Skip to content

Does health aid to governments make governments spend more on health?

If you’re not an economist, you might reasonably assume that the answer to this question is yes. The story might go something like this: aid agencies give money to poor country governments to distribute bed nets or give vaccinations, and those additional funds are added to whatever money the country was able to scrape together to spend on health before the donor came along. As a result of the health aid, the total amount of money spent on health increases.

There is new evidence, from a study from the Institute for Health Metrics and Evaluation published in the Lancet last week, showing that this story doesn’t describe what’s really going on. Overall, global public health financing shot up by 100 percent over the last decade, but the study’s authors found that on average, for every health aid dollar given, developing country government shifted between $.43 and $1.17 of their own resources away from health. The trend is most pronounced in Africa, which received the largest amount of health aid.

The finding that health aid substitutes for rather than complements existing government health spending has caused a mini– scandal in the press precisely because it runs so counter to people’s optimistic expectations, perpetuated by aid agencies’ fund-raising campaigns, about the level of control that donors can exert over the spending of developing country governments.

Economists, on the other hand, have been beating the dismal drum for a long time on this issue. In 1947, Paul Rosenstein-Rodin, then a deputy director at the World Bank, famously said, “When the World Bank thinks it is financing an electric power station, it is really financing a brothel.” Economists expect that aid will be at least partially fungible (that is, that aid money intended by donors for one sector or project can and will be used by governments interchangeably with funding for other priorities), and this prediction is borne out by empirical studies from the late 1980s on. The authors of a 2007 paper in the Journal of Development Economics observed, “While most economists assume that aid is fungible, most aid donors behave as if it is not.”

You might argue (as Owen Barder does in depth here) that recipient governments are acting rationally in response to erratic donor funding, which ebbs and flows according to donor priorities and how well the global community mobilizes fundraising around a particular issue in any given year. After all, doesn’t the donor community’s insistence on country ownership mean that they want poor country governments to be able to set their own budget priorities?

The problem is that aid agencies have long used the argument that earmarking aid for a specific project or sector is a credible way to force recalcitrant recipient country priorities into line with donor priorities—to coerce bad governments into making good decisions.

If  governments that don’t prioritize their people’s welfare respond to an influx of aid money by simply shifting their existing resources around to circumvent donor priorities (and we don’t know what is happening to the resources shifted away from health—they could be going to private jets and presidential palaces, or to education, infrastructure, or loan repayments, or really anything at all),  then the aid agency argument for project aid falls apart. The burden of proof correctly lies with the aid agencies to show that aid isn’t freeing up funds for bad governments to use badly.

The Lancet findings are scandalous, relative to the naïve but widespread belief that donors can use earmarked aid to force bad governments to behave.

This entry was posted in Aid policies and approaches, Economics principles, Global health, Organizational behavior and tagged , . Bookmark the permalink. Follow any comments here with the RSS feed for this post. Both comments and trackbacks are currently closed.


  1. Sam Gardner wrote:

    If aid is all about transferring money, it is indeed “kind of” a scam.

    However, if aid is about searching together for solutions for practical problems, than the money transfer on its own is not always the main contribution of aid.

    In a recent ODI briefing (What does an effective Multilateral donor look like?), the depth of commitment is seen as the main characteristic, combined with the capacity to adapt to the specific needs.

    This looks very much like the expertise for search, willingness to go for it and to learn on the fly, are more important than the shear amount of money.

    Apart from the money, a donor can contribute a lot:
    – expertise
    – attention (see posting on the somnambulist)
    – focus
    – feedback loops to higher level government that internal systems might lack
    – evaluation loops
    – accountability and anti-corruption attention.

    Most of these contribution though, are nowadays considered as transaction costs.
    Of course, this added value can only be delivered if the project is not ” managed by committee” e.g. as a trust fund for health sector support.

    Posted April 13, 2010 at 1:23 am | Permalink
  2. Blaise wrote:

    I’m wondering if we’ll be able to find out where the money goes. It doesn’t have the same impact if it goes to private jet or education. My guess is that a huge part remains in the national budget. I follow Owen on this point.

