Purpose
Produce a defensible, entity-by-entity build of the group's blended effective tax rate — reconciling from statutory rates to the actual ETR — and surface where jurisdictional or income-type divergence, permanent differences, or Pillar Two top-up taxes are moving the rate. Replaces the fragile spreadsheet most tax teams rebuild every quarter with a structured, auditable calculation.
When to Use
Invoke this skill when the user asks to:
- Calculate or forecast the group's blended effective tax rate
- Model Pillar Two (GloBE) top-up tax exposure across low-taxed jurisdictions
- Explain why the group ETR differs from a headline statutory rate
- Prepare an ETR reconciliation for the board, audit committee, or tax provision
- Assess the rate impact of a new entity, jurisdiction, or income mix shift
- Attribute ETR movement to specific drivers (permanent differences, incentives, top-up tax)
Inputs Required
Before starting, collect:
- Entity list with jurisdictions: Each legal entity, its tax residence jurisdiction, and functional currency
- Pre-tax book income by entity: Profit before tax per entity (or per jurisdiction), same currency or with FX rates provided
- Income type breakdown: Split of income by category where rates differ — trading, passive/investment, IP/royalty, capital gains, etc.
- Applicable statutory rates: Corporate rate per jurisdiction, plus any preferential/incentive rates or ruling-based rates
- Permanent differences: Non-deductible expenses, exempt income, tax credits, and other book-to-tax adjustments (best available)
- Pillar Two data: Jurisdictional coverage — which entities are in-scope (group revenue ≥ €750M), and any Substance-Based Income Exclusion (SBIE) inputs (payroll, tangible asset carrying value)
- Reporting basis: Whether the target ETR is accounting (IFRS/GAAP book) ETR, cash ETR, or GloBE ETR
If income-type splits or permanent differences are unavailable, calculate a first-pass ETR on statutory-rate-times-book-income and flag the omitted precision. If Pillar Two scope is unknown, test the €750M revenue threshold and state the assumption.
Instructions
Step 1: Normalize Inputs to a Common Basis
Establish the calculation foundation before any rate math:
- Convert all entity pre-tax income to a single reporting currency using provided FX rates (note rate and date used)
- Confirm the reporting basis (book ETR vs. cash ETR vs. GloBE ETR) — these are different denominators and cannot be mixed
- Aggregate entities by jurisdiction where multiple entities share a tax residence (Pillar Two operates on a jurisdictional blend, not entity-by-entity)
- Flag any entity with negative pre-tax income — losses distort blended ETR and must be shown separately, not netted silently
Step 2: Build the Entity/Jurisdiction Rate Table
For each entity (or jurisdictional group), compute:
- Applicable rate: Statutory rate, or preferential rate if a documented incentive/ruling applies
- Income-type adjustment: Where categories carry different rates (e.g., IP box, capital gains), split income and apply the correct rate to each tranche
- Current tax before permanent differences: Pre-tax income × applicable rate
- Present as a table: Entity | Jurisdiction | Pre-tax income | Statutory rate | Applicable rate | Tax at applicable rate
Step 3: Apply Permanent Differences and Credits
Adjust from statutory to real cash/book tax:
- Add back non-deductible expenses (× applicable rate) → increases tax
- Deduct exempt income (× applicable rate) → decreases tax
- Apply tax credits and incentives as direct reductions to tax, not to the base
- Compute each entity's adjusted tax charge and its entity-level ETR (adjusted tax ÷ pre-tax income)
- Note that temporary differences do NOT change the book ETR — exclude them unless computing cash ETR
Step 4: Calculate the Blended Group ETR
Roll up to the group figure:
- Blended ETR = Total adjusted tax charge ÷ Total group pre-tax income
- Show the weighting: each jurisdiction's contribution to the blended rate = (its pre-tax income ÷ group pre-tax income) × its ETR
- Present the walk from the largest statutory rate to the blended ETR as a reconciliation bridge (rate reconciliation format)
Step 5: Pillar Two (GloBE) Top-Up Assessment
For each in-scope jurisdiction where the GloBE ETR is below 15%:
- GloBE ETR = Covered taxes ÷ GloBE income for that jurisdiction (jurisdictional blend)
- Top-up percentage = 15% − GloBE ETR
- Substance-Based Income Exclusion (SBIE): Exclude a carve-out equal to the applicable % of payroll costs plus the applicable % of tangible asset carrying value (use current transition-year percentages)
- Top-up tax = Top-up % × (GloBE income − SBIE), less any Qualified Domestic Minimum Top-up Tax (QDMTT) already levied locally
- Flag each jurisdiction with a top-up liability and the estimated amount
- State clearly this is an estimate, not a filing — GloBE has adjustments (deferred tax recast, transition rules) that require full data
Step 6: Divergence Flags
Identify and explain material rate variance:
- High divergence (entity ETR more than 5 points from its statutory rate): name the driver — permanent difference, incentive, loss position, or income-type mix
- Low-taxed jurisdiction risk: any jurisdiction with ETR below 15% flagged for Pillar Two exposure even if not yet in scope (forward-looking)
- Blended-vs-statutory gap: explain the top 3 drivers moving the group blended ETR away from the weighted-average statutory rate
- Rank flags by materiality (impact on group ETR in basis points)
Step 7: Sensitivity Note
Provide a short sensitivity read:
- Which single jurisdiction most influences the blended rate, and the ETR impact if its income shifts ±20%
- Which permanent difference or incentive, if removed, moves the group ETR most
Output Format
- Structure: Entity/jurisdiction rate table → permanent-difference adjustments → blended ETR calculation → rate reconciliation bridge → Pillar Two top-up table → divergence flags → sensitivity note
- Length: 700–1,000 words plus tables
- Format: Structured document with two required tables (entity rate table; Pillar Two top-up table) and a rate reconciliation bridge; prose reserved for driver explanations and flags
- Sections: Assumptions & basis → Entity rate build → Adjustments → Blended group ETR → Rate reconciliation bridge → Pillar Two assessment → Divergence flags → Sensitivity
- Tone: Tax director — precise, conservative, assumption-explicit; every number traceable to an input or a stated estimate
Quality Checklist
Before delivering output, verify:
- [ ] Every entity appears in the rate table with jurisdiction, income, and applied rate
- [ ] Reporting basis (book / cash / GloBE) is stated once and applied consistently — no mixing of denominators
- [ ] Blended ETR equals total adjusted tax ÷ total pre-tax income (arithmetic checks out and is shown)
- [ ] Loss-making entities are shown separately, not silently netted into the blend
- [ ] Every Pillar Two top-up figure nets QDMTT and applies SBIE — and is labeled an estimate
- [ ] Each divergence flag names a specific driver and its basis-point impact — not "rate differs"
Notes
- The single most common error is blending on the wrong basis: book ETR excludes temporary differences, cash ETR includes them — pick one and hold it.
- Pillar Two operates jurisdiction-by-jurisdiction, not entity-by-entity — always aggregate entities within a jurisdiction before testing the 15% floor.
- A low blended ETR is not itself a problem; an unexplained one is. The divergence flags are the deliverable's real value — they turn a number into a defensible position.
- This is a management-estimate tool, not a tax filing. GloBE top-up figures should be confirmed against full GloBE Information Return data before any provision is booked.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills