Ternium: Latin America's Steel Empire Built from Post-War Vision

I. Introduction & Episode Roadmap

On a stretch of flat scrubland outside Monterrey, in the municipality of Pesquería, a 300-tonne electric arc furnace is being assembled next to the rolling mills it was designed to feed.¹ When it starts melting steel, it will close a loop that has taken Ternium more than a decade and roughly $7.5 billion to draw: iron ore pellets arriving at a dedicated port facility, natural gas stripping the oxygen out of them, scrap and iron melted into liquid steel, slabs rolled into coils, and coils pickled, cold-rolled and galvanized into the skin of North American cars, all on one site.²³

The company first said the furnace would run in the first half of 2026.⁴ On the call in August 2026, the chief executive said start-up was expected in early 2027.⁵ That small slip in the calendar is a good place to begin, because it captures the whole company: enormous ambition, long horizons, and a habit of discovering that the world does not move on the timetable a steelmaker would like.

Ternium is Latin America's largest flat-steel producer, and it is very hard to describe it without describing a family. It makes the flat and long steel that goes into cars, refrigerators, buildings and pipelines from production centers in Mexico, Brazil, Argentina, Colombia, Guatemala and the United States.⁶ In 2025 it shipped about 15 million tons of steel and booked about $15.6 billion of revenue.⁷

Roughly half of those tons went to Mexican customers, a quarter to Brazil, and most of the rest to Argentina and its neighbours.⁷ Yet the company is controlled, through a chain of holding companies, by the heirs of an Italian engineer who arrived in Buenos Aires after the Second World War with a telex code for a company name and a plan to industrialize a continent.⁸⁹

The stock market values Ternium's equity at about $10.6 billion, which works out to roughly 15 times the last twelve months of earnings and a little less than the accounting value of its equity.¹⁰ That is a cheap-looking price for Latin America's steel champion, and the cheapness is itself the puzzle. Is the market missing a nearshoring winner at the bottom of a steel cycle, or is it correctly pricing a business whose profits flow through minority shareholders, related-party contractors and political risk before they reach anyone holding an American depositary share?

This story follows Ternium through five turning points that each answer part of that question. There is the privatization wave of the 1990s, when the Rocca family's Techint group bought state steel mills in Argentina and Venezuela and then a Mexican pioneer of gas-based ironmaking. There is the 2008 seizure of its Venezuelan mill by Hugo Chávez. There is a costly, litigated foothold in Brazil's Usiminas that began in 2012 and is still in court today.

There is the 2017 purchase of a giant Brazilian slab plant from a German conglomerate that had lost around €8 billion trying to make it work. And there is Pesquería, the bet that the United States–Mexico–Canada Agreement will reward whoever melts steel inside North America. The themes running underneath are privatization arbitrage, family capitalism, the promise and the politics of nearshoring, and what it actually costs to operate a heavy industrial business in countries where the government can change the rules overnight. The story starts, as Ternium's does, in wartime Italy.

Subscribe on LinkedIn

↑ Back to Top

II. Techint Origins: The Rocca Family & Post-War Industrial Vision (1945-1990s)

In November 1943, with the Allies pushing up the Italian peninsula and Mussolini installed as a German puppet at Salò, an engineer named Agostino Rocca was offered the job of minister of war production in the new fascist republic. He turned it down.¹¹ By the spring of 1944 the Salò regime had arrested him, and it took German intervention to get him released.¹¹

A year later, on April 27, 1945, as partisans took Milan, a warrant was issued for his arrest from the other side; it was withdrawn, and in February 1946 he was formally cleared.¹¹ Few industrialists have been distrusted by both a dictatorship and its liberators in the space of twelve months. Rocca's response to the experience shaped everything that came after it: he left.

He had been, by any measure, one of the most powerful men in Italian heavy industry. Born in Milan in 1895, he joined the tube maker Dalmine in 1922, became its chief engineer three years later, and in the 1930s rose through the state holding company IRI to run Dalmine, the shipbuilder Ansaldo and other engineering firms at once.¹¹ When IRI created Finsider in 1937 to organize Italy's state steel industry, Rocca became its director general, and he resigned two years later when the steel program he had designed was scaled back.¹¹

He had joined the fascist party in 1923 and sat in Mussolini's corporate chamber, and the biographical record describes him as a convinced corporatist who believed that only the state could modernize an industry as capital-hungry as steel.¹¹ It also records a striking piece of advice he wrote in the summer of 1945: that Italy could never become a modern steel country with a "nationalist mentality," and needed to integrate into a wider economic area.¹¹ The man who would build a steel empire across Latin American borders had already concluded that steel did not respect them.

In November 1945, in Milan, he founded the Compagnia Tecnica Internazionale, a name soon shortened to Techint after the company's telex code.¹¹⁸ By 1947 Techint was in Argentina, drawn by Juan Perón's industrialization program, and the first big job was not a steel mill but a gas pipeline from Patagonia to Buenos Aires, completed in December 1949.¹¹⁸ Engineering and construction would remain a Techint business for the next eight decades, and it remains a supplier to Ternium today, a relationship that becomes important later in this story.

The steel came next. In 1954 Techint built and began operating two seamless tube plants, Tamsa in Veracruz, Mexico, and Siderca at Campana, Argentina.⁸ Those tube plants were the seed of what became Tenaris, the global supplier of pipe to the oil and gas industry that listed in New York, Milan and Mexico in 2002.⁸

In 1969 the group added a flat-steel cold-rolling plant at Ensenada, near La Plata, called Propulsora Siderúrgica.⁶ Cold rolling is a finishing step: a mill takes thick hot-rolled steel and squeezes it thinner, smoother and stronger, the kind of sheet that becomes a car door or a washing machine. Propulsora, renamed Siderar and then Ternium Argentina, was the seed of the flat-steel side of the family business.⁹⁶

Agostino Rocca died in Buenos Aires on February 17, 1978.¹¹ He left a son, Roberto, whose name is now on the group's technical schools, and a group that would be run by his descendants with an unusual combination of engineering culture and patience.¹¹⁸ Today Paolo Rocca is chairman of Ternium and, at the same time, chairman and chief executive of Tenaris and a director of San Faustin, the family holding company.¹²⁹ The Techint group as a whole reports about $33 billion of revenue across six companies and nearly 90,000 employees, but that group figure is not Ternium's: Ternium is one listed company inside the empire, with its own outside shareholders and its own accounts.⁸

The two branches of the family business grew up with very different customers. Tenaris sells pipe to oil and gas companies around the world, which ties its fortunes to drilling budgets and global energy prices. Ternium sells sheet and long steel to carmakers, appliance makers, builders and distributors in its own region, which ties its fortunes to Latin American industry and to the trade rules that protect it. Keeping them as separate listed companies let investors choose which cycle to own, and let the family hold both through a single holding structure; Tenaris also owns a stake in Ternium directly.⁹

The structure matters because it explains how the company thinks. A family that controls a business through a Dutch foundation and a Luxembourg holding company does not have to answer to the next quarter, and Ternium's history is full of decisions that only make sense on a twenty-year view. The same structure means outside shareholders are passengers.

