--- title: the protection gap is a condition gap canonical_url: https://ensurance.app/guide/the-protection-gap-is-a-condition-gap markdown_url: https://ensurance.app/guide/the-protection-gap-is-a-condition-gap.md subtitle: "insurance measures the gap in premiums. underneath it is a gap in forests, marshes, and slopes nobody is paid to keep working" category: ensurance --- # the protection gap is a condition gap *insurance measures the gap in premiums. underneath it is a gap in forests, marshes, and slopes nobody is paid to keep working* If you work in reinsurance you have said this number out loud in the last ninety days. If you haven't, here is the definition the people who publish it use. **The protection gap** is the difference between the losses that occur and the losses that are insured. Swiss Re Institute expresses it in premium-equivalent terms: the difference between insurance premiums currently written and those required to fully cover expected economic losses. For 2025 they estimate it at USD 424 billion. That is an honest number, produced by people who are not hiding from it. It is also a number about premiums. Underneath it sits a second gap, measured in something else entirely. The watershed, the forest, the reef, and the meadow exist and do their work — holding snow, breaking waves, keeping soil on the slope — whether or not an industry forms around them. Insurance formed around ships that were already sailing. **[Ensurance](/guide/what-is-ensurance?from=guide)** forms around living systems that are already functioning. It is how they get funded, not what they are. ## the number, as the industry states it Every figure in this section is Swiss Re Institute's, from its natural catastrophe protection gap research and its sigma 1/2026 study on 2025 losses. None of it is ours. - The global natural catastrophe protection gap **widened to USD 424 billion in 2025**, up from USD 395 billion a year earlier, in premium-equivalent terms. The model behind it excludes wildfire for data reasons — worth holding onto, since wildfire is where this post ends up. - Swiss Re's Natural Catastrophe Insurance Resilience Index — the share of modelled protection need actually covered by insurance — **rose to about 27.3% in 2025, from 25.3% in 2015**. A modest but real decade of improvement, and also a statement that almost three quarters of global exposure is still uninsured. - In emerging economies, **80–90% of catastrophe losses are typically not covered** by insurance. - Natural catastrophes in 2025 caused **USD 220 billion in economic losses**, of which **USD 107 billion was insured** — a **49% insured share, the highest on sigma records**. - The Los Angeles wildfires alone produced around **USD 40 billion in insured losses**, the largest insured wildfire loss event on sigma records. Read those together and the picture is not an industry walking away. The insured share of 2025 losses was a record. Coverage has broadly kept pace with exposure for a decade. Swiss Re's own explanation for why the absolute gap keeps widening fits in half a sentence: "the protection gap continues to grow, as there is simply more to protect." More to protect. More value standing in the places where the hazard arrives. ## what "more to protect" leaves out Exposure is one half of that sentence. The other half is whatever stands between the hazard and the exposure — and that is not holding still either. The marsh that used to sit in front of the town is a parking lot. The forest on the slope above the reservoir carries several times the stem density it held a century ago, and burns differently because of it. The floodplain that used to take the crest is drained, diked, and planted. The reef that broke the wave is bleached down to rubble. None of that registers as exposure growth. It registers as severity — the same storm, the same acre, a bigger bill. A catastrophe model carries it as hazard and vulnerability; a premium carries it as a rate. Neither has a line that says *marsh*. Swiss Re sees the same thing from the loss side: for some perils and regions, it writes, exposure alone no longer explains the speed of loss growth, and hazard intensification and shifting vulnerability are becoming material. This is the gap underneath the gap. Call it the **condition gap**: the distance between the function a place still performs and the function it used to perform for the people and property downstream of it. [Condition](/natural-capital?from=guide) is a measurable thing — stocks of forest, wetland, grassland, and reef, and the flows of flood attenuation, fire moderation, and water regulation they produce. It is just not a thing anyone writes a premium against. | | the protection gap | the condition gap | |---|---|---| | **measured in** | Premiums — written versus required | Condition — the function a place still performs | | **closed by** | More coverage, more capacity, more capital | More function — the marsh, the forest, the floodplain working again | | **whose it is** | The insurer, the reinsurer, and the capital behind them | The steward on the land, and everyone downstream of it | | **when it acts** | After the loss, as an indemnity against a future event | Before the loss, as maintenance on a system working now | | **who is paid to close it** | An industry with 340 years of practice | Water funds, land trusts, and public programs — grant by grant, project by project, at a fraction of the scale | ## the retreat is rational When the condition gap widens far enough in one place, insurance prices it, and at some point the price stops clearing. That is not a failure of underwriting. That is underwriting working correctly on a deteriorating physical input. Carriers have pulled back from wildfire country and from flood-exposed ground, and the people who made those calls were reading the same hazard data as everyone else. Both stories are written up elsewhere in this guide: [why carriers stop renewing in fire country](/guide/wildfire-insurance-non-renewal?from=guide), and [what it takes for a flood-exposed property to become insurable again](/guide/flood-insurance-disappearing-how-to-become-insurable?from=guide). Notice what the retreat tells you. Insurance is the instrument that *prices* the condition of a place. It was never the instrument that *funds* it. No carrier gets paid to go restore the marsh, and no carrier should be expected to — that work has a different duration, a different cash flow, and a different owner. So the condition gap sits there with no industry assigned to it — a water fund here, a land trust there, a public grant cycle when one comes through. That is not a gap in insurance. That is a missing industry. ## swiss re already said the hard part The most useful line in the 