Climate change and ecosystem degradation are widening the global protection gap faster than private insurance markets can absorb. In 2025, global economic losses from natural disasters reached USD 220 billion, with only around half of them insured. Insurers are retreating from the most exposed markets, premiums are rising beyond the reach of those who need cover most, and the structural logic of risk pooling is breaking down as correlated catastrophes become more frequent. The industry has not been idle in response: a range of financial innovations has emerged to extend coverage where conventional insurance cannot reach, and to bring new sources of capital into a system under increasing strain. What do these instruments address successfully, where do their limits lie, and is closing the protection gap ultimately a question of better product design or of something more structural?
At Broadpeak, we collaborate with industry experts, impact-driven investors, and academic institutions to address urgent global challenges. Through our articles and trilogies, we aim to share the insights we have gained from these projects with our network. Explore all of our published articles and trilogies in the blog section of our website.
The most established response to rising primary-market exposure is to expand the layer of capital sitting behind it. Reinsurers absorb portions of the risk that primary insurers underwrite, allowing those insurers to write policies they would otherwise have to decline or price out of reach. Global reinsurance capital reached a record USD 760 billion in 2025. But expanding capacity is not the same as expanding insurability. As climate losses become more frequent and more correlated, reinsurers face the same problem as primary insurers, only at a larger scale: tail events are arriving often enough that diversification stops working as it used to. The response has been to raise rates or withdraw capacity. To address this, a variety of further financial innovations have emerged precisely to extend coverage where conventional indemnity insurance, even with deeper reinsurance backing, cannot reach.
Catastrophe bonds: including capital markets
The first response moves outside the insurance sector entirely. Catastrophe bonds (or cat bonds) tap capital markets rather than the reinsurance sector to absorb peak losses. Investors receive above-market yields in exchange for bearing the risk of loss if a defined trigger is breached; if no qualifying event occurs, the principal is returned. This brings a pool of capital into the risk-transfer system whose balance sheet is uncorrelated with insurance underwriting cycles, and the instrument has demonstrated genuine capacity to absorb large losses from individual events.
What it has not done is change the geography of coverage. Leigh Johnson is an Associate Professor of Geography and Environmental Studies at the University of Oregon, and her work focuses on climate risk finance and the political economy of disaster risk management. Her research has shown that cat bonds deepen liquidity where the market already exists rather than extending coverage to the uninsured. Analyzing nearly USD 37 billion in cat bond issuances from 1997 to 2011, Johnson found that US hurricane and earthquake risks alone absorbed an average of 67 percent of total capital invested, with European wind and Japanese earthquake risks accounting for most of the remainder. Following the 2008 financial crisis, this concentration deepened rather than broadened.
The instrument works for the upper tail of catastrophe risk in markets that can already structure, issue, and service it, precisely the markets where reinsurance was already deepest. The communities and regions where the protection gap is widest sit outside that geography. Cat bonds thus expand the system’s capacity to absorb peak losses; they do not extend its perimeter. This is the limitation that parametric insurance was designed to address.
Parametric insurance: speed over precision
Parametric products pay out on the occurrence of a measurable trigger such as wind speed, rainfall, or earthquake magnitude rather than on assessed damage. Because there is no claims adjustment process, payouts arrive in days rather than months, and administrative costs are low enough to make coverage viable in contexts where conventional indemnity insurance cannot reach. The instrument is increasingly technology-driven, with satellite data and remote sensing forming the backbone of newer products designed for markets where the infrastructure for traditional underwriting does not exist. When Hurricane Maria struck Dominica in 2017, the country received USD 19.3 million from the Caribbean Catastrophe Risk Insurance Facility within fourteen days. In Puerto Rico, which relied on conventional indemnity policies, more than USD 1.6 billion in conventional claims remained unsettled two years after the same storm.
But the reach comes with two structural limits. The first is legal. Jaroslav Mysiak, principal scientist at the Euro-Mediterranean Center on Climate Change and coordinator of the Horizon Europe NATURANCE project, points out that “you are not looking at on-the-ground evidence of damage, only at on-the-ground evidence of the extreme event triggering the claim. You might end up with cases where people or businesses have not suffered any damage and you are paying the compensation, which is prohibited in certain governance regimes in Europe.” Where indemnity without demonstrated damage is not legally permitted, parametric products cannot operate at all.
