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

A+| A| A-

Goods and Services Tax: Some Progress towards Clarity

The first discussion paper on Goods and Services Tax released by the Empowered Committee is an important step in signalling the consensus and commitment to harmonise the indirect taxes levied by the states and the centre and in traversing some distance in clarifying the design and implementation aspects of the new tax regime. The efforts of the EC must be complimented for building a consensus on many a contentious issue in the process of evolving the GST in the country. However, much more remains to be done, many of the design and implementation issues are yet to be negotiated and settled and it would take considerable time and effort before they are finalised.


Goods and Services Tax: Some Progress towards Clarity

M Govinda Rao

the recent reforms to introduce the VAT at both the central and state levels, a considerable amount of cascading continues. The high threshold and large list of exemptions in the central value added tax (CENVAT) and non-inclusion of a number of services in the service tax do not allow

The first discussion paper on Goods and Services Tax released by the Empowered Committee is an important step in signalling the consensus and commitment to harmonise the indirect taxes levied by the states and the centre and in traversing some distance in clarifying the design and implementation aspects of the new tax regime. The efforts of the EC must be complimented for building a consensus on many a contentious issue in the process of evolving the GST in the country. However, much more remains to be done, many of the design and implementation issues are yet to be negotiated and settled and it would take considerable time and effort before they are finalised.

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

here has been considerable speculation on the architecture, engineering and management aspects of the proposed goods and services tax (GST) in India. The reason for this has to be found in the fact that, so far, the stakeholders have not been taken into confidence on the deliberations in the Empowered Committee (EC) of State Finance Ministers. From this perspective, the first discussion paper on GST released by the EC is an important step in signalling the consensus and commitment to harmonise the indirect taxes levied by the states and the centre and in traversing some distance in clarifying the design and implementation aspects of the new tax regime. This will facilitate better understanding and discussion by various stakeholders and facilitate meaningful interaction before finalising the design of the tax.

At the outset, it is necessary to note that the GST will not be a new tax. It is only a further improvement over the prevailing consumption tax systems at the centre and states. At present, there is a value added tax (VAT) at the manufacturing stage on goods and a separate tax on selected services at the centre and an intra-state VAT up to the retail stage at the state level. R eforms over the years have gone some distance in reducing cascading by providing for input tax credit. Nevertheless, the tax bases at both the central and state l evels are narrow, making the VAT chain incomplete with significant cascading e lements, remaining at both the levels.

At the centre, the tax base does not include value added at stages subsequent to manufacturing and leaves out a number of services. The states, on the other hand, cannot tax services except the ones that are specified in the state list under the Seventh Schedule to the Constitution such as the entertainment tax, electricity duty and passengers and goods tax. In spite of

december 19, 2009

the input tax credit chain to be completed and this causes considerable distortions due to cascading elements in the tax. S imilarly, at the state level, non-inclusion of services prevents completion of the chain as goods are used in providing services and services enter into the manufacture of goods, not taxing services prevents providing input tax credit to them and in the process some cascading has continued even after VAT was introduced.

The introduction of GST is expected to expand the base of the tax and reduce distor tions by eliminating input taxes further. Thus, GST is not a new tax, but s imply a more comprehensive VAT on goods and services. The discussion paper devotes considerable space to clarifying this.

There are at least three reasons why the first discussion paper put out by the EC must be welcomed. First, it reiterates the commitment to reform the consumption tax system towards evolving the GST. Second, it helps to initiate discussion on the new tax regime with various stakeholders especially, various business groups, tax experts and people at large. Finally, the discussion paper goes some distance in clarifying the design and i mplementation aspects of the new tax r egime. Although this is short of expectations, the process initiated is important and hopefully there will be more discussion papers as the reform process advances and various issues of the new tax r egime get finalised.

