Basis Reduction and MACRS: The Half-Credit Haircut, and the Date That Decides Your Bonus Depreciation
Through 2023 and 2024, a great many developers signed equipment supply contracts earlier than they otherwise would have, to establish beginning of construction and lock in a credit rate before the rules moved. Series A covered why that was rational and how the evidence file gets assembled.
Here is the part nobody was optimising for at the time. The One Big Beautiful Bill Act restored 100% bonus depreciation permanently — but only for property acquired and placed in service after 19 January 2025. Property acquired on or before that date, and placed in service after it, stays on the old phase-down schedule.
A binding supply contract signed in 2024 to protect the credit may therefore have locked the project out of 100% bonus depreciation. On a $250m project, that is roughly $4m of present value, given away in pursuit of something else entirely.
This post closes Series B with the mechanism underneath the whole series: what the ITC does to depreciable basis, how that basis is recovered, what OBBBA changed, and why 100% bonus makes the tax equity basis problem harder rather than easier.
ℹ️ Note: This describes how these rules work in practice. It is not tax advice — acquisition dates, binding contract analysis and the transitional elections are fact-specific.
What Does the Basis Reduction Do?
It removes half the credit from depreciable basis. Where the ITC is claimed, the property's tax basis is reduced — for all tax purposes, including depreciation and the calculation of gain on a sale — by one-half of the credit amount.
Stoel Rives puts the result in the form worth memorising: qualifying solar components "typically have a tax basis of 85 percent of their original cost when the ITC is claimed."
The arithmetic is simple enough to do in your head. A 30% credit, halved, is 15% of cost. Subtract that from 100% and 85% is what remains to depreciate.
This is the half-credit rule, and it is specific to the energy credit. It is worth stating plainly because it is routinely modelled wrongly in both directions: models that reduce basis by the full credit understate depreciation by fifteen points of cost, and models that ignore the reduction entirely overstate it by the same.
How Is the Property Depreciated?
Over five years, on a double declining balance method, under MACRS.
Qualifying components of a solar facility, and qualifying storage, are five-year MACRS property — "greatly accelerated depreciation deductions... typically over a five-year period based on the double declining balance method." That is against a building's 27.5 or 39 years, which is the comparison that makes the treatment remarkable.
Five-year MACRS under the half-year convention runs across six tax years:
| Year | MACRS % |
|---|---|
| 1 | 20.00% |
| 2 | 32.00% |
| 3 | 19.20% |
| 4 | 11.52% |
| 5 | 11.52% |
| 6 | 5.76% |
ℹ️ Note: The half-year convention is the default, not a guarantee. Where more than 40% of the year's additions are placed in service in the fourth quarter, the mid-quarter convention applies instead and the year-one deduction shrinks. For a portfolio owner with a December-heavy COD schedule this is a real effect, and it is one of the few reasons to push a commissioning date earlier rather than later.
What Goes Into the Basis in the First Place?
Before either rule applies, something has to decide how large the basis is — and that question drives both the credit and the deductions off the same number.
Basis is generally the cost of the property, and may also include a capitalised portion of related costs such as permitting, engineering, and interest during construction. On the equipment side, eligible property covers panels, inverters, racking and balance of system, step-up transformers, circuit breakers and surge arrestors, state and local sales and use taxes on equipment and installation, and storage devices of 5 kWh or greater.
Two boundaries are worth knowing precisely.
Land is out. The portion of a purchase price reasonably assigned to land is eligible for neither the credit nor depreciation. In an acquisition of an operating project this allocation is not a formality — it moves both numbers at once, and it is exactly the sort of line an appraisal has to support.
Interconnection costs have a size cliff. The IRA made interconnection property a qualified expense, but for projects of 5 MW AC or less. A 4.9 MW project can include the interconnection property costs it incurred to enable distribution or transmission; a 6 MW project cannot.
That threshold deserves a moment. Interconnection costs do not scale smoothly with project size, and network upgrade obligations can be substantial for a project just over the line. A developer sizing a facility at 5.5 MW to capture a little more energy may be giving up credit and depreciation on the entire interconnection spend to get it — a trade that is rarely run explicitly, because the interconnection budget and the tax model usually live in different workbooks.
