Transfer Pricing at Year-End: One Comparable, One Allocation—and Several Red Flags

 By CA Surekha S Ahuja

What the TVS Motor ruling teaches before financial statements and TP positions are finalised

Transfer pricing should be stress-tested at closing—not defended for the first time after the TPO changes the numbers.

That is the real practical lesson from the recent ITAT Chennai decision in TVS Motor Co. Ltd. v. DCIT, order dated 9 September 2026 for AY 2022-23.

At first sight, the case is about a narrow question:

Can a taxpayer using a single internal comparable still claim the statutory tolerance band where the range concept under Rule 10CA does not apply?

The Tribunal held that it can.

But for companies currently finalising their financial statements, the larger lesson is more important:

A transfer-pricing position should not be reviewed only on the numbers produced by the business. It should be stress-tested for the changes a TPO could make—and for the other AE exposures that sit outside the tolerance-band calculation altogether.

What happened in TVS Motor?

TVS Motor applied TNMM with OP/OC as the PLI and used its own non-AE/domestic segment as the internal comparable.

The TPO's principal intervention was the allocation of depreciation between the AE and non-AE segments. That changed the AE segment margin materially.

The assessee argued that even on the TPO's recomputed figures, the difference remained within the notified 3% tolerance band.

The relevant figures were:

	Margin
Internal comparable / non-AE segment	1.31%
AE segment after TPO's recomputation	–0.46%
Variance	Within ±3%

The Tribunal accepted the tolerance-band argument and deleted the adjustment.

How can one comparable have an “arithmetic mean”?

This was the central legal question.

Rule 10CA(1)–(6) provides the range concept in specified cases where six or more comparables are used.

Rule 10CA(7) applies where the range mechanism does not apply.

The Revenue argued that an arithmetic mean necessarily required more than one value.

The Tribunal's answer was straightforward:

The arithmetic mean of a dataset containing one value is that value itself.

Rule 10CA(7) does not prescribe a minimum number of comparables before the tolerance mechanism can operate.

Therefore, a single internal comparable is not automatically excluded from the Rule 10CA(7) tolerance-band analysis.

The more important lesson: the comparable and the allocation travel together

The internal comparable in TVS Motor was not just a number in a TP study.

Its usefulness depended upon the underlying segmental results and allocation of costs, particularly depreciation.

This is where the year-end risk begins.

If depreciation or common costs are allocated differently, the AE margin can change.

And if the margin changes, the transfer-pricing conclusion can change.

The Tribunal ultimately did not decide whether the TPO's depreciation allocation itself was correct because the tolerance-band issue disposed of the adjustment.

So the ruling should not be read as approving any particular allocation methodology.

The practical lesson is:

Do not merely test the TP conclusion. Test the assumptions and allocations that produce the TP conclusion.

And the TP review should not stop at the operating margin

This is particularly important in group structures.

A company may be comfortable with its operating-margin analysis and still have separate TP exposures arising from other dealings with its AEs.

Corporate guarantee

A corporate guarantee given on behalf of an AE should be reviewed independently.

It should not simply be folded into the operating-margin tolerance-band analysis.

The attached guidance note identifies corporate guarantee as a separate benchmarking exposure and specifically states that it is not covered by the tolerance-band ratio.

At year-end, therefore, ask:

• Have all guarantees given for AEs been identified?
• Are the amount, tenure, beneficiary and underlying borrowing documented?
• Has the applicable TP treatment and benchmarking been considered?
• Is the guarantee exposure being analysed separately from the operating-margin tolerance?

Outstanding AE receivables

Delayed realisation from an AE is another area requiring separate attention.

The attached note identifies outstanding receivables beyond the credit period as a potential international-transaction exposure and states that this issue is not covered by the TVS Motor tolerance-band reasoning.

So the year-end review should also ask:

Are there significant AE receivables outstanding beyond the agreed credit period, and has the resulting TP exposure been separately examined?

The year-end TP red-flag test

Before the financial statements and TP documentation are frozen, review the complete AE relationship:

Red flag	Closing-stage question
Internal comparable	Is it genuinely comparable and supported by reliable segmental data?
AE vs non-AE margin	What is the variance on the relevant PLI?
Depreciation allocation	Can another allocation materially change the AE margin?
Common costs	Is the allocation basis rational, documented and consistently applied?
Comparable count	Does the range concept apply, or should Rule 10CA(7) be examined?
Tolerance band	Has the percentage applicable to the relevant AY been correctly verified?
Corporate guarantee	Has the guarantee been identified and independently benchmarked?
AE receivables	Has delayed realisation been separately examined?
Documentation	Would the TP file explain the position without depending on a future litigation argument?

The attached guidance note itself identifies internal comparables, segmental allocations, corporate guarantees and outstanding receivables as separate year-end areas requiring attention.

Six questions before signing off the TP position

1. What is the most appropriate method and PLI for each AE transaction?

2. What exactly is the comparable set—and if there is one internal comparable, is its segmental basis defensible?

3. Can depreciation or common-cost allocation materially change the result?

4. Does the applicable tolerance mechanism cover the case, and has the percentage for the relevant AY been verified?

5. Have corporate guarantees been separately identified and benchmarked?

6. Have delayed AE receivables and other separate TP exposures been reviewed?

The tolerance percentage should be verified against the applicable notification for the relevant assessment year rather than simply carried forward from the preceding year.

Where the position falls within the applicable tolerance, the conclusion should be documented contemporaneously in the TP study and Form 3CEB.

Where the position is outside the tolerance or remains uncertain, the related tax provision and contingent-liability implications should be considered during financial-statement finalisation.

A favourable Tribunal ruling is not a year-end strategy

TVS Motor gives an important legal answer where Rule 10CA(7) applies.

But the purpose of a year-end TP review should not be to prepare for a long litigation journey.

The better objective is:

Identify → stress-test → correct where possible → document → assess the remaining exposure.

That review should happen before the accounts and TP documentation are frozen.

Once an adjustment is proposed, the focus necessarily shifts from prevention to defence.

Concluding words

The real lesson of TVS Motor is not merely that one comparable can still qualify for the tolerance band.

It is that transfer pricing has to be reviewed as a complete AE relationship—not just as an operating-margin calculation.

Before closing the year, look at:

the comparable → the segmental margin → depreciation → common costs → tolerance → corporate guarantee → receivables → documentation.

A favourable Tribunal ruling may provide protection in an appropriate case.

It should not become the year-end strategy.

The better strategy is to identify the red flags while the numbers can still be reviewed, the allocation can still be examined, the documentation can still be strengthened and the financial statements can still be finalised with the exposure understood.

Transfer pricing should be stress-tested at closing—not defended for the first time after the TPO changes the numbers.

Important

TVS Motor is a Tribunal-level, fact-specific decision. Its reasoning concerns the non-range framework under Rule 10CA(7) and should not be mechanically extended to cases where the range provisions apply or to materially different methods and facts. The applicable tolerance percentage must be verified for the relevant assessment year.

