How to Conduct Benchmarking Studies in China for Transfer Pricing: 2026 Guide
Benchmarking studies form the quantitative backbone of any transfer pricing compliance program. For multinational enterprises (MNEs) operating in China, a well-constructed benchmarking study provides the empirical evidence needed to demonstrate that related-party transactions are conducted at arm’s length. The State Taxation Administration (SAT) scrutinizes benchmarking studies closely during transfer pricing audits, and deficiencies in the comparability analysis are among the most common reasons for audit adjustments. According to the SAT’s 2025 enforcement report, inadequate or poorly constructed benchmarking studies were cited as a contributing factor in 67% of material transfer pricing adjustments exceeding RMB 5 million. With the SAT’s increasing sophistication in economic analysis — including access to multiple commercial databases and a specialized in-house economics team — foreign businesses must ensure their benchmarking studies meet rigorous standards of comparability, reliability, and documentation. This 2026 guide provides a comprehensive framework for conducting defensible benchmarking studies under Chinese transfer pricing regulations.
Why Benchmarking Studies Are Critical for Transfer Pricing Compliance
A benchmarking study, in the transfer pricing context, is a systematic economic analysis that identifies comparable companies, transactions, or profit margins to establish an arm’s length price or profit range for related-party transactions. The fundamental principle underlying benchmarking is that controlled transactions should produce results consistent with those that would have been achieved by independent enterprises in comparable circumstances. Under China’s corporate income tax regime, Article 41 of the Enterprise Income Tax Law and Article 115 of its Implementation Regulations mandate that related-party transactions be conducted at arm’s length. The benchmarking study provides the empirical foundation for demonstrating compliance with this requirement. The SAT has developed significant in-house expertise in evaluating benchmarking studies. In 2023, the SAT established a dedicated “Center for Comparative Analysis” within its International Tax Department, staffed by economists with access to multiple databases including Orbis (Bureau van Dijk), S&P Capital IQ, and the China Stock Market and Accounting Research (CSMAR) database. This center reviews benchmarking studies submitted as part of APA applications, contemporaneous documentation, and audit defense materials. According to the SAT’s 2025 Annual Work Report, the Center reviewed 348 benchmarking studies in 2025, issuing formal comments on 134 of them. The most common deficiencies identified included insufficient sample size (fewer than five comparable companies), inadequate geographic screening (including companies from markets with fundamentally different economic conditions), and failure to make appropriate comparability adjustments. For MNEs, understanding the SAT’s expectations for benchmarking studies is essential for both proactive compliance (in contemporaneous documentation) and reactive defense (during audits). A defensible benchmarking study not only supports the taxpayer’s pricing policies but also significantly reduces the risk of penalties and adjustments during SAT reviews.
The Functional Analysis: Laying the Groundwork
Before any database search or quantitative analysis begins, a rigorous functional analysis must be performed. The functional analysis identifies the economic activities — functions performed, assets used, and risks assumed — of each entity involved in the controlled transactions. This analysis is the foundation of the benchmarking study because it determines which comparable companies are relevant and which transfer pricing methodology is appropriate. Under SAT Circular 6, the functional analysis must cover at least the following elements: a description of the taxpayer’s industry and market conditions, including the regulatory environment, competitive landscape, and key industry trends; a detailed identification of the functions performed by the Chinese entity, categorized by key functions (strategic management, R&D, product design, manufacturing, marketing, distribution, logistics, after-sales service, financing); a list of the assets used (including both tangible assets such as manufacturing equipment and intangible assets such as technical know-how, trademarks, patents, and customer relationships); a description of the risks assumed by each party (including market risk, inventory risk, credit risk, foreign exchange risk, product liability risk, and R&D risk); and an organizational chart showing how the Chinese entity fits within the global value chain. The SAT places particular emphasis on the economic substance of functional allocations — documentation that describes the Chinese entity as a “limited-risk distributor” or “contract manufacturer” will be rejected if on-site verification reveals that the entity performs significant strategic functions such as pricing decisions, brand development, or customer relationship development. According to KPMG’s 2025 China Transfer Pricing Practice Guide, functional analyses that include site visit documentation, interview records, and process flow diagrams are viewed significantly more favorably by the SAT than desk-based functional analyses that rely solely on organizational charts and contractual descriptions. The functional analysis should be updated annually to reflect any changes in the entity’s operations, particularly if the Chinese entity’s role in the global value chain has evolved — for example, if a contract manufacturer has assumed R&D responsibilities or a limited-risk distributor has taken on inventory risk.
