PESTEL Analysis of Mexico (2026): record FDI, no new money

Mexico reported its best foreign-investment half-year on record in September 2026: US$34.97 billion, up 2.1%. That is the number most PESTEL analyses of Mexico would put in the Economic cell, and on that number the cell scores favourable.

The same release says something else. 88.5% of that record was profit that companies already operating in Mexico chose to reinvest there. New investment, meaning money from firms putting capital into Mexico for the first time or opening new operations, was US$2.726 billion, down 13.4% from US$3.149 billion a year earlier. Banxico's balance-of-payments measure of net FDI fell 11.7%. The firms already in Mexico are still committing capital. New arrivals have slowed.

A PESTEL analysis scans six external macro forces: Political, Economic, Social, Technological, Environmental and Legal. It asks whether each one favours the decision in front of you. For Mexico in 2026 the answer depends on whether you already have a plant there, more than on what you do or where you sell. This is the eighth country page in the series. The USA had one force holding up the other five, Canada one force rewriting them, and China and India had cells that split by ledger and by who can move them. Mexico's cells split by when you committed.

Position being analyzed

The decision this analysis is built for: a manufacturer or investor deciding in late 2026 whether to build new capacity in Mexico now, expand what it already has there, or wait for the 1 July 2027 USMCA review.

That is the live question because of what happened on 1 July 2026. The three governments met for the agreement's first required joint review, and Mexico and Canada both supported extending it. The United States Trade Representative's statement was two sentences long: "The United States did not agree to renew the USMCA in its current form. As a result, the USMCA is not renewed." The agreement did not lapse. It stays in force to 1 July 2036, with preferential tariffs, rules of origin and dispute settlement unchanged, but it is now reviewed every year rather than every six. US-Mexico bilateral rounds have covered auto rules of origin, steel and aluminium, economic security, agriculture, labour, the environment and regulatory compatibility.

Indicator	Latest reading	As of
USMCA status	Not renewed: in force to 2036, reviewed annually; next review 1 July 2027	USTR, 1 July 2026
GDP growth, Q1 / Q2 2026 (year on year)	0.2% / 2.1%; Q2 +1.4% q/q (flash 1.5%), best since Q4 2020	INEGI, 24 August 2026
Share of Q2 q/q growth from World Cup spending	~two-thirds	Banco Base, 30 July 2026
Policy rate	6.50%, third straight hold	Banxico, 24 September 2026
Inflation, 1H September (headline / core)	3.42% / 3.79%	INEGI via Reuters, 24 September 2026
FDI, H1 2026	US$34.97bn, +2.1%: a record	Economy Ministry, 1 September 2026
…of which reinvested earnings / new investment	88.5% / 7.8%; new investment US$2.726bn, −13.4%	Economy Ministry
Net FDI (balance-of-payments basis), H1 2026	US$24.81bn, −11.7%	Banxico
Remittances, H1 2026 / full-year 2025	US$30.76bn, +3.1% / US$61.79bn, −4.6%	Banxico
Auto exports to the US, Q1 2026	US$38.1bn, −11.3%	US Commerce data via El Economista
Public-sector borrowing requirement, 2026 est. / 2027 budget	4.1% / 3.9% of GDP; April had promised 3.5%	Paquete Económico 2027, 8 September 2026
Population / total fertility rate	130.9m, growing 0.7% / 1.2	INEGI Intercensal Survey 2025

Political — an agreement that is now reviewed every year

The domestic political picture is the most stable it has been in years. President Claudia Sheinbaum's coalition passed the 40-hour-week reform unanimously, carried Mexico's own tariff law 281-24, and delivered a 2027 budget on schedule. Whatever else is uncertain about Mexico, nobody is waiting to find out who governs it.

The uncertainty comes from the United States, in four specific ways:

• USMCA's annual review. The agreement is unchanged, but every 1 July from now to 2036 is a date on which the United States can ask for more. A plant whose payback period runs four or seven or ten years now runs through that many reviews.
• Section 232. The Supreme Court's 20 February 2026 ruling ended the IEEPA tariffs, including the 25% fentanyl tariff on non-compliant Mexican goods. It left Section 232 untouched. Mexican-built vehicles still pay 25% on their non-US content. Rules of origin for autos are on the agenda of the bilateral rounds.
• The 1944 Water Treaty. The 2020-25 delivery cycle closed in October 2025 with Mexico still owing just over 865,000 acre-feet, almost half its five-year obligation. Under US pressure, Mexico committed in February 2026 to deliver at least 0.35 million acre-feet a year and to produce a plan for the arrears.
• Mexico's own China tariffs, which a Political cell has to count as partly a US event. The tariffs of up to 50% on non-FTA imports that took effect on 1 January 2026 were designed under visible pressure over Chinese goods entering the US through Mexico, and they are the measure Mexico can point to in every review.

