Safe Harbor Protects Your Tax Credit. It Doesn't Protect Your Module Price.
In June, Wood Mackenzie published its US Solar Market Insight Q2 2026 outlook, produced with SEIA. The headline was a puzzle: demand is strong, the pipeline is deep, and yet the forecast is flat.
Their base case called for roughly 43 GWdc of annual solar additions through 2031 — enough to double the US fleet in five years, but a plateau in annual volume. The previous doubling took three.
Analyst Zoë Gaston named the constraints holding it there: interconnection queues, permitting, expiring tax credits, and trade uncertainty. On that last one, she flagged a specific open question — an anticipated Section 232 action on polysilicon that "could severely constrain manufacturing activity for some domestic producers."
That question got answered on August 7.
What the June outlook established
Three things from Wood Mackenzie's analysis are worth carrying forward, because they frame what the tariff actually lands on:
The demand side is not the problem. Solar made up 60% of all new US generating capacity in Q1 2026. Paired with storage, the two accounted for 91%. Contracting activity grew 15% year over year. Projects are getting built on schedule.
There is a large safe-harbored buffer. Wood Mackenzie estimates 216 to 240 GWdc of utility-scale capacity is safe harbored, most of it locked in before FEOC requirements took effect. Even after normal attrition, that's enough to carry utility-scale buildout through 2030.
The cell bottleneck was already the weak link. Domestic module assembly has scaled to roughly 70% of 2025 installation volume. Domestic cell capacity sits at about 3 GWdc. US module factories run on imported cells, and six countries subject to trade action supplied 78% of those cell imports last year.
That third point is the one that matters most right now.
What changed on August 7
The Section 232 order applies a 15% tariff plus minimum import prices across the chain — including $0.22/W on cells and $0.38/W on modules — effective December 4, 2026.
Read that against Wood Mackenzie's June picture and the interaction is clear. A domestic module industry that depends on imported cells just had the cost of imported cells set by floor rather than by market. The 3 GWdc domestic cell gap that was already constraining things is now a gap with a price attached to it.
Wood Mackenzie's flat forecast was built with this as an open variable. It resolved toward the constraining end of the range.
What that means if you're procuring
The safe-harbored pipeline is a schedule, not a shield. Safe harbor protects tax credit eligibility. It does not protect equipment cost. A project that safe harbored in 2025 and orders modules in Q1 2027 pays post-232 prices on that equipment. If your financial model assumed 2025-era module pricing because the project was safe harbored, that model needs a second look.
The December 4 line runs through your pipeline. Orders landing before it price under current rules. Orders after it do not. For anyone with flexibility on timing, that's the highest-leverage decision available between now and then.
Cell sourcing is the question behind the question. Wood Mackenzie's point about 3 GWdc of domestic cell capacity was already the structural weak point in the domestic manufacturing story. Anyone specifying domestic modules should understand where the cells in those modules come from, because that determines both cost exposure and qualifying status.
Expect the forecast to move. Wood Mackenzie noted their outlook had shifted only 1.4% from the prior quarter — remarkable stability. That stability was priced with 232 unresolved. The Q3 update is worth watching.
The practical read
The strong-demand, flat-installation puzzle Wood Mackenzie described in June has not gone away. What's changed is that one of the four constraints they identified now has a number and a date attached.
For the next four months, that makes procurement timing a live strategic decision rather than a logistics detail. After December 4, it makes supplier breadth and documentation quality worth considerably more than they were worth in June.
Neither of those is solved by having more product sitting in a warehouse. They're solved by knowing where qualifying product actually is, what it qualifies for, and how quickly it can reach a site.
Where we fit
Exel Solar US is a sourcing and logistics partner. We work across a broad supplier network rather than a single line card, with hubs in California, Texas, and Ohio.
If you're re-running numbers on a safe-harbored project, or working out what your cell sourcing exposure actually looks like, we're glad to think it through with you. Schedule a conversation.
Market data and forecasts in this article are from Wood Mackenzie's US Solar Market Insight Q2 2026 report, produced in partnership with SEIA and summarized here. Those figures represent Wood Mackenzie's estimates and were published on June 10, 2026, prior to the Section 232 announcement. Tariff details are from the White House fact sheet of August 7, 2026. This article is provided for general information and does not constitute pricing, availability, or procurement advice for any specific project.
