Cost-recovery in the health sector: an inappropriate policy in complex emergencies
by Timothy Poletti, London School of Hygiene and Tropical Medicine March 2004

The introduction of cost-sharing mechanisms as part of healthcare programmes in complex emergencies has become a source of increasing concern to many humanitarian relief agencies. Cost recovery seems contrary to the humanitarian principle of impartiality and the allocation of assistance based on need alone.

Critics argue that, in already difficult healthcare environments, charging users fees compounds inequities in access to treatment and contributes to the destitution of the most vulnerable. Yet donors have increasingly made their funding contingent on having these mechanisms. Both donors and national governments see such policies as developmental; they believe that their introduction is inevitable, and that bringing them in at an early stage will contribute to building a sustainable, locally financed health system in the longer term.

The rationale for introducing cost-sharing in complex emergencies has not been well articulated, and the arguments in favour of it are largely ideological. The research base on cost-sharing in complex emergencies is extremely limited; there is insufficient evidence to develop more empirically based approaches. However, given the evidence available on cost-sharing in other resource-poor settings, and the results of NGO evaluations, there is a strong case that, regardless of whether cost-sharing mechanisms should remain a longer-term development goal, their introduction in complex emergencies is inappropriate and should be abandoned.

Cost-sharing in developmental settings

While cost-sharing is a recent innovation in complex emergencies, there is substantial experience with such schemes in the development sector. Cost-sharing became widely accepted as a necessary element of healthcare financing in the developing world in the mid-1980s. At that time, governments were unable to adequately fund public services including health, and out of pocket expenditure on health was growing rapidly as people (including the poor) were forced to seek care in the private sector.

Alternative sources of financing were clearly needed, and the World Bank began pushing for the inclusion of national cost-sharing mechanisms as a way of bridging what is known as the health sector resource gap – the shortfall between the funding provided by governments and donors and the level of funding required to provide a basic level of healthcare of acceptable quality.

The World Bank’s arguments in favour of cost-sharing were given added weight by the Bamako Initiative, developed by WHO and UNICEF and adopted by African ministers of health in 1987. This focused on the potential of user fees to increase the resources available for primary healthcare.

Three basic arguments have been developed to support cost-sharing:

  1. Increased revenue. User fees are one of the few feasible ways of raising revenue to bridge the health sector resource gap in resource-poor environments. There are other ways of raising revenue: increased donor funding; increased private philanthropy; economic growth and a consequently increased tax base; taxes on health-damaging products such as tobacco and alcohol; or increasing the share of government expenditure spent on health. However, none of these is likely to be achievable in the developing world, especially in Sub-Saharan Africa, in the near future.
  2. Increased efficiency. User fees, if well designed, should mean that resources are used more efficiently within the health system. They discourage unnecessary use, and can create incentives for providers and patients alike to shift the focus towards cost-effective high-priority care for disease prevention; they can also, via differential pricing, move the delivery of care away from expensive hospital-based treatment to more cost-efficient primary healthcare.
  3. Increased equity. If the income they generate is used to improve service quality, user fees could have positive equity outcomes. Even with user fees, a public health system that delivers high-quality care close to where people live would offer poor people cheaper and better care than they would be able to get in the private sector.

The verdict on cost-sharing in resource-poor settings

The results of cost-sharing in resource-poor settings have been disappointing. It has failed as a revenue-raising tool: although the World Bank had hoped for 15%–20%, user fees have raised an average of 5% of total recurrent health system expenditure, and even this is an over-estimate because it does not take account of the cost of collecting the fees.

User fees have, however, been able to generate a large proportion of non-salary recurrent expenditure, ranging from 10% to more than 100% in Sub-Saharan Africa. This can be significant in well-directed and well-managed health systems, where otherwise effective systems were failing because there was insufficient money for drugs and other medical supplies.

