Mind the Gaps: Takeaways from Emerging Research and Policy Implications for Aid Transition in Health and Development

Introduction

As countries grow economically, governments face rapidly growing demands for quality, affordable, accessible, and equitable healthcare and other social services. At the same time, many middle-income countries face the prospect of transitioning away from donor aid, adding pressure to already-constrained public budgets to fill gaps as donor support ramps down. As a result, governments must contend with difficult trade-offs and make tough decisions about whether to continue funding the same interventions. An emerging body of research describes the historical effects of aid declines on the amount and composition of public spending and development outcomes in partner countries. One particular question researchers seek to understand is how aid transition may affect the sustained delivery of essential public health services historically financed—or co-financed—by aid. Another understudied dimension is whether current policy approaches to transition are adequately addressing potential adverse effects.

Earlier this year, the Center for Global Development (CGD) convened a workshop with leading experts from multilateral and bilateral institutions, government departments and agencies, research organizations, and NGOs to discuss findings from recent research on upcoming aid transitions in middle-income countries. The group also considered the policy implications for sustained heath and development impact, public financing, fiscal management, and aid eligibility policies and their implementation. This workshop—the first in a series of meetings—closely relates to CGD’s ongoing research exploring transitions in global health financing.

Key Takeaways from Recent and Ongoing Research

Taken together, the research presented at the workshop shed light on what we know and don’t know about aid transitions in health and development. While the work shared is subject to many caveats and limitations regarding research design and available data, most studies suggest total spending in social sectors, including health, has been adversely affected by aid declines. Below, we summarize seven key takeaways that emerged from the workshop discussions.

1. Countries see a sizeable drop in aid as they grow wealthier, and there may not be a compensatory effect in borrowing or government spending. 

The shift in a country’s lending status from International Development Association (IDA) to International Bank for Reconstruction and Development (IBRD) is correlated with an overall decline in World Bank lending.^[1] Forthcoming work by CGD colleagues also suggests crossing the IDA income threshold (as defined in GNI terms, versus a change in a country’s lending status) is associated with sizeable drops in official development assistance (ODA)—equivalent to 4 percentage points of GNI, on average.^[2] The decline is primarily from sources other than IDA, as many funding mechanisms tend to observe the IDA threshold more strictly. Yet, preliminary results suggest there is no apparent increase in non-aid international borrowing nor an effect of the decline in aid on the overall level or composition of government spending (i.e., it neither declines nor increases to compensate for aid loss). ^[3] However, the authors caution that the data are sparse, and findings should be considered illustrative at this stage.

2. There are disproportionate effects on social sectors, including health.

As countries move to IBRD status, where lending terms are “harder” than those of IDA, there is a disproportionate decline in the share of social sector lending.^[4] Relatedly, forthcoming analysis suggests roughly 50-75 percent of the total drop in ODA when countries cross the IDA threshold is driven by a decline in social sector ODA.^[5] This can, in part, be explained by the observed tendency for transitioning countries to favor investments in infrastructure development and other non-social sectors.^[6] These trends raise the concern that transitioning countries could face a sudden drop or “fiscal cliff” in social sector spending, which may threaten hard-earned progress if domestic priorities and resources are not sufficiently directed towards closing the funding gap. Indeed, actual behavior tends to reflect a common belief among many country policymakers that borrowing for social sectors on concessional terms is a bad idea. Varying time horizons for returns, limited visibility into spending, and perceptions of inefficiency could all affect country government decisions to borrow for different sectors. Some experts refute this dichotomy between sectors, arguing that decisions to borrow for social sectors should be driven by the same rationale as those for “hard sectors” like infrastructure or industry.^[7]

3. The upcoming cohort of transitioning countries is less prepared to manage donor transition than previous ones.

The upcoming cohort of transitioning countries is experiencing a much broader set of transitions, collectively known as the 4Ds: demographic shifts, changing disease burdens from infectious to noncommunicable diseases (NCDs), transition from development assistance for health, and evolutions in domestic health financing.^[8] They are also more disadvantaged compared to previous cohorts; on average, these countries have higher baseline levels of inequality, worse overall health outcomes, and a larger proportion of total health spend that comes from donor aid.^[9] For immunization in particular, fiscal space may be a greater issue for the upcoming cohort of transitioning countries that face higher vaccine co-financing requirements amidst competing priorities and slower economic growth projections, as well as challenges related to inequities in immunization coverage.^[10]

4. For some countries, the loss of health aid (sometimes from unexpected sources) could have significant fiscal impact—with implications for key and vulnerable population groups.