The wider point is the one that runs through this series: basis is not an accounting output arrived at after the deal. It is the input that determines the credit, the depreciation, the DRO and the step-up simultaneously, which is why the appraisal supporting it is the most diligenced document in a hybrid and why the excessive transfer penalty attaches to getting it wrong.
Does the Same Apply if You Elect the PTC?
No, and that asymmetry is a genuine input to the election.
The half-credit basis reduction is a feature of the investment credit. A project electing the production tax credit claims no investment credit, so there is nothing to halve — the full cost remains depreciable.
Put the two side by side on a $250m project:
| ITC elected | PTC elected | |
|---|---|---|
| Depreciable basis | $212,500,000 | $250,000,000 |
| Basis foregone | $37,500,000 | — |
| PV of that difference at 21% / 8% | −$7,291,667 | — |
| Credit timing | Year 1, one amount | Over 10 years, on generation |
| Recapture exposure | Five-year vesting | None |
The ITC is usually still larger in present value on a high-capex, moderate-yield project, which is why solar has historically elected it. But the honest comparison has three components, not one: the credit itself, the depreciation given up to claim it, and the recapture tail the previous post priced at roughly two and a half cents in the transfer market.
Run all three and the answer moves. A high-capacity-factor project selling credits to a buyer who will charge for recapture risk, insurance and indemnity can find the PTC nets more — while a model comparing gross credits alone will report the opposite, confidently, every time.
What Did OBBBA Change?
It brought 100% bonus depreciation back, and made it permanent.
The One Big Beautiful Bill Act, enacted 4 July 2025, permanently reinstated 100% bonus depreciation under §168(k). BDO's summary of the effective date is the operative sentence: it applies to property "acquired and placed in service after January 19, 2025."
That replaced a schedule that had been winding down for years — 80% in 2023, 60% in 2024, 40% in 2025, 20% in 2026, and nothing after 2026. Any model, memo or website still carrying that phase-down is describing law that no longer governs most new property, and there are a great many of them.
For a five-year MACRS asset, 100% bonus means the entire depreciable basis is deducted in the year the property is placed in service. The six-year table above becomes a single line.
The Acquisition Date Trap
Here is where the earlier Series A behaviour comes back.
The test has two limbs, not one. Property must be acquired after 19 January 2025 and placed in service after that date. BDO is explicit about what happens otherwise: property "acquired on or before January 19, 2025, and placed in service after that date remains subject to the bonus depreciation phase-down rules as provided under prior law" — the old declining percentages of 40%, then 20%, then zero.
Now consider the standard safe-harbouring pattern. A developer signs a binding supply contract in mid-2024 to establish beginning of construction and protect its credit rate. The equipment is delivered in 2025 and the project reaches commercial operation in 2026.
On those facts the property was acquired in 2024. The project is placed in service in 2026. It is not eligible for 100% bonus depreciation, and under the prior-law schedule it gets 20% — with the remaining 80% of basis recovered over the ordinary six-year MACRS table.
The two rules were designed for different purposes and nobody reconciled them. The beginning-of-construction rules reward contracting early. The bonus depreciation transition rule punishes it. A developer who did exactly what the credit rules encouraged in 2024 may find the depreciation rules treating that same contract as the reason it gets a fifth of the bonus instead of all of it.
There is also a transitional election running the other way: taxpayers may elect 40% bonus instead of 100% for the first tax year ending after 19 January 2025, with 60% available for long production period property. That is an election to take less, which sounds perverse until you remember the basis gates — and we will come back to it.
Why 100% Bonus Makes the Tax Equity Problem Worse
This is the counterintuitive result, and it runs directly into the four gates from the outside basis post.
A full deduction in year one is worth more in present value terms. It is also a much larger year-one loss allocation, and a tax equity investor allocated 99% of it has to be able to use it. The basis post set out the sequence of gates it must pass — §704(d) outside basis, §465 at-risk, §469 passive activity, §461(l) excess business loss — and a deduction that cannot clear them is not worth its face value.
Three consequences follow.
The DRO gets bigger. The deficit restoration obligation exists to support allocations that drive a capital account negative. Compressing six years of depreciation into one drives it negative faster and deeper. This is the mechanism behind the range the basis post reported — DRO caps running from 15% to well over 100% of investment, driven by the depreciation election.
