Spain in Latin America: The Thirty-Year Cycle From Conquest to Retreat

No European country bet more of its corporate balance sheet on Latin America than Spain, and none is now unwinding that bet more visibly. Between 1993 and 2001, Telefónica, Santander, BBVA, Repsol, Endesa, Iberdrola and Iberia bought controlling stakes in the utilities and banks that Argentina, Brazil, Chile, Peru and Mexico were selling, and Spain became, in less than a decade, the region’s second-largest investor. In 2025 and 2026 the same companies sold Argentine, Peruvian, Chilean, Colombian and Mexican businesses at fractions of what they once paid. This case study follows the full arc. It is part of the Spain Company Stories hub.

Why did Spanish companies go to Latin America in the first place?

Because the assets were for sale and Spanish companies, newly privatised themselves, needed somewhere to grow. Latin American governments sold telephone monopolies, state banks and power utilities in the 1990s to fund deficits, and Spanish buyers spoke the language, had the regulatory experience and faced almost no domestic growth.

Telefónica bought into Chile and Argentina in 1990 and Peru in 1994, paying around $2bn for the Peruvian monopoly alone. Endesa took control of Enersis in Chile in 1999, Repsol bought YPF the same year for roughly $15bn, and Santander and BBVA bought Banespa, Bancomer and a dozen smaller banks between 1997 and 2001. Iberia had already bought Aerolíneas Argentinas in 1990, a purchase that lost it money for a decade.

The logic was simple. Spanish companies were being pushed by the state into competition after decades of protection; the Spanish domestic market was mature; and Latin America offered under-penetrated demand for telephones, credit and electricity in countries where a Spanish manager could arrive on Monday and run the business by Friday. The Banco de España estimates that services — banking, telecoms and utilities — still account for the bulk of the Spanish investment stock in the region.

How large did the exposure become?

By the early 2000s Spain had accumulated a stock of Latin American direct investment that no other European country matched, and the region represented more than a third of the Spanish corporate sector’s foreign assets. The Banco de España’s 2025 review puts the figure at over 30% of total outbound FDI stock, concentrated in Brazil and Mexico.

The concentration inside individual companies was even greater. At its peak, Latin America generated close to half of Telefónica’s revenue, roughly half of BBVA’s profit, and a similar share at Santander once Brazil was added in 2007. Endesa derived a large part of its EBITDA from Enersis; Repsol’s reserves were majority Argentine after YPF.

That exposure was profitable for a time. Argentina, Brazil and Mexico grew faster than Spain through the commodity boom of 2003–2012, and the Spanish parents booked the growth. What they did not book, until the currencies moved, was the cost of holding assets in pesos and reais while paying dividends in euros.

What did the 2001 Argentine crisis teach Spanish boards?

That a contract written in dollars in Buenos Aires is not a dollar contract. The 2002 pesification converted dollar-denominated tariffs and deposits into devalued pesos overnight, and Spanish utilities and banks absorbed several billion euros of losses on assets they had bought only two or three years earlier.

Telefónica, Endesa, Repsol and the two banks all wrote down Argentine positions. BBVA had merged its Argentine unit into Banco Francés; Santander owned Banco Río; both stayed but shrank. Iberia had already sold most of Aerolíneas in 2001 to the Marsans group after losing an estimated €2bn on it over the decade.

The lesson that stuck was about tariffs rather than exit. Spanish groups learned to hedge the currency, to avoid dollar-linked tariffs that governments could unilaterally break, and to prefer Mexico, Chile, Brazil and Peru over Argentina. What they did not do was leave. Argentina recovered, the commodity boom arrived, and by 2008 the region was again the growth engine.

When did the cycle peak?

Around 2010–2013, when Latin America was generating record euro profits for Spanish groups while Spain itself was in recession. In those years the region was not a diversification; it was the reason Santander, BBVA and Telefónica stayed solvent and paid dividends while their domestic businesses lost money.

Santander’s Brazilian bank, floated in 2009, was by 2011 the largest single contributor to group profit. BBVA’s Bancomer contributed around a third of group earnings even before the Spanish real estate losses were counted. Telefónica’s Latin American revenue passed that of Spain, and the company’s Brazilian acquisition of GVT in 2015 for €7.5bn was the last large Spanish purchase in the region.

The peak coincided with a peak in the currencies. The Brazilian real traded near 2.2 to the euro in 2011; by 2020 it was over 6. The Mexican peso went from around 16 per euro to more than 24 at the worst point of the pandemic. Argentina’s peso became unquotable. Every one of those moves subtracted from euro-denominated profit while the peso-denominated businesses were still growing.

Why did the 2020s turn into a retreat?

Because three things changed at once: the currencies stopped recovering, European regulators started charging capital for emerging-market risk, and investors began valuing Spanish stocks on their European earnings only. A Latin American profit that arrived in euros at a 40% discount to five years earlier, and carried a higher capital charge, was no longer worth the balance sheet it occupied.

