Profit Level Indicator Selection: How to Choose the Right PLI for Your Benchmarking Study
Borys Ulanenko
CEO of ArmsLength AI
TL;DR - Key Takeaways
Match your PLI to the tested party's value driver: revenue (Operating Margin), costs (Net Cost Plus), or assets (ROA/ROOA).
Berry Ratio is appropriate only in limited cases—when COGS is pass-through and operating expenses capture the value-add. Document the OECD conditions carefully.
OECD doesn't prescribe a single preferred PLI; selection depends on which indicator provides the most reliable measure in your specific circumstances.
Operating margin is the most commonly used PLI in practice—accounting for ~60% of US APAs where CPM was applied to tangible/intangible transfers.
Your denominator definition often matters more than the PLI name—document what goes into OPEX, COGS, and operating assets clearly.
Get the latest transfer pricing insights, AI benchmarking tips, and industry updates delivered straight to your inbox.
Quick Answer: Which PLI Should I Use?
The right Profit Level Indicator depends on what drives profitability in your tested party's transaction:
Operating Margin (OM) → For revenue-driven businesses (distributors, service providers, sales entities)
Net Cost Plus (NCP) → For cost-driven operations (contract manufacturers, R&D services, shared services)
Berry Ratio → For pass-through activities where operating expenses are the value driver (limited-risk distributors, agents)—but only when OECD conditions are met
Return on Assets (ROA/ROOA) → For asset-intensive businesses where capital employed drives returns
OECD doesn't prescribe a single "best" PLI. The selection should provide "a reliable measure" based on your specific circumstances (OECD Guidelines ¶2.90-2.92). Always match the PLI to the tested party's functions, assets, and risks.
What is a Profit Level Indicator?
A Profit Level Indicator (PLI) is the ratio used to measure profitability in a TNMM or CPM benchmarking analysis. The PLI you choose determines how you'll compare your tested party's results against independent companies.
US Treasury Regulations §1.482-5(b)(4) defines PLIs as "ratios that measure relationships between profits and costs incurred or resources employed." The key insight is that different PLIs capture different aspects of what makes a business profitable—and the right choice depends on your tested party's economic substance.
The PLI must match the tested party's value driver. A distributor earning returns on sales volume needs a different PLI than a manufacturer earning returns on production costs or assets employed.
The Five Main PLIs Explained
1. Operating Margin (OM)
Formula: Operating Margin = Operating Profit ÷ Net Revenue
Operating Margin measures profitability as a percentage of sales. It's the most widely used PLI globally—accounting for approximately 60% of US APAs where CPM was applied to tangible and intangible transfers (per Deloitte's summary of IRS 2023 APA statistics).
Best For:
Full-fledged and limited-risk distributors
Service providers with revenue-based pricing
Sales agents and marketing entities
Any entity where sales volume is the primary value driver
Advantages:
Financial data widely available
Easy to calculate and understand
Intuitive interpretation for auditors
Universal acceptance across jurisdictions
Limitations:
Sensitive to revenue fluctuations and pricing strategy
May not capture asset-intensive operations well
Pass-through revenue can distort results (agency/commissionaire arrangements)
Typical Application: A European distribution subsidiary buys products from its US parent and resells to third-party customers. Its value comes from sales execution—customer relationships, logistics, and market knowledge. Operating margin captures whether its return on sales is arm's length.
2. Net Cost Plus (NCP) / Return on Total Costs
Formula: Net Cost Plus = Operating Profit ÷ Total Costs = Operating Profit ÷ (COGS + Operating Expenses)
Net Cost Plus (also called Full Cost Plus or Return on Total Costs) measures the markup earned on the total cost base. It's the standard PLI for cost-driven operations.
Markup vs. Margin Confusion: NCP is a markup on costs (OP ÷ Costs), not a margin (OP ÷ Sales). China SAT Notice No. 6 explicitly labels this "full cost mark-up" with EBIT divided by full cost. Don't confuse the two.
