Nicholas Institute for Environmental Policy Solutions
Proceedings
From Fragmentation to Coordinated Action: Building a US Resilience Finance System
Contents
The Core Problem: The System Isn't yet Set up for Resilience Investment
Why the Current System Falls Short
Two Systems Side-by-Side: What the Panels Revealed
Three Building Blocks of a Coordinated US Resilience Finance System
What the Room Designed: Five Models and Three Open Questions
Why This Moment May Be Actionable
Acknowledgments and More Information
Authors and Affiliations
Citation
How can we ensure resilience funding consistently reaches the communities and projects that need it? Explore the challenges and potential solutions.
The Core Problem: The System Isn't Yet Set up For Resilience Investment
Communities across the country are losing ground to floods, fires, heat, and storms. Insufficient resilience investment is often framed as a simple shortage of money; however, there is existing capital across federal agencies, state programs, pension funds, philanthropy, community lenders, and private infrastructure investors. Better coordination is needed, not just access to funding.
On April 20, 2026, Duke University and the Milken Institute brought together more than 40 senior practitioners—from green banks, infrastructure investment, commercial finance, think tanks, academia, housing and community development finance, local and state government, climate adaptation and resilience organizations, and international development—to examine what financial infrastructure is needed so that capital consistently reaches the communities and projects that need it.
Participants at the DC Climate Week convening unpacked why the resilience finance system is not working as well as it should and what a more coordinated approach could look like. A clear theme emerged: this is a coordination problem. Many of the organizations essential to improving the existing system were represented in the room.
Why the Current System Falls Short
Four structural barriers emerged as the core reasons capital is not reaching resilience projects at the scale needed:
- The difficulty of valuing resilience benefits and asking the right investment questions
- The absence of a funded pre-development function
- Projects that are too small and fragmented to attract large investors, requiring either regional bundling into coherent, locally grounded solutions or standardization across regions so similar projects can be pooled and evaluated at scale
- A governance gap that leaves no coordinating body to organize projects, manage early-stage risk, sequence capital, and connect public and private financing programs
Taken together, these point to a coordination problem more than a capital shortage. The money exists, but the architecture to connect it to projects does not.
“The ounce of prevention worth the pound of cure is very difficult to generate revenue from.”
—US resilience leader and academic
Two Systems Side-by-Side: What the Panels Revealed
Practitioners from the World Bank, IDB Invest, the Global Facility for Disaster Reduction and Recovery, and World Wildlife Fund described a consistent international pattern: resilience finance remains in the tens of billions annually, while published estimates place adaptation needs in developing countries in the hundreds of billions per year, rising as high as $500 billion annually by 2050 (Tall et al. 2021; OECD 2026).
Their clearest lesson was that the limiting factor is not the number of financial instruments, but rather the ability to aggregate enough projects with clear value and payback mechanisms into a portfolio that large investors can act on.
Two observations shared by the international development practitioners in the workshop illustrate the scale challenge:
- The Climate Investment Funds' (CIF) Forest Investment Program (FIP) committed $71.6 million in concessional finance for Cerrado conservation over more than a decade through the Brazil Investment Plan. The World Bank and the Inter-American Development Bank (IDB) implemented the financing at CIF’s typical cofinancing ratio of roughly 1:10 (CIF 2025).
- A separate $70 million CIF plan for clean energy grid integration in Brazil pro-jected mobilizing $9.1 billion from partners—a ratio of 1:130—demonstrating that where value is clearly defined and payback mechanisms are legible to investors, even modest upfront preparation capital can catalyze investment at a scale that no single institution can achieve alone (CIF 2023).
The United States has no equivalent project-preparation and aggregation function operating at scale. While the World Bank can finance smaller loans, its sovereign lending model generally operates at a much larger transaction scale: in FY2025, the International Bank for Reconstruction and Development averaged roughly $294 million per operation, making a standalone $10 million local resilience need difficult to reach efficiently without aggregation, intermediation, or dedicated project-preparation support (World Bank 2025).
