This report does not constitute a rating action.

Key Takeaways

• Steel tariffs in the U.S. should support the credit quality of American producers with higher domestic prices, volume gains at the expense of imports, and stronger profitability.

• Aluminum tariffs boost the profitability of the four smelters operating in the U.S., with only modest credit benefit to any corporates because of modest prospects for more domestic output.

• Steel prices in the U.S. are up 25%, and the U.S. Midwest aluminum premium hit an all-time high before tariffs were implemented, so profitability downstream already depends on higher prices for fabricated products.

• Comparative advantage matters: The U.S. has large resources of iron, coal, scrap, and spare capacity to make more steel profitably. But aluminum smelters in the U.S. have been closing for decades, requiring cheap electricity and years of capital investment to displace imports.

• Imports accounted for 22% of U.S. steel consumption in 2024 and 88% of aluminum.

Steel And Aluminum Tariffs Affect Credit Differently

Steelmakers in the U.S. have spare capacity and access to inputs to increase volumes by displacing imports, with the profitable umbrella of higher prices for every ton. Tariffs since 2018 have supported the competitive positions and profitability of steel producers in the U.S., which has resulted in numerous upgrades. On the other hand, our ratings on U.S. aluminum producers haven’t moved much in the same timeframe, because primary output continues to decline in the face of lower-cost global supply, while downstream aluminum fabricators consume capital to increase domestic output.

With reliable trade barriers, steel mills in the U.S. could displace half of imports by increasing capacity utilization to a profitable 85% from a sluggish 75%. But imports from Argentina, Brazil, or South Korea could limit that upside by absorbing tariffs with their quota-free access to the U.S. market. Prices in those countries are under pressure from Chinese exports, making the U.S. relatively attractive even with tariffs. And, we believe economic risks are rising as the U.S. trade agenda takes shape (see “Growth Prospects Strained After The U.S. Takes The Tariff Plunge”, published March 5, 2025). Nevertheless, the price of hot-rolled coil steel in the U.S. jumped 25% off a cyclical low in the four weeks from Feb. 4-March 4, while benchmark prices in China and Europe stayed flat. Downstream metals fabricators typically operate on margin-over-metal, so we assume that higher costs get passed through with some lags, working capital increases despite softer demand, and competition limits profit gains in these fragmented subsegments.

The U.S. has little unused aluminum capacity that could displace imports, so all-in costs in the U.S. now rival the supply chain disruptions of 2021-2022. Buyers are paying a 30% premium to the London Metal Exchange (LME) benchmark, as indicated by the U.S. Midwest premium. Higher prices boost the profitability of the four smelters operating in the U.S., but only one or two small facilities could be restarted soon. We estimate that the entire fleet of operating and mothballed primary aluminum smelters at 100% capacity utilization could only replace about 20% of imports, probably with the highest costs in the world. And doing that would require enough electricity to power a small city.

(Editor's note: S&P Global Ratings believes there is a high degree of unpredictability around policy implementation by the U.S. administration and possible responses--specifically with regard to tariffs--and the potential effect on economies, supply chains, and credit conditions around the world. As a result, our baseline forecasts carry a significant amount of uncertainty. As situations evolve, we will gauge the macro and credit materiality of potential and actual policy shifts and reassess our guidance accordingly (see our research here: spglobal.com/ratings).)

Electricity accounts for 30%-40% of the production costs of primary aluminum, so more output will need to outbid data centers for power. Aluminum smelters in the U.S. have been closing permanently for decades as power shifts from this commodity into higher-value manufacturing, residential use, and technology applications. We estimate that less than 1 million metric tons of incremental output is possible from every operating and dormant aluminum smelter in the U.S., compared to 4 million of imports. (Aluminum is measured in metric tons, which is 1,000 kilograms or 2,205 pounds; steel in the U.S. is measured in short tons, which is 2,000 pounds.) Alcoa and Century Aluminum own the four operating smelters. Alcoa is the largest aluminum producer domiciled in the U.S., but most of its primary metal output is in Canada, so it faces a tariff on volumes it ships into the U.S. Century Aluminum has most of its aluminum capacity in the U.S., so it should get a significant profit uplift from tariffs. U.K.-based Rio Tinto is the largest aluminum producer in Canada and exports to the U.S., where it produces no aluminum.

Steel Tariffs Deepen The Competitive Moat Around U.S. Producers

Tariffs under Section 232 of the Trade Expansion Act have supported the competitive position of U.S. steel assets since 2018, along with factors such as consolidating domestic production and, until recently, slower production growth from China. Consequently, we’ve raised the ratings on both Cleveland-Cliffs and U.S. Steel three times in three years, and on Steel Dynamics twice in five years. Even with customary high volatility, steel prices stayed about 20% higher than in past downturns, so the profitability of the highest-cost, lowest-margin producers in the U.S. is better than in previous downturns.