    On the other hand I’m surprised that nothing is done in health to prevent the fungibility issue. In the fast-trach initiative for education gvt have to meet targets in order the donors give the money. These targets mean that a certain share of the gdp/c is invested in the education sector. Even if I have a lot of discontents with this approach, I have to say that it prevents the gvt to shift the funds.

    Posted April 13, 2010 at 4:13 am | Permalink
  3. Rachael Burke wrote:

    It’s not scandalous, and it’s sensationalist to suggest that that’s the only explanation (Barder’s blog that you link to is interesting). The accompanying editorial in the lancet sets out three reasons for this, none of which are necessarily scandalous.
    Fungibility – as you’ve discussed. I don’t see why this has be to be problem… of course in a world were everyone really meant Paris and Accra and whatever it wouldn’t be such an issue, but when donor aid is so focused on donor priorities it’s not reasonable to expect countries to use their own limited resources as well. The other point, which the editorial didn’t make, is that the authors found a positive elasticity on spending on NGOs and govt. health spending…. but disbursing ODA to NGOs and ignoring governments has so many problems we don’t want to go down that road.
    The second was using domestic resources to iron out aid flucuations. So if last year donors were all over agriculture and now they’ve decided health is their baby it’s prudent to switch domestic funding to make up the agricuture shortfall and/or to spend less of your money on health this year so that when donors have a notion they’d rather spend money on what’s in vogue next you can make up the shortfall.
    And lastly it could be that donor resources are not spent all at once is this is an artefact from how the data was collected. Health metrics group is great, but there’s only so much you can do with the data that exists. If a country gets given £10 in DAH it could decide not to spend it all in that economic year, and keep it’s domestic spending the same. But when it comes to balancing the books that £10 DAH disbursement ‘looks’ like it was all spent in that year, so to make the numbers add it up ‘looks’ like domestic spending went down.

    I think the solution, as other posters have noted, is more to do with realising the ideals of Paris and Accra and not spending on things that countries don’t want. If there’s money on the table developing country governments aren’t stupid… they’ll take it. But if it’s offered for donor priorities, on donor initatives, under donor disbursement pressure to ‘look good’ then it’s completely rational that domestic resources get moved around.

    Posted April 13, 2010 at 4:42 am | Permalink
  4. Justin Kraus wrote:

    Certainly donors should live up to their Paris and Accra agreements. But that doesn’t mean we should condone recipient countries who hold their hands out for sector-specific aid while having ulterior motives. We shouldn’t be so accepting of dysfunction, especially when every can see it so clearly.

    Posted April 13, 2010 at 8:02 am | Permalink
  5. April wrote:

    though I doubt we’ll ever be able to figure out how a government is spending the marginal funding that donors freed up by giving funding for health, it’s probably not a bad proxy to look at the total budget (as executed, say, last year, not planned). When this is done, we will see that many of the most aid-dependent countries spend their money in ways that we certainly don’t like, ways that bear little resemblance to preferences of their population, and of course, often the funding is spent ineffectively in addition.
    That is, it’s not just that they take the money and spend the freed up funds on, say, weapons, but they do poor (often corrupt) weapons procurement.

    To me, the two main takeaways from The Lancet Lu et al piece are:
    1) there are serious problems generated by health aid donors pretending health aid is not fungible and that it is all spent on precisely their priorities

    2) donors (whatever their priorities) should ALL be focusing a lot more attention and support on public expenditure management capacity building and monitoring. If our aid effectiveness is largely determined by how well those marginal dollars are spent, for everyone’s sake, we should try to make it better.

    Posted April 13, 2010 at 8:13 am | Permalink
  6. geckonomist wrote:

    Did you really need an article in the Lancet to figure that out ?

