Should we worry about fungibility of health aid?

A new article published in The Lancet by Chunling Lu with Chris Murray, Dean Jamison and others, has caused quite a stir in development circles.  They use data on health aid and government spending on health to estimate that for every $1 given in health aid, the recipient government shifts between 43 cents and $1.14 of their own spending to other priorities. (If the aid goes to NGOs, by contrast, government health spending appears to increase.)

Even if the quantitative analysis is correct (which is by no means certain, given huge gaps in information), it is far from clear that this is a problem that needs to be solved. Furthermore, of the five recommendations in the paper, three are irresponsible sectoral special pleading which deserve to be rapidly dismissed.

This story has spilled over into the mainstream press (for example, in The New York Times) as a result of a sensationalist AP story headed “Health Aid Made Some Countries Cut Budgets“. The story breathlessly reveals:

After getting millions of dollars to fight AIDS, some African countries responded by slashing their health budgets, new research says. For years, the international community has forked over billions in health aid, believing the donations supplemented health budgets in poor countries. It now turns out development money prompted some governments to spend on entirely different things.  … “When an aid official thinks he is helping a low-income African patient avoid charges at a health clinic, in reality, he is paying for a shopping trip to Paris for a government minister and his wife,” said Philip Stevens, of the London-based think tank International Policy Network.

The language used by the authors is less inflammatory, but the opening sentence makes it clear they think there is a problem:

Government spending on health from domestic sources is an important indicator of a government’s commitment to the health of its people, and is essential for the sustainability of health programmes.

As summarized in their press release, the authors make five recommendations to deal with this alleged problem:

  • adoption of a clear set of reporting standards for government health spending as source and spending in other health-related sectors
  • establishment of collaborative targets to maintain or increase the share of government expenditures going to health
  • investment in developing countries’ capacity to effectively receive and spend health aid
  • careful assessment of the risks and benefits of expanded health aid to non-governmental sectors
  • study of the use of global price subsidies or product transfers as mechanisms for health aid

The first recommendation is fine: I’m all for the adoption of reporting standards for spending by donors and by governments, and for those standards to specify the source as well as the destination of all spending. (The authors may not be aware of the progress that is being made globally on this under the International Aid Transparency Initiative).   It is also hard to be against investing in the capacity of developing countries to receive and spend health aid, though I wonder what this means in practice.  The other three recommendations are irresponsible, for reasons we shall come to below.

Let’s start with the problem we are trying to solve.  It is far from clear that the behaviour of developing countries described in the paper is anything we should be concerned about.  Of course health advocates who earn their living from health spending in developing countries are up in arms at the news that their various wheezes to capture a big chunk of available development finance and redirect it to their cause may not have been a complete success.   But those of us who take a more objective view of the relative priorities of different types of development spending can be more sanguine.

There are at least four reasons why the findings of the paper should not be a cause for concern.

First, it suggests that governments are reprioritising their spending in the light of the aid they are receiving. I think this is a good thing.    Exercises to find out what poor people actually care about, such as Voices of the Poor, routinely find that the poor place put a lot of value on security (of person and property), but this does not usually excite people who work in development.  Donors find it more attractive to finance health services than to pay for essential services such as a national statistical office or the efficient functioning of courts.  If we are willing to pick up the bill for health care then it is not only reasonable but desirable that developing countries should use the fiscal space we have created to invest more in important national priorities that don’t happen to be of interest to their donors.

Second, increases in aid for health may well come at the expense of other forms of aid which developing countries are right to try to offset.  (I say “may well” because of course we don’t know what would have happened to total aid if health aid had not increased so rapidly.)  Donor fads come and go: this year it is agriculture.  When developing countries see health aid rising, but the donors losing interest in infrastructure, the most sensible thing they can do is make an offsetting shift in their own budget allocations.  When the donor pendulum swings back again, recipient countries will have to make the corresponding shift in the opposite direction.

Third, as eloquently pointed out by Sridhar and Woods in the Lancet, the desire to force changes in the spending priorities of recipient countries runs directly contrary to the evidence about what makes aid effective, and a series of international agreements, especially the Paris Declaration (2005) and Accra Agenda for Action (2008). In the face of evidence that aid is most effective when there is ownership by the recipient country, donors and multilateral agencies committed themselves to align their aid with the systems and priorities of recipient countries.  It is not OK for health sector lobbyists to ignore this because they don’t like the priorities actually chosen by developing countries.