Several global health mechanisms have developed transition plans and policies that outline—with varying degrees of formality and explicitness—the eligibility criteria and timelines to gradually scale-up co-financing requirements and simultaneously drawdown support. For example, many middle-income countries, including top recipients like Nigeria, India, and Pakistan, are projected to transition from Gavi support over the next 10–20 years.^[11] None of the Global Fund’s largest or most aid dependent recipients are projected to transition by 2040, given its relatively inclusive eligibility criteria. Nevertheless, even if countries remain eligible, co-financing requirements are increasing rapidly. On the other hand, the Global Polio Eradication Initiative (GPEI) and PEPFAR lack explicit transition frameworks or eligibility criteria, but are still expected to phase out support in several recipient countries over the next several years.^[12]

According to CGD research, the fiscal impact of these anticipated global health transitions could be significant for some countries—assuming existing levels of aid spending would need to be borne by domestic sources to sustain current levels of service provision. Eleven countries are at highest fiscal risk from global health transition, where cumulative lost funding through 2040 would amount to at least 10 percent of total government expenditure on health.^[13] Overall, countries at greatest fiscal jeopardy are those facing PEPFAR non-acceleration and/or GPEI withdrawal (five countries are predicted to face both simultaneously), which represent large sources of funding that are typically off-budget.^[14]

External support, although a small share of overall health sector spending in some instances, may represent a significant source of financing for specific disease or program areas—and particularly for key and vulnerable populations. For example, more than half of people with HIV live in countries where development assistance for health (DAH) accounts for more than 50 percent of care and treatment.^[15] In many countries, additional government spending for HIV programs will not be able to adequately fill the gaps left by declining DAH.^[16] Heavy reliance on donor assistance could result in access gaps, with important equity implications. Further, financing of health products represents a large share of external funding—40 percent of the Global Fund’s grant disbursements is spent on procurement of health products, for example.^[17] This reliance could have large fiscal impacts as countries begin to self-finance the purchase of health products; there may also be implications for quality-assurance as procurement functions gradually shift from the global to the national level, as CGD research has highlighted.^[18]

5. Most development partners do not have clear plans to manage and coordinate transition; and the transition outlook is poorly understood by country governments.

For the most part, donor approaches to transition are currently fragmented. Many donors approach transition in the context of their bilateral relationship with a country rather than collaborating with other development partners.^[19] The lack of coordination can be particularly challenging when a country faces drawdown from multiple financing mechanisms at the same time. As part of planning for transition, donors are increasing investments in health systems strengthening (HSS), but the size of this support remains relatively small, it is implemented in a fragmented or piecemeal way, and the overall results are unclear. The recently launched Global Action Plan’s Financing Accelerator aims to enhance coordination among multilateral global health and development organizations at the country level.^[20] Yet, the extent to which it will enable more coordinated financing, especially around HSS, remains unknown.

At the country level, policymakers lack complete information about the transition outlook, its implications, and effective strategies to manage it. Transition planning is oftentimes limited and collaboration between different governmental units—notably ministries of finance and health—can be ad hoc and disjointed. One notable challenge is the lack of expenditure and costing data. Such information enables governments to understand what will need to be funded in the absence of donor aid and assess cost-effectiveness, which is critical to transition planning. Improved forecasting of expenditure requirements could also help donors understand country-level capacity and commitment to financing specific health programs. Another obstacle is limited coordination for cross-cutting functions, such as public financial management (PFM), procurement and supply chains, and health information systems, among others.^[21] Broadly, not enough is being done to adequately plan for the capacity and resource needs for these critical areas.

6. The overall evidence base is thin; the limited (observational) evidence that is available shows mixed—but mostly negative—effects of aid transition on country responses, health systems, and outcomes.