LLP to Private Limited Company: Do You Really Need a Merger- 8 Smarter Restructuring Routes

8 restructuring routes promoters should examine before converting, transferring or merging

By CA Surekha S Ahuja

A promoter has an LLP and a private company. The businesses overlap.

The instinctive question is: “Can we merge the LLP into the company?”

That may not be the right starting point. The better question is:

What exactly must the restructuring achieve?

Is the objective ownership, cooperation, investor readiness, succession, selective transfer of a business or assets—or genuinely one legal entity carrying everything forward?

The answer can change the entire route, cost, tax position, stamp-duty exposure, compliance burden and timeline.

First decide what needs to change

An LLP and a company are separate legal entities. Their restructuring mechanisms are not interchangeable. The LLP Act contains provisions dealing with reconstruction and amalgamation of LLPs, while company mergers operate within the Companies Act framework.

The relevant framework may involve, depending on the proposed route, the LLP Act, Sections 366 and 230–232 of the Companies Act, and the applicable income-tax provisions governing conversion, succession and tax-attribute continuity.

Therefore, an LLP-to-company merger should not be assumed to be an ordinary company-to-company merger.

Eight routes promoters should examine

Objective	Possible route	What changes?	Principal issue
Ownership	LLP becomes shareholder	LLP holds shares in company	Business stays in LLP
Participation	Company becomes LLP partner	Company acquires partnership interest	LLP agreement becomes critical
Cooperation	JV / commercial arrangement	Contractual relationship only	Tax, GST and related-party implications
Corporate form	LLP converted into company	Legal form changes	Tax-neutrality conditions
Selective integration	Conversion + transfer	Selected business/assets move	Capital gains, GST, stamp duty, contracts
Complete consolidation	Conversion + merger	One company survives	NCLT, valuation, cost and time
Direct LLP → company	Specialist restructuring route	Proposed direct consolidation	Statutory availability must be established
Multiple LLPs	LLP amalgamation + conversion	LLPs consolidate first	Two-stage restructuring

These are not merely eight technical alternatives.

They represent eight different levels of disruption to legal identity, assets, liabilities and tax attributes.

Ownership does not require moving the business

If the objective is simply to bring the LLP and company under a common ownership structure, the business itself may not need to move.

An LLP can hold shares in a private company.

The LLP can continue to own its business, assets, contracts and liabilities while participating in the company's ownership.

Dividend, share transfer, buyback and other shareholder-level consequences should, however, be separately examined.

The reverse structure can also be relevant: a company may become a partner in an LLP where commercial participation in the LLP business is intended.

In that case, capital contribution, profit sharing, voting, reserved matters, exit and deadlock provisions need careful drafting.

Change the ownership without unnecessarily changing the business.

Cooperation may be enough

Sometimes the businesses need to work together, not become one entity.

Management services, shared services, licensing, financing, use of intellectual property or other commercial arrangements can achieve substantial integration without transferring the underlying business.

That may avoid a major restructuring altogether.

But “simple” does not mean undocumented.

Pricing, valuation, GST, income-tax, related-party requirements, TDS and transfer of resources must still be examined on the facts.

Conversion and merger solve different problems

If the LLP structure itself has become unsuitable—for example, because of equity investment, conventional corporate ownership, succession or future investor requirements—conversion into a company may be appropriate.

But conversion does not automatically require merger.

Conversion changes the legal form.
Merger eliminates the separate entity.

If conversion achieves the commercial objective, merging immediately afterwards may simply add another layer of cost and procedure.

Where only a particular business, undertaking or asset needs to move, selective transfer may be more appropriate.

That route requires separate modelling of:

• capital gains;
• GST;
• stamp duty;
• valuation;
• contract novation or consents;
• financing arrangements; and
• assumption or retention of liabilities.

When does complete consolidation justify a merger?

A full consolidation becomes relevant when the commercial requirement is genuinely:

one legal entity + one balance sheet + one ownership structure + consolidated assets and liabilities.

A possible broad restructuring sequence is:

LLP → Company → NCLT merger → Surviving company

This can achieve much greater legal consolidation, but it is also the most elaborate route.

The planning analysis may involve 8–15 months, depending on the facts and approvals. This is a practical planning range, not a statutory timeline.

Valuation, professional fees, notices, creditor considerations, hearings, documentation, stamp duty and post-merger implementation can materially affect both cost and timing.

Therefore:

A merger should solve a real consolidation or succession problem—not merely the inconvenience of having two entities.

The direct LLP-to-company route requires a separate legal opinion

This is an area where promoters should resist “standard template” advice.

The LLP Act provides a framework for reconstruction and amalgamation of LLPs. Company mergers, on the other hand, operate within the Companies Act framework.

That does not automatically establish a routine direct LLP-to-existing-company merger mechanism identical to a company-to-company merger.

Accordingly, where a direct LLP → company route is proposed, the first step should be to obtain a specific written legal opinion on the statutory route, approvals, tax consequences and implementation mechanism.

Establish the route first. Spend on implementation second.

Tax neutrality must be tested separately

A legally valid conversion does not automatically make the transaction tax-neutral.

The applicable income-tax provisions need to be tested for the particular conversion, including their conditions and consequences for tax attributes.

The transition to the Income-tax Act, 2025 with effect from 1 April 2026 should also not be misunderstood as wiping out conditions attached to earlier transactions or transactions whose tax treatment depends on prescribed continuity requirements.

The conversion file should answer five questions

Tax area	Question to establish
Capital gains	Does the applicable exemption or transition provision protect the conversion?
Losses	Can brought-forward business losses continue?
Depreciation	What happens to unabsorbed depreciation?
Tax credits	What is the statutory basis for their continuation?
Continuing conditions	What must remain true after conversion?

Pending assessments, demands, refunds, appeals and litigation also need an entity-by-entity transition plan.

Business continuity is not, by itself, proof of tax-attribute continuity.

Where a five-year continuity condition applies, it should be treated as a post-conversion compliance obligation to be monitored, not as a box to be ticked and forgotten.

A better restructuring sequence

Do not begin with:

Merger → documentation → tax consequences

Begin with:

Commercial objective
↓
Which entity should ultimately survive?
↓
What actually needs to move?
↓
What should remain where it is?
↓
Which tax attributes must be preserved?
↓
What are the entry, implementation and exit costs?
↓
Select the simplest lawful structure

This sequence often prevents unnecessary restructuring.

Calculate the total cost—not just the professional fee

The cost of a restructuring is not the merger fee.

A proper model should consider:

Professional and statutory costs + tax cost + stamp duty + GST impact + financing consequences + compliance cost + management time + contractual disruption + future exit or unwind cost + risk to tax attributes

A structure that is inexpensive to create may be expensive to maintain.

And a structure that looks elegant on paper may create unnecessary tax, compliance or contractual friction.

The restructuring decision

If the objective is ownership: examine ownership structures.

