Mexico's M7.3 Quake Missed Its Cat Bond Trigger by One Tenth of a Magnitude

The magnitude 7.3 earthquake that ruptured 58 kilometers off Puerto Madero on July 17 landed inside a live trigger box of the IBRD CAR Mexico 2024 catastrophe bond, in a cell where the $70 million Class B notes activate at M7.4 (USGS, July 2026; Artemis, July 2026). One tenth of a magnitude unit kept $17.5 million of principal in investors' hands. Mexico's own seismological service measured 7.4.

The United States Geological Survey settled the event at moment magnitude 7.3, a thrust-interface rupture at 14:48 UTC, 18.6 kilometers deep, centered at 14.60 north, 92.95 west near the Guatemala border (USGS, July 2026). Artemis, which published the near-miss analysis on July 20, placed the epicenter inside one of the bond's parametric trigger boxes and noted that the same cell would have needed a far stronger M7.9 event to touch the $225 million Class A notes (Artemis, July 2026). Chiapas civil protection crews reported no structural damage on their first inspection sweeps (N+, July 2026). The bond did what its map said it would do, and the map is where the actuarial content lives.

How the 2024 Grid Reads an Earthquake

The earthquake classes of IBRD CAR Mexico 2024 run on what the World Bank calls a cat-in-a-grid trigger. Covered territory is divided into boxes, each box carries its own minimum magnitude for each class of notes, and a qualifying event pays a share of principal that steps up with intensity. Verisk's AIR unit built the risk model and serves as calculation agent, reading each event's magnitude, depth, and location against the grid (World Bank, February 2025). The 2024 renewal redrew that geometry deliberately: the World Bank's case study records that the grid gained an explicit depth condition, smaller boxes, and smoother transitions between adjacent cells, changes the Bank describes as closing Mexico's coverage gap while holding down basis risk (World Bank, February 2025).

The grid prices geography the way an exposure-rating actuary would. The case study's published Class A payout map shades each cell by the magnitude that pays 100 percent of principal, in bands running from 6.7 at the low end to 8.8 and above at the high end (World Bank, February 2025). A magnitude 6.7 in the most sensitive cells pays the full $225 million tranche, while the offshore cell where July's rupture landed required 7.9 for Class A and 7.4 for the riskier Class B. A subduction-zone event off the Chiapas coast, however energetic, projects modest modeled loss onto the federal asset base the bond protects, so its cell carries a high bar; a smaller, shallower event under a population center clears a much lower one. Attenuation, depth, and distance to exposure are all priced into the thresholds before the event, which is what lets the payout decision collapse to a table lookup after it.

The payout function is stepped rather than binary. The World Bank describes the 2024 design as a stepped linear function running from 30 to 70 percent and from 70 to 100 percent of principal as intensity climbs; Artemis's reading of the Class B terms for this specific cell put the first activation at 25 percent of principal for an M7.4 (World Bank, February 2025; Artemis, July 2026). On $70 million of notional, that first step is worth $17.5 million. The two descriptions bracket the same design intent, a ladder of partial payouts in place of a single zero-to-total cliff, meant to scale the recovery roughly with severity. July 17 stopped one rung below the ladder.

Two Agencies, One Rupture

Every parametric trigger nominates a reporting agency, and the nomination decided this event. The USGS solution settled at Mww 7.3, and the agency's own catalog shows the spread of estimates beneath that single decimal: one contributing network's Mwp solution read 7.4, the W-phase inversion 7.3, a body-wave method 7.0 (USGS, July 2026). Mexico's Servicio Sismologico Nacional published magnitude 7.4, locating the event 135 kilometers southwest of Ciudad Hidalgo (N+, July 2026). The bond does not insure the earthquake; it insures the nominated agency's estimate of the earthquake. Moment magnitudes routinely differ by a tenth between agencies because they are inversions of different waveform sets through different velocity models, and a tenth is exactly the width of this miss. Artemis noted the note terms allow a window for parameter revisions, and that as of July 20 the USGS had revised nothing in either direction (Artemis, July 2026).

Mexican cat bond investors have already lived the mirror image. In June 2020 a USGS M7.4 struck near La Crucecita, Oaxaca, inside the trigger zone of the prior IBRD/FONDEN 2020 bond, killed at least six people, and damaged buildings across Oaxaca state (NBC News, June 2020). The bond paid nothing because that cell's minimum sat higher: "With a magnitude of 7.4, yesterday's earthquake does not meet the required parameters to trigger these bonds" (Twelve Capital via Artemis, June 2020). Six years apart, the same printed number lands on opposite sides of the payout line depending on which cell contains it. Both outcomes follow from the same premise, that each cell's threshold encodes the modeled federal loss of an event there, and both resolved the way the map was drawn.

