By R. Suryamurthy
Nine years after GST replaced a maze of central and state indirect taxes, the most revealing measure of its success may no longer be the monthly collection number. The harder question is whether India has built a tax system in which the government can distinguish between a taxpayer who is trying to cheat it and one who is simply trying to navigate a complicated set of rules. That is the question the GST Council should confront when it meets on October 7.
The transformation in revenue mobilisation is difficult to dispute. Gross GST collections touched about ₹12.46 lakh crore in April-September 2026, 11.6 percent higher than a year earlier, while September collections rose 14.7 percent to around ₹2.04 lakh crore. A tax that was once described as an uncertain experiment has become a dependable pillar of government finances.
But revenue buoyancy can conceal institutional weaknesses. A tax system is not successful merely because it collects more. It is successful when the cost of collecting that revenue is proportionate, when legitimate businesses can predict their liabilities, when disputes are resolved without excessive litigation and when enforcement resources are concentrated on deliberate evasion rather than dispersed across millions of routine transactions.
GST has not yet fully made that transition. Its first phase was necessarily defensive. The government had to build a common tax architecture across a vast economy while confronting widespread fears of revenue leakage. Technology became the principal instrument of control: registrations, invoices, returns, input tax credits and payments were increasingly brought within a digital framework.
The result was a tax administration with an extraordinary amount of information. The next phase should be about what the administration does with that information. The temptation will be to use more data to impose more scrutiny. The better objective would be to use better data to reduce scrutiny where risk is low.
That distinction is at the heart of GST 3.0. For a compliant business, the ideal tax system should become almost invisible. Transactions should generate the appropriate credit automatically, refunds should move without prolonged intervention, minor errors should be capable of correction and a genuine commercial disagreement should not immediately become a confrontation with the enforcement machinery.
Instead, many businesses continue to experience GST as a system in which compliance is accompanied by uncertainty over whether a technical mismatch, a supplier’s default or an interpretation of a provision could eventually produce a demand, penalty or prolonged dispute.
That uncertainty has an economic price. Consider input tax credit, the central mechanism through which GST is supposed to eliminate cascading taxation. In theory, tax paid on inputs should flow through the chain and ultimately attach to final consumption. In practice, credits can become stranded because of inverted duty structures, disputes over eligibility or problems elsewhere in the supply chain.
The result is a contradiction at the heart of a value-added tax: a system designed to prevent tax from becoming a cost can itself create costs through the inability to use the credit efficiently.
The answer cannot be to keep expanding exceptions indefinitely. It should be to restore the principle on which the credit mechanism was built. Where a legitimate tax credit exists, the taxpayer should either be able to use it against an appropriate liability or recover it within a clearly defined framework. If the law deliberately creates situations in which this is impossible, the policy should be explicit rather than allowing businesses to accumulate credits indefinitely.
This is particularly relevant after rate rationalisation. Lowering the GST rate on finished products may provide a direct benefit to consumers, but if the input structure continues to generate excess credit, part of the intended benefit remains trapped within corporate balance sheets. A rate reform that ignores the credit architecture is therefore incomplete.
But the bigger problem is not the existence of disputes. It is the way risk is distributed. GST increasingly makes one business’s compliance dependent on another business’s behaviour. A purchaser may maintain invoices, make payments through banking channels and comply with every requirement expected of it, only to find itself exposed because a supplier subsequently fails to discharge its tax obligations.
The state’s objective is understandable: a credit should not be available against fictitious or unpaid tax. Yet the enforcement mechanism cannot reasonably expect every purchaser to possess the investigative powers of the tax department.
If buyers are made responsible for detecting every possible supplier default, the tax system creates an unintended incentive to consolidate transactions among large, easily monitored companies while making smaller suppliers more difficult to engage. That could undermine precisely the formalisation of small businesses that GST was expected to encourage.