The Global Facility for Disaster Reduction and Recovery was specifically established to simultaneously build capacity on both sides of this gap: helping countries develop project proposals and helping institutions develop the ability to evaluate them and create investable portfolios of projects. The United States also has no equivalent.
This challenge is not unique to the United States. Around the world, resilience projects often struggle to attract funding because they are not organized into clear, ready-to-finance packages. Money tends to go to projects that are already well developed, with clear costs, benefits, and repayment plans. Right now, the United States does not have a strong enough system to prepare and group resilience projects in that way. The panel discussions showed that the US system remains fragmented, while some international approaches—though far from perfect—offer useful models for how to better organize projects and connect them to funding.
Table 1. Comparing the Portfolio Bottleneck Approaches of the United States versus International Resilience Finance
The US System | The International System |
|---|---|
The Core Difference | |
| A collection of resilience finance programs that were not designed to work together, run by agencies that often coordinate unevenly, with financial risk sitting on the people least equipped to carry it. | A more deliberately engineered set of institutions that can absorb early resilience finance risk and draw private capital behind it—still reaching only a small share of what is needed, but with clearer design and sequencing. |
Reimbursement-Based, Not Investment-Based | Uses Cascade Logic: Public Money Goes First to Make Private Money Possible |
State Revolving Funds—the primary federal vehicle for water infrastructure—are loan programs that require costs to be incurred before funds are released, creating a significant cash-flow burden for anyone working on a resilience project without deep working capital. Access to planning and design funding varies widely by state, and many offer nothing before construction. Where it does exist, funding requires being on a competitive annual priority list first. Further, the rates are so low (often 0%) that private investors cannot compete, and therefore do not try. One participant described running a $10 million-plus working capital line off their own balance sheet just to survive the reimbursement lag. These factors mean the system depends on individual workarounds rather than predictable financing pathways. | The World Bank and other multilateral development banks take what one participant described as the “first dollar of loss”—they absorb the earliest and highest risk so that private capital can follow behind with confidence. That sequencing is intentional and governed by agreed-upon rules. In the United States, by contrast, the sequencing is accidental: whoever shows up first gets the risk. |
Siloed Programs Make Coordination Difficult | Blended Finance with Explicit Governance Guardrails |
The US Environmental Protection Agency manages water, the Federal Emergency Management Agency runs disaster risk, and the Department of Housing and Urban Development contends with community development. Each has its own criteria, timelines, and funding cycles. States have similar silos. A resilience project that touches multiple categories—like green infrastructure that also addresses flooding and serves a low-income neighborhood—must navigate all of these agencies separately, and coordination is uneven. The Chesapeake Bay was lifted up as an illustrative example: multiple state agencies were all funding riparian buffers with different goals, different metrics, and no unified outcome in mind. | IDB Invest’s model—mixing public, philanthropic, and private money—operates under established rules for when and how subsidized capital gets deployed, managed by an independent body with clear principles. The IDB participant described a public finance continuum: permanent subsidy on one end, fully market-investable on the other, with catalytic capital bridging the middle. International systems face their own coordination challenges; fragmentation across multilateral institutions is a significant constraint. However, the US middle zone is navigated informally, inconsistently, and without agreed rules. Some international models have at least made explicit governance a named goal. |
Two Critical Investment Signals Are Weakening or Absent in The United States: Insurability and Community Acceptance | Community Acceptance: An International Lesson The United States Has Not yet Applied |
In markets where resilience investment works, two things tend to be true. First, investors can get a reliable read on whether a project is insurable, which tells them something real about physical risk. Second, they can assess community acceptance early enough to price it into their returns. Neither is functioning well in the United States today. Insurance, which used to be a stable, predictable cost over the life of an asset, is no longer a reliable signal. Insurers are exiting markets, repricing annually, and in some cases refusing to cover assets at all. When insurance becomes unpredictable, one of the clearest indicators of whether a project is financeable disappears with it. Community acceptance has the same problem from the other direction. The United States has formal public engagement requirements, but they are process requirements—they do not translate into a financial signal investors can use. | International development finance has learned through hard experience to treat community resistance as a financial risk, not a values question. Of 200 infrastructure projects examined in one study, 36 were canceled as a result of community resistance and 162 were significantly delayed. Community engagement is now treated as a core financial risk to be managed from the outset, priced into project timelines and returns from day one. |
