GST without a seamless input tax credit system fails its core principles

We are losing touch with the soul of the indirect tax

Indirect taxation, when designed poorly, is an enemy of economic growth. We were part of the early dreams of cleaning this up by going to the “value-added tax” (VAT), an internationally established idea, which was rebranded for India by us as goods and services tax (GST). This is intended to be a clean consumption tax, which is neutral to the methods of production or international trade. It is the right way to do indirect taxation in any country. 

The engine of GST is input tax credit (ITC). Comprehensive ITC eliminates the cascading of taxes, ensures that tax is levied only on the value added at each stage of production and therefore (ultimately) only on consumption, and creates a unified national market. 

Today’s GST, however, repeatedly violates this core principle. The system has become an example of what Lant Pritchett calls “isomorphic mimicry”: It has adopted the external form of a modern GST to seek the legitimacy that comes with the name, but without the underlying function. The ITC mechanism has been increasingly obstructed by legal and procedural blockages. In many respects, the Indian GST has become a tax on production, similar to the previous system of excise and service tax.

  A functional GST is built on a seamless ITC system. When a business pays tax on its inputs — materials, services, or capital goods — it should receive a full and immediate credit. This ensures the tax is a pass-through. When ITC is blocked, the economic logic is undermined. The tax ceases to be a levy on consumption and becomes a tax on business inputs, embedding itself as a cost that cascades through the supply chain. The effective level of taxation can become very high. 

One obstruction came up in the treatment of refunds under an “inverted duty structure”, which arises when inputs are taxed at a higher rate than the final output. In this situation, excess credit should be refunded (as is done with zero-rating of exports). But, under Rule 89(5) of the Central GST Rules, refunds are restricted to the tax paid on input goods only. The ITC accumulated on input services and capital goods is not refunded. 

The divergence between the nominal GST rate and the effective tax burden shows the impact. For a business in the 5 per cent slab, the blocked ITC can mean the tax incidence is higher than the nominal rate. A supplier of goods might face an effective rate of 14 per cent, while a service provider’s burden can increase to 28.4 per cent. In this form, GST is not a modern consumption tax; it is an old-style tax on production. 

The departure from a true VAT is not a theoretical problem; it inflicts broad, systemic damage across the economy, penalising our most dynamic sectors. 

This rule disproportionately affects the services sector. A manufacturer can claim refunds on its raw materials but must still accumulate ITC for services and capital goods. In contrast, a software or logistics firm, even though treated equitably in law, does not receive the same relief since their inputs are primarily other services and capital equipment. As a result, costs remain elevated in both manufacturing and services, but the burden is heavier in services where refunds are effectively most restricted.

  The broken credit chain has a bias in favour of imports. A domestic product may have a nominal 5 per cent GST, but its final price is inflated by uncredited taxes on its inputs, pushing the effective tax burden towards 14 per cent or higher. An imported equivalent, however, is taxed cleanly at the border at the nominal rate. This puts domestic producers at a structural disadvantage. The disadvantage is sharper for services, where blocked credits on inputs and capital goods push the effective incidence even higher, further eroding competitiveness. 

GST has become a tax on investment. Credit for GST paid on capital goods is often deferred until a project is operational and generates its own tax liability. But a credit of ₹100 today is worth more than a credit of ₹100 in three years. This delay raises the cost of capital. 

Within the logic of the flawed system, the “GST 2.0” reforms of September 2025 have unfortunately made things worse. By merging the 12 per cent slab into the 5 per cent slab, the range of goods under cascading taxes went up, and high effective production taxation rates have emerged. 

Exporters under GST are not fully neutralised for embedded taxes, since refunds exclude ITC on capital goods. The international best practice is to zero-rate exports completely, so that we never tax foreigners. Our rules try to tax foreign buyers and hamper the competitiveness of Indian exporters. 

The ITC blockage weighs far more heavily on micro, small, and medium enterprises (MSMEs). Larger firms can manage refunds, restructure supply chains, or absorb financing costs. MSMEs, operating on thin margins and limited access to credit, lack such flexibility. Their costs stay elevated, their competitiveness erodes, and they face a deeper import bias. The tax system, therefore, penalises smaller firms disproportionately. 

The traditional GST reform agenda has focused on two goals: Achieving a single rate with a comprehensive base and improving the treatment of imports and exports. A third priority must now be added: The restoration of a purist ITC mechanism, to create a genuine GST/VAT. 

The path forward requires a three-pronged approach. First, the law must be amended to guarantee flow-based, immediate, and full ITC across all inputs: Goods, services, and capital goods. Second, once the credit chain is fixed, the rate structure should be rationalised by merging the lower rates upwards into a single, revenue-neutral rate of 12-14 per cent. Finally, the administration of refunds requires improvement. Refunds are a persistent problem in VAT systems, but India has the technology to address this. A GST-CPC (Central Processing Centre), modelled on the successful Income-Tax CPC, could automate return processing and issue refunds based on risk-based checks. 

The authors are affiliated to XKDR Forum and the Pune International Centre