CUP Method: Complete Guide with Worked Examples (2026)

The Comparable Uncontrolled Price (CUP) method is the most direct test of the arm’s length principle: it compares the price actually charged in a controlled transaction with the price charged in a comparable uncontrolled transaction. Where a genuine uncontrolled price exists, CUP is the OECD-preferred method — no PLI, no margin, no pool statistics, just the price.

How CUP works

The mechanics in four steps:

1. Find the comparable uncontrolled transaction — an independent buyer paying an independent seller for the same (or an equivalently comparable) product or service, under comparable conditions.
2. Establish the comparability — the goods/services must be the same or equivalent; any difference that would affect the price must be adjusted for or the comparison discarded.
3. Align the price terms — delivery terms (EXW/FOB/CIF), payment terms, quantity, quality grade, warranty, packaging, and market conditions must match or be normalised.
4. Compare — the controlled price against the uncontrolled price (or a narrow band of uncontrolled prices), with any residual difference explained.

CUP is a transaction-level method: it tests this transaction against that transaction. It does not produce a range the way TNMM does; it produces a price (or a tight price band) and a comparison.

Internal vs external CUPs

Type	What it is	Strength	Weakness
Internal CUP	The same entity sells the same product to independent third parties	The cleanest possible comparability — same product, same entity, same market	Rare in captive structures; the third-party sales must be material and un-managed
External CUP	An independent seller charges an independent buyer in a comparable market	Available for commodities, standard goods, published-price services	Product identity and market conditions must be defended; price databases help

Internal CUP is the gold standard when it exists — an Indian manufacturer that sells the same component to overseas third parties at a published price has a defensible CUP that no TPO pool-substitution can touch. Check internal sales before reaching for any other method.

When CUP is strong

• Commodities — steel, chemicals, bulk agricultural products, metals: exchange prices, spot prices and published quotations give the uncontrolled price directly.
• Standard goods with published prices — some industrial goods, spare parts, standardised services (freight, telecom, utilities).
• Arbitrage markets — where the market itself enforces one price, the “comparability” question nearly disappears.
• Internal third-party sales — as above.

When CUP fails

• Unique products — where no uncontrolled sale of the same product exists, CUP has nothing to compare (this pushes the analysis to TNMM or profit split).
• Contract terms that dominate the price — where payment terms, volume, exclusivity or technical specifications drive the price and cannot be matched, the adjustment burden overwhelms the method.
• Different market conditions — selling in India vs the US at different competitive intensities; the market-adjustment argument becomes the whole case.

Adjustments under CUP

CUP adjustments are price-level, which is exactly what makes them hard — most databases give margins, not prices. The standard adjustments:

• Delivery terms — normalise FOB vs CIF (freight and insurance differences).
• Payment terms — the financing cost of credit periods (a working-capital adjustment at the price level).
• Quantity — volume discounts and break points; compare like-with-like order sizes.
• Quality/specification — grade, purity, tolerance differences.
• Market conditions — regional price levels where defensible.

Each adjustment must be quantified and documented; the unadjusted comparison with an unexplained residual is not a CUP result, it is an assertion.

Worked example: commodity chemicals

An Indian entity exports a standard industrial chemical to its US affiliate at ₹410/kg CIF. The same chemical (identical grade, same delivery terms) sells on the Indian spot market and to independent US buyers at:

Uncontrolled sale	Price (₹/kg CIF-equivalent)
Indian spot market (same grade)	405
US independent buyer A (same grade, CIF)	409
US independent buyer B (same grade, CIF)	412

The controlled price (₹410) sits inside the uncontrolled band (₹405-₹412). No adjustment needed; the CUP test passes with the price comparison itself as the evidence. The documentation shows: product identity (grade, specification), delivery-term alignment, the three uncontrolled prices with sources, and the conclusion. That is the entire exhibit.

Contrast with the same fact pattern priced at ₹440: the 6-8% residual is now the case. Either find the functional or market explanation (a specification upgrade, a volume commitment) with evidence, or accept that CUP fails and the transaction needs a different method or an adjustment the TPO will test.

Common mistakes

• Calling a TNMM pool a CUP. Comparing the controlled price to a price reconstructed from margins of a pool of companies is not CUP — it is an unsupported hybrid. CUP needs actual uncontrolled prices.
• Mismatched delivery terms. Comparing an EXW controlled price to a CIF uncontrolled price without the freight bridge — the single most frequent CUP technical error.
• Ignoring quantity effects. A captive’s 50,000-tonne volume priced against a spot-market 500-tonne price, with no volume adjustment.
• A stale price. CUP is sensitive to price movement; the uncontrolled prices must be from the same period as the controlled transaction (or the movement must be documented).

Documentation checklist

1. Product/service identity — specification, grade, terms.
2. The uncontrolled transactions — counterparties (independent), dates, volumes, prices, sources (contracts, invoices, published prices, exchange data).
3. Each adjustment — formula, inputs, source of the rate.
4. The comparison — controlled vs uncontrolled, with the residual explained.
5. The conclusion — arm’s length, with the price (or band) stated.

Where CUP is genuinely available, use it — it is the method a TPO has the least room to re-cut. The method selection framework covers the order in which to test methods; the CUP vs TNMM comparison covers when the direct price test yields to margin benchmarking.