Working Group Paper

Project: Centre on Global Health Security

Trygve Ottersen, Aparna Kamath, Suerie Moon and John-Arne Røttingen
Shared Responsibilities for Health: A Coherent Global Framework for Health FinancingPhoto by EdStock/iStock.

The challenge

The past two decades have witnessed a tremendous increase in total development assistance for health (DAH). Now, however, the DAH system is challenged on several fronts: by the economic downturn and the stagnation of DAH, by the epidemiological transition and the rise of noncommunicable diseases, and by the economic transition with the rise of many middle-income countries (MICs). These trends all challenge the normative framework for DAH and call for a careful assessment of allocation criteria and contribution norms. The need for such an assessment is further highlighted by the upsurge of proposals for new financing mechanisms for global health and by the ongoing discussion about the post-2015 development agenda.

The general objective of this paper is to examine allocation criteria and contribution norms for DAH. More specifically, it:

  • Reviews the allocation criteria stated by major institutions and estimates distributional implications related to different criteria; and
  • Reviews recognized contribution norms and, in the context of these, estimates total need and total available funds for DAH.

Criteria stated by major institutions

We examined the DAH allocation criteria explicitly emphasized by five multi- or polylateral institutions and 10 bilateral institutions charged with the distribution of DAH or aid more generally. This set of stated criteria is a useful starting point for any discussion on the normative framework for DAH.

We found that many institutions did not have specific criteria publicly available, and this was especially the case for the bilateral institutions. Given the current emphasis on transparency and accountability and the substantial resources involved, increased use of explicit, detailed criteria is needed.

More generally, all institutions seemed to employ criteria related to need as well as effectiveness. However, the relative emphasis given to each of these criteria varied. Even more pronounced was the variation in the specification of the two criteria or, more generally, the specific criteria and indicators emphasized by the various institutions.

Amid the variation, at least one specific criterion was explicitly emphasized by nearly all institutions. This was the criterion related to gross national income per capita (GNIpc), and this was particularly central to the determination of eligibility. However, the GNIpc threshold value, above which countries are deemed ineligible for aid, varied considerably: from $1,175 to $12,616. In comparison, low-income countries (LICs) and high-income countries (HICs) are currently classified by the World Bank as having GNIpc ≤ $1,035 and ≥ $12,616, respectively.

As for what the institutions did not emphasize, there were at least two commonalities. Typically, institutions not specifically devoted to health did not have specific criteria for DAH and generally had few, if any, criteria directly related to health. Moreover, none of the reviewed institutions emphasized criteria directly related to economic inequalities or inequalities in health or health care.

Distributional implications of different criteria

We further looked at which countries should receive development assistance for health, based on specific criteria. For any discussion of and search for appropriate allocation criteria, it is crucial to understand how the criteria would potentially change the distribution of DAH. We therefore estimated, as a rough illustration, how DAH would be distributed according to each of 12 different allocation criteria. More specifically, we examined how each of the different criteria would redistribute the total amount of DAH currently available across countries and country categories.

We found, not surprisingly, that the estimated distribution of DAH across countries and country categories varied substantially depending on which criteria were applied. Compared to the current distribution, most criteria shifted DAH towards lower-middle-income countries (LMICs) and the top 10% of most populous countries, which also included many upper-middleincome countries (UMICs). More specifically, criteria related to absolute health needs (underfive mortality rate, years of life lost and burden of disease) and health inequality (inequalityadjusted life expectancy) advantaged LMICs and disadvantaged LICs compared to the current distribution. Moreover, when the Gini index for income was used as a criterion of need, huge amounts of DAH shifted towards UMICs, compared to the current distribution as well as to a GNIpc baseline distribution. These findings underscore, among other things, how the middleincome countries (MICs) challenge the normative framework for DAH. In particular, different criteria deal very differently with the MICs and the choice of criteria can have tremendous impact on the amount of DAH going to these countries as opposed to LICs.

Potential contribution norms

We also looked at the other side of the coin, i.e., which countries should contribute to development assistance for health, and how much they should contribute. In that regard, we describe two widely recognized contribution norms relevant for DAH: the 0.7% ODA/GNI target and the scale of assessments for the apportionment of the expenses of the United Nations. We further note that health ODA and DAH have recently represented 12% and 19% of total ODA, respectively. Against that background, we examine a norm according to which countries should provide DAH equivalent to at least 0.1% of their GNI. In 2010, only four OECD-DAC members met this 0.1% DAH/GNI target. If, instead, all of today’s HICs had met this target in 2010, the total amount of DAH available would be at minimum $43 billion. This would have constituted more than a 50% increase in DAH, compared to the $28 billion actually available in 2010.

Required contributions among donors should plausibly also depend on the intensity of recipient need. We demonstrate different ways to estimate total need for DAH. For example, taking need to be represented by the gap between a $86 target of government health expenditure per capita (GHEpc) and current GHEpc, total need for DAH was estimated at $144 billion. Considering instead the shortfall from $86 if GHE in every country represented 5% of GNI, total need for DAH was estimated at $79 billion. These figures can be contrasted with the $28 billion of DAH that was actually available in 2010 and the $43 billion that would be available if all HICs met the 0.1% DAH/GNI target described. When a range of other metrics of need was also considered, estimated total need for DAH varied from $79 billion to $436 billion. This indicates how total need may vary with the choice of metric as well as whether LMICs and UMICs are included among the potential recipients of DAH.

Another crucial choice in the development of contribution norms is the donor inclusion threshold, i.e. that above which a country should become a donor. Assuming full compliance with the 0.1% DAH/GNI norm, we show how minimum total DAH available would vary from $40 billion to $58 billion depending on the threshold and to what extent MICs are included as donors. Moreover, we show how certain thresholds would imply that many required donors are among the present DAH recipients. For example, if all HICs and UMICs were obliged to contribute, 50 out of 124 required donor countries would have been actual DAH recipients in 2010.


The system of development assistance for health is challenged on several fronts, and wellfounded allocation criteria and contribution norms are more important than ever. This paper provides three kinds of input to the assessment and improvement of the normative framework for DAH: the criteria emphasized by major distributors of DAH, the distributional implications of potential criteria and the implications of different contribution norms.

The wide variation in criteria emphasized by different institutions and the wide variation in implications from the different criteria and norms underscore the importance of more critical reflection on the normative framework for DAH. In particular, clarifying the role that MICs should play in that framework is crucial, not only for those countries, but for the entire DAH system.