Suspended losses become more likely. A loss that fails a gate in year one is suspended, and the basis post's closing point was that suspended losses are extinguished on exit. Accelerating deductions into a single year concentrates them precisely where the gates are tightest, because outside basis is at its lowest before any income has been allocated.
The transitional election starts to make sense. Electing 40% instead of 100% spreads the deduction and may produce more usable benefit for an investor whose gates bind. That is a genuinely counterintuitive answer that only a model running both ledgers can produce, and it is the reason the election exists as a choice rather than a default.
The general point is one worth carrying beyond this series: acceleration is only valuable to a taxpayer who can absorb it. In a partnership with constrained partners, the faster schedule can be the worse one.
What Happens on a Sale?
The basis reduction comes back as ordinary income.
Section 50(c) provides that for purposes of §§1245 and 1250, the basis reduction "shall be treated as a deduction allowed for depreciation." The fifteen points of basis that were never depreciated are nonetheless treated as though they had been, so gain on a sale is recaptured as ordinary income to that extent rather than being taxed as capital gain.
Combine that with 100% bonus and the position on an early sale is stark. The entire depreciable basis has been deducted, the basis-reduction amount is treated as deducted, and the property's adjusted basis is at or near zero. A sale in year three produces ordinary income on essentially the whole price.
This is not an argument against the deductions. It is an argument for modelling the exit correctly: a project sold inside the depreciation life converts a large accelerated deduction into a large ordinary income event, and a model that shows the year-one shield without the exit recapture is showing half the transaction.
And If the Credit Is Recaptured?
The basis reduction reverses, to the extent of the recapture.
The previous post covered the mechanics and the honest assessment of what that offset is worth. It is a deduction rather than a credit, so it returns the tax rate rather than a hundred cents. It arrives over the remaining recovery life rather than immediately, while the recapture tax is due now. And after a §6418 transfer it lands in a different entity from the one paying the tax.
Under 100% bonus the timing problem is sharper still, because there is no remaining recovery life to speak of — the deduction has already been taken. The restored basis is recovered against a schedule that has effectively finished.
How Do You Model This in Excel?
As two separate questions: how much basis is there, and how fast is it recovered.
The inputs
Assumptions, labelled as such:
Eligible cost $250,000,000
ITC rate (with PWA) 30%
Gross credit $75,000,000
Basis reduction = 50% × credit $37,500,000
Depreciable basis $212,500,000 (85% of cost)
Tax rate 21%
Discount rate 8%
What the credit costs in depreciation
Lost depreciation = 37,500,000 × 21% = $7,875,000
PV at 100% bonus, yr 1 = 7,875,000 ÷ 1.08 = $7,291,667
The credit is $75m and it costs about $7.3m of depreciation value to claim. Net, call it $67.7m. That is the number to compare against a PTC election, not the gross credit — and it is a comparison most models never make because the basis reduction sits in a different tab from the credit.
The two bonus cases
100% bonus (acquired after 19 Jan 2025):
Year 1 deduction = $212,500,000
Tax value = $44,625,000
PV at 8% = $41,319,444
20% bonus (acquired on or before 19 Jan 2025, PIS 2026):
Year 1 42,500,000 + 170,000,000×20.00% = $76,500,000
Year 2 170,000,000 × 32.00% = $54,400,000
Year 3 170,000,000 × 19.20% = $32,640,000
Year 4 170,000,000 × 11.52% = $19,584,000
Year 5 170,000,000 × 11.52% = $19,584,000
Year 6 170,000,000 × 5.76% = $9,792,000
------------
$212,500,000
Tax value, discounted at 8% = $37,228,000
The cost of the acquisition date
Difference in PV = $4,091,000
Four million dollars of present value, turning on whether a supply contract was signed before or after 19 January 2025 — a date that did not exist as a concept when most of those contracts were negotiated.
ℹ️ Note: Total deductions are identical in both cases. Only the timing differs. Everything in this comparison is a discounting effect, which means the answer moves with the discount rate: at 12% the gap widens to roughly $5.7m, and at 4% it narrows to about $2.4m. Run it at the rate that actually applies to the party taking the deduction, not at the project discount rate.