The Key Implication for the United States | The Sobering Benchmark |
| The international experience shows that deliberate design—even imperfect design—can outperform a system where every resilience project starts from scratch, absorbs its own startup costs, and has no coordinated pathway to become part of a fundable portfolio. The question is whether the right organizations can raise the capital and commit to designing a coordinated system intentionally rather than waiting for it to emerge on its own. | Even with all the international machinery—multilateral development banks, dedicated climate facilities, and 20 years of effort—tracked adaptation finance flows reached $63 billion annually in 2021–22, all sources combined, against estimated needs of $215–387 billion per year in developing countries alone by 2030, with higher-end estimates reaching $500 billion annually by mid-century. International public finance from developed countries—the deliberately engineered portion of the system—reached roughly $32 billion in 2022, representing less than 10% of even the lower-bound need. This is not a reason for pessimism; it is a reason for urgency (Climate Policy Initiative 2023; Tall et al. 2021; UNEP 2023). The gap is enormous, but many of the tools to close it are already available. |
Three Building Blocks of a Coordinated US Resilience Finance System
During the convening, five design groups worked in parallel to sketch out solutions. Despite starting from different angles, the groups converged on three functional requirements. No single new institution is necessarily required, but these functions need to be deliberately built, assigned, and connected:
Table 2. Three Functional Requirements for US Resilience Finance Solutions
1 Get Projects Ready | Resilience projects need a funded, ongoing predevelopment function that moves them from a community’s idea to something an investor can evaluate. This function should cover early-stage risk assessment, technical assistance, preconstruction funding, risk valuation and financial planning, and grouping projects together. In practice, this means building regional capacity—offices serving counties and small towns that lack their own staff for this work—backed by a running inventory of available funding and a single entry point that routes projects to the right programs rather than requiring communities to navigate each one separately. |
2 Build Investable Project Portfolios | For resilience projects to attract large investors, they need to be bundled into pools large enough to evaluate and deploy capital against. There are two distinct paths to get there, and the system needs both. Applying consistent lending criteria across many states can make similar projects easier to compare, bundle, and sell to larger investors. Climate First Bank and OneEthos offer a clean-energy example: Climate First reports more than $555 million in residential and commercial solar financing across all 50 states, while OneEthos uses a fintech platform to help mission-driven lenders originate solar loans at scale. This reflects a broader lesson from clean energy finance: standardized underwriting, documentation, and performance data are essential for aggregating small projects and connecting them to secondary markets. Community banks and credit unions are well-suited to finance standardized resilience upgrades because they often provide lower-cost loans than private equity. That can make smaller projects more affordable without relying on tax credits. But lenders need a clear way to know what counts as a “resilient” building. Standards from the Insurance Institute for Business & Home Safety can help provide that benchmark. When a project meets a clear, verifiable standard that insurers recognize and lenders can underwrite, it becomes easier to finance one building at a time—and eventually bundle many similar projects into a larger investment pool An important corollary: if an asset cannot be insured, it cannot be financed. Building in resilience from the start so assets qualify for insurance could become a powerful organizing principle for this approach. The second pathway is systems bundling. Most resilience investment will not be standardizable—it will be local, complex, and cut across water, housing, energy, and transportation concerns simultaneously. The path to scale is therefore integration: for example, combining upstream wetlands, stormwater parks, elevated infrastructure, and flood-resilient development into a coherent package that addresses risk across an entire area. Reducing risk at a systems level increases development value and economic activity. That increased value is what pays for the investment. |
3 Align the Money | Public money, philanthropic grants, and private investment should be organized around clearer, more transparent criteria—what the field calls capital coordination—so that the projects coming out of the first two functions are ones investors can actually fund. This does not require building a new national financial institution. It requires investors with resilience mandates to publicly state what they need from a project, organizations that can translate between the language communities use and the language investors speak, and a shared data system that tracks how projects perform and what lenders require. The pay-for-results model piloted in the Chesapeake Bay shows one way capital alignment can work in practice. In this model, governments pay for outcomes rather than individual projects, private developers take on execution risk, and costs drop as the model scales. |
What the Room Designed: Five Models and Three Open Questions
During the design session, participants produced five distinct resilience finance models. They are best understood less as competing alternatives than as different entry points into the same system. What follows is a summary of each, and the three design questions that must be resolved before any of them moves from concept to institution.
