ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Goods and Services Tax: The 13th Finance Commission and the Way Forward

The Thirteenth Finance Commission was required to look into the revenue impact of the introduction of the goods and services tax. Its report, based on the recommendations of a task force constituted to study the issue, recommends a highly uniform and centralised format that does not adequately recognise a tax reform exercise in a multi-level fiscal system that involves compromises and trade-offs. While several flaws can be pointed out in its design, developments that have taken place before and since the report was submitted have to a large extent rendered the commission's recommendations irrelevant. All this underlines the need for a model that goes beyond uniform rates of tax and allows states to vary beyond a floor, with a fixed classification of commodities and services, so that they can choose an appropriate rate to ensure that their revenue requirements are met.


Goods and Services Tax: The 13th Finance Commission and the Way Forward

R Kavita Rao

The Thirteenth Finance Commission was required to look into the revenue impact of the introduction of the goods and services tax. Its report, based on the recommendations of a task force constituted to study the issue, recommends a highly uniform and centralised format that does not adequately recognise a tax reform exercise in a multi-level fiscal system that involves compromises and trade-offs. While several flaws can be pointed out in its design, developments that have taken place before and since the report was submitted have to a large extent rendered the commission’s recommendations irrelevant. All this underlines the need for a model that goes beyond uniform rates of tax and allows states to vary beyond a floor, with a fixed classification of commodities and services, so that they can choose an appropriate rate to ensure that their revenue requirements are met.

R Kavita Rao ( is at the National Institute of Public Finance and Policy, New Delhi.

Economic & Political Weekly

november 27, 2010 vol xlv no 48

1 Introduction

he introduction of the goods and services tax (GST) is e xpected to be a major reform in the arena of domestic i ndirect taxes in India. While this new regime will change the tax base of both central and state taxes, the extent of its impact on the economy as well as the finances of various levels of g overnment was not clear. The Thirteenth Finance Commission (THFC) was assigned the task of incorporating the impact of the proposed implementation of the GST into its overall recommendations. To obtain an assessment of the GST’s likely impact, the THFC constituted a task force. The task force submitted its report (Finance Commission 2009a) to the THFC on 15 December 2009 and its recommendations were endorsed in the final Report of the THFC (2010-2015) (Finance Commission 2009b, subsequently Report).

The recommendations of the THFC, which says that “both the Centre and the States should conclude a Grand Bargain to implement the model GST”, comprise five elements.

  • (1) There will be a prescribed design for the GST as a consumption-based tax subsuming the bulk of all indirect taxes, including stamp duty and state excises. It will be a single rate regime with a few exemptions,1 have a uniform threshold for the centre and all states, uniform rates of tax across all states, and include a proposed design for treatment of interstate transactions.
  • (2) There will be an emphasis on harmonised tax laws and procedures for administration, and common dispute resolution and advance ruling mechanisms, as well as simultaneous implementation by all the states.
  • (3) There will be a binding agreement between the centre and the states on the design as well as the rates of tax to be adopted, and the conditions under which the rates can be altered. The THFC’s proposed scheme says that for decreasing the rates, all states will have to agree, but for increasing them, only two-thirds need agree. The power of veto will be vested in the centre.
  • (4) In the event of the agreement being violated, disincentives in the form of withholding state-specific grants and GST compensation grants will apply.
  • (5) There will be a phased implementation allowing for delayed incorporation of real estate transactions into the base and allowing for a two-rate tax in place of one in the initial years of implementation, with an agreement to move forward before 31 December 2014.
  • Any GST adopted, the THFC emphasises, has to be consistent with all the elements of the “Grand Bargain” listed above. If not, the compensation package of Rs 50,000 crore proposed by it will not be disbursed.


    There are three fundamental problems with the approach adopted by the THFC. First, since it was required to look into the revenue impact of the introduction of the GST, it had to work on the assumption of some design for the tax. It would have been prudent for the THFC to have explored alternative designs and assessed their corresponding revenue impact. Thus, it would have generated a bouquet of options for the policymakers, allowing them to evaluate the choices available. However, it preferred a route where it sought to identify and prescribe a model for the country. While this is a very meaningful academic exercise, it does not fit into the overall scheme of a union finance commission’s (UFC) domain, especially because it is not the ultimate decision-maker in this policy reform exercise. By the time the THFC Report was submitted, it should be noted, the Empowered Committee of State Finance Ministers (EC) had already hammered out a consensus on some of the design issues and put out a discussion paper. Reopening these issues and suggesting a radically different view on some of them will not make the Report/design acceptable to the states, who are key partners in this tax reform. Further, the proposed design, by attempting to be comprehensive and complete, incorporates some difficult areas for taxation. It also proposes to radically alter the fiscal space, and the roles of the centre and the states, in the country. The difficulties with the proposed design are discussed in the next section.

