Spanish capital is not leaving Mexican generation: it is changing hands — and risk profile

Two sales, one seller

In twenty-six months Iberdrola sold its entire Mexican generation business. On 26 February 2024 it closed the sale of 13 plants — 8,539 MW, mostly combined-cycle — to a trust of México Infrastructure Partners (MIP) backed by Fonadin and other Mexican public and private financial institutions, for US$6.2bn; those assets contributed 55% of the company's EBITDA in the country FACT. On 24 April 2026 it closed the second transaction: the remaining 2,600 MW — 1,368 MW of combined-cycle and cogeneration plants and 1,232 MW of wind and solar — together with the retail business serving more than 500 large customers and a development pipeline of around 12,000 MW, went to Cox for US$4.0bn, US$200m below the US$4.2bn enterprise value announced on 31 July 2025 FACT. Neither company's closing statement explains the difference FACT.

The political reading of the first sale is well known: the Mexican state bought back thermal generation with public money. The second is more interesting for anyone who thinks in terms of capital, because the buyer is neither the state nor another large utility but a mid-sized Spanish group whose first-half 2025 EBITDA of roughly €82m was a fraction of what the acquired business generates. On Cox's own numbers, the business would have contributed pro forma 2025 revenues of €2,551m and EBITDA of €786m FACT. Iberdrola, for its part, framed the exit as a decision to prioritise "regulated network businesses or generation with long-term contracts" in the United States and the United Kingdom FACT.

This is the second instalment of a series that opened with the record in pipeline gas imports from Texas: that gas becomes electricity in plants like these, and who owns them and how they are financed matters as much as the price of the fuel.

The framework that changed between the two sales

Between the two transactions, the Constitution changed. The decree published in the Diario Oficial de la Federación on 31 October 2024 reformed Articles 25, 27 and 28 and established that private participants in electricity "shall in no case prevail over the state public enterprise" FACT. The Electricity Sector Law (Ley del Sector Eléctrico), published on 18 March 2025, turned that principle into a number: the state must retain at least 54% of the average energy injected into the grid in a calendar year, and in mixed-investment projects the public enterprise must hold a direct or indirect stake of at least that same 54% FACT.

What the market feared — a closed door — did not happen. On 7 June 2026 CFE awarded its first mixed-development tender: 7,411 MW across 37 projects from 31 developers, 114% of the 6,500 MW requested, with 46 lots declared void; the response was almost entirely solar (6,710 MW against 700 MW of wind) FACT. The conclusion is not that the reform is irrelevant. It is that the resulting framework — CFE as majority partner, private developers as minority partners or contracted generators — is known and bankable, and private capital has responded to it INTERPRETATION.

Cox therefore bought a business whose value lies not in building merchant capacity, as in the 2010s, but in operating 2,600 MW already in service, supplying more than 25% of the qualified-user market and holding a project pipeline that will only materialise inside the mixed scheme or under contract INTERPRETATION.

From utility balance sheet to leveraged structure

This is the substantive change. When those 2,600 MW belonged to Iberdrola, they were financed by a group earning more than €4.3bn in a half-year; Mexico risk was diluted in a global balance sheet. Cox funded the purchase with a US$2.65bn syndicated loan from seven banks, signed on 26 January 2026, and on 6 May refinanced two-thirds of that bridge with its debut issue in the U.S. market: US$2.0bn in two tranches, five-year at 7.125% and ten-year at 7.75%, on reported demand of US$8.0bn FACT. The remaining third sits in a term loan of roughly five years FACT.

The first results with Mexico consolidated, published on 28 July 2026, give the scale: EBITDA of €245m for the half-year against around €82m a year earlier; a net loss of €4.67m on extraordinary financing costs of the acquisition; net debt of about €3.3bn, 4.9x EBITDA; and, for the Mexican business, revenues of US$870m (+24%) and adjusted EBITDA of US$302m (+7%) FACT. These are the company's published figures; this article passes no judgement on them beyond noting that they describe a different risk profile from the one the asset carried a year ago.

Data point. Same asset, two financiers. Under Iberdrola: a utility balance sheet with €7,010m of half-year investment. Under Cox: a US$2.65bn syndicated loan, a US$2.0bn bond at 7.125–7.75% and net debt of 4.9x EBITDA at June 2026.

In 2025 Cox announced a plan of more than US$10.7bn in Mexico to 2030 — the acquisition, over US$4.0bn in new energy assets and up to US$1.5bn in water FACT. At a dollar cost of debt of 7–8%, that plan depends on cash generation from the Mexican business and a cheaper refinancing arriving before the investment commitments do FORECAST. That is not a prediction about the company; it is the arithmetic of any leveraged buyer.

What changes for capital

For the investor in Mexican private generation, the precedent cuts both ways. Iberdrola booked a net capital gain of €953m on the transaction and confirmed that an operating Mexican asset has a buyer and a price even after the reform FACT. But the multiple was paid by a mid-sized group using debt, not by a large utility using cash: the exit price of a Mexican asset is now set by whoever is willing to lever up to buy it OPINION.

For the lender, the signal is that the U.S. bond market accepted leveraged Mexican generation risk at 7.125% for five years, oversubscribed, barely two weeks after closing FACT. That is the reference cost for the next transaction of comparable size — and, in all likelihood, above the average cost at which an investment-grade utility financed the same plants INTERPRETATION.

For the industrial offtaker — the qualified user who contracted with Iberdrola México and now has Cox as counterparty — the contract has not changed, but the credit quality standing behind it has. It is worth re-reading the guarantees, the change-of-control provisions and the replacement-supply clauses; not because a problem exists, but because the counterparty of 2027 is not the counterparty of 2024 INTERPRETATION.

The Spain–Mexico cross-border reading

Does Iberdrola's sale mean Spanish capital is withdrawing from Mexico? The data say no. Spain was Mexico's second-largest foreign investor in 2025, with US$4,431m, 10.8% of the total; in the first half of 2026 it held that position with US$4,954m, 14.2%, although 16.6% below the same period a year earlier FACT. Reinvested earnings accounted for 88.5% of the half-year's foreign investment, not new capital: foreign money in Mexico is, above all, money that was already there FACT.

The rotation is sectoral, not geographic. Iberdrola closed the half-year with net profit of €4,336m (+21.7%), investment of €7,010m and €4,384m in networks, up 42% FACT. The capital leaving Mexican generation is going into regulated networks in the United States and the United Kingdom; the capital entering Mexican generation is Spanish capital of a different size, structure and cost. For an adviser working between Madrid and Mexico City, the relevant question is no longer whether Spain invests in Mexican energy, but with which balance sheet OPINION.

That is the thesis: private investment in Mexican generation to 2030 will be defined not by the text of the reform, which is known and has already been tested in one tender, but by the cost and structure of the capital willing to enter under that framework OPINION. It is a question of private capital more than of regulation, and the next instalment of this series — the real balance of Pemex's mixed contracts — shows that in hydrocarbons the equation is the same. More on our practice in Madrid–Mexico advisory.