Best For:
Contract manufacturers
Toll manufacturers
R&D service providers
Shared service centers
Any entity where cost efficiency is the value driver
Advantages:
Appropriate when the tested party's value comes from cost control
Less affected by pricing strategy variations
Works well for cost-plus service arrangements
Limitations:
Requires accurate cost allocation and segmentation
Pass-through costs must be identified and treated consistently
Affected by accounting treatment differences
Cost Base Design Matters: If subcontracted costs are pure pass-through (no value added by the tested party), including them in the denominator mechanically depresses the NCP. Define your cost base clearly and be consistent with comparables.
3. Berry Ratio
Formula: Berry Ratio = Gross Profit ÷ Operating Expenses
The Berry Ratio, named after economist Dr. Charles Berry who testified in Sundstrand Corporation v. Commissioner (1991), measures gross profit relative to operating expenses. It's designed for situations where operating expenses—not revenue or assets—drive the tested party's value.
Denominator Definition: China SAT Notice No. 6 defines the Berry ratio denominator as "Operating Expenses + G&A." The key is that COGS is excluded—the denominator captures value-adding activities, while COGS is assumed to be pass-through.
When Berry Ratio is Appropriate (OECD ¶2.99-2.101):
The tested party's value-add is captured in operating expenses
COGS is largely pass-through (the tested party doesn't add significant value to the goods themselves)
No significant intangibles or risks beyond routine activities
The tested party isn't performing additional functions that warrant a different return
When Berry Ratio is NOT Appropriate:
The tested party bears meaningful inventory risk or pricing risk
COGS includes value-adding activities (manufacturing, processing)
Operating expenses don't capture the full scope of value creation
The comparable set has inconsistent cost classifications
The Classification Sensitivity Problem:
Berry ratio is "very sensitive" to cost classification, as the UN Practical Manual warns. Consider this example:
Firm
COGS
OPEX
Gross Profit
Berry Ratio
Firm A
850
120
150
1.25
Firm B (same economics, 40 reclassified from OPEX to COGS)
890
80
110
1.375
Same operating profit, same economics—different Berry ratio solely due to accounting classification. This is why Berry disputes are common in jurisdictions like India.
Audit Risk: Berry ratio is accepted in the US and under OECD Guidelines (with conditions), but it's heavily litigated in India and less commonly accepted in Germany. If you use Berry, document explicitly how the OECD conditions in ¶2.101 are satisfied.
4. Return on Assets (ROA)
Formula: Return on Assets = Operating Profit ÷ Total Assets = EBIT ÷ Average Total Assets
China SAT Notice No. 6 specifies using average total assets: (opening balance + closing balance) ÷ 2.
Best For:
Asset-intensive manufacturers
Warehousing and logistics operations
Distributors with significant inventory and warehousing facilities
Any entity where capital investment drives returns
Advantages:
Captures asset intensity that other PLIs miss
Appropriate when returns should relate to capital employed
Useful when comparables vary significantly in asset intensity
Limitations:
Affected by depreciation policies (straight-line vs. accelerated)
High audit risk if balance sheets aren't comparable across accounting standards
PLI Selection by Transaction Type
Use this decision matrix as a starting point, then validate against your specific facts and circumstances:
Transaction Type
Primary PLI
Alternative
Rationale
Limited-risk distributor
Operating Margin
Berry Ratio
OM aligns to sales-based return; Berry only if OPEX is the true value driver
Full-fledged distributor
Operating Margin
ROA/ROOA
Inventory and working capital may justify asset-based PLI
Commissionaire / agent
NCP or Berry
Operating Margin
Revenue often not comparable across structures; cost-based focus more stable
Contract manufacturer
Net Cost Plus
ROA/ROOA
Cost efficiency is value driver; asset-based if intensity varies materially
Toll manufacturer
Net Cost Plus
-
Define cost base carefully (conversion costs vs. total if materials are pass-through)
Routine service provider
Net Cost Plus
Operating Margin
UN Manual explicitly notes services often fit profit-to-costs PLI
R&D service provider
Net Cost Plus
-
Tight cost base definition critical; exclude pass-through subcontracting
Shared services center
Net Cost Plus
-
Standard cost recovery with markup
Asset-intensive manufacturer
ROA/ROOA
NCP
Significant capital employed justifies asset-based PLI
The Decision Rule: Ask "What does this entity get paid for?" If it's sales execution → OM. If it's cost efficiency → NCP. If it's capital deployment → ROA. If it's low-risk intermediation with value in OPEX → Berry (maybe).