Fourth and finally, we say that we want to see capable, accountable and responsive states in developing countries.  Making, passing and executing budgets is the very heart of a capable and accountable state. That is why in the UK, as in many other western-style democracies, a government which cannot pass its budget (“carry supply”) is deemed to be unable to govern.  If resource allocation priorities are determined elsewhere, then the government is one in name only.  We cannot expect governments to be accountable to their citizens for decisions that they have not made.  If we want accountable states rather than puppet client states, we should rejoice, not complain, when they demonstrate a willingness to make choices of their own.

Sectoral advocates may say that we should not accept the priorities determined by developing countries, especially in countries in which there are weaknesses in democratic accountability or technical ability to execute budgets.   They might say that the government represents the interests of an elite, not the majority of the country’s poor.  Of course that may be true in some countries: but there is no reason to think that donors’ priorities, also driven by vocal lobby groups and vested interests, reflect the real needs of a country or its poorest people.  We should avoid getting into the situation in which well-heeled foreign academics and lobbyists from international NGOs with no accountability to people in developing countries are treated as a more representative voice of the poor than their own government.

What is most shocking about this paper is that it betrays a combination of ignorance of, or indifference to, decades of experience about what works in development.  The three most egregiously inappropriate recommendations amount to setting input targets, bypassing government by using NGOs, and giving aid in kind rather than in cash.  The paper’s authors should pause to reflect on the fact that progressive development thinking has fought a long, slow, painful campaign to shift away from exactly this kind of aid, and for very good reasons.  Aid that leads to long-term, sustainable change must be based on real ownership of the developing country and help build rather than undermine or marginalise national institutions.

To be fair to the authors, the press release is quite measured, and it begins by highlighting the commitment to health by developing country governments.  It also highlights the most important and sensible of their recommendations, the need for greater transparency.   But the paper also irresponsibly creates the impression, amplified by the Associated Press, that health aid has somehow been wasted, and that donors should try to address this in ways that would be a couple of steps backwards on the long slow road to more effective aid.

19 comments on “Should we worry about fungibility of health aid?”

  1. I agree Owen that it’s important not to misinterpret these findings or overreact. And let’s be honest, most development experts have known about fungibility for a long time.

    That being said, the estimates of the proportion of reallocation in the countries receiving the largest amounts of health development assistance are alarming. The quality of spending and the rationality of what gets done vs not are very different where a country’s health expenditure is largely the sum of a bunch of programs, than when they are more sensibly allocated to generate the range of things a government and it’s people want from their health system. It’s one thing if the gov’t spending covers the operation of the basic service, and programs come in to add more around that (I think this is what health donors generally pretend they are doing). It’s another if the funds allocated to the basic stuff get shifted, as donor program funds come in. The systemic distortions in this scenario are profound. Many of the health systems in the countries receiving the largest amounts of aid have been stagnant or declining in terms of performance for many years – even where, in some cases, other sectors have started to improve. You have to ask, is health aid holding them back in this regard. Lu et al’s results suggest that the answer may be “yes”.

  2. I also think it matters if you raise money by telling the potential donor that their money will be spend on health (save a child from malaria, save a woman from dying in delivery)and then the funds essentially get spent on something else ( a dollar of whatever Malawi or Ghana spends their marginal dollar on).

    Aid funding needs to be accountable in two directions, does it not?
    That’s not to say we should be complacent about the distortions that come from the fact that funds are much easier to raise for some causes than others – but surely the answer is not to just raise the money by misrepresentation.

  3. April, you seem to suggest that by giving so much money in health aid, we are somehow undermining health systems. Is it not more to do with the kind of health aid that has been given that is the problem? The vast majority of aid for the health sector either bypasses entirely country systems or is loosely “coordinated” with them. Only more recently, in reaction to the heavy criticisms, are donors beginning to focus a bit more on strengthening national systems, rather than focusing on treating specific diseases in an isolated way.

