Comparability analysis is the methodological foundation of every transfer pricing examination. It is the process of identifying whether a controlled transaction — one between associated enterprises — is priced consistently with transactions that would have been agreed between independent parties under comparable circumstances. The obligation derives from the arm’s length standard as expressed in Article 9 of the OECD Model Tax Convention and its domestic equivalents, such as Article 57 of the French General Tax Code or equivalent provisions in Hungarian, Polish, Romanian, and Ukrainian law. The analysis requires systematic comparison across five key factors: the characteristics of the property or services transferred, the functions performed and risks assumed by each party, the contractual terms, the economic circumstances of the parties and markets, and the business strategies pursued.

Disputes arise when the tax authority and taxpayer differ on which transactions or companies constitute valid comparables, which transfer pricing method is most appropriate, and what adjustments must be made to achieve genuine comparability. In the Issey Miyake case, French authorities rejected the taxpayer’s chosen comparables and substituted their own database selection. In the Kenya Beta Healthcare case, authorities replaced the taxpayer’s TNMM analysis with a CUP approach. In Poland’s C. spółka z o.o. dispute, the tax authority rejected the cost-plus method for lack of supporting evidence and applied the TNMM using its own comparable set. The typical contest is therefore threefold: method selection, comparable selection, and the adequacy of comparability adjustments.

The governing framework is Chapter III of the OECD Transfer Pricing Guidelines (2022), which addresses comparability analysis at paragraphs 3.1 through 3.62, including the five comparability factors at paragraphs 1.36 to 1.105 of Chapter I. Chapter II governs method selection and hierarchy. The 2010 Report on the Attribution of Profits to Permanent Establishments is also relevant where functional analysis interacts with branch structures. National courts frequently reference these guidelines as interpretative aids, as seen in the Hungarian Kúria decisions on the LoanConnector database and electronic components cases.

Courts and practitioners examine the search methodology used to construct a comparable set, the filtering criteria applied to databases such as Bureau van Dijk or Thomson Reuters LoanConnector, and whether adjustments for working capital, accounting differences, or geographic risk have been performed and documented. The most contested questions are whether a proposed comparable truly performs similar functions and bears comparable risks, and whether the resulting interquartile range was correctly constructed and applied.

The cases in this category demonstrate that procedural rigour in comparable selection is as legally consequential as the economic analysis itself, and practitioners must ensure that every filtering and adjustment step is fully documented and defensible.