    After this exercise, the THFC assumes that whichever be the design adopted, it will be revenue neutral to both the centre and the states. By making this assumption, however, it eliminated the need to incorporate the impact of the GST on the rest of its recommendations. In the same stroke, it also made its recommendations irrelevant to the debate on the GST in India. The extremely deterministic nature of the recommendations, along with the stipulation that the incentive grants for the GST will be accessible only if the entire model is accepted, allowed the centre to accept the Report and yet allow the GST debate to take a completely different course. As developments since the submission of the R eport indicate, the evolving design of GST in India is somewhat removed from that recommended by the THFC.

    Second, apart from recommendations on the design of the GST, the task force suggested rates for the centre and the state that would be adequate to raise the revenues required. The rate suggested in this exercise, 12% for both levels of government put t ogether, is substantially lower than the current rates of more than 20% in the case of manufactured goods. By suggesting the feasibility of such a low rate, the task force built a case for a s ingle-rate tax regime. It is quite commonly understood that at somewhat higher rates a single rate appears politically infeasible to policymakers. It is therefore important to ask whether the methodology adopted by the task force provides a reliable estimate of the underlying base for the tax. The third section explores the methodology adopted and identifies some of its limitations. This is particularly important because in all subsequent discussions on the GST between policymakers and potential tax payers, this rate has b ecome the benchmark for comparisons and both levels of government are hard-pressed to defend proposing distinctly higher rates.

    Third, the model GST of the THFC recommends a uniform rate across all states. While this approach is proposed on the grounds that it will aid in forming a common market across all the states in India, it is useful to understand what one means by a common market, what aids or hinders forming a common market, and what the downside is of such a recommendation. The fourth section explores these issues. Some concluding remarks on the way forward for the GST in India are presented in the last section.

    2 Design Difficulties

    The design of the GST proposed by the task force and adopted in the THFC Report is a classic comprehensive value added tax (VAT), which has a lot to commend it. As is well accepted, a comprehensive VAT is desirable to ensure that distortions to economic decision-making arising from cascading in the tax regime are minimised. While exempting final consumption does introduce some distortions, the effect is more damaging if an exemption is a ccorded to a sector that provides inputs to other taxable sectors. In the design proposed by the EC, electricity and transportation of goods are two important examples of input-providing sectors left out of GST coverage. If electricity is kept out of the GST structure, inputs used by this sector will be subject to GST, for which no mechanism of credit will be available – these taxes will be added to the cost of production. Further, there will be some other standalone taxes, for which credit will not be available within the GST. This will generate cascading, and result in distortions such as attempts by users of electricity to generate their own power instead of relying on supplies from the grid. In such cases, the sector and the economy do not reap any benefits from economies of scale. It is therefore desirable that all such sectors be incorporated into the GST design – an approach the task force and the THFC Report have taken, unlike the EC.

    However, there are two sectors that pose some difficulties in being incorporated into a GST regime, financial services and real estate-related transactions. Unlike in other supplies, the value addition in some financial services is embedded in a margin – the value addition by a bank through financial intermediation is captured in the interest margin that the bank earns. Taxing this margin is not difficult, but assigning the tax collected to the agents benefiting from these services so that they, in turn, can claim i nput tax credit for taxes paid on financial services is difficult. Therefore, in some countries, these services are exempt with limited tax credit for inputs purchased, while in others, the tax law mandates that where the charges for the services provided are in the form of a fee, a tax needs to be charged, but in the case of margin-based services, no taxes are payable. This is the case with the Indian service tax as well. The latter approach, however, encourages financial service providers to construct services where the charges are only implicit. The slew of bank accounts with a minimum balance that an account holder has to maintain in r eturn for a wide spectrum of free services is an example of such financial engineering. Similarly, for incorporating non-life insurance services, it is important to identify the true value of the service supplied by the service provider.

    Including financial services in the GST framework requires d esigning and adopting technical solutions. The task force report mentions three options, but leaves the choice to the policymakers. The need to bring financial services within the tax net

    november 27, 2010 vol xlv no 48


    conceptually relates to distortions introduced by exempting these services – since service providers will not have access to input tax credit for inputs purchased by them, they are likely to resort to self supply or imports to avoid a tax liability. Further, some revenue could be lost because the services to final consumers will also escape taxation.

