. Many countries, he argues, remain trapped in self-reinforcing cycles of poverty: poor populations generate low incomes, low tax revenues and limited investment in healthcare, education, infrastructure and institutions, in the form of administrative capacity. Foreign aid can therefore provide the “big push” necessary to break this cycle and initiate sustained economic growth, since it can improve both infrastructure and institutional capacity and thus enable individuals to act upon their knowledge (Sachs, 2005).

Sachs correctly identifies that poverty can become self-perpetuating. Yet his conclusion rests upon the assumption that resources constitute the primary constraint on development. If this were true, decades of aid transfers should have produced more consistent economic success than they have; instead, cross-country evidence has found little robust relationship between aid inflows and economic growth (Rajan and Subramanian, 2008).

Sachs and Hayek fundamentally disagree about the nature of poverty itself. Sachs treats poverty as a resource constraint, whilst Hayek treats poverty as a knowledge problem. If Sachs is correct, aid can provide the missing capital required for growth, whereas if Hayek is correct, no quantity of external resources can substitute for the dispersed knowledge held by local individuals.

Although they seem mutually exclusive, Haushofer and Shapiro’s work (2016) suggests that both are partly correct. They demonstrated that direct cash transfers increased consumption, improved psychological well-being and enabled productive investment. Unlike traditional aid programmes, cash transfers do not attempt to determine what poor people should value; instead, they transfer resources while leaving decisions in the hands of the recipients. The success outlined by Haushofer and Shapiro illustrates that aid becomes more successful as it relinquishes control. Therefore, whilst Sachs correctly identifies that aid may be necessary to break self-perpetuating poverty cycles, many traditional aid programmes remain insufficient because, as Hayek argued, donors lack local knowledge.

This is captured particularly well by William Easterly’s (2006) distinction between “planners” and “searchers”. Planners attempt to transform societies from above, whereas searchers begin with local conditions and adapt to feedback. The central weakness of foreign aid is that it too often behaves like a planner, which is shown by the stark contrast between the TOMS programme and cash transfers, both in their execution and in their effects. While TOMS attempted to determine recipients’ needs in advance and delivered a prearranged solution, cash transfers allowed recipients themselves to discover and pursue opportunities that outsiders could not foresee.

The eradication of smallpox, the reduction of polio and the decline in child mortality clearly stand as genuine humanitarian achievements. Yet these successes should not obscure the limitations of aid as a development strategy, since alleviating suffering is not the same as eliminating poverty. Neither “poor people” nor “foreign aid” constitute uniform categories, making any universal answer impossible; poor people do not share identical interests, and aid programmes differ enormously in both their objectives and outcomes.

Foreign aid’s greatest successes have come when its goals are narrow, its outcomes are measurable, and its implementation does not rely on knowledge that outsiders cannot possess. Humanitarian relief can save lives, but development requires something donors cannot supply: the dispersed, contextual knowledge held by poor people themselves. The evidence examined here, from the failures of the TOMS programme and the intervention in Haiti’s rice market to the contrasting success of cash transfers, shows that aid helps poor people when it expands their agency and hurts them when it replaces their judgement.

Development rarely succeeds through prescriptive, donor-designed solutions; it must emerge from the decisions of those whose lives are at stake. Until aid is structured around that principle, it will continue to substitute good intentions for good outcomes.

By Leona Bastianic

Bibliography

Dreher, A., Lang, V. and Reinsberg, B. (2024) ‘Aid effectiveness and donor motives’, World Development, 176, article 106501. doi:10.1016/j.worlddev.2023.106501.

Duflo, E. (2017) ‘The economist as plumber’, American Economic Review, 107(5), pp. 1–26. doi:10.1257/aer.p20171153.

Easterly, W. (2006) The White Man’s Burden: Why the West’s Efforts to Aid the Rest Have Done So Much Ill and So Little Good. New York: Penguin Press.

Foreign, Commonwealth & Development Office (2026) Statistics on International Development: provisional UK Official Development Assistance spend 2025. GOV.UK. Available at: https://www.gov.uk/government/statistics/statistics-on-international-development-provisional-uk-official-development-assistance-spend-2025/statistics-on-international-development-provisional-uk-official-development-assistance-spend-2025 (Accessed: 2 June 2026).

Hansard (1998) ‘European Structural Funds’, HC Deb, 22 October, vol. 317. Available at:https://publications.parliament.uk/pa/cm199798/cmhansrd/vo981022/debtext/81022-24.html (Accessed: 20 June 2026).

Haushofer, J. and Shapiro, J. (2016) ‘The short-term impact of unconditional cash transfers to the poor: experimental evidence from Kenya’, The Quarterly Journal of Economics, 131(4), pp. 1973–2042. doi:10.1093/qje/qjw025.

Hayek, F.A. (1945) ‘The use of knowledge in society’, American Economic Review, 35(4), pp. 519–530.

HM Treasury and Department for International Development (2015) UK aid: tackling global challenges in the national interest. GOV.UK. Available at: https://www.gov.uk/government/publications/uk-aid-tackling-global-challenges-in-the-national-interest (Accessed: 1 April 2026).

Organisation for Economic Co-operation and Development (n.d.) Untied aid. Available at: https://www.oecd.org/en/topics/sub-issues/oda-standards/untied-aid.html (Accessed: 20 June 2026).

Oxfam (2008) ‘Haiti no longer grows much of its own rice’, Oxfam America, 3 June. Available at: https://www.oxfamamerica.org/explore/stories/haiti-no-longer-grows-much-of-its-own-rice/ (Accessed: 1 June 2026).

Rajan, R.G. and Subramanian, A. (2008) ‘Aid and growth: what does the cross-country evidence really show?’, The Review of Economics and Statistics, 90(4), pp. 643–665. doi:10.1162/rest.90.4.643.

Rostow, W.W. (1960) The Stages of Economic Growth: A Non-Communist Manifesto. Cambridge: Cambridge University Press.

Sachs, J. (2005) The End of Poverty: Economic Possibilities for Our Time. New York: Penguin Press.

Sen, A. (1999) Development as Freedom. Oxford: Oxford University Press.

TOMS (n.d.) Impact. Available at: https://uk.toms.com/pages/impact (Accessed: 1 June 2026).

Trust for London (2026) ‘Poverty 2026: London has highest poverty rate in the country again’. Available at: https://trustforlondon.org.uk/news/poverty-2026-london-has-highest-poverty-rate-in-the-country-for-second-year-in-a-row/ (Accessed: 20 June 2026).

United Nations Office for Disaster Risk Reduction (2025) Global Assessment Report 2025: Resilience Pays: Financing and Investing for Our Future. Available at: https://www.undrr.org/gar/gar2025 (Accessed: 15 May 2026).

World Health Organization (n.d.) History of smallpox vaccination. Available at: https://www.who.int/news-room/spotlight/history-of-vaccination/history-of-smallpox-vaccination (Accessed: 2 June 2026).

World Health Organization (2025) Poliomyelitis. Available at: https://www.who.int/news-room/fact-sheets/detail/poliomyelitis (Accessed: 2 June 2026).

World Health Organization (2026) Child mortality (under 5 years). Available at: https://www.who.int/news-room/fact-sheets/detail/child-mortality-under-5-years (Accessed: 2 June 2026).

Wydick, B., Katz, E., Calvo, F., Gutierrez, F. and Janet, B. (2018) ‘Shoeing the children: the impact of the TOMS shoe donation program in rural El Salvador’, The World Bank Economic Review, 32(3), pp. 727–751. doi:10.1093/wber/lhw042.