If it is cooperation: consider a commercial arrangement.

If it is investor readiness: examine conversion.

If it is selective integration: consider selective transfer.

If it is complete succession or consolidation: evaluate conversion followed by merger.

If a direct LLP → company merger is proposed: establish the statutory route before implementation.

Do not merge because you have two entities.

Merge only when the business actually needs one surviving legal entity.

The best restructuring is not necessarily the one that achieves the maximum integration.

It is the one that achieves the commercial objective with the minimum unnecessary movement of legal identity, assets, liabilities, tax attributes and compliance burden.

Casahuja Perspective

Restructuring should begin with the business objective—not the legal form.

Before asking “How do we merge the LLP into the company?”, ask:

“What is the minimum structural change required to achieve what the promoter actually wants?”

That one question can determine whether the right answer is no merger at all, ownership restructuring, conversion, selective transfer—or a full consolidation.

Proprietor Dies? GST Can Transition the Business. Income Tax Requires a Succession Strategy — 2026 Guide

 By CA Surekha S Ahuja

What happens to GST registration, ITC, tax liabilities, losses, inherited assets and the business when a sole proprietor dies? A practical 2026 guide to succession, tax continuity and the critical differences between GST and Income Tax.

A sole proprietor's death can happen in a day. The business cannot.

Customers still need invoices. Employees need salaries. Receivables and payables remain. GST returns may be due. Tax notices may be pending. Assets may carry years of tax history.

But the person behind the business is no longer there.This creates an important distinction: 

GST has a transition mechanism. Income Tax has a succession framework. The real challenge is connecting the person, the business, the period, the tax attributes and the assets.

That is the real succession problem.

Two Tax Laws. Two Different Succession Questions

The mistake is to treat the proprietor's death as one common "tax succession" exercise.

It is not.

GST	Income Tax
Primarily deals with continuation of the business	Primarily deals with the deceased taxpayer and succession
Provides a route for GST registration and eligible ITC transition	Determines who represents the deceased and who succeeds to the business
ITC-02 provides the mechanism for eligible ITC transfer	Loss carry-forward depends on the loss, succession and statutory conditions
Old GST liabilities remain relevant	Pre-death and post-death income require separate treatment
Focus: business transition	Focus: person, period, succession, tax attributes and assets

This distinction should determine the entire workflow.

The Succession Map

              BEFORE DEATH
           PLAN THE SUCCESSION
                    ↓
             DEATH OF PROPRIETOR
                    ↓
      FREEZE → SEPARATE → TRANSFER
                    ↓
             PROTECT → CONTINUE
                    ↓
             NRI HEIR?
       TAX + FEMA + BANKING
          BEFORE MONEY MOVES

Before death: plan the succession

Death cannot be planned. Succession can.

Where the business is expected to continue, establish in advance:

• who is expected to continue;
• what assets and liabilities belong to the business;
• where books and tax records are maintained;
• what registrations, bank accounts and contracts exist;
• what tax disputes or notices are pending; and
• whether the intended successor is resident or NRI.

The objective is simple:  Death should not become the first day on which the family discovers how the business operates.

After Death: First Establish What Existed

The first step should not be cancellation. It should be establishing what existed on the date of death.

Prepare a Date-of-Death and Succession Statement covering:

Area	Establish
Business	Stock, receivables, creditors, contracts
GST	Returns, ITC, notices and demands
Income Tax	Income, TDS, payments, losses, proceedings
Assets	Property, investments, machinery
Liabilities	Loans, creditors, tax exposure
Records	Books, title documents, agreements
Succession	Legal representative and business successor

This creates the factual bridge between the deceased and the successor.

GST: The Law Provides a Transition Route

GST is principally concerned with continuing the registered business.

For the death of a sole proprietor, the GST framework provides a relatively clear transition mechanism.

Where the business continues, the successor deals with the GST registration and eligible unutilised ITC. CBIC Circular No. 96/15/2019-GST provides that FORM GST ITC-02 should be filed before applying for cancellation of the deceased proprietor's registration; on acceptance, eligible ITC is credited to the successor's electronic credit ledger.

The GST sequence

DEATH → SUCCESSION → SUCCESSOR GST REGISTRATION → ITC-02 → ELIGIBLE ITC TRANSFER → CANCELLATION / TRANSITION

The practical message: Do not make GST cancellation the first step. Deal with the eligible ITC transition first.

GST Transition Does Not Erase GST Liability

Moving the business does not wipe out its tax history.

Section 93 of the CGST Act provides the statutory framework for tax, interest and penalty where a person liable to tax dies and the business is continued. The liability position therefore needs to be established before the successor takes over.

Prepare an: OLD GST LIABILITY REGISTER

Capture:  unpaid tax, interest and penalty; notices and assessments; appeals; refund disputes; and

• periods still open for action.

The GST registration may change. The GST history does not.

Income Tax: Succession Starts With the Person, Not the Registration

Income Tax begins with a different question. It must establish: 

Who represents the deceased?

What income belongs to the deceased's period?

Who succeeded to the business?

What happens to eligible losses?

What happens to the assets and their tax history?

This makes an important distinction necessary: Legal representative and business successor are not automatically the same legal capacity.

A person may represent the deceased for tax compliance while another person becomes the person continuing the business.

The documentation should make that distinction clear.

Income Tax Has a Date-of-Death Cut-Off

Unlike GST's registration-and-ITC transition, Income Tax requires the taxpayer's period and capacity to be identified.

UP TO THE DATE OF DEATH : The tax affairs of the deceased.

AFTER DEATH / SUCCESSION : The applicable position of the estate, successor or other person entitled to the income, depending on the facts and legal arrangement.

The Income-tax Act, 2025 separately addresses legal-representative liability and succession to a business or profession.

The practical rule: Create a clean date-of-death cut-off instead of simply continuing the deceased proprietor's books under a new identity.

Income-tax Act, 2025: Continuity, But Not Automatic Transfer

The new Act does not by itself erase valid historical tax positions from 1 April 2026.

The repeal-and-saving framework preserves applicable earlier rights, liabilities, proceedings and other tax consequences, subject to the relevant conditions. This is particularly relevant for historical losses.

But two questions must remain separate:  Did the loss survive?

and Can the successor use it?

The second question depends on the particular loss and the manner of succession.

The analysis should therefore be:  LOSS → VALIDITY → TYPE → MODE OF SUCCESSION → CONDITIONS → REMAINING PERIOD

The law distinguishes succession by inheritance from certain other forms of succession.

So neither blanket statement is safe: "All losses disappear on death."

"All losses automatically transfer."

The correct answer is loss-specific and succession-specific.