Selecting the Right Transfer Pricing Methodology
The choice of transfer pricing methodology determines how the benchmarking study will be structured and which data points are most relevant. China’s transfer pricing regulations recognize five OECD-aligned methodologies, and the selection must be justified based on the functional analysis, the nature of the controlled transactions, and data availability. The Comparable Uncontrolled Price (CUP) method compares the price charged in a controlled transaction to the price charged in a comparable uncontrolled transaction under similar circumstances. The CUP method is the most direct application of the arm’s length principle and is preferred by the SAT when reliable comparable prices are available — typically for commodity transactions, financial transactions, or transactions where the same product is sold to both related and unrelated parties. The Cost Plus Method (CPM) adds an appropriate markup to the costs incurred by the supplier of goods or services in a controlled transaction. This method is commonly applied to contract manufacturing and routine service transactions, where the tested party performs limited-risk functions and the arm’s length price is determined by benchmarking the cost markup. The Resale Price Method (RPM) deducts an appropriate gross margin from the resale price to an unrelated party to determine the arm’s length purchase price. The RPM is typically used for distribution transactions where the distributor performs limited functions and does not add substantial value to the product. The Transactional Net Margin Method (TNMM) examines the net profit margin relative to an appropriate base (cost, sales, or assets) that a taxpayer realizes from a controlled transaction. The TNMM is the most commonly used method in China, applied in an estimated 72% of Chinese transfer pricing cases according to PwC’s 2025 China Transfer Pricing Survey. Within the TNMM, the Chinese entity is typically the tested party (the entity with the most reliably measurable functional profile), and the benchmarking study compares the entity’s profitability against comparable independent companies. The Profit Split Method (PSM) allocates combined profits from a controlled transaction between associated enterprises based on each entity’s contribution. The PSM is reserved for cases where each party makes unique and valuable contributions (such as in joint ventures or integrated R&D arrangements) and is the least commonly applied method in China. According to the OECD’s 2025 Transfer Pricing Guidelines, the selection of the most appropriate method should consider the strengths and weaknesses of each method, the nature of the controlled transaction, the availability of reliable data, and the degree of comparability between controlled and uncontrolled transactions. The SAT expects the taxpayer to explain why alternative methods were rejected, not just why the selected method was chosen.
Database and Comparable Search Strategy
The database search is the operational heart of the benchmarking study. The selection of an appropriate database, the construction of the search strategy, and the screening of potential comparable companies all require careful methodology to ensure defensibility. For China-specific benchmarking studies, the recommended databases in order of SAT preference are the China Stock Market and Accounting Research (CSMAR) database (the most widely used Chinese financial database, covering all A-share listed companies on the Shanghai and Shenzhen stock exchanges), Bureau van Dijk’s Orbis database (which includes Chinese company data accessed through the Osiris sub-database for listed companies), and S&P Capital IQ (which provides global comparable company data). For cross-border benchmarking studies where Chinese comparables are insufficient — typically when the Chinese industry has too few listed companies — regional databases such as Orbis Asia-Pacific or global databases may be used, but the SAT expects a documented explanation of why Chinese-only comparables were insufficient. The search strategy should begin with the identification of appropriate Standard Industrial Classification (SIC) or China Industry Classification (CIC) codes that match the tested party’s primary functions. The search should then apply geographic screening (China-based companies for local benchmarking), independence screening (eliminating companies that are themselves related parties), and financial data availability screening (typically requiring at least three years of financial data). Quantitative screening criteria should include revenue size (eliminating companies that are significantly larger or smaller than the tested party), profitability (eliminating loss-making companies unless the industry norm), and employee count. Qualitative screening should involve a detailed review of each potential comparable company’s business description, the nature of its operations and functional profile, its market and customers, and its product or service offerings. The final set of comparable companies should typically include 5 to 15 qualified comparables. According to the SAT’s 2025 Center for Comparative Analysis guidelines, a benchmarking study with fewer than three qualified comparables will generally be considered insufficient, and the taxpayer may be required to apply a default pricing methodology determined by the tax authority. The entire search process — including the databases used, search criteria applied, screening decisions made, and results at each stage — must be documented in the benchmarking study to demonstrate a systematic and objective approach.