Salient Features

The discussion paper details the basic features of the GST structure and implementation aspects. Of course, some of the d etails were already known and a few

  • o thers help to clarify some details of the proposed design. Some of the salient features of the proposed GST reform are:
  • (i) The reform entails a dual GST – one at the central level (CGST) and another at
  • vol xliv no 51

    Economic & Political Weekly


    the state level (SGST). Each taxpayer will be allotted a PAN-linked taxpayer identification number with 13/15 digits. The dual GST will be implemented through multiple statutes – one for CGST and one for each SGST. The new tax system would have a broader base to include wholesale and retail trade in the case of the CGST and all services in the case of the SGST. The basic features of the acts, including definitions and the classification system, are expected to be uniform among the various state statutes “as far as practicable”. To the e xtent feasible, a uniform procedure will be followed for the collection of both CGST and SGST.

    (ii) There will be separate tax administrations at the centre and in the states. I nput tax paid on CGST will be credited against the output tax on CGST and those on SGST will be credited against the SGST. Cross-utilisation of input tax credit (ITC) between CGST and SGST will not be allowed. Thus, there will be some cascading in the tax system.

    (iii) The CGST will have a threshold of Rs 1.5 crore. The threshold for the SGST is proposed at Rs 10 lakh, which would be uniform for all states and union territories. A simplified tax at 0.5% of the turnover will be levied for dealers with a turnover up to Rs 50 lakh. In the case of CGST, however, the threshold is proposed to be kept at Rs 1.5 crore in the interest of small traders and small-scale industries.

  • (iv) The CGST will subsume central excise duty, additional excise duty, excise duty on medicinal and toilet preparations, service tax, additional customs duty (countervailing excise duty), special additional duty on customs, surcharge cesses and surcharges. The SGST would subsume VAT/sales tax, entertainment tax, luxury tax, taxes on lottery, betting and gambling, state cesses and surcharges and entry tax not in lieu of octroi. A majority of the states would like to have the purchase tax also subsumed under the SGST, whereas some states which receive substantial r evenue from such taxes do not want it to be subsumed. The decision of the EC is that in case it is subsumed, the centre should give “adequate and continuing” compensation.
  • (v) Alcoholic beverages will be kept out of the CGST. The states can continue to levy sales tax/VAT as per the existing
  • Economic & Political Weekly

    december 19, 2009

    p ractice in addition to levying state excise duties. However, tobacco products will be included in the SGST with proper input tax credit. In addition, the central government can levy excise duty.

    (vi) The prevailing practice in regard to the taxation of crude, motor spirit (including aviation turbine fuel) and high speed diesel (HSD) will continue and these items will be kept outside the SGST and the tax would continue to be levied on these products with a floor rate. A final view on whether natural gas should also be kept outside the SGST will be taken after further deliberations by the EC. Thus, cascading on account of this will continue. Since, on an average, states receive over 30% of sales tax revenue from these products and as these are largely marketed through the public sector oil marketing companies, a dministering the tax on these items is easy and compliance of the tax is high. Surely, the states have decided to live with distortions in the interest of revenue.

    (vii) States will have concurrent powers to levy the tax on services. In the case of services of an inter-state benefit span, if these are intermediate services, a model of integrated GST (IGST) has been introduced to ensure seamless trade while making the system destination-based. However, if these are in the nature of final consumption, it is not clear how the revenues will be apportioned between the states where the service is produced, transacted and consumed.

    (viii) The important feature of the GST scheme is the mechanism to ensure a common market. To ensure seamless trade across the country, the discussion paper puts forth the IGST model. In this model, the inter-state seller of goods and services will pay the IGST, which will be equal to the total of CGST and SGST a fter taking credit of the input taxes to the central agency, to be created especially to administer the IGST. The exporting state credits the input tax revenue to the account of the importing state. The central agency will credit the SGST c ollected by it to the account of the i mporting State and the importing dealer will collect the SGST on his sales after taking credit of the tax already paid and the chain will continue. The process not only ensures seamless trade and

    vol xliv no 51

    u ninterrupted ITC chain but also enables the levy to be destination-based.