Jurisdictional Preferences
Tax authorities have different "comfort zones" for PLI selection. Understanding these preferences helps you anticipate audit questions.
United States (IRS)
Primary Method: CPM under §1.482-5
Common PLIs: Operating margin (~60% of APAs), Berry ratio for appropriate distributor cases, ROA
Guidance: §1.482-5(d) provides detailed definitions for sales revenue, gross profit, operating expenses, and operating profit
Berry Acceptance: Explicitly recognized; IRS practice units include Berry formula examples
Key Rule: "Non-operating items are generally excluded" (¶3.11); points to OECD ¶2.92+ for denominator selection
Preference: Operating margin commonly used; Berry ratio appears less common in practice
Sample Size: German tax authorities sometimes expect larger comparable sets
United Kingdom (HMRC)
Key Warning: HMRC criticizes TNMM applications "based on little more than a list of supposedly comparable companies" (INTM421080)
Best Practice: Start with internal comparable net margin first, then external if unavailable
Acceptance: TNMM widely used, but justify your PLI choice clearly
India
Legal Basis: Rule 10B(1)(e) permits net profit margin relative to costs, sales, assets, or "other relevant base"
Range: Rule 10CA uses 35th-65th percentile (narrower than OECD IQR)
Berry Ratio: Accepted but frequently challenged by TPOs—denominator logic is heavily litigated
Key Cases: Mitsubishi, Marubeni Itochu, Sumitomo (Berry ratio disputes)
China (SAT)
Guidance: SAT Public Notice [2017] No. 6, Article 20 explicitly lists acceptable PLIs with official formulas
Recognized PLIs: Operating margin, full cost mark-up, ROA, Berry ratio
Key Principle: "The profit level indicator selected should reflect the functions performed, risks assumed and assets used"
Practical Examples
Example 1: Same Distributor, OM vs Berry Give Different Answers
Facts:
Sales: 1,000
COGS: 850
Gross Profit: 150
Operating Expenses: 120
Operating Profit: 30
PLI Results:
PLI
Calculation
Result
Operating Margin
30 ÷ 1,000
3.0%
Berry Ratio
150 ÷ 120
1.25
Comparable Ranges (illustrative):
Operating Margin: 1.5% – 3.5% → Tested party (3.0%) is in range
Berry Ratio: 1.35 – 1.55 → Tested party (1.25) is below range
Interpretation: If COGS isn't truly pass-through (the distributor bears inventory risk, pricing risk, or adds value to products), Operating Margin may be the appropriate PLI. Berry ratio could wrongly "penalize" the high COGS intensity.
Example 2: Contract Manufacturer—NCP vs ROA Diverge
Tested Party:
Operating Profit: 20
Total Costs: 400
Total Assets: 300
PLI Results:
PLI
Calculation
Result
Net Cost Plus
20 ÷ 400
5.0%
ROA
20 ÷ 300
6.67%
Comparable Analysis:
Comparable A (asset-light): NCP 5.0%, ROA 10.0%
Comparable B (asset-heavy): NCP 5.0%, ROA 5.5%
Takeaway: If your comparables vary in asset intensity, NCP can "hide" major differences. ROA may align better with value driver when assets are material to the business model.
Common PLI Selection Mistakes
Using the "industry standard" without analysis — There's no universal "right" PLI for an industry. Match it to your tested party's specific facts.
Ignoring denominator definition — Your accounting mapping memo (what counts as OPEX vs. COGS vs. operating assets) often matters more than the PLI name.
Selecting Berry ratio without meeting conditions — If COGS isn't pass-through, Berry will produce unreliable results. Document how OECD ¶2.101 conditions are satisfied.
Inconsistent PLI between tested party and comparables — The same PLI and same definitions must apply to both.
Switching PLIs mid-analysis — If working capital differences affect OM, make adjustments rather than switching to a different PLI.
What profit level indicator should I use for a limited-risk distributor?
For limited-risk distributors, Operating Margin is the standard choice because sales volume drives the entity's value. However, if the distributor truly has pass-through COGS (no inventory risk, no pricing risk), Berry Ratio may be appropriate—but only if you can demonstrate that operating expenses capture the value-add and the OECD conditions in ¶2.101 are satisfied. Document your reasoning carefully; Berry ratio selection is frequently challenged.
When is Berry Ratio actually appropriate?
Berry Ratio is appropriate when: (1) the tested party's value-add is captured in operating expenses, (2) COGS is largely pass-through without significant value creation, (3) there are no significant unique intangibles or non-routine risks, and (4) the comparable set has consistent cost classifications between COGS and OPEX. These conditions come from OECD Guidelines ¶2.99-2.101 and were confirmed by the EU General Court in the Apple state aid case (2020).
How do I choose between Operating Margin and Net Cost Plus?
Ask: "What does the tested party get paid for?" If the entity earns returns based on sales execution (volume, customer relationships, market access), use Operating Margin. If it earns returns based on cost efficiency and delivering services/products at a markup on effort, use Net Cost Plus. For many distributors, OM is appropriate; for most contract manufacturers and service providers, NCP is preferred.
What exactly counts as "operating profit" for PLI calculation?
Operating profit should reflect profits from ordinary operating activities—typically EBIT (Earnings Before Interest and Taxes). OECD ¶2.86-2.87 states that net profit indicators "should generally be applied to profits from ordinary activities." Exclude interest income/expense, investment income, foreign exchange gains/losses (unless integral to operations), and extraordinary items. Be consistent between tested party and comparables, and document any inclusions/exclusions.
Can I use different PLIs for different comparables?
No. The same PLI must be applied consistently to both the tested party and all comparables. If some comparables don't have data for your chosen PLI, exclude them rather than mixing PLIs. Consistency is essential for reliable benchmarking.
How do I handle pass-through costs in a Net Cost Plus analysis?
Pass-through costs (where the tested party adds no value—just passes costs through at no markup) should be excluded from the cost base denominator, or included only if comparables treat them the same way. If you include pass-through subcontracting costs in "total costs," it mechanically depresses your NCP result. Define your cost base clearly and document the treatment.
What if my PLI analysis gives different results than another PLI?
This is common and expected—different PLIs measure different things. The question is which PLI best reflects the tested party's value driver. If OM shows the tested party in range but Berry shows it below range, you need to determine whether revenue or operating expenses is the true value driver. Document your analysis and explain why your chosen PLI is most reliable for the specific transaction.
How do tax authorities view ROA vs. ROOA?
Both are accepted, but ROA is more commonly used because data is more widely available. Use ROOA when the tested party has significant non-operating assets (excess cash, financial investments) that would distort a straight ROA comparison. ROOA requires clean operating asset mapping, which increases documentation burden but may produce more reliable results for capital-intensive operations.
Should I use a 3-year rolling average for PLI calculations?
Multi-year averaging is standard practice and recommended by OECD (¶3.75-3.79). A 3-year weighted average smooths economic cycles and anomalies. The typical formula is: (OP₁ + OP₂ + OP₃) ÷ (Revenue₁ + Revenue₂ + Revenue₃) for Operating Margin, or the equivalent for your chosen PLI. This approach is commonly used in APA structures.
How do I defend my PLI choice in an audit?
Document three things: (1) What drives profitability for the tested party—link to your functional analysis; (2) Why this PLI is most reliable—reference OECD ¶2.90-2.92 and explain why other PLIs would be less appropriate; (3) How you applied it consistently—show the same definitions and formulas for tested party and comparables. Courts focus on whether the PLI reflects the economic substance of the transaction, as seen in the EU Apple case where Berry ratio was upheld because the conditions were properly documented.