Table 3. Five Resilience Finance Models
| Group | Model and Core Idea | Primary Orientation |
|---|---|---|
Group 1 Guided navigation + regional resilience offices | A concierge-style service routes communities to the right funders, backed by a live inventory of what projects exist. Regional offices serve counties and small towns that do not have their own staff to navigate the system. The goal: “air traffic control” between projects and money, knowing not just where funding exists but the specific strings, timelines, and constraints attached to each source. | State-led · Capacity-first · Strong focus on rural and under-resourced communities |
Group 2 Long-term strategy + linked components | Cities and states develop 10–15-year investment plans that address whole-system risk, not just a list of individual projects. Those plans feed into: (1) capacity-building for states that are earlier in building this function; (2) a shared pool of capital with clear, public criteria; and (3) financial intermediaries who translate between local needs and investor language. Designed to address systemic risk rather than one crisis at a time. An example from Maryland serves as a working model. | Strategy-first · Multi-actor · Long investment horizon |
Group 3 Single application + independent platform manager | A project applies once and gets routed to the most relevant funders—like a single permit application that addresses permit requirements across all programs. Run by an independent nonprofit with a government charter so it can accept both public and private money and survive political transitions. Key design principle: transparency over a single standard. Funders publicly state what they’re looking for, so projects can find the right match without needing an insider to navigate. | Government-chartered · Independent governance · Focus on transparency and routing |
Group 4 Privately run platform with AI built in | A privately led platform—so it survives election cycles—covers insurers, developers, lenders, and investors in one place. Year 1 focus: Reduce friction through standardization only. Years 1–3: Use AI to learn what deal structures succeed, turn those into templates, and build a replicable playbook. Government is a partner and funder, not the owner or operator. Explicitly designed around the reality that the four-year political cycle is incompatible with resilience project finance timelines. | Private-led · Technology-enabled · Explicitly insulated from political cycles |
Group 5 Shared lending standards + open performance data | Shared lending criteria let projects be consistently described, bundled, and sold to larger investors. Community banks and credit unions are the lowest-cost source of private money and do not need government incentives to make the math work. Open performance data is what makes projects underwritable at scale and breaks the “conspiracy of silence” where all parties have incentives to keep project outcomes private. Prince George’s County green infrastructure program serves as the proof of concept. | Market-based · Data transparency · Community banks as primary capital source |
Three Design Questions That Need Owners
The proposed resilience finance models point toward a common architecture, but they diverge on three important design questions. Until each question has an owner and a process for resolution, the ideas are unlikely to move from concept to implementation.
1. Who Runs Infrastructure Investment—and How Do You Keep It from Changing Every Four Years?
The Tension: Infrastructure investment needs a 10–15-year horizon, while four-year election cycles can disrupt continuity. But a purely private entity has no mandate to serve low-resource communities and no mechanism to align public funding programs. Three governance structures have the potential to mitigate this perennial problem:
- Independent Nonprofit with a Government Charter
- Can accept public and philanthropic money
- Flexible to operate
- Modeled on Congress-chartered conservation foundations like the National Fish and Wildlife Foundation
- Private-Sector-Led
- Insulated from election cycles
- Aligned with investor timelines
- Government stays as partner, not owner
- Faster moving but less accountable to public interest
- Investor-Led Coalition
- Pension funds and institutional investors set shared public criteria
- Governance follows where the money comes from
- Requires sufficient investor alignment to hold together
Suggested Next Steps: Compare three to four governance structures side-by-side. Look closely at IDB Invest’s model—independently run, jointly owned by countries that provide and receive funding—as a template for staying accountable without being politically controlled. FAST-Infra, a sustainable infrastructure platform run by Bloomberg and used actively by major infrastructure investors, offers another reference point for how a privately operated platform with clear standards can bridge public goals and private capital at scale.
