The problem the 101st Amendment was built to solve was not economic. Economists had agreed for decades that a destination-based value added tax across the whole supply chain was better than the layered system India actually had, and nobody serious defended the alternative. The problem was constitutional, and it was severe: under the division of taxing powers as originally drafted, neither the Union nor any state was capable of levying such a tax, and no combination of ordinary legislation by both of them could produce one either. The Union could tax the manufacture of goods but not their sale within a state. The states could tax the sale of goods but not services and not manufacture. Both levied taxes at different points on the same commercial transaction without being able to give credit for the other’s tax, and the Constitution gave neither the competence to fix it.

That is the tension inside this amendment, and it explains everything unusual about its design. To create a single tax on the supply of goods and services, the Constitution had to give two different levels of government power over the same subject at the same time, which the Seventh Schedule scheme does not otherwise do, and it had to create a body in which they would agree on the details, which the Constitution had never done for a tax before. The result is Article 246A, which is unlike any other taxing provision in the Constitution, and Article 279A, which created the Goods and Services Tax Council. Understanding those two articles and the voting rule inside the second one is the whole of what this amendment did, and it is where most published explanations go wrong.

This article covers the constitutional amendment and the institution it created. The tax statutes enacted under it, the levy mechanics, input tax credit, registration, returns and the rate structure belong to the article on India’s GST laws, and the wider fiscal federalism argument about what states gave up and what they received belongs to the article on GST and Centre-state fiscal relations. What follows is the constitutional layer underneath both.

What existed before, and why it could not be fixed by legislation

To see why an amendment was unavoidable, it helps to set out the pre-existing division of indirect taxing power precisely, because the boundaries were the source of the dysfunction.

The Union’s principal indirect taxes rested on three foundations. Excise duty was levied on the manufacture or production of goods in India under an entry in the Union List, so the taxable event was manufacture and the tax attached before the goods were ever sold. Customs duties covered imports and exports. Service tax had no entry of its own for most of its life and was levied under the residuary entry, which gave Parliament power over any matter not enumerated in the State List or the Concurrent List. A specific entry for taxes on services was inserted into the Union List by an earlier amendment, together with an article providing for the levy to be collected and appropriated by the Union and the states, but neither was ever brought into force, so service tax continued to be levied under the residuary power throughout. That is a small and instructive detail: a constitutional provision can sit in the text, fully enacted, and never operate, because commencement was left to notification and no notification came.

The states’ principal indirect tax was the tax on the sale or purchase of goods within the state, which after 2005 most states levied in value added form. Around it sat a cluster of smaller levies: tax on the entry of goods into a local area for consumption, use or sale, which in some states took the form of octroi collected at municipal boundaries; taxes on luxuries, on entertainments and amusements, on advertisements other than in newspapers and broadcasts; and purchase tax. Central sales tax on inter-state sales was levied under a central statute but collected and retained by the exporting state, and it was not creditable anywhere, so it stuck as a cost on every inter-state transaction.

Four structural failures followed from this division, and each of them was a constitutional failure rather than an administrative one.

The first was cascading. Because excise attached at manufacture and sales tax attached at sale, and because neither government could give credit against the other’s tax, the state tax was levied on a price that already included central tax, and central tax on inputs could not be set off against state tax on outputs. Tax was therefore levied on tax repeatedly along a chain, and the effective burden depended on how many times a good changed hands rather than on the value actually added.

The second was the classification war. When the Union may tax manufacture, a state may tax sale of goods, and the Union may tax services, then every commercially ambiguous transaction becomes a jurisdictional dispute. Is a works contract a sale of materials or the provision of a service, and if it is both, how is the consideration split? Is packaged software goods or a service, and does the answer change if it is downloaded rather than delivered on a disc? Is supplying food in a restaurant a sale of food or a service of serving it? Decades of litigation went into these questions, and the answers were frequently unsatisfying because the transactions were single commercial events being forced into two constitutional boxes drawn for a mid-twentieth-century economy.

The third was the fragmentation of the national market. Entry tax and octroi created internal customs frontiers. Check posts at state borders existed to verify documents for taxes that were levied differently on each side. Central sales tax made an inter-state sale more expensive than an identical intra-state sale for reasons unrelated to cost. Business structures, warehouse locations and distribution networks were designed around tax boundaries rather than logistics, which is a pure deadweight cost.

The fourth was the impossibility of a legislative fix. Suppose the Union and every state had unanimously wished to create a single tax on supply covering goods and services with seamless credit. Parliament could not have enacted it, because it had no power to tax intra-state sales of goods. No state could have enacted it, because it had no power to tax services or manufacture. A central statute plus twenty-nine state statutes could not have achieved it either, because the credit chain requires that tax paid at one stage be set off against tax due at the next stage, and no legislature can direct another legislature’s tax to be credited against its own. The competence to build the mechanism did not exist anywhere. It had to be created.

Why did GST need a constitutional amendment at all?

Because the taxing powers required did not exist. The Union could not tax the sale of goods within a state and the states could not tax services or manufacture, and no combination of central and state legislation could create a credit chain across both. Only an amendment could confer power over the same subject on both levels simultaneously.

The long approach: why it took so many years

The passage record of the 101st Amendment is unusually long and unusually instructive, because almost every design feature of the final text can be traced to a specific negotiation.

The intellectual groundwork was laid by the reports of the task force on indirect taxes and on implementation of the fiscal responsibility legislation, chaired by Vijay Kelkar, which recommended a comprehensive goods and services tax covering the whole chain with credit at every stage. The institutional groundwork was laid by the Empowered Committee of State Finance Ministers, a body created to coordinate the introduction of state value added tax in the early 2000s. That committee had no constitutional status whatever. It was an informal forum of ministers, and it turned out to be the single most important institution in the entire GST story, because it demonstrated that finance ministers of states governed by opposing parties could agree on a common tax design and hold to it. The value added tax rollout of 2005 was its achievement, and the GST Council is, in institutional terms, that committee given constitutional form.

A target date for introducing GST was announced in a Union budget speech in the middle of the 2000s and was missed, as were several successors. A first discussion paper setting out a dual model, with a central and a state component levied on the same base, was published by the Empowered Committee at the end of that decade and framed the design that eventually prevailed.

The first legislative attempt was the Constitution (One Hundred and Fifteenth Amendment) Bill, introduced in 2011. It was referred to the departmentally related standing committee on finance, which reported on it, and it lapsed on the dissolution of the fifteenth Lok Sabha in 2014. Lapsing is the ordinary fate of a bill pending in the Lower House at dissolution, and it is one reason constitutional amendments that miss their political moment often have to be begun again from the start.

The second attempt was the Constitution (One Hundred and Twenty-second Amendment) Bill, introduced in the Lok Sabha in December 2014. The Lok Sabha passed it in May 2015. In the Rajya Sabha, where the government did not have a majority, it was referred to a select committee, which reported in July 2015, and it then stalled for over a year while the government and the principal opposition negotiated. The Rajya Sabha passed it with amendments on 3 August 2016, the Lok Sabha agreed to those amendments on 8 August 2016, more than half the state legislatures ratified it as Article 368 required for an amendment altering the distribution of taxing powers, and the President assented on 8 September 2016.

Why did half the states have to ratify this amendment?

Because it altered the distribution of legislative powers between the Union and the states and amended the Seventh Schedule, which brings it within the proviso to Article 368(2). An amendment of that kind requires ratification by the legislatures of not less than one-half of the states before it is presented for assent, in addition to the special majority in each House.

The ratification requirement, and the three routes by which the Constitution can be amended, are set out in the guide to the constitutional amendment process, and this amendment is the cleanest modern illustration of the proviso operating exactly as designed.

The negotiation that shaped the text

Three episodes in the passage record determined what the amendment finally said, and each is worth understanding because each explains a feature readers frequently misattribute.

The first is the additional one per cent tax on inter-state supply. The bill as introduced in 2014 provided for an additional tax of up to one per cent on the supply of goods in the course of inter-state trade, to be levied by the Union for two years or longer as the Council recommended, and assigned to the state from which the supply originated. It was a concession to manufacturing states, which feared that a destination-based tax would move revenue from where goods are made to where they are consumed and would leave states with large industrial bases worse off.

Economically the provision was indefensible, because a non-creditable origin-based levy on inter-state supply reproduces in miniature exactly the distortion that central sales tax had created and that GST was designed to remove. The select committee and the opposition both pressed for its removal, and it was dropped in the amendments the Rajya Sabha passed in August 2016. The episode is the most instructive single item in the passage record, because it shows a genuine conflict between the economics of the reform and the fiscal interests of a subset of states being resolved in favour of the economics, and it shows what the price of that resolution was: a stronger compensation guarantee.

The second is the compensation guarantee itself. The bill as introduced left compensation to states for revenue loss as something Parliament may provide for, up to five years, in amounts and on a basis the Council recommended. The final text made it mandatory: Parliament shall, by law, on the recommendation of the Council, provide for compensation to the states for loss of revenue arising on account of implementation of the goods and services tax for a period of five years. The change from a permission to a command is small on the page and large in effect, and it is why the compensation legislation that followed was not optional.

Notice where that guarantee sits. It is section 18 of the amending Act, not an article of the Constitution. It is a binding statutory direction to Parliament contained in a constitutional amendment, which is an unusual instrument, and its location matters for a reason discussed later: a guarantee written into the Constitution itself would have been considerably harder to let lapse.

The third is what the opposition asked for and did not get. Two demands were pressed hard and refused. One was a cap on the GST rate written into the Constitution, so that the rate could not be raised beyond a stated ceiling without a further amendment. The government resisted on the ground that a constitutional rate cap would make the tax impossible to administer and would require an amendment every time a rate needed to move. The other was a binding independent dispute resolution mechanism between the Union and the states, rather than leaving the Council to decide its own modalities. The final text contains a provision permitting the Council to decide about the modalities for resolving disputes arising out of its recommendations, which is a permission addressed to the body whose decisions would be disputed, and no standing mechanism has been established under it.