They get the patience, and they also get the conflicts of interest that come from a controlling family owning the engineering firm, the gas supplier and the power plant that sell to the steel company. Both sides of that bargain began to matter in the 1990s, when Latin American governments decided to sell their steel mills and the Roccas were standing at the front of the queue.

↑ Back to Top

III. The Privatization Wave: Building the Latin American Steel Empire (1992-2005)

Picture the steelworks at San Nicolás on the Paraná River in the early 1990s. Somisa, Argentina's state steel company, had been built as a symbol of national industrial independence and had become, like many of the region's state enterprises, a symbol of everything that had gone wrong with it. When President Carlos Menem's government put it up for sale in 1992, Techint bought it.⁶ It then merged Somisa with its own Propulsora cold-rolling plant to create Siderar, so that a single company now made the liquid steel and finished it.⁶ In 1997 it folded in a second plant at Canning.⁶

The logic is worth spelling out, because it recurs for the next thirty years. A cold-rolling plant without its own steel has to buy hot-rolled coil from someone else, and in Argentina that someone was Somisa. A state steelmaker with no private finishing arm sells a commodity. Put the two together and one company captures the margin along the whole chain, from raw steel to the coated sheet a carmaker will pay a premium for. Privatization let Techint buy that integration from a government that needed cash and wanted out of the business, typically at a price no private seller would accept.

Venezuela came next. In the late-1990s privatization of Sidor, the giant state steelworks in the Orinoco basin, a consortium called Amazonia won a 70% stake, with the state keeping the rest.¹³ Amazonia was a coalition of the Techint group, Siderar, the Mexican steelmaker Hylsamex, Brazil's Usiminas and the Venezuelan steelmaker Sivensa.¹³⁶ After a restructuring in 2003, the consortium held about 60% of Sidor.¹³ Sidor produced about 4.3 million tons of liquid steel in 2006, enormous by regional standards, and it sat on top of Venezuela's iron ore and cheap natural gas.¹⁴ For a few years it looked like the best of the privatization prizes. It turned out to be the most dangerous.

The deal that defines Ternium's identity, though, was Mexican. Its roots go back to the Second World War, when a Monterrey industrial group called Valores Industriales, the owner of a brewery, could no longer import the sheet steel it needed for bottle caps.¹⁵ In 1943 it founded its own supplier, Hojalata y Lámina, which everyone came to call Hylsa.¹⁵ Some accounts put the founding in 1942; either way, Latin America's technologically inventive steelmaker began as a way to keep beer flowing.

Hylsa's great contribution came from scarcity. Making steel in an electric furnace needs a steady supply of scrap or a substitute for it, and Mexico in the 1950s had neither in reliable quantities. So the company's engineers, under Juan Celada, worked out how to strip the oxygen from iron ore using natural gas instead of coal, producing what the industry calls direct reduced iron, or DRI.¹⁵ The process proved itself on an industrial scale in 1957, and the plant in Monterrey was the world's first commercial gas-based direct reduction plant feeding an electric arc furnace.¹⁵¹⁶

By 1980 Hylsa had sold the rights to its "HYL" technology in more than 50 countries.¹⁵ The idea is simple once explained. Think of iron ore as rust, iron bonded to oxygen. A blast furnace pulls the oxygen off with coke made from coal, at very high temperatures, and makes liquid iron. A direct reduction plant does it with hot gas at lower temperatures, and makes solid pellets of nearly pure iron that can then be melted with scrap in an electric furnace. The technology lives on today as Energiron, developed jointly by the Italian groups Tenova and Danieli.¹⁶

Hylsa's financial history was less glamorous. It became the steel division of Grupo Alfa in 1974, borrowed heavily, and by the spring of 1988 owed about $1.2 billion to 68 lenders in 11 countries before a restructuring cut the debt roughly in half.¹⁵ By 2005 Alfa was ready to sell. In May of that year Techint agreed to buy Alfa's 42.5% stake in Hylsamex, then extended the offer to every other shareholder.¹⁷ By August Techint held more than 99% of the company, for a total purchase price of about $2.25 billion.¹⁸ The combined Latin American steel business it was assembling had roughly 12 million tons of capacity and about $5 billion of revenue.¹⁷

The strategic logic was a pan-regional champion: Argentine finishing know-how, Venezuelan raw steel, and a Mexican franchise with its own iron ore technology sitting next door to one of the largest steel markets in the world. It was also a bet that a way to build an empire is to buy the pieces from owners who no longer want them, whether those owners are governments or indebted conglomerates. That bet had paid handsomely so far. The next step was to turn three national champions into one listed company, and to sell part of it to the public.

↑ Back to Top

IV. Birth of Ternium & NYSE IPO (2005-2007)

The formal machinery had been built quietly. Ternium S.A. was organized in Luxembourg on December 22, 2003, as a holding company that conducts all its operations through subsidiaries.¹³ In 2005 the Techint group combined Siderar, Sidor and the newly bought Hylsa under that roof.⁶ Luxembourg was not a sentimental choice. A holding company there sat in a stable European legal system, outside the reach of any single Latin American government, with tax treatment designed for exactly this kind of entity; Ternium's tax-exempt holding status ran until the end of 2010.¹³ For a group that had watched governments nationalize and privatize the same assets within a generation, putting the parent company in a neutral jurisdiction was a form of insurance. It would soon be tested.

In January 2006 Ternium launched its initial public offering of about 24.8 million American depositary shares, each representing ten ordinary shares, and in February 2006 the shares began trading on the New York Stock Exchange under the ticker TX.¹³⁶ The listing gave the family a currency and a market price without giving up control. Two decades later the arithmetic of control looks almost exactly the same.

As of February 2026, Techint Holdings owned about 62% of Ternium, a Tenaris affiliate about 11.5%, and another family vehicle about 3%, so San Faustin's beneficial ownership was about 65%; treasury shares were about 2% and the public held about 21%.⁹ San Faustin, in turn, is controlled by a Dutch foundation, Rocca & Partners Stichting Administratiekantoor Aandelen San Faustin, which holds the voting rights.⁹ Outside investors own about a fifth of Ternium and have no realistic path to influencing who runs it.

The freshly listed company did not sit still. On May 1, 2007, Ternium agreed to buy Grupo Imsa, a Monterrey steel processor, for about $1.7 billion, or $6.40 a share, financed primarily with debt.¹⁹ Imsa's majority holders, with about 90% of the shares, agreed to sell at the same price.¹⁹ Paolo Rocca, explaining the deal, said it would strengthen Ternium's position in North America, where more than 60% of the company's revenue would now be concentrated.¹⁹ Imsa added coated and value-added steel lines in Mexico and the United States.¹⁹

Then came a move that shows the family's style better than any mission statement. In December 2007, only months after closing, Ternium's Imsa subsidiary agreed to sell Steelscape and Imsa's other American building-products businesses to BlueScope Steel for $730 million, cash-free and debt-free.²⁰ Those businesses had about $1.2 billion of sales in 2006, and management said they were not a strategic fit with the production system.²⁰

Ternium had bought a company, kept the steel processing that fed its mills, and sold the parts it did not want, recovering a large slice of the purchase price within a year. This is the template of a disciplined acquirer: know which link of the chain you want, and do not fall in love with the rest.