2. Does the Approach Work State-By-State, or Does it Need National Standards—and Could Both Be True?
The Tension: State-level approaches are politically realistic and can use existing institutions. But all these potential different sets of funding rules and criteria are a major barrier to projects from being bundled into pools large enough to attract big investors.
- State-Led Investment Strategies
- Some states already have the institutional capacity to develop and execute long-term resilience investment strategies—Maryland is a working example. Building this capacity more broadly, using existing regional structures like councils of government rather than creating new institutions, is a realistic starting point. With 36 gubernatorial races on the ballot in 2026, incoming governors represent a practical near-term opening to prioritize this strategy early, before budgets and agendas are set.
- Nationally Consistent Lending Criteria
- Applying the same lending criteria across all 50 states—the approach Climate First Bank used for clean energy—would allow similar resilience projects to be bundled together and sold to larger investors through secondary markets, regardless of where they were built. Technology makes this feasible at scale without requiring new legislation in every state. The limitation is that this approach works best for project types that are replicable and measurable (e.g., standardized retrofits, specific construction standards) and is harder to apply to the complex, locally specific systems investments that make up much of resilience work.
- State Compacts
- A group of willing states could agree to adopt shared standards and build scale together without waiting for national consensus. The US Climate Alliance structure presents a model for how this works in practice.
Suggested Next Step: Identify five to eight states with the legal framework and institutional readiness to move first. Accomplish this by working through existing networks like the US Climate Alliance, the National Governors Association, and the National Conference of State Legislatures, which already have relationships across state governments and could help coordinate a multistate agreement on shared resilience lending criteria. Connect this work to upcoming work on state-level legislative playbooks.
3. What Gets Built First—and What Does a Win Look Like in Year One?
The Tension: The sequencing challenge is clear: the platform cannot be built without capital commitments, and capital commitments cannot be received without the platform. Sequencing will strongly shape whether this launches or stalls. Three different initial steps could promote action:
- Start with Shared Standards
- Agree on what a resilient project looks like
- Align insurers and lenders around that definition
Build the platform on top of criteria that already have buy-in
Note that shared standards work best for replicable project types. More complex, locally specific investments will need the systems bundling approach rather than standardization to reach investment scale.
- Start with a Project Inventory
- Map what resilience projects already exist
- Match them to available funding
- Show investors there is an existent pipeline to look at before asking for commitments
- Start with One Real Deal
- Close one transaction that others can copy, building from a proven example rather than a design document. Two live candidates emerged from the convening:
- A homebuilding resilience initiative in Los Angeles focused on bringing insurance-linked construction standards into new development
- A county-level resilience financing effort in Monterey County, CA, that is testing how local infrastructure investment can be structured to attract private capital
Suggested Next Step: Rather than waiting for a fully designed national or state resilience finance platform, the field should identify one transaction-ready opportunity and use it as a live test case. Participants identified the Monterey County, CA, resilience financing effort as one opportunity to test how local infrastructure investment can be structured to attract private capital. Pairing that effort with an insurance-linked underwriting approach would allow lenders, insurers, developers, and public partners to define what evidence is needed to show that a project reduces risk, remains insurable, and can support financing. The goal would be to learn from an actual transaction what standards, data, approvals, capital sources, and governance arrangements are needed to make similar resilience investments easier to finance in other places.
Why This Moment May Be Actionable
Two factors make the current moment unusually actionable for resilience finance.
First, the federal government has pulled back from some coordinating roles in infrastructure investment, but the work has not stopped.
Nonprofits, states, and community organizations are stepping in, absorbing projects and relationships that might otherwise stall. In clean energy, this shift has already produced institutional infrastructure—financing vehicles, lending standards, and project pipelines—that did not exist a decade ago.