Score: domestically favourable and low-volatility; externally adverse and now volatile on an annual cycle. The external half is what sets the price of new capital.

Economic — a strong quarter driven by a one-off

The second quarter was the strongest since 2020: +1.4% quarter on quarter and +2.1% year on year, after 0.2% in the first quarter. But Banco Base's Gabriela Siller attributed about two-thirds of the quarterly growth to World Cup spending (construction in April and May, then retail and services during the tournament) and expected a quarter-on-quarter contraction in Q3. The IMF's forecast for 2026 as a whole is 1.2%. The World Cup 2026 PESTEL published in June called Mexico the tournament's relative winner among the three hosts, because a smaller economy turns the same event into a bigger share of GDP. Q2 agrees. The Q3 number will show how much of that was pulled forward rather than added.

Monetary policy is on hold rather than loosening. Banxico kept its rate at 6.50% on 24 September, unanimously and for the third meeting in a row, and stated that Mexican rates should not mechanically follow the Fed. Headline inflation picked up to 3.42% in the first half of September, driven by a 22.8% two-week jump in tomato prices. Core inflation eased to 3.79%, inside Banxico's 3% ± 1 point range.

The fiscal side changed direction on 8 September. The 2027 economic package cuts federal support for Pemex's debt payments to 81.1 billion pesos, from 263.5 billion this year, a reduction of nearly 70%, on the expectation that Pemex runs a financial surplus of about 95 billion pesos. Yet the broad deficit target came in at 3.9% of GDP, not the 3.5% promised in April, with debt rising to 55% of GDP and the 2027 growth assumption cut to 1.5-2.5%.

And then there is investment, which is where the cell splits:

Investment measure, H1 2026	Value	Change
Total FDI (Economy Ministry)	US$34.97bn	+2.1%, record
Reinvested earnings	~US$30.9bn (88.5%)	up
New investment	US$2.726bn (7.8%)	−13.4%
Inter-company accounts	3.7%	—
Net FDI (Banxico, balance-of-payments)	US$24.81bn	−11.7%

Score: favourable for firms already in Mexico, which are reinvesting at record levels. Neutral-to-adverse for new entrants, and the drop in new money is the market's own reading of that.

Social — fewer workers coming, and a shorter week by law

INEGI's 2025 Intercensal Survey counted 130.9 million people, growing at 0.7% a year, the slowest rate in two decades. It put the total fertility rate at 1.2 children per woman, down from 1.9 in 2019, and as low as 0.8 in Mexico City. The median age rose from 29 in 2020 to 32. For a country whose nearshoring pitch rests partly on a young, cheap workforce, this changes the long-run reading: the workforce exists today, but the cohorts entering it after 2040 are being born now, and there are fewer of them.

The second Social shift is legislated. Mexico amended its constitution in March 2026 to reduce the maximum working week from 48 to 40 hours with no loss of pay, phased in over 2027-2030 (46, 44, 42, then 40 hours). The secondary law took effect on 1 May 2026 and requires electronic time records. For a factory running continuous shifts, that adds up to a 17% cut in hours per worker at the same wage.

Remittances, which support household consumption in large parts of the country, recovered after their first fall in more than a decade. They fell 4.6% to US$61.79 billion in 2025, then rose 3.1% to US$30.76 billion in the first half of 2026. That was despite a new 1% US tax on cash-funded transfers from 1 January. Transfers funded from bank accounts and cards are exempt.

Score: neutral now, adverse over time. The demographic trend is slow but has already started.

Technological — demand for power is outrunning the grid

The technology story in Mexico in 2026 is mostly about data centres, and data centres are mostly about electricity. National operational data-centre capacity went from 115 MW in 2024 to 235 MW in 2025 and 279 MW in 2026, and Querétaro holds about 72% of it. Hyperscaler commitments continue. The constraint is the grid run by the state utility CFE, which according to September reporting has not expanded fast enough to serve new industrial loads in northern Mexico. Firms describe constraints on grid connection, power quality and clean-energy contracts, and often need private backup generation.

On the policy side, the Kutsari semiconductor-design programme launched under Plan México in February 2025, with hubs in Puebla, Jalisco and Sonora and a target of 3,000 trained engineers by 2030. It is a design programme, not a fab. The Economy Ministry has discussed AI-supercomputer manufacturing with Nvidia, but those talks had not become committed capacity by September.

Score: favourable for firms already connected to the grid; adverse for a new site that needs megawatts in the next two years.

Environmental — water and power in short supply

Two environmental constraints matter to an investor, and both are about supply rather than regulation.

