The Inflation Reduction Act created a bonus tax credit for using American-made equipment. For roughly the first year of its existence, few projects could claim it in practice, and the obstacle was not equipment availability. It was that the calculation required information no manufacturer was willing to hand over.
That problem has now been solved, twice, most recently last month. For any commercial or municipal project entering design this year, the domestic content bonus has moved from theoretical to bankable, and it is worth understanding why.
For an investment tax credit project meeting prevailing wage and apprenticeship requirements, satisfying domestic content adds 10 percentage points. A 30 percent credit becomes 40 percent. On a $3 million storage installation, that is $300,000.
One point of frequent confusion: for production tax credit projects, the bonus increases the credit by 10 percent as a multiplier rather than adding 10 percentage points. For investment credits it is 10 points added, or 2 points if prevailing wage and apprenticeship requirements are not met.
Qualifying requires clearing two separate bars.
The first is the steel and iron requirement, and it is absolute. All manufacturing processes for any structural steel or iron component must occur in the United States, from melting through coating. For a ground-mount project that captures steel racking, piles and ground screws, and reinforcing products in foundations. There is no percentage and no partial credit. Nuts, bolts, screws, clamps, and cabinets are excluded as non-structural.
The second is the adjusted percentage rule, which applies to manufactured products. A project must show that domestically produced manufactured products and components account for at least a specified share of total manufactured product cost. For investment credit projects that threshold is 40 percent. The tech-neutral production credit carries an escalating schedule that rises toward 55 percent later in the decade.
Notice 2023-38, issued in May 2023, established the framework. The domestic cost percentage was defined as domestic manufactured product costs divided by total manufactured product costs, where both figures used only direct costs, meaning direct materials and direct labor, as incurred by the manufacturer.
That last clause is the problem in a sentence. The calculation did not use what the developer paid. It used what the manufacturer spent. A developer claiming the bonus therefore needed each supplier to disclose its own direct material and labor costs at the component level.
Manufacturers treat that data as among their most closely held commercial information. A foreign manufacturer had a particularly clear reason to decline, since the information would be used to help an American buyer claim a tax benefit premised on not buying from them.
The result was a bonus that existed on paper and was effectively unbankable. Tax equity investors and insurers would not underwrite a credit resting on a calculation the taxpayer could not substantiate. Projects plainly using American racking and American-assembled modules could not prove it to the standard the guidance required.
Notice 2024-41, issued in May 2024, solved this by removing the manufacturer from the equation.
The Department of Energy developed tables assigning a fixed cost percentage to each listed component of a representative project. A taxpayer using the safe harbor needs no cost data from anyone. They identify which listed components are US-manufactured, add up the assigned percentages, and compare the total to the threshold.
Notice 2025-08, issued January 16 of this year, updated those tables. It split solar photovoltaic into separate ground-mount and rooftop tables, revised battery storage percentages using 2024 cost data from three national laboratories, and added a second set of columns available to projects using domestically produced crystalline silicon cells made from domestically produced wafers.
Taxpayers may elect either the 2024 tables or the new ones, but must apply one consistently.
For a ground-mounted tracker project, the current table assigns 38.0 percent to photovoltaic module cells, 11.0 percent to the tracker torque tube, 6.0 percent each to the module frame and front glass, 4.77 percent to module production, and smaller shares down through bypass diodes at 0.4 percent. The full column sums to approximately 100.
Four rules catch people out.
Components not listed in the table contribute zero, regardless of where they were made. Production line items are all-or-nothing: module production counts only if the production occurred domestically and every listed component going into that module is also domestic, so domestically assembling foreign cells earns nothing on that line. Mixed sourcing is prorated by nameplate capacity. And the election applies at the project level, so you cannot use safe harbor percentages for some components and actual costs for others.
The new domestic wafer columns look like an unambiguous upgrade. They are not, and the examples in Notice 2025-08 make the point directly.
Take a 100 MWdc tracker project with 60 MWdc of US modules and US torque tubes on 80 percent of the trackers. Electing the domestic wafer columns, where module cells are valued at 51.6 percent rather than 38.0 percent, the project scores 50.9 percent.
Now take the same project, same facts, and decline the election. It scores 63.5 percent.
Nothing changed but the election, and the lower score belongs to the version using the more generous-looking table. The reason is that electing the domestic wafer columns requires treating domestically produced cells that do not exclusively use domestic wafers as not being US components, which makes the modules containing them non-US manufactured products. You forfeit them entirely.
A developer with mixed wafer sourcing can therefore score materially higher by declining an election that appears designed to reward them. Run both calculations before filing.
The practical shift is that domestic sourcing has become a design decision with a quantifiable value, rather than a preference.
On a ground-mount project, module cells and the tracker torque tube together represent roughly half the assigned percentage. Those two line items are where the bonus is won or lost, and they are procurement decisions made early. A project team that gets to detailed design without having asked the domestic content question has usually already foreclosed it.
For storage, the table is dominated by battery pack cells, which makes the analysis simpler and the sourcing harder.
Three things worth doing on any project in development:
Ask suppliers for their domestic content position at the quoting stage. A supplier who cannot document it is not a cheaper supplier; they are a supplier whose price does not include the bonus you were counting on.
Run the safe harbor arithmetic before the equipment package is locked, not after. The difference between clearing 40 percent and missing it is often one component.
For public agencies, remember that elective pay lets a district monetize the credit directly, which means the bonus is real cash rather than a tax attribute you cannot use.
