When Aid Falls, How Can Philanthropy Rise?

vak427 Tue, 06/30/2026 - 15:24 June 30, 2026 International aid is falling just as global needs are growing. In 2025, aid from OECD donors fell by more than $40 billion, about 23 percent, the steepest annual decline on record, with the United States cutting its aid by nearly 57 percent . The human consequences are stark: the shutdown of USAID has been linked to hundreds of thousands of deaths from infectious disease and malnutrition worldwide. For philanthropy, this is both a moment of pressure and of possibility. Philanthropic actors are stepping in to fill some of this gap and rethinking how to use their limited resources for the greatest impact. At the same time, a growing share of development spending now comes from developing countries themselves, signaling a shift of power to local actors and new opportunities to strengthen local institutions and leaders. In this moment of global change, the Harvard Center for International Development (CID) used its 2026 Global Empowerment Meeting (GEM) as a platform to spotlight the pivotal role of philanthropy. As part of this year’s programming, CID convened GEM Foundations, a roundtable on the margins of GEM, bringing together a diverse group of philanthropic actors to grapple with one central question: "How can philanthropy deliver outsized social returns and maximize social impact per dollar spent?" The discussion began with participants emphasizing that philanthropy can fund what markets and governments often cannot, including high‑risk experiments, long‑term bets, and early‑stage ideas. Beyond funding, it can also use its soft power to shape agendas, legitimize bold approaches, and convene the networks, policies, and partnerships needed to take promising solutions from pilot to scale. However, realizing this potential depends on addressing several persistent questions and debates with no easy answers. What Does ‘Scale’ Mean and For Whom? Participants pushed hard on what "scale" actually means, and whether every initiative should even aim for it. They distinguished between "scaling what works" adapting and diffusing effective approaches through existing systems and "working at scale," where an organization directly operates large programs or services. Several warned against premium pilots: high-touch, high-cost models that look impressive in a controlled setting but are structurally impossible for governments or communities to adopt. Instead, they argued designing minimum viable projects from the outset; lean versions that preserve the core value for end users and can realistically be sustained by public systems or local actors. This led to a discussion about roles and tradeoffs. Not all funders need to be scalers: some are best placed to back early-stage innovation, others to underwrite rigorous testing, and others to help embed proven models into government or market systems. The real requirement is clarity and alignment. Foundations were encouraged to be explicit about whether they seek to scale at all, what kind of scale they mean, and who is ultimately meant to take over. Claiming something is scalable is not the same as actually scaling it; success depends not just on the model, but on whether the right capabilities, incentives, and institutions are in place to carry it forward. Asim I. Khwaja presents solutions during CID’s GEM Foundations roundtable. How Should Philanthropy Decide and What Should It Measure? Investment logic and outcome measurement surfaced as key areas for refinement. The discussion kept circling back to the balance between trust-based giving, which relies on strong relationships with governments and local communities, and evidence-based giving, which leans on data and a clear theory of change. As a new generation steps into leadership at family foundations, many described a shift from ad hoc, heart-driven charity toward more structured, investment style decisions. Yet, they also stressed how hard it is to actually use the evidence. Many topics are understudied, and even strong randomized controlled trials rarely explain their design choices and context in enough detail to guide real world funding decisions or show others how to replicate a program. Foundations often end up improvising or trying to track down the original researcher. For many in the room, this is where collaboration with local and global researchers could add real value: capturing and sharing the practical "how to" details that do not fit into academic papers but are critical for institutional memory. There was broad agreement that clearer theories of change and stronger measurement strategies would make philanthropic capital more effective, especially if learning is shared so others can build on it. At the same time, the group acknowledged, evaluation itself is a tension point: donors and implementers often track different things, funder metrics can skew behavior, and grantee data is uneven. It is also widely recognized that evaluations are expensive, but as one participant put it, "they are one of the most cost-effective investments you can make in the long run." Ultimately, the group agreed that trust and evidence both matter and that shifting toward more high-quality evidence, shared openly, can deepen trusted relationships. Local Actors, Partnerships, and System Responsibility Local actors such as implementers, universities, and researchers are closest to the problems, understand needs, hold legitimacy, and drive implementation. Yet they are still too often last in line for direct, flexible funding and timely, usable evidence. The discussion underscored that progress depends on strong partnerships of many kinds, for example between implementers and government and between local and global experts, with technology as just one enabler among many. Many noted that philanthropy too often runs in parallel to public systems instead of through them, partly because engaging ministries is slow, political, and resource intensive. At the same time, the government is the only actor that can take most interventions to a true scale. One foundation, for example, found that integrating community development was far harder to scale than early childhood programs, because public subsidies and infrastructure simply weren’t there. The group also recognized that outcomes were generated by a complex system and not any single organization. Scholarships for higher education, for instance, only work if students first get strong basic education, if someone helps them navigate choices at the end of secondary school, and if employers are willing to offer internships and hiring pathways. This can make achieving positive outcomes challenging, which participants framed this less as a failure of any one program and more as a call for better orchestration: being deliberate about which partners are needed along the chain. Taken together, these reflections sharpened a question that ran through the day: where does a foundation’s responsibility end, and where should the state and the wider system take over? The way forward The roundtable surfaced a practical agenda for what comes next. Participants saw room for CID to use its convening power, research base, and networks to help philanthropy act efficiently: exploring the feasibility of a few concrete tools, such as an ongoing forum for funders, practical evidence summaries and case studies, and matchmaking between philanthropies, researchers, and technical talent. GEM27 will pick up these threads and focus on testing concrete tools and partnerships that move from isolated projects toward systems level impact. When Aid Falls, How Can Philanthropy Rise? Image Credits Matt Teuten and Bharti Sharma Contact Emy Janssen and Bharti Sharma