Water. The drought behind the 1944 treaty arrears is the same drought that affects industrial water in Mexico's northern manufacturing belt. Mexico ended the last treaty cycle almost 50% short. Its 2026 commitment of at least 0.35 million acre-feet a year to the US comes from the same Rio Grande basin that Chihuahua, Coahuila, Nuevo León and Tamaulipas draw on. Texas Public Radio summarised the situation in February: both sides are running out of water.

Power. The grid constraint in the Technological cell is also an environmental one. A plant that cannot get a clean-energy contract and runs on backup generation has a worse emissions profile than one that can. That matters more each year for any supplier inside a US or European customer's Scope 3 reporting.

Score: adverse, and slowly worsening. The water obligation is now a treaty commitment to a neighbour, not only a domestic policy choice.

Legal — three big changes since 2024

Mexico has changed more of its business law in two years than in the previous ten:

Change	In force	What it does	Who it costs
Judicial reform	Enacted 15 Sept 2024; first full year of elected judges 2026	Federal judges, including the Supreme Court, are elected; a 2026 amendment pushed the next judicial elections to 4 June 2028 and added mandatory professional exams	Firms with contracts already exposed to Mexican courts
40-hour week	Constitution amended Mar 2026; law in force 1 May 2026; phase-in 2027-2030	48 → 40 hours, no pay cut; overtime capped; electronic time records	Employers with existing 48-hour shift patterns
Tariffs on non-FTA imports	1 Jan 2026	5-50% on 1,463 tariff lines in 13+ sectors from China, India, South Korea, Thailand, Indonesia and others	Manufacturers with Asian input chains

Outside Mexico's own statute book, the US Section 232 auto tariff is the legal fact with the biggest effect on volume: Mexico's auto exports to the US fell 11.3% in Q1 2026, and passenger-car exports fell 22%.

Jones Day's advice to foreign investors on the judicial reform is the clearest sign of how the Legal cell splits. Investors should check whether their corporate structure qualifies for protection under an investment treaty, and restructure "before any alleged breach occurs or becomes foreseeable." That advice is easy to follow for a firm that has not yet invested and much harder for one that already has.

Score: adverse for incumbents, who must absorb all three changes. Manageable for entrants, who can design around all three from day one.

PESTEL scorecard for Mexico, September 2026

Force	Incumbent (already operating)	Entrant (deciding whether to come)	The fact that sets it
Political	Favourable: USMCA in force, preferences intact	Adverse: every year of payback runs through a review	USTR, 1 July 2026: "the USMCA is not renewed"
Economic	Favourable: reinvesting at a record pace	Neutral-to-adverse: new investment −13.4%	FDI H1: 88.5% reinvested earnings
Social	Neutral, adverse long-run	Neutral, adverse long-run	Fertility 1.2; 40-hour week by 2030
Technological	Favourable: connected	Adverse: queue for megawatts	279 MW of data centres, 72% in one state
Environmental	Adverse	Adverse	~50% short on the last water-treaty cycle
Legal	Adverse: three changes to absorb	Neutral: three changes to design around	Elected judges, 40-hour week, 5-50% Asia tariffs

Four of the six rows score differently depending on the column, and they do not all point the same way. That is what the next section explains.

The Entrant Test

Standard PESTEL assumes the reader is a single, timeless decision-maker. Mexico in 2026 shows why that assumption matters: firms that committed before July 2026 and firms deciding after it read the same six forces and reach different answers. The investment data shows them doing so.

The Entrant Test: score every PESTEL cell twice, once for a firm already operating in the country (the incumbent) and once for a firm deciding whether to arrive (the entrant). Then sort the cells where the two readings differ by one question: can an entrant design around this cell before it commits? Cells an entrant can design around are a cost of arriving. Cells it cannot design around are a reason to wait, and if the binding one has a scheduled date, waiting until that date is cheap.

Applied to Mexico:

Cell	Incumbent	Entrant	Can an entrant design around it?	Set by
Legal: judicial reform	Adverse	Neutral	Yes: structure through an investment treaty before any dispute is foreseeable	Mexico
Legal: 40-hour week	Adverse (re-roster existing shifts)	Neutral (staff for 40 from day one)	Yes	Mexico
Legal: 5-50% Asia tariffs	Adverse if inputs are Asian	Favourable: competitors with Asian inputs pay more	Yes: source in North America	Mexico
Political: USMCA annual review	Favourable (capital already sunk)	Adverse	No: depends on Washington, every 1 July	United States
Technological: grid connection	Favourable (connected)	Adverse	Only partly: backup generation at a cost	CFE build-out

Two findings come out of this table.