    Today in Mozambique, the government has resorted to dividing up the provinces and assigning international organisations – mostly NGOs- to each province to provide HIV/AIDs treatment. While they coordinate loosely with the district health services, they are still independent structures. Of course in the long term this is going to mean weak health systems – but donor/NGO policy is as much to blame for this as government budget (re)allocations.

  4. In an ODI study about perception of effectiveness by recipients, the depth of the commitment was apparently the main factor.

    If aid would be only a transfer of money from donor A to recipient B, the Aid system would be as simple as writing a money transfer.

    It is remarkable how many aid decision makers talk about aid as if it is only about transferring money. Why don’t they ever take this to its logical consequence?

  5. Lucy, I agree that the negative effect I’m talking about is generated partly, perhaps even mostly, from HOW the aid flows.
    As you note, health aid tends to flow through vertical channels, and often separate administrative structures. These separate channels are in some ways very effective. Implementation can be much more controlled by donors and their agents/ implementing bodies. They can deliver measurable outputs, and even sometimes impact (as GAVI funded immunization campaigns do). And people and agencies feel good about funding them. Buy a bednet. Save a kid.
    Ask the AIDS advocates who are criticized for causing too much health funding to flow their direction; they will say, we changed the game in getting people and agencies to be willing to fund a health issue in poor countries. Wouldn’t happen if you tried to raise money for say “health systems’ strengthening” let alone “increasing on the margin” the budget in country’s budget.

    So yeah, it has to do with how the money flows, but I don’t see honestly how it will ever change (notwithstanding all the talking about harmonization and “country ownership”.

    I hope someone will post after me and give insight from the frontline that shows my pessimism is unfounded.

  6. I think Sam has a point here. Although I am very sympathetic to the argument that donors give based on their own preferences rather than those of recipient countries and that this can be a bad thing, I also like accountability and transparency. Turning a blind-eye to recipient countries when they essentially lie to donors by spending health (or any other cause)-directed funds on other priorities may be occasionally forgivable given how imperfect donors are, but it should never be directly defended.
    A while back you argued for what I termed a “global welfare system.” Here your willingness to excuse the re-allocation of donor funds by recipient countries sounds like its coming from a similar line of thinking, i.e. you’re taking Mr. Gardner up on his offer to “take this to its logical consequence.” Am I connecting the dots correctly?

    Justin – I think you’ve misinterpreted the conclusions of the paper. Developing countries took money from donors and spent it on health, exactly as they said they would. No lying involved there. At the same time, they made choices about their own budget allocations in the light of the rising donor interest in funding health, and chose to spend their own money on something else. There is no suggestion of deception or even lack of transparency.

    It is possible, of course, that some recipients may have given some kind of commitment to donors about the levels of domestic funding for health, in which case you would have a point if those commitments were subsequently broken. But there is nothing in the study about whether domestic resource allocation commitments were given or broken as a consequence of health aid.

  7. Just a note for clarification: I am fully with Owen on this issue. I think as a donor, you are partnering to a recipient “for good and bad days” and help them to solve development issues in a certain sector. This is what is depth of commitment. It means also respecting the recipients’ priorities.

    In my experience, this depth of commitment (I wrote a post on it)is more important than the money. I advocate strongly for committed donor interventions, with expertise and respect for the local systems and needs.

    However, the development circus has been focused lately mostly on funding by committee. Donors have been scaling down their in house expertise on substance, replacing it with expertise on funding arrangements. However, if a donor has no substantial expertise anymore, his contributions will be administrative instead of substantial. In that case, it would be better if he would go out of the way and simply transfer the money, perhaps to an intermediary who does care.

    I do think it is good practice for recipient countries to reassign the budget according to the needs. As accountability, focus and capacity building within a project is in general better than outside of it, even with some fungibility, the impact of the project can be important.