Both refusals are defensible and both have consequences. The absence of a rate cap is the reason rate changes are an administrative matter rather than a constitutional one. The absence of an operative dispute mechanism is the reason disputes between the Union and states over compensation and over Council decisions have gone to the political arena and to the courts rather than to a forum designed for them.

How the amendment is structured

The Constitution (One Hundred and First Amendment) Act, 2016 does five distinct things, and separating them is the fastest way to understand the whole instrument.

It confers a new taxing power on both levels of government simultaneously, through Article 246A. It creates a special regime for supplies in the course of inter-state trade, through Article 269A. It defines the tax and the term “services”, through insertions in Article 366. It constitutes the Goods and Services Tax Council, through Article 279A. And it clears the ground by amending or omitting every existing provision that would otherwise have obstructed the new scheme, which means consequential changes to a string of articles and a substantial rewriting of entries in the Seventh Schedule.

Two further provisions sit in the amending Act rather than in the Constitution: the compensation obligation and a transitional clause. Both matter more than their placement suggests.

The provisions came into force in stages. The article constituting the Council was brought into force first, on 12 September 2016, so that the Council could be constituted within the sixty-day period the article itself prescribes, and the remaining provisions followed a few days later. That sequencing is not a technicality; it is the reason the constitutional clock discussed below began when it did.

Article 246A: the provision that has no parallel in the Constitution

Article 246A provides that, notwithstanding anything contained in articles 246 and 254, Parliament and, subject to the second clause, the legislature of every state have power to make laws with respect to goods and services tax imposed by the Union or by that state. The second clause provides that Parliament has exclusive power to make laws with respect to goods and services tax where the supply takes place in the course of inter-state trade or commerce. An explanation defers the application of the article to petroleum crude, high speed diesel, motor spirit, natural gas and aviation turbine fuel until the date the Council recommends.

Four features make this unlike anything else in the Constitution’s scheme of legislative competence.

The first is that it does not work through a List. Every other head of legislative power in the Constitution is allocated by an entry in the Union List, the State List or the Concurrent List, and Article 246 then tells you who may legislate on it. Article 246A allocates a subject directly, in the article itself, without placing it in any List. A reader looking for goods and services tax in the Seventh Schedule will not find it, which surprises people constantly.

The second is that the power is simultaneous rather than concurrent. On a Concurrent List subject, both Parliament and a state legislature may legislate on the same matter and Article 254 resolves conflicts by making the central law prevail over the repugnant state law. Article 246A opens with a non obstante clause excluding both Article 246 and Article 254, so the ordinary repugnancy mechanism does not apply. What exists instead is two parallel legislative powers over the same field, each operating in its own sphere: the Union levies central GST and the state levies state GST on the same supply, and neither law overrides the other because they are not, in the constitutional sense, in conflict.

The third is the exclusivity carve-out for inter-state supply. Where a supply is in the course of inter-state trade or commerce, only Parliament may legislate, which is what makes the integrated tax on inter-state supply a purely central levy even though its proceeds are shared.

The fourth is the deferred application to five petroleum products. Those goods are within the definition of goods and services tax and within Article 246A, but the article does not apply to them until the Council recommends a date. This is a materially different position from the exclusion of alcoholic liquor for human consumption, which is carved out of the definition of the tax altogether. The distinction is the single most consequential technical point in this whole area and is examined below.

What makes Article 246A different from every other taxing provision?

Article 246A confers power to make laws on goods and services tax on Parliament and on every state legislature at the same time, without routing that power through any List in the Seventh Schedule, and it expressly overrides both Article 246 and the repugnancy rule in Article 254. Parliament alone may legislate on inter-state supply.

Article 269A: the integrated tax and how its proceeds move

Article 269A provides that goods and services tax on supplies in the course of inter-state trade or commerce shall be levied and collected by the Government of India and apportioned between the Union and the states in the manner Parliament provides by law on the recommendations of the Council. An explanation deems supply in the course of import into India to be supply in the course of inter-state trade or commerce, which is how imports are brought into the same mechanism. A further clause provides that the amount apportioned to a state shall not form part of the Consolidated Fund of India, which is a drafting necessity: money that never enters the Consolidated Fund does not require an appropriation by Parliament to leave it. Another clause empowers Parliament to formulate principles for determining the place of supply and when a supply takes place in the course of inter-state trade or commerce, which is the constitutional basis for the place of supply rules that decide which state receives the tax on any given transaction.

The economic function of this article is to make a destination-based tax workable across state borders without check posts. A supply from one state to another attracts a single integrated levy collected centrally, the recipient claims credit for it against output tax in the destination state, and the tax is apportioned so that the destination state ends up with the state share. The credit chain crosses the state boundary because the tax that crosses it is a central tax.

How does the integrated levy reach the destination state?

Article 269A provides that GST on inter-state supply is levied and collected by the Union and apportioned between the Union and the states as Parliament provides by law on the Council’s recommendations, with imports deemed inter-state. The state’s apportioned share does not form part of the Consolidated Fund of India, so no separate appropriation is needed to release it.

Article 366: the two definitions that set the boundaries

The amendment inserted two definitions. Goods and services tax is defined as any tax on supply of goods, or services, or both, except taxes on the supply of alcoholic liquor for human consumption. Services is defined as anything other than goods.

Both definitions are doing heavy work. The definition of the tax fixes the taxable event as supply, which is a deliberate move away from manufacture, sale and provision of service, and it is the constitutional foundation for ending the classification wars described earlier. If the taxable event is supply, it does not matter whether a works contract is a sale or a service, because both are supplies.

The residual definition of services as anything other than goods is the widest possible formulation and is what allows the tax to reach transactions that were previously outside both the sales tax and the service tax net.

The exclusion of alcoholic liquor for human consumption from the definition is the crucial structural point. Because it sits in the definition rather than in a deferral clause, alcohol is outside goods and services tax as a constitutional matter, and no recommendation of the Council and no ordinary legislation can bring it in. Only a further constitutional amendment could. Petroleum products, by contrast, are inside the definition and inside Article 246A, and are merely deferred until the Council recommends a date, so bringing them in requires a Council recommendation and a notification, not an amendment.

That distinction is misstated in an enormous quantity of published material, which treats petroleum and alcohol as though they were excluded in the same way. They are not, and the practical difference is the difference between a decision the Council can take and a decision that would require ratification by half the state legislatures.

The consequential amendments

The rest of the amendment is clearing work, and although it is dull it is where a careful reader can see the scale of what was being displaced.

The residuary power in Article 248 was made subject to Article 246A, so that the new taxing power takes priority over Parliament’s power over unenumerated matters, which is where service tax had lived. The provisions enabling Parliament to legislate on State List subjects in the national interest on a Rajya Sabha resolution, and during a proclamation of Emergency, were extended to cover goods and services tax, so that the new power sits inside the ordinary emergency and national-interest machinery. Article 268 was trimmed by removing excise duties on medicinal and toilet preparations from the duties levied by the Union and collected by the states. Article 269 on inter-state sales was made subject to the new Article 269A. Article 270, which governs the distribution of central taxes, was amended so that central GST and the Union’s share of integrated GST enter the divisible pool shared with the states on the Finance Commission’s recommendations. Article 271, which allows the Union to levy a surcharge on any tax for its own purposes with the proceeds retained entirely by the Union, was made inapplicable to goods and services tax, which prevents the Union from taking back through a surcharge what the sharing arrangement gives to the states. Article 286, which restricts state taxation of supplies in the course of inter-state trade, import and export, was rewritten in the vocabulary of supply. And the article inserted by an earlier amendment to provide for service tax levied by the Union and collected by the Union and the states, which had never been brought into force, was omitted.

The Seventh Schedule restructuring

The entries carrying the subsumed taxes were rewritten rather than simply deleted, and the pattern of what survived is worth attention.

In the Union List, the excise entry was narrowed so that central excise survives only on petroleum crude, high speed diesel, motor spirit, natural gas, aviation turbine fuel and tobacco and tobacco products. Tobacco is therefore subject to goods and services tax and to central excise at the same time, which is deliberate and is the only category treated that way. The entry on taxes on the sale or purchase of newspapers and advertisements in them was omitted, as was the never-commenced entry on taxes on services.

In the State List, the entry on taxes on the entry of goods into a local area was omitted outright, which is what abolished entry tax and octroi. The sales tax entry was narrowed to sales of the five petroleum products and alcoholic liquor for human consumption, excluding inter-state and international sales. The entry on taxes on advertisements other than those in newspapers and broadcasts was omitted. And the entry on taxes on entertainments and amusements was narrowed so that it survives only to the extent such taxes are levied and collected by a panchayat, a municipality, a regional council or a district council.

That last item deserves a moment, because it connects two clusters of this series. The 101st Amendment did not abolish entertainment tax; it moved what remains of it into the exclusive fiscal space of local bodies. Whether local bodies actually receive and use that space depends on the state legislation and the devolution record examined in the article on the 73rd and 74th Amendments in practice. A tax entry reserved to local government is only as valuable as the local government it is reserved to.

The structure of the three Lists, how entries are read, and how competence is determined when entries appear to overlap are set out in the article on the Union, State and Concurrent Lists.

Article 279A: the Council, and the arithmetic that defines it

The Goods and Services Tax Council is the central creation of this amendment, and almost everything argued about GST federalism is an argument about its design.

Article 279A required the President to constitute the Council by order within sixty days of the commencement of the amending Act. The Union Cabinet approved its creation within days of the assent, the Council was notified, and it held its first meeting later that month. Its secretariat was established in New Delhi, with the Revenue Secretary as ex officio secretary and the head of the central indirect tax administration as a permanent non-voting invitee.

Composition

The Council consists of the Union Finance Minister as chairperson, the Union Minister of State in charge of revenue or finance, and the minister in charge of finance or taxation or any other minister nominated by each state government. The state members choose one among themselves as vice-chairperson for such period as they decide. Where a proclamation under Article 356 is in operation in a state, the member is a person nominated by the Governor.

Two features of this composition are worth noting because they are frequently glossed over. The Union has two members, not one, and the chair is held permanently by the Union Finance Minister rather than rotating. But the second Union member does not increase the Union’s voting weight, because weight is allocated to the Union as a whole rather than by head. And every state has one member regardless of population, revenue or economic size, so a state with a very large economy and a state with a very small one have identical voting weight.

Functions

The Council makes recommendations on the taxes, cesses and surcharges to be subsumed; the goods and services that may be subjected to or exempted from the tax; model laws, the principles of levy, the apportionment of integrated tax and the principles governing place of supply; the threshold turnover below which goods and services may be exempted; the rates including floor rates with bands; special rates for a specified period to raise additional resources during a natural calamity or disaster; special provision for a listed group of north-eastern and hill states; and any other matter the Council decides. It is also required to recommend the date on which the tax is to be levied on the five petroleum products. And it is directed to be guided by the need for a harmonised structure of the tax and for the development of a harmonised national market.

Two further clauses complete the picture. One permits the Council to decide about the modalities for resolving disputes arising out of its recommendations, between the Union and one or more states, between the Union and any state on one side and one or more other states on the other, or between two or more states. Another provides that the Council’s acts and proceedings are not invalid merely because of a vacancy, a defect in its constitution, a defect in the appointment of a member, or a procedural irregularity not going to the merits.

The voting rule

This is the provision that defines the institution, and it is the one most often misdescribed.

Half the total number of members constitutes the quorum. Every decision must be taken at a meeting by a majority of not less than three-fourths of the weighted votes of the members present and voting. The vote of the Union has a weight of one-third of the total votes cast in that meeting. The votes of all the state governments taken together have a weight of two-thirds of the total votes cast in that meeting.

Read that carefully, because the weighting is of votes cast in the meeting and not of the total membership. States that are absent or that abstain do not dilute the state bloc; their share redistributes among the states that are present and voting. That single drafting choice has a large effect on the arithmetic, and almost no published account of the Council mentions it.

The GST Council voting calculator

The table below works out what the rule actually requires in each situation a reader is likely to care about. Throughout, the state share is divided equally among the state members present and voting.

Situation	The arithmetic	Threshold in fractions	What it means in practice
Union in favour, how many states are needed	Union contributes one-third; the balance to reach three-fourths must come from the state pool of two-thirds	States supporting must be at least five-eighths of the states present and voting	With the Union in favour, a proposal passes only if at least sixty-two and a half per cent of participating states also support it
Union opposed, can states pass anything	Even with every state in favour, the state pool tops out at two-thirds of votes cast, which is below three-fourths	Unreachable	The Union holds an absolute veto that it can exercise alone, with no state support at all
States blocking a proposal the Union supports	The opposing or abstaining states must deny the proposal the five-eighths it needs	More than three-eighths of the states present and voting must withhold support	Any group amounting to more than thirty-seven and a half per cent of participating states can block, but no smaller group can
A single large state acting alone	One state’s weight is two-thirds divided by the number of states present and voting	Far below the blocking threshold in any realistic meeting	No individual state can block anything, whatever its size or revenue contribution
Absent or abstaining states	Weight is calculated on votes cast, not on total membership	The state pool remains two-thirds and is shared among fewer states	Absence strengthens the states who attend rather than weakening the state bloc, which is the opposite of the usual assumption
Quorum	One half of the total number of members	Half	A meeting can validly decide with half the members present, and the weighting then operates among those who vote

Three conclusions follow, and they are the corrections this article most wants a reader to carry away.

The Union does not have a majority in the Council. It has one-third of the weighted vote, which is a minority. What it has is a blocking third, because one-third exceeds the one-quarter needed to defeat a three-fourths requirement. The frequently repeated claim that the Union controls the Council through a majority is simply wrong, and it is wrong in a way that misdescribes the whole institution.

The states do not have a veto individually. They have one collectively, exercisable only by a coalition amounting to more than three-eighths of those participating. Building such a coalition requires states with different party affiliations, different revenue profiles and different consumption patterns to agree, which is precisely what the design was intended to make difficult.

Nothing can pass over the Union’s objection, and nothing can pass over the objection of a sufficiently large group of states. That is the double-veto design, and it is the single most important fact about the Council. Every argument about GST federalism, whether made by a state finance minister complaining of central dominance or by a central official pointing to the collective state weight, is ultimately an argument about the asymmetry between a veto one party can exercise alone and a veto the other party can exercise only by coalition.

What majority does a Council decision actually require?

Decisions require at least three-fourths of the weighted votes of members present and voting. The Union’s vote carries one-third of the votes cast and all states together carry two-thirds, shared equally among the states voting. The Union can therefore block alone, while states can block only as a group exceeding three-eighths of those participating.

Why a body with a blocking third has rarely used it

The arithmetic describes what the Council could do. What it has actually done is different, and the difference is the most interesting thing about it as an institution.

For several years after it was constituted, the Council took every decision by consensus. Its first formal vote came years into its life, on the rate applicable to lotteries, and even then the vote was recorded as an exception to a practice rather than as the beginning of a new one. Rate structures, exemption lists, threshold limits, the design of the composition scheme for small taxpayers, the treatment of specific sectors and a great many procedural decisions were settled in discussion and adopted without a division.

Three explanations for that practice are worth distinguishing, because they carry different implications.

The first is institutional inheritance. The Council grew directly out of the Empowered Committee of State Finance Ministers, which had no voting rule at all because it had no legal authority, and which therefore worked by consensus as its only available method. The people in the room in the Council’s early years were largely the people who had been in the room for value added tax, and they brought the working method with them.

The second is the fragility of the whole arrangement. A tax that requires the Union and every state to legislate in parallel on a common design collapses if participants begin defecting. A decision carried over the objection of a bloc of states would have set a precedent that the losing states could invoke later, and every participant had an interest in not establishing that precedent early.

The third is that the arithmetic itself pushes towards consensus. Because nothing can pass without the Union and without five-eighths of the participating states, the range of proposals that can pass at all is narrow, and a chair who wants decisions rather than deadlock will bring to the table only proposals already close to acceptable. A demanding voting rule does not usually produce frequent close votes; it produces pre-negotiated proposals and unanimous adoptions.

Whether that practice is a strength or a weakness is genuinely contested. Read one way, consensus is evidence of cooperative federalism working, with a permanent forum in which the Union and every state negotiate tax policy continuously rather than fighting it out through litigation and unilateral legislation. Read another way, consensus is what happens when the smaller party knows it cannot win a vote and therefore negotiates for the best available outcome inside a framework it did not choose, which is a description of accommodation rather than agreement. Both readings are supported by the record, and which one a reader prefers usually tracks their prior view of Indian fiscal federalism rather than any fact about the Council.

The compensation bargain

The compensation guarantee is the consideration the states received for surrendering taxing powers, and understanding it precisely is essential to understanding the federalism argument.

Section 18 of the amending Act provides that Parliament shall, by law, on the recommendation of the Council, provide for compensation to the states for loss of revenue arising on account of implementation of the goods and services tax for a period of five years. Parliament enacted the compensation legislation before the tax was introduced. That statute fixed a base year, guaranteed each state a protected rate of growth in its subsumed revenues from that base, provided for the shortfall to be computed at fixed intervals and paid, and created a cess levied on a small set of goods to fund a compensation fund from which the payments would be made.

Four features of this design determined everything that followed.

The guarantee was for a fixed period of five years running from the introduction of the tax, not indefinitely. Every state entered the arrangement knowing the date on which the protection would end.

The protected growth rate was fixed in the statute at a level substantially above what most states had been achieving, which made the guarantee generous in nominal terms and meant that the size of the required compensation would grow every year automatically, since the protected figure compounded from a fixed base while actual collections did not.

The funding source was ring-fenced. Compensation was payable out of a fund credited with the proceeds of a specific cess, not out of general revenues, which meant that when the cess yielded less than the guarantee required, there was no automatic mechanism to make up the difference.

And the guarantee was placed in the amending Act and in ordinary legislation rather than in the Constitution. A constitutional obligation could not have been altered without another amendment ratified by half the states. A statutory one can be altered by Parliament, and a five-year period written into a statute expires on schedule without anyone having to decide to end it.

The consequence became visible when collections fell short of the protected level by a wide margin, which happened both because economic growth was weaker than the protected rate assumed and because of an extraordinary contraction during the pandemic period. The compensation fund was insufficient, the Union initially took the position that the guarantee was payable from the fund rather than from its own resources, states argued that a guarantee that pays only when the ring-fenced source suffices is not a guarantee at all, and the eventual resolution involved borrowing arranged to meet the shortfall and repaid from the continued levy of the cess beyond the original five-year window. That episode is the strongest evidence available for the proposition that the states’ side of the bargain was weaker than it appeared at the time of ratification.

What exactly did the compensation guarantee promise?

For five years from the introduction of the tax. Section 18 of the amending Act obliged Parliament to legislate for compensation for loss of revenue arising on account of implementation of the tax for a period of five years, and the compensation statute fixed a protected growth rate on a base year and funded payments from a dedicated cess. The guarantee was time-limited by design.

The fiscal detail of the compensation mechanism, the arguments about the protected rate, and the position after the guarantee period belong to the article on GST and Centre-state fiscal relations. What matters at the constitutional level is the structure of the bargain: permanent surrender of taxing powers in exchange for a time-limited statutory guarantee and a permanent seat at a table where the state bloc can block but cannot pass.

The constitutional clock nobody talks about

One provision of the amending Act did more to force the pace of implementation than any political commitment, and it is almost never mentioned in accounts of the reform.

Section 19 provided that any provision of any law relating to tax on goods or services in force in a state immediately before the commencement of the amendment, which was inconsistent with the amended Constitution, would continue in force until amended or repealed, or until the expiry of one year from commencement, whichever was earlier.