Twenty Bonds Since 2006, Three Checks Written

Mexico built the sovereign parametric market. Its first cat bond came in 2006, the World Bank counts 20 Mexican issuances since with the Bank as advisor or issuer on every one, and the current $595 million program, placed as four bonds in April and May 2024, is the largest single-country sovereign cat bond transaction on record (World Bank, February 2025). More than 30 investors bought it, with ILS funds taking 68 percent of the paper (World Bank, February 2025). Across those two decades, three events have converted principal into fiscal relief, and two more have now tested the trigger without moving money.

Event	Bond and tranche	Governing parameter	Outcome
Hurricane Patricia, October 2015	MultiCat Mexico 2012 Class C, $100m	Landfall pressure 932mb, inside the 920-932mb band	50% payout, $50m, resolved February 2016
Chiapas M8.1 quake, September 2017	IBRD/FONDEN 2017 Class A, $150m	Magnitude and location box	100% payout, $150m, confirmed October 11, 2017
Oaxaca M7.4 quake, June 2020	IBRD/FONDEN 2020 quake classes, $235m	In-zone, below the cell's magnitude minimum	No payout
Hurricane Otis, October 2023	IBRD/FONDEN 2020 Class D, $125m	Minimum central pressure 922mb near landfall	$60m payout, paid April 2024
Chiapas M7.3 quake, July 2026	IBRD CAR Mexico 2024 Class B, $70m	Mww 7.3 against a 7.4 cell minimum	No payout, 0.1 short

The 2017 payout is the speed benchmark. The IBRD/FONDEN 2017 bond closed in August 2017 with a $150 million Class A earthquake tranche carrying a 3.43 percent expected loss at a 4.5 percent risk margin; the magnitude 8.1 Chiapas earthquake of September 8, 2017 exhausted it, and the full payout was confirmed on October 11, five weeks after the rupture and roughly two months after investors funded the notes (Artemis, October 2017). No loss adjuster visited anything. The published event parameters were the claim file.

The hurricane side has run slower, because named-storm triggers key off National Hurricane Center best-track pressure data, and the NHC's definitive tropical cyclone reports arrive months after landfall. Patricia's $50 million payout waited on the February 2016 report that fixed landfall pressure at 932 millibars, the last millibar inside the band MultiCat 2012's Class C required (Artemis, February 2016). Otis, a Category 5 landfall at Acapulco in October 2023, produced a $60 million payout that reached Mexico in April 2024, after the NHC's March 2024 report put minimum central pressure at 922 millibars (NHC, March 2024; World Bank, February 2025). The 2024 redesign attacked that lag directly: storm payouts now read the NHC's B-deck data, available about two weeks after the final public advisory, rather than waiting for the full report (World Bank, February 2025).

The Premium Ledger Behind the Near-Miss

Mexico pays the risk margin component of the coupon, with the World Bank's funding rate riding on top for investors, and the disclosed terms let anyone reconstruct the ledger (World Bank, April 2024). The margins run 4.0 percent on Class A, 11.0 percent on Class B, 13.5 percent on the Atlantic hurricane Class C, and 12.0 percent on the Pacific hurricane Class D (World Bank, February 2025). Applied to the tranche sizes, the program costs Mexico roughly $54.6 million a year, of which the two earthquake classes account for about $16.7 million against $295 million of quake limit. In the 27 months between settlement and July's event, Mexico had spent about $38 million of earthquake risk margin and recovered nothing, the unremarkable base case for remote-layer protection and still the number a budget hearing will hear first.

Class	Size	Peril	Trigger	Expected loss	Attachment prob.	Risk margin	Margin / EL
A	$225m	Earthquake, low risk	Cat-in-a-grid	0.90%	1.17%	4.0%	4.4x
B	$70m	Earthquake, high risk	Cat-in-a-grid	5.84%	8.30%	11.0%	1.9x
C	$125m	Atlantic hurricane	Cat-in-a-gate	5.69%	7.96%	13.5%	2.4x
D	$175m	Pacific hurricane	Cat-in-a-gate	4.09%	6.26%	12.0%	2.9x

Risk statistics and margins are the World Bank's disclosed figures; the multiple is the quotient of the two columns. The shape is the standard cat bond result, remote layers earning the richest compensation per unit of modeled risk: Class A's 1-in-85 attachment probability commands 4.4 times its expected loss while Class B's roughly 1-in-12 layer earns 1.9 times. The vintage comparison shows what renewing into the post-Ian hard market cost the sponsor. The March 2020 bond's equivalent Class A carried the identical 0.90 percent expected loss at a 3.5 percent margin, a 3.9x multiple, and its high-risk quake tranche paid 1.6x against 2024's 1.9x (Artemis, March 2020). Pricing has since rolled over: the site's Q2 2026 cat bond coverage tracked secondary spreads compressing to 5.71 percent in late June with record demand absorbing record issuance. A 2028 renewal priced at current multiples would buy the same grid materially cheaper, an option the sponsor holds and the 2024 investors have already been paid to write.