By the end of 2007 Ternium was consolidated in Mexico, listed in New York, controlled from Luxembourg, and producing steel in three countries.⁶ What it had not yet learned was how quickly one of those countries could disappear from the map.

↑ Back to Top

V. INFLECTION POINT #1: The Venezuela Nationalization Crisis (2008)

On April 8, 2008, Venezuela's vice president, Ramón Carrizales, announced that the negotiation over Sidor's labor contract had failed. The president, he said, had instructed him to inform the company that the government was taking control of the business.²¹¹⁴ The negotiations had dragged on for about 15 months over a contract covering almost 12,000 workers, with months of short strikes and clashes at the plant gates along the way.¹⁴²¹

Paolo Rocca had seen it coming, or at least feared it. In a letter written before the announcement, he urged Chávez to help find a solution and put a new offer on the table: higher salaries, higher pensions, and permanent jobs for 600 contract workers.²¹ "We advance this proposal to make possible an agreement, in spite of Sidor's financial limitations," he wrote.²¹

It did not matter. The issue had stopped being about wages. In New York, Ternium's shares fell more than 10% in a day, to about $35.²¹ Sidor's Venezuelan sales were running at about $2.4 billion a year.²¹ A quarter of the company's steelmaking had been declared the property of the state by a man who did not need a court to agree.

The episode is often told as a straightforward expropriation, and in substance it was. What followed was a slow, legalistic negotiation over price, conducted by a company with very little leverage. In early May, Chávez said Techint was demanding $3 billion to $4 billion in compensation, while the government was offering $800 million.²²

Venezuela's National Assembly had already declared the mill a matter of national utility, and a decree-law dated April 30, 2008, ordered Sidor turned into a state enterprise.²²¹³ On July 12, 2008, the state holding company CVG took operational control.²³¹³ Ternium kept legal title to shares in a plant it could no longer run, which is roughly as useful as holding the deed to a house someone else has moved into.

The final settlement came on May 7, 2009, when Ternium transferred its entire 59.7% stake to CVG for $1.97 billion.²⁴²³ Only $400 million arrived at closing. The rest came in installments: $945 million over six quarters, and the remainder due in late 2010, with early repayment triggers linked to the price of crude oil.²⁴ Ternium was, in effect, lending the purchase price to the government that had just seized its asset, and betting that Venezuela's oil revenue would keep the cheques clearing. By early 2010 CVG had paid about $1.5 billion; one installment that fell due in February 2010 arrived 23 days late.¹³

So how should the Sidor episode be judged? The consensus story is that Ternium was robbed. The record supports a more nuanced verdict. Ternium recovered nearly two and a half times Venezuela's opening offer, which says something about the value of patient, legalistic bargaining backed by a parent company outside Venezuela's jurisdiction. But it also lost a low-cost integrated steelworks sitting on its own iron ore and gas, permanently, at a price set under duress.

There is a second lesson, about labor. Sidor's crisis grew out of a long dispute over pay, pensions and contract workers, and the company's last offer conceded much of what workers wanted.²¹ A steelworks is a political object in a way a software company is not; it employs thousands of people in one town, it consumes subsidized energy, and its output feeds national infrastructure. That makes the relationship with the state a part of the cost structure, whether or not it appears in the accounts.

The loss also changed the shape of the company's growth. In place of one giant state-built works, Ternium spent the following decade adding smaller, more controllable pieces. In 2010 it bought a controlling interest in Ferrasa, a Colombian steelmaker and distributor with a business in Panama, and in 2021 it opened a new mill at Palmar de Varela in Colombia able to make 520,000 tons a year of rebar and other long products.⁶ None of these were Sidor-sized. All of them were in countries where Ternium could run the plant itself.

With Venezuela gone, Ternium's center of gravity shifted decisively to Mexico, where the Hylsa and Imsa plants now carried the company, and to Argentina, where Siderar remained the dominant flat-steel supplier. But a company built by buying the region's steel mills was not going to stay out of Latin America's largest steel market for long. Its next move was into Brazil, and it proved more expensive than anything Chávez had done.

↑ Back to Top

VI. The Brazil Play: Usiminas Investment (2011-2023)

On November 27, 2011, Ternium announced that it had agreed to pay 36 reais a share for a slice of Usiminas, the flat-steel maker based in Minas Gerais and one of the pillars of Brazilian industry.²⁵ Together with its Argentine subsidiary Siderar and Tenaris's Brazilian pipe unit Confab, Ternium bought about 140 million ordinary shares, 27.7% of Usiminas's voting capital, from the construction group Camargo Corrêa, the Votorantim conglomerate and the Usiminas employees' pension fund.²⁵

Ternium and Siderar's share of the bill was about 4.1 billion reais, roughly $2.2 billion, paid from cash and a new $700 million loan.²⁵ The price was about 41% above the average of Usiminas's share price over the previous six months; measured against the price on the screen the Friday before, Brazilian press put the premium at about 83%.²⁵²⁶

What Ternium bought was a seat at a crowded table. The control group of Usiminas after the deal was split between Nippon Steel's group with about 46%, the Ternium–Tenaris group with about 43%, and the pension fund with about 11%.²⁵ Ternium and Tenaris joined that group in January 2012.⁶ It was a shared house with a powerful Japanese co-owner who had been there for decades, and the paper agreement between them assumed the two sides would agree on how to run it.

They did not. In September 2014 Usiminas's board, after audits that Nippon Steel said had found improper payments to executives, voted to dismiss the chief executive, Julián Eguren, and two vice presidents, all of them nominated by Ternium.²⁷ Ternium defended the executives and argued that its partner had broken the shareholders' agreement; the fight moved into the courts.²⁷ A widely read Brazilian account of the dispute later summed it up in a line: both sides lacked humility.²⁷ For the next several years Usiminas was a company run by a divided board in the middle of a Brazilian recession.

The financial verdict arrived in the write-downs. Ternium cut the value of its Usiminas investment by about $275 million in 2012, about $740 million in 2014, and about $192 million in 2015.²⁸ At the end of 2015 Usiminas shares closed at about 4 reais, and Ternium's stake was worth about $171 million on the market, against more than $2 billion paid four years earlier.²⁸ This is the strongest disconfirming evidence in the company's own record against the idea that the Roccas always buy at the bottom. In Usiminas they paid a large control premium near a cyclical peak for a minority position in a company they could not run.

