VoxEU Column Development

NGO aid – well targeted to the needy and deserving?

How do NGOs spend their development assistance? This column discusses research showing that NGO aid is no better targeted to the neediest countries than state aid agencies either by choosing needier countries, or by entering un-chartered waters and trying to excel where state aid is most likely to fail.

Exposing the flaws of official development aid (ODA) is fairly easy. State aid agencies provide critics with the data needed, for example, to figure out the extent to which ODA allocation is distorted by selfish donor motives such as export promotion or forming political alliances. It’s different with aid from non-governmental organisations (NGOs). The allocation of NGO aid is of considerable importance in quantitative terms: The annual budget of World Vision International, for example, exceeds the official foreign aid budget of Italy. Plan International spends more than Greece on development projects each year and the Save the Children Alliance more than Finland. Overall grants by NGOs based in OECD countries amounted to almost $15 billion annually in 2005 and 2006, thus exceeding bilateral ODA from every individual donor country except for the United States. Given the huge amount of aid involved, it is surprising that NGO aid has hardly been mapped, let alone explained. Arguably, the main reason is that sufficiently detailed data are hard to come by. Maybe it’s for good reasons that NGOs are reluctant to open their books: Critical analysis could tarnish the favourable image of superior donors that they enjoy.

Strong beliefs, weak evidence

It has long been an article of faith that NGOs provide well-targeted aid (e.g., Tendler 1982). NGOs are widely believed to be closer to the poor than state aid agencies are. They may directly deal with local target groups, circumventing (frequently corrupt) recipient governments. This may reduce leakage and result in better alignment with the needs of the poor. Furthermore, bilateral and multilateral ODA has been shown to be biased for political reasons (e.g., Kuziemko and Werker 2006, Dreher, Sturm, Vreeland 2008). The allocation of NGO aid should be less distorted by commercial and political self-interests of donor governments.

Even donor governments appear to share the view that NGOs have an important role to play for aid to reach the poor and render it more effective. The share of bilateral ODA channelled to or through NGOs reached up to 20 percent in recent years for some donor countries (see Dreher, Mölders, Nunnenkamp 2008). But already by the 1990s, some critics suspected that the case for NGO aid largely rests on ideological grounds. In particular, the view that NGOs have a clear focus on the poor has come under attack. Rather than trying to outperform state agencies and risk failure, NGOs may prefer the quieter life of implementing the policy agenda of their home government, especially if they depend financially on official “backdonors”. Both the critics’ attempts to demystify NGO aid and the proponents’ articles of faith have hardly been subjected to empirical scrutiny.

Some interesting insights may be gained, however, on the basis of data we collected for some 60 quantitatively important NGOs based in various OECD countries. Taken together, they granted aid on the order of $ 5.7 billion in 2005, nearly as much as the sum of ODA by all four Scandinavian countries. The representativeness of the NGO sample may be questioned, and aid flows refer to just one year. Yet some of our findings for this sample put into serious doubt the view that NGO aid is generally superior to ODA with respect to targeting the needy and deserving countries.

This is despite the fact that NGOs tend to be more active in recipient countries with more pressing need for aid. But the poverty orientation of NGO aid does not appear to be particularly strong, for example when looking at the extent to which rising per-capita income of recipients reduces NGO aid. Employing a multivariate regression framework and controlling for various factors (a country’s level of democracy, the amount of bilateral ODA it receives, the number of other NGOs present in the same country, joint religion of the NGO’s home country and the recipient country, population size of the recipient country), we find that NGO aid significantly decreases with per-capita GDP, but not dramatically so (see the fitted values of the partial leverage plot in Figure 1) (Koch et al. 2008).