Resilience is at an earlier stage. The field is still building the foundational systems—common standards, coordinated financing, investable pipelines—that clean energy took years to develop. The question is whether the organizations that were present at the convening can accelerate that process, and whether resilience finance practitioners can capitalize on the momentum before it dissipates.
Second, with 36 gubernatorial races on the ballot in 2026, there is a specific near-term opening: the chance to define changes to state law that could unlock significantly more resilience investment. A practical guide for governors does not currently exist. Building it is achievable, timely, and could give incoming governors a concrete set of actions to consider in their first year.
“Don’t over-engineer a single solution. Keep the channel going. Do not wait for perfect conditions. We’ve got almost all the pieces in the room already today.”
—Milken Institute participant
What Comes Next
The convening made clear that the pieces of a more coordinated US resilience finance system already exist, but they are not yet connected in a way that consistently moves projects from local need to investable opportunity. The next phase of work is therefore less about creating a single new institution and more about building the connective tissue the field is missing.
Several near-term priorities emerged:
- The field needs a clearer way to identify and prepare resilience projects before they reach the financing stage. This includes earlier support for project definition, risk analysis, predevelopment, benefit valuation, and capital planning, especially for communities and local governments that do not have the staff or resources to navigate multiple funding systems on their own.
- Resilience finance needs better shared data and common language. Lenders, investors, insurers, public agencies, and communities need more transparent information about what makes projects fundable, how funded projects perform, and which standards or criteria can help similar projects be compared, bundled, and scaled (UNEP FI 2024).
- Participants emphasized the importance of testing real transactions rather than waiting for a fully designed platform. Early pilots can help clarify what governance, underwriting, insurance, data, and capital-stacking functions are needed, and can create examples that others can adapt.
- The group identified state policy as a practical near-term opportunity. With many states entering new political and legislative cycles, there is an opening to define what changes in state law, funding rules, procurement, revenue authority, and public finance tools could unlock more resilience investment.
Duke University and the Milken Institute will continue to synthesize the outputs from the convening and support follow-up conversations with organizations ready to lead specific pieces of this work. The goal is to keep momentum moving from discussion to design, from design to pilots, and from pilots to a more coordinated system for resilience investment.
Acknowledgments and More Information
This convening was organized by Duke University’s Nicholas Institute for Energy, Environment & Sustainability and the Milken Institute. It was aligned with and funded by the Duke Climate Commitment.
Authors and Affiliations
- Victoria Salinas, Climate Leader in Residence, Duke University
- Lydia Olander, Ph.D., Program Director, Nicholas Institute for Energy, Environment & Sustainability, Duke University
- Sara Mason, Senior Policy Researcher, Nicholas Institute for Energy, Environment & Sustainability, Duke University
- Elizabeth Losos, Ph.D., Executive in Residence, Nicholas Institute for Energy, Environment & Sustainability, Duke University
- Rachel Halfaker, Director, Finance, Milken Institute
- Sarah Ortner, Associate Director, Finance, Milken Institute
Use of Artificial Intelligence Tools
Artificial intelligence tools, including Claude and ChatGPT, were used during the development of these proceedings to synthesize meeting notes from multiple individuals, assist in drafting, and refine language and structure. Consistent with Duke University guidance on responsible AI use, AI-generated outputs were treated as provisional and subject to verification. All analysis, strategic direction, conclusions, recommendations, and final editorial decisions were led by the authors. This document reflects a human-led development and review process informed by practitioner insights shared during the convening, interdisciplinary collaboration, and responsible use of AI-enabled tools.
Citation
Salinas, V., L. Olander, S. Mason, E. Losos, R. Halfaker, and S. Ortner. 2026. From Fragmentation to Coordinated Action: Building a US Resilience Finance System. Durham, NC: Nicholas Institute for Energy, Environment & Sustainability, Duke University. https://nicholasinstitute.duke.edu/publications/fragmentation-coordinated-action.
Contents