Consider what that meant. The amendment stripped the states of their power to tax the sale of goods generally, abolished the entry tax entry, and narrowed several other entries. Existing state sales tax and entry tax statutes were, from the moment of commencement, inconsistent with the amended Constitution. Section 19 kept them alive for one year and no longer.

The commencement provisions were notified in September 2016. The one-year saving therefore expired in September 2017. If GST had not been in force by then, the states would have lost their principal source of tax revenue with nothing to replace it, because their old statutes would have ceased to operate and the new regime would not have begun. There was no constitutional mechanism to extend the period without a further amendment.

That is the reason the tax was introduced in the middle of a financial year, on 1 July 2017, rather than at the start of one. It is the reason the design work was compressed into roughly nine months of intensive Council meetings. And it is the reason a great many transitional and procedural questions were settled after the tax began rather than before, with the compliance difficulties that followed.

The provision is worth studying on its own account as a piece of legislative technique. A transitional saving with a hard expiry is the strongest commitment device available to a reformer, because it makes non-implementation more costly than implementation for the party that would otherwise delay. Compare it with the local government amendments examined elsewhere in this series, where a similar one-year conformity clause forced states to legislate but where the substance of what they legislated was left to discretion. A deadline compels action; it does not determine the content of the action. Here the content was determined by the Council and the deadline was determined by section 19, and the two together are why a reform that had failed to launch for over a decade launched within a year of the amendment.

What the amendment did not do

Several things widely believed to be part of the 101st Amendment are not in it, and a pillar article should be explicit about the boundaries.

It did not fix any rate. There is no rate, no cap and no band in the Constitution. Rates are recommended by the Council and given effect by notification under the tax statutes, which is why rate changes happen without any constitutional process.

It did not create the tax. The amendment confers the power; the levy is created by central and state legislation enacted under that power. A reader who says the Constitution imposes goods and services tax has confused competence with levy.

It did not subsume every indirect tax. Basic customs duty on imports sits outside, and the entries covering it were untouched. Stamp duty on instruments remains with the states. Electricity duty remains a state tax under an untouched entry. Property tax and other local levies remain. Excise on the five petroleum products and on tobacco remains with the Union, and state sales tax on those petroleum products and on alcohol remains with the states.

It did not abolish entertainment tax, although it is commonly said to have done so. It narrowed the entry so that what survives is confined to entertainment taxes levied and collected by local bodies.

It did not create a dispute resolution mechanism. It permitted the Council to decide about the modalities for one, which is a different thing, and no standing mechanism has been established under that permission. Disputes have therefore been resolved politically or through litigation.

It did not make Council recommendations binding, and it did not say they were not binding either. The text is silent, which is why the question had to be answered by a court.

Which taxes were subsumed and which survived?

Subsumed on the Union side were central excise on most goods, service tax, and the additional duties that stood in for them, along with central sales tax on inter-state sales. Subsumed on the state side were state value added tax and sales tax on most goods, entry tax and octroi, luxury tax, purchase tax, advertisement tax, and entertainment tax except the portion reserved to local bodies. Surviving are basic customs duty, excise and state sales tax on the five petroleum products, state sales tax on alcoholic liquor for human consumption, central excise on tobacco alongside GST, stamp duty, electricity duty, property tax, and the local entertainment tax.

The judicial treatment: what the Supreme Court actually held

The constitutional text is silent on whether the Council’s recommendations bind the governments that receive them. That silence was left deliberately, since the demand for a binding mechanism had been pressed during the passage of the bill and refused, and it produced the question that eventually reached the Supreme Court.

The occasion was not a federalism case at all. In Union of India v Mohit Minerals Pvt Ltd, decided on 19 May 2022 by a bench of three judges, the dispute concerned whether integrated tax could be levied on the importer of goods, on a reverse charge basis, in respect of ocean freight where the goods had been imported on terms under which the foreign supplier arranged and paid for the shipping. The Court held that the levy could not stand, because the importer was already liable to tax on a composite supply that included the transportation element, and a separate levy on the same transportation as a distinct supply of service amounted to taxing the same element twice within a scheme that treats a composite supply as a single supply.

That is the ratio, and it is a tax case. What made the judgment consequential for constitutional law was the reasoning the Court reached on the way there, because the Union had argued that the levy was traceable to a recommendation of the Council and that the recommendation was binding.

The Court rejected the premise. Reading Article 246A and Article 279A together, it held that neither article is subject to the other, that Article 246A confers simultaneous legislative power on Parliament and on state legislatures, and that the Council is a constitutional body created to make recommendations rather than a superior legislature whose output binds those who receive it. The recommendations, the Court held, have persuasive value. Where a statutory provision expressly makes a recommendation a precondition for the exercise of a delegated power, the recommendation operates within that statutory scheme; it does not operate as an independent source of obligation on the legislatures.

The Court drew support from the drafting history. An earlier version of the amendment had contemplated a dispute settlement authority whose decisions would have had a different character, and that mechanism was not carried into the enacted text. The Court also placed weight on the constitutional description of India’s federalism as involving both cooperation and contestation, and treated the Council as an institution of collaborative dialogue in which the Union and the states negotiate rather than one in which the Union directs.

Do the Council’s recommendations carry legal force?

No. The Supreme Court held in 2022 that the Council’s recommendations have persuasive value and do not bind either Parliament or the state legislatures, because Article 246A confers simultaneous legislative power on both and neither that article nor Article 279A is subordinate to the other.

Why that ruling changed less than the headlines suggested

The reporting of the judgment was, in places, close to the opposite of what it decided in practice. It was widely presented as freeing states to set their own goods and services tax rates and to depart from the Council’s design. Nothing of the kind followed, and the reasons are structural rather than political.

The first reason is the credit chain. Goods and services tax works because tax paid at each stage is creditable against tax due at the next. A state that unilaterally changed a rate, an exemption or a classification would break the correspondence between what its taxpayers charge and what taxpayers in other states can claim. The immediate loser would be businesses in the departing state, whose customers elsewhere would face unmatched credits and whose own inputs would carry credits that no longer align. A state can inflict this on itself; it cannot inflict it on anyone else.

The second reason is the structure of the statutes themselves. The central and state tax laws were drafted on a common model precisely so that the definitions, the taxable event, the place of supply rules, the return architecture and the classification system would be identical. A state departing from that model would have to rewrite a great deal more than the rate schedule, and would need administrative systems that no longer matched the shared technology platform on which returns, invoices and credits are processed.

The third reason is the integrated tax. Inter-state supply is exclusively within Parliament’s competence under Article 246A, and the apportionment of the integrated levy is governed by central law made on the Council’s recommendations. A state has no legislative purchase on the tax that governs its trade with every other state, which is a large share of any state’s commerce.

The fourth reason is political. A state that departed would be blamed by its own businesses for compliance chaos and would find its departure used as an argument for further centralisation. The strategic value of the judgment to states lies not in departure but in negotiation: a state that can credibly say the Council’s recommendation does not bind it has more leverage inside the room than one that cannot.

This is the general lesson worth extracting, and it applies far beyond tax. A formally non-binding recommendation can be effectively binding when the cost of departure falls on the party that departs. The Council’s authority does not rest on legal compulsion; it rests on the fact that the tax only works if everyone applies the same design, which makes unilateral variation self-harming. Institutional designers should notice that this is a more durable form of authority than a binding rule, because it does not depend on an enforcement mechanism and cannot be defied without immediate cost to the defier.

The judgment did change three things, and they should not be understated. It settled that a state legislature retains its own legislative competence and is not a delegate of the Council, which matters for how state amendments are drafted and defended. It removed the argument that a levy is valid merely because the Council recommended it, so a delegated levy must still be traceable to the statute and consistent with it. And it changed the atmosphere of Council meetings, because a state minister who disagrees can now assert a legal position rather than only a political one.

The federalism argument, stated fairly on both sides

This is the most argued-about aspect of the amendment, and it is a genuine constitutional disagreement rather than a factual dispute, so both cases deserve their strongest form.

The case that GST damaged state fiscal autonomy

States surrendered, permanently and irrevocably, their principal source of independent tax revenue. Before the amendment a state legislature could set its own sales tax rates, grant its own exemptions, design its own incentives to attract investment, and adjust its revenue effort in response to its own fiscal position. After the amendment it can do none of those things in any meaningful way, because the rate structure, the exemptions and the classification are settled in a forum where it holds a fraction of a vote.

What it received in exchange was time-limited. The compensation guarantee ran for five years and was funded from a ring-fenced cess, and when the cess proved insufficient the states discovered that the guarantee’s practical content depended on the Union’s willingness to fund it from elsewhere. A permanent surrender exchanged for a temporary guarantee is a poor bargain by any ordinary standard.

The residual taxing powers left to states are the least attractive ones. Petroleum products and alcohol remain state-taxable, which means states are fiscally dependent on precisely the two categories whose taxation carries the greatest social and political cost, and are structurally discouraged from supporting the inclusion of petroleum in GST even where the economic case for inclusion is strong.

The Council’s design compounds this. The Union has a veto it can exercise alone; the states have one they can exercise only in a coalition of more than three-eighths. Coalitions of that size across party lines are difficult to assemble and easy to fracture, and the Union has no equivalent coordination problem.

The case that states retained more than they surrendered

Every state gained the power to tax services, which no state had before, and services are the largest and fastest-growing part of the economy. Describing the amendment as a one-way surrender ignores that the states acquired a base that had been exclusively central.

The powers that were surrendered were, in practical terms, already constrained. Tax competition between states on sales tax rates was a genuine phenomenon and its effect was to bid rates down and to fragment the market without producing net investment gains. A state’s freedom to set its own rate was, in a competitive federation, a freedom to be undercut.

The states hold two-thirds of the weighted vote and every state holds an equal share of it regardless of size. A small state has the same weight as the largest. That is a substantial protection for smaller states that no other Indian fiscal institution provides, and it means that the state bloc’s collective weight exceeds the Union’s by a factor of two.