2026 sigma is not a number. It is Balz Grollimund, Head Catastrophe Perils at Swiss Re, on why loss growth is structural: "it is critical to identify the risk drivers behind this to manage and reduce risks before losses occur." Before losses occur. Published by a reinsurer, in a loss report. Swiss Re puts economics on it too. Its review of US adaptation projects approved between 2010 and 2022 that disclosed a benefit-cost ratio found a **median BCR of 1.86** — close to two dollars of expected benefit for every dollar invested — and it notes that those appraisals often exclude life-safety benefits, broader economic effects, and ecosystem services. Its conclusion is explicit: adaptation and insurance are *both* essential to narrowing the gap. We agree with all of it. The only thing we would add is a category. Adaptation is usually pictured as built work: a levee, a dike, land-use planning, a hardened roof to a Fortified standard. Those are real and they perform. But some of the highest-performing adaptation is not built at all. It is a living system kept in working order — the floodplain reconnected, the forest thinned and burned on purpose, the wetland that is wet again, the reef that still has structure. That kind of adaptation almost never has a maintenance contract or a standing payor. Where it does — a water utility paying into the forest above its reservoir — it is a partnership negotiated once, not an industry. It is nobody's capital project, because it is nobody's asset. Ensurance is the layer that pays for it: continuously, to the people doing the work, on places that are still functioning, before anyone files a claim. :::johnson **a gap measured in premiums sits on a gap measured in condition.** Insurance prices the second one, and when it widens far enough, has to withdraw from it. Funding it is a different job — and the industry that does that job is still being built. [read next: the jobs the ensurance industry needs →](/guide/the-jobs-the-ensurance-industry-needs?from=guide) ::: ## what an insurer can do today Nothing here asks a carrier to stop underwriting. Hold both: the book still has to be priced, and the gap still needs capacity. The move that is actually new is to hold a position in the condition above the book — a **certificate** on one named place, bought from the account that place holds in its own name, where the proceeds fund the stewardship that keeps it working. [What that looks like from a carrier's seat](/solutions/insurers?from=guide&topic=ensurance-industry) — wildfire and flood risk mitigation, risk assessment, measurement and verification, syndicate structures — is set out there by use case. We are not going to quote you a loss-ratio benefit. Nobody has measured one for this instrument yet, and inventing a figure would make this post worth less than the number it opened with. What we will defend is the mechanism: [the risk never transferred in the first place](/guide/there-is-no-risk-transfer?from=guide) — it was redistributed. A risk only gets smaller when the physical condition producing it gets better, and somebody has to be paid to make that happen. If the question on your mind is parametric cover and nature-linked policies, that argument — and why a payout is still not a reef — is [written up separately](/guide/nature-based-insurance-is-still-insurance?from=guide). ## frequently asked questions ### what is the protection gap? The protection gap is the difference between the economic losses an event causes and the share of those losses that insurance covers. Swiss Re Institute measures it in premium-equivalent terms: the difference between the premiums currently written and the premiums that would be required to fully cover expected economic losses. ### how big is the natural catastrophe protection gap? Swiss Re Institute estimates the global natural catastrophe protection gap at USD 424 billion in 2025, up from USD 395 billion in 2024. Its resilience index — the share of protection need covered by insurance — stood at about 27.3%, meaning almost three quarters of global exposure is uninsured. In emerging economies, 80–90% of catastrophe losses are typically uninsured. ### why is the protection gap growing? Largely because there is more to protect. Swiss Re finds that insurance coverage has broadly kept pace with rising exposure — its resilience index improved from 25.3% in 2015 to about 27.3% in 2025 — but the dollar value of exposed assets keeps climbing, so the gap widens in absolute terms even while the covered share holds. Underneath that, the condition of the systems that used to absorb hazard has also declined, which raises severity without ever appearing as exposure growth. ### can insurance close the protection gap on its own? No, and Swiss Re does not claim it can. Its position is that adaptation and insurance are both essential: coverage has to broaden and expected losses have to fall. Lowering expected losses is adaptation — built and living. The built kind has sponsors and appraisal standards. The living kind, the condition of the places upstream of the exposure, is a funding problem, not an underwriting one. ## sources [Swiss Re Institute — natural catastrophe protection gap](https://www.swissre.com/institute/research/topics-and-risk-dialogues/climate-and-natural-catastrophe-risk/growing-exposure/Natcat-protection-gap.html) — USD 424 billion gap in 2025 (from USD 395 billion), resilience index 27.3% from 25.3% in 2015, premium-equivalent definition (modelled gap excludes wildfire), median adaptation benefit-cost ratio of 1.86 [Swiss Re Institute — sigma 1/2026, press release 19 March 2026](https://www.swissre.com/press-release/Wildfires-storms-floods-contribute-to-record-92-of-global-insured-losses-in-2025-says-Swiss-Re-Institute/7b39b1a5-b878-4a55-a5ff-bf5aa561a675) — USD 220 billion economic and USD 107 billion insured losses in 2025, 49% insured share, LA wildfires at around USD 40 billion, 80–90% uninsured in emerging economies, hazard intensification and shifting vulnerability becoming material for some perils and regions, Balz Grollimund quotation ## the series Read first: [what is ensurance](/guide/what-is-ensurance?from=guide) - [what the ensurance industry actually is](/guide/what-the-ensurance-industry-actually-is?from=guide) - [how an industry gets born](/guide/how-an-industry-gets-born?from=guide) - [the word is older than the industry](/guide/the-word-is-older-than-the-industry?from=guide) - [the protection gap is a condition gap](/guide/the-protection-gap-is-a-condition-gap?from=guide) - [the jobs the ensurance industry needs](/guide/the-jobs-the-ensurance-industry-needs?from=guide) - [how to take part in the ensurance industry](/guide/how-to-take-part-in-the-ensurance-industry?from=guide)