    Posted April 13, 2010 at 8:42 am | Permalink
  7. Rachael Burke wrote:

    I don’t think it’s nearly as simple as countries “holding their hands out for sector specific aid.” Sector-specific, narrowly-earmarked aid is what’s on the table for the time being, and there’s a huge amount of disbursement pressure on donors and foundations to spend money. To an extent, it’s an example of smart people in situations with silly incentives leading to silly results. For example (although it’s all within health) in India about 90Rs crore is spent on blindness healthcare and the same on HIV care – but HIV is funded 98% externally and blindness 14% externally. Which is an example of donor priorities not matching up to country priorities.
    Here’s a report from a Oxford Global Economic Governance working group about how developing countries are able and unable to influence global health agenda in their own countries –

    I’m not saying this is a full explanation, or explains everything. But it’s certainly one of the reasons for this finding.

    The other thing is that a lot of these countries are in Africa and have signed up to Abuja declaration to spend 15% of domestic govt revenue on health and very few have some close to meeting that. Which is a problem, regardless of how donor money affects it.

    Posted April 13, 2010 at 8:47 am | Permalink
  8. Owen Barder wrote:

    A minor point, perhaps, but I want to say how bothered I am by Justin’s use this metaphor when he says: “recipient countries who hold their hands out for sector-specific aid”.

    These countries are very poor and need money on pretty much any terms that donors condescend to give it to them.

    Sectoral earmarking is a donor disease, not one caused by recipient countries.

    I know of no instances, and the Lancet article reports no cases, in which recipient countries have made misleading commitments about their own resource allocations.


    Posted April 13, 2010 at 9:58 am | Permalink
  9. Rachael Burke wrote:

    I am, however, a little surprised that debt relief had no effect on govt domestic health spending according to the Lu et al paper.

    Posted April 13, 2010 at 10:31 am | Permalink
  10. Justin Kraus wrote:


    With all due respect, you’re reading more into my words than what is there. I certainly meant no condescension when I stated that recipient countries “hold out their hands…” My comments here and on your blog have consistently stated that donors must be more attentive to and respectful of recipient country’s priorities. But that doesn’t mean I must excuse recipient countries for “playing the game” that donor intransigence presents them with. Frankly I’m not sure why this is controversial. Doesn’t everyone agree that it would be better if donors did align themselves with recipient government priorities (or at least knew what they really were) so that reallocation of government funds due to external funding would be minimized? Surely reallocation, even when/if justified, is a second-best option?

    Posted April 13, 2010 at 10:46 am | Permalink
  11. Robert Tulip wrote:

    The fungibility debate over health funding illustrates how emotional priorities of charities have become the tail that wags the aid dog. When pacifying lobby groups has become the determinant of public policy, skewing overall support toward aid for social sectors and away from market instruments for economic growth, something is very wrong.

    Charities routinely accept fungibility in their own programs, for example with money given for child sponsorship used instead for community development, because they admit they can get more donations that way.

    Distortion of aid priorities has reached the extent that some charities claim that basic health and education are the decisive sectors for economic growth. The result is that economic infrastructure, public financial management and generation of tax revenue have been relatively neglected.

    Official donors should think about leaving the health sector more to the charities so that government aid funds can focus on generating economic growth and tax revenue. Aid that causes an increase of local tax funding for sustainable service delivery is less susceptible to the problems of fungibility and dependence. It is a corruption of process to allow aid priorities to be determined by donor political preference rather than the measured needs of poor countries.

    Posted April 13, 2010 at 11:18 pm | Permalink
  12. Geir wrote:

    Rachel, you say that you are surprised that the article states that debt relief has had no effect on health spending. This is also what we found in an analysis of the Tanzanian budget before and after the start of the PRSP (as a conditionality of HIPC), as a part of the DFID financed Patterns of Accountability study (

    Whereas spending on social services went up as the total budget increased, the spending as a proportion of the budget did not change. In fact, when we ran the numbers, the PRSP did not register at all on the graph of relative budget allocation to social services. This meant that the HIPC conditionality that debt relief be spent disproportionally on social services was not honoured.

    The finding was not picked up by the donor supported Poverty Monitoring System.

    Posted April 14, 2010 at 3:44 am | Permalink
  13. Raphael wrote:

    As noted above, there is positive news in this Lancet article. Some health aid is apparently NOT fungible: “However, DAH [Development Assistance for Health] to the non-governmental sector had a positive and significant effect on domestic government health spending.”