The collective veto has rarely been used because it has rarely been needed, not because it does not exist. A chair who knows that five-eighths of participating states must agree brings forward proposals that can attain that support, which is exactly how a supermajority requirement is supposed to operate.

And the Union surrendered something real as well. It gave up the ability to set central excise and service tax unilaterally, it is barred from levying a surcharge on goods and services tax under the amended Article 271, and it accepted that a substantial part of its indirect tax policy would be made in a forum where it cannot act alone.

What the record settles and what it does not

The record settles that the Union does not hold a majority in the Council, that no state can be outvoted individually, and that nothing passes without the Union. It settles that the compensation guarantee was time-limited and statutory rather than constitutional, and that it proved less robust than states expected when collections fell short. It settles that Council recommendations do not legally bind, and that no state has meaningfully departed from them.

It does not settle whether the exchange was a good one, because that depends on how a reader weighs a unified national market against a state’s ability to set its own tax policy, and reasonable people weigh those differently. It does not settle whether the compensation period should have been longer or should have been constitutionalised, because that is an argument about how much protection a federation owes its units when it asks them to surrender revenue. And it does not settle whether a dispute resolution mechanism should be established under the permissive clause, because the arguments about who would appoint it and what its decisions would bind are unresolved.

This article takes no position on those three questions. It takes a firm position on the factual claims that circulate around them, several of which are wrong.

How the Council actually works between the constitutional lines

The article creating the Council says nothing about how it prepares its business, and the working method that developed around it is a substantial part of the institution.

Meetings are convened by the chairperson and are preceded by the work of a committee of officers drawn from the central and state tax administrations, which examines proposals, prepares options and identifies where agreement is likely. Sectoral and thematic groups of ministers are constituted from time to time to examine particular questions, report back, and give the Council a worked position rather than an open question. The secretariat is staffed by officers on deputation from both the Union and the states, which matters because an institution whose staff come only from one side tends to produce agendas that reflect that side.

That apparatus is not in the Constitution and could be dismantled without amending anything. Its existence explains why the Council decides as much as it does: by the time a proposal reaches a meeting it has usually been through an officers’ committee, a group of ministers, and informal circulation, and the meeting itself confirms an outcome already negotiated. It also explains a recurring criticism, which is that a design settled by officers and ratified by ministers concentrates real influence in the central tax administration, which is larger, better resourced and permanently staffed compared with the tax departments of most states.

The counter-argument is that the alternative is worse. A body of thirty or more ministers meeting for a day cannot design a rate structure from first principles, and a forum without technical preparation would either decide badly or not decide at all. Every intergovernmental body of this kind concentrates influence in its secretariat, and the answer is to strengthen state capacity to engage with the technical work rather than to abolish the technical work.

The Council also has a documentary practice that is unusual for an Indian institution and worth noting. It publishes agendas and detailed minutes, so its deliberations are traceable in a way that few Indian executive bodies are. A researcher who wants to know why a rate was set at a particular level can usually find the argument recorded. That transparency is not required by Article 279A and could be withdrawn, but while it lasts it makes the Council one of the more scrutinisable institutions in Indian fiscal governance.

The petroleum question, resolved precisely

No single topic in this area produces more confused writing, so the position is worth setting out in full.

The five products are petroleum crude, high speed diesel, motor spirit, natural gas and aviation turbine fuel. They are within the definition of goods and services tax in Article 366(12A), because that definition excludes only alcoholic liquor for human consumption. They are within Article 246A, because the explanation to that article defers its application to them rather than excluding them. And Article 279A expressly requires the Council to recommend the date on which the tax is to be levied on them.

Three consequences follow. First, bringing petroleum products into goods and services tax requires no constitutional amendment. It requires a recommendation of the Council fixing a date, followed by the ordinary machinery under the tax statutes. Second, until that date, central excise and state sales tax on those products continue under the entries that were narrowed rather than deleted, which is why fuel prices carry a layered tax structure that the reform was supposed to eliminate. Third, the obligation on the Council to recommend a date is expressed as a requirement, but no time limit attaches to it, so the requirement has no operative force until the Council chooses to act.

The politics of that choice are straightforward and worth stating without editorialising. Taxes on the five products constitute a large share of both central and state revenue, they are collected efficiently, and the base is inelastic in the short run. Bringing them into goods and services tax would subject them to a rate structure decided collectively, would make the tax creditable to businesses that consume fuel, and would sharply reduce revenue at existing rate levels unless a substantial compensation cess were levied on top. Neither level of government has an incentive to move first, and the design ensures that neither has to.

Alcoholic liquor for human consumption sits in a completely different position. Because it is excluded from the definition of the tax itself, no recommendation of the Council can bring it in and no ordinary statute can. Only a constitutional amendment, requiring a special majority in each House and ratification by at least half the state legislatures, could. States that regard alcohol revenue as their last substantial independent tax base would have to vote for that amendment in their own legislatures, which makes the prospect remote.

Can petroleum be brought under GST without another amendment?

Yes. Petroleum crude, diesel, petrol, natural gas and aviation turbine fuel are inside the definition of the tax and inside Article 246A, with application merely deferred until the Council recommends a date. Alcoholic liquor for human consumption is excluded from the definition itself, so only a further constitutional amendment could bring it in.

The dispute mechanism that was permitted and never built

Article 279A contains a clause allowing the Council to decide about the modalities for resolving disputes arising out of its recommendations, covering disputes between the Union and one or more states, between the Union and some states on one side and other states on the other, and between states.

The clause is permissive, it is addressed to the body whose recommendations would be in dispute, and no standing mechanism has been established under it. The reason is not mysterious. Establishing a mechanism requires the Council to agree on who would constitute it, what its decisions would bind given that the Council’s own recommendations do not bind, and how a decision adverse to the Union would be enforced. Each of those questions divides the Council along the same lines that produce the disputes in the first place.

The consequence is that disputes have gone elsewhere. Compensation disagreements were argued politically and settled by negotiation and borrowing. Questions about the character of the Council’s recommendations were settled by the Supreme Court in ordinary tax litigation. Disagreements about specific rates and exemptions have been pursued through public statements by state finance ministers and through the Council itself.

Whether a mechanism should be built is contested. The case for it is that a federation which requires continuous intergovernmental agreement on a shared tax needs somewhere for disagreements to go that is neither the political arena nor a court deciding a private tax dispute. The case against it is that any mechanism would either bind, in which case it would create an authority above both Parliament and the state legislatures that the Constitution does not otherwise contemplate, or not bind, in which case it would add a stage without resolving anything. That objection is serious and has not been answered by the mechanism’s advocates.

There is a structural echo here that connects this article to the local government cluster. In both cases a constitutional provision uses a permissive verb to address the institution that would have to give something up in order to act, and in both cases the permission has gone unused for the whole life of the provision. Where a constitutional drafter wants something to happen, the verb has to command and a consequence has to attach; where the drafter writes a permission addressed to the reluctant party, the safest prediction is that nothing will happen.

The first year, and what a constitutional deadline does to design

The Council was constituted in September 2016 and the tax began on 1 July 2017, which gave it roughly nine months to settle a complete design for the largest indirect tax reform any federation has attempted. The compression is visible in the outcome, and it is worth understanding because it explains several features that are usually attributed to bad policy judgment rather than to time.

In that period the Council had to agree the model laws that the Union and every state would enact in parallel, the threshold turnover below which small suppliers would be outside the net, the composition scheme for taxpayers just above that threshold, the rate structure and the assignment of every category of goods and services to a rate, the exemption list, the treatment of the sectors that had been taxed under special regimes, the place of supply rules, the administrative division of taxpayers between central and state authorities, the transitional credit arrangements for stock held on the appointed day, and the return architecture.

Several of the choices made under that pressure were revisited afterwards, and the revisions tell you where the compression bit hardest. The return system as originally designed proved unworkable at scale and was replaced with a simplified interim arrangement that then became the operating system. Rate assignments for a large number of items were changed within the first two years as anomalies surfaced, particularly where a rate difference between an input and its output produced an accumulated credit that could not be used. Threshold and composition limits were revised upward. None of that is surprising for a reform of this size; what is notable is that the revisions happened through the same Council, at the same pace, without any of the participants leaving the arrangement.

The constitutional deadline deserves the credit and the blame for both halves of that record. Without section 19 the reform would probably have slipped again, as it had slipped repeatedly for over a decade. With section 19, a design that would have benefited from another year of preparation was launched with parts of it unfinished, and the finishing was done in public on live taxpayers.

The general lesson for anyone designing a constitutional transition is that a hard deadline is a powerful instrument and an indiscriminate one. It compels the reluctant, which is what it is for. It also compels the unready, and the drafter cannot choose which of the two is being compelled.

What the amendment means for a business operating across states

The constitutional analysis has a practical face, and it is worth stating because it is the clearest evidence of what changed.

Before the amendment, a business selling goods across state lines faced a different tax position in every state, a non-creditable central sales tax on inter-state movement, entry tax or octroi at some destinations, and the possibility that its product would be classified as goods in one state and as a service by the Union. Warehouse networks were designed to avoid inter-state sales rather than to minimise transport cost, which meant a distribution centre in every state whether or not the volumes justified one.

After the amendment, the taxable event is supply, the same definition applies everywhere because the statutes were drafted on a common model, tax on inter-state supply is a single central levy that the recipient can credit, and entry tax and octroi are gone. A business can locate a warehouse where the logistics make sense. The classification question that used to be constitutional is now merely a question of rate.

Three difficulties remain and are worth naming honestly. Registration is state-wise, so a business operating in many states holds many registrations and files many returns, which is a compliance burden the reform did not remove and in some respects increased. Place of supply rules determine which state receives the tax and generate disputes in sectors where the place of consumption is genuinely ambiguous. And because petroleum products remain outside, a business that consumes fuel heavily carries a non-creditable tax cost that breaks the credit chain at exactly the point where transport-intensive industries are affected most.