First, the cells that favour entrants are all Mexican; the cells that favour incumbents are American or infrastructural. Each of Mexico's own reforms can be planned for in a new site: a treaty-protected holding structure, a 40-hour shift design, a North American supply chain. None of them can be planned around once the plant exists. Mexico has spent two years changing exactly the laws an entrant can absorb most easily. The cell no entrant can design around is set in Washington.

Second, the binding cell has a date. A firm that waits until 1 July 2027 gives up one year of production and learns something it cannot learn any other way: whether the second annual review produces renewal, new conditions, or another refusal. That is the logic the H1 data shows. Incumbents, whose capital is already committed, reinvested US$30.9 billion. Entrants, for whom waiting is cheap, put in US$2.7 billion, down 13.4%. On this reading the decline in new investment is a sensible response to how the annual review is scheduled, not a loss of confidence in Mexico.

This differs from the Two-Ledger Test used for China, which splits the scan by which side of the economy a decision touches (production or household). It also differs from the Imported-Cell Test used for India, which tags cells by who can move them. The Entrant Test splits by when you committed. A cell can be imported and still identical for incumbents and entrants (Mexico's water obligation is), or domestic and still read in opposite directions by the two (the 40-hour week).

One prediction, written down before the data

The test makes a checkable claim, so it is recorded here with the date it will be scored:

Placement	Mexico, 2026: favourable to incumbents, adverse to entrants, until a renewal signal from the annual USMCA review
Prediction	Full-year 2026 new investment comes in below full-year 2025's, while reinvested earnings keep total FDI near record levels
Falsifier	Full-year 2026 new investment at or above the 2025 figure without any renewal signal, which would mean entrants are not treating the review date as a reason to wait
Scored on	The Economy Ministry's full-year 2026 FDI release (expected February 2027), then the 1 July 2027 review

Counter-argument: isn't reinvested earnings always most of Mexico's FDI?

The strongest objection is that reinvested earnings have dominated Mexico's FDI for years, because a large stock of foreign-owned plants generates a large flow of profit. If 88.5% is roughly normal, the composition says nothing about 2026.

The objection is right about the share and wrong about what the test reads. The Entrant Test does not depend on the share. It depends on the change in new investment, which fell 13.4% year on year, and on the fact that Banxico's independent net-FDI measure fell 11.7% over the same months. A high reinvestment share would explain a record total. It does not explain new money falling.

The honest limit is the one this site applies to any single-window reading: one half-year is not a trend. H1 2026 includes the run-up to the 1 July review, when waiting was obviously sensible. If new investment recovers in the second half without a renewal signal, the test's prediction fails, and the table above says in advance what failure would look like.

Key takeaway

On headline numbers, Mexico in September 2026 looks like a country that won nearshoring: record FDI, the best quarter since 2020, inflation inside the target band, a budget that finally reduces support for Pemex. The scan changes once you ask who those numbers describe. The record is incumbents reinvesting. New arrivals fell 13.4%. The reason is not anything Mexico changed. It is the one cell no entrant can design around, which has been reviewed annually in Washington since 1 July 2026.

If you already operate in Mexico, the environment is favourable and the main risk is legal: the 40-hour phase-in from 2027 and the judicial reform apply to you whether you planned for them or not. If you are deciding whether to come, every Mexican-made change can be designed around, and the US-made one has a date. The decision is less "Mexico or not" than "now or after 1 July 2027." The answer depends on whether a year of lost production costs more than a year of review risk.

The three releases that will update this analysis cheapest are Q3 GDP (late October, which will show how much of Q2 was the World Cup), the full-year 2026 FDI composition (February 2027, which scores the prediction above), and the second annual USMCA review on 1 July 2027.

Want to run a PESTEL on your own market? Framework for iPhone & iPad ships a PESTEL worksheet with all six forces and AI assistance for the scan.

Want to go deeper

• PESTEL framework: the six forces, and how to score them
• How to do a PESTEL analysis: the step-by-step method
• PESTEL analysis of the USA (2026): the other side of the annual review, and the Load-Bearing Force Test
• PESTEL analysis of Canada (2026): the third USMCA economy, one political force rewriting the other five, and the Reversal Test
• PESTEL analysis of China (2026): the supplier Mexico's 50% tariffs are aimed at, and the Two-Ledger Test
• PESTEL analysis of India (2026): also caught by Mexico's non-FTA tariffs, and the Imported-Cell Test
• World Cup 2026 PESTEL analysis: the June scan that called Mexico the tournament's relative winner
• PESTEL analysis of Germany (2026): a different reason to wait for a date, and the Two-Clock Score
• PESTEL vs Porter's Five Forces: when the macro scan is the wrong tool

Cover photo: Juan Pablo Lara on Unsplash.