Updated 2026-09-16
The brief's Mexico call still stands and is slightly firmed, because the June 2026 ASF programme change turns out to have eliminated project- and contractor-level audits in favour of whole-entity integral reviews. The correction cuts in the direction of the forecast on thread one: with 332 granular audits replaced by 279 entity-level reviews, ASF findings will attach to public entities and their officials rather than to named contracts, making a charged private contractor entity before February 2027 less likely, not more. But it also costs the brief part of its ranking argument. The load-bearing assumption flagged at publication, that announced-mandate breadth proxies for exposure, takes a direct hit: the mandate is fast but the audit lens just got coarser, and coarser findings are harder to convert into anything at the Fiscalia Especializada. Twenty-one complaints filed with the FGR on 26 June is consistent with high volume and thin conversion, and nothing in the evidence moves it past referral. The pivotal open question is unchanged in form but sharper in stakes: does any of that complaint flow reach formal charging, and if the integral-audit model structurally cannot isolate a contractor, does Mexico's practical exposure converge toward Brazil's faster than the brief allowed. Brazil and Chile are simply unobserved here.
Forecast: Mexico carries the most practical exposure. The gap widens through end-2027.
Fastest-growing detection reach, no settlement option, and the weakest recourse drive this ranking.
Mexico: speed reaches cash flow — Referral speed, not conviction risk, is what hits your portfolio. — ASF opens direct investigations and files complaints in real time. No corporate settlement channel exists to buy finality. Interim relief is gone in concession and permit disputes, so an ASF-derived complaint is immediately cash-flow effective and only retrospectively contestable.
Brazil: finality is purchasable — High enforcement frequency is offset by codified, predictable resolution. — Exclusive CGU competence, published agreement terms, and dated January 2026 settlements create a defined counterparty and price. Bilateral rescission procedure limits discretionary reopening. Three-year CNEP bar on non-compliance is the real commercial risk, not settlement size.
Chile: counterparty and timing dominate — Direct corporate liability is rare; schedule disruption and intermediary risk are not. — Contraloría's ex ante legality gate delays projects but blocks illegal acts before they crystallise. Criminal exposure under the 2023 statute is severe but infrequent and slow. Fines in practitioner commentary are described as capable of exceeding USD 200 million, but no corporate conviction has been recorded.
In Mexico, ASF referrals produce no charged private contractor entity (Likely) — ASF has inverted the statutory baseline under which audit of the Cuenta Pública followed the close of each fiscal year. It now opens direct investigations, sanctions refusal to hand over requested information and files complaints in real time, paired with integral audits, data intelligence and citizen-complaint follow-up. In May 2026 the Auditor Superior convened state finance secretaries, anti-corruption bodies and legislators around the same model, pulling the federalised spending layer into scope. There is no consolidated corporate settlement channel to buy finality, and interim relief is gone in permit and concession disputes.
Brazil's CGU leniency agreements stay near January 2026 ticket sizes (Likely) — Brazil's detection perimeter already includes private legal persons via the 2013 Anti-Corruption Law; the 2026 change is procedural consolidation through a CGU and AGU interministerial normative ruling dated 19 December 2025 codifying negotiation, execution and rescission of leniency agreements. It is the only one of the three with a dated 2026 conversion record. A portfolio company facing an anti-corruption problem has a defined counterparty, a defined price, a defined publication regime and a defined non-compliance penalty, which is diligenceable and insurable in a way real-time referral risk is not.
No corporate conviction under Chile's economic-crimes statute through 2027 (Likely) — Chile's perimeter is bifurcated: the Contraloría General's legality and audit gate over state transfers, and criminal exposure for firms under the 2023 economic-crimes statute. The binding gate is ex ante legality review, not real-time investigation, and it cuts both ways: it delays projects but blocks illegal acts before they crystallise, independent of executive will. The exposure vector for a growth-stage investor is a portfolio company whose revenue depends on state or regional-government transfers of the type at issue in the Caso Convenios perimeter, plus director-level criminal liability requiring a genuinely implemented compliance programme.
Practical enforcement exposure, not statutory quality: detection mandate, conversion capacity, negotiated resolution, recourse, selectivity, applied identically to all three. Mexico carries the most exposure for growth-stage investors through 2027 and the gap widens. Brazil ranks second on frequency, first on predictability, so lowest practical risk. Chile: lowest frequency, highest per-event severity. Three instruments, not one scaled template.
Used to be: Mexico's audit of the Cuenta Pública followed the close of each fiscal year, giving counterparties time to prepare.
Now: ASF opens direct investigations, sanctions refusal to hand over information, and files complaints in real time, with the federalised spending layer pulled into scope.
ASF has inverted the statutory baseline. Where audit once followed the fiscal year's close, it now runs in parallel with spending. The May 2026 national coordination meeting pulled state finance secretaries, anti-corruption bodies and legislators into the same model, extending that real-time reach across the federalised spending layer where growth-company public-sector revenue concentrates.
The 2025 Cuenta Pública audit programme was modified on 5 June 2026 to add audits including port-administration entities. There is no consolidated corporate settlement channel to buy finality, and interim relief is gone in permit and concession disputes, so an adverse action is immediately cash-flow effective and only retrospectively contestable.
For Brazil, the change is procedural rather than perimeter-widening: the interministerial normative ruling dated 19 December 2025 codified how CGU and AGU negotiate, execute and rescind leniency agreements, producing dated 2026 settlements and a bilateral rescission procedure that limits discretionary reopening. The framework was already there; it now has a published operating manual.
Five dimensions applied identically to each jurisdiction: detection mandate and timing, conversion capacity, negotiated-resolution channel, adjudicatory recourse, selectivity risk. Each country then run through base, deterioration and stabilisation. Confidence moderate. The Mexico leg rests on 2026 primary institutional material and drives the directional call; the Brazil leg on dated, quantified 2026 enforcement actions. The Chile leg is weakest: figures on the Caso Convenios perimeter come from a congressional committee document referencing 2024 Fiscalía reporting, and no verified 2026 Chilean enforcement-volume data point was retrieved, so Chile's ranking is structurally rather than empirically derived. No market-pricing source was retrieved, so no credit, curve or currency indicator informs the ranking.