Anyone reasoning from the constitutional structure to a particular commercial position should note that the answer in any actual case depends on the statutes, the notifications and the facts rather than on the constitutional articles alone, and that the general framework described here does not substitute for advice on an individual matter.

Who and what the amendment binds

A pillar article should be explicit about scope, because a constitutional amendment binds differently from a statute.

It binds Parliament and every state legislature, by conferring a power and by restricting the powers they previously held. A state legislature cannot enact a sales tax on goods outside the two surviving categories, because the entry that supported such a tax no longer covers them. Parliament cannot levy a surcharge on goods and services tax for its own purposes, because the amended article excludes it.

It binds the President, by requiring the constitution of the Council within sixty days, and it binds the Union and the state executives in the sense that their tax administrations operate under statutes enacted within the constitutional limits it sets.

It does not bind a taxpayer directly. No person pays tax because of the 101st Amendment; people pay tax because of the central and state statutes enacted under it. A challenge to a levy is therefore usually a challenge to the statute or to a notification under it, with the constitutional provisions operating as the measure of competence rather than as the source of liability. Anyone dealing with an actual demand should note that the constitutional analysis in an article of this kind describes the framework and not the answer in a particular assessment, and that limitation periods for appeals under the tax statutes are short and strictly applied.

It binds the Council to a purpose. The requirement that the Council be guided by the need for a harmonised structure and a harmonised national market is a directive on the exercise of its functions, and although no case has turned on it, it is the kind of provision that can become operative when a recommendation is challenged as arbitrary.

What this amendment got right and wrong as institutional design

Read as a piece of constitutional engineering rather than as tax policy, the 101st Amendment is the most interesting Indian constitutional amendment of its generation, and its successes and failures are separable.

What it got right, first.

It solved a genuine competence problem with a minimum of constitutional disturbance. Article 246A is a single provision that creates simultaneous power without disturbing the architecture of the Seventh Schedule for any other subject. A cruder solution would have moved the subject to the Concurrent List, which would have imported the repugnancy rule and made the central law prevail, and states would never have ratified it.

It specified the decision rule in the constitutional text. The composition of the Council, the quorum, the weighting and the three-fourths threshold are all in Article 279A rather than in a statute or a rule, which means none of them can be altered without a further amendment ratified by half the states. This is the single most important protective feature the states obtained, and it is the feature most often overlooked by people describing the Council as centrally dominated.

It created a permanent forum rather than a one-off settlement. A tax that has to be adjusted continuously needed an institution that meets continuously, and the Council has met, deliberated and decided at a pace that no other Indian intergovernmental body approaches.

It attached a hard deadline through the transitional saving, which converted an aspiration that had been missed repeatedly into a reform that launched within a year.

What it got wrong, second.

It placed the compensation guarantee in the amending Act and in ordinary legislation rather than in the Constitution. Had the guarantee been an article of the Constitution, its duration and its funding would have been alterable only by a further amendment ratified by half the states. Placing it in a statute meant that when the guarantee proved expensive, the argument about whether it was payable from general revenues or only from a dedicated fund could be conducted as a question of statutory construction rather than of constitutional obligation. If the states were negotiating this bargain again, the location of the guarantee is the first thing they would change.

It left the dispute mechanism to a permission addressed to the disputing body, which was a predictable failure and was predicted at the time by those who had asked for a binding mechanism.

It left the character of the recommendations unstated, which produced six years of uncertainty resolved only by litigation. A single clause saying either that recommendations bind or that they do not would have saved a great deal of argument, and the omission was a deliberate deferral of a disagreement rather than an oversight.

And it left the petroleum decision to a requirement with no time limit, which is a requirement in name only.

The pattern across those four failures is consistent and is the analytical claim this article offers alongside the double-veto framing. Where the amendment specified a rule, the rule has held. Where it deferred a disagreement into a permission or an unenforced requirement, the disagreement has remained unresolved for the whole life of the provision. Constitutional drafting that postpones a conflict does not dissolve it; it relocates it into politics or litigation.

The errors that recur in published accounts

Six mistakes appear repeatedly, and a reader who can identify them will be able to judge a source quickly.

The claim that the Union has a majority in the Council is wrong. The Union has one-third of the weighted vote, which is a blocking minority, not a majority.

The claim that Council decisions are binding is wrong, and has been since the constitutional text was settled, whatever assumptions were made before the Supreme Court said so.

The claim that the amendment created the goods and services tax is wrong. It created the power; the tax was created by central and state statutes.

The claim that petroleum and alcohol are excluded in the same way is wrong, and it is the most consequential of the six, because it misstates what would be needed to change either position.

The claim that entertainment tax was abolished is wrong. The entry was narrowed to what local bodies levy and collect.

The claim that the 122nd Amendment Bill became the 101st Amendment Act by a simple renumbering is technically right and analytically misleading, because the enacted text differed from the introduced bill in two respects that mattered: the additional one per cent inter-state levy was dropped and the compensation obligation was strengthened from a permission to a command. Bills are numbered on introduction and Acts on enactment, and treating the two as the same document conceals the negotiation.

The honest assessment

The 101st Amendment did what it set out to do, and the thing it set out to do was narrower than the reform it enabled.

As a solution to a constitutional problem it is close to fully successful. The competence that did not exist now exists, it was created without disturbing the rest of the Seventh Schedule scheme, the classification disputes that consumed decades of litigation have very largely disappeared because the taxable event is now supply, the internal customs barriers created by entry tax and octroi are gone, and the credit chain crosses state boundaries. Those are structural achievements and they are irreversible without a further amendment.

As a piece of federal bargaining it is more ambiguous, and the ambiguity is genuine rather than a matter of perspective. States acquired the power to tax services, an equal vote regardless of size, and a collective blocking position entrenched in the constitutional text. They surrendered permanently the power to set their own rates on the largest part of their tax base, and received in exchange a guarantee that was time-limited, statutory, ring-fenced to a dedicated cess, and less robust in a downturn than it appeared in a boom. Whether that is a fair exchange depends on a judgment about the value of a unified national market against the value of fiscal autonomy, and this article does not pretend that the record settles it.

As an institution the Council has performed considerably better than the pessimistic predictions made when it was created. It has met regularly, decided a very large number of questions, published its deliberations, and operated for most of its life by consensus rather than by division. The predictions that it would deadlock, or that the Union would simply outvote the states, have both been falsified, the second one because the Union cannot outvote the states and the first because a demanding threshold produces negotiation rather than paralysis when the alternative to agreement is worse for everyone.

The verdict, stated as this series requires: the 101st Amendment is the most successful piece of constitutional engineering in modern Indian practice at solving the problem it addressed, and the least successful at protecting the party that gave up the most. The double-veto design is the reason for both halves of that sentence. It gave the Union a veto it can exercise alone, which prevents the states from imposing costs on the Union; it gave the states a veto they can exercise only collectively, which prevents any single state from obstructing; and it left the states’ compensation in an instrument the Union’s Parliament could allow to lapse on schedule. A reader who understands the arithmetic understands the politics.

Where to go next

This article owns the constitutional layer: the amendment, the articles it inserted, the Council, the voting rule and the judgment on the character of the recommendations. Three neighbours own the rest.

The levy itself, the structure of the central, state and integrated statutes, registration and thresholds, input tax credit, returns, the composition scheme, classification and rates belong to the article on India’s GST laws. The fiscal federalism analysis, the compensation arithmetic, the revenue positions of individual states and the argument about vertical and horizontal balance belong to the article on GST and Centre-state fiscal relations. The mechanics of the Lists and how competence is determined belong to the article on the Seventh Schedule.

For readers building their own material, the voting calculator is the object worth keeping, because it answers most Council questions without further reference, and you can keep a private annotated copy of the voting table and your case notes on VaultBook as you work through the underlying provisions. Aspirants should note that this amendment is examined on Article 246A, the voting weights and the binding-force question far more often than on the list of subsumed taxes, and can test that against previous-year questions on the ReportMedic explorer before moving on to the tax statutes.

The shortest summary, for anyone who needs one line: the 101st Amendment gave two governments power over the same tax and one room in which to agree, and then gave one of them a veto it can use alone.

The 101st Amendment in the company of other competence changes

Placing this amendment alongside the other occasions on which the Constitution has been amended to move legislative competence sharpens what is distinctive about it.

The Emergency-era transfer of five subjects from the State List to the Concurrent List moved competence in one direction and gave the states nothing in return, which is why it required no institution and why it survives without a single state having consented to it since. It is the model of a unilateral competence change made possible by an extraordinary parliamentary majority, and it is examined in the comparison of the 42nd and 44th Amendments.

The local government amendments created institutions without transferring competence at all, leaving the transfer to state discretion, which is why the institutions exist and the competence did not follow. That record is set out in the article on the 73rd and 74th Amendments in practice.

The 101st Amendment did something neither of those did. It transferred competence in both directions at once, took something from each level and gave something to each, and created an institution with a specified decision rule in which the exchange would be administered continuously. That is a structurally different kind of amendment, and it is the reason it required the consent of half the state legislatures and got it, when a purely extractive amendment of the same magnitude would not have.

Three features distinguish it and are worth carrying to any future proposal of the same kind.

It made the exchange simultaneous. The states surrendered sales tax competence and acquired services competence in the same instrument, so neither surrender preceded the compensating gain. Sequenced exchanges, where one side gives first and is promised something later, are the ones that fail, and the compensation guarantee is the part of this amendment that was sequenced and the part that proved weakest.

It put the decision rule in the Constitution. A voting weight and a threshold written into an article cannot be altered by the party that benefits from altering them. Had the composition and voting of the Council been left to a statute, the states’ protection would have been worth very little.

And it accepted a permanent institution rather than a one-off settlement. A tax that changes constantly cannot be settled once, and an amendment that had merely reallocated entries without creating a forum would have produced thirty-one legislatures drifting apart within a few years.