    Although the discussion paper goes to a considerable distance to provide conceptual clarity on a number of issues the available information on the proposed GST system brings out some inadequacies. It is well known that in all federal fiscal systems, tax assignment has to strike a compromise b etween fiscal autonomy and tax harmony and some inefficiency may have to be a ccepted. However, there are others which hopefully, after due consideration will be ironed out.


    Some of the important shortcomings of the proposed GST are summarised in the following.

  • (a) First, the discussion paper does not specify the list of exempted goods and services. The list of exempted goods and services is yet to be finalised and it is quite likely that some discretion may be allowed to individual states. This is a matter relating to the fiscal autonomy of the states, but to the extent that there is no uniformity, both administrative and compliance costs will be higher.
  • (b) Second, two rates are proposed for the SGST – one a low rate for essential items and another a standard rate for the remaining goods and services. The paper also advocates a similar approach to CGST. This increases both administrative complexity and compliance costs besides creating classification disputes. Indeed, there is considerable evidence across the world to show that levying a GST at multiple rates does not improve equity. First, the classification of goods and services is done according to judgments on income elasticity of demand. Even if they are correct, in general equilibrium terms, employment intensity of a good or service may be different from income elasticity of demand. Thus, taxing goods and services at multiple rates instead of a single rate (in addition to exemption) may decrease rather than increase employment. Surely this is a socio-political choice exercised by governments, but it is necessary to know the economic cost of this decision.
  • (c) At the state level, the proposal still leaves open the possibility of levying entry tax in lieu of octroi as also octroi. Similarly,

    it does not include stamp duties and registration fees. Furthermore, entertainment tax, if levied by local bodies (Kerala) will continue. Thus, while the proposal goes a long way in unifying multiple taxes, it still leaves out some taxes. Indeed, it is important to ensure that revenue sources of local bodies are protected. The more r ational course would be to add an additional percentage point to SGST as a local levy and distribute the proceeds to urban and rural local governments based on their consumption shares. Maharashtra, the only state in which municipal corporations are allowed to levy octroi too can abolish it.

  • (d) This is the opportune time to correct some of the design faults that exists in the prevailing state VAT. One of the problems with the present design is the distinction made between inputs and outputs and levying the tax on the former at 4% and the latter at 12.5% even as it is well known that the essential principle of VAT is providing credit for input taxes. First, taxing inputs and outputs at different rates is unscientific for, what is input in one use can be an output in another. Sugar, for example, is an input for a restaurant whereas it is an output for households. Second, an 8.5 percentage point margin of difference in the rates provides sufficient incentive to evade the output tax. Thus, a manufacturer of steel furniture, for example, will buy his input – steel – and pay the tax at 4%, but can suppress his output of steel furniture and evade paying the tax at 12.5% of the output value.
  • The fact that most states do not yet have reliable information system to match input and output transactions reduces the probability of detection. In any case, no tax administration can match each transaction and when South Korea tried to do it, it created a chaotic situation. If the inputs too are taxed at the same general rate of 12.5%, the incentive to evade will be much less because the tax saved from evasion will be only on the value added at that stage. Hopefully, this design fault will be corrected in the GST.
  • (e) For the reasons explained above, under GST, the concept of “declared goods” in the Goods of Special Importance Act, does not have a place and the only criterion for rate differentiation to be followed,
  • if at all, is on the basis of income elasticity of demand. As mentioned above, even this need not ensure overall equity in the general equilibrium sense; but tax design is less of a science and more of exercising socio-political judgments. Since under GST input tax will get the credit, there is no need to maintain a special treatment for “goods of special importance”.