Telefónica grouped its Spanish-speaking American units into a “Hispam” division in 2019 and declared them non-core. Under Marc Murtra, appointed in January 2025 after the state re-entered the shareholder register, the sale became total: Argentina to Telecom Argentina for $1.245bn, Peru to Integra Tec for a nominal sum after the unit entered insolvency, Uruguay, Ecuador and Colombia to Millicom, Chile to NJJ and Millicom for up to $1.2bn in July 2026, and Mexico to a consortium for $450m. The company booked €2.27bn of Hispam-related losses in 2025 and a €4.3bn net loss overall. The full story of that shareholder-driven reversal is in our Telefónica case study.

The utilities moved the same way. Endesa had already transferred its Latin American holdings to its Italian parent Enel in 2014 for €8.25bn. Iberdrola sold 55% of its Mexican business to a state-backed trust for $6.2bn in February 2024 and the remainder to Cox for $4.2bn in a deal signed in August 2025 and closed in April 2026, redeploying the proceeds into US and UK grids — a pivot we detail in Iberdrola’s Anglosphere strategy.

Why have the banks stayed when the telecoms and utilities left?

Because banking in Mexico and Brazil is still extraordinarily profitable, and the banks never depended on tariffs a government could rewrite. A regulated utility earns what the regulator allows; a bank in a concentrated market with 20% returns on equity earns what its customers pay.

BBVA México produced €5.26bn of net attributable profit in 2025, roughly half of the group’s record €10.5bn. Santander earned a record €14.1bn in 2025 with Brazil, Mexico, Chile and Argentina together contributing something like a third. Neither bank can replace those earnings in Europe; BBVA’s failed €16bn bid for Sabadell in 2025 was, in large part, an attempt to dilute its Mexican dependence, as our analysis of BBVA México as the group’s profit engine explains.

Santander has pruned rather than retreated. It sold its Polish bank to Erste in 2025, is folding TSB into a UK exit, and has concentrated Latin American capital on Brazil and Mexico while keeping Chile and a small Argentine franchise. The strategy, described in our account of Santander’s global model, is to run the region as a portfolio of retail and consumer platforms rather than a set of national universal banks.

How did the profits compare with the capital deployed?

Unevenly. On a thirty-year view, the banks earned their money back several times over; the telecoms and utilities did not, once currency and expropriation losses are counted. Telefónica invested well over €40bn in Latin America between 1990 and 2015 and sold its Hispanic American operations for a combined sum in the low single-digit billions.

Repsol paid around $15bn for YPF in 1999, lost 51% of it to expropriation in 2012 and settled for $5bn in Argentine bonds in 2014 — a case we examine in Repsol, YPF and the sovereign-risk lesson. Endesa recovered its Enersis investment only when Enel bought it out. Iberia never recovered its Aerolíneas losses.

The banks are the exception because their Latin American businesses compounded: BBVA paid roughly €5bn in total for Bancomer across 2000 and 2004 and has since extracted well over ten times that in dividends. Santander’s Brazilian purchase price of around €10bn in 2007 has been repaid many times. The difference is not geography; it is the durability of a franchise that governments regulate but do not price.

What does “refocus” mean for the next decade?

It means Spanish groups keeping Brazil and Mexico, treating everything else as optional, and moving fresh capital to the United States and the United Kingdom instead. The old logic of language and proximity has been replaced by the logic of regulated returns and currency stability.

Iberdrola’s 2025–2028 plan allocates €37bn of €58bn to the UK and US; Brazil gets €7bn and the rest of Latin America nothing. ACS gets 63% of its revenue from America, most of it from the United States. Ferrovial moved its headquarters to Amsterdam and its listing to New York to get closer to US infrastructure capital. Santander is expanding in the US through Openbank and consumer finance while selling in Poland.

The reverse flow, meanwhile, has become significant: Latin American families and companies hold roughly €67bn of direct investment in Spain, according to ICEX, and Mexican capital alone accounts for close to half of it. Madrid has become the place where Latin American money is parked, a phenomenon covered in our article on Latin American capital buying Spain.

Was the thirty-year bet a mistake?

Not in aggregate, but the returns were narrower and more concentrated than the strategy implied. Spain got two world-class banks with genuinely global franchises; it also got a telecoms group that spent a generation building operations it sold for less than a year’s capital expenditure.

The honest scorecard is that Mexico and Brazil paid for everything else. Argentina destroyed capital in every sector on every occasion; Venezuela was written to zero; Peru, Chile and Colombia were profitable but small; and the currency losses across the region since 2012 almost certainly exceed the cumulative dividends from the non-bank businesses.

The strategic lesson Spanish boards have drawn is not to avoid emerging markets but to price them properly: to hold only businesses whose local returns are high enough to survive a halving of the currency, and to keep the balance-sheet weight of any single country below what a bad election could destroy. That is the logic behind the 2020s refocus, and it is why Spain’s next thirty years of foreign investment will look more like Iberdrola in Ohio than Telefónica in Lima.

Frequently Asked Questions