There is limited evidence about the impact of aid transition on program implementation. A recent analysis found that early and extensive planning is needed to mobilize financing to cover costs previously funded by donors—especially for programs serving key populations subjected to stigma and discrimination.^[22] Further, looking at spending on health products across a snapshot of countries, there is some evidence that out-of-pocket expenditures increase as countries move from low-income to lower-middle-income status; nevertheless, more in-depth analysis is needed to tease out a causal pathway and study long-term time trends.^[23] There is also very little research available in the published literature (or that is easily accessible in the public domain) exploring the impact of aid transition on health outcomes.^[24] Existing methodological approaches to assess the effects of aid transition are generally weak. Available observational research points to mixed but mostly negative outcomes—including negative impacts on the health workforce, access to essential medicines and vaccines, and outcomes for vulnerable groups.^[25]

7. Current uses—and the inefficient allocation—of aid could complicate allocative efficiency in the wake of transition.

Aid currently finances many of the best buys in healthcare, however not all aid is dedicated to the most cost-effective uses.^[26] Co-financing of aid-funded health technologies and interventions that do not represent best value-for-money according to a country’s own budget constraints may distort the allocative efficiency of public spending on health, in addition to diminishing health impact. However, less efficient uses of public expenditure on health may be “sticky” (i.e., hard to reallocate to more cost-effective interventions). While fiscal space increases alongside economic growth in most countries, this growth is generally modest.^[27] Considering these dynamics, many unanswered questions remain: What are the most essential aid-supported interventions that should be rolled over to public financing? What is their price tag? How will countries absorb the financing requirements for the most cost-effective, aid-financed services during and after transition? How can we evaluate existing uses of aid to assure essential services are protected during and after transition?

In addition, there is a lack of clarity on how to best approach allocative efficiency and the targeting of resources ahead of transition. For example, if donors disproportionately fund prevention, especially for poor and vulnerable populations, these services may be at risk of collapsing with the drawdown of donor support. The case of Romania is a cautionary tale: the departure of the Global Fund resulted in a significant financing gap for HIV prevention; just 1 percent of domestic HIV expenditure is dedicated to prevention services.^[28] One approach would be for local advocacy organizations to engage the government earlier and more effectively to ensure the scale up of spending for prevention.^[29]

However, some research findings suggest slightly nuanced takeaways regarding current policy approaches to managing transition:

Ongoing research suggests that coverage of certain key interventions may be unaffected when countries cross the IDA threshold, despite drops in total spending; preliminary analysis shows no clear sign of the aid drop on selected health (vaccination) and education (enrollment) outcomes, though these findings must be caveated due to sparse data.^[30]

Further, there is some evidence of fungibility. In the case of Gavi, a recent study found that while the probability of a country receiving Gavi assistance declined after crossing the GNI threshold, there was no significant impact on vaccination coverage (e.g., Hepatitis B, DPT) among middle-income countries near the eligibility threshold. Yet, it should be noted that these findings are specific to vaccines and, since these effects were observed among countries near Gavi’s eligibility threshold, the findings might not be as generalizable or relevant to other (poorer) countries.^[31]

Fungibility is also closely related to fiscal space, and there are some indications of additional fiscal space for health. Economic growth is a key driver for increasing per capita public financing for health in low- and middle-income countries—although growth tends to be modest.^[32] Modeling exercises point to the potential for additional government spending to be mobilized to help make up for the decline in DAH. One study suggests additional resources could be mobilized for HIV, although low-income countries have very low potential and a majority of the additional resources would be concentrated in 10 middle-income countries (e.g., Nigeria, Argentina, South Africa, etc.).^[33] Yet, it remains unclear whether findings from these modeling exercises translate into practice. For instance, research in Kenya, South Africa, and Nigeria shows that increases in tax revenues may not be accompanied by an increase in the relative share of the government budget allocated to health; factors driving this are manifold and include intersectoral competition in priority setting and low perception of the health sector’s absorptive capacity, among others.^[34]

Initial Reflections on Policy Implications

The research presented at the workshop pointed to several policy implications. The ideas presented below are still nascent and deserve further discussion, elaboration, and vetting. Yet, we hope these broad considerations will be top of mind among the global health—and broader development—community going forward.