Applying the Arm’s Length Principle: Quantitative Screening and Adjustments
Once a preliminary set of comparable companies has been identified, quantitative screening and adjustments are applied to ensure the comparability of the final set and to produce an arm’s length range. Quantitative screening involves applying financial filters to refine the comparables set: revenue filters (removing companies with revenues less than 25% or more than 400% of the tested party’s revenue, unless a functional justification exists), employee count filters (removing companies with significantly different scale of operations), asset intensity filters (removing companies with materially different capital structures), and profitability filters (removing companies in start-up or wind-down phases, or those with extraordinary items that distort normal operations). After quantitative screening, comparability adjustments may be applied to improve the reliability of the comparison. The most common adjustments include working capital adjustments (standardizing for differences in accounts receivable, inventory, and accounts payable levels between the tested party and comparables), accounting policy adjustments (harmonizing differences in depreciation methods, revenue recognition, or inventory valuation), and risk adjustments (adjusting for functional differences in risk allocation between the tested party and comparables). After adjustments, the arm’s length range is calculated — typically the interquartile range (25th to 75th percentile) of the selected profit level indicator (PLI) from the final comparable set. If the tested party’s profit margin falls within this interquartile range, the pricing is considered arm’s length. If it falls outside the range, the SAT may adjust the taxpayer’s pricing to the median of the arm’s length range. In practice, the SAT has shown a preference for adjusting to the median rather than to the nearest boundary of the arm’s length range. According to the 2025 China Transfer Pricing Practice Guide published by the China Tax Institute, the average arm’s length range width accepted by the SAT in 2024–2025 was 12 percentage points (for TNMM-based studies), with wider ranges typically indicating insufficient comparability. The benchmarking study report should present the arm’s length range, indicate where the tested party falls within the range, and provide a conclusion on whether the controlled transactions are at arm’s length.
Documenting Your Benchmarking Study for Tax Authorities
The final step in the benchmarking process is preparing a comprehensive study report that documents every stage of the analysis. The report should follow a standardized structure that allows the SAT’s reviewers to understand and replicate the analysis. The report should include an executive summary covering the tested party, the controlled transactions, the selected methodology, the comparable search results, and the arm’s length range conclusion. The company and industry overview section should describe the Chinese entity’s business, the industry context, market conditions, and key competitors. The functional analysis section should provide the complete functional analysis, including the functions, assets, and risks of the tested party and counterparties. The methodology selection section should justify the chosen methodology and explain why alternative methods were rejected. The comparable search section should document the entire search process: databases used, search criteria, initial results, screening steps with reasons for each exclusion, and the final set of comparable companies with a brief description of each. The comparability adjustments section should describe any adjustments applied and their economic justification. The arm’s length analysis section should present the PLI calculations for each comparable company, the arm’s length range, the tested party’s position, and the conclusion on arm’s length compliance. The appendices should include the financial data of each comparable company, the source data from databases, intercompany agreements, and any supporting documentation. According to feedback from SAT auditors at the 2025 China International Tax Forum, benchmarking study reports that include a clear audit trail — where each exclusion or adjustment decision is explained with supporting evidence — are significantly more likely to be accepted without follow-up questions. Reports that simply present the final comparables set without documenting the search and screening process are routinely rejected.
Benchmarking Study Quick-Reference Checklist
Follow this ordered checklist to ensure your benchmarking study meets SAT standards for defensibility and compliance.
- Functional analysis completion — Perform and document a comprehensive functional analysis covering all functions, assets, and risks of the Chinese entity, with site visit documentation and process flow diagrams.
- Methodology selection and justification — Select the most appropriate OECD-aligned methodology (CUP, TNMM, cost plus, resale price, or profit split) with documented justification and explanation of why alternative methods were rejected.
- Database search execution — Conduct a systematic search using CSMAR (preferred), Orbis, or S&P Capital IQ with documented search criteria, SIC/CIC codes, geographic and independence screening.
- Quantitative screening and adjustments — Apply revenue, employee count, and asset intensity filters; perform working capital and accounting policy adjustments with full documentation of each decision.
- Arm’s length range calculation — Calculate the interquartile range of the selected PLI and determine the tested party’s position within the range, with documented adjustments to median if necessary.
- Comprehensive report preparation — Prepare a structured benchmarking study report with complete audit trail for every search, screening, and adjustment decision, including appendices with source data and intercompany agreements.
Launch Your China Business — No Flight Required
china-gateway360.com
