ROUND TABLE
Does earmarked donor funding make it more or less likely that developing countries will allocate their resources towards programmes that yield the greatest health benefits?
L'affectation des fonds par les donateurs favorise-t-elle plus ou moins l'allocation des ressources des pays en développement aux programmes apportant les plus grands bénéfices pour la santé ?
¿Los fondos de donantes para fines específicos hacen más o menos probable que los países en desarrollo asignen sus recursos a los programas que generan los mayores beneficios sanitarios?
Catriona Waddington
Head of Learning, Institute for Health Sector Development, 27 Old Street, London EC1V 9HL, England (email: catriona.waddington@ihsd.org)
ABSTRACT
It should not be assumed that earmarked donor funding automatically increases the allocation of developing-country resources towards programmes that yield the greatest health benefits. Sometimes it does, sometimes it does not how the funding is designed can influence this. This is true particularly in the longer term, once the earmarked funding has ended. Even in the short term, total funding does not necessarily increase because of fungibility (i.e. recipient governments adjust their spending to offset donor funding preferences). The author explores six problems with earmarked funding: the multiplicity of earmarked funds confuses the situation for decision-makers; earmarking works against the spirit of the sectorwide approach; from the national perspective, it makes sense not to double-fund activities; local ownership of an activity is often compromised; earmarking can lead governments to accept interventions which they cannot afford in the longer term; and earmarking can distort local resource allocation.
Keywords: Financing, Organized/organization and administration; Resource allocation; Health expenditures; Health priorities; Health services needs and demand/economics; Financial management; Developing countries (source: MeSH, NLM).
RÉSUMÉ
On ne doit pas supposer que l'affectation des fonds par les donateurs majore automatiquement le volume des ressources affectées par les pays en développement aux programmes apportant les plus grands bénéfices pour la santé. C'est parfois le cas, parfois non - la façon dont le financement est conçu pouvant influer sur ce résultat. C'est particulièrement vrai à long terme, une fois les fonds affectés entièrement dépensés. Même à court terme, le montant total des fonds n'augmente pas nécessairement en raison de la fongibilité (c'est-à-dire que les gouvernements bénéficiaires ajustent leurs dépenses pour compenser les préférences en matière de financement des donateurs). L'auteur examine six problèmes liés à l'affectation du financement : la multiplicité des fonds affectés rend la situation confuse pour les décideurs, l'affectation des fonds va à l'encontre de l'esprit des approches sectorielles, du point de vue national, il est logique de ne pas financer doublement les activités, la prise en charge locale des activités est souvent compromise, l'affectation peut conduire des gouvernements à accepter des interventions qu'ils ne peuvent se permettre à plus long terme et elle peut introduire des distorsions dans l'affectation locale des ressources.
Mots clés: Organisation financement/organisation et administration; Allocation de ressources; Dépenses de santé; Priorités en santé; Besoins et demande services santé/économie; Gestion financière; Pays en développement (source: MeSH, INSERM).
RESUMEN
No debe darse por sentado que los fondos asignados por los donantes para fines específicos se traduzcan automáticamente en un aumento de la asignación de los recursos de los países en desarrollo hacia los programas que generan los mayores beneficios sanitarios. Unas veces es así, y otras no, todo depende de cómo se planifique la financiación. Esto es válido particularmente a más largo plazo, una vez concluida la financiación para el fin particular. Incluso a corto plazo, los fondos totales no necesariamente aumentan, debido a la sustituibilidad (los gobiernos beneficiarios ajustan su gasto para compensar las preferencias de financiación de los donantes). El autor examina seis problemas asociados a la financiación con fines particulares: la multiplicidad de los fondos destinados a fines específicos induce a confusión a los decisores; esa forma de financiación es contraria al espíritu del enfoque sectorial; desde una perspectiva nacional, carece de sentido financiar doblemente las actividades; a menudo se pone en peligro la identificación de la población local con la actividad; la financiación con fines específicos puede llevar a los gobiernos a aceptar intervenciones que no puedan costearse a largo plazo; y existe el riesgo de distorsionar la asignación de los recursos locales.
Palabras clave: Organización del financiamiento/organización y administración; Asignación de recursos; Gastos en salud; Prioridades en salud; Necesidades y demanda de servicios de salud/economía; Administración financiera; Países en desarrollo (fuente: DeCS, BIREME).
