Economic efficiency requires that taxes fall on final household consumption, never on intermediate inputs or capital formation. GST is an accrual-based flow tax requiring symmetric treatment of positive and negative net liabilities. Each tax period, liability equals tax on outward supplies less ITC on legitimate business inputs, including goods, services and capital goods. When output tax exceeds ITC, the difference is payable to the government; when ITC exceeds output tax, the excess is paid by the government.
This is the gold standard of a consumption-type VAT. ITC or its refund is not a privilege granted to certain groups: it is the essence through which we graduate from the old Indian excise tax to GST with all the consequential economic gains. Capital-goods ITC must also be immediately available: delayed recovery imposes financing costs on investment and undermines neutrality. A credit not promptly recovered is worth less than its nominal amount, and it is tantamount to not getting ITC correctly.
At its October 8 meeting, the GST Council recommended extending inverted-duty refunds under Section 54(3) to input services from November 1, 2026, and including capital-goods ITC in refunds for zero-rated and qualifying inverted-duty supplies from April 1, 2027, subject to apportionment over 60 months. The apportionment applies to refunds, not to credits usable immediately against output tax. It also proposed removing selected restrictions under Section 17(5) and accelerating refunds through automated, risk-based processing. These decisions are in the correct direction and represent a part of the deficiencies identified in our paper.
Yet five-year apportionment introduces an income-type feature into a consumption-type VAT. Established businesses with sufficient output-tax liability (typically brownfield projects) can utilise capital-goods credits immediately, while newly established businesses with little initial output-tax liability (typically greenfield projects) depend on deferred cash refunds. The rule, therefore, discriminates against new enterprises undertaking identical investments. Like depreciation-based relief under an income-type VAT, deferred recovery reduces capital-goods credit in present-value terms, effectively taxing investment.
The GST Council has incrementally tried to address the refundability of ITC without changing the refund architecture. Cash refunds remain exceptions available in specified circumstances rather than the automatic consequence of negative net tax liability. Consider two businesses producing supplies taxed at 5 per cent and accumulating identical legitimate ITC. One qualifies for a refund because its material inputs attract a higher GST rate; the other does not because its excess credit arises mainly from services taxed at 18 per cent before outward supplies begin. The distinction is fully arbitrary when seen through the lens of economics. Under a consumption VAT, refund entitlement depends on excess legitimate credit, not the composition of inputs — goods, services or capital expenditure.
The 60-month rule also treats firms able to utilise credits immediately differently from those dependent on refunds, raising the latter’s effective investment costs. Selective relaxation of blocked-credit provisions is welcome but incomplete: restrictions on legitimate business expenditure, including factory civil works, still interrupt the credit chain and burden firms making substantial initial investments when output-tax liabilities are modest.
The October announcements represent parametric improvement within a conceptually confused framework. Refunds remain conditional while 60-month apportionment retains an income-type feature for businesses dependent on refunds. Unrecovered ITC raises production costs, discourages investment, and disadvantages domestic value addition relative to imported finished products, which cross the border free of domestic input-tax cascading. Faster processing cannot compensate for denying refunds of legitimate credits. The measure of reform is not the number of additional categories made refundable but whether tax on business inputs and investment is eliminated in economic terms.
We now have a decade of experience with GST and there is a lot of understanding of how it works. The reform priority now is to replace exception-based refunds with a comprehensive, symmetric mechanism. All legitimate business-input taxes should qualify for immediate credit, with positive net liabilities paid to the government and negative liabilities promptly refunded, subject to anti-abuse safeguards.
This requires the GST architecture proposed by us in the Thirteenth Finance Commission Task Force architecture: a broader consumption base, fewer exemptions, and a low uniform rate around 12 per cent to reduce classification disputes and manipulation. Much of the immediate fiscal burden from refunds would be transitional rather than permanent. Previously blocked credits, however, carry a separate recurring fiscal cost. Eliminating taxes on production and investment would improve resource allocation, stimulate investment, and expand the tax base. Transitional costs must be managed, and not used to justify permanent economic distortions.
A comprehensive GST with a single 12 per cent rate, including gold consumption but excluding investment bullion, could generate sufficient revenue to finance the transition and produce a fiscal surplus thereafter. A lower uniform rate would also weaken the justification for exemptions, expanding the tax base and strengthening revenue productivity.
States ultimately receive 70.5 per cent of GST — 50 per cent directly as state GST and 20.5 per cent via the 41 per cent devolution of central GST. They also retain substantial taxing powers over alcohol, petroleum products, motor vehicles and property. Reforms of these taxes could offset transitional GST losses. Odisha, Karnataka, Tamil Nadu and Maharashtra are working on revenue reforms, demonstrating the path to strengthen state finances without perpetuating GST distortions.
The writers are, respectively, a researcher at the XKDR Forum, and affiliated with the Pune International Centre and XKDR Forum.
 "GST")