    Here, it is useful mention that from the input-output tables for 2006-07, more than 73% of supplies of financial services are for business purposes. The loss of revenue from exemptions will therefore be from the remaining 27%. To the extent there is some revenue captured through taxation of inputs, the revenue loss will be lower. To get a more precise idea, it will be useful to

    o btain alternative estimates of the extent to which these services are utilised by final consumers and the extent to which they support business enterprises. If business enterprises emerge as the more important and significant players, it will be adequate to find mechanisms to address the concerns of these segments and financial sector firms instead of attempting a more comprehensive taxation approach. A partial set-off, as proposed in Australia and Singapore, or a partial zero-rating, as proposed in New Zealand, are viable options to consider. The answers on how these services should be treated need to be ironed out before proposing a “broad” base by incorporating these services into the GST base.

    The other sector which has received plenty of attention in the task force report as well as in the THFC recommendations is real estate. The task force argues for incorporating this sector – that is, all transactions related to sale and purchase as well as renting of commercial and non-commercial property – into the GST. Apart from arguments on equity, it asserts that this sector now faces very high tax liability in the form of sales taxes on the material used and a separate stamp duty on the sale and purchase of property. The high tax liability encourages undervaluation of property. Incorporation of these transactions into the GST regime will reduce cascading and diminish distortions in this sector. While this is the basic premise of any GST regime, implicitly, this argument presumes that the bulk of the distortions in the sector arises out of indirect taxes. However, income tax provisions and pre-existing valuation difficulties queer the pitch for such an a rgument. While the first sale of a new construction faces higher valuation, there is no incentive to record augmented values in any subsequent sale. Apart from a GST liability, such a valuation would trigger income tax liability as well. This can serve as a persistent disincentive to improved valuation. As long as this i ssue does not resolve itself, incorporating real estate transactions into the GST will not be prudent.

    One of the conditions under which some resolution is possible is the rapid expansion of the proportion of transactions that are financed through loans from formal financial institutions. Today, if one compares the value of transactions reported through purchase of non-judicial paper (assuming an average tax rate of 7%) with the value of turnover reported in income tax returns by builders and developers, the latter accounts for 80% of the former. While there is no information available on how many transactions these represent of the total, this fraction will a ccount for the bulk of fair value transactions. Introducing GST in these cases will potentially reduce their tax liability as a result of

    Economic & Political Weekly

    november 27, 2010 vol xlv no 48

    input tax credit. For the others, the valuation issue remains. With the Direct Taxes Code proposing to harden its stance by allowing rollover of receipts into an alternative asset only in the case of the first house for an individual, any fair valuation would trigger c apital gains tax, a potential liability which will neutralise any gains from the GST input tax credit.

    Here, the Indian context provides an additional dimension of conflict. At present, the state governments have the right to levy and collect taxes on such transactions in the form of stamp d uties. Incorporation of these transactions into the GST would r equire subsuming stamp duty into the basket of taxes to be r eplaced by the GST. Given the highly uniform and centralised format of the GST proposed in the THFC’s Report, the states perceive this as a mechanism to reduce their fiscal autonomy. Given this, the states will endeavour to keep some other levies outside the purview of the GST to retain their sense of fiscal autonomy. Extension of the GST to include real estate will therefore have to be left to future reforms – a factor the task force as well as the THFC’s Report does not recognise.

    The task force report makes very heroic and often rosy a ssumptions while providing an assessment of the increments to tax base by incorporating real estate transactions into it. It a ssumes that all household sector gross fixed capital formation is in housing. The household sector also comprises partnership firms, proprietary firms and self-employed individuals. All these units will be investing some amount on their business. For i nstance, if unregistered manufacturing accounts for almost a quarter of total manufacturing, it will be investing some amount towards capital formation. These amounts will be recorded as household investment. If all this investment is assumed to be investment in housing, and the land component of the GST base is assumed to be equally large, it will be an overestimate for these reasons alone. As a point of comparison, while the report assumes the land component of real estate transactions to be Rs 4,29,260 crore, the total base for stamp duty at 7% is barely Rs 4,02,179 crore – and this includes both land and buildings. Clearly, incorporating such a substantial estimate into the projected tax base will be way off the mark.