The gate test that must sit underneath
Year 1 loss allocated to investor (99%) = $44,175,000 (at 100% bonus)
Investor outside basis at year 1 = to be computed
Binding gate = §704(d) / §465 / §469 / §461(l)
Usable deduction = MIN(allocated, gate)
If the usable amount is materially below the allocated amount, test the 40% transitional election. A deduction spread across six years that is fully usable beats a deduction taken in one year that is half suspended and extinguished on exit.
To run the full comparison — basis reduction, both bonus cases, the gate test and the transitional election — prompt Dezzmond with your cost basis and acquisition dates.
What Do Sponsors and Investors Actually Check?
- The acquisition date of each material component. Not the COD. The 19 January 2025 line is the one that decides the bonus percentage.
- Whether basis was reduced by half the credit, not the whole credit and not zero.
- Whether the mid-quarter convention applies, if the COD schedule is fourth-quarter heavy.
- Whether the investor can actually use a year-one deduction of that size, running all four gates.
- Whether the 40% transitional election produces more usable benefit than 100% does.
- What the DRO has to be to support the accelerated allocation.
- What the exit looks like, with §1245 ordinary income recapture on a near-zero basis.
Frequently Asked Questions
How much is depreciable basis reduced by the ITC?
By half the credit. At a 30% credit that is 15% of cost, leaving 85% of cost to depreciate — the half-credit rule under §50(c).
Is bonus depreciation still phasing down?
No. The One Big Beautiful Bill Act, enacted 4 July 2025, permanently restored 100% bonus depreciation for qualifying property acquired and placed in service after 19 January 2025. The old 80/60/40/20 schedule still governs property acquired on or before that date.
Can a project acquired in 2024 get 100% bonus?
No. Both limbs of the test must be met. Property acquired on or before 19 January 2025 remains on the prior-law phase-down even if it is placed in service years later — which for a 2026 placed-in-service date means 20%.
Why would anyone elect less bonus depreciation?
Because an accelerated deduction is only worth its face value to a taxpayer who can use it. Where the §704(d), §465, §469 or §461(l) gates bind, spreading the deduction can produce more usable benefit than concentrating it — and suspended losses are extinguished on exit.
Does the PTC reduce depreciable basis?
No. The half-credit reduction applies to the investment credit only, so a project electing the PTC depreciates its full cost. On a $250m project that is $37.5m more basis, worth about $7.3m in present value at a 21% rate.
Are interconnection costs in the basis?
For projects of 5 MW AC or less, yes — the IRA made interconnection property a qualified expense at that scale. Above 5 MW AC they are not, which is a cliff worth checking before sizing a facility just over the threshold.
What happens to the basis reduction on a sale?
It is treated as a depreciation deduction for §1245 and §1250 purposes, so the corresponding gain is recaptured as ordinary income rather than taxed as capital gain.
Closing: Series B, and the Thing That Connects It
Eight posts on tax structuring end at a fairly unglamorous place: a basis calculation and a recovery schedule. But it is the right place to end, because almost every difficulty in the preceding seven posts traces back to it.
The partnership flip exists to move deductions to someone who can use them. The DRO is sized by how fast those deductions arrive. The four gates decide whether they are usable at all. The hybrid exists to make the basis bigger before the credit is computed on it. Recapture reverses the reduction that this post describes. And the choice between a transfer and direct pay is a choice about monetising a credit whose true cost includes the $7.3m of depreciation it quietly removed.
The date trap is the sharpest illustration of why this matters commercially rather than technically. Nothing about the project changed. The panels are the same, the credit is the same, the deductions in total are the same. A signature date on a supply contract, chosen for a completely different reason eighteen months earlier, moved four million dollars of present value — and no model that treats depreciation as a schedule to be filled in after the deal is agreed would ever have surfaced it.
Series C starts next, on market structure: what basis risk actually is, and why a P50 case can survive the model and not the settlement statement.
Sources: BDO — One Big Beautiful Bill Act Expands 100% Depreciation Expensing Opportunities · Stoel Rives — The Law of Solar: Navigating Tax Issues in Solar Energy Projects · Cornell LII — 26 U.S. Code §50 · SEIA — Depreciation of Solar Energy Property in MACRS