The cross-cycle ledger is blunter. Mexico's triggers have paid $260 million ever: $50 million for Patricia, $150 million for Chiapas, $60 million for Otis. The current program alone will cost around $218 million in risk margin over its four years. Mexico's stated objectives for the 2024 structure were closing the coverage gap, holding down basis risk, and accelerating hurricane payouts, and its behavior since, doubling the separate traditional parametric insurance program to roughly $575 million for the June 2026 to May 2027 term and carrying about $1.17 billion of total parametric protection, says the government prices those objectives well above expected-value breakeven (Artemis, June 2026).

What the Box Buys That Indemnity Cannot

An indemnity trigger would dissolve the near-miss problem by definition: pay actual loss, no cells, no dueling magnitude estimates. Primary carriers increasingly structure exactly that, and Travelers' record $750 million Long Point Re placement this spring runs on an indemnity trigger and per-occurrence basis (Artemis, May 2026). Indemnity works for an insurer because its claims systems produce an auditable ultimate. A sovereign's do not. Mexico's covered exposure is a federal asset base spread across ministries, states, and municipalities, and settling an indemnity claim on it would mean adjusting public infrastructure losses to an agreed ultimate across all of them, a process measured in years and contestable at every step. The parametric bond pays while emergency-phase spending is still happening, five weeks after the 2017 earthquake against the multi-year tail of a public-works loss adjustment.

The price of that speed is exactly the geometry this month put on display, and the Otis arithmetic states it plainly: a $60 million payout against the NHC's $12 to 16 billion total damage estimate (NHC, March 2024). The bond is a liquidity layer scaled to first response, two orders of magnitude below reconstruction cost, sitting inside a stack that now includes the doubled Agroasemex-fronted parametric insurance program and ordinary budget instruments (Artemis, June 2026). Judging the instrument against reconstruction cost, or against a single near-miss, mistakes which layer of the stack it occupies.

Where the 0.1 Lands in Basis Risk Pricing

The reflex is to read July 17 as basis risk made visible. The classification runs the other way on the facts. Basis risk is the mismatch between payout and need, and a parametric program's history sorts into four quadrants: payout with loss, payout without loss, loss without payout, and neither. July's event produced no reported structural damage and no payout, which files it in the quiet quadrant where trigger and need agree. Had the SSN's 7.4 been the nominated datum, investors would have funded $17.5 million against near-zero damage, a false positive on the other side of the ledger. The genuine sponsor-side failure in Mexico's record is June 2020, real deaths and widespread building damage with no recovery, and the genuine structural compromise is Otis, a sized-for-liquidity layer meeting a $12 to 16 billion tail event. An actuary evaluating parametric value for a sovereign client should populate that two-by-two from program history before repricing anything off a near-miss, because a trigger that almost fires against no ground loss has tracked the loss exactly.

The design record shows the sponsor's actuaries doing precisely that. Each generation of Mexico's grid has been redrawn against accumulated events: the 2024 structure shrank the boxes, added the depth condition, smoothed the transitions between adjacent cells, and moved storm payouts to B-deck timing, four separate corrections aimed at the quadrants where trigger and need had disagreed (World Bank, February 2025). Trigger design is experience rating with a four-year update cycle. The July event now sits in that experience file alongside a subtler exposure the site has covered from the modeling side: if vendor earthquake models are systematically understating tail hazard, as the supershear research suggests, the expected-loss figures that anchor parametric spreads inherit the understatement, and the sponsor's margin-to-EL multiple is being computed on a soft denominator.

For investors, the near-miss is informationless for modeled EL and informative about everything around it. The 2024 quake tranches remain loss-free with 21 months to run, Class B collecting 11 percent a year on a layer that just demonstrated its trigger discipline under a live M7.3. Spreads on new sovereign parametric paper will price off the soft market documented at the mid-year renewals, not off this event. The working questions between now and the April 2028 renewal are concrete: whether the USGS revision window closes without movement, whether Mexico extends the grid refinements another generation, and whether a softening market compresses the 4.4x remote-layer multiple toward its 2020 level of 3.9x. Actuaries advising either side of that table should treat the trigger map, not the headline magnitude, as the object under negotiation.