Where the Two Laws Still Need Different Treatment

Question	GST	Income Tax
What is being transitioned?	Registered business, including eligible ITC	Taxpayer's affairs and, separately, the succeeded business
Who matters?	Successor / person continuing the business	Legal representative, estate and business successor, depending on the issue
Key dividing point	Registration transition	Date of death and succession
Tax attribute	Eligible unutilised ITC	Losses subject to their own conditions
Past liability	GST dues remain relevant	Deceased's tax liability continues through the statutory legal-representative framework
Asset history	Relevant to business records	Critical for future tax computation
NRI issue	Not primarily a GST issue	Tax + FEMA + banking may arise on inherited assets and remittance

The difference can be reduced to one line:

GST asks, "How does the business move?" Income Tax asks, "Who is responsible, for which period, in what capacity, and with what tax history?"

Inherited Assets: The Title Moves, the History Must Follow

An inherited property may be sold years later.

Preserve: original acquisition documents; succession and title documents; historical cost; improvement expenditure; valuation evidence where relevant; and earlier tax records.

The chain is: INHERIT → OWNERSHIP → HOLDING → SALE → CAPITAL GAINS

Succession must preserve history, not merely title.

NRI Heir: Add FEMA Before the Money Moves

For an NRI heir, the succession file acquires another layer.

The sequence becomes:  INHERIT → TITLE → TAX REVIEW → SALE → TAX / TDS → BANKING → REPATRIATION

RBI's framework permits an NRI/PIO, subject to applicable conditions and documentation, to remit up to USD 1 million per financial year from specified NRO balances, sale proceeds and eligible assets acquired through inheritance or legacy. Applicable FEMA, tax and authorised-dealer requirements must be satisfied.

The practical rule: Do not sell first and ask the FEMA question later.

A Practical 30-Day Working Model

This is a working model, not a statutory deadline.

Stage	Action
1–7	FREEZE — death date, records and Day-Zero position
8–15	SEPARATE — pre-death and post-death transactions
16–20	TRANSFER — succession, GST registration and ITC-02
21–25	PROTECT — notices, liabilities, losses and assets
26–30	CONTINUE — successor books, banking and operations

NRI heir: add TAX + FEMA + BANKING before sale or remittance.

Five Errors That Turn Succession Into a Tax Problem

1. Treating GST and Income Tax as one succession exercise

Their objectives and mechanisms are different.

2. Cancelling GST before dealing with eligible ITC

The ITC-02 mechanism must form part of the transition sequence.

3. Mixing pre-death and post-death transactions

Establish a clean accounting and tax cut-off.

4. Assuming losses automatically disappear or transfer

Examine the particular loss and succession route.

5. Selling inherited assets before planning the NRI tax/FEMA/banking route

The sale and movement of money should be considered together.

Final Perspective

Death Is Sudden. Succession Should Be Structured.

A proprietor's death is not one tax event. It is a connection problem.

GST : BUSINESS → REGISTRATION → ITC → LIABILITIES → TRANSITION

Income Tax : DECEASED → LEGAL REPRESENTATIVE → DATE OF DEATH → SUCCESSION → LOSSES → ASSETS → SUCCESSOR

The Income-tax Act, 2025 preserves important historical tax positions subject to applicable conditions. But the practical challenge remains identifying who is acting, for whom, for which period and in respect of which business, loss or asset.

The best succession file should answer:

What existed on the date of death?

Who became responsible, and in what capacity?

Who succeeded to the business?

Which liabilities and tax attributes survived?

What evidence will be required years later?

For an NRI successor:

How will the money legally move after the asset is sold?

GST can transition the business. Income Tax can address succession. The real work is connecting the person, the business, the period, the tax attributes and the assets.

The practical discipline

BEFORE DEATH
PLAN THE SUCCESSION

AFTER DEATH
FREEZE → SEPARATE → TRANSFER → PROTECT → CONTINUE

NRI HEIR
PLAN TAX + FEMA + BANKING BEFORE MONEY MOVES

Death may be unexpected. Succession should not be unstructured.

Gold Monetisation Scheme 2.0: How India Can Turn 34,000 Tonnes of Gold into Income, Security and Economic Growth

 By CA Surekha Ahuja

India does not have a gold shortage. It has a gold mobilisation problem. The next Gold Monetisation Scheme should solve the owner's problem first — and the economy's problem as a result.

India's households are estimated to hold around 34,000 tonnes of gold. Yet only a fraction has entered formal monetisation channels. The Medium and Long Term Government Deposit components of the Gold Monetisation Scheme were discontinued from 26 March 2025, while banks may continue to offer Short Term Bank Deposits at their discretion.

That experience offers a simple policy lesson.

The next scheme should not begin by asking:

How much gold can the Government collect?

It should begin by asking:

Why would a household, temple, trust or institution willingly give up physical possession of its gold?

That is the real design question for GMS 2.0.

The problem is not gold. It is trust.

Gold is not an ordinary financial asset.

A family may hold jewellery inherited over generations. A temple may hold ornaments with religious significance as well as monetary value. A trust may hold gold whose ownership, accounting and governance cannot simply be treated like an individual's investment.

Therefore, asking people to melt, assay and surrender gold in return for a conventional financial return may not be enough.

A successful GMS 2.0 should make the owner confident that:

My gold is safe.
My ownership is protected.
Its value is transparent.
I can redeem it when genuinely required.
I am not giving away the benefit of future gold appreciation.
And I am not creating an unnecessary tax or documentation problem.

Once these concerns are addressed, mobilisation becomes much more realistic.

What should GMS 2.0 actually offer?

Design issue	Recommended approach	Why it matters
Gold linked return	Include	The depositor should retain meaningful economic participation in gold
Easy redemption	Include	Liquidity is central to the attractiveness of gold
Compulsory melting	Avoid as the only route	Jewellery and institutional ornaments can have value beyond metal content
Verified custody	Include	Owners must be comfortable surrendering physical possession
Digital gold registry	Include	Ownership, custody, pledge and redemption need an auditable record
Temple and institutional window	Include separately	Legal ownership, authority, accounting and governance are different
Jeweller network	Use with regulation	Existing networks provide reach and convenience
Insurance	Mandatory	Custody risk must not remain with the depositor
Independent audit	Mandatory	Physical gold and digital records must reconcile
Tax safe harbour	Narrow and time-bound	Documentation gaps should not prevent genuine household participation
Large HNI holdings	Enhanced scrutiny	Larger holdings require stronger source, ownership and governance checks
Gold backed lending	Phase 2	Lending should follow reliable ownership and valuation
Blockchain or similar technology	Use only if useful	Legal enforceability and auditability matter more than technology branding

The biggest change: give the owner a choice

GMS 2.0 should not force every owner into one product.

It could have three distinct routes.

1. Gold Deposit

The owner deposits eligible gold and receives a gold-linked financial claim with clearly defined return and redemption rights.

This is the mass-market route.

2. Secure Gold Custody

The owner retains economic ownership while the physical gold is placed in professionally controlled and insured custody.

This could be particularly useful for:

• inherited jewellery
• high-value family holdings
• temple ornaments
• institutional gold
• bullion

The objective is important:

Monetise the financial utility of gold without unnecessarily destroying the form in which the owner values it.

