Update: Mexico Remittances Fall 4.6% to $61.8B — First Drop in 11 Years

What changed since the research vintage

In early August we mapped the theme’s baseline: aging hosts run heavy public pensions while remittances hit $685B. That post answered the stock question — how age and migration show up in money flows across a dependency×remittance scatter, corridor pipes, and OECD pension shares. This update answers the flow question markets and finance ministries actually trade on: what moved in the newest official vintage, and does the two-ledger story still hold once Mexico’s remittance streak breaks?

Three information events force a refresh. Banxico’s December 2025 remittance bulletin (published early February 2026) prints full-year family remittances at $61.791 billion, a −4.6% decline from $64.746 billion in 2024 — the first annual drop in 11 years and the largest since 2009. That Banxico 2024 actual also restates our research post’s Brief 41 Mexico estimate ($68B) down by about $3.3 billion. BBVA’s Migration Observatory shows the rest of Latin America and the Caribbean still racing: remittances to Honduras (+25.3%), Guatemala (+18.7%), El Salvador (+17.8%), Colombia (+10.6%), and the Dominican Republic (+10.3%) imply an ex-Mexico LAC average above 16%. OECD Pensions at a Glance 2025 (November) keeps Italy near 16% of GDP on public cash pensions and projects the OECD-32 average from 8.8% (2023–24) to 10.0% by 2050 — a +1.2 pp host-side path while the US→Mexico private pipe cools.

The dashboard above is built as a vintage delta: Mexico’s annual series with YoY overlay, LAC growth divergence bars, prior→new dollar dumbbells, an age×remittance scatter with Mexico’s 3.7%→3.4% GDP-share shift, and the new OECD pension path.

The headline table: Brief 41 / research vs newest prints

Treat the Mexico −4.6% carefully. It is a true period delta on Banxico’s books — not a measurement residual like some ownership restatements elsewhere on this site. The Brief 41 $68B → $64.7B move is a restatement: the research post used the World Bank’s 2024 estimate; Banxico’s closed year is lower. Stack them and the cumulative gap versus the research narrative is roughly $6B of Mexico dollars that are no longer on the prior path.

Mexico’s streak break is the cleanest remittance delta

For eleven consecutive years, Mexico’s family remittances expanded — nearly tripling across the streak, per BBVA’s reading of Banxico history. 2025 ends that run. December alone printed $5.32 billion (+1.9% YoY), breaking an eight-month decline streak inside the year, but the annual total still fell to $61.8 billion. Electronic transfers remain ~99% of the flow; the average December transfer rose slightly (about $408 vs $375 a year earlier), so the decline is primarily a volume of remitters / transfers story, not a collapse in ticket size.

That matters for the theme’s core question. In the research post, Mexico sat as the bridge remittance origin: youngish age structure (old-age dependency ~13), material remittance/GDP (3.7%), thin public pensions (~3%), and the world’s largest bilateral corridor from the United States. A −4.6% print does not flip Mexico into an aging-host ledger. It does pull remittance dependence left on the scatter — to about 3.4% of GDP — and it stresses the US labor-market / immigration-enforcement engine that Brief 41 credited for LAC strength. Banxico and market analysts point to a softer US labor market, stricter enforcement, and a strong peso as the proximate drivers. Inter-American Dialogue work estimating remitter and migrant-flow losses puts the missing dollars in the low billions — the same order of magnitude as Banxico’s annual gap.

LAC peers refuse to copy Mexico’s print

The vintage is not “remittances are rolling over everywhere.” BBVA’s Observatory estimate that ex-Mexico LAC grew more than 16% on average in 2025 is the divergence panel’s thesis. Honduras (+25.3%), Guatemala (+18.7%), and El Salvador (+17.8%) are still compounding from US payrolls and newer migrant cohorts. Guatemala’s approximate level move (~$21.5B → ~$25.5B) alone offsets a large slice of Mexico’s dollar decline in regional accounting even if it cannot replace Mexico’s absolute size.

Read that beside the global remittance corridors post: country-pair plumbing can keep expanding in Central America while the largest pair (US→Mexico) cools. Concentration risk and dependence risk are different meters — the same lesson India vs Tajikistan taught on the research scatter. Mexico can shed $3 billion and remain the Western Hemisphere’s remittance giant; Honduras can grow 25% and still be a fraction of Mexico’s stock.

Host-side pensions: PaG 2025 adds the path, not just the level

The research post’s host ledger leaned on OECD levels: Italy and Greece above 16% of GDP, OECD average 8.1%, Japan ~9.3%, US ~7.1%. Pensions at a Glance 2025 largely confirms those levels — Italy’s country note still puts public pension expenditure at about 16% of GDP, second only to Greece — and adds the projection the prior post treated as background. Across OECD-32 countries with full paths, public pension spending rises from 8.8% of GDP in 2023–24 to 10.0% by 2050 (+1.2 pp). Italy’s long-run path actually bends down later in the century as NDC rules mature; the near-term message is still high expenditure plus a shrinking working-age base (Italy’s 15–64 population projected to fall by more than a third by 2060).

