The Ranking Moved.
Did Mexico?

Mexico looks different from a distance

From Canada, Mexico can easily look like a single economic idea: one country, one market, one destination for investment, manufacturing, tourism or expansion.

Up close, that picture begins to break apart.

A project landing in northern Mexico does not encounter the same labour market, infrastructure, security conditions or productive ecosystem as one established in the centre of the country or on the Yucatán Peninsula. The same national border contains very different local realities.

That changes the question. Before asking which Mexican state ranks highest, it is worth asking what someone is actually trying to do there.

The question comes before the ranking

A Canadian manufacturer looking for a location may care about formal labour, wages, skills, exports, logistics and energy. A tourism project may care about air connectivity, the size of the tourism economy, labour availability and security. A technology operation may care more about higher education, internet access, productive complexity and talent.

None of those questions produces exactly the same map.

So how do you compare 32 different environments without reducing Mexico to a national average?

There is a remarkably detailed source that lets us begin doing exactly that.

We went into the data

For two decades, the Instituto Mexicano para la Competitividad (IMCO) has compared the conditions under which Mexico's states compete. Its 2026 State Competitiveness Index — the Índice de Competitividad Estatal (ICE) — evaluates all 32 entities through 53 variables grouped into six subindices.

IMCO defines competitiveness as the capacity to generate, attract and retain talent and investment. That makes the dataset useful for much more than identifying who finished first.

We were less interested in the ranking than in what was happening underneath it.

So we began changing the question.

Change the question and the map changes

What follows is not a new index. It is a way of reading the existing evidence. Each question pulls a different set of variables to the foreground.

These variables do not make an investment decision, a travel decision or a location decision on their own. They do something more useful at this stage: they reveal which questions deserve a deeper comparison.

This is where the dataset becomes more interesting than the headline. A reader interested in talent may stop at higher education. A manufacturer may move immediately to informality, wages, exports and energy. A tourism investor may connect air traffic with tourism GDP. Someone studying workforce inclusion may ask why female participation and the gender income gap do not necessarily move together.

One dataset can produce many different investigations. The useful comparison begins when the question becomes specific.

And then we noticed something else

We could have stopped there and compared states one question at a time.

But while moving through the data, another pattern appeared.

The ranking itself had moved. A lot.

At first, the explanation seems obvious

Quintana Roo moved up twelve places. Nayarit gained eight. Campeche gained six. Yucatán gained three. At the other end, Baja California lost six positions, while Chihuahua, Hidalgo and Sinaloa each fell seven.

Read quickly, the explanation seems obvious: some states improved and others deteriorated.

Except that rankings do not work that way.

The national movement becomes more revealing when placed beside IMCO’s own description of what changed underneath it.

IMCO reports that 26 states increased the share of adults with higher education and 30 improved average years of schooling. Yet only five states improved growth in employment registered with Mexico’s social-security system. Average state-level formal job growth moved from +0.4% to −0.9%, while labour informality remained at 54.6%.

That is not a minor detail. It means a state could improve educational capacity and still struggle to convert that capacity into formal employment. The variables are moving in different directions.

The same tension appears in security. IMCO reports that 29 states reduced homicide rates, yet only 27.4% of adults felt safe and the national “dark figure” — crimes not reported or not investigated — remained extremely high at 92.9%.

This is where the ranking becomes useful again — not as an answer, but as a clue.

The three traps hidden inside a ranking

1. Improvement is not the same as relative movement

If a state improves its own indicator but other states improve faster, it can lose places. The ranking records the loss; the underlying series records the improvement. Both are true.

2. Stability can be an achievement — but only in context

A state that remains almost unchanged while peers deteriorate may gain relative ground. That can reasonably be described as resilience. But stability alone is not resilience. We need evidence that the surrounding environment actually weakened.

3. A gain can contain deterioration

A state can move upward even while some of its own indicators worsen. That does not make the ranking wrong. It tells us that competitiveness is comparative and multidimensional.

Look at the states that broke the pattern

Quintana Roo’s twelve-place rise is the most visible movement on the board. Nayarit gained eight and Campeche six. IMCO separately identifies those three states as the largest advances and reports that all three increased their economic-complexity measure in innovation sectors by between 12.7 and 26.9 points.

That is stronger evidence than rank movement alone: it identifies at least one underlying capability that moved with the ranking. But it still does not mean every component improved, or that the same mechanism explains all three states.

Now look in the other direction. Chihuahua, Hidalgo and Sinaloa each lost seven positions. Baja California lost six. Those changes are large enough to demand the opposite question: did the states deteriorate, or did the field move around them?

The boletas — the state scorecards published with the ICE — are where that question can be investigated. They place the overall ranking beside the underlying indicators and make visible a distinction the headline ranking cannot resolve on its own.

Yucatán shows why the distinction matters

Yucatán is not the largest mover in 2026. It gained three positions, from tenth to seventh. Its recent path is even more revealing: 8 → 10 → 7 from 2024 through 2026.

That is not a staircase. It is a fall followed by a recovery.

And when its 53 indicators are read against their own prior values, the internal picture is mixed: some improve, some deteriorate and some barely move. In our review of the Yucatán scorecard, 27 indicators moved in a favourable direction, 19 moved unfavourably and seven were essentially unchanged at the precision shown in the source.

That makes Yucatán useful not because it “proves” the national story, but because it exposes the mechanism. A state can contain genuine improvement and deterioration at the same time — and still finish higher in a relative ranking.

The useful question is no longer “Who won?”

For an investor, strategist or project team looking at Mexico from Canada, the temptation is understandable: start at number one and work downward.

But the 2026 movement suggests a better sequence.

First ask what the project needs. Then identify which ICE dimensions matter most. Then examine the state’s underlying indicators and their direction of travel. Finally, use the overall rank as context rather than as the decision itself.

A manufacturing project may care intensely about formal labour, skills, logistics, exports and productive complexity. A technology operation may place more weight on advanced talent, connectivity and innovation. Tourism may make air connectivity, security, labour conditions and the existing tourism economy more relevant.

The index does not eliminate due diligence. It tells you where due diligence should begin.

And that is the point of going beneath the headline ranking. Tourism does not produce the same map as skilled labour. Skilled labour does not produce the same map as infrastructure. Infrastructure does not produce the same map as institutional conditions.

What are you trying to understand about Mexico?