RoyaltyStat Blog

Gross Profit Methods are Unreliable

Posted by Ednaldo Silva on Jun 14, 2016 5:59:10 AM

If an enterprise makes or buys goods to sell, the cost of goods sold (COGS) can be deducted from net receipts. However, to determine COGS, the inventory at the beginning and end of each tax year must be valued. Consider the symbols:

Vt – 1 = Inventory at beginning of year, Pt = Purchases, Lt = Cost of labor, Mt = Materials and supplies, Qt = Other costs, Vt = Inventory at end of year, Ct = Cost of goods sold (COGS), and t = tax year.

Using the symbols above, we can define the accounting equation:

     (1)     Ct = Pt + (Lt + Mt + Qt) + (Vt – 1 – Vt)

     (2)     Ct = Pt + Ht + ∆Vt, where

     (3)     Ht = (Lt + Mt + Qt).

E.g., see

Prior to the 1994 U.S. transfer pricing regulations and the 1995 OECD Guidelines, which introduced comparable operating profits methods (TNMM, profit split), transfer pricing audit was made supposedly by comparing related party purchases (Pt) with unrelated party purchases. However, unless a CUP (comparable uncontrolled price) existed, information about unrelated party purchases could rarely be found. As a result, transfer pricing audit was based on investigation of the aggregate COGS equation (1).

Equation (1) relies on two ceteris paribus assumptions that (i) Ht and (ii) ∆Vt are similar between the related and comparable unrelated parties (maledictum “transactions”). This first ceteris paribus assumption (i) can’t be tested because listed companies are not required to disclosed information about the cost of labor, materials and supplies, and the amorphous (impossible to compare and thus tacitly assumed to be “immaterial”) “other costs”. Perhaps only in "pure" distribution and retail activities can the Ht factor become immaterial. But we can test the second assumption in which the transfer pricing shifting of taxable income can occur by violation of assumption (ii), including transfers of obsolete inventory via controlled manipulation of ∆Vt. Although information about inventories is disclosed on the balance sheet of listed companies, our experience suggests that audit of ∆Vt is rarely made.

We don’t find equation (1) in economics. Even if we make an inventory valuation adjustment to approximate an economic concept of variable costs:

     (4)     Ĉt = (Ct − ∆Vt) = (Pt + Ht),

we can’t isolate and thus compare related and unrelated party purchases. Purchases are bound together with the nebulous Ht factor. In short, gross profit analysis is vulnerable to this heroic assumption (i) that Ht is similar between related and “comparable” unrelated parties. In practice, the comparability of Ht is untested, assumed to be immaterial.

We find gross profits (SALE – COGS) to be an unreliable measure of transfer pricing adjustments. Here, our objection to transfer pricing methods based on gross profits (including the resale price, cost plus, and the Berry ratio) is based on the practical difficulty of testing the ceteris paribus assumption (i). Unless proven otherwise, we are persuaded that the unknown (and likely incommensurable) Ht factor impeaches the accounting and legal fiction of transfer pricing adjustments based on comparing gross profits. Calling gross profit methods “transactional” is perfidious, because it can be measured only in the aggregate at the level of COGS and not by comparing individual transactions. As an aside, the "T" of TNMM is a solecism because operating profits methods are also aggregate and not transactional measures. We should abandon this eccentric humpty-dumpty lingo and use the word "transactional" with etymological meaning.

Read article about gross profit methods!

Ednaldo Silva (Ph.D.) is founder and managing director at RoyaltyStat. He helped draft the US transfer pricing regulations and developed the comparable profits method called TNNM by the OECD. He can be contacted at:

RoyaltyStat provides premier online databases of royalty rates extracted from unredacted agreements, normalized company financials (income statement, balance sheet, cash flow), and annual reports. We provide high-quality data, built-in analytical tools, customer training and attentive technical support.

Schedule an Online Demo 

Topics: Comparability Analysis