  • (f) Another design fault in the state VAT pertains to the special cases of taxing pharmaceuticals and works contract. At present in the case of pharmaceuticals, the state VAT is still levied at the first point by taking stamped maximum retail price (MRP) as the retail value. Similarly, there is special treatment of works contract in which, the value added is derived in a presumptive manner. It would be desirable to avoid breaking the VAT chain by introducing such special cases and treat all goods and services alike under GST.
  • (g) There are far too many design problems in the prevailing CENVAT and a lot of work needs to be done before a clean and scientific GST is put in place. It has too many rates; it treats services and goods separately; it treats commodities covered under the MRP differently from others. The tax refund on exports does not follow a clean zero-rating mechanism. There are various schemes to refund the excise duty paid on exports like “duty draw-back” and “duty entitlement pass book” (DEPB) which is essentially presumptive. A p resumptive scheme is necessary when the system is complicated and when the tax administration does not have the r equired information and invariably the scheme will be adopted by the businesses only when they stand to gain more from such a scheme than from receipt of the a ctual refunds. Once created, it generates special interest groups to continue the scheme and not surprisingly, there has been considerable pressure to continue the scheme by the businesses even after its expiry and it had to be extended from time to time.
  • Challenges in Implementation

    While the discussion paper traverses some distance in providing clarification on the new tax regime, and this is particularly true of the treatment of inter-state transactions, there are many other areas which

    december 19, 2009

    are still unclear. Furthermore, much more work needs to be done before a clear picture of the new tax system emerges. Some of the challenges to be faced in implementing the GST regime are summarised in the following:

  • (1) The first major hurdle to be crossed in implementing the GST is enacting the Constitutional amendment to enable the centre to expand the tax base to include wholesale and retail transactions and to empower the states to levy the tax on services. Since, both the centre and the states are agreed that the switchover is desirable, this may not be a major hurdle, but the procedure itself will take time.
  • (2) It is clear from the discussion paper that the most important requirement for introducing the GST is the setting up of the centralised agency to settle the IGST claims and work as a clearing house. The system to track inter-state transactions in goods and services is critical to ensure the success of SGST and unfortunately, we have to begin this from scratch.
  • The attempt to develop it through the Tax Information Exchange System (TINXSYS) was not successful as only some states opted for it. The problem is not merely one of setting up the IT system. It is necessary to evolve a system to achieve a total solution to deal with inter-state trade in goods and services. This is a major challenge and the entire success of GST will hinge on this. An agency like the National Security Depository (NSDL) which has done commendable work in the case of setting up the tax information network (TIN) in income tax, which led to a sharp increase in revenues should be contracted early so that they can work out a competent system of tracking inter-state transactions and ensure smooth working of the clearing house mechanism.
  • (3) So far the taxation of services levied by the centre has been a selective tax. This has made the estimation of a revenue neutral tax rate difficult as some services with significant revenue potential continue to be excluded from the tax system, the most prominent being a tax on railway fares and freights. Besides the challenge of extending the tax to all services, there are also issues of defining “service”.
  • (4) It is necessary to estimate the revenue neutral rate of the tax at the centre
  • vol xliv no 51

    Economic & Political Weekly


    and in each of the states. This is necessary not only to arrive at convergence on the tax rates to be levied but also to negotiate and finalise the compensation plan. In any case, to ensure success of the tax and evoke high degree of compliance, it would be useful to keep the rates low, but that would require the centre taking a lead in assuring the comfort of insurance against fall in revenue at least for the next three years by promising to compensate the states for the revenue loss.

    It may be recalled that when Thailand introduced VAT in January 1992, the estimated revenue neutral rate was 10%, but they decided to levy the tax at 7%. Not only did they realise the full revenue at this lower rate, but their income tax revenue too went up by 25%! During the Asian financial crisis they had to increase the rate to 10% under International Monetary Fund pressure, but once they came out of the crisis, they reverted back to 7%!

    Eventually, the success of GST will depend on improved compliance, higher productivity growth leading to faster economic growth and buoyant revenues. Hopefully, both the centre and states will realise this and keep the rates low.