3. Gold-Backed Financial Asset

Once ownership, custody and valuation systems are reliable, the verified gold balance can support regulated lending and other financial products.

This is where the larger economic opportunity lies.

But it should come after the trust infrastructure is established.

Temples need a separate lane

The institutional opportunity should not be reduced to:

“Temples should deposit their gold.”

The better proposition is:

“Give temples and eligible institutions a safer way to protect, account for and earn from gold that can lawfully be monetised.”

Before institutional gold enters the system, GMS 2.0 should clearly establish:

• who legally owns the gold
• who is authorised to deposit it
• who can redeem it
• how it is independently valued
• how income is accounted for
• who approves the transaction
• how changes in trustees or management affect the deposit
• how the asset remains identified in institutional records
• how religious and legal restrictions are protected

This could create an important double benefit.

Less gold physically stored on vulnerable premises, and a potential financial return on otherwise idle assets.

But the legal character and ownership of institutional gold must remain protected.

Security should be the foundation

When physical gold leaves the owner's premises, custody becomes the heart of the scheme.

A credible GMS 2.0 should therefore provide:

Digital ownership records
Every deposit should have a secure record of quantity, purity, owner, custody status, pledge status and redemption rights.

Multi-party controls
No single custodian or intermediary should be able to move or encumber gold unilaterally.

Mandatory insurance
Responsibility for loss, theft or custody failure should be clearly allocated.

Independent audits
Physical stocks and digital records should be periodically and independently reconciled, with surprise verification where appropriate.

Clear liability
The depositor should know in advance who pays if something goes wrong.

Technology can support this architecture.

But technology is not the architecture.

A blockchain ledger cannot compensate for weak ownership rules, poor custody or uncertain redemption.

Do not make melting the price of admission

For bullion, conventional assaying and refining may be straightforward.

For family jewellery and institutional ornaments, it is different.

A better system could have two tracks.

Track A — Assay and refine

Eligible gold is standardised and credited to the depositor.

Track B — Verified custody

Eligible gold remains identifiable, securely stored and digitally recorded, subject to prescribed valuation, inspection and custody safeguards.

Track B is more complex.

But it addresses a fundamental psychological barrier:

People may be willing to give up possession of their gold without being willing to destroy its identity.

Tax certainty can unlock household gold

Another major barrier is documentation.

Gold accumulated over generations may not always have invoices or a complete purchase trail.

The answer should not be a permanent blanket amnesty.

A better approach would be a narrow, clearly defined and time-bound safe harbour for genuine household holdings, subject to prescribed conditions.

For large HNI, business and institutional holdings, the approach should be different:

more disclosure, stronger beneficial ownership verification, better valuation evidence and stronger governance documentation.

The objective should be formalisation — not a back door for unexplained assets.

Gold-linked returns could change the proposition

The owner needs an economic reason to participate.

If gold prices rise substantially, a conventional fixed rupee return may make the depositor feel that the principal economic attraction of gold has been surrendered.

GMS 2.0 should therefore examine whether a transparent gold-linked return mechanism can be structured within the applicable monetary, banking, tax and securities framework.

The precise product can be debated.

The principle should not be ignored:

The depositor should not feel that monetisation means giving away gold's future economic value.

That could fundamentally change the psychology of the scheme.

Lending should come second

Gold-backed credit offers a potentially powerful economic channel.

But GMS 2.0 should resist the temptation to begin with lending.

The sequence should be:

Gold → Verified Ownership → Secure Custody → Transparent Valuation → Digital Registry → Redemption → Lending

Not:

Gold → Loan

Once the infrastructure is proven, a regulated lending layer could address:

• pledge creation
• valuation
• loan to value
• margin requirements
• default
• notice
• auction
• recovery
• surplus return to the owner
• release of pledge

This sequencing would allow the Government to build the financial ecosystem on reliable collateral rather than unresolved custody problems.

Jewellers can provide the last mile

India already has a vast gold ecosystem.

GMS 2.0 should use that infrastructure rather than recreate it.

Authorised jewellers, refiners, assayers and other regulated participants could potentially become collection and service points, while ownership records, custody and settlement remain subject to strong regulatory controls.

The objective should be simple:

Make depositing gold almost as convenient as holding it.

Convenience will matter as much as return.

What does the economy gain?

A successful GMS 2.0 can be much more than a gold deposit scheme.

Lower dependence on fresh imports

Mobilising domestic gold can potentially reduce the need to meet every incremental demand through fresh imports, although the actual macroeconomic effect would depend on participation, product design and how mobilised gold is used.

Productive use of household wealth

Physical wealth that currently sits idle can become part of the formal financial system.

Safer institutional assets

Temples and eligible institutions could potentially reduce physical-security exposure while retaining lawful economic ownership.

Deeper financial markets

A reliable gold registry and custody system could support new regulated products.

More formal credit

Once ownership and valuation are reliable, gold can become higher-quality collateral for regulated lending.

Greater formalisation

A carefully designed disclosure and documentation framework can bring genuine household gold into the formal economy without creating a permanent amnesty.

The Government should measure more than tonnes

The success of GMS 2.0 should not be judged simply by the quantity of gold collected.

A scheme could mobilise thousands of tonnes and still be economically inefficient if custody, refining, insurance, administration and redemption costs are excessive.

The Government should instead monitor:

Measure	What it tells policymakers
Gold mobilised	Scale of participation
Active depositors	Household and institutional adoption
Repeat deposits	Whether trust is sustained
Redemption time	Actual liquidity
Cost per gram mobilised	Economic efficiency
Custody losses and claims	Quality of infrastructure
Imports affected	Actual rather than assumed macro impact
Credit generated	Financial productivity
Institutional participation	Success beyond households
Tax compliance	Quality of formalisation
Customer satisfaction	Whether the product works for the owner

The GMS 2.0 decision test

Before introducing any feature, policymakers should ask:

Owner's question	Scheme's answer should be
Is my gold safe?	Independent custody, insurance and audit
Is it still legally mine?	Clear ownership and enforceable rights
What if gold prices rise?	Transparent participation in gold economics
Can I get it back?	Simple and predictable redemption
Will my jewellery be destroyed?	More than one eligible deposit route
What about old family gold?	Clear and limited tax and documentation safe harbour
What about temple gold?	Separate institutional governance framework
Who controls the record?	Central, secure and auditable registry
What if something goes wrong?	Clearly allocated liability and compensation
Can my gold support credit?	Regulated lending after the core infrastructure works

Casahuja Perspective

GMS 2.0 should not be designed as a Government scheme asking Indians to part with their gold. It should be designed as a financial system that makes Indians want to participate.

India's gold is already owned.

The challenge is to make ownership, security, liquidity and productivity work together.

For households:

Keep the economic value. Reduce the storage risk. Gain liquidity.

For temples and institutions:

Protect the asset. Preserve lawful ownership. Reduce physical-security exposure. Earn where legally permissible.