According to the World Bank, NGO aid has an important complementary role to play when governance is particularly bad and the local environment is highly distorted – a view echoed in various UN and national aid policy documents (e.g., World Bank 1998 and United Nations Millennium Project 2005). But NGOs hardly make use of their perceived comparative advantage of engaging in countries with “difficult environments,” i.e., under conditions in which government-to-government transfers are unlikely to work. As Figure 2 shows, NGO aid is not targeted towards less democratic countries (again controlling for other factors), possibly because NGOs refused to accept the role assigned to them by official donors and focus on countries “left over” by bilateral ODA.

Keeping a low profile

Rather than trying to excel by working in difficult environments, NGOs appear to have taken the easy road of replicating the allocation behaviour of their home government. The strong tendency to mimic official backdonors revealed in Figure 3 supports critics pointing to the limits to autonomous decision-making of NGOs that depend on government funds.

NGOs also tend to follow their peers, so NGO aid becomes clustered (see Figure 4). Some so-called “donor darlings” receive over 20 times more NGO aid per capita than “donor orphans”. Hiding in the crowd may help ensure future funding since the conformity of location choices renders it difficult for backdonors to assess the performance of any particular NGO (Fruttero and Gauri 2005). NGOs may also prefer engaging in countries where other NGOs are already active because this allows them to piggyback on the efforts of their peers in training local staff and strengthening local organisations. Taken together, NGOs are contributing to the divide between donor darlings and aid orphans.

NGOs may draw some comfort from the finding that the distribution of NGO aid is hardly affected by commercial and political self-interests that state agencies are often found to have when granting aid. Yet there is one similarity between official donors and NGOs even in this respect. Both tend to favour former colonies and/or recipient countries with common traits related to religion and language (Koch et al. 2008). While this preference may be justified by comparative advantages of working under such conditions, picking recipient countries according to colonial ties or cultural similarities may well violate a needs-based allocation of aid.

All this invites the conclusion that NGO aid is no panacea to provide aid that is better targeted to the neediest countries. In contrast to what one might expect, NGOs seem to prefer keeping a low profile. They do not distinguish themselves from state aid agencies either by choosing needier countries, or by entering unchartered waters and trying to excel where ODA is most likely to fail.

Some open questions remain: What exactly lies behind NGOs’ tendency to replicate the location choices of both official donors and their peers? Do NGOs relying strongly on official backdonors behave differently from those being financially autonomous? There is an urgent need for better data and more research on the increasing number of quantitatively important private donors, as these first findings on NGO aid are rather discomforting in light of reaching the Millennium Development Goals.


Dreher, A., F. Mölders and P. Nunnenkamp (2008). Aid Delivery through Non-Governmental Organizations: Does the Aid Channel Matter for the Targeting of Swedish Aid? KOF Swiss Economic Institute and Kiel Institute for the World Economy, mimeo.
Dreher, A., J.-E. Sturm and J. Vreeland (2008). Development Aid and International Politics: Does Membership on the UN Security Council Influence World Bank Decisions? Journal of Development Economics, forthcoming.
Fruttero, A., and V. Gauri (2005). The Strategic Choices of NGOs: Location Decisions in Rural Bangladesh. Journal of Development Studies 41 (5): 759-787.
Koch, D.J., A. Dreher, P. Nunnenkamp and R. Thiele (2008). Keeping a Low Profile: What Determines the Allocation of Aid by Non-Governmental Organizations? Working Paper 191, KOF Swiss Economic Institute, ETH Zurich.
Kuziemko I and E. Werker (2006). How Much is a Seat on the Security Council Worth? Foreign Aid and Bribery at the United Nations. Journal of Political Economy 114 (5): 905-930.
UN-Millennium Project (2005). Investing in Development: A Practical Plan to Achieve the Millennium Development Goals. Overview. New York: United Nations Development Programme.
World Bank (1998). Assessing Aid: What Works, What Doesn’t, and Why. Oxford: Oxford University Press.
Tendler, J. (1982). Turning Private Voluntary Organizations into Development Agencies. Questions for Evaluation. AID Program Evaluation Discussion Paper 12. Washington, D.C.

525 Reads