The U.S. has imported 20%-30% of its steel for many years . This has dropped to 23% of U.S. consumption since 2019 from 29% in the five years prior to 2018. Concurrently, hot-rolled coil steel prices averaged $914/ton versus $621/ton, a 47% jump. Importantly, U.S. steelmakers have excess capacity and continue building minimill capacity that can quickly and profitably take share from imports.

Tariffs Alone Can’t Spark U.S. Aluminum Output

In 2019, we noted that, “Tariffs probably won't increase U.S. primary metal output much, because of limited low-cost electricity to spark production” (see “Global Trade At A Crossroads: Section 232 Tariffs Leave Dents In U.S. Aluminum Credit Quality”, published April 1, 2019). U.S. aluminum capacity and output were lower in 2024 than in 2017. The U.S. produced 670,000 metric tons of primary aluminum in 2024 and imported 4 million metric tons, mostly from Canada.

Compared to steel, increasing aluminum output in the U.S. will be difficult and expensive. Using current capital projects for reference, replacing 4 million metric tons of imported primary aluminum at $5,000/metric ton of capacity would cost at least $20 billion for smelters alone. As examples, Rio Tinto is spending $1.1 billion over 2½ years for a 160,000-metric-ton expansion in Quebec, about $7,000/metric ton, and Indonesian coal producer PT Adaro is constructing a 500,000-metric-ton smelter for $2 billion over two years, about $4,000/metric ton. Any large new smelter in the U.S. would likely take longer to gain permits and be more expensive than both, so $20 billion - $30 billion over 10 years for that much smelting capacity is probably conservative. And that’s before any investment in power.

The International Aluminium Institute (IAI) estimates that 13,000-16,000 kilowatt hours of electricity are required to produce a metric ton of aluminum, so 4 million metric tons of new annual smelting capacity could need 50 terawatt hours (TWh) of power every year. For comparison, New York City consumed about 50 TWh of electricity in 2024. S&P Global Ratings estimates incremental U.S. power demand from data centers could be 150-250 TWh from 2024-2030, potentially requiring about $60 billion of investment in generation and $15 billion in transmission (see “Data Centers: Surging Demand Will Benefit And Test The U.S. Power Sector”, published Oct. 22, 2024). New smelters to replace U.S. aluminum imports could add one-third to that.

Tariffs Add To Costs For U.S. Metal Buyers

Steel and aluminum prices have consistently been 20% above the 10-year average in the wake of the COVID-19 pandemic, which aligns with most metals and elevated production costs. Now, steel prices in the U.S. are 30% higher than other global markets. Also, the Platts spot 99.7% P1020 U.S. Aluminum Transaction Premium (i.e., the Midwest premium) has more than doubled since tariffs were first floated in November 2024 (Platts is part of S&P Global Commodity Insights). The Midwest premium has historically accounted for normal frictional costs such as transportation from the coasts to manufacturing inland, but the record premium of $900/metric ton equates to 35% of a $2,500 LME global benchmark price. Tariffs have boosted metals costs for U.S. manufacturers since 2018, on top of a decade of rising labor costs and a shift in manufacturing to lower-cost jurisdictions (see “Evolving Risks For Credit Quality In U.S. Capital Goods”, published July 18, 2024).

Stacking tariffs, such as another 25% on Canada and Mexico imports, might support higher prices, but it more likely enables lower-tariffed material from other countries, such as Brazil or South Korea, to supplement higher-cost U.S. output. We also believe this would increase production costs, considering scrap metal flows between the U.S., Mexico, and Canada. Most of the steel in the U.S. is produced using scrap in electric arc furnaces, and an increasing share of aluminum rolled products consumes scrap aluminum because scrap consumes 90% less energy with excellent metallurgical properties.

Canada produced 5x more primary aluminum than the U.S. in 2024, and China produced 10x more than Canada and the U.S. combined. China dominates global output with 43 million metric tons in 2024. Canada has nine smelters operating and China has more than 50 versus the four in the U.S. China is by far the largest producer and exporter of steel and aluminum. Its production has been a dominant driver for aluminum prices for several decades. The four U.S. smelters combined to produce 670,000 metric tons of primary aluminum in 2024 whereas each of the four largest in Canada can produce at least 400,000 metric tons. In total, Canada produced 3.3 million metric tons in 2024 and exported 90% of that to the U.S. to satisfy more than half of U.S. aluminum consumption of 5 million metric tons.