    Apart from the specific coverage-based issues discussed, one distinct feature of the GST regime proposed is “uniform rates, uniform procedures”, with a significant lock-in to initial decisions taken. The model proposed suggests that all changes will require two-thirds of the states to agree and the centre will have veto power. While, there is a broad consensus that GST is best levied and administered at the central level – most countries have such a system – it must be noted that a tax reform exercise in a multilevel fiscal system involves compromises and trade-offs. The main trade-off here is between tax harmonisation towards uniformity and fiscal autonomy of the states. The THFC’s recommendations, however, represent a rather centralised model for India. The model, and its implementation, is in favour of a complete lock-in to uniformity and centralisation.

    One pointer to uniformity and centralisation is the élan with which the Report proposes a model distinct from that proposed by the EC for dealing with interstate transactions. It suggests a “modified bank model”, which comes very close to having an


    autonomous tax administration that collects the tax and transfers the designated amount to respective levels of government. The modified bank model envisages that some nodal bank will be able to set up and run a comprehensive information system for the GST system. When compared to the Integrated GST (IGST) model, the modified bank model, as proposed in the Report, seems more restrictive, just as the overall design of the GST. The IGST model potentially allows the states to choose different rates of tax for local sales – a feature that makes it more flexible w ithout losing any of the benefits of the “modified bank model” proposed by the THFC.

    The Report makes a very eloquent case for harmonisation of the law, rules, procedures, forms and rates across all states. While there is convenience in harmonisation, the decision to have uniform rates should be taken by the EC, and not be mandated by anyone else. The need to emphasise this point is the radical change in the fiscal space proposed by the Report and the EC’s discussion paper. It is important to realise that the Report explicitly argues that the states give up the notion of autonomy in taxes and focus on the broader concept of autonomy in fiscal decisions. Whether this is an acceptable objective to the states needs to be widely debated; it cannot be considered a foregone conclusion.

    3 Revenue Neutrality

    For assessing the base of the tax, the Report provides five alternative estimates. Two are based on estimates of consumption e xpenditure, one based on information compiled from income tax returns of assessees, one based on the “Shome Index” and, finally, one based on revenue collection by the union government, corrected for the incremental base that will result from moving to a comprehensive GST.

    To begin with, let us look at the most conservative estimate in the Report, the one derived from the “Shome Index”.2 This index is basically a rule of thumb that says for a given rate of tax, the revenue (as a percentage of gross domestic product, GDP) a tax regime will generate varies between a third of that rate and half of that rate. In other words, if the rate of tax is 12%, the revenue will vary between 4% and 6% of GDP. Regimes are expected to approximate 6% if they have a broad base, good compliance and a transparent administration. From this, the task force presumes it appropriate to infer that the country will be at the upper end of the range suggested by the index and correspondingly, the base for the tax, will be half of GDP. After incorporating the effects of design, administrative effectiveness and degree of compliance into the Shome Index, which is basically a rule of thumb, it will be incorrect to consider the upper limit as an estimate or even an indicator of the size of the base. This appears more like wishful thinking and less like an estimator of any kind. Further, if India happens to lie closer to the lower end of the spectrum, and not the upper end as hypothesised, then the suggested tax base would be significantly smaller, a factor the Report does not want to consider. Following the format in the Report, if India is at the lower end, the base for the tax would be about Rs 15 lakh crore and not the Rs 21 lakh crore cited, and the corresponding revenue neutral rate will be more than 23%.

    Now let us look at the estimates based on income tax returns. This estimate is based on the profit and loss accounts filed by more than 28 lakh business entities for the financial year 2007

    08. These accounts provide information on local supplies of goods and services, as well as on purchases of goods and services, including capital goods, all of which entitle the firms to input tax credit. The difference between the value of supplies and the value of purchases would therefore be an estimate of the tax base. However, since the units would not have been entitled to tax credit on purchases from the informal sector, the Report makes some assumptions on the extent of such purchases. Further, corrections are made for sectors that are under-represented in the income tax data base, for instance, railway charges for transportation of goods.

    A look at the data presented in this section suggests that there is a fundamental discrepancy between the income tax data and the Central Statistical Organisation (CSO) data on national income. The Report does mention that sales as reported in income tax data are higher than sales reported in the CSO data. However, the GDP estimates as per the CSO are significantly higher than those reported in the income tax data. In other words, the “deductions” from gross output to derive the value added in the case of income tax data are substantially higher than those reported in the CSO data. As a result, the value added in the CSO data is significantly higher than that reported in the income tax figures

    – Rs 33,54,597 crore in the CSO figures against Rs 16,03,564 crore in the income tax figures. The income tax data therefore shows higher turnover but lower value added, but this feature is not explored or explained anywhere. It is not clear why the income tax data should be considered any more reliable than the CSO data.