The dilemma faced by developing countries which are offered the prospect of applying for international aid for a specific intervention such as vaccination, for example, are similar to those that beset a town council planning its budget. Imagine a council meeting in a small poor town: the council considers applying for some earmarked funds one of 15 such initiatives it has discussed during the past year which, together, are worth US$ 100 000. Now picture another scenario: at its annual discussion, the council has good news: a consortium of donors has agreed to increase its budget by US$ 100 000, and papers have already been signed describing in broad terms the processes to be used for deciding how the money will be spent. Which council meeting is more likely to end up with a decision to allocate resources to activities that yield the greatest benefits? Obviously, there is no correct answer, as it depends on a variety of factors.
In the short term, it is perhaps likely that spending will rise on the earmarked activity. But even then, it depends: is the application process too complicated to make applying worthwhile? When the money is received, does the council react by moving its own money away from that activity, meaning that overall spending has not increased as much as may initially appear? In the long term, the earmarked funding will cease. What happens then depends on what the council thinks about the activity. Is it a council priority? Does the council "own" the programme? Has the council become accustomed to not funding the earmarked activities itself?
Note that the assumption is made that the earmarking is sensible, in that it is for highly beneficial activities. In practice, however, this may not be the case, especially as different countries have radically different needs. Over time, and juggling with 15 separate earmarked funds, the council may face a wider problem. All donors want to fund preventive activities, and none of them is interested in bearing the costs of salaries, maintenance, or supporting hospitals.
How these difficult issues are dealt with depends on the nature of the council. Is it democratically accountable? Efficient? Corrupt? Dominated by narrow interests? A well-run council will try to match its priorities to the available funding, while a short-sighted one will be likely to accept what money is on offer, without plans to continue activities when the donors withdraw.
Now consider the situation with developing countries and donor aid, where similar issues are faced. The six considerations described below mean that earmarked funding does not directly cause greater allocations for priority health activities. The overall conclusion is not a blanket condemnation of intervention-specific earmarked funds, but rather the point that it is too simplistic to assume that more earmarked funding means more local resource allocation. Earmarked funding has to be applied intelligently, in a way that is sensitive to individual country circumstances.
Multiple earmarked funds
The multiplicity of earmarked funds confuses the situation for decision-makers. The amount of resources allocated by government for, say, immunization is essentially decided at two levels: government-wide, in terms of how much is allocated to the ministry of health, and within the ministry in terms of how much is allocated to the activity. Ideally, the decisions will be based on a reasonably transparent mix of political and technical considerations about cost and relative priority. Earmarked donor funds operate by altering the cost to the government; donors in effect say, "do this activity, because we will subsidize it".
The Commission on Macroeconomics and Health, after considering data it had received from two United Nations agencies, the multilateral development banks and two bilateral organizations, described the composition of an estimated US$ 1.7 billion to support disease-specific activities as follows: 20% for HIV/AIDS (including sexually transmitted diseases), 15% for vaccine-preventable childhood diseases, 10% for maternal and perinatal conditions, 5% for malaria, 4.5% for tuberculosis, 3% for noncommunicable diseases, and 40% for other disease-specific activities (1, p. 17).
Who is to say that this allocation is rational? Is the benefit of spending on maternal and perinatal conditions twice that of spending on malaria? If the earmarking cannot be justified internationally, there is no reason to suppose that it would lead to rationality at the national level.
Prioritizing in the health sector is notoriously difficult, compounded by acute shortages in many developing countries of people who are skilled in this area. Multiple earmarked funds can send out confusing signals, both in terms of why certain activities are priorities and also in practical terms of understanding the various rules for different funds.
Sectorwide approaches
Sectorwide approaches emerged in response to the problems associated with numerous donor projects and earmarking arrangements, which made sector coordination difficult as regards overall prioritizing and practical management. The idea of sectorwide approaches is that government and donors sign up to one sectoral plan and work together on prioritizing and monitoring the activities (1).
Earmarked donor funds are not incompatible with the sectorwide approach; the extreme view that such an approach requires all donors to contribute to one common funding basket is not supported here.
In practice, however, earmarked funds often operate against the spirit of the sectorwide approach. By their very nature, earmarked funds do not address priorities across a sector. Rather than fitting in with government procedures, earmarked funds often come with their own application procedures, timetables and monitoring requirements, which can take up a disproportionate amount of the time of scarce senior government staff.
Donors with earmarking often want to work quickly. Sectorwide work proceeds more slowly. The whole point of the sectorwide approach is that countries are in the driving seat for prioritization and coordination. Donors with a strong incentive to supply funds for predetermined activities work against this, and it can be very difficult for poor countries to say "no" to proffered resources.
Duplication
There is a sense in which developing countries cannot win when it comes to funding priorities. In the 1980s, donor funds were overwhelmingly channelled to projects for primary health care. Far less international money was available for hospitals, yet these were still needed and demanded by the population, so national governments predominantly funded curative care, leading to the self-fulfilling accusation that governments in developing countries neglected primary health care.