    So what can we draw from this? Perhaps civil society is a better means of providing social services than govt (health, edu, etc). Certainly, there are some problems with this (e.g. staff moving from govt to the NGO sector, thus further weakening govt capacity). But, on balance, does this not point to a model of aid where civil society plays a more prominent role in basic service delivery?

    Posted April 14, 2010 at 4:59 am | Permalink
  14. Owen Barder wrote:


    I think it is funny that you say that there is “one bit of good news” in the article, as if it is bad news that governments are spending their money on national priorities.

    Leaving that aside, I strongly dispute your conclusion about using NGOs.

    As I pointed out here the other day there is no reason in principle why aid to NGOs should lead to a smaller shift of resources than aid through governments.

    Perhaps one explanation of the finding in the Hu et al is that spending through NGOs so disjointed and disorganised, and disconnected with real needs, that the government judges that this kind of aid does not lead to effective health care and so does not create space for them to shift resources elsewhere?


    Posted April 14, 2010 at 6:39 am | Permalink
  15. Raphael wrote:

    @ Owen. On your first point, I think it depends on what those national priorities are. Too many govts neglect the social needs of their populations and allocate resources to unproductive ends. I think there is a difference between “national” priorities and “government” priorities in many, many cases. Of course, donor priorities are not always national priorities either…. As always, context matters.

    On your second point, there is ‘principle’ and then there is ‘evidence.’ You make a fair theoretical point but there are many causal theories that may explain the Lancet findings. One of them is that civil society operates outside the information horizon of governments so they can’t account for it in their resource allocation decisions.

    Posted April 14, 2010 at 10:07 am | Permalink

11 Trackbacks

  1. By Fungibility and sloppy thinking on April 13, 2010 at 9:37 am

    […] allocations in response to aid, so I will not repeat that discussion here.  Alanna Shaikh and Laura Freschi have interesting things to say on this issue as well.  I would add only that I am not naively […]

  2. […] their thoughts on the findings; I’d recommend reading Alanna Shaikh’s UN Dispatch post and Laura Freschi’s Aid Watch post. All have an interesting perspectives and are worth a […]

  3. By » The future of health aid on April 13, 2010 at 6:05 pm

    […] If you are a wealthy government, how do you make sure the aid money you give to a poor country gets spent on something worthwhile? A popular way is to designate the money for a particular purpose, and one good purpose is health. If the poor country government is already spending a certain amount on health services, and you give them the same amount again, designated for health services, then the total expenditure on health care doubles, doesn’t it? Unfortunately not. […]

  4. By Other notes « Cousin Dampier's Blog on April 13, 2010 at 10:07 pm

    […] AidWatch: Does health aid to governments make governments spend more on health? […]

  5. […] health aid given to governments of developing countries cause those governments to cut their own health […]

  6. […] Freschi at Aid Watch makes the other contribution we should pay attention to, though I’m not uncritical of it. She says that donors use project […]

  7. […] outrage. Thoughtful, if not fully aligned, responses came from Laura Freschi and Owen Barder. But we can thank Ranil at Aid Thoughts for getting to the nub of the problem. As […]

  8. By The fungible and the furious « Aid Thoughts on April 16, 2010 at 9:35 am

    […] paper made some headlines and upset aid critics and much of the global health community. A part of this has to do with a misunderstanding of what […]

  9. […] Street and Aid Watch have great comments on a recent Lancet article. The article found that when donors increase […]

  10. […] yes, but the health advocates don’t like that idea much. As Laura Freschi on Aid Watch […]

  • About Aid Watch

    The Aid Watch blog is a project of New York University's Development Research Institute (DRI). This blog is principally written by William Easterly, author of "The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics" and "The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good," and Professor of Economics at NYU. It is co-written by Laura Freschi and by occasional guest bloggers. Our work is based on the idea that more aid will reach the poor the more people are watching aid.

    "Conscience is the inner voice that warns us somebody may be looking." - H.L. Mencken

  • Archives