    Further, even if one takes the income tax figures at face value, it is important to point out that the ratio of intermediate inputs (B2) to sales (A1), as reported in Table 5 of the Report, suggest a margin of only 15.96% to cover both profits and wages and salaries. Depending on the extent of specialisation in any industry, the ratio of profits to sales can vary. Information available from the Prowess database for 2007-08, however, suggests that the r atio of profits to sales for the manufacturing sector alone is 9% of sales. Together with wages and salaries and depreciation d eductions, it amounts to more than 18% of sales. For other services, the ratios are significantly higher at 15.2% for profit before tax and 36% for profits, depreciation and wages as a proportion to total sales. These figures raise some questions about the usefulness of the database sourced from income tax returns for the estimation of the GST tax base.

    Given the data, the Report makes corrections for purchases from the informal sector. It assumes that of the purchases reported in the profit and loss accounts, a fraction is from segments of the economy that are, and will continue to be, outside the tax net because of the exemption threshold. This approach assumes that there is a vibrant link between these two segments of the economy. While for sectors that are already in some form of a VAT system it would not make sense to continue to purchase from units that cannot provide a VAT invoice for purchases, it is possible that there are hitherto untaxed sectors that have such a link. However, how large this link is, is an important question. The Report on average assumes that more than 20% of total purchases are from the informal sector. While this number is as good

    november 27, 2010 vol xlv no 48


    or as bad as any other number, the derived tax base is quite sensitive to where the level is pegged. For instance, if the purchases from the informal sector are assumed to be 10%, not 20%, of total purchases, the total tax base would decrease from Rs 30 lakh crore to about Rs 20 lakh crore, and the revenue neutral rate of tax for the country would increase to more than 16%.

    Justifying these numbers in terms of the average size of the unorganised sector in the economy is rather arbitrary because, given the operation of various tax systems, there is an inherent interest in the unorganised sector to purchase from itself and hence become self-contained. Further, a chunk of the unorganised sector in India, as classified by the National Academy of S ciences (NAS), would comprise partnership and proprietary firms, which would be a part of the tax system as proposed by the GST design.

    The third estimate presented is based on information from n ational income accounts. This alternative is based on the notion that the tax would apply on final consumption of goods and services by government and private individuals. Since this would include some exempted goods and/or purchases from the informal sector, some adjustments are made for it. A third component of the base in this approach is the gross domestic capital formation by the household sector, net of labour inputs in construction. The Report argues that all investment expenditure by the household sector should be included here because this segment has no scope for claiming any input to tax credit.3 This approach then goes on to make an erroneous correction – as in the income tax approach, it assumes that purchases from the informal sector would add to the base. The approach based on the income tax data works through approximating the extent of value added, and if there are tax-free purchases by a firm, the way VAT operates, these purchases are added to the tax base and taxed as a part of the output of the firm. However, in the consumption approach, the estimates are figures for final consumption and not value added. So there is no additional liability on account of such purchases. But some other correction is definitely required – if there are purchases of organised sector output to be used as inputs in the unorganised sector/exempt sectors, then there will be tax payable on these purchases, which will not be reflected in the figures of final consumption from the organised sector.

    Taking the figure for exempted sales/sales by the organised sector, applying the average ratio of purchases to sales, and assuming 18.8% of total purchases to be from the informal sector (as assumed in the case of income tax data), or more conservatively, assuming half the purchases to be from the informal sector, it is possible to derive the value of purchases from the formal sector. With these revised numbers, the base for the tax is significantly lower at Rs 25 lakh crore against the estimate provided by the Report of Rs 37 lakh crore. If the second assumption is considered more appropriate, then the figure falls further to Rs 23 lakh crore. Corresponding to these estimates, the revenue neutral rate for the country would be 13.4% and 15% respectively.

    The revenue method of the Report is also based on some optimistic assumptions. One such assumption is that all the revenue reported as forgone can actually be counted for determining the base or the potential for taxation. There are two qualifications

    Economic & Political Weekly

    november 27, 2010 vol xlv no 48

    that this number needs to be subjected to. The first, not all activity that exists under an exemption will survive if a tax is introduced on the activity. Some of the economic units will become unviable, and the scale of economic activity will be lower. This is a well-recognised issue when discussing interpretations of figures of revenue forgone. Studies on tax expenditures therefore make a distinction between revenue forgone and revenue earned (Villela et al 2010).