• Donors should evaluate the cost-effectiveness of the interventions they currently fund to inform what to transition to domestic spending. By prospectively evaluating cost-effectiveness before transitions take place, donors can encourage the gradual or phased adoption of the most cost-effective interventions. And by basing resource allocation on cost-effectiveness, donors can identify what exactly they are transitioning and what needs to be sustained by country governments. For example, such analyses could consider whether the health products purchased by donors are the best value for money, or most efficiently procured.^[35] Some experts have suggested ways for how the Global Fund can think about transitioning cost-effective services to governments, for instance.^[36] Amidst growing momentum around Universal Health Coverage (UHC), donors are well-positioned to engage with country policymakers on the role of cost-effectiveness in informing benefit packages, along with considerations of equity to increase coverage.^[37] The International Decision Support Initiative (iDSI), a global network now housed at CGD that engages with policymakers on how to efficiently allocate healthcare budgets, could serve as a potential platform for further engagement and collaboration on these issues.

• Focusing on efficiency in health spending should be a core strategy to increase domestic financing. Since the overall efficiency of health spending directly impacts the availability of domestic financing, budget asks are more compelling when accompanied by evidence on the cost-effectiveness of health programs. Nevertheless, although donors might expect domestic financing to cover unmet needs, the tax-to-GDP ratio is still low in many countries and could be slow to progress for many years.^[38] Some experts also propose soft earmarking public revenues for the health sector, potentially through taxes on tobacco, alcohol, and sugary beverages, as a way to partially address the drop in social sector spending that countries experience after transition.^[39]

Reform for cross-cutting functions—including procurement and supply chains, PFM, and information systems—can also drive efficiencies across health systems. Addressing health system bottlenecks and institutional constraints helps stretch scarce resources further and ensures both financial and programmatic sustainability after transition. For example, as countries transition from donor aid, procurement of health products will be increasingly financed and managed by countries. Accordingly, countries—with dedicated support from development partners—will need to undertake reforms to address barriers to inefficient procurement policies.^[40] Doing so will be critical to advancing UHC.

• Transitions would benefit from early, coordinated engagement and planning with ministries of finance as a central counterpart alongside ministries of health. As part of this engagement, timelines and budget asks to country governments could be streamlined and clearly articulated. These could be negotiated as a package as part of medium-term expenditure frameworks, for example. Further, to the extent possible, it will be important to connect vertical programs with all-sector health financing reforms and arrangements. Nevertheless, the incentives that drive resource allocation decisions at the national and subnational levels differ from those of donors and involve complex political economy considerations that should be acknowledged in engaging with country policymakers.

• Enhanced collaboration and communication between development partners is also critical to facilitate smoother transitions. Partnership in this area helps to avoid risks associated with the drawdown of donor support, specifically where a country might experience simultaneous transition from several donors. Relatedly, development partners can also structure phaseouts to include transitional support that is flexible and adapted to each country’s context. For example, Bolivia, Sri Lanka, and Vietnam graduated from IDA at the end of FY2017 and are receiving transitional support for the FY2018–2020 period.^[41] And even after traditional financial support ends, development partners should consider continuing technical support in key health system areas; procurement is just one such area, as highlighted by CGD’s Working Group on the Future of Global Health Procurement.^[42]

• To facilitate shared learning, development partners, country governments, research organizations, and civil society must advance research collaborations and learning opportunities by connecting work streams related to transition, resource mobilization, and fiscal space. This would help fill critical outstanding knowledge gaps—such as the need for analysis on the fiscal impact of the decline in donor aid beyond the health-focused multilateral institutions—and inform policy approaches to transition going forward. The vaccine space provides potential models, including the Learning Network for Countries in Transition—a community of practice focused on Gavi transitions—and Immunization Economics—a dissemination platform and forum to exchange relevant research methods and approaches on issues related to vaccine economics.^[43] Looking ahead, advancing shared learning on these issues will be important—but it will be even more critical to push forward holistic efforts across program areas and technologies.

Looking ahead, there is a need to better understand the extent to which the priorities of governments and donors align; the support and types of coordination transitioning countries desire; and how development partners can best work with countries before, during, and after transition. Stay tuned for more as CGD continues to conduct research and policy analyses in this space, including a workstream on how countries can manage the convergence of vertical health programs and UHC plans. In the meantime, we welcome your feedback and suggestions.

The authors thank all workshop participants for their engagement and thoughtful comments. The content of this note is based on the discussions and research presented by workshop participants, however participants do not necessarily endorse all components. All errors and omissions are those of the authors.

See here for a full list of participants.