  • (5) Estimation of the revenue-neutral general rate would require the estimation of the tax base in each of the states and since the threshold for the centre and states are different, of the tax base of the centre as well. This requires a decision on the list of exempted goods and services, goods and services subject to a lower rate of tax and the lower rate of tax. Further, the levy of GST at the state level entails giving concurrent powers to the states to levy a tax on services. As there are many inter-state services which are produced in some state, sold in some other state and consumed in yet another state, it is necessary to finalise the rules of revenue appropriation to determine the size of the tax base. The matter could involve protracted negotiation and bargaining before it is settled.
  • (6) The discussion paper is silent about how the “hard to tax” sectors will be dealt with. In particular, taxation of the financial sector, real estate and housing, is complex and unless carefully designed and implemented, it can be a source of distortion. Often, it is difficult to separate
  • Economic & Political Weekly

    december 19, 2009

    financial intermediation, which is a genuine development activity, from financial services. In some cases, it is possible to identify the service charges and in others it is inextricably mixed up with profits. In the case of general insurance, for example, it is inappropriate to consider the premium paid as is being done now, for this does not constitute the service. It is necessary to net this out with the settlement of claims to arrive at the right base.

  • (7) The tax administrations, both at the central and state levels will have to prepare themselves to deal with transitional issues. Assessments and refunds arising from the prevailing tax systems will continue even as the transition is made to GST and settlement of these would require simultaneous administration of both the new and the old systems.
  • (8) A major challenge in the introduction of GST in the country is building the capacity of the administration to implement the new tax. The tax administration at both central and state levels will have to be augmented to deal with the new tax regime. Although GST is a natural transition from the VAT system, both CENVAT and state VAT prevailing at present are in substantial variance with the standard VAT system seen internationally, both in their design as well as in its implementation.
  • The administration will have to be geared to administer the destinationbased consumption type VAT on goods and services. Furthermore, there are hardly any clear cases of international experience of levying sub-national GST except in B razil and Canada and neither of these can be emulated in India. Preparing the administration to implement the GST would require considerable effort at building capacity on matters such as the a ccounting system, forms and procedures, assessment, auditing and computerisation. Unfortunately, very little administrative changes were made when the CENVAT at the centre and VAT at the state level were introduced and the administration continues to be organised on geographical rather than functional lines.
  • (9) Another important measure needed is to create a permanent institutional arrangement to negotiate, harmonise and monitor the reform process as well as the working of the new tax system. A permanent
  • vol xliv no 51

    bargaining and dispute resolution mechanism is important to ensure compliance of the agreements by all the parties concerned. So far, the EC has done a commendable task of forging consensus on many a contentious issue. Never theless, the institution is informal and needs to be placed on a much more secure footing.


    Reforming the consumption tax system in an intergovernmental context is an extremely difficult issue. In a globalising environment, it is necessary that the tax system generates sufficient revenues to provide efficient infrastructure. However, these revenues should be raised by minimising the three costs – the cost of collection, the cost of compliance and the economic costs arising from distortions. In an intergovernmental context, all these have to be achieved while ensuring fiscal autonomy to subnational jurisdictions. A completely harmonised tax system which is very much desired by the businesses can be achieved only at the cost of fiscal autonomy of the states. Complete harmonisation is equivalent to uniformity and this takes away the powers of the states to vary their public services and taxes.

    In Indian fiscal federalism, uniformity is being achieved by building consensus among the states by the EC. Nevertheless, this will reduce the ability of the states to vary the levels of public services and taxes. This is the trade off exercised between fiscal autonomy and tax disharmony,

    Thus, the first discussion paper of EC reaffirms the commitment to shift to a GST and helps to understand the broad contours of the tax. However, much more remains to be done and many of the design and implementation issues are yet to be negotiated and settled and it would take considerable time and effort before they are finalised. The opportunity should be used also to overcome some of the shortcomings in the existing design of VAT.

    It is important to prepare well before switching over to the new system to ensure sound fundamentals to the new tax system. The efforts of the EC must be complimented in weeding through the most difficult terrain of building a consensus on many a contentious issue in the process of evolving the GST in the country.

    Dear Reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Back to Top