    Second, in the Indian context, the figures reported by the d epartment of revenue are erroneous – while central VAT (CenVAT) reports figures for revenue collected net of input tax credit, the figures for revenue forgone are gross revenue. These figures are not comparable with the CenVAT figures, and cannot be added to CenVAT either. A similar difference exists between CenVAT and countervailing duties. The derived figures for output tax and input tax therefore need to be corrected. This approach also adds on potential revenue from services such as financial services and railways. The bulk of these services are used by the goods sector and hence taxes are here a contra entry, triggering a tax credit. To this extent, adding these services will be overestimating the base. Even assuming that only half the revenue forgone actually would accrue to the exchequer as additional revenue, and excluding additionalities from financial services and railways, the tax base will decline to Rs 25 lakh crore from Rs 29 lakh crore.

    If the more conservative estimates suggested here are adopted, the average as computed in the task force report would drop to around Rs 23 lakh crore instead of the Rs 30 lakh crore in the Report. The corresponding rate of tax would be 15%. However, when alternative estimates are provided in any context, it is appropriate to rank these estimates and choose from among them. The approach adopted by the Report of taking an average of these estimates as the base for the tax, especially when they are fairly divergent, beats reason. This approach does not provide any rationale for including widely different estimates in the pool, and does not attempt to assess the robustness of these estimates to changes in assumptions. A more pragmatic approach would have been to take the smallest estimate if no rational basis can be found to choose one of the estimates over the others.

    Given the sensitivity of the numbers to underlying assumptions, and further, since the recommendation of a uniform rate of tax is closely dependent on the rate of tax emerging from this e xercise, the recommendation needs to be reconsidered.

    4 Common Markets and Uniform Rates

    As mentioned above, one of the important arguments driving the reform towards a comprehensive GST in India is to remove all impediments to forming a single common market covering all the states. The present regime with incomplete coverage at the state level in VAT, along with limitations on input tax credit and sourcebased taxation of interstate transactions, has segmented the market, thereby denying the benefits of economies of scale to producers in the country. While the desirability of such a goal is not being disputed, it is important to understand what is absolutely essential for such a regime.

    Most definitions of a common market refer to a group of entities, usually nations, which eliminate or reduce barriers on the


    movement of goods and services as well as productive inputs – capital and labour within the group. As the above suggests, a structure of taxes introduces some barriers to the movement of goods and services in a country. For instance, in India, since i nterstate sales are subject to central sales tax (CST), but consignment transfers were not subject to CST, it was optimal for firms to set up depots in all the states so as to convert all supplies to the state into consignment transfers and avoid CST. Similarly, with the introduction of state VAT with input tax credit limited to local purchases, every dealer likes to purchase locally, thereby discriminating between local purchases and interstate purchases. What these suggest is that these distortions to investment decisions can be avoided if taxes on purchases remain the same regardless of where the purchases are sourced from, and taxes on all sales in a jurisdiction remain the same regardless of the location of the supplier.

    The above will imply that an investor will buy inputs from wherever they are best sourced, without consideration of tax costs. In the context of the GST or VAT, it will require that input tax credit is available for all input taxes, no matter in where the taxes are paid. Further, an investor will choose a location for the enterprise based on economic parameters. There is no additional market access advantage any particular location provides. Wherever the enterprise is located, the tax costs of supplying to any given state will remain the same. These features are often r eferred to as a destination-based tax regime – where the tax on any transaction accrues to the place where the good or service is finally consumed, irrespective of where it is produced.

    The IGST design provides a good mechanism for setting up a destination-based tax regime. While taxes are collected on interstate sales by the exporting dealer, the revenue so collected is transferred to the importing state, where the importing dealer is allowed an input tax credit against subsequent sales/supplies, thus making them equivalent to local purchases in terms of tax treatment. For sales, all supplies made in a state will be taxed at the same rate.

    It should be noted that while all sales in a state should be subject to the same taxes, so far, the argument does not require that the taxes be the same in all the states. As long as the integrity of the tax system in any given state can be maintained, the above system can allow the rates of tax to vary across states without compromising on the destination principle. A higher rate in one state when compared to another does not provide an opportunity for relocation because wherever the firm may be located, its sales in the two will face higher and lower taxes respectively. In other words, uniform rates are not necessary for the sustenance of a common market.

