Section 174A: how the new domestic R&E expensing regime works

Section 174A is the OBBBA's answer to three years of mandatory research amortization: for tax years beginning after December 31, 2024, domestic research or experimental expenditures are deductible in the year paid or incurred, with an option to capitalize over at least 60 months instead. Foreign research got no such rescue — it remains under old Section 174, capitalized over 15 years.

The new section is not a simple repeal-and-restore. It splits the pre-2022 world into two statutes, adds an election with real planning content, and forces every multinational R&D operation to answer a question that used to be trivial: where, exactly, is this research performed?

The default rule: immediate domestic expensing

For tax years beginning after December 31, 2024, a taxpayer deducts domestic research or experimental expenditures currently. The definition of what qualifies carries over from Section 174 practice: costs in the experimental or laboratory sense, incurred to eliminate uncertainty about the capability, method, or design of a product, process, or software. Software development remains in scope by statute — the TCJA's categorical inclusion of software development costs was preserved, so domestic software engineering is deductible under 174A rather than falling back to pre-2022 authorities like Rev. Proc. 2000-50. (The history of that inclusion is covered in our brief on software development.)

The mechanics of identifying the cost pool — direct research labor, allocable overhead, contract research the taxpayer bears risk on — did not change with the OBBBA. Notice 2023-63's framework remains the reference point; see identifying SRE expenditures.

Deduction under 174A is the default method for costs first incurred in post-2024 years. Taxpayers coming off the 2022–2024 capitalization regime change methods to get there, and their existing unamortized balances are handled by transition rules discussed below and in detail in the transition guide.

The election: capitalize over 60 months or more

Section 174A preserves the pre-2022 architecture's second track: an election to capitalize domestic R&E expenditures and amortize them ratably over a period of not less than 60 months, beginning when the taxpayer first realizes benefits from the expenditures.

Why elect slower recovery? Several fact patterns make it rational:

• NOL management. Post-TCJA, NOLs arising after 2017 offset only 80% of taxable income in a carryforward year. A startup that expenses $5 million of research into a loss builds an NOL it can use only at 80 cents on the dollar later; capitalizing preserves flexibility.
• Section 382 exposure. A company anticipating an ownership change — a financing round, an acquisition — may prefer not to enlarge pre-change NOLs that will be limited anyway.
• Expiring attributes. Foreign tax credits and pre-2018 NOLs expire. Income now, sheltered by expiring attributes, can beat deductions now.
• State and financing covenants. Some taxpayers manage to book-like taxable income for covenant or state-tax reasons.

What stays behind: foreign research under Section 174

Section 174 did not go away. It survives as the mandatory regime for foreign research: SRE expenditures attributable to research performed outside the United States must still be capitalized and amortized over 15 years, half-year convention in year one. Congress kept the TCJA's disincentive for offshoring R&D fully intact.

Domestic vs. foreign: the situs test

The line is geographic and activity-based: research is domestic if performed within the United States, its possessions, and territories (including Puerto Rico). Points that matter in practice:

• Where the work happens, not who pays or who owns the IP. A U.S. parent funding development performed by engineers in Poland has foreign research. A foreign parent's U.S. subsidiary doing the work stateside has domestic research.
• Contract research follows the contractor's activity. Paying a U.S.-headquartered consulting firm does not make the research domestic if the firm's delivery team sits in Bangalore. Contracts rarely say where the work occurs; invoices and staffing records have to.
• Remote and hybrid teams. An employee on the U.S. payroll working remotely from abroad performs foreign research for those hours. Companies with distributed teams need location tracking they mostly did not keep before 2022.
• Mixed projects require allocation. A project split between San Francisco and Toronto teams needs a reasonable allocation of costs — typically by labor hours or headcount — between the 174A deduction and the 174 15-year pool.

The traps here are treated at greater length in foreign research and the 15-year rule. The audit posture is predictable: with a first-year deduction of 100% versus 3.33%, the IRS's incentive is to find foreign situs, and the taxpayer's burden is documentation.

Interaction with Section 41 and Section 280C

Section 174A slots into the credit architecture where old Section 174 used to sit. Three points of contact:

Eligibility gateway. Qualified research expenses under Section 41 must be eligible for treatment under the R&E deduction rules. Restoring expensing does not expand or shrink the QRE pool — Section 41(d)'s four-part test and its cost-category limits still control. The 174/174A pool remains broader than the QRE pool, mainly because 41 excludes general overhead and haircuts contract research to 65%. See Section 174 vs. Section 41 for the five-minute version, and the R&D credit primer for the credit itself.

Section 280C coordination. With a current deduction back in place, 280C reverts to its classic operation: a taxpayer claiming the research credit must reduce its otherwise-allowable deduction by the amount of the credit, or elect under Section 280C(c) to take a reduced credit (the gross credit cut by the top corporate rate, 21%). During the amortization era, the required reduction applied against capitalized amounts and mattered less in-year; under 174A the choice is live again every year. For most C corporations the reduced-credit election remains the administratively cleaner path; taxpayers with NOLs or low marginal rates should run both.

Foreign QREs, largely a null set. Section 41 has always required qualified research to be conducted in the United States, so the 15-year foreign pool generally produces no credit — a reminder that the foreign classification is pure cost with no offsetting benefit.

The credit-versus-deduction interaction is also treated in our comparison of the R&D credit and Section 174.

Getting onto 174A: transition and method mechanics

For a calendar-year taxpayer, 2025 is the first Section 174A year. Two distinct workstreams:

1. Current-year costs. Domestic 2025 research is deducted (or capitalized under the 60-month election). Implementing this from a 2022–2024 capitalization method is an accounting-method change, generally with automatic consent — see Section 481(a) and Section 174 and the broader Form 3115 guide.
2. Legacy balances. Unamortized domestic amounts from 2022–2024 are recovered either through the elective one- or two-year catch-up deduction beginning in the first post-2024 year, or — for small businesses under the $31 million gross-receipts test — through retroactive amendment of the 2022–2024 returns. The modeling is the subject of the transition rules guide.

Foreign balances from 2022–2024 get no catch-up; they continue amortizing over their original 15-year schedules.

Where 174A does not help

Neutrality requires the ledger's other side:

• Foreign-heavy R&D. A company whose development sits offshore sees essentially no change. Re-shoring decisions have a new tax input, but 15-year amortization continues for existing foreign operations and balances.
• Loss companies. Immediate deductions have time-value only if they offset income. A pre-revenue company converts 174A deductions into NOLs subject to the 80% limitation — real value, but deferred and discounted. Such companies should genuinely evaluate the 60-month election rather than default to expensing.
• Book earnings. 174A changes cash taxes and deferred tax balances, not ASC 730 book expense, which was current all along. The financial-statement effects — chiefly the release of deferred tax assets built up during 2022–2024 — are covered in R&D capitalization and financial statements.
• IRS attention. Expect examination focus on domestic/foreign allocation, on whether 2025 deductions include amounts properly part of legacy capitalized balances (no double-counting), and on the procedural validity of transition elections.

The bottom line

Section 174A gives domestic research its pre-2022 treatment back, with a 60-month capitalization election for taxpayers whose attribute posture favors income. The price of the restoration is complexity at the seams: a geographic line with a 100%-versus-3.33% first-year stakes differential, a transition menu with method-change mechanics attached, and a credit coordination rule that is once again an annual decision. The substantive definition of research did not change; the map did.