Recycling Aluminum Looks Like A Winner

The lowest-cost path to displacing some aluminum imports is likely domestic scrap. The Aluminum Association in the U.S. estimates that only 43% of consumer aluminum beverage cans are recycled, putting 61 billion cans, or about $1.2 billion worth of aluminum in 2023, into landfills. That waste is a valuable, low-cost resource in a $12 billion-$15 billion market dominated by imports. Further, the IAI estimates the value of that recycled aluminum at $1,338/metric ton in 2023 compared to the LME price plus Midwest premium today of about $3,500/metric ton. Tariffs appear to provide a bigger price incentive to divert aluminum to recycling from American landfills. Recycling consumes about 90% less energy than primary aluminum. The U.S. remelted 1.3 billion pounds of cans in 2023, or about 600,000 metric tons. If the U.S. increased its beverage can recycling rate to Canada’s 71% or Maine’s 83%, those 400,000-500,000 extra metric tons could displace about 15% of total imports, or the equivalent of one large smelter in Canada.

Even with higher costs, downstream aluminum customers continue investing in rolling capacity to satisfy demand tailwinds in transportation and packaging. Most capacity for these facilities will rely on scrap aluminum, including post-consumer used beverage cans. Novelis and Steel Dynamics are both building world-scale aluminum rolling mills in the U.S., which would be the first greenfield mills in North America in about 40 years. Steel Dynamics is building a 650,000-ton aluminum flat rolled mill and two recycled aluminum slab sites for an estimated $2.7 billion in Columbus, Mo. The site is expected to start later in 2025 with recycled aluminum as the primary feedstock.

Steel Dynamics has experience executing projects of this scale, but aluminum rolling is a departure from electric arc furnaces, with products that serve different customers. Novelis is building a 600,000-ton recycling and rolling mill in Bay Minette, Ala. Highlighting the complexity of these large-scale greenfield projects, the company revised its estimated capital costs to $4.1 billion from $2.5 billion since breaking ground in October 2022. The facility is slated to start serving the can sheet, automotive, and specialty markets in the second half of 2026. The escalated capital cost will increase Novelis’ cash flow deficits and could raise debt leverage.

Steel And Aluminum Are More Different Than Some Think

Steel and aluminum are being lumped together in trade policy, but these markets are different. The U.S. consumes about 100 million tons of steel, and we assume 2025 prices of about $800/ton for industry revenue of about $80 billion (before value-added processing). The U.S. consumes about 5 million metric tons of aluminum, and we assume about $2,500/metric ton for industry revenue of about $12.5 billion. The metals differ in application and demand, because they differ in properties including weight, strength, malleability, corrosion, and conductibility. They also differ in output and supply chains. They compete on the margin in industries such as automotive and construction, but steel dominates in heavy, large-volume applications because it’s less expensive and robust. Aluminum leads in more precision, lighter weight applications. And the two markets are fed by a primary stream and a recycled stream of metal, both with different costs and economic sensitivities.

Steel is cheaper and easier to make than aluminum, especially in the U.S. Iron ore and metallurgical coal are the ingredients for carbon steel, using coal to reduce iron inside blast furnaces. The recipe dates back 3,000 years, and the U.S. has a natural endowment of raw materials with a short domestic supply chain. For example, the U.S. produced about 80 million tons using iron from Minnesota, coal from Appalachia, and scrap from across the nation. About 70% of the steel in the U.S. is made from scrap. By comparison, China produced about 1 billion tons of steel in 2024 using iron and coal from Australia, Brazil, and Canada to supplement domestic resources. About 10% of the steel in China is made from scrap.

Large-scale aluminum production is recent compared to steel, with the first commercial output using Hall-Héroult electrolytic reduction in about 1900. Supply chains for aluminum are global, complex, and much longer than for steel. Aluminum’s core element, bauxite, is mined from regions of elevated concentration such as Australia, Guinea, China, and Brazil. Bauxite is a freely trading commodity, but much of the world’s production is refined near its source into alumina (also called aluminum oxide) using caustic soda. Alumina is then transported or sold to areas of lower-cost, reliable electricity for smelting into aluminum.

Here, we provide a primer on various subsegments of the commodity steel and aluminum markets.

Commodity steel and aluminum market breakdown 

Related Research

• S&P Global Ratings Metal Price Assumptions: Holding Firm Through First Waves Of Tariffs, March 5, 2025
• Growth Prospects Strained After The U.S. Takes The Tariff Plunge, March 5, 2025
• U.S. Capital Goods Brief: Tariffs Would Test Pricing Power, Feb. 6, 2025
• Data Centers: Surging Demand Will Benefit And Test The U.S. Power Sector, Oct. 22, 2024
• Evolving Risks For Credit Quality In U.S. Capital Goods, June 18, 2024
• Global Trade At A Crossroads: Section 232 Tariffs Leave Dents In U.S. Aluminum Credit Quality, April 1, 2019