Then there was the lawsuit. Companhia Siderúrgica Nacional, the rival Brazilian steelmaker, owned a stake in Usiminas and argued that Ternium's 2012 purchase had amounted to a change of control. Under Brazilian corporate law, that would have entitled minority holders like CSN to a "tag-along" offer at 80% of the price Ternium paid.⁹ For years the courts disagreed with CSN. A first-instance court dismissed the case in 2013, an appeals court affirmed that in 2017, and in March 2023 the Superior Court of Justice, Brazil's highest court for non-constitutional matters, rejected CSN's appeal.⁹ Brazil's securities regulator had also concluded that no offer was required.²⁶

In June 2024 the same court reversed itself. By three votes to two, a panel of the Superior Court of Justice ruled that Ternium owed CSN compensation, about 5 billion reais at the time.²⁶²⁹ Ternium called the decision contrary to substantive law and due process.²⁹ In December 2024 the court partly granted Ternium's motion for clarification, changing the inflation-adjustment mechanism and capping legal fees, which lowered the amount without changing the outcome.⁹³⁰

In 2025 Ternium filed further appeals, including to the Supreme Federal Tribunal.⁹ The accounting consequence is concrete. Ternium booked about $410 million of litigation expense in 2024 and another $117 million in 2025, carrying a provision of about $528 million at year-end 2025 that had grown to about $599 million by June 2026 as interest and inflation accrued.⁹³¹ The 2024 charge is why Ternium reported a small loss attributable to its shareholders that year, about $53 million, despite running its mills at a profit.⁹ A dispute over a 2012 share purchase is, fourteen years later, still accruing.

Brazil did not get easier. Ternium kept buying, and its position changed from co-owner to controller. In July 2023 it bought an additional block of about 69 million shares from Nippon Steel for about $119 million and signed a new shareholders' agreement that gave Ternium the right to nominate a majority of Usiminas's board and its chief executive.⁹³²

With 51.5% of the control group and about 25% of Usiminas's total shares, Ternium began consolidating Usiminas line by line in July 2023.³² The accounting cost of that step was a non-cash loss of about $1.1 billion, mostly from releasing years of accumulated currency translation losses into the income statement, which is why Ternium's third quarter of 2023 showed a large loss.³²

On the day it took control, Ternium also took on the full weight of Brazil's import problem: flat-steel imports into Brazil rose by about 30% in 2025.⁹ In the third quarter of 2025, Usiminas wrote down $405 million of deferred tax assets, an admission that it might not earn enough taxable profit to use them, and the auditors flagged the recoverability of the remaining tax assets as a critical audit matter.⁷⁹

The final act so far came in February 2026. Ternium closed the purchase of Nippon Steel's and Mitsubishi's remaining 153.1 million Usiminas shares for about $315 million, or $2.06 a share.³³³⁴ That took Ternium to about 83% of the control group and left minority shareholders with about 62.5% of Usiminas's economics, meaning Ternium's own economic interest is roughly 37.5%.³³³¹ Put the two prices next to each other and the arc of the Brazil play becomes vivid. In 2011 Ternium paid about $19 a share for Usiminas.²⁵ In 2026 it paid about $2 for the same kind of share, from the partner that had once outvoted it.³³

The verdict is mixed rather than damning. The first purchase destroyed value on any reasonable measure, and its litigation tail is still costing money. The later purchases bought control cheaply, and Ternium now runs a large, integrated Brazilian steel and iron ore business with some 9 million tons a year of mining capacity.⁹ Moody's Local Brasil, rating Usiminas in early 2025, credited exactly those strengths, the integrated flat-steel and mining position and solid liquidity, while warning about weak prices, import pressure and investment needs.³⁵ The same agency measured Usiminas's adjusted gross leverage at about 3.3 times at the end of 2024, a reminder that the Brazilian subsidiary carries a heavier debt load relative to its earnings than its parent.³⁵

Usiminas's own ratings, Ba2 on Moody's global scale with a stable outlook as of July 2026, belong to Usiminas, not to Ternium, which does not publish a parent-level rating rationale in its annual filing.³⁶⁹ What Ternium shareholders actually own of that business, though, is little more than a third. Before Ternium took control of Usiminas, it had already found a cheaper way into Brazil, by buying a plant from a German company that wanted nothing more to do with it.

↑ Back to Top

VII. INFLECTION POINT #2: CSA Brazil Acquisition from ThyssenKrupp (2017)

Start with the scale of the wreckage. In the mid-2000s, the German conglomerate thyssenkrupp decided to build a steel business in the Americas. The plan was elegant on paper: make low-cost slabs in Brazil, near the iron ore and on the coast, then ship them to new rolling mills in Alabama and to Europe for finishing.

A slab is the steel industry's semi-finished brick, a thick rectangle of solid steel, 20 or so centimetres deep, that rolling mills turn into coil. The Brazilian end of the plan was CSA, Companhia Siderúrgica do Atlântico, a giant integrated steelworks in the state of Rio de Janeiro with its own deep-water port and power plant, built with the mining company Vale as a partner.³⁷³⁸

The plan went wrong in almost every way a steel project can. Costs ran far over budget, the start-up was troubled, and the complex contractual ties with Vale took until May 2016 to unwind; CSA only received its definitive operating licence in September 2016.³⁸ The Alabama end of the project had already been sold, which is why Ternium's later paperwork refers to thyssenkrupp's "former" Calvert mill.³⁷ When it finally agreed to sell CSA, in February 2017, it said that even after the sale proceeds its net loss from the attempt to expand into the Americas stood at around €8 billion.³⁹

Ternium's side of that announcement, dated February 21, 2017, was a model of understatement. It would buy CSA for an enterprise value of €1.5 billion, with the economics backdated to September 30, 2016, and because CSA carried about €0.3 billion of debt to the Brazilian development bank BNDES, Ternium expected to pay out about €1.26 billion.³⁷ For that, it got a plant with capacity to make 5 million tons a year of high-end slabs, a deep-water harbour, a 490-megawatt combined-cycle power plant, and a contract, inherited from thyssenkrupp, to supply 2 million tons a year of slabs to the former thyssenkrupp rolling mill in Calvert, Alabama.³⁷

In 2016 those assets had shipped about 4.3 million tons and earned about €256 million of EBITDA on €1.6 billion of sales.³⁷ At that rate, Ternium was paying about six times one year's operating cash earnings for a plant that had cost its builder many times the purchase price. The sale closed in September 2017, and CSA was renamed Ternium Brasil.³⁸⁶

Why did it make sense for Ternium when it had made no sense for thyssenkrupp? Daniel Novegil, then Ternium's chief executive, said the slab plant combined with the company's capacity in Mexico and Argentina "will support new integration opportunities."³⁷ Translated: Ternium's Mexican rolling mills needed steel to roll, and Mexico did not make enough of its own.

Thyssenkrupp had built a slab plant with no captive customer close enough to make the logistics work. Ternium already owned the customer. The slabs could cross the Gulf of Mexico to its own hot-rolling mills instead of being sold into the open market. The same asset is worth more to a buyer who can use it internally, which is the whole economics of vertical integration in one sentence.