The solution is not to abandon verification. It is to make responsibility proportionate to knowledge and control. That principle becomes even more important when enforcement crosses into criminal law. The debate over arrest powers under Section 69 of the CGST Act should therefore not be reduced to the familiar argument that strong enforcement is necessary to combat fraud. Of course it is. Fake invoicing, organised tax evasion and deliberate manipulation of input credits require a forceful state response.
But the credibility of enforcement depends on knowing where to stop. A tax administration that possesses the power to arrest must exercise that power with a degree of institutional restraint greater than that required for issuing an ordinary tax notice. Criminal sanctions should attach to demonstrable criminal conduct, not become an additional weapon in resolving fundamentally fiscal or interpretational disputes.
The distinction matters because the power imbalance between the state and taxpayer is enormous. The possibility of arrest changes the character of an investigation even before guilt has been established. It can influence whether a business contests a demand, accepts an adverse interpretation or chooses settlement simply to eliminate the risk of prolonged exposure.
That may make the system appear efficient. It does not necessarily make it fair or economically efficient. India needs a tax administration in which enforcement becomes more selective as technology becomes more powerful.
The government now has access to information from e-invoices, returns, payments and other digital records on a scale that earlier tax administrations could not have imagined. That should make it possible to identify unusual patterns rather than treating entire categories of taxpayers as inherently risky.
The principle should be simple: the greater the evidence of deliberate fraud, the stronger the intervention; the lower the risk, the lower the administrative burden. Such a model would also change the treatment of MSMEs.
A large multinational with sophisticated tax advisers and an organised finance department cannot be treated in exactly the same manner as a small enterprise trying to understand an increasingly complex compliance system. This does not justify lower standards of honesty. It does justify proportionate administration.
A genuine error should be corrected. A repeated pattern of manipulation should be investigated. An organised fraud should be prosecuted. Treating all three as variations of the same problem is poor administration.
The Council should therefore resist the temptation to measure reform by the number of new rules, notices or enforcement actions. The real benchmark should be the amount of unnecessary friction removed from the system.
Can a low-risk business obtain registration quickly? Can a legitimate refund be processed without repeated intervention? Can an honest taxpayer correct a mistake without entering a prolonged dispute? Can input credits be reconciled without forcing businesses to become investigators of their suppliers? Can tax authorities identify serious fraud without using criminal powers as a routine extension of assessment? These are not demands for preferential treatment. They are demands for a functioning tax system.
There is also a broader fiscal argument for such reform. Money trapped in unresolved credits is money unavailable for business investment. Resources spent repeatedly responding to notices are resources diverted from productive activity. Professional fees generated by avoidable disputes are costs borne by the economy even though they do not appear in the government’s revenue accounts.
The state sees the revenue collected. Businesses see the cost of obtaining certainty. GST 3.0 must bring those two perspectives together. The Council has already demonstrated that it can undertake politically difficult rate reform. The next challenge is institutional rather than political. It must decide whether the tax administration will continue to evolve primarily as an instrument of surveillance and revenue protection or whether it can become a system that combines technological intelligence with administrative trust.
That does not mean trusting everyone. It means learning to identify whom not to trust. The distinction is crucial. A tax system that suspects every taxpayer eventually burdens the honest majority in its attempt to catch the dishonest minority. A system that uses data to identify risk can reverse that equation: fewer intrusive controls for compliant businesses and greater concentration of enforcement on those deliberately gaming the system.
That should be the real meaning of GST 3.0. The first nine years were about building the tax. The next phase should be about building confidence in it. India does not need a GST that merely collects more. It needs one that businesses can understand, finance ministers can rely on, courts are not overwhelmed by and honest taxpayers do not fear.
The ultimate measure of reform will not be another record collection month. It will be whether an honest business can pay the tax it owes, claim the credit it is entitled to and get on with its business without wondering whether a routine compliance problem could turn into a battle with the state. That is the institutional bargain GST now needs. (IPA Service)