For the financial system:

Create verified collateral and a trusted gold ecosystem.

For the economy:

Turn a portion of dormant physical wealth into productive financial capital.

For Government:

Build the rails, trust and safeguards — rather than merely setting a target for tonnes of gold.

The first question should no longer be:

“Why won't Indians deposit their gold?”

The better question is:

“What would make an Indian willingly deposit gold today?”

If the answer is better returns, safe custody, easy redemption, tax certainty, preserved ownership and genuine convenience, GMS 2.0 has the ingredients for scale.

If the proposition remains simply “deposit your gold because the economy needs it,” mobilisation may remain limited.

India does not need to persuade its people to stop loving gold.

It needs to make their gold safer, more liquid and more productive — without making them feel that they have lost control of it.

That is the opportunity for Gold Monetisation Scheme 2.0

India Semiconductor Industry 2026: Where the Real Business Opportunity Lies Beyond Chip Manufacturing

 By CA surekha Ahuja

If you are looking for the next ₹5–20 crore business, do not start by asking how to manufacture a chip. Ask what the chip industry will be unable to operate without.

SEMICON India 2026 is being held in New Delhi from 17 to 19 September with the theme Silicon to Systems. Building the Ecosystem, covering manufacturing, equipment, materials, packaging, testing, supply chain, design and related technologies. Applied Materials has also announced plans for a US$5 billion investment in India over the next decade, including R&D and supply-chain expansion.

For Indian founders, the important opportunity may therefore sit around the chip, not inside the chip.

The question is:  What will semiconductor companies have to buy repeatedly, qualify carefully and replace reluctantly?

That is where a business can find a defensible entry point.

Who should enter?

Not everyone should start a semiconductor business.

Business profile	Logical starting point
Existing industrial business	Adapt an existing capability and enter the supply chain
Existing technical-services business	Upgrade into specialised semiconductor applications
Technology startup	Solve a narrow, high-value problem with limited CAPEX
New manufacturing startup	Validate customer and qualification before building capacity
Capital-rich promoter	Invest only after technology, customer and economics are validated

The key insight

Existing businesses may have the strongest starting position.

A company that already has engineering capability, machinery, technical manpower, quality systems and working capital may need an upgrade rather than an entirely new business.

A startup should generally look for a specific problem before looking for a factory.

Capital should be the last part of the equation, not the first.

Do not chase the chip. Chase the bottleneck.

The semiconductor ecosystem requires much more than fabs.

Look for dependencies around:

precision engineering | equipment maintenance | testing and calibration | specialised materials | clean-room services | water and utilities | packaging | logistics | automation | technical services

But need does not equal opportunity. A commercially usable opportunity requires:

customer + specification + qualification + recurring demand + economics.

The wrong model is: CAPEX → factory → product → customer

The disciplined model is: Customer problem → specification → prototype → qualification → order → CAPEX → scale

Customer first. Machinery later.

The existing-business advantage

An existing business should ask one question: What do we already know how to make, test, maintain or manage that the semiconductor ecosystem will increasingly need?

For example, an existing precision-engineering company may explore semiconductor-grade components. 

A testing company may develop specialised testing or calibration capability.

An automation company may adapt its systems to semiconductor manufacturing.

A water-treatment company may explore specialised process-water requirements.

The objective is not to abandon the existing business.

It is to move one step closer to a high-value supply chain.

That can mean lower incremental CAPEX, existing manpower, established systems and potentially faster commercial validation.

What about a startup?

A startup should not enter because the semiconductor market is large.

It should enter because it has found a specific problem worth paying to solve.

Potential areas include: inspection | testing | automation | monitoring | specialised software | equipment support | process technology | technical services

The startup sequence should be: Problem → prototype → pilot → customer validation → qualification → commercial order

Only then should substantial capital be committed.

What about a capital-intensive project?

This is where the greatest discipline is required.

Before committing substantial CAPEX, establish:

Who is the customer?

What exactly will they buy?

What is the specification?

How long is qualification?

What is the expected volume?

What is the payment cycle?

Can the facility serve another industry?

Then stress-test:

lower volume + delayed approval + higher CAPEX + slower collections + higher rejection + delayed incentive.

If the project fails under reasonable downside assumptions, the answer is not to increase the investment.

The Semiconductor Opportunity Test

Before investing, get six answers:

Test	Question
Customer	Who will pay?
Specification	What exactly must be supplied?
Qualification	What must be proven before approval?
Demand	Is there repeat demand?
Cash	Can the business survive the qualification and ramp-up period?
Second market	Can the capability serve another industry?

If these six answers are not clear, the business is not ready for major CAPEX.

The CA's role is before the investment

The financial review should cover:

GST — classification, ITC, exports and job work.

Customs — classification, valuation, duties and exemptions.

Income tax — depreciation, R&D expenditure, capitalisation and applicable incentives.

FEMA and transfer pricing — where foreign investment, technology, royalties or associated enterprises are involved.

And one discipline is critical: 

Prepare the project both with and without incentives.

An incentive can improve a viable project.

It should not be used to disguise an unviable one.

Get qualified before everyone starts looking

An existing business considering this opportunity should create a Semiconductor Vendor Readiness File:  capability | machinery | technical team | quality systems | testing | capacity | compliance | traceability | financial strength | safety | business continuity

The objective is simple: Do not wait for the procurement opportunity to appear before becoming vendor-ready.

Qualification can itself become the competitive barrier.

The five mistakes to avoid

1. Confusing an investment announcement with an order.

2. Using the total semiconductor market as your revenue projection.

3. Assuming the lowest price will win qualification.

4. Calculating profit without calculating working capital.

5. Building capacity before validating demand.

The most expensive mistake is usually the last one.

The 90-Day Founder Move

Days 1–15: Identify 10 supply-chain bottlenecks.

Days 16–30: Select three based on customer pain, qualification barrier, recurring demand and capital requirement.

Days 31–45: Meet actual customers and obtain specifications.

Days 46–60: Test technical and financial feasibility.

Days 61–75: Pilot and begin qualification.

Days 76–90: Make the investment decision.

Not before.

Final Words

The semiconductor opportunity should not be reduced to:

“India is building semiconductor capacity. Therefore, build a semiconductor factory.”

That is too simplistic.

For an existing business, the opportunity may be to upgrade an existing capability and become a qualified supplier.

For a startup, it may be to solve one specialised problem without carrying unnecessary CAPEX.

For a capital-intensive promoter, the sequence should be:

customer → technology → specification → qualification → economics → capital.

Not:  capital → factory → hope.

The real competitive question is: Which dependency will become more important as India's semiconductor ecosystem expands, and can your business become difficult to replace?

Find the bottleneck before the crowd finds it.

Validate the customer before buying the machinery.

Build the capability before scaling the capital.

The next semiconductor opportunity may not be the company making the chip.

It may be the company that the chipmaker cannot afford to operate without.