This was the privatization playbook adapted for a new kind of seller. In the 1990s the motivated sellers were governments. In 2017 it was a European conglomerate under pressure to shrink, which had already written down most of the asset's value and needed the plant off its balance sheet. Ternium had the patience, the balance sheet and the strategic use. The phrase "masterclass in buying distressed assets" gets applied to the CSA deal a lot, and for once the evidence mostly supports it: the price was low relative to the plant's capacity and its operating earnings, and the asset fit a specific gap in Ternium's system.

There is one serious caveat, and it shapes the rest of the story. Steel made in Brazil is not steel made in North America. When the United States, Mexico and Canada renegotiated their trade agreement into the USMCA, its automotive rules of origin pushed carmakers toward steel that is "melted and poured" in the region.⁹ A Brazilian slab rolled in Mexico does not meet that test.

The very plant that solved Ternium's steel shortage in 2017 could not, by itself, supply the most valuable customers in its biggest market under the rules that came into force a few years later. To sell North American steel to North American carmakers, Ternium would have to melt it in North America. That is the bet now rising out of the ground at Pesquería.

↑ Back to Top

VIII. INFLECTION POINT #3: The Pesquería Mega-Investment & USMCA Bet (2013-Present)

On Wednesday, September 18, 2013, Mexico's president, Enrique Peña Nieto, stood beside Paolo Rocca in Pesquería, Nuevo León, to open the Ternium Industrial Center, a $1.1 billion complex of cold-rolling and galvanizing lines built to make steel for car bodies.⁴⁰ Part of it was a joint venture called Tenigal, 51% owned by Ternium and 49% by the Japanese steelmaker Nippon Steel, with a roughly $300 million galvanizing line able to coat 400,000 tons a year of automotive sheet, including the exposed outer panels that must be flawless.⁴¹ That partnership, with the same Japanese group that was simultaneously fighting Ternium for control of Usiminas, is a reminder that in the steel industry rivals and partners are often the same people in different rooms.

The 2013 opening was the first phase of a site that has kept growing. In 2016 a 900-megawatt gas-fired power plant, Central Eléctrica Pesquería, opened in the Monterrey area, together with a technical school named for Roberto Rocca.⁶ In May 2021, more than two weeks ahead of schedule, a new hot-rolling mill produced its first coils.⁴² With capacity of about 4.4 million tons a year, it gave Ternium Mexico the ability to turn slabs, including the Brazilian ones, into the hot-rolled coil that feeds every downstream line.⁹ On its Mexican website, the company now counts about $7.5 billion invested across the center's three phases.²

The third phase is where the strategy becomes a bet. On February 14, 2023, Ternium's board approved an upstream project: an electric arc furnace steel shop with capacity of about 2.6 million tons a year and a direct reduction module making about 2.1 million tons a year of DRI, plus a port facility for raw materials, at an expected cost of about $2.2 billion.³ In June 2023 the company confirmed it would build the mill at Pesquería itself, integrated with the existing downstream lines, and said construction would begin in December 2023 with operations in the first half of 2026.⁴

Together with a new cold-rolling mill, galvanizing and finishing lines on the same site, the whole program was announced at $3.2 billion.⁴³¹ Máximo Vedoya, the chief executive, put the thesis plainly: Mexico had to seize the opportunities of nearshoring, and steel was "the first link of this value chain."⁴³ He also said Ternium had received no incentives for the project.⁴³

Vedoya, an engineer who has run Ternium since 2018, is the face of the Pesquería era, and the project's record is the best available evidence of how his management team handles promises.⁹ It is a mixed record. The downstream half has largely been delivered: the pickling line and new finishing lines are running, and the cold-rolling mill and hot-dip galvanizing line had begun operating and were ramping up by early 2026.⁹ The upstream half has slipped. A year after the announcement, Ternium raised its budget to $3.5 billion, up 9%, blaming equipment inflation and currency moves.³²

The start date moved from the first half of 2026 to the fourth quarter of 2026 in the annual report filed in March 2026.⁴⁹ And on the August 2026 call, Vedoya said the slab facility was progressing well "with start-up expected in early 2027," and warned analysts not to expect a large contribution to earnings in 2027 because the ramp-up would take several quarters.⁵ None of these slips is unusual for a greenfield steel shop. Together they say that the earnings payoff from Pesquería sits further out than the original timetable implied, and that investors are being asked to wait.

Why build it at all? The answer is the trade map. The United States raised its Section 232 tariff on steel imports to 25% in early 2025 and then to 50% in June 2025, removing the country exemptions Mexico had enjoyed.⁹ That hit Ternium's customers, the Mexican factories that make appliances, car parts and machinery for the American market, and Mexican apparent steel demand fell by about 10% in 2025.⁹

On the August call, Vedoya said the tariffs were "hard to justify" because the United States runs a large steel trade surplus with Mexico, and that talks between the two governments had advanced without producing concrete results.⁵ The USMCA's first joint review fell due on July 1, 2026.⁹ Mexico, for its part, raised import tariffs on more than 1,400 tariff lines for countries without a free trade agreement, covering most steel products, which helped Ternium win back share from Asian imports at home.⁷⁵

So the Pesquería bet is really two bets stacked on top of each other. The first is that North American rules of origin will keep rewarding steel melted and poured inside the USMCA, so that a Mexican mill making high-specification automotive steel can sell into the integrated car supply chain of all three countries.

The second is that the United States will not use tariffs to treat Mexican steel as foreign anyway. The first bet is underwritten by the treaty's text. The second depends on politics in Washington, and the gap between hot-rolled coil prices in Mexico and the United States, which an analyst put at about $300 a ton on the August call, measures how far apart the two markets currently sit.⁵

There is also a question an activist investor would ask out loud. Who is building Pesquería? Part of the answer is the Rocca family. In 2025 Ternium paid about $406 million in engineering, construction, labor and supervision fees to companies controlled by San Faustin, up from about $178 million in 2024, mainly because of the Pesquería expansion; it also bought about $317 million of electricity from the group's Techgen power plant.⁹ Material related-party transactions are reviewed by an audit committee of three independent directors, but only three of the eight board members are independent, and the chairman sits on the boards of both Tenaris and San Faustin.⁹

Ternium does not publish a project-level reconciliation showing what Pesquería has cost relative to independent benchmarks. The fees may well be fair. The structure means outside shareholders have to take that on trust, at exactly the moment when capital spending is at its peak: from about $1.5 billion in 2023 to a peak of $2.5 billion in 2025, before falling, management says, to about $1.2 billion in 2027.⁹⁵

The verdict on Pesquería cannot be written yet, because the decisive part has not started. What can be said is that the downstream half has been delivered, the upstream half is late and more expensive than first announced, and its returns depend on a trade relationship that changed for the worse after the project was approved. The steel shop's other selling point is environmental, and that claim deserves its own examination.

