Growth in low- and middle-income countries (LMICs), combined with declines in aid spending, means the vast majority of development spending, except in the poorest countries, comes from governments and individuals in the countries themselves. This has been true for some time, but recent aid cuts underline the change and raise new questions for bilateral donors about how to have the most impact in LMICs with fewer aid dollars. Increasingly, these donors are recognizing that the most impact comes from helping to improve the effectiveness of LMIC government spending and systems. For example, rather than building and financing specific aid-funded schools, it is much more impactful to improve the teaching methods used in all government schools.
Aid funding for systems strengthening is not new. A concern that aid-funded vertical health programs that cover specific diseases or approaches set up systems parallel to those established by LMIC governments led to an increasing share of vertical funds spending going to “systems strengthening.” For example, the Global Fund projected to spend $6 billion, or 38 percent, of its funding on systems strengthening in its 2024-26 cycle.
Published strategies and private conversations with bilateral donors suggest systems strengthening is becoming a much bigger focus. In the UK, for example, one of four key strategic shifts is from “service delivery to systems support.” Yet while the trend appears to be growing, there is no agreement on what system strengthening is or how to do it well.
What is systems strengthening?
There is no consensus on what systems strengthening involves. The Global Fund includes providing medical oxygen, funding health workers, building laboratories, supporting community health worker programs, and setting up digital monitoring systems. Some of this, such as providing oxygen or paying for health workers, is direct service provision, and one on the list (community health workers) is highly evidenced and cost-effective.
How do donors seek to “strengthen systems”? Often, through a combination of external technical assistance (TA) and funding for equipment. A survey of over 900 civil servants across Africa by my colleague Biniam Bedasso found considerable skepticism among those in the system of the effectiveness of TA, and a concern that it often builds parallel systems through, for example, foreign-staffed implementation units, which undermine the normal processes of government. My colleague Lee Crawfurd reviews the evidence on TA and recommends long-term embedded advisors. But is TA even the best way to provide systems strengthening, and if it is given, what should it be advising on?
What should (evidence-based) systems strengthening be?
Stepping back, the best “system” is one that spends its limited resources in a way that achieves the most impact. The best way to do that is to spend in line with the evidence on what is most cost-effective, given a country’s needs and constraints. To strengthen a system, therefore, is to help it move to be more in line with the evidence on what is cost-effective and scalable. This can be done in three ways:
Supporting prioritization. Support analysis, ideally by a country itself, to help determine the most impactful way to spend its own (and other donor) resources. For example, in health, this could involve combining data on the country’s burden of disease with evidence on the cost-effectiveness of different approaches to addressing those burdens, and prioritizing health spending accordingly. Some countries can do the analysis on their own but welcome funding for more data to help them optimize more effectively; others might welcome an embedded advisor. Ideally, this prioritization is not just for governments’ own funding, but donors and multilateral agencies agree to be part of the process and follow the prioritization too. When this approach was tried in Ethiopia, we estimated it increased the combined impact of government and donor health spend by 15 percent.
Cost-effectiveness modeling can also help countries prioritize infrastructure spending—for example, rural electrification versus more reliable electricity to firms, where the latter almost always wins out. This can and should be done for all sectors, from education to social spending. LMIC governments clearly need to lead this process, but donors can fund data collection on needs, embedded experts on the evidence, or analysts for the budget optimization process. Supporting specific parts of a “system” without this prioritization can make things worse: if a donor funds a laboratory or a road that is not a priority, the government may end up paying maintenance costs with funding that could have been used more effectively elsewhere.
Prioritization has long been a focus of TA and implementation units, but they have too rarely been grounded in detailed knowledge of global evidence on cost-effectiveness. Recent global synthesis of evidence and cost-effectiveness, for example, on health, education, and social protection, makes the process of tailoring this to a country’s needs much easier. CGD has experience providing this type of support in health and has found that countries can secure a return on investment of 9:1 through improvements in health system efficiency.
- Supporting the transition from less to more effective policies. Once an evidence-based prioritization process suggests there should be a change in policy or practice, “systems strengthening” programs can help cover transition costs, for example, retraining teachers in the new (more effective) pedagogy or revising textbooks in line with that pedagogy. In education, training and materials on their own have been found to be ineffective, but most effective programs include training and materials to support the new teaching system. Similarly, setting up data systems before deciding what is the most cost-effective approach may well be a waste of money, but setting up appropriate data is always part of the transition to a new, highly cost-effective program or approach.
- Generating evidence about which systems are most cost-effective. Systems strengthening and prioritization rely on a body of evidence about the impact and cost of different systems. This ranges from whether to reward tax collectors, health workers, and teachers based on performance, to whether to charge for preventative medicine. This evidence is a global public good, and donors have been the main funders. Evidence generation involves working on the details of how systems work. For example, the UK’s FCDO-funded International Growth Centre research on property tax reform involved detailed work on how to value properties (and how to get tax authority buy-in), which led to reforms across the country and a big increase in taxes collected. Evidence generation on social protection programs has yielded many lessons about the precise design of government systems, such as the best way to target the poor, and dramatically improved the effectiveness of LMIC spending in the sector.
Conclusion
Done well, a shift by donors to fund “systems strengthening” rather than direct provision of services could dramatically improve the positive impact of aid. But to generate these benefits, it must be focused on helping each country do what evidence has found is cost-effective for the challenges specific to them. This means systems strengthening should involve evidence-based prioritization of country-level resources, based on local needs and cost-effectiveness; supporting the transition to more cost-effective policies; and generating evidence which often directly leads to systems strengthening.
