One in Three Homeowners Lack Disaster Coverage — C.D. Howe Proposes Federal Fix

A protection gap is not the same as being underinsured on your own home. It is a system-level measurement: the share of economic losses from natural disasters that the insurance industry does not cover. The C.D. Howe study estimates that share at roughly 37% across Canada — about two-thirds of natural-disaster risk insured, one-third left to households, governments, or absorbed as private loss.

That definition tracks closely with how federal housing research describes the same concept. A CMHC-commissioned comparative study defines the natural-catastrophe insurance protection gap as the difference between total economic losses from a disaster and the amount actually insured, and pegs Canada's 2020 catastrophe losses at roughly US$3.9 billion against US$2.1 billion insured — leaving a national protection gap of about US$1.8 billion in a single year. The 37% figure is not a one-off. It is a stable shape of the Canadian market.

Two structural realities push it up. The first is geography. High-severity perils — earthquakes in British Columbia and Quebec, overland flooding in Atlantic Canada and Manitoba's Interlake — are concentrated in regions where take-up of the relevant coverage is uneven. The second is behaviour. Households consistently underestimate low-frequency, high-severity risk and buy less coverage than the underlying hazard would justify. Koeppl's report treats both as evidence that voluntary take-up alone is unlikely to close the gap.

The proposed mechanism is reinsurance — insurance that insurers buy to limit their own catastrophic exposure. Under Koeppl's design, Ottawa would assume industry-wide natural-disaster losses above defined thresholds in exchange for risk-based premiums from insurers. Routine claims would stay with private companies. Tail risk — the kind that would otherwise threaten solvency — would migrate to the federal balance sheet, which is cheaper to fund because of the government's lower borrowing cost.

Two design features distinguish it from existing arrangements. First, a Crown corporation reinsurer rather than a one-off appropriation: a permanent institution that would price tail risk and collect premiums in advance, not a relief envelope opened after the fact. Second, the report proposes that natural-hazard coverage be mandatory or included by default in property policies, with homeowners free to opt out only if they also forgo eligibility for ongoing government disaster assistance. That is the lever aimed directly at the protection gap. Coverage stops being an active purchase and becomes the default state.

The report is careful to spread the catastrophe pool nationally rather than concentrate it in high-risk regions. That structure matters because risk pooling collapses if only the most exposed households participate. A pool that includes a homeowner on the Prairies, an apartment owner on the West Coast, and a property on the Atlantic shoreline is solvent. A pool that only includes the highest-risk addresses is not — that is the adverse-selection problem federal policy has run into before, including the continuing hesitation in Ottawa over a dedicated national flood insurance program.

The case for a federal backstop rests on a hard fact about the existing industry safety net. It is small. The Property and Casualty Insurance Compensation Corporation, the industry-funded body that steps in when an insurer fails, holds a compensation fund of roughly $60 million and has authority to raise up to about $1.27 billion per year from member insurers through special assessments. That sounds substantial in isolation. Against the scenarios Canadian actuaries already model, it is not.

PACICC's own modelling concludes that the compensation fund would need to be roughly four times larger — on the order of $225 million to $250 million — to manage the contagion risk of a top-70 insurer failure or an earthquake generating $30 billion to $35 billion in insured losses. PACICC has already added a $250 million standby line of credit from six major Canadian banks to increase liquidity in the first 48 hours after a failure, but the basic numerical mismatch remains. An industry pool sized in the low hundreds of millions is calibrated to ordinary insolvency, not to a major catastrophe.

The C.D. Howe argument is not that the industry has failed. It is that asking insurers to self-insure against once-in-a-generation losses produces premiums that price catastrophe coverage out of reach for the households most exposed. Moving the top of the risk tower to a government balance sheet, in this view, is what makes the bottom of the tower affordable again. The same dynamic is already visible at the retail level, where the market has been tightening with $10,000 deductibles and coverage pullbacks becoming routine for high-risk addresses.

The international framing in the report is not rhetorical. Most of Canada's G7 peers already have some version of what Koeppl is proposing. France operates a state-backed catastrophe scheme. The United States runs national programs for flood and earthquake risk. The United Kingdom backs Flood Re, a reinsurance pool for high-risk homes. A separate Insurance Business analysis reports that Canada is the only G7 country without any form of government-supported climate or natural-catastrophe backstop for its property-and-casualty sector, with Lloyd's Americas president Mark Lipman calling the country an "outlier" among peers.

A 2024 C.D. Howe benchmarking study made a related point: in many developed countries, more catastrophe risk is shifted to governments via state-run pools or backstop mechanisms, and the absence of those arrangements in Canada may help explain why homeowner premiums here run relatively high compared to peer markets. The current report builds on that observation. If similar economies have decided that catastrophic tail risk does not belong on private balance sheets alone, the question is not whether Canada should consider the same architecture. It is why it has not. The financial backdrop is hard to dismiss — insured climate losses in Canada nearly tripled to $37 billion over the past decade, and 2024 alone delivered roughly $9.2 billion in catastrophe payouts.

For a homeowner reading this, the abstract policy debate has a concrete entry point: the premium notice. Home insurance premiums rose 8.6% nationally year-over-year in Q1 2026, with Alberta leading at 16.2% and several other provinces posting double-digit increases. That is the affordability backdrop the C.D. Howe report lands against. Households are already paying more, often without a clear sense of what the extra premium is buying.

The proposed backstop would not lower premiums tomorrow. It would change the structure that determines them. Two effects are plausible. First, mandatory or default natural-hazard coverage broadens the pool of insured households, which lowers per-policy cost for the high-risk perils where take-up is currently thin. Second, moving tail risk off insurer balance sheets reduces the capital insurers need to hold against catastrophic scenarios, which feeds back into pricing over time. Neither effect is immediate, and the report does not promise specific premium reductions. The argument is structural: a federal backstop changes what insurers have to price for, not just how the bill is split.

For households in regions with explicit catastrophe exposure — earthquake zones on the West Coast and in Quebec, flood-exposed neighbourhoods across the Prairies and Atlantic Canada — the practical question implied by the proposal is whether existing coverage already reflects the actual hazard. The 37% gap is not distributed evenly. Some households are well inside the insured two-thirds. Others are firmly in the uninsured third without realizing it.

C.D. Howe reports are inputs to policy, not policy itself. The proposal will move through the same channels as any major fiscal or regulatory idea in Canada: industry consultation, federal review, provincial response. The Insurance Bureau of Canada and PACICC have both flagged the structural gap for years; this report gives them a specific architecture to point at rather than a general call for federal involvement. Whether the proposal advances depends on political appetite for a new Crown corporation, on the federal fiscal position, and on whether the affordability pressure on households tips the issue from industry concern into ballot-box concern.

For homeowners, the timeline is the most important caveat. Even if Ottawa moved on the proposal immediately, a Crown reinsurer would take years to design, capitalize, and operationalize. Mandatory or default coverage would phase in through provincial regulatory channels, not through a single federal switch. The 37% protection gap will be a market reality for years yet. Climate exposure is also reshaping the underlying risk faster than the policy framework is keeping up, a dynamic that runs through our broader coverage of climate-driven uninsurability. The value of the C.D. Howe report is not that it solves that problem this quarter. It is that it names it specifically enough — with a number, a mechanism, and a Canadian benchmark — that homeowners and policymakers are now arguing about the same thing.