↑ Back to Top

IX. Decarbonization & Sustainability Strategy

In the hills around Olavarría, in Buenos Aires province, 22 wind turbines now turn above the pampas. Ternium finished installing them in early 2025, and the Vientos Olavarría wind farm, 99 megawatts of capacity built for more than $220 million, has been fully operational since March 2025.⁴⁴⁴⁵ It is designed to generate about 480 gigawatt-hours a year and replace nearly 90% of the electricity Ternium buys from Argentina's grid.⁴⁴ In its first year it generated about 434 gigawatt-hours.⁵

The company says the turbines are built with about 85% recyclable components and that the farm will cut carbon dioxide emissions by about 111,000 tons a year, and the World Steel Association gave the project its sustainability award in October 2025.⁴⁵ The economics are as interesting as the ecology. Argentina's grid electricity has been unreliable and, through repeated currency crises, unpredictably priced; owning the generation turns a volatile input cost into a fixed capital outlay, which is the same integration logic Ternium applied to slabs, only with wind.

The wind farm is real, finished and measurable, which makes it a useful contrast to the bigger decarbonization claim, the one that rests on chemistry. Steel is one of the most carbon-intensive materials in the world. The route Hylsa pioneered in Monterrey uses natural gas, which is mostly methane and releases far less carbon per ton of iron than coal.

Ternium describes the new Pesquería steel shop as a low-emission-intensity project, and says the DRI module will have carbon capture and will be ready to switch from natural gas to hydrogen.⁴⁹ With hydrogen in place of methane, the only by-product of stripping oxygen from iron ore would be water.

That last sentence is where a skeptic should pause. "Hydrogen-ready" means the equipment can use hydrogen; it does not mean hydrogen is available at a price a steel mill can afford, and the company has not said when, or at what cost, a hydrogen supply for Pesquería would exist. Carbon capture equipment captures only what it is designed to capture and still needs somewhere to send the carbon dioxide. The near-term decarbonization benefit of Pesquería comes from replacing imported slabs, many made in blast furnaces, with steel made from gas-based iron and scrap, not from hydrogen.

And most of Ternium's steel is still made the old way. The company's installed capacity at the end of 2025 included about 13.5 million tons of basic oxygen furnaces fed by about 11.3 million tons of blast-furnace iron, almost all of it in Brazil and Argentina, against about 4.5 million tons of electric arc furnace capacity, nearly all of it in Mexico.⁹

The Usiminas consolidation and the CSA purchase made Ternium, on balance, more of a blast-furnace company, not less. Its formal target is modest: a 15% cut in emissions intensity by 2030 compared with 2023, backed by an environmental investment plan of about $556 million over the decade.⁹ That is a respectable industrial target. It is not a transformation, and it would be a mistake to let the language of hydrogen-readiness stand in for one.

The more interesting point is economic. The DRI-plus-electric-furnace route is most attractive where natural gas is cheap, so a mill with access to North American gas is the one where that route pays. If carmakers and their customers begin paying a premium for lower-carbon steel, or if border carbon taxes spread from Europe, Pesquería's chemistry becomes a commercial edge. Until then it is an option, valuable but unpriced, sitting on top of a business whose current results are driven by something far more ordinary: the steel cycle.

↑ Back to Top

X. Current Operations & Financial Performance

Ternium's 2025 annual results, published in February 2026, read like a company at the bottom of a cycle. Steel shipments fell about 4%, to about 15.1 million tons; revenue fell about 12%, to about $15.6 billion; and adjusted EBITDA, the company's preferred measure of cash operating earnings, fell to about $1.5 billion, a margin of about 10%.⁷ Net income attributable to Ternium's shareholders was about $425 million, or $2.17 per ADS.⁷ Mexico took about 7.4 million tons, Brazil about 3.9 million, Argentina and its neighbours about 2.2 million and other markets about 1.5 million.⁷ The company says it has about 15.4 million tons of raw steelmaking capacity and about 22.3 million tons of finishing capacity.⁹

To see how far the business had fallen, go back four years. In 2021, when steel prices exploded after the pandemic, Ternium earned an operating margin of about 33% on about $16 billion of revenue, and about $3.8 billion of net profit for its shareholders.⁹ Operating income then fell for four years in a row, by roughly half in 2022 and more than 40% in 2025, until the operating margin stood at about 4.5%.⁹

Revenue was about the same in 2025 as in 2021, partly because Usiminas had been added to the consolidated accounts in the meantime; the profit had evaporated. Steel revenue per ton fell about 10% in 2025 alone.⁹ This is what a commodity business looks like: the same mills, the same customers, and operating profit that swung by a factor of about seven depending on the price of a ton of coil.

The cash story is subtler, and better. Over the twelve years from 2014 to 2025, Ternium generated about $20.6 billion of operating cash flow against about $11.4 billion of net profit attributable to its shareholders.⁹ Part of that gap is real: steel mills carry heavy depreciation, which is subtracted from profit but not from cash, and in a downturn working capital releases cash as inventories shrink.

Part is an accounting mismatch, because the cash flow statement includes the minority holders' share of Usiminas and Ternium Argentina, while the profit figure does not.⁹ In 2025 the effect was dramatic. A consolidated profit of about $300 million became $2.3 billion of operating cash, helped by about $790 million of depreciation and a working-capital release of about $1 billion.⁹ Then capital spending of $2.5 billion swallowed all of it and more, leaving free cash flow slightly negative for the year.⁷

The balance sheet behind that cash is conservative but not inert. At the end of 2025 Ternium had about $2.4 billion of financial borrowings against roughly $3.1 billion of cash and investments, a net cash position of about $712 million.⁹⁷ About two thirds of the borrowings carried floating interest rates, about $600 million was due within a year, and a $1.25 billion green loan facility for Pesquería had only $300 million drawn, with a leverage covenant attached; the company reported that it complied with its covenants.⁹

Its customers pay on time by steel-industry standards. Of about $1.5 billion of trade receivables at year-end 2025, only about $90 million was past due, and large portions were guaranteed or covered by credit insurance.⁹ The auditor, PricewaterhouseCoopers, signed a clean opinion on the 2025 accounts and on internal controls, and singled out one judgment for special attention: whether Usiminas will earn enough future taxable profit to use the roughly $580 million of deferred tax assets still on its books after the 2025 write-down.⁹ That is the accounting estimate most likely to move Ternium's reported profit without any change in its cash.

