Tax

Benchmarking in TP documentation: data sources

Overhead view of charts and graphs with magnifying glass on a wooden table

Benchmarking is the empirical core of a transfer pricing file: everything else is narrative until a set of independent companies’ results is placed next to the tested party’s margin. Decree 255/2026/ND-CP matters here because it prescribes, for the first time in such detail, a hierarchy for the data sources a benchmarking study may use.

The data sources, in hierarchy order

At the top sit the company’s own comparable independent transactions (internal comparables) — same conditions, same accounting. Next come commercial databases of independent companies with public financial statements, screened by industry, function and independence; exchange quotations and published price statistics serve CUP-type comparisons. Decree 255/2026/ND-CP tightens this into an explicit hierarchy of benchmarking data sources, with preference for data from Vietnam and markets at a comparable development level, and a duty to explain departures. The practical consequence: a study built entirely on distant-market companies with no justification paragraph is structurally weak under the new decree, whatever its numbers say.

What makes a source acceptable to the authority

Acceptance follows process, not brand names. The study must show the screening journey: database and search date, initial population, each screening criterion (industry code, independence test, data sufficiency, loss and special-event exclusions), who was removed and why, and the final set — then the statistical treatment, usually the interquartile range. Authorities test the final set’s functions hard: a ‘distributor’ set containing manufacturers or retailers collapses the study. Financial data must be for the same fiscal year as the tested transaction and the comparables must be genuinely independent; circular use of other group members’ published data is an error auditors spot immediately.

Cost, timing and the annual refresh

Benchmarking is commissioned work: advisory firms charge by scope — number of transaction groups, methods tested, database access — and no regulated tariff exists, so the article deliberately gives no figure. What buyers should fix in the engagement is not only price: the screening criteria, the data hierarchy applied, delivery of the full screening log (not just conclusions), and the update duty. A benchmarking set ages fast: margins move with the market, so the study must be refreshed for each fiscal year, in step with the local file. Reusing a three-year-old set is the single most common reason an otherwise good file loses at audit — the authority’s fresh set simply looks more honest.

Frequently Asked Questions

Where does benchmarking data for TP documentation come from?

From commercial company databases of independent enterprises, exchange quotations for CUP cases, and the company’s own internal independent transactions. Decree 255/2026/ND-CP now prescribes a hierarchy of data sources, favouring Vietnamese and comparable-market data with reliable financial disclosure.

Which data sources do tax authorities accept?

Sources meeting the hierarchy and screening discipline of Decree 255/2026/ND-CP: independent companies in the same industry and function with audited, public financial data, screened by transparent criteria. A set picked without a screening record is routinely replaced by the authority’s own set.

How much does it cost to hire a firm for benchmarking?

No regulated price exists; cost depends on the number of transaction groups, the method and the database licences involved. Fee quotes only become comparable when the scope states the screening criteria and update duties — and benchmarking must be refreshed for each fiscal year.


Related articles

Reviewed by the Legal Practice Group – FLAT LAW FIRM

This legal analysis has been verified against current Vietnamese regulations and cross-border FDI transaction practices by the attorneys at FLAT LAW FIRM.