  8. Perhaps “lie” is a harsh characterization, but it is not entirely inaccurate. When I said “turning a blind-eye…” I recognize that recipient countries aren’t directly spending health aid on agriculture (or whatever) but that doesn’t mean they are being transparent either. You call the AP story “sensationalist” and argue that there is no deception occurring. But obviously some people are surprised, and unhappy, about the degree of fungibility going on. When donor’s give money to whatever cause they are doing so because they believe it is a necessary “supplement,” not because they want to free up that government’s budget for other projects. As I said, donor priorities can be (often are) wrong, but then you should change the priorities not condone fungibility. Just north of me in the DPRK South Koreans have long been uncomfortable with the fact that their humanitarian aid has allowed Kim Jung-il to allocate more resources to military armament.
    To add a little personal experience, an enterprising man in the Gambia once asked me to help him secure funding for an AIDs awareness workshop at his village school. He was not particularly interested in AIDs but had worked out a deal with the local headmaster whereby if he could secure an AIDs workshop the headmaster would be free to allocate other funds for a school project in which this man had a material interest. Although I admired his honesty with me, and even sympathized with the other project, I turned him down because I couldn’t ethically go to the AIDs donor knowing that funding for the AIDs workshop was motivated by, and making possible the implementation of, a project that had nothing to do with AIDs. Are you saying that I should have had no qualms about securing the man his AIDs workshop?

    @Sam, I misunderstood your comment. I completely agree with you that “deep” commitment is necessary and think that if donor’s had such commitments it is likely that recipient countries would be able to ask them for funding in the sectors and amounts that truly mattered to them instead of playing to donor fads and then reallocating the money afterwards.

    Justin: your argument places a lot of weight on your supposition that this reallocation is in some way a deception. Clearly the global health lobby is up in arms about the reallocation, but that doesn’t tell us that the recipient countries have given and then breached undertakings about how their own budget allocations will respond to these aid inflows.

    An article about recipient countries having given and then failed to meet commitments on global aid spending, with suitable evidence, would be interesting, and there would be a case to answer. But there is no such evidence in this paper, which merely complains that when we give health aid we prefer that recipient governments suspect their critical faculties when making their own budget choices.

  9. “reallocation is in some way a deception.” I’m not sure how one can see it in any other terms. Unless the recipient countries state clearly “We want your health aid because it will free up our money for other projects” shouldn’t the (naive?) assumption be that the recipient country’s desire for the health aid is motivated by a need to supplement health priorities and not in order to free up money for agriculture (or nukes in the case of the DPRK)? And if we are worried about implementation at all, what are the chances that donor supplied health funding will be well spent if the original motivation for acquiring it was simply to free up money for another project? This was one of my key concerns in the Gambia.
    I agree that paternalistic donors who try to either run, or alternatively side-step, recipient governments are problematic, but lets not pretend that recipient countries don’t engage in a lot of deception in regards to how donor funds are spent and that “reallocation” is not a central tool of that deception.

    Justin – This just isn’t right. Donors and recipient countries have signed both the Paris and Accra Declarations saying that donors will align with recipient country systems, including their budget priorities. There are occasional specific commitments to e.g. sustain social sector spending above a certain level, but other than where such agreements exist, donors are not entitled to assume that recipient countries will not change their budget priorities.

  10. and let’s not pretend that if most individuals making donations to various health causes were asked to allocate money to: the marginal budget priority for X country, they would be giving much much less.

    And if aid agencies had to get their budget approval with clear reference to the implications of this fungibility, likewise, probably a lot less money would be forthcoming. Probably USAID is an outlier, but much of their funding is earmarked and with quite specific deliverables for dollars spent.

    As an aside, this discussion is making me think that a lot more development assistance should be devoted to improving public expenditure management. If our money is mostly going to the marginal government priority, I expect we should care a lot more about how those priorities are determined, and how well the spending is executed.

  11. Perhaps I am not being clear. Paris and Accra say that donors should align their funding with recipient country budget priorities. I agree. But clearly this isn’t always happening because recipient country’s feel it neccesary to reallocate their funding. Where we seem to disagree is how one corrects this problem. Rather than condoning such reallocations I would prefer to blame the donors for not aligning themselves with recipient government priorities. You, on the other hand, do not seem to see such reallocations as problematic. I gave a few reasons why I thought they were. 1. Donors have the right to expect that recipient countries do not have ulterior motives when applying for sector-specific aid. You disagree and say they have no such right (Fine, though I am unnconvinced). 2. Donors are right to be concerned if their funding may be indirectly supporting unknown programs. 3. I am skeptical how well sector-specific donor funding will be spent if recipient countries do not percieve a real need for it in that sector. Do points 2 and 3 not concern you?

