Domestic Aluminum Production Viability Constraints: Industrial Electricity Access as Limiting Factor in US Reshoring

Executive Summary

US aluminum smelting expansion faces insurmountable electricity pricing bottlenecks that render tariff protection secondary to power contract availability. Aluminum smelters require stable, long-term power contracts at approximately $40 per megawatt-hour over 10-20 year terms, but most US utilities price electricity at $60-80 per MWh while technology companies capture available capacity at $115 per MWh. The fundamental constraint is structural energy economics, not trade protection, aluminum production requires 11 terawatt-hours annually per smelter (equivalent to powering Boston or Nashville) while competing against data centers and AI infrastructure for grid access. Power purchase agreements have proven determinative for production decisions, with Century Aluminum's Mt. Holly restart enabled by Santee Cooper contract extension through 2031 rather than Section 232 tariff levels.

The Electricity Pricing Constraint

The primary bottleneck preventing aluminum smelter expansion is the fundamental mismatch between industry electricity requirements and US utility pricing structures. Aluminum smelting requires approximately 14,821 kilowatt-hours per metric ton of production, making electricity roughly 40% of total production costs. The industry's maximum viable power price of $40 per MWh creates an insurmountable economic barrier when competing against utilities charging $60-80 per MWh for industrial customers.

This pricing constraint becomes more severe when considering the competition for available electrical capacity. Technology companies, particularly those operating data centers and AI infrastructure, routinely commit to power purchase agreements at $115 per MWh, nearly triple the aluminum industry's maximum viable rate. The Energy Information Administration estimates a 50 terawatt-hour energy deficit by 2035, intensifying competition for limited grid capacity.

The scale requirements compound this challenge. A single aluminum smelter requires approximately 11 TWh of electricity annually, equivalent to powering major cities like Boston, Nashville, or Detroit for an entire year. Most regional grids cannot provide this "city-size scale of clean, affordable energy through stable, long-term contracts," as Century Aluminum executives note.

Long-Term Power Agreements As Determinative Factors

Power purchase agreement availability has proven more determinative than tariff policy for aluminum production decisions. Century Aluminum's Mt. Holly smelter restart in 2025, involving $50 million investment to restore 50,000+ tonnes of capacity, resulted directly from securing power contract extension with Santee Cooper through 2031. This demonstrates that long-term electricity access, not trade protection, enables production viability.

Conversely, Century's Hawesville smelter closure exemplifies how market-rate electricity exposure eliminates production regardless of tariff protection. Despite Section 232 tariff benefits, Hawesville faced power costs reaching "upper-70¢ per MWh" through MISO grid pricing, exceeding viable thresholds and forcing complete curtailment. The facility's market-based electricity arrangement created direct exposure to wholesale power volatility that made operations uneconomical.

The electricity deregulation beginning in 1977 has created the structural foundation for this pricing pressure. The Aluminum Association identifies electricity deregulation as "the single most important factor leading to the near total demise of the primary aluminum industry in the United States". This regulatory shift eliminated the stable, long-term utility contracts that previously supported aluminum smelting operations.

Section 232 Tariffs Vs. Power Contract Fundamentals

The April 2026 restructuring of Section 232 tariffs demonstrates the secondary nature of trade protection relative to electricity access constraints. The modified framework applies 50% tariffs to aluminum articles and 25% to derivatives based on full customs value, representing significant increases from previous metal-content calculations. However, these tariff modifications address import competition rather than the fundamental cost structure that makes US aluminum production uneconomical.

Section 232 tariffs coincided with rising electricity pricing, creating a policy disconnect. While import protection increased, the underlying electricity cost disadvantage, Canadian smelters operating at $25-40 per MWh versus $60-80 per MWh in the US, remained unaddressed. This explains why tariff protection failed to prevent continued smelter closures or enable significant new capacity development.

The tariff restructuring included temporary 15% rates for "metal-intensive industrial equipment and electrical grid equipment" through December 2027, aimed at supporting domestic metal production capacity. However, this provision focuses on equipment imports rather than addressing the electricity infrastructure constraints that prevent new aluminum production facilities.

Grid Access Competition And Infrastructure Constraints

The expanding digital economy creates systematic displacement of energy-intensive manufacturing from available grid capacity. Data centers and artificial intelligence infrastructure capture power contracts at prices aluminum smelters cannot economically match. As industry analysts note, "aluminum producers are being scooped by data centers and hyperscalers" that "can simply pay more for the power".

This competition extends beyond pricing to capacity allocation. The Energy Information Administration's projection of a 50 TWh energy deficit by 2035 indicates structural power shortages that will intensify competition for available electricity. Aluminum smelters requiring 11 TWh annually for single facilities face systematic disadvantage when competing against distributed technology infrastructure.

Grid infrastructure limitations compound capacity constraints. Lead times for electrical equipment necessary for smelter construction "currently exceed five years," creating additional barriers to new facility development. The combination of limited generation capacity, infrastructure delays, and price competition from technology sectors creates multiple overlapping bottlenecks.

Regional variations in power purchase agreement pricing demonstrate geography-specific constraints. Northeast regions average $116.06 per MWh (NYISO) and $111.50 per MWh (ISO-NE), while Texas ERCOT offers $44.00 per MWh. However, even the most favorable regional pricing approaches rather than meets aluminum industry requirements, and available capacity remains limited by technology sector competition.

Counterarguments

Indicators To Watch

Decision Relevance

Scenario A (~70%): Continued electricity pricing pressure without fundamental infrastructure expansion — Recommended: aluminum-consuming industries should diversify supply chains toward Canadian production and secure long-term procurement contracts. Manufacturing companies should evaluate alternative materials where technically feasible.

Scenario B (~25%): Regional electricity market improvements enable limited expansion — Recommended: monitor Texas and Pacific Northwest power markets for capacity availability below $50/MWh. Consider staged procurement strategies that can benefit from limited new domestic capacity while maintaining international supply arrangements.

Scenario C (~5%): Technology sector demand collapse creates electricity capacity availability — Recommended: prepare rapid expansion plans for domestic aluminum procurement if power pricing fundamentally improves. Maintain capability for quick contract negotiations with regional utilities.

Analytical Limitations

• Electricity market pricing data reflects current conditions; energy market volatility could significantly alter cost structures within 6-12 months
• Section 232 tariff effectiveness analysis limited by short implementation period; long-term trade policy impacts may differ from initial market responses
• Regional grid capacity assessments based on public utility data; proprietary power arrangements between major industrial users may not be fully captured
• Technology sector electricity demand projections assume continued AI infrastructure expansion; significant changes in digital economy growth would alter competitive dynamics
• Long-term power purchase agreement availability depends on utility regulatory environments that could change through state-level policy modifications