A government that responds to earmarked donor funds by moving its own money away from that activity is often behaving rationally and responsibly: it is avoiding duplication and putting the freed-up resources to another use. The risk is that these new funding patterns become established, and that they reduce the likelihood that funding for the externally supported activity will continue beyond the period of donor earmarking.
Recipient governments adjusting their spending to offset donor funding preferences is called fungibility. Fungibility markedly reduces the usefulness of earmarking.
Local ownership
Ownership is strong domestic support for a priority, policy or process. For continued government support, political and administrative ownership are crucial. Securing it often requires long-term, in-country involvement, which is frequently absent from earmarked funding.
The Commission on Macroeconomics and Health identified some indicators of ownership, including: ministry of health chairmanship of major sectoral and subsectoral meetings; ministry of health control of drafting and finalization of major policy documents; and ministry of health leadership of donor coordination processes. (1, p. 31).
In practice, earmarked funds often bypass these policy and coordination processes. A single-issue interagency coordination committee or similar body is no substitute, particularly if it operates largely independently of other high-level bodies, thereby jeopardizing local ownership and the longer-term willingness to fund.
Affordability
For continued government support, the earmarked activity must be viewed as affordable and a top priority among many competing demands. There are many health interventions that yield great health benefits. Some of them are highly cost-effective, but nevertheless unaffordable to some countries.
Affordability can be an issue when funds are earmarked. In Kenya, the annual cost per capita of the Expanded Programme on Immunization was US$ 0.23 this almost trebled to US$ 0.63 when external funds enabled hepatitis B and Hib (Haemophilus influenza type b) to be added to the immunization schedule (2).
Ministries of health in developing countries face a cacophony of demands to regard particular interventions as a high priority. The population wants good quality, affordable hospital care; there are many special-interest groups; and the international community has a long list of "essential" activities. Beyond health, there are equally forceful demands for clean water, school textbooks and adequate food. It is naïve to assume that government will take over funding an activity because it is important and cost-effective. Many activities fit this description poor countries cannot afford them all.
Resource allocation
Earmarking can distort resource allocation in unintended ways. Earmarked funds rarely cover the whole cost of an activity. Typically, it is assumed that a network of staff and health facilities already exists, whereas earmarked funds are designated for items such as drugs, vaccines and transport. For example, a meeting in 2003 on the economics of vaccination in low- and middle-income countries did not discuss staff costs at all, and these were generally omitted from immunization costs. District health administrations and individual facilities generally receive very small cash budgets from government and have to rely on donor funds to keep going. This inevitably distorts the use of their staff time and other resources in favour of the donor-supported activities, which are not necessarily their own priorities.
In the long term, the very priorities that earmarked funds are intended to support can be compromised by this distortion. While funding for drugs, vaccines and other consumables is relatively abundant, in many developing countries there are simply not enough trained, motivated and reasonably paid staff to use these resources efficiently.
Conclusion
It should not be assumed that earmarked donor funding automatically increases the allocation of developing country resources towards programmes that yield the greatest health benefits. Sometimes it does, sometimes it does not how the funding is designed can influence this. This is true particularly in the longer term, once the earmarked funding has ended; but even in the short term, fungibility means that total funding does not necessarily increase.
In favourable policy environments, with reasonably effective governments, donors should work through national systems of prioritization, financial management and monitoring. Money is best provided on a sectorwide or even government-wide basis: earmarking should be avoided as much as possible. When there is earmarking, it should be in line with government priorities and should adhere to existing procedures. New interventions should be appraised carefully, with strong local involvement and without undue haste.
In less favourable policy environments, it must be accepted that there is really no guarantee at all that the government itself will devote resources to the earmarked activity (3): "Whilst it is difficult to withdraw support, attempts to work around a poor-quality public sector are unlikely to produce anything worthwhile or lasting. In these cases donor support should be geared less to financing and more to activities that in the long run may lay the groundwork for institutional and policy reform."
References
1. Diabre Z, Lovelace C, Norberg C. Report of Working Group 6 of the Commission on Macroeconomics and Health. Geneva: World Health Organization; 2002.
2. Kaddar M, Lydon P, Levine R. A critical review of financial sustainability. London: London School of Hygiene and Tropical Medicine; 2003. Draft paper for the Workshop on The Economics of Immunization, p. 8.
3. Assessing aid: what works, what doesn't and why. New York: Oxford University Press; 1998. World Bank Policy Research Report, p. 21.
(Submitted: 2 February 2004 Final revised version received: 12 May 2004 Accepted: 14 May 2004)