Every transfer-pricing rule in the U.S. code traces back to one idea: the arm's-length principle. It says that when two related companies transact — a parent and its foreign subsidiary, two sister entities under common ownership — they should price the deal the way two unrelated companies dealing at arm's length would. If an independent supplier would charge $100 for a widget, your controlled affiliate should pay about $100 for the same widget. The principle is the benchmark behind IRC §482 and the yardstick the IRS uses to decide whether your intercompany pricing shifts profit improperly. It sounds abstract until you try to apply it — which is where the transfer-pricing methods come in.
The methods are simply structured ways to prove an arm's-length price. Each one answers the question "what would an unrelated party have paid?" from a different angle, using different data. The U.S. regulations under §482 do not force you to pick a favorite: they apply a best method rule, meaning you use whichever method most reliably produces an arm's-length result given the comparables you can actually find. That last part — the comparables — is where the real work lives. A method is only as good as the third-party data behind it. Below, the five methods you will hear named most often, in plain English.
Comparable Uncontrolled Price (CUP) — the gold standard
The CUP method is the most direct: find a real, unrelated transaction for the same product under comparable terms, and use that price. If your company sells the identical component to an unrelated distributor for $100, that $100 is a strong arm's-length benchmark for what your foreign affiliate should pay. When a genuine comparable exists, CUP is the most reliable method because it compares price to price with no assumptions in between.
CUP is preferred whenever you have it, but it is demanding — the comparable transaction has to be truly similar in product, volume, terms, and market. A near-match is not a match. Small differences (a different currency, a bundled service, a volume discount) can undermine a CUP comparison, which is why many smaller multinationals end up relying on the profit-based methods instead.
Resale Price and Cost Plus — working from one side of the deal
When you can't find a clean price comparable, two methods build the arm's-length price from a known margin instead. They approach the same transaction from opposite ends.
Resale Price Method
- Best for a distributor that buys from an affiliate and resells
- Starts from the price the affiliate resells to outside customers
- Subtracts an arm's-length gross margin the reseller should earn
- The remainder is the arm's-length intercompany purchase price
- Benchmark: gross margins of independent distributors
Cost Plus Method
- Best for a manufacturer or service provider selling to an affiliate
- Starts from the supplier's cost of producing the good or service
- Adds an arm's-length markup for the function performed
- The total is the arm's-length intercompany sale price
- Benchmark: gross markups earned by independent producers
Resale Price fits a buy-and-resell affiliate: you know what it resells for, so you back into the intercompany price by subtracting the margin a comparable independent distributor would keep. Cost Plus fits a maker or service provider: you know its cost, so you add the markup a comparable independent producer would charge. Both depend on comparable margins rather than comparable prices, which makes them workable when identical-product data is scarce.
Transactional Net Margin Method (TNMM) — the workhorse
TNMM is the most widely used method in practice, especially for smaller companies, because it needs the least perfect data. Instead of comparing prices or gross margins, it compares the net profit margin the tested party earns to the net margins of comparable independent companies performing similar functions. If independent contract manufacturers in your industry earn a net operating margin of, say, 4–6% on sales, your affiliate's intercompany pricing should leave it earning within that range.
| Method | What it compares | Best when |
|---|---|---|
| CUP | Price vs. a real third-party price | A truly comparable transaction exists |
| Resale Price | Gross margin of a reseller | The affiliate buys and resells goods |
| Cost Plus | Gross markup of a producer | The affiliate makes goods or performs services |
| TNMM | Net profit margin vs. comparable firms | No clean price or gross-margin comparable exists |
| Profit Split | Division of combined profit | Both parties add unique, valuable contributions |
TNMM's popularity comes from data availability: comparable company profit margins from public financial databases are far easier to assemble than identical transactions. Its weakness is that net margins absorb many factors — operating efficiency, cost structure, one-off expenses — so a defensible TNMM study screens the comparable set carefully and often adjusts for differences.
Profit Split — when both sides are indispensable
The four methods above test one party against outside benchmarks. Profit Split takes a different tack: it combines the total profit from the intercompany dealings and divides it between the related parties according to the relative value each contributes. It is the right tool when both entities bring something unique and valuable — proprietary technology on one side, a distinctive market position on the other — so that neither can be benchmarked in isolation against an ordinary third party.
The U.S. regulations deliberately avoid a rigid ranking. Under the best method rule you choose the method that, given your facts and the quality of your comparables, most reliably approximates arm's-length. That means you can't just default to the easiest method — you have to be able to explain why the method you chose is more reliable than the alternatives. Documenting that reasoning is part of what protects you from penalties.
Comparables: the foundation under every method
Whichever method you pick, its credibility rests on comparables — the third-party prices, margins, or companies you benchmark against. Weak comparables sink an otherwise sound method; strong ones make even a simple method hold up. Building them is the substantive work of a transfer-pricing analysis.
- Similar product, service, or function to the controlled transaction
- Comparable market, geography, and level of the supply chain
- Adjustments made for material differences (volume, terms, currency)
- A screened set of multiple comparables, not a single cherry-picked point
- Documented sources so the analysis can be reviewed and reproduced
You do not need to master these methods yourself — you need to pick the right one and document it before you file. For most smaller multinationals the practical answer is a TNMM or Cost Plus study backed by a clean comparable set, prepared contemporaneously. That combination usually satisfies the best method rule and shields you from the §6662(e) penalty if the IRS later disagrees on the margin.
Price it right, prove it once
Pick the right method and back it with real comparables
Talk to our team about your situation — if your business has intercompany transactions, solid documentation now is what makes pricing hold up if the IRS ever asks.
Talk to a tax proSources
- Treasury Regulations Section 1.482-1(c) — Best Method Rule
- Treasury Regulations Section 1.482-3 — Methods for transfers of tangible property (CUP, Resale Price, Cost Plus)
- Treasury Regulations Section 1.482-5 — Comparable Profits Method / Transactional Net Margin Method
- Treasury Regulations Section 1.482-6 — Profit Split Method
- IRC Section 482 — Allocation of income and deductions among taxpayers
Frequently asked questions
It means related companies should charge each other the same price two unrelated companies would agree to for the same deal. If an outside supplier would charge $100 for a product, your foreign affiliate should pay roughly $100 too. It is the benchmark the IRS uses to test whether intercompany pricing is honest.
The five most common are: Comparable Uncontrolled Price (CUP), which compares your price to a real third-party price for the same item; Resale Price, which starts from the resale price and works back; Cost Plus, which adds a market markup to cost; the Transactional Net Margin Method (TNMM), which compares net profit margins; and Profit Split, which divides combined profit between the related parties.
The U.S. rules use a 'best method' rule — there is no fixed hierarchy. You use whichever method most reliably produces an arm's-length result given the data you have. CUP is preferred when a truly comparable third-party price exists; TNMM is common when it doesn't, because comparable company profit margins are easier to find than identical transactions.
Comparables are third-party transactions or companies similar enough to yours to serve as a benchmark. Every transfer-pricing method depends on them — the whole point is to show what an unrelated party would have charged. The quality of your comparables largely determines whether your pricing survives an IRS review.