GST Refund SOP: A Practical Internal System to Build, Review and Defend Every Refund Claim

 By CA Surekha Ahuja

A company has ₹4.80 crore of eligible ITC on its books.

During the period, it has ₹18 crore of exports under LUT and ₹9 crore of domestic supplies, including supplies affected by an inverted-duty structure.

The CFO asks a simple question. How much GST refund can we claim?

The wrong place to start is Form RFD-01.

The right place is the accounts. Before a refund number is produced, the business needs to establish which ITC belongs to which activity, which statutory refund category applies, whether any ITC has already supported an earlier refund, how common ITC should be dealt with, whether the books and GST returns reconcile, and which statutory formula or documentary test applies.

RFD-01 should be the last step, not the first calculation.

GST refund is a control exercise, not just a filing

GST provides different refund mechanisms for different situations. Export refunds, inverted-duty refunds, deemed exports, SEZ supplies, excess cash balances, excess tax payments and consequential refunds do not all operate through one common formula.

What can be common is the internal control system used to establish the amount claimed.

A practical refund process should therefore follow this sequence:

Build → Reconcile → Classify → Attribute → Review → Defend → File

In detailed terms:

Transaction → eligibility → statutory refund category → ITC classification → reconciliation → attribution → previous refund check → applicable formula or documentary test → review → RFD-01

This changes the central question from:

How much ITC do we have?

to:

How much of that ITC is legally available for this particular refund claim?

Refund category and ITC classification are different

These two concepts should not be mixed.

Question	What it establishes
Why is refund legally available?	Statutory refund category
Which inward credits are involved?	ITC classification
Is the credit eligible?	Eligibility
To which outward activity does it relate?	Attribution
Has it already supported a refund?	Refund consumption
Which computation applies?	Statutory formula or test

For internal control, a business can use four simple codes.

Z — Direct zero-rated ITC

ITC directly attributable to qualifying zero-rated supplies.

I — Direct inverted-duty ITC

ITC attributable to qualifying inverted-duty supplies.

C — Common ITC

ITC supporting more than one business activity and requiring appropriate statutory treatment.

X — Excluded ITC

Blocked, reversed, ineligible or otherwise outside the relevant refund computation.

These are internal accounting and review codes, not additional legal refund categories.

Build the Refund Master before preparing the claim

For a substantial refund, maintain one Refund Master for the relevant period.

The objective is to create one continuous trail:

Purchase invoice → supplier GSTIN → GSTR-2B → books → GSTR-3B → ITC classification → outward activity → refund category → earlier refund usage → current claim

A practical Refund Master can contain:

Control field	Purpose
Supplier GSTIN	Supplier identification
Invoice number and date	Primary audit trail
Taxable value	Reconciliation
Tax amount	ITC reconciliation
GSTR-2B period	Portal evidence
ITC booked	Books reconciliation
ITC availed	GSTR-3B reconciliation
Reversal	Net eligible credit
Z / I / C / X	Internal classification
Earlier refund usage	Consumption control
Current eligible amount	Claim computation
Supporting evidence	Review trail

The Refund Master should not be confused with the prescribed refund statement or filing utility.

The statutory statement supports the application. The Refund Master controls how the application was built.

The ITC bridge is reconciliation, not a new refund formula

Consider this position:

Particulars	Amount
Eligible ITC available for analysis	₹4.80 crore
Less. ITC utilised	₹1.20 crore
Less. ITC already consumed in earlier refund	₹0.80 crore
Less. Other applicable exclusions	₹0.20 crore
Residual ITC requiring analysis	₹2.60 crore

The ₹2.60 crore is not automatically refundable.

It is only the residual amount requiring further classification and statutory analysis.

Suppose the internal mapping gives:

Classification	Amount
Direct zero-rated	₹1.15 crore
Direct inverted-duty	₹0.90 crore
Common	₹0.55 crore
Total	₹2.60 crore

The next question is not:

Can we claim ₹2.60 crore?

It is:

What portion, if any, becomes refundable under the statutory mechanism applicable to each category?

That is the difference between an ITC reconciliation and a refund computation.

Keep export and inverted-duty refund tracks separate

Both situations may involve accumulated ITC.

That does not make their refund calculations interchangeable.

Export without payment of IGST

For qualifying zero-rated supplies without payment of integrated tax, Rule 89(4) provides the prescribed formula involving zero-rated turnover, Net ITC and adjusted total turnover. The statutory formula determines the admissible amount; the total ITC appearing in the books does not automatically become the refund.

The internal working should separately establish:

• qualifying zero-rated turnover
• adjusted total turnover
• eligible Net ITC
• relevant period
• common ITC treatment
• earlier refund consumption
• reconciliation with books and returns

Thus, ₹4.80 crore of eligible ITC and ₹18 crore of exports do not by themselves establish a ₹4.80 crore refund.

Export with payment of IGST

This is a different route.

For exported goods, the refund of IGST paid is linked to the prescribed customs and GST reporting mechanism. For export of services, the relevant export conditions and supporting evidence must be established through the applicable refund process.

The internal system should therefore maintain a separate IGST-paid export register, rather than mixing it with the LUT refund computation.

Supplies to SEZ

Supplies qualifying as zero-rated supplies to an SEZ unit or developer require the prescribed evidence of receipt or admission for authorised operations.

The SEZ register should therefore separately capture:

Invoice → SEZ recipient → authorised operations → prescribed endorsement/evidence → return → refund

Deemed exports

Deemed exports are a separate statutory category and should not be treated as ordinary zero-rated exports.

The claim must follow the applicable conditions and documentation. Depending on the prescribed framework, the supplier or recipient may be entitled to claim, but the same supply should not generate a dual benefit.

Inverted-duty refund requires a separate working

Inverted-duty refund operates under the Rule 89(5) framework.

The working should establish:

• qualifying outward supplies
• applicable input and output rate structure
• eligible inputs
• excluded or reversed credits
• period-specific rate changes
• relevant turnover
• output tax
• ITC already used or refunded elsewhere

The Rule 89(5) computation must be applied for the relevant period.

Input services and capital goods should not simply be inserted into the Rule 89(5) Net ITC calculation because they are otherwise eligible credits in the books.

The Supreme Court decision in Union of India v. VKC Footsteps India Pvt. Ltd. forms part of the judicial background to the inverted-duty refund framework. Subsequent amendments and notifications also make period-specific testing important.

Do not run an inverted-duty claim through the export refund working merely because both claims involve accumulated ITC.

One Refund Consumption Register for every category

This is one of the most useful controls for businesses making repeated refund claims.

Instead of maintaining separate records for export, inverted duty and other refunds, maintain one consolidated Refund Consumption Register.

For example:

Period	Category	ITC considered	Refund sanctioned	ITC consumed	Reference
Q1	Export / LUT	₹1.10 cr	₹0.75 cr	₹0.75 cr	ARN 01
Q2	Inverted duty	₹0.90 cr	₹0.42 cr	₹0.42 cr	ARN 02
Q3	Export / LUT	₹1.15 cr	Under review	To be determined	Current

This prevents a common problem.