    If there are differences in rates of tax between neighbouring states, it is possible that consumers may choose to travel to the low-tax jurisdiction and make purchases. This, in turn, could lead to rate wars among states. Most arguments for uniform rates arise from apprehensions of such an eventuality. It should be mentioned that while states in India did go through an extended period of rate wars and competition through tax-based industrial incentives, they could hammer out a solution to the problem in the form of the floor rates regime, first implemented in January 2000. While the regime was not completely foolproof and there were some violations by states, by and large, it is considered a fairly successful reform initiative, which resulted in improved revenue performance for states. These deviations, as well as d eviations from the original design of state VATs through the adoption of higher rates of tax in the years beyond the compensation period, do suggest that while floor rates will be adequate to prevent rate wars, there is the need for some mechanism to e nsure that the states conform to them.

    While the above suggests that a common market does not r equire uniform rates of tax, it is quite commonly accepted that uniform rates are “desirable”. Uniform rates and procedures would of course reduce compliance and administrative costs. However, imposing uniformity on the states would severely limit their fiscal capacities. The revenue base varies considerably across states. If one attempts to derive a revenue neutral rate for different states assuming that the present classification of goods remains unaltered, the lower rate for goods is 6% and for services, 8%. One would find considerable variation in the rates required to protect the revenues of individual states (Table 1). One primary feature that the table highlights is that while for more than half the states a rate of 9% to 10% does provide a dequate

    Table 1: Revenue Neutral Rates for GST: A Comparison

    Scenario I Scenario II
    Single Rate Two Rates Single Rate Two Rates
    Andhra Pradesh 8.93 13.47 8.37 12.06
    Arunachal Pradesh 7.17 8.69 6.21 5.23
    Assam 7.29 9.05 6.39 5.95
    Bihar 10.10 17.73 8.02 11.10
    Chhattisgarh 10.51 17.75 9.46 15.36
    Delhi 9.63 15.61 8.32 12.07
    Goa 11.31 19.76 10.28 17.65
    Gujarat 10.01 16.36 9.06 14.14
    Haryana 10.72 18.22 9.72 16.04
    Himachal Pradesh 9.12 14.13 7.99 10.97
    Jharkhand 12.00 22.08 10.53 18.98
    Jammu and Kashmir 8.27 11.73 7.50 9.56
    Kerala 10.01 17.72 7.75 9.93
    Karnataka 9.28 14.43 8.37 12.11
    Madhya Pradesh 8.65 12.81 7.54 9.55
    Maharashtra 8.79 13.23 7.64 9.87
    Manipur 6.54 6.52 5.30 0.75
    Meghalaya 7.87 10.63 6.52 5.68
    Mizoram 5.98 4.16 4.58 -4.12
    Nagaland 6.57 5.67 4.78 -5.52
    Orissa 9.55 15.24 8.45 12.41
    Punjab 10.02 16.60 8.79 13.51
    Rajasthan 8.71 12.93 7.82 10.49
    Sikkim 8.08 11.25 6.85 7.11
    Tamil Nadu 9.81 15.88 8.79 13.40
    Tripura 6.45 6.09 5.13 -0.47
    Uttar Pradesh 9.30 14.64 8.12 11.40
    Uttarakhand 7.53 9.75 6.74 7.21
    West Bengal 8.96 13.97 7.28 8.31
    Average Rate 9.34 14.71 8.23 12.10

    Figures derived using finance account data on tax collections and estimates of service tax from Rao and Chakraborty (2010). Both scenarios assume that real estate transactions remain outside the GST. Further, scenario I reflects no taxes from the IT sector and full credit for financial services. Scenario II has some local demand from IT yielding revenue and partial credit for financial services.

    november 27, 2010 vol xlv no 48


    revenue, there are at least five states that require rates higher than 12%. Choosing a high rate will provide a bonus to the lowtaxed states, while choosing a low rate will mean a revenue loss for some other states.