The failures, correspondingly, are all in the parts that were not given a rule: the compensation location, the dispute mechanism, the petroleum date and the binding character of recommendations. That is a clean division and it is the most portable finding in this article. Constitutional amendments succeed at what they specify and fail at what they defer.

The special provision for the north-eastern and hill states

One item in the Council’s mandate is easy to overlook and carries a federal significance out of proportion to its length. Article 279A directs the Council to make recommendations on special provision with respect to a named group of states: Arunachal Pradesh, Assam, Jammu and Kashmir, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim, Tripura, Himachal Pradesh and Uttarakhand.

The reason for naming them is that a destination-based consumption tax works differently for a small state with a limited manufacturing base, a high dependence on transfers, difficult terrain and a small tax administration. Under the previous system several of these states had used exemptions and incentives to attract industry, and those instruments largely disappeared with the reform, because an exemption granted by one state breaks the credit chain for everyone else.

The constitutional response was not to give these states a different tax but to require the Council to consider special provision for them, which has in practice taken the form of higher exemption thresholds and adjusted registration requirements rather than a different rate structure. The design choice is instructive. The amendment could have written a differential regime into the Constitution, which would have entrenched it and made it inflexible. Instead it placed the question on the agenda of a body that meets continuously and can adjust, at the cost of leaving these states dependent on the Council’s willingness to keep considering it.

Whether that was the right trade is a live question in fiscal federalism, and it is a smaller version of the same question the whole amendment poses: is a state better protected by a rule written into the Constitution or by a permanent seat in a room where the rule can be revisited? The answer this amendment gives is different for different items. The voting weight is in the Constitution and is therefore secure. The special provision for these states is a direction to consider and is therefore not.

Why an informal committee mattered more than any report

The intellectual case for a goods and services tax was made in expert reports, but the institutional case was made by an informal body with no legal standing at all, and that is the part of the story most accounts leave out.

The Empowered Committee of State Finance Ministers was created to coordinate the introduction of state value added tax in the early 2000s. It had no constitutional status, no statutory basis, no power to bind anyone and no enforcement mechanism. It was a forum in which finance ministers of states governed by opposing parties met, argued, and agreed on a common design that each then took home and enacted.

Its achievement was to prove that the thing the 101st Amendment would later require was possible. Before value added tax, the standard objection to any coordinated tax reform in India was that thirty state governments with different parties, different revenue positions and different election cycles could never sustain a common design. The Empowered Committee demonstrated that they could, and it built the personal and professional relationships among state finance ministers and their officials that the Council later inherited.

There is a general point here about institution building that the local government cluster illustrates in the negative. An institution created by constitutional fiat without a working culture behind it tends to become a notification. An institution created to give constitutional form to a practice that already exists tends to function, because the people in the room already know how to use it. The Council worked from its first meeting because it was, in substance, a body that had been meeting for over a decade.

What a future amendment to this scheme would have to do

Three changes are argued for regularly, and each would require a different instrument, which is worth setting out because the arguments frequently confuse them.

Bringing petroleum products into the tax requires no amendment. It requires a Council recommendation fixing a date and the ordinary statutory machinery. Anyone who says a constitutional amendment is needed for petrol and diesel is mistaken.

Bringing alcoholic liquor for human consumption into the tax requires a constitutional amendment to Article 366(12A), with a special majority in each House and ratification by at least half the state legislatures. Since the states hold the revenue that would be affected and would have to ratify it in their own legislatures, this is the least likely of the three.

Changing the compensation architecture, whether by extending a guarantee, altering its funding or writing it into the Constitution, could be done by ordinary legislation if the intention is a further statutory guarantee, or would require an amendment if the intention is to entrench it. The states’ experience in the compensation episode is the strongest argument for entrenchment, and the Union’s reluctance to accept an open-ended constitutional liability is the strongest argument against it. That disagreement is unresolved and is the most likely subject of the next serious constitutional argument about this scheme.

A fourth change, altering the voting weights or the three-fourths threshold, would require an amendment with ratification, and no proposal to do so has attracted significant support from either side. That silence is itself evidence: both the Union and the states appear to regard the current arithmetic as acceptable, which is the strongest available indication that the double-veto design is doing what its drafters intended.

Union territories, and the third definition people forget

The amendment inserted a third definition into Article 366 alongside those of the tax and of services, and it is the one most often omitted from summaries. For the purposes of the articles dealing with the new taxing power, the integrated levy and the Council, the expression “State” includes a Union territory with a legislature.

That clause does a surprising amount of work. Without it, a Union territory with its own legislature would have had no power to levy the state component of the tax and no seat in the Council, which would have left a gap in the credit chain wherever such a territory sits in a supply route. With it, those territories legislate on the tax as states do and their finance ministers sit in the Council as members with an equal share of the state voting weight. Union territories without a legislature are covered instead by a separate central statute enacted for them, since there is no local legislature to enact a state law.

The clause also has a quiet effect on the arithmetic set out earlier. Because the state pool of two-thirds is divided equally among the members voting, every Union territory with a legislature that participates increases the number of shares and reduces each state’s individual weight. That does not change the blocking fraction, which is expressed as a proportion of those present and voting, but it does mean the number of members required to assemble a blocking group is not fixed and moves with the composition of the meeting.

The cascade, shown rather than described

The clearest way to see what the amendment achieved is to follow a single good through a chain under both systems, without numbers, since the mechanism rather than the magnitude is the point.

Under the previous system, a manufacturer paid central excise on the goods at the factory gate. That duty entered the price. A distributor buying from the manufacturer paid state sales tax on a price that already included the excise, and could not set the excise off against the sales tax because the two were levied by different governments under different powers. If the distributor sold across a state boundary, central sales tax attached and was creditable nowhere. If the goods entered a local area, entry tax or octroi attached. A retailer then paid state sales tax on a price containing all of the above. At each stage the base included tax paid at the previous stage, and the total burden depended on the number of stages rather than on the value added at each.

Under the present system, tax attaches at each supply, and each recipient claims credit for the tax charged on the supply to it, so tax is effectively paid only on the value added at each stage. When the supply crosses a state boundary the tax charged is the integrated levy, which the recipient credits in the destination state, and the apportionment moves the state share to the destination. The number of stages no longer determines the burden.

That is the whole economic case for the reform, and it is a case that could not have been implemented without the constitutional change, because the credit at each stage requires that the tax being credited and the tax being paid arise under a single coherent scheme rather than under two competing competences. This is why the reform is properly described as a constitutional achievement first and a tax achievement second.

The reverse charge and the constitutional question underneath it

One design feature of the tax has produced more constitutional litigation than any other, and understanding why requires returning to the definition of the taxable event.

Ordinarily the supplier collects the tax and pays it. Under a reverse charge, the recipient is made liable instead. The mechanism exists because there are situations in which the supplier is outside the administration’s reach, most obviously where the supplier is located abroad, or where a large number of small unregistered suppliers deal with a small number of registered recipients and collecting from the recipients is more practical.

The constitutional question is whether a reverse charge on a recipient is a tax on supply at all within the meaning of the definition inserted into Article 366. The answer accepted by the courts is that it is, because the levy still attaches to the event of supply and the identification of the person liable to pay is a matter of collection machinery rather than of the charge itself. A tax may be levied on one person and collected from another without ceasing to be a tax on the same event, and Indian tax law has long accepted that distinction.

Where the litigation has succeeded is at the next step, and this is what the ocean freight case turned on. If the transaction has already been taxed as a composite supply in the hands of the recipient, a second levy on one element of it, charged to the same person under a reverse charge, is not a levy on a separate supply at all. It is a second levy on the same supply, and the statute treating a composite supply as a single supply does not permit it. The failure was therefore not in the reverse charge mechanism but in identifying a supply that the scheme had already treated as part of another one.

That distinction matters beyond the specific dispute, because it shows how the constitutional definition and the statutory scheme interact. Article 366 fixes the taxable event as supply. The statutes then define when two things are one supply and when they are two. A levy that offends the statutory definition of a single supply is bad even though the constitutional power to tax supply plainly exists, and a levy that would offend the constitutional definition would be bad even if every statute permitted it. Keeping the two levels of analysis apart is the discipline that this area most rewards, and conflating them is the commonest error in argument about it.

What the amendment tells a reader about Indian federalism

Three propositions about the federal structure are supported by this amendment better than by almost any other constitutional episode, and they are worth stating because they generalise.

The first is that Indian states will trade competence for a rule they can rely on. The standard account of Indian federalism treats states as resisting every transfer of power to the Union, and the ratification of this amendment by more than half the state legislatures within weeks falsifies that account in its strong form. What the states obtained was not more power but a decision procedure entrenched in the Constitution, and they judged that worth a permanent surrender of taxing autonomy.

The second is that the value of such a trade depends entirely on what is entrenched and what is left to ordinary law. The voting weights sit in Article 279A and are secure. The compensation guarantee sat in the amending Act and in a statute and expired on schedule. Same bargain, two components, and the difference in outcome is explained wholly by the instrument each was placed in.

The third is that cooperative federalism in India works where there is a forum with a rule and fails where there is only a forum. The Council decides because Article 279A tells it how to decide. Bodies established for intergovernmental coordination without a decision rule tend to become consultative and then ceremonial, which is the fate of most such bodies in Indian practice. A drafter designing the next intergovernmental institution should take the voting clause more seriously than the mandate.

Frequently Asked Questions

Q: What did the 101st Amendment change in the Constitution?

It inserted Article 246A, giving Parliament and every state legislature simultaneous power to make laws on goods and services tax, with exclusive Union power over inter-state supply. It inserted Article 269A for the levy, collection and apportionment of the integrated tax on inter-state supply, and Article 279A constituting the Goods and Services Tax Council. It added definitions of the tax and of services to Article 366. It made consequential changes to the articles governing residuary power, legislation on State List subjects in the national interest and during an Emergency, the distribution of central taxes, surcharges and restrictions on state taxation, and omitted a never-commenced article on service tax. It also rewrote entries in both the Union and State Lists to remove the subsumed taxes.

