Rosa Del Mar

Daily Brief

Issue 78 2026-03-19

Reshoring Pattern: Downstream First, Upstream Constrained

Issue 78 Edition 2026-03-19 8 min read
General
Sources: 1 • Confidence: Medium • Updated: 2026-04-11 17:30

Key takeaways

  • Current U.S. module assembly capacity is described as sufficient to supply roughly 40–50 GW per year of U.S. demand.
  • First-of-kind U.S. wafer/cell factories are described as slowed by lack of experienced contractors and permitting authorities, requiring education and capability-building to proceed smoothly.
  • The Inflation Reduction Act is characterized as the key change that made reshoring U.S. solar manufacturing economically feasible beyond what prior trade policy could achieve.
  • China’s property-sector crash is described as prompting a pivot toward export-oriented manufacturing in solar, batteries, and EVs that oversupplied global markets and reduced the follow-on pace of U.S. solar manufacturing investment.
  • U.S. polysilicon capacity is characterized at roughly 10–20 GW and is supplied by only a few major suppliers.

Sections

Reshoring Pattern: Downstream First, Upstream Constrained

  • Current U.S. module assembly capacity is described as sufficient to supply roughly 40–50 GW per year of U.S. demand.
  • Module assembly is described as easier to onshore because shipping bulky panels is costly and module assembly is faster and less capital- and infrastructure-intensive than wafer/cell production.
  • Most solar panels sold into the U.S. market are now at least domestically assembled.
  • U.S. wafer manufacturing is described as limited to two companies: Qcells and Hemlock Corning, with Hemlock Corning about to start.
  • U.S. cell manufacturing capacity is described as larger than U.S. wafer manufacturing capacity.

Cost Drivers And Execution Bottlenecks In U.S. Factories

  • First-of-kind U.S. wafer/cell factories are described as slowed by lack of experienced contractors and permitting authorities, requiring education and capability-building to proceed smoothly.
  • U.S. capital markets are described as less suited to China-style rapid factory churn and oversupply shakeouts because investors underwrite assets as 20-year projects and expect longer-duration certainty.
  • Higher U.S. wafer/cell/module costs are described as driven largely by higher construction and project development costs (structure, steel, contracted labor, permitting) rather than uniquely higher equipment costs.
  • Manufacturers are described as needing long-term offtake assurances on the order of decades to justify U.S. factory investment given rapid economic shifts in the industry.
  • Because global solar manufacturing is already mature and scaled, especially in China, new U.S. factories are described as needing to compete on cost immediately rather than relying on early-stage technology premiums.

Policy As Economic Enabler; Implementation Uncertainty As Drag

  • The Inflation Reduction Act is characterized as the key change that made reshoring U.S. solar manufacturing economically feasible beyond what prior trade policy could achieve.
  • Even with 45X incentives, U.S. manufacturing is described as not yet reaching prevailing global module pricing of roughly 7–9 cents per watt.
  • Domestic solar manufacturing is described as currently requiring industrial policy support because customers will not voluntarily pay substantially higher prices, exemplified as about 30 cents/W for U.S. modules versus about 10 cents/W globally, without demand-shaping mechanisms and continuity.
  • Reshoring progress is described as slowed by delayed and changing policy implementation, including slow issuance of domestic-content guidance and subsequent changes tied to OBBBA.

Global Competition, Oversupply, And Price Spreads

  • China’s property-sector crash is described as prompting a pivot toward export-oriented manufacturing in solar, batteries, and EVs that oversupplied global markets and reduced the follow-on pace of U.S. solar manufacturing investment.
  • Domestic solar manufacturing is described as currently requiring industrial policy support because customers will not voluntarily pay substantially higher prices, exemplified as about 30 cents/W for U.S. modules versus about 10 cents/W globally, without demand-shaping mechanisms and continuity.
  • The hardest choke point in scaling U.S. ingot/wafer/cell capacity is described as economics under global competition, including competition from imported inputs and foreign-headquartered firms with U.S. factories.
  • Two solar industry rules of thumb are described: demand tends to grow faster than expected while prices tend to fall faster than expected, producing rapid shifts in factory economics.

Upstream Materials: Limited Polysilicon Scale And Supplier/Product Mix

  • U.S. polysilicon capacity is characterized at roughly 10–20 GW and is supplied by only a few major suppliers.
  • REC Silicon is described as having shifted production toward silane gas rather than solar-grade polysilicon.
  • Wacker is described as sourcing most solar-grade polysilicon from Germany rather than the U.S.
  • Polysilicon manufacturing economics are described as driven predominantly by electricity costs, advantaging U.S. regions with cheap power.

Watchlist

  • Solar PPA prices are described as rising over the past two years.

Unknowns

  • What is the current verified operating and utilized U.S. module assembly capacity, and how does it compare to actual annual U.S. installations (not forecasts)?
  • What are the verified operating, under-construction, and near-term commissioning timelines and ramp performance for U.S. wafer and cell lines (including the two wafer players named)?
  • What are the current U.S. vs global realized module ASPs and full costs (with and without 45X), and how stable is the cited 7–9 cents/W global benchmark over time?
  • How much of the U.S. manufacturing cost disadvantage is attributable to permitting duration, labor/steel costs, and construction execution versus scale effects and learning curves?
  • What specific domestic-content guidance delays and OBBBA-linked changes occurred, and what were their effective dates and practical impacts on project financing and factory FIDs?

Investor overlay

Read-throughs

  • Near term U.S. solar manufacturing localization may concentrate in module assembly, while deeper domestic content remains constrained by wafer and possibly polysilicon scale, shifting value and bottlenecks upstream rather than downstream.
  • Execution capacity and permitting friction may be a repeatable cost and timeline driver for first of kind wafer and cell plants, making project schedules and ramp performance more determinative than technical feasibility.
  • Global oversupply and low pricing may suppress follow on U.S. investment pace even with IRA support, increasing sensitivity of factory final investment decisions to realized U.S. versus global price spreads and incentive stability.

What would confirm

  • Verified operating and utilized U.S. module assembly capacity aligns with or exceeds actual annual U.S. installations, while wafer availability remains a limiting input relative to cell and module capacity.
  • Documented commissioning timelines and ramps for U.S. wafer and cell lines show delays attributable to permitting, contractor capability, and first of kind execution, with measurable improvements as stakeholders gain experience.
  • Observed realized U.S. versus global module ASPs and full costs, with and without 45X, show persistent spreads and volatility consistent with oversupply dynamics affecting investment timing.

What would kill

  • Verified data shows U.S. module assembly utilization materially below implied levels due to demand, financing, or other constraints, undermining the view that downstream capacity is sufficient for near term needs.
  • U.S. wafer and cell projects demonstrate on time delivery and rapid ramp without notable permitting or contractor bottlenecks, weakening the claim that execution capacity is a primary constraint.
  • Price and cost data indicate the global benchmark and U.S. versus global spreads are not persistent, or policy execution issues around domestic content guidance do not measurably affect project financing or factory FIDs.

Sources

  1. 2026-03-19 traffic.megaphone.fm