Why does Mexico not export beef or chicken to Europe?

The Modernised Global Agreement covers political dialogue, cooperation, trade and investment. It is a mixed agreement, so it requires ratification by all 27 member states as well as by Mexico. The Interim Trade Agreement covers only what falls under the European Union's exclusive competence, which is why it does not need to pass through 27 parliaments.

On the European side the interim agreement's procedure is complete. The European Parliament gave its consent on 8 July 2026 and the Council adopted the conclusion decision on 14 July. The text was published in the Official Journal on 31 July, running to 975 pages.

Its Article 33.9 provides that the agreement enters into force on the first day of the second month after both parties notify each other that their internal procedures are complete. The Union has notified. Mexico has not. Put precisely: the only outstanding step anywhere in the world for the trade agreement to enter into force is approval by the Mexican Senate.

Preference utilisation measures how much of the eligible value actually took advantage of it. It reached 68.7% in 2008, held near that level for a decade, and fell to 48.3% in 2025. In absolute terms, 5.9 billion euros of eligible goods paid tariffs unnecessarily.

The sector breakdown explains the mechanism. To claim the preference you must prove the origin of the goods, which means documenting where every input came from. In products assembled from components from many countries, that paperwork costs more than the tariff it avoids, so the rational exporter pays and moves on.

This qualifies the thesis rather than contradicting it. The barrier is not the tariff, but neither is it sanitary in this stretch: it is administrative. And a new agreement that does not simplify rules of origin will reproduce the same outcome.

For 2025, the European Union reports importing 38.2 billion dollars from Mexico. Mexico reports exporting 19.3 billion dollars to the European Union. The gap is 18.9 billion dollars, a factor of 1.98x.

Part of it is method: the Union values its imports including insurance and freight, while Mexico values its exports without them. Another part is a recording artefact, crude oil leaving Mexico declared to the continent 'Europe' rather than to a country. But two thirds of the gap is neither of those.

The rest follows from a different criterion. Mexico assigns the trading partner by the destination declared on the export document; the European Union assigns it by the origin of the goods. When a Mexican product leaves for the United States and ends up consumed in Europe, Mexico counts it as a sale to the United States and Europe counts it as a purchase from Mexico. Both administrations are right within their own framework.

The proof is in the detail. Germany, where goods travel directly, matches within four percent. The Netherlands, a port of entry, differs by 4.6 times. And in medical and surgical instruments, the Union reports importing from Mexico 28 times what Mexico says it sold, while declaring an enormous volume of that same code to the United States.

Regulation (EU) 2023/1115 bans the sale on the European market of seven commodities, among them coffee, cocoa, rubber and wood, if they come from land deforested after 31 December 2020, and requires declaring the geolocation of every plot of origin. After two postponements it applies from 30 December 2026 to large and medium companies, and from 30 June 2027 to small ones.

Mexico ends up at standard risk, and it is worth saying how. Implementing Regulation (EU) 2025/1093 published the list on 22 May 2025, but that list names only the 140 low-risk countries and the 4 high-risk ones. The rule provides that the standard level is maintained for every country not listed. That is, Mexico was not assessed and put in a middle box: it simply is not among the low-risk ones, where Chile, Costa Rica, Uruguay, the United States, Canada, the Dominican Republic and Cuba are.

The difference between being on the list and not is not paid by the Mexican exporter but by its buyer. Article 13 exempts from assessing and mitigating risk anyone sourcing from a low-risk country; not so with Mexico. And Article 16 requires member states to check every year at least 3 per cent of the operators sourcing from a standard-risk country, against 1 per cent for low risk. Buying Mexican coffee costs more in paperwork than buying Costa Rican, for reasons that have nothing to do with the coffee.

How much it weighs, counted line by line on the same Comext file. In 2025, between 0.83% and 1.32% of the value the European Union bought from Mexico falls within the scope of the rule. Against total bilateral trade that is little, and it should be said. Against agri-food trade it is not: it is between 19% and 30.3% of what Mexico sells to Europe in chapters 01 to 24.

The range exists because Annex I lists some headings in full and others prefixed with 'ex', meaning only the part that contains the commodity. Counting those in full overstates and discarding them understates, so both figures go here and never just one.

The bulk is coffee: 171 million euros in 2025, more than half of everything exposed. And there the circle closes with the geographical indications. Two of the 20 the agreement protects for Mexico are Café Chiapas and Café Veracruz. The trade chapter gives Mexican coffee a protected name in Europe, and seven months later another European rule asks it for the coordinates of the plot.

The first layer is exports by state. In 2025, Chihuahua leads with 109.5 billion dollars, followed by Coahuila de Zaragoza and Nuevo León. Mexico's export geography is northern and Bajío, and it answers to the automotive and electronics chain, not to agriculture.

The second layer is agricultural production, which is a different thing and must not be added to the first. In 2025 the value of national agricultural production was 812.2 billion pesos, spread across 50 crops. A state can produce a lot and export little, or the reverse: producing is not selling abroad.

The third layer does not exist. No official Mexican source crosses federal entity with destination country under the criterion of where the good was produced. State exports carry no country; the trade balance carries country but its coverage is national in one hundred percent of records. On the European side the gap is symmetrical: the European customs record does not keep the partner's subnational origin.

There is a source that does cross state with country, but it measures something else: it attributes the sale to the exporting company's fiscal address, not to the place of production. The difference is measurable. Mexico City accounts for 16.4 percent of national exports under that criterion, and 0.5 percent under the productive-origin criterion. That is why this report publishes no figure for state exports to Europe.