Through all of this Ternium kept paying dividends, and then, in April 2026, it blinked. Its February results had proposed a dividend of $2.70 per ADS for 2025, about $530 million.⁴⁶ On April 15, 2026, the board cut the recommendation to $2.20 per ADS, about $432 million, citing caution over the potential global economic impact of the conflict in the Middle East and the prudence of reinforcing the balance sheet.⁴⁶ Shareholders approved the lower dividend on May 12, 2026.¹²

Even after the cut, Ternium paid out more than its 2025 profit. Over twelve years, dividends absorbed about 45% of the free cash flow the business generated, with the rest accumulated as cash: about $477 million at the end of 2013 and about $3.1 billion at the end of 2025.⁹ That cash is not all equally liquid. About $803 million sits inside Ternium Argentina, much of it in US-dollar Argentine sovereign bonds, and Argentina's exchange controls still restrict moving it out.⁹ Only about a third of the group's liquid financial assets were in investment-grade instruments at the end of 2025.⁹

The 2026 numbers so far show a recovery underway. Adjusted EBITDA rose to about $480 million in the first quarter and about $717 million in the second, a margin of about 16.5%.³⁴³¹ Steel revenue per ton climbed to about $1,070, and Mexican shipments rose sharply from a year earlier as distributors rebuilt inventories and imports retreated.³¹⁵

Net income attributable to shareholders in the first half was about $557 million, already more than all of 2025.³¹ Management has been consistent in its quarterly guidance this year. It said in February that EBITDA would rise in the first quarter; it did.⁷³⁴ It said in May that EBITDA would rise in the second quarter; it did.³⁴³¹ In August it said EBITDA would rise again in the third quarter on higher shipments and better margins.³¹

The recovery has a cost in cash. Higher prices and volumes require more working capital, and in the second quarter Ternium tied up about $418 million in inventories and receivables.³¹ Combined with capital spending and the May dividend payment, that moved the company from net cash of $327 million in March to net debt of about $112 million at the end of June.³¹ That is a trivial level of debt for a company of this size. It is also the first time in years that the balance sheet has moved in the wrong direction while the operating business was improving, and it shows how little room there is between "investment peak" and "dividend commitment."

On the August call, Vedoya agreed with the World Steel Association's forecast of about 4% growth in Mexican steel demand in 2026, after the 10% decline of 2025, but cautioned that the market "is growing at a pace that still needs to improve."⁵ In Brazil he described import protection as improving, with quotas renewed until June 2027 and an antidumping case on Chinese hot-rolled coil due for a decision, while saying "more is still needed."⁵

In Argentina, his view had not changed: manufacturing remained weak, held back by soft demand and imports.⁵ When Goldman Sachs's analyst asked whether dividends could return to earlier levels, the chief financial officer, Pablo Brizzio, called an increase "a possibility" but said the company was "not yet at the moment of a definition."⁵ That is the honest version of where Ternium sits in late October 2026: earnings recovering, capital spending past its peak, and the payoff from the largest project in its history still a year or more away.

↑ Back to Top

XI. The Bull and Bear Case

Imagine two investors standing on the same Pesquería site in the autumn of 2026. One sees the most modern flat-steel complex in Latin America, nearly finished, sitting inside the world's largest integrated manufacturing region, owned by a family that has bought assets cheaply for three generations and is about to stop spending heavily on them. The other sees a commodity producer whose profits have fallen for four years, whose biggest market is hostage to American tariff politics, whose Brazilian arm is a minority-owned, import-battered company with a court judgment hanging over it, and whose controlling family is also its contractor. Both are looking at the same facts. The job is to weigh them.

Start with the industry, because steel's economics set the ceiling for everyone in it. The threat of new entrants is low in the classic sense, because a new integrated steel shop costs billions and takes years, as Pesquería's own timetable shows. But the real entrant in Latin American steel is not a new mill; it is the import ship. Chinese and other Asian steel, and goods made from it, flood into markets whenever local prices rise above world levels, which is why flat-steel imports into Brazil rose about 30% in 2025 and why Ternium spends so much management attention on antidumping cases and tariff quotas.⁹⁵

Supplier power is moderate: iron ore and coal are bought at world prices, though Ternium's own mines in Mexico and Brazil and its gas supply from a related party soften the exposure.⁹ Buyer power is real but fragmented. No customer accounted for more than 10% of 2025 sales, and much of the commercial market is priced close to spot, which means Ternium has little ability to hold prices up when the cycle turns.⁹⁵

Substitutes, mainly aluminium and plastics in cars and appliances, chip at the edges. And rivalry with other regional producers, from Brazil's Gerdau and CSN to ArcelorMittal's local subsidiaries, is fierce in the commoditized products and gentler in the high-specification automotive grades. On balance the five forces describe a tough, cyclical industry in which local producers earn good returns only when governments keep imports out.

Now use Hamilton Helmer's seven powers to ask whether Ternium has any durable advantage within that industry. Network economics, switching costs in the software sense, and brand do not really apply to steel. Counter-positioning does not either. Three powers are worth testing.

The first is scale economies, and the case for them is real but regional. Ternium is the largest flat-steel producer in Mexico, its mills are new, and the Pesquería complex concentrates hot rolling, cold rolling, galvanizing and soon steelmaking on one site, cutting transport and handling costs. But global steel scale belongs to Chinese, Japanese and Indian producers many times Ternium's size, and in a world of open trade those producers set the marginal price. Ternium's scale advantage is real only inside the protected perimeter that trade policy draws.

The second is cornered resource, and here the claim is strongest and most fragile at the same time. The cornered resource is not an ore body; it is the combination of a USMCA rules-of-origin regime, Mexican tariffs against non-treaty imports, and a near-monopoly position in domestic high-specification flat steel. That combination is genuinely hard to replicate, and it is why Ternium could win share against imports in 2026.⁵ It is also a resource granted by governments, which means it can be withdrawn by governments, as Section 232 demonstrated in 2025 and as Chávez demonstrated in 2008. History narrows this power to its smaller version: a protected regional franchise whose value rises and falls with politics.

The third is process power, the accumulated know-how that lets a company do the same thing better than rivals who copy its equipment. The evidence for it is Hylsa's long engineering tradition, the hot-rolling mill started ahead of schedule in 2021, and the Tenigal partnership with Nippon Steel for exposed automotive panels.⁴²⁴¹ The evidence against it is the record on margins: when the cycle turned, Ternium's operating margin fell from about 33% to about 4.5%, and its Brazilian assets under Ternium's operational control still needed a large tax-asset write-down.⁹⁷

Process power that cannot protect margins in a downturn is a competence, not a moat. The honest verdict is that Ternium has a defensible regional position, mostly built on integration and trade protection, and that its returns over a full cycle look like a good steelmaker's rather than a great company's. Its return on equity was about 3.6% in 2025 and about 5.7% over the twelve months to June 2026.⁹³¹

What, then, is the stock price assuming? At about $54 per ADS on October 29, 2026, Ternium's market value was about $10.6 billion and its enterprise value about $10.9 billion.¹⁰ That is roughly 15 times trailing earnings, under six times trailing EBITDA, and about 0.86 times book value, with a dividend yield of about 4%.¹⁰ Some data services multiply the ADS price by the number of ordinary shares rather than ADSs, which inflates the market value and the price-earnings ratio tenfold; one ADS represents ten shares.¹³

The nearest listed peers trade in the same neighbourhood: Gerdau at about 19 times earnings and under five times EBITDA, ArcelorMittal at about 27 times earnings and about ten times EBITDA, and CSN, still loss-making, at about six times EBITDA and about a third of book value.⁴⁷⁴⁸⁴⁹ Analysts' consensus, at about seven times forward earnings, assumes the 2026 recovery continues into much higher profits.¹⁰ The market is not paying a premium for Ternium's franchise. It is valuing a cyclical steelmaker near book value, with a modest bet on recovery, and applying a discount for the layers between the mills and the shareholder.

