The US\u2013Mexico sprint beats Canada \u2014 and the cheese, steel and auto carve-outs will outlast the midterms

Mexico\u2019s economy ministry and the Office of the US Trade Representative resumed talks in early September after the Canada\u2013US track collapsed on August 28, and both sides now want a political win before the November 3 midterm elections. A Mexican source told Reuters on September 11 that \u201ctime is of the essence for the Mexican government,\u201d citing weak growth and credit-rating pressure, while Mexico\u2019s economy ministry said only that \u201cthere are no specific deadlines at this time.\u201d Commerce Secretary Howard Lutnick met President Claudia Sheinbaum virtually on September 11 to keep the sprint on schedule.

With Mexico sending more than 80% of its exports to the US and Canada retaliating dollar-for-dollar after Ottawa\u2019s August 28 breakdown, the negotiating asymmetry is stark. Trump already cut deals with Japan, the EU and South Korea at a 15% auto rate and with the UK at 10%; the Canada framework being mirrored toward Mexico envisions a 15% vehicle duty with an additional US-content discount that would bring the effective rate to about 7% \u2014 an order of magnitude below the current 25% Section 232 hit.

Ternium, the Luxembourg-headquartered steelmaker with the bulk of its mill base in Mexico, is the cleanest beneficiary of any steel carve-out. Its stock trades at $57.61 with an EV/EBITDA of 5.6x \u2014 a deep cyclical-discount multiple on what is essentially a direct USMCA-region steel franchise. The Q2 2026 earnings release already showed a sequential recovery in profitability; a 50%-to-10% Section 232 cut on Mexican-origin steel could rerate the stock meaningfully because Ternium\u2019s North American book is concentrated in Mexico.

Grupo M\u00e9xico, the BMV-listed mining and infrastructure parent whose rail arm Ferromex hauls much of Mexico\u2019s steel, is already compounding: Q2 2026 revenue jumped 33.8% year-over-year to $11.28B with EBITDA up 49.6% to $6.85B, driven by copper, not steel \u2014 so a steel-demand recovery from a tariff cut is upside, not the base case.

The single biggest risk to the sprint is not the sprint itself \u2014 it is the 2036 USMCA sunset clause. The Trump administration formally declined on July 1, 2026 to extend USMCA for another 16 years, kicking off an annual review cycle that runs through 2036. Any party can still invoke Article 34.6 and exit with six months\u2019 notice; if that happens, the regional rules-of-origin regime that today makes 92% of Mexico\u2019s auto-parts exports tariff-free disappears overnight.

Ford and Stellantis are the most exposed individual names. Ford already paid roughly $1.3B in tariff-related costs in Q1 2026 and posted a $1.3B Q2 2026 net loss driven by a $3.6B BlueOval SK disposition charge, even as it raised full-year adjusted EBIT guidance to $10\u2013$11B on the assumption that some of the Mexico pressure relents. Stellantis\u2019s North American footprint is concentrated at plants in Coahuila, Mexico and Ontario, Canada \u2014 meaning a sprint-deal that helps Mexico but not Canada would split its supply chain economics in two. General Motors absorbed an estimated $1.1B tariff hit in Q2 2026 and guided to a $2.5\u20133.5B full-year EBIT-adjusted impact, but used the same call to confirm it is \u201conshoring significant production\u201d to reduce that exposure.

The Mexican peso is its own loser-in-reverse: it has rallied to 16.97 per dollar on September 11 \u2014 a ~20% surge since January 2025 \u2014 exactly because markets are pricing in the bilateral deal. A sprint failure would weaken the peso but also raise the dollar cost of every Mexican import US consumers were about to enjoy cheaper.

Mexico\u2019s auto-parts industry exports roughly $185.79B a year and 92% of every part crosses into the US under USMCA preferences at zero tariff. The non-compliant 8% is what gets caught by the 25% Section 232 wall. So the bilateral deal\u2019s real economic value is restoring that 8% \u2014 roughly $15B of cross-border auto-parts flow \u2014 to the same preferential treatment without changing the 92% that already works.

The chart captures the asymmetry: the bilateral sprint can fix a chunk of the steel number and most of the auto-export number, but it cannot fix the manufacturing PMI contraction, which is already baked into the August 2026 print.

No bilateral text has been published; \u201cinterim\u201d means exactly that \u2014 a partial agreement on the run-up to the midterms, not a substitute for USMCA. The 7% effective auto rate and the 10% steel rate are framework proposals mirrored from the Canada track, not signed commitments. The $1B US cheese exposure figure traces to industry estimates of US shipments to Mexico, not to a tariff schedule. And Mexico\u2019s projected $2.20B Q2 2026 net profit at Grupo M\u00e9xico is a copper story, not a steel one \u2014 a sprint-deal steel rerate is upside on top of an already-strong base, not the base itself.