Choose a predictive method, not a convenient reserve
Appendix C to Ind AS 12 governs uncertainty over income-tax treatments. Where acceptance is not probable, use the most-likely amount or expected value according to which better predicts resolution. Checked on 28 October 2026, this guide develops that measurement workpaper, distinct from loss-DTA recoverability and current-tax offsetting. A mathematical expected value is not automatically the required accounting answer in every dispute. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Define the unit of analysis and examination assumption
Paragraph 6 asks whether separate or grouped consideration better predicts resolution. Document the relationship between issues and the expected examination approach. Paragraph 8 assumes examination of amounts the authority has a right to examine, with full knowledge of related information. Do not reduce an exposure merely by multiplying it by a low estimated detection probability. Retain evidence about substantive acceptance, rather than a prediction that the filing will escape review. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Assess acceptance before selecting measurement
If acceptance is probable, paragraph 10 uses the treatment adopted or planned in the filing for the relevant tax inputs. Otherwise paragraph 11 requires reflecting uncertainty. A binary or highly concentrated outcome can favour most-likely amount; a dispersed range can favour expected value. These are predictive judgements, not a free choice of whichever produces the smaller liability. The examples stipulate the acceptance and predictive-method conclusions and do not supply a percentage definition of probable acceptance. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Make probabilities and amounts internally consistent
For expected value, list mutually exclusive, complete outcomes on the same basis and verify probabilities total 100%. Distinguish an incremental tax amount from an adjustment to taxable profit. Apply the relevant supported tax rate only if the model begins with a profit adjustment. Do not apply the rate again to amounts already expressed as tax. The examples use fictional incremental tax amounts or an explicitly illustrative 25% measurement rate, not an asserted Indian company rate. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Keep linked tax inputs consistent
Paragraph 12 requires consistent judgements where uncertainty affects current tax and deferred tax. A denied deduction can have different consequences if it is permanently disallowed or merely deferred. Establish the actual legal effect before creating a corresponding tax base or deferred asset. The isolated current-tax models do not automatically establish an equal deferred-tax offset or the recoverability of any resulting asset. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Reassess on evidence and preserve disclosure conclusions
Paragraphs 13–14 address changed facts or information, Ind AS 8 estimates and the applicable Ind AS 10 review for later events. Mere authority silence is not, in isolation, the change in evidence described in application guidance A3. Keep the prior estimate, new evidence, revised outcomes and movement together. Review the applicable judgement, estimation and tax-contingency disclosures; do not treat a booked accounting estimate as an admission that the tax position is legally wrong. ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5.
Applicable primary references
ICAI Ind AS 12, Appendix C: paragraphs 6–14 and application guidance A1–A5
Six worked illustrations
All facts and amounts below are fictional. Assumptions describe the model, not additional statutory rules.
Illustrative example 1: Binary outcome and the most-likely amount
Assume acceptance is not probable and most-likely amount better predicts a binary incremental tax outcome: ₹0 with 30% probability or ₹10 lakh with 70%. The selected most-likely amount is ₹10 lakh. The ₹7 lakh arithmetic weighted mean is not substituted where that method is not the better predictor.
Illustrative example 2: A dispersed range
Assume expected value better predicts incremental tax outcomes ₹0/₹10/₹20 lakh with probabilities 20%/50%/30%. The amount is 0 × 20% + 10 × 50% + 20 × 30% = ₹11 lakh. Probabilities sum to 100%.
Illustrative example 3: Do not apply tax twice
Assume expected value is appropriate for incremental tax amounts ₹4/₹8/₹12 lakh with probabilities 25%/50%/25%. The estimate is ₹8 lakh. Multiplying that tax amount by an illustrative 25% again would incorrectly produce ₹2 lakh.
Illustrative example 4: Convert a profit adjustment once
Assume expected value is appropriate for additional taxable-profit outcomes ₹0/₹20/₹40 lakh, probabilities 25%/50%/25%. Expected additional taxable profit is ₹20 lakh. With the explicitly illustrative supported-rate assumption 25%, the isolated tax effect is ₹5 lakh, not ₹20 lakh.
Illustrative example 5: Exclude a detection discount
Assume a properly selected substantive measurement of incremental tax ₹12 lakh. An internal 10% detection estimate does not turn it into ₹1.20 lakh under the full-knowledge examination assumption. The difference is ₹10.80 lakh; no separate legal penalty or interest is modelled.
Illustrative example 6: Reassess a changed distribution
Assume expected value initially uses tax outcomes ₹0/₹10/₹20 lakh with 20%/50%/30%, producing ₹11 lakh. Relevant new evidence changes probabilities to 10%/30%/60%, producing ₹15 lakh. The increase is ₹4 lakh; determine its accounting treatment and event timing from the applicable standards.
Ten-step verification table
| Step | Action | Evidence |
|---|---|---|
| 1 | Identify the uncertain income-tax treatment | Actual or planned filing position and governing law |
| 2 | Define the unit of analysis | Separate/grouped conclusion and predictive basis |
| 3 | Apply informed examination assumption | No detection-risk discount |
| 4 | Assess probable acceptance | Substantive legal and factual support |
| 5 | Select the better predictive method | Most-likely or expected-value justification |
| 6 | Define consistent outcomes | Tax amounts versus taxable-profit adjustments |
| 7 | Validate probabilities and arithmetic | Complete distribution, 100% total and rate applied once |
| 8 | Check linked current/deferred inputs | Consistent assumptions and actual legal timing effect |
| 9 | Review changed facts and later events | New evidence, estimate movement and reporting-date analysis |
| 10 | Document disclosures and reconciliation | Judgements, estimation uncertainty, contingency review and closing amount |
Sixteen practitioner FAQs
1. Is every tax dispute measured at expected value?
No. Select the method that better predicts resolution under Appendix C.
2. When may most-likely amount be useful?
Binary or concentrated outcomes may support it, subject to the actual predictive analysis.
3. When may expected value be useful?
A dispersed range of outcomes can favour probability-weighted measurement.
4. Can management choose the lower answer?
No. The criterion is predictive quality, supported by evidence.
5. Can related issues be grouped?
Paragraph 6 requires the separate or grouped approach that better predicts resolution.
6. Does a low detection estimate reduce the exposure?
The examination assumption requires full knowledge for amounts the authority has a right to examine.
7. What if acceptance is probable?
Use the filing treatment for the relevant inputs under paragraph 10, while reviewing applicable disclosures.
8. Do the examples define probable by a numerical threshold?
No. They stipulate the conclusion without inventing a universal percentage test.
9. Must expected-value probabilities total 100%?
Yes. Use a complete and internally consistent outcome population.
10. Should a tax rate be applied to an amount already stated as tax?
No. That would apply the tax effect twice.
11. Does the illustrative 25% prescribe an Indian statutory rate?
No. Actual accounts require their supported applicable measurement rate.
12. Does an uncertain current-tax charge automatically create equal DTA?
No. Establish the legal timing effect and independently applicable recognition requirements.
13. Must current and deferred estimates be consistent?
Paragraph 12 requires consistent judgements for linked effects.
14. Does tax-authority silence automatically resolve uncertainty?
Application guidance A3 says silence alone is unlikely to constitute the relevant new information.
15. Can new evidence change the amount?
Yes. Reassess under paragraphs 13–14 and review the relevant estimate and reporting-date treatment.
16. Is the accounting estimate a concession of legal liability?
No. Keep the accounting measurement and the legal tax position clearly documented separately.
Conclusion
Choose the method that better predicts resolution, assume informed tax examination and keep current-tax and deferred-tax estimates consistent.