    Justin – I don’t agree that it is a “problem” that developing countries reallocate their funding; so I don’t agree that we need to correct it.

    You are making a serious allegation that developing countries are taking sectoral aid on false pretences. You need either to stand that up or withdraw it.

  12. I have no knowledge of or expertise in development theory, but I was once involved in the practicalities of giving — sorry, ‘delivering’ — aid, including at one stage budget support for a developing country that was chronically short of foreign exchange. All of us concerned in this had very much in mind that money itself is ‘fungible’: i.e. one unit of currency is interchangeable with another unit of the same currency or its equivalent in another: once money enters a pool such as the consolidated fund, it doesn’t carry ID enabling us to trace it back to its origins.

    It seems to me therefore misleading to discuss whether development aid ear-marked for (e.g.) health should be ‘fungible’, or can be more or less fungible. Unless the aid takes the form of hospital beds or actual drugs shipped out to the recipient country, health aid like any other aid is fungible, whether we like it or not. Even if we physically ship out hospital beds as aid, the beds are likely to free up for the recipient government the funds in its own treasury that it would otherwise have spent on the same number of beds. If the recipient government chooses to spend the funds so freed up on agricultural development, or even on a private jet for the President, I don’t see how the donor of the beds can complain: it’s their money, not the donor’s. As Owen keeps having to point out, it’s wrong to describe such expenditure decisions by a recipient government as a ‘transfer’ of the donor’s money to a sector for which it was not intended, unless a condition of the gift of the beds was that the recipient government’s health spending would be kept at a pre-determined level regardless of the arrival of the beds, and the diversion of the counterpart funds causes a breach of that condition. Otherwise, it’s impossible to say that a specific new expenditure item, whether on a new presidential jet or on a new agricultural research college, has been paid for with the specific money released by the supply of hospital beds. You can’t mark up another government’s bank notes so that you know how they are spent.

    In the case of budget support aid, we (the UK government at the time) agreed with the African government concerned a list of development goods on which the foreign exchange we were giving was to be spent, seeking to satisfy ourselves as far as possible that these would be additional to what the recipient government would have bought if the extra foreign currency had not been available. We also agreed with the recipient government on the uses, in broad sectoral terms, to which the local currency counterpart of our aid would be put, to make sure as far as we could that it would be spent for broadly developmental purposes. Negotiation of all these understandings or agreements started from the position that it was up to the recipient government, not us as the donors, to propose the purposes for which they would spend the foreign currency and local currency counterpart funds, in accordance with their own development priorities, although obviously we were free to query their proposals if they seemed patently perverse or irrational (which they were not, in practice). But I would accept the obvious criticism, with hindsight and in the light of the way development aid practice has developed in the decades since, that the safeguards we insisted on in those days involved a significant amount of time and paperwork for the already over-stretched recipient government departments concerned; if five or six other donor governments had been making them jump through all the same hoops, the resulting burden could well have been intolerable. However, at the time, UK budget support represented by far the greatest part of all the bilateral aid then being provided to this particular country and I don’t think that our requirements were regarded by those concerned as unreasonable.

    Times have changed, though, and it seems to me that donors now need to accept that recipient governments must be free to spend their own money as they choose. Of course if their policies are manifestly perverse or corrupt, donors will not give them so much aid, or any at all. But the last thing we should be doing, surely, is return to the bad old days when the IMF, with enthusiastic support from the western donor community, imposed structural adjustment policies on dirt-poor countries, with such disastrous results that there was rioting in the streets in protest, and governments fell like ninepins.

    No doubt these thoughts would be more persuasively expressed in developmentese. But I no speak de language.

    Brian (declaring but uninfluenced by the paternal relationship)

  13. In my experience reallocation is not deception. Authorities in virtually all the countries I have worked with have been very clear about their budget constrains and needs with the donor community. From discussions of annual budgets which include reallocation of resources away from overfunded programs and or sectors towards underfunded ones, to reallocations to respond to sudden influx or gaps in donor funding (as you know year by year funding is highly volatile. This is responsible financial management and public policy. Also in my experience most sensible bilateral and multilateral donors support it, knowing as they know that their constituencies will only respond to certain things (very few people will provide support for revamping the justice system, or even for rural roads. That is why a lot of bilateral aid goes through budget support or flows relatively untied to an overall sector. Those same bilateral donors are the biggest financiers of what I will call “the issues programs” which account for the lion share of the substantial increase in health aid. Those issue programs are the ones that sell well with the public and pacify the big NGOs. So in my opinion if anyone has to be accused of deception is the bilateral donors that are talking from both sides of their mouths on the one to do what is fiscally and aid responsible and on the other what will pacify the big national NGOs and interest groups.

    However, my preference will not be to accuse anyone but rather as Owen points out to do a more thorough analysis including one that will shed light on whether that reallocation resulted in better or worse health outcomes, because at the end of the day that is what matters. Many of the other sectors towards which the responsible governments allocate domestic funds have as much and at times more impact in health as health services. As a minister of finance told us “..Madame, there is not a global fund for feeder roads ”.
    I applaud the courage of governments to manage their country and strive to serve the multiple often conflicting needs of their populations while advocating for transparency across all government spending and evaluation of results, not to punish but to redress policies.

    Owen replies: I agree with this.

  14. As per Montse’s mention of “the issues programs which account for the lion share of the substantial increase in health aid… that sell well with the public and pacify the big NGOs,” this whole fungibility debate illustrates how the emotional priorities of aid charities have become the tail that wags the dog of aid spending priorities. When “pacifying” aggressive rich country lobby groups has become the determinant of public policy something is very wrong. Charities routinely accept fungibility in their own programs, for example with money given for child sponsorship used for community development, because they admit they can get more donations that way.

    Emotional distortion of aid priorities has reached the extent that some charities even 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 neglected. Official donors should think about leaving health to the charities so that more aid funds can be used to generate economic growth, tax revenue and local funding for sustainable service delivery. It is a corruption of process to allow policy to be determined by donor preference rather than the needs of the poor.

  15. I think you are misunderstanding my argument. I will try again to be clearer. My contention is that whenever a recipient government reduces it’s funding for sector (A) in order to fund another sector (B) because it knows it will receive outside funding specified for sector (A) (which is what the Lancet report says is happening, as does Dan’s link), then they are engaging in a kind of deception IF the donor is unaware at the time that the funding reduction will occur because of its donation (and obviously some people, as the AP article shows were/are unaware). We seem to be disagreeing over my characterization, not the data. You are putting the onus on the donors to secure explicit guarantees (if they think it appropriate) from recipient countries while I am saying such guarantees are (should be) implicit. In any case we will probably have to agree to disagree here as I sense (tell me if I am wrong) that your position is founded upon a more general understanding of the relationship between donors and developing countries that you have written about in the “global welfare system” post that I alluded to earlier.
    Now certainly if the donor is aware that its funding for sector A will cause the recipient government to reduce funding for sector A and increase it in sector B, then there is no deception occurring. And it appears (according to Montse at least) that donors are often aware. But if so, why isn’t it simply explicitly funding sector B (or sector’s A and B)? This is the point on which we agree, namely that this kind of reallocation arises because donors aren’t (but should be) aligning themselves more with recipient country’s budget priorities.
    Finally when you say you don’t think that “developing countries reallocating their funding” is a problem, I think you need to be more specific. Certainly I am not arguing that ALL reallocations are problematic, just as I assume that you are not stating that ALL reallocations are unproblematic. I am simply saying that this kind of reallocation is problematic, and that “it takes two to tango.” Donor intransigence is setting up what Rachel Burke (who agrees with you over on AidWatch) calls a “silly game” and unfortunately developing countries are playing along. I’m simply saying, “shame on both of them.”

Leave a Reply

Your email address will not be published. Required fields are marked *

Published by

Owen Barder

Owen is Senior Fellow and Director for Europe at the Center for Global Development and a Visiting Professor in Practice at the London School of Economics. Owen was a civil servant for a quarter of a century, working in Number 10, the Treasury and the Department for International Development. Owen hosts the Development Drums podcast, and is the author Running for Fitness, the book and website. Owen is on Twitter and