Q: What is Article 246A and why is it unusual?

Article 246A confers power to legislate on goods and services tax on Parliament and on every state legislature at the same time, and gives Parliament exclusive power where the supply takes place in the course of inter-state trade. It is unusual for four reasons. It allocates a subject directly in the article rather than through an entry in any List, so goods and services tax does not appear in the Seventh Schedule. It opens with a non obstante clause overriding both Article 246 and the repugnancy rule in Article 254, so the ordinary mechanism by which a central law prevails over a state law does not operate. It creates simultaneous rather than concurrent power. And its application to five petroleum products is deferred until the Council recommends a date.

Q: How does voting work in the GST Council?

Every decision requires a majority of not less than three-fourths of the weighted votes of the members present and voting, with a quorum of one half of the total membership. The Union’s vote carries a weight of one-third of the total votes cast in that meeting, and the votes of all state governments together carry two-thirds, shared equally among the states voting. Because the weighting is of votes cast rather than of total membership, absent or abstaining states do not dilute the state bloc. With the Union in favour, at least five-eighths of participating states must also support a proposal. Nothing can pass over the Union’s objection, since the state pool alone cannot reach three-fourths.

Q: Are GST Council recommendations binding on states?

No. The Supreme Court held in 2022 that the Council’s recommendations have persuasive value and do not bind Parliament or the state legislatures. The reasoning rests on Article 246A conferring simultaneous legislative power on both levels, with neither that article nor Article 279A subordinate to the other, so the Council is a body of collaborative dialogue rather than a superior legislature. Where a statutory provision makes a recommendation a precondition for exercising a delegated power, the recommendation operates within that statutory scheme. In practice no state has meaningfully departed from the Council’s design, because the input tax credit chain, the common statutory model and the exclusively central integrated levy make unilateral variation costly to the state that attempts it.

Q: What happened to entry tax and octroi after the 101st Amendment?

Both were abolished. The State List entry permitting taxes on the entry of goods into a local area for consumption, use or sale was omitted outright by the amendment, and octroi was a form of that levy collected at municipal boundaries. Their removal is one of the clearest practical achievements of the reform, because they were the taxes that created internal customs frontiers within India, required check posts at state and municipal boundaries, and were not creditable anywhere in the chain. Their abolition is also why the amendment is credited with reducing freight transit times, since the delays arose from documentary verification at those boundaries rather than from the tax itself.

Q: Why did the GST amendment need approval from state legislatures?

Because it altered the distribution of legislative powers between the Union and the states and amended the Seventh Schedule, which brings it within the proviso to Article 368(2). An amendment of that character requires, in addition to a majority of the total membership of each House and a two-thirds majority of those present and voting, ratification by the legislatures of not less than one-half of the states before it can be presented to the President for assent. More than half the states ratified it in the weeks after both Houses had passed it, and assent followed on 8 September 2016. It is the cleanest modern example of the ratification proviso operating as designed.

Q: What is Article 269A and how is IGST divided?

Article 269A provides that goods and services tax on supplies in the course of inter-state trade or commerce is levied and collected by the Government of India and apportioned between the Union and the states in the manner Parliament provides by law on the Council’s recommendations. An explanation deems supply in the course of import into India to be inter-state supply. A separate clause provides that the amount apportioned to a state does not form part of the Consolidated Fund of India, which removes the need for a parliamentary appropriation before the money is released. Parliament is also empowered to formulate principles for determining the place of supply, which is the constitutional basis for the rules deciding which state receives the tax on a given transaction.

Q: How long were states promised compensation for GST revenue loss?

For five years from the introduction of the tax. Section 18 of the amending Act obliged Parliament to legislate, on the Council’s recommendation, for compensation to states for loss of revenue arising on account of implementation of the tax for a period of five years. The compensation statute fixed a base year, guaranteed a protected rate of growth in subsumed revenues, and funded payments from a dedicated cess credited to a compensation fund. Three features shaped what followed: the period was fixed, the protected growth rate was set well above what most states had been achieving, and the funding was ring-fenced to the cess rather than payable from general revenues, so a shortfall in the cess created a gap with no automatic remedy.

Q: Who are the members of the GST Council and who chairs it?

The Union Finance Minister is the chairperson. The other members are the Union Minister of State in charge of revenue or finance, and the minister in charge of finance or taxation, or any other minister nominated by the state government, from each state. The state members choose one among themselves as vice-chairperson for a period they decide. Where a proclamation under Article 356 is in operation in a state, the member is a person nominated by the Governor. The Revenue Secretary serves as ex officio secretary and the head of the central indirect tax administration attends as a permanent non-voting invitee. Each state has one member regardless of population or revenue, so voting weight among states is equal.

Q: Can the Union be outvoted in the GST Council?

No. Because a decision requires at least three-fourths of the weighted votes cast and the states together hold only two-thirds, the states cannot reach the threshold even with complete unanimity if the Union votes against. The Union therefore holds an absolute veto exercisable alone. The reverse is not symmetrical: the Union cannot pass anything by itself either, since one-third falls well short of three-fourths, and it needs at least five-eighths of the participating states. The asymmetry is that the Union’s veto requires no coalition while the states’ veto requires a group exceeding three-eighths of those present and voting.

Q: Why are petrol and diesel still outside GST?

Because the Council has not recommended a date for their inclusion, and nothing compels it to do so within any period. Petroleum crude, high speed diesel, motor spirit, natural gas and aviation turbine fuel are within the definition of the tax and within Article 246A, but the explanation to that article defers its application to them until the Council recommends a date, and Article 279A requires the Council to make that recommendation without setting a time limit. The reason for the delay is fiscal: these products yield a large, efficiently collected and short-run inelastic revenue for both the Union and the states, and inclusion at prevailing rate levels would reduce that revenue sharply while making the tax creditable to fuel-consuming businesses.

Q: Why is alcohol treated differently from petroleum under GST?

Because the two exclusions sit in different places and have different legal consequences. Alcoholic liquor for human consumption is carved out of the definition of goods and services tax in Article 366(12A) itself, so it lies outside the tax as a constitutional matter and no Council recommendation or ordinary statute can bring it in. Only a further constitutional amendment, requiring a special majority in each House and ratification by at least half the state legislatures, could. Petroleum products are inside the definition and inside Article 246A, with application merely deferred. Treating the two as equivalent exclusions is the most consequential error in published writing on this amendment, because it misstates what changing either position would require.

Q: What did the Supreme Court decide in Mohit Minerals?

The dispute concerned whether integrated tax could be levied on an importer, on a reverse charge basis, in respect of ocean freight where the foreign supplier had arranged the shipping. The Court held in May 2022 that the levy could not stand, because the importer was already liable on a composite supply that included the transportation element and a separate levy on the same element amounted to taxing it twice within a scheme that treats a composite supply as one supply. The constitutionally significant part of the reasoning was the rejection of the argument that the levy was valid because the Council had recommended it: the Court held that Council recommendations have persuasive value, that Article 246A confers simultaneous power on both levels, and that neither article is subordinate to the other.

Q: Did the 101st Amendment abolish entertainment tax?

Not entirely, though it is commonly described that way. The State List entry on taxes on entertainments and amusements was narrowed rather than omitted, and what survives is confined to such taxes to the extent they are levied and collected by a panchayat, a municipality, a regional council or a district council. The state-level entertainment tax was subsumed into goods and services tax; the local-body portion was preserved and placed in the exclusive fiscal space of local government. Whether local bodies actually receive and use that space depends on state legislation and on the wider devolution record, since a tax entry reserved to local government is only as valuable as the local government it is reserved to.

Q: Why was the additional one per cent inter-state tax dropped from the bill?

Because it reproduced the distortion the reform was designed to remove. The bill as introduced in 2014 allowed the Union to levy an additional tax of up to one per cent on inter-state supply of goods, assigned to the originating state, as a concession to manufacturing states worried that a destination-based tax would shift revenue towards consuming states. Since the levy would not have been creditable, it would have stuck as a cost on every inter-state transaction, which is precisely what central sales tax had done. The select committee and the opposition pressed for its removal and it was dropped in the amendments the Rajya Sabha passed in August 2016, with a strengthened compensation obligation offered in its place.

Q: Does the GST Council have a dispute resolution mechanism?

It has a permission to create one and has not used it. Article 279A contains a clause allowing the Council to decide about the modalities for resolving disputes arising out of its recommendations, whether between the Union and states or among states. No standing mechanism has been established under it. The difficulty is structural: any mechanism would either bind, creating an authority above Parliament and the state legislatures that the Constitution does not otherwise contemplate, or not bind, in which case it would add a stage without resolving anything. Disputes have therefore gone to political negotiation or, as in the litigation on the character of the recommendations, to the ordinary courts.

Q: Why did GST have to start on 1 July 2017?

Because of a transitional provision in the amending Act rather than a policy choice. Section 19 provided that any state law relating to tax on goods or services which was inconsistent with the amended Constitution would continue in force only until it was amended or repealed, or until one year from commencement, whichever was earlier. The commencement notifications were issued in September 2016, so the saving expired in September 2017. Had the new tax not been in force by then, states would have lost their principal tax revenue with nothing to replace it and no constitutional route to an extension. The mid-year launch and the compressed design work both follow from that hard deadline.

Q: Is GST listed in the Seventh Schedule?

No, and this surprises readers who look for it there. Every other head of taxing power in the Constitution sits in an entry in the Union List or the State List, with Article 246 determining who may legislate on it. Goods and services tax is allocated directly by Article 246A without any corresponding entry, which is one of the features that makes the article unlike anything else in the constitutional scheme. What the amendment did to the Seventh Schedule was subtractive rather than additive: it narrowed the central excise entry to five petroleum products and tobacco, omitted the entry tax and advertisement tax entries, narrowed the state sales tax entry, and confined the entertainment tax entry to local bodies.