Those layers are the bear case's best material. The first is minority leakage: outside shareholders own about 62.5% of Usiminas and 37.4% of Ternium Argentina, so a large slice of consolidated EBITDA belongs to someone else.³¹ The second is related-party spending, which reached about $874 million of purchases of goods and services in 2025, including the construction fees and Techgen electricity at the heart of Pesquería.⁹

The third is the CSN provision, now about $600 million and still growing while appeals continue.³¹ The fourth is Argentine cash that cannot easily leave Argentina.⁹ An activist looking at this company would not ask for a new strategy. They would ask for disclosure: a project-level accounting of what Pesquería cost against independent benchmarks, an explicit capital-return policy for the post-investment years, and individual executive pay, which the company does not disclose.⁹ A controlling shareholder with 65% of the votes need not grant any of it.

Management credibility deserves its own weighing, because the bull case rests on trusting the people building Pesquería. The quarter-to-quarter record in 2026 is good: three consecutive outlooks for rising EBITDA, the first two of which were delivered.⁷³⁴³¹ The multi-year record is weaker. The slab mill's start has moved from the first half of 2026 to early 2027, its budget has risen, and the 2025 dividend announced in February was cut eight weeks later.⁴⁵³²⁴⁶

Explanations were given each time, inflation and currency for the budget, Middle East uncertainty for the dividend, but the pattern is of a team that is accurate about the next quarter and optimistic about the next few years. That narrows the management-quality claim rather than rejecting it: the operators are credible, the long-range calendar less so.

Three numbers will tell investors which story is winning. The first is adjusted EBITDA margin, which captures price, cost and mix in one figure: it fell to about 10% in 2025 and recovered to about 16.5% in the second quarter of 2026, rising.⁷³¹ The second is Mexican steel shipments, the volume measure of whether nearshoring and import protection are real: about 7.4 million tons in 2025, and rising strongly in 2026, with second-quarter shipments well above a year earlier.⁷³¹

The third is free cash flow after capital spending, the measure of whether the investment cycle is over: slightly negative in 2025 and about negative $364 million in the first half of 2026 as working capital rebuilt, improving only if capex falls as promised.⁷³¹ If all three rise together through 2027, the bull case is being proven. If margins rise while free cash flow stays negative, the recovery is being spent before it reaches shareholders.

↑ Back to Top

XII. Lessons

The first lesson comes from Techint’s purchases of Hylsa and CSA. The best price in steel is paid to a seller who has stopped believing. Ternium's great deals were made when governments, conglomerates and creditors wanted out, and when Ternium owned a downstream customer that made the asset worth more in its hands than anyone else's.

Its worst deal, the 2012 Usiminas stake, came when a confident seller was paid a large premium at a cyclical peak for a seat at a shared table. For investors, the lesson is that capital allocation in cyclical industries is judged by the seller's mood more than the buyer's spreadsheet.

The second lesson comes from Venezuela's nationalization and the tariff shifts of 2025 and 2026. In Latin American steel, the government is a shareholder that never appears on the register. Ternium's moat and its biggest risk are the same object: a trade perimeter drawn by politicians. Anyone who values the company without a view on that perimeter is valuing a different company.

The third lesson comes from Ternium's later purchases of Usiminas shares. Control bought late is cheap; control bought early is ruinous. The second purchase looks brilliant only because the first one was so expensive, and the court judgment from the first purchase is still accruing interest. Founders and acquirers tempted to pay a premium for a minority seat in a shared house should remember that a seat is not a key.

The fourth lesson comes from Ternium's 2026 dividend cut. Ternium's patience is subsidized by its shareholders' patience. A family with a twenty-year horizon can build a steel shop through a downturn, and has. The outside investor funds that patience with lower dividends and a heavier balance sheet, and gets the reward only if the project earns its cost of capital after the family's own contractors have been paid.

The fifth lesson comes from Hylsa's gas-based ironmaking and the hydrogen-ready DRI module at Pesquería. Ternium's best technology was born of necessity, not of marketing. The HYL process spread because it solved a cost problem; hydrogen-readiness will matter only when hydrogen solves one too. For investors weighing any industrial company's green promises, the question is the same one Hylsa answered seventy years ago: what does this lower the cost of, and for whom?

↑ Back to Top

XIII. Epilogue

Tonight, in late October 2026, Ternium stands at a hinge. Its mills are running harder than they were a year ago, its Mexican customers are restocking, Mexico's tariff wall is keeping out much of the cheap Asian steel that crushed 2025, and management has guided for a third straight quarter of rising EBITDA.³¹ Its third-quarter report is the next results day. Its balance sheet carries almost no net debt, and its capital spending is past the peak.³¹⁵ The share price has risen by about 43% over the past year.¹⁰

The next moments that will decide the story are already on the calendar. The first is the start-up of the Pesquería steel shop, which management now expects in early 2027; the first heat of liquid steel will be a technical milestone, but the commercial milestone is the certification of automotive grades and the multi-quarter ramp that follows.⁵ A smooth ramp would validate the most expensive decision in the company's history. Another delay, or another budget increase, would push the payoff further away and test investors' willingness to keep waiting. The second is the outcome of the US–Mexico talks over Section 232 and the USMCA review.

A deal that restores Mexico's exemption would narrow the price gap between the two markets and make Pesquería's steel more valuable to everyone who buys from it; a failure would leave Ternium selling North American steel into a North America that treats it as foreign. The third is the next round in Brazil's courts, where an adverse final ruling in the CSN case would turn a provision into a cheque, and the antidumping decision on Chinese hot-rolled coil, which will decide whether Usiminas's recovery is protected. And the fourth is the 2026 dividend, which will show whether a family that cut the payout in April is ready to share the recovery.

Each outcome maps onto one of the questions this story has been asking. Is the 2026 rebound a cycle or a change in the earnings base? Pesquería's ramp and the margin it sustains will answer that. Will the capital spent earn its keep? Free cash flow in 2027 and 2028, after capex falls to around $1.2 billion, will answer that. How much of the value reaches the outside shareholder? The dividend, the minority share of Usiminas, and the related-party bill will answer that. None of these answers is in yet, which is precisely why the stock trades near book value.

↑ Back to Top

XIV. Outro

Agostino Rocca's post-war, cross-border vision shaped the company his descendants built over eight decades. Now, on that patch of Nuevo León scrubland, they are building a steel shop that inverts it: a bet on being inside the right border, made valuable by a treaty and threatened by a tariff.

After decades of buying from sellers ready to leave, Ternium has bet billions on staying put and waiting for the world to come to it.