The second structural limit is calibration. Because the trigger is a proxy for damage rather than a measurement of it, the policy can fail to pay even where damage is severe. Mysiak gives the clearest illustration: “parametric schemes in the Caribbean against tropical cyclones have been calibrated to atmospheric pressure as a proxy for wind strength, and we have seen cases where the model showed cells that would have triggered the parametric insurance, but the cell was just outside the island, and the island was completely devastated without the triggering threshold being reached.” Designing a trigger that captures damage accurately requires granular hazard data of a kind that is often unavailable in precisely the regions most exposed to climate risk: “Parametric insurance needs to be taken with care and really well calibrated to ensure it delivers the protection it sets out to deliver.”
Leigh Johnson argues that this is not a technical problem to be optimised away but a structural feature of any index-based product. The instrument’s central promise, automating away difficult human decisions, holds only most of the time: “you can automate it maybe 60, 70, 80 percent of the time in a best-case scenario, but that remainder is the piece where trust is built or lost. And you don’t know in advance which cases will fall on which side. For a product to really work, and for you to be confident that it’s working, you still absolutely have to have networks and relations with people on the ground who are speaking back to you. There is no escaping certain sets of human deliberation and decision-making that parametrics are often sold as eliminating “.
Parametric insurance thus extends reach, but managing the failure modes it introduces still requires the institutional capacity, data, and human judgment whose absence the instrument was designed to work around.
Nature-based Solutions: preserving insurability
Each of the three instruments above expands the capacity of risk transfer in some way: more capital behind the primary market, new sources of capital outside it, faster and cheaper payouts where conventional insurance cannot operate. The most conceptually ambitious development in climate risk finance moves in a different direction. Rather than transferring more risk more efficiently, it uses the insurance architecture to fund the natural systems that prevent risk from growing beyond what insurance can absorb in the first place.
The most cited example is the coral reef policy in Quintana Roo, Mexico, launched in 2018 by The Nature Conservancy together with the state government and structured by Swiss Re. A parametric trigger releases funds for reef restoration when hurricane wind speeds exceed a defined threshold, and the scheme was activated in 2020 after Hurricane Delta. By 2023, the model had scaled into the Mesoamerican Reef Insurance Programme, covering eleven sites across Mexico, Belize, Guatemala, and Honduras. The mechanism is parametric, but the object of insurance is not. The reef itself is treated as risk-reduction infrastructure, and the payout funds its repair so that it can continue to absorb wave energy before that energy reaches insured assets behind it.
Mysiak identifies the underlying logic: “by investing into disaster risk reduction, we are keeping the business viability of insurance rather than removing the need for insurance in the first place.” Where conventional risk transfer responds to losses after they occur, NbS-linked products work to keep expected losses inside the range that insurance can still price. Mario Wilhelm of Rift Partners sees the broader category as still nascent: “people don’t know you can actually insure a coral reef. The industry has to innovate and show that it can build products that solve a need that is not yet articulated as demand.”
But this shift in logic comes with a heavier set of preconditions, not a lighter one. Structuring an NbS-linked policy requires scientific quantification of the ecosystem service being insured, legal vehicles capable of holding natural assets in trust, technical capacity to design the trigger, and political agreement among multiple stakeholders on who pays the premium and who controls the payout. The Coastal Zone Management Trust in QuintanaRoo exists because no standard legal structure was available for a state government to insure a reef. Johnson warns that the push to scale these products risks repeating a pattern already visible in parametric insurance, where the elegance of the instrument overshadows the needs it is meant to serve: “when you’re trying to make a market, you often end up with targets: we need to reach this many people, or cover this many hectares of wetland, or this monetary value of risk by 2030. And that’s a recipe for making poor decisions, because then you lose track of the specifics of design and context that need to be considered to make this work in any particular case.” The best products in this space, she argues, are those built around the specific hazards, social arrangements, and recovery needs of the communities they cover, rather than around the volumes that make the market scalable.
What financial markets can and cannot do
Reinsurance provides depth of capital, catastrophe bonds bring in funding uncorrelated with insurance cycles, parametric products cut administrative cost and accelerate payouts, and nature-based solutions fund the ecosystem services that hold risk within insurable ranges. But all of them require the same conditions to function: legal and financial infrastructure capable of supporting complex contracts, data to calibrate triggers and price risk, institutional capacity to structure and service the instruments, and counterparties already integrated into formal financial systems. Those conditions concentrate where insurance markets are already deep and recede where they are thin. Across many developing countries, where more than 90 percent of climate-related losses remain uninsured, the instruments either fail at the point of payout or do not arrive at all. The limits are not in the instruments but in the infrastructure they presuppose, and that distribution is a political and infrastructural fact rather than a product-design problem. As Anthony Chaillet of Rift Partners, puts it, “insurance is part of a broader risk management strategy.”