    The central government as well as the THFC argue that adequate provisions will be made in the budget to compensate for any loss in revenue. However, it is important to emphasise that this revenue shortfall, if any, results from a structural deficit for the state and cannot be treated as a transitional issue. Compensation packages are short-term hand-holding exercises to provide comfort and confidence to the affected agent so that the reform process can be smooth. However, if there is an underlying structural deficit that a state faces, then adopting uniform rates of tax will be akin to suggesting that this state continue to remain dependent on transfers from various arms of the central government on a sustained basis. A state like Jharkhand, for instance, will lose more than Rs 500 crore if CST is eliminated. The bulk of this is derived from mineral exports from the state. Since compensating revenue is to come from services, if the services base is not large enough, Jharkhand will not be compensated in terms of its revenue potential. There are no answers available if one looks for them in a uniform rates regime. However, if the rates are a llowed to vary beyond a floor, with a fixed classification of commodities and services, the states can be assigned the responsibility of choosing an appropriate rate to ensure their revenue r equirements are met. In other words, states can be made more responsible for their finances.

    5 Concluding Remarks

    Given the approach adopted by the THFC and the developments that have taken place before and since its Report was submitted, its interventions in the GST arena have been rendered irrelevant. While the union government did initiate discussions on the lines suggested by the THFC, the discomfort voiced by states has changed the trajectory of the talks. In the dialogue until the s econd draft of the constitutional amendment bill, the centre has given up the veto power proposed for the union finance minister and also agreed to dilute the demand for uniformity. While what exactly this entails will be revealed in the days to come, what is apparent is that the highly centralised model of the THFC is not in play anymore.

    While there is some open-endedness on what the GST regime will mandate as essential for all states, it is desirable to retain some commitment or conformity on the following issues.

  • Classification of goods and services: While the states may be allowed to choose the rates associated with different categories of goods or services, the classification of goods and services into different categories should remain the same across all states and the central government.
  • Homogenisation of forms and procedures: This would facilitate improved compliance and easy administration.
  • Floor rates: These should be prescribed for all activities to minimise the damage from tax competition. International experience suggests that such regimes can be implemented, Canada and the European Union (EU) are two clear examples. The Canadian Harmonised Sales Tax (HST), initially introduced at a
  • Economic & Political Weekly

    november 27, 2010 vol xlv no 48

    u niform rate in three provinces, now covers two more provinces and works with three rates of tax. While the rate in most of the provinces is 13%, one chose to join the HST regime with a tax rate of 12% and another to raise it to 15%. All these taxes are implemented by a single agency. On the other hand, the EU experience is based on the floor rates principle. The Sixth Directive and its recast version of 2006 prescribe a floor on the rates of tax and on the classification of commodities into different categories. For instance, while these directives allow one or two lower rates to be in operation, the commodities or services on which they can be applied are listed.


    1 The prescribed list of exemptions are unprocessed food and public services provided by all governments, excluding railways, communications and publicsector enterprises, service transactions between an employer and an employee, and health and education services.

    2 No reference to this index could be found in the literature.

    3 This would be an overestimate to the extent some household investment is in a business enterprise in the form of a partnership or proprietorship. While this would be called household or unorganised sector for other purposes, it could be a part of the tax net, given the manner in which the exemption threshold is defined.


    Finance Commission (2009a): “Report of the Task Force on Goods and Services Tax”, Thirteenth Finance Commission,

    – (2009b): “Report of the Thirteenth Finance Commission” (2010-2015), http:// Rao, R Kavita and Pinaki Chakraborty (2010): “Goods and Services Tax: An Assessment of the Base”, Economic & Political Weekly, 2 January, pp 49-63.

    Villela, L, Andrea Lemgruber and Michael Jorratt (2010): “Tax Expenditure Budgets: Concepts and Challenges for Implementation”, Inter-American Development Bank, Institutional Capacity and Finance Sector, Washington DC, http://


    On Remembering Lohia – Yogendra Yadav Lohia’s Socialism: An Underdog’s Perspective – Sachchidanand Sinha

    Understanding Capitalism through Lohia – Sunil Understanding Lohia’s Political Sociology: Intersectionality

    of Caste, Class, Gender and Language – Anand Kumar Context, Discourse and Vision of Lohia’s Socialism – Rajaram Tolpadi Many Lohias Appropriations of Lohia in Karnataka – Chandan Gowda

    Lohia as a Doctoral Student in Berlin – Joachim Oesterheld

    What Is Living and What Is Dead in Rammanohar Lohia? – Yogendra Yadav

    For copies write to: Circulation Manager,

    Economic and Political Weekly,

    320-321, A to Z Industrial Estate, Ganpatrao Kadam Marg, Lower Parel, Mumbai 400 013. email:

    Dear Reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here


    To gain instant access to this article (download).

    Pay INR 50.00

    (Readers in India)

    Pay $ 6.00

    (Readers outside India)

    Back to Top