That is the demographic cash-flow asymmetry in one sentence: origin-country private pipes can reverse in a single Banxico year; host-country public pension shares move on multi-decade actuarial rails. Policy that watches only remittance headlines will miss the host fiscal path; policy that watches only pension commissions will miss the US→Mexico enforcement shock.

For the US on-budget twin of that working-age mobility, the Social Security trust-fund depletion path still clocks combined OASDI reserves under intermediate assumptions. Immigration assumptions are already inside the Trustees’ demography; Banxico’s 2025 print is the origin twin of how those labor flows are performing in the newest year.

What the scatter shift is really saying

The age×remittance scatter does not need a new geometry. Aging hosts still cluster at high old-age dependency and near-zero remittance/GDP. Remittance origins still cluster young and remit-heavy. What changed is Mexico’s horizontal position on remittance intensity: 3.7% → 3.4%, with absolute dollars down. Tajikistan’s 45% of GDP dependence (Brief 41) remains the extreme; India’s absolute $129B inflow remains the volume champion. The update’s job is not to redraw the map — it is to show that the largest North American pipe can bend without breaking the global remittance > FDI narrative that Brief 41 established at $685B.

Three mistakes still follow from ignoring the joint ledger:

1. Reading Mexico’s −4.6% as an LMIC remittance recession. LAC peers are accelerating; Brief 41’s LMIC aggregate has not been restated downward in a newer World Bank brief we can cite here.
2. Reading OECD pension stability as “no news.” The +1.2 pp path to 2050 is the newsa slow host-side cash claim that remittance YoY prints cannot cancel.
3. Treating remittances as public finance. They remain private household transfers. A Banxico decline hurts origin consumption and FX buffers; it does not automatically repair US OASDI math or Italian NDC transition costs.

What would rewrite this update

1. A World Bank Migration & Development Brief 42/43 that restates 2024 LMIC totals away from $685B or prints a 2025 LMIC growth rate that incorporates Mexico’s Banxico miss.
2. Banxico 2026 H1 showing either a sharp rebound (guest-worker / H-2 expansion thesis) or a second annual decline that turns −4.6% into a trend.
3. OECD PaG revisions moving Italy materially off ~16% or the OECD-32 2050 anchor off 10.0%.
4. US employment composition prints that reverse the foreign-born vs native-born recovery gap Brief 41 used as the LAC engine.
5. Corridor matrix refresh showing US→Mexico bilateral dollars falling in line with Banxico’s national total (KNOMAD ranks in the research post remain 2021-vintage pipes).

Until those print, the live frame versus the research post is narrow: Mexico’s remittance streak is over (−4.6% to $61.8B); Brief 41’s $68B Mexico estimate was high versus Banxico; LAC ex-Mexico is still hot; OECD host pensions are on a path to 10% of GDP by 2050; the age×migration money map still holds.

Caveats and methodology

1. Banxico ≠ World Bank Brief 41. National family-remittance concepts and World Bank/IMF BOP remittance aggregates differ. We use Banxico for Mexico YoY and restatement; we retain Brief 41 for the LMIC $685B anchor until a newer brief restates it.
2. BBVA LAC growth rates are Observatory estimates based on national sources; treat peer +16% / +25% figures as disclosed secondary synthesis, not a World Bank regional table.
3. Mexico remittance/GDP ~3.4% for 2025 follows market analysis around Banxico’s print; it is not a Brief 41 cell.
4. OECD pension paths mix disclosed anchors (2023–24 / 2050 OECD-32) with illustrative country interpolations for chart continuity; Italy’s ~16% level is from PaG 2025 notes.
5. Old-age dependency ratios remain UN WPP 2024 anchors from the research postthis update does not invent a new demographic vintage.
6. Corridor dollar ranks (US→Mexico ~$52B in the research post) are not re-estimated here; Banxico’s national total is the YoY meter.
7. Informal remittances still mean true flows exceed recorded totals in both vintages.

Primary sources: Banxico Ingresos y Egresos por Remesas (December 2025 / full-year 2025); BBVA Research Migration and Remittances Observatory (3 Feb 2026); OECD Pensions at a Glance 2025 public expenditure and long-term projection chapters; World Bank People Move / Migration & Development Brief 41 (prior vintage); prior theme baseline in demographic cash flows research.

The shareable takeaway

Versus the research print, Mexico’s remittances fall 4.6% to $61.8B — ending an 11-year growth streak — after Banxico also restates 2024 to $64.7B versus Brief 41’s $68B estimate. LAC peers excluding Mexico still grow more than 16% on average. OECD-32 public pensions are projected to rise from 8.8% to 10.0% of GDP by 2050, with Italy still near 16%. Age and migration still show up as two ledgers — private origin pipes that can reverse in a year, and public host pensions that grind higher for decades. For corridor plumbing see global remittance corridors; for the baseline essay see the research post.