The evidence that cost-sharing improves efficiency is weak. There is no evidence to support the argument that user fees discourage unnecessary use, or that they have significantly altered patterns of service delivery in ways that increase efficiency. There is evidence that user fees can create perverse incentives that actually decrease efficiency by reducing the use of preventive services; reducing access for the poorest (and often sickest); and forcing people to wait until they are very sick, and then seeking more expensive treatment in hospitals.

Evidence as to the equity impact of user fees has been mixed, in part because it is difficult to disentangle the effects of price, quality of care and affordability on service use. In many cases, the introduction of user fees has led to significant and sometimes dramatic decreases in utilisation, despite mechanisms exempting certain sections of a population from having to pay. There is also evidence of an increase in utilisation following the removal of fees.

At the same time, there are examples that suggest that, if cost-sharing is linked to improvements in the quality of care, and particularly to the availability of drugs, user fees may benefit everyone (including the poor) by providing cheaper access to public care of higher quality. On balance, the introduction of cost-sharing will in many instances create a significant barrier to poor people accessing care.

There is also the question of willingness and ability to pay. Poor people may be willing to pay to access care, but they may be unable to do so without sacrificing their longer-term economic well-being through unsustainable borrowing or selling productive assets. This is referred to as catastrophic health expenditure.

By 1993, the World Bank’s position on user fees in healthcare had become more neutral, and there is now widespread consensus that the most efficient provision of healthcare involves free service at the point of delivery. Nevertheless, policy-makers within donor organisations and governments, both in the West and in the developing world, have bought into the World Bank’s mid-1980s arguments in favour of cost-sharing, and still retain a widespread belief that it is a necessary component of healthcare financing in the developing world – largely because there are no other options, since taxation or insurance systems are too complex to operate in these environments.

Cost-sharing in complex emergencies

A similar logic (that there is no other option) drives cost-sharing in complex emergencies where donor pressure has been crucial in pushing cost-sharing onto the policy agenda in these environments.

The majority of funding for health programming comes from donors, which means that donors have significant leverage over health policy. In its approach to supporting primary healthcare in the Democratic Republic of the Congo (DRC), for instance, the European Commission Humanitarian Aid Office (ECHO) has developed guidelines for partner NGOs which promote cost-sharing. As a consequence, ECHO-funded projects in the DRC include cost-sharing mechanisms.

The arguments in favour of cost-sharing in complex emergencies are the same as in other resource-poor settings: it can raise revenue; increase efficiency; and, with appropriate exemption mechanisms, increase or have minimal effect on equity. Cost-sharing is regarded by its proponents as developmental: the underlying logic is that, given the widespread introduction of cost-sharing mechanisms throughout the developing world, it is more than likely that such a scheme will be introduced in any case once peace has been restored.

The introduction of user fees is seen as a necessary step towards rebuilding a sustainable health system. This does not, of course, solve the problem of when a complex emergency might reach the point where developmental programming is appropriate, and opponents often criticise cost-sharing as an example of premature developmentalism.

Many people working in humanitarian relief have an instinctive resistance to the introduction of user fees in complex emergencies. It runs counter to the ethos and principles of humanitarianism, under which assistance should be rendered to people affected by conflict on the basis of need alone. To many, it seems absurd that people who are struggling to survive in difficult and unstable circumstances should have another financial burden placed on them. The potential for catastrophic health expenditure is self-evident in complex emergency settings, where people’s asset base is typically extremely vulnerable and their health needs are grossly elevated.

Critics point out that the characteristics of complex emergencies are such that transferring the policy from a development context is inappropriate: poverty is widespread, needs are high, governance is absent, per capita incomes are already insufficient to meet essential needs and the skills and capacity to support the introduction of user fees are missing.

There is very limited published literature on the impact of user fees in complex emergencies. In-house NGO assessments suggest that their capacity to raise significant amounts of money is limited. Utilisation rates indicate that, in already disrupted and inequitable healthcare environments, user fees compound inequities in access to treatment and contribute to the destitution of the most vulnerable. Little is known about who is discouraged from seeking treatment, or what the impact might be, but community-based surveys suggest that the poor are prevented from accessing care.

There is limited information regarding the impact of cost-sharing on the health of populations, particularly its implications for the control of infectious disease, which is a major source of morbidity and mortality in complex emergencies. However, if user fees do create an access barrier for the poor, it is likely that some individuals with infectious diseases will fail to get effective treatment because they cannot afford it. They then act as a reservoir of infection, with the potential for triggering and sustaining epidemics.


Definitively answering the myriad of complex questions that the introduction of user fees in complex emergencies raises would require well-designed academic studies. Such operational research is the only way to develop a foundation for more empirically based policymaking in this area. In the meantime, there is no evidence to support cost-sharing mechanisms in complex emergencies, and there are good grounds to argue that they should not be introduced. It is likely that cost-sharing in complex emergencies will:

  • Raise little money.
  • Have a significant negative impact on equity, which cannot be effectively mitigated via exemption mechanisms.
  • Have a negative impact on efficiency.
  • Result in unequal access to care.
  • Potentially tip individuals and families into destitution via catastrophic health expenditure.
  • Potentially hamper efforts to control epidemic infectious disease.
  • Needlessly increase the complexity of programming in already challenging environments, potentially damaging the motivation of local staff and the relationship between local and expatriate staff.

This constitutes a reasonable basis for arguing that cost-sharing should not be introduced in complex emergency settings. If donors reject this argument on the basis that the evidence is insufficient, the onus should be on them to fund further operational research to establish the basic conditions for the successful introduction of user fees. It should not be acceptable to make claims for any benefits from user fees when the basic underlying conditions do not exist for their introduction; where doing so has predictable negative impacts; where prior experience in more favourable settings is negative; and where levels of need are high.

If donors fail to change their policies, some NGOs and humanitarian agencies will be forced to continue to compromise their principles and include cost-recovery in programme design; others may withdraw altogether from the provision of health services in complex emergencies. If cost-sharing remains part of the donor policy agenda, we may perhaps need to redefine the humanitarian ethic as an obligation to prevent suffering and protect life and health – but only for people who can pay.

Timothy Poletti is a member of the research staff in the Conflict and Health Programme, Health Policy Unit, at the London School of Hygiene and Tropical Medicine (LSHTM). Olga Bornemisza and Egbert Sondorp, both of the Conflict and Health Programme, as well as Austen Davis of MSF-Holland, all rendered significant assistance in the writing of this article. Any correspondence about this article should be sent to

This article is based on a literature review and background study on cost-sharing in complex emergencies (Timothy Poletti, Healthcare Financing in Complex Emergencies: A Background Issues Paper, LSHTM, October 2003). This paper is available upon request from or The work was funded by MSF-Holland and the Andrew W. Mellon Foundation.

References and further reading

Implementing the 20/20 Initiative: Achieving Universal Access to Basic Social Services, a joint publication of the UNDP, UNESCO, UNFPA, UNICEF, WHO and the World Bank, September 1998.

D. Arhin-Tenkorang, Mobilising Resources for Health: The Case for User Fees Revisited, Commission on Macroeconomics and Health Working Paper WG3:6, November 2000.

A. Creese and J. Kutzin, Lessons from Cost-Recovery in Health, WHO Discussion Paper No. 2, 1995.

L. Gilson, ‘The Lesson of User-Fee Experience in Africa’, Health Policy and Planning, vol. 124, no. 4, 1997, pp. 273–85.

B. McPake, ‘User Charges for Health Services in Developing Countries: A Review of the Economic Literature’, Social Science and Medicine, 36, 1993, pp. 1,397–1,405.

M. Whitehead, G. Dalgren and T. Evans, Equity and Health Sector Reforms: Can Low-Income Countries Escape the Medical Poverty Trap’, The Lancet, 358, 8 September 2001, pp. 833–36.