The export team may prepare one ITC working.

The GST team may prepare another.

Accounts may prepare a third.

Each spreadsheet may look correct individually while the same ITC is inadvertently considered more than once.

One ITC Master. One Refund Consumption Register. Separate statutory computation tracks.

Earlier refund does not automatically settle the next refund

The Madras High Court in VSM Weavess India Pvt. Ltd. v. Assistant Commissioner (ST) considered the relationship between an earlier zero-rated refund and a subsequent inverted-duty claim.

The Court did not treat the earlier refund, by itself, as automatically extinguishing the subsequent claim. The taxpayer was required to substantiate the ITC attributable to the inverted-duty supplies.

The practical lesson is more important than the litigation.

Do not merely preserve the earlier refund sanction order. Preserve the underlying computation showing which ITC was consumed in that refund.

That creates the evidence required for the next claim.

Common ITC is where weak refund workings become vulnerable

Suppose a manufacturer has:

• exports under LUT
• domestic inverted-duty supplies
• ordinary domestic taxable supplies

Common expenditure may include rent, electricity, software, professional services and other business costs.

It is not enough to say:

₹55 lakh is common ITC, so allocate 50 percent to exports and 50 percent to inverted duty.

A convenient percentage is not automatically a defensible attribution methodology.

The applicable statutory mechanism should first be identified. The underlying business data should then support the working.

Common ITC is a reconciliation problem before it becomes a formula problem.

All refund categories need the same control discipline

The calculation may differ, but the control questions remain similar.

Refund situation	Primary control question	Principal evidence
Export without IGST	Does the supply qualify and is the export/LUT trail established?	LUT, invoices, returns and export evidence
Export with IGST	Was IGST actually paid and is the export correctly linked?	Tax invoice, returns and customs/export data
SEZ supply	Is the supply eligible and supported by prescribed SEZ evidence?	Invoice and endorsement/admission evidence
Inverted duty	Does the supply satisfy the statutory test and applicable Rule 89(5) computation?	Rate mapping, purchase data and returns
Deemed export	Does the supply satisfy the notified conditions?	Prescribed evidence and undertakings
Excess cash balance	Is the balance genuinely refundable?	Electronic cash ledger and return reconciliation
Excess tax payment	What caused the excess and what is the appropriate correction/refund route?	Books, returns and payment records
Order or appeal-related refund	What order or statutory payment created the entitlement?	Order, appeal record and payment evidence
Specified persons	Does the claimant and supply fall within the notified Section 55 framework?	Eligibility and prescribed documents
Specified unregistered-person cases	Does the transaction fall within the notified refund mechanism?	Agreement, invoices, supplier certificate and prescribed evidence

The purpose of this matrix is not to replace the detailed law governing each category.

It is to ensure that the correct legal route is identified before the calculation begins.

The reverse audit test

Before filing a substantial claim, start with the final refund number and work backwards.

Ask:

Where did this number come from?

Then trace:

Refund figure → statutory formula or test → ITC pool → GSTR-3B → GSTR-2B → purchase invoice → supplier → underlying business transaction

For export turnover:

Refund figure → export computation → zero-rated turnover → invoice → export evidence → GSTR-1 → books

For inverted duty:

Refund figure → Rule 89(5) working → eligible inputs → purchase invoice → tax rate → outward supply → GSTR-1 → GSTR-3B

If the chain breaks, the claim is not ready.

Three registers are better than one spreadsheet

For substantial claims, maintain three linked records.

ITC Register

What credit arose?

Refund Consumption Register

What credit has already been used for a refund?

Evidence Register

What document supports the current claim?

Together they answer three different questions:

Is the credit real?

Has it already been used?

Can the claim be proved?

That is a stronger control framework than maintaining only the final refund calculation.

Stop the claim if these red flags appear

Red flag	Risk
ITC differs between books and GSTR-3B	Unstable claim base
GSTR-2B differences remain unexplained	Documentary weakness
Same ITC appears in two refund workings	Double-counting risk
Common ITC is allocated without a documented basis	Attribution challenge
Export turnover differs between books and returns	Formula risk
Earlier refund consumption is unidentified	ITC availability cannot be demonstrated
Inverted-duty working includes inappropriate credit categories	Rule 89(5) computation risk
Period includes rate changes without separate analysis	Period-specific computation risk
Refund category was selected before transaction analysis	Wrong legal route
RFD-01 differs from the approved internal working	Filing control failure

Four reviews before filing

A substantial refund claim should pass four separate reviews.

Books review

Does the claim reconcile with the accounting records?

Returns review

Does it reconcile with GSTR-1, GSTR-3B and GSTR-2B?

Legal review

Is the correct statutory category, formula and relevant-period rule being applied?

Evidence review

Can the claim be understood and supported from the documents without reconstructing the taxpayer's entire business?

Only after these reviews should the application be filed.

The GST Refund SOP

The entire process can be reduced to ten steps:

1. Identify the transaction

What actually happened?

2. Identify the statutory refund category

Why is the amount refundable?

3. Determine the relevant period and limitation

Which statutory clock applies?

4. Reconcile the data

Books, returns, ledgers and supporting records.

5. Establish eligible ITC or refundable tax

Remove what the applicable law excludes.

6. Classify the ITC

Z, I, C or X for internal control purposes.

7. Check previous refund consumption

Identify ITC already used in earlier claims.

8. Apply the correct statutory test

Use the formula and documentary requirements applicable to that category and period.

9. Conduct the reverse audit

Trace the final number back to the underlying transaction.

10. File the refund application

RFD-01 should record a number that has already been independently established.

What changes when refund becomes a system

Conventional approach	Controlled approach
Start with RFD-01	Start with transactions
Calculate total ITC	Establish eligible ITC
Select a refund category	Determine category from facts
Prepare separate spreadsheets	Maintain one Refund Master
Ignore earlier claims	Track refund consumption
Allocate common ITC casually	Document attribution
Reconcile after mismatch	Reconcile before filing
Defend the final number	Build the evidence trail first

Conclusion

A GST refund should not be viewed simply as:

ITC available → formula → RFD-01

It should be viewed as:

Transaction → legal entitlement → eligible tax or ITC → classification → reconciliation → attribution → previous consumption → statutory computation → evidence → refund

For a business having exports, SEZ supplies, inverted-duty supplies or multiple refund situations, the stronger system is:  One ITC Master. One Refund Consumption Register. Separate statutory computation tracks. One final independent review.

The professional rule is simple: Build. Reconcile. Classify. Attribute. Review. Defend. File.

The strongest GST refund claim is not the one that produces the largest number on a spreadsheet.

It is the one where every rupee claimed can be traced, explained and defended.

The first half establishes what and why before touching numbers. The second half is where classification, prior-refund checks, and the reverse audit sit.

            One more worth showing: the document's